From bobx at linuxmail.org  Sun Feb 17 10:02:20 2002
From: bobx at linuxmail.org (Bob)
Date: 17 Feb 2002 07:02:20 -0800
Subject: ActiveState version of 2.2???
Message-ID: <1001ff04.0202170702.6ebaaf55@posting.google.com>

When?


From pinard at iro.umontreal.ca  Mon Feb 18 12:31:17 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 18 Feb 2002 12:31:17 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: <a4racj$bm2$1@panix2.panix.com>
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
	<roy-05958C.07303117022002@reader2.panix.com>
	<ubsenehjm.fsf@online.no> <a4psgi$5jo$3@peabody.colorado.edu>
	<a4racj$bm2$1@panix2.panix.com>
Message-ID: <oqofim3dei.fsf@titan.progiciels-bpi.ca>

[Aahz Maruch]

> In article <a4psgi$5jo$3 at peabody.colorado.edu>,
> Fernando =?ISO-8859-1?Q?P=E9rez?=  <fperez528 at yahoo.com> wrote:
> >
> >Would you folks mind sticking to English in public discussions? Use private 
> >email for other languages, please.

> I think the occasional non-English post simply reflects the international
> nature of Python.

Exactement.  Bravo!  Bien dit! :-)

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



From gcordova at hebmex.com  Mon Feb 11 16:15:25 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Mon, 11 Feb 2002 15:15:25 -0600
Subject: Best python way to do 'object messaging'
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A27@SRM201002>

Have to taken a look at some of the available MUDs? I believe
some of them are scriptable as you say, you define an object
class, which can be anything in the game; you define lotsa
properties and methods which control the way it interacts
with the environment.

The objects you create can be weapons, plants, scrolls,
monsters, seemingly anything.

Good luck on your quest :-)
-- 
Gustavo C?rdova Avila
*  gcordova at sismex.com <mailto:gcordova at sismex.com> 
* 8351-3861 | 8351-3862



From sholden at holdenweb.com  Fri Feb  1 16:46:27 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 16:46:27 -0500
Subject: reading sendmail alaises
References: <m34rl2jh7m.fsf@moroni.pp.asu.edu>
Message-ID: <i3E68.110115$8b.4168188@atlpnn01.usenetserver.com>

"David Bear" <iddwb at moroni.pp.asu.edu> wrote in message
news:m34rl2jh7m.fsf at moroni.pp.asu.edu...
> I would like a quick way to resolve sendmail aliases.  I've thought
> about reading and parsing /etc/aliases myself but thats too much
> work.  Is there some module/interface to sendmail that would allow
> something like
>
> realaddress = sendmail.resolve-alias(name)
>
> Even though I could build a dictionary from aliases, the file may
> actually be in db format -- so I'm looking for a 'sendmail' way of
> grabbing them.
>
[root at headrat tmp]# /usr/lib/sendmail -bv testalias
sholden... deliverable: mailer local, user sholden
root... deliverable: mailer local, user root

so:

#! /usr/bin/python

import os, sys, string
f = os.popen("/usr/lib/sendmail -bv %s" % string.join(sys.argv[1:], " "))
names = []
while 1:
    line = f.readline()
    if not line: break
    name = string.split(line)[0]
    names.append(name[:-3])

print names

Sorry about the 1.5isms - nothing more recent to hand :-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From philh at comuno.freeserve.co.uk  Fri Feb 22 09:58:05 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 22 Feb 2002 14:58:05 +0000
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4> <d8cbfc61.0202220406.271bcb18@posting.google.com>
Message-ID: <slrna7cn3t.v6u.philh@comuno.freeserve.co.uk>

On 22 Feb 2002 04:06:42 -0800, Clive Page <cgp at le.ac.uk> wrote:
>Philip Swartzleonard <starx at pacbell.net> wrote in message news:<Xns91B6DFD6A36CDRASXnewsDFE1 at 130.133.1.4>...
>> 
>> Steve has mentioned that UNIX people are used to it... i think it's a 
>> little deeper than that, b'cause as far as i know, _every_ programming 
>> language that has ever supported octal does it this way...
>
>Not quite all of them.  Fortran (at least since Fortran90) allows the
>forms O"01234" or O'1234' i.e. an upper-case O and a quoted string of
>digits (but they are only allowed in data statements, unfortunately).
>
>Prior to Fortran90 there was no _standard_ way of expressing octal
>constants, but lots of vendors had their own extensions.  But I don't
>recall ever coming across anything as error-prone as the C notation
>with a leading zero.  Fortran is a much safer language than C; I
>recommend it as the nearest thing to a compiled form of Python.  :-)

I like the way Smalltalk does it: BrN

Where B is a base, 2 to 36, and N is a number in that base, e.g:

2r11110000, 8r377, 16rFF, etc.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From threeseas at earthlink.net  Sun Feb 17 12:57:51 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 17 Feb 2002 17:57:51 GMT
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk>
Message-ID: <12393.813T2860T7905870threeseas@earthlink.net>

On 16-Feb-02 13:13:54 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Sat, 16 Feb 2002 01:45:42 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>>On 12-Feb-02 10:46:49 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>>On Mon, 11 Feb 2002 18:37:51 -0600, Chris Gonnerman
>>><chris.gonnerman at newcenturycomputers.net> wrote:
>>>>
>>>>Indeed.  He uses a nonstandard definition of "language" and applies
>>>>"physics"
>>>>to programming.  He can't program, you know; ten years on this project and
>>>>nothing to show for it.  If I had tried to develop something like this for
>>>>ten years and had nothing to show for it but a website, I'd give up.
>>
>>>At least his code is all neatly lined up!
>>
>>Chris's comprehension level is greatly lacking and you've shown your
>>superiority attitude.
>>
>>Ten years???? it was Feburary 1988 when I identified the nine commands.

>14 years then. You do realise, that makes you look even worse.

>>I never claimed to be a programmer

>14 years is plenty of time to lesarn, if you want to.

>>and never wanted to make the VIC but
>>rather someone to use it.

>This implies you'd rather someone else coded it...

>>But I also don't, and never will want to be
>>abused by any programmer wanting to take credit and property rights to the
>>VIC.

>...but if they do, you want to take the credit and copyright for their
>code.

>>The computer industry has alot of ethical problems in this area of taking
>>everything the can from the users, from their ideas, to their debugging
>>efforts (feedback), to their money in payment for what came from them, and
>>now rental....and always it is never quite right but lacking always... for
>>upgrading recycling of consumer entrapment abuse.

>If you refer to Microsoft, I agree. But free software / open source *isn't*
>like that, in fact it is deliberately in opposition to that sort of thing.

>>It's good to know that such practice is going to soon come to an end.

>Because of VIC? Or because of something else?


You have made it perfectly clear that you do not like me or this project
and that you will even commit acts of deception in your taking out of
context things I have said and in the wrongful insinuations you make.

It's an open project Phil, under the GPL (and I reserve the right to
change that to a more secure license should there be found a weakness in
the GPL - Lawrence Lessig seems to be now going in a direction in
consideration of what I've said regarding a fundamental change in
intellectual property laws, and this includes Copyrights, in the
"Creative Commons"
http://www.sfgate.com/cgi-bin/article.cgi?file=/gate/archive/2002/11/creatcom.DTL
soooo there will be plenty of choices.) And if that ain'tr enough:
http://www.eetimes.com/story/OEG20020208S0079

I have already stated that I believe in giving credit where it is
deserved and more so than as it is explained in:
http://tuxedo.org/~esr/writtings/cathedral-bazaar where the lead of a
project gets more credit than they actually do, but from my stand
in that by using usenet and the unbiased date time stamped public
accessable archives it is possible to insure credit is given to where it
belongs, to those who openly help.

In the movie "The Matrix" (wonderful collection of metaphors) Neo goes
after Morpheus. Into a FEDERAL BUILDING... It's never been done before....
that's why it is going to work....

Usenet is that Federal Building Phil. Unbiased Date Time Stamped Public
Recorder enabler built by the federal government.

You want to contribute to this project and be assured you'll get credit
for it? Simple, post you work! International copyright law states that the
moment a work is published (Usenet counts) copyright law applies.

The 14 years you mentioned Phil, it doesn't make me look bad Phil, it
makes the computer industry look bad. How is it that such a collection and
integrated configuration of common functionality not have been recognized
and already done, unless the computer industry has gotten so detached from
genuine science? (most everyone knows it got very detached) And what was
it that cause it to become detached? Easy Money, all you have to do is be
dishonest, as Bill Gates said, "make people need you". And you do that by
not giving people the tools to allow them to do things for themselves, but
rather have them rely on, or need, you to do it for them (for a price, as
you have expressed here yourself.)

I took electronics in High School (early 70's when ICs began to really
filter into the market - But tube technology was still being taught) and I
aced it. Late 70's I took a "Minicomputer and Microprocessor technology"
course thru Capitol Radio Engineering Institute, where I aced thru as much
of the course [2/3] as I completed in a summer (before this world wide top
electronics corespondance school went belly up.) In 1990 I went back to
school [Dekalb Tech] to learn C programming, mainly for this project.
While working full time in a job that had nothing to do with computers or
programming. I made Deans List twice and it would have been 3 times
only that last term simply didn't have enough credit to qualify. I have a
4.0 in all computer electronic and programming courses I have ever taken.

But there seems to always be something missing from the needed resources
for me to do this project. Then it was the fact that I began to do it in
Minix but really wanted to do it on the Amiga, where the Amiga already
had a growing application, device, library list of things that had the
third user interface built in, as standard (The Application side door port
- or as I'm now calling it, the Automation Port Interface (API).)

OK so I learned C on a PC with all it single tasking junk. But wanted to
do the project on the multi-tasking Amiga, only that's a different beast
to write for and there ain't no school classes, only hard to use RKMs (ROM
Kernel Manuals) which I tried to use. I even tried to get a job in
programming, figuring it would help me to get and keep my mind geared in
the programmer mindset, only at that period of time we were dealing with
a recession and jobs were far fewer and employers where far more selective
knowing they could get greater experience and education for less money.
Certainly I made nobodies potential employees list, even in consideration
of grade and willingness to continue school. So I continued on working as
cabinet maker and due to lack of work got back involved in the Tradeshow
business of building booths. Not something that is supportive of the
programmers mindset. At some point in time while working as a cabinet
maker I set out to do some programming that would help the company
generate drawings and cut list, etc.. In the process I found a freely
available C library that could help me do this. In pursuing this library I
found it had been sold and some of it was being converted to assembly
while the rest of it was being rewritten and greatly improved and
enhanced. I had the opportunity to buy it before it was released at good
discount, so I did. The time went by and it wasn't released, more time and
still not released.... two years later I had to insist the company send me
a copy of the finally release version of what I had paid for, even though
I could not use it due to the system requirement to run the library having
changed to system upgrade I could not afford. Not to mention that the
project died a year earlier due to failure to deliver by the company by
set date.

There are alot of people in this world Phil, Not everyone can be a
programmer, or we'd all starve. But that doesn't mean that programmers
have a right to abuse the rest of us, treat us like coppertops, in not
doing their job to automate the things we need automated and most
certainly can be, in order to allow us to do alot more. And that includes
programming.

Programmers and in some cases even robotics (controlled by programming)
have taken jobs away from people. Sooooo What, people who lost their jobs
found other ones. And in the software industry, programmers changing jobs
is not uncommon. Sooooo What! that auto-coding would cause the elimination
of jobs. In fact another kind of "autocoding" is going on by programmers
that is resulting in the loss of jobs in the field of medicine. Medical
Transcription is being replaced by autocoding that is created by
programmers.

It's time for psuedo Programmrs to get in line for losing their jobs. I'm
not heartless, they will find other jobs (perhaps as automaters).
Heartless is promising something that will help and empower others to do
things for themselves but never really delivering. There is no obscene
enough language to describe some of the deceptitive actions that go on in
the computer industry, and we all know it's level of dishonesty is high,
thanks to cases like MS vs. DOJ. There even exist public record of such
statement as "sure we are greedy" by well known people in the computer
industry. There is public recoder of a great deal of dishoinesty and
deception in the computer industry.

It's really long past the time for this childish sandbox greed to stop.

There is more to this project than just the tool. The tool is really very
small and simple to all that is possible to convert, build and create for
and thru this tool. Consider a compiler, what good is it without all the
support libraries, such as even .net and C# are small in themselves but
limited without all the support stuff.

But the difference here with the VIC is that it opens the door up, takes
them off their hinges (blows the doors off the coders exclusive elitist
elevator), and allows alot more people to become involved, like typical
users even, or should I say the ones who need to and have a right to know
and do for themselves.

Phil, you think 14 years makes me look bad? But the truth is that you have
a really messed up bad attitude and intent (even regardless of how much
proof you have supplied to show that). What you should be seeing is that
I've given the computer industry 14 years to correct their crap. You
really should see and understand the last thing Neo Said in the Movie "The
Matrix". .

What good is it to the users/consumers if the industry holds control over
what the users/consumers can and can't do for themselves?

The VIC is a small tool, smaller than many shells. To have it put together
out in the openness of the automatic credit giving machinery of Usenet,
under the GPL side of the spectrum, with the help of those in the
industry...

Like I said before, there will never be enough programmers to do all the
things the users of computers want to and need to have done. But there is
the exact number of users of computers to do all the things the users of
computers want to and need to have done.

And to knock that psuedo pedistal out from under you Phil, you too, are a
user of computers. You are NOT Excluded!

How will I know that sandbox greed of the industry has come to it's end?

How do you suppose I can help it come to an end?

I've been telling people what you and those like you and follower of such,
do not want the people to hear, and after a decade there is some good
evidence I'm clearly being heard and even that people, the masses, are
ready for the greedy of the computer industry to be looking for another
job.

So stop being manipulatingly greedy, Phil. There is plenty for everyone to
do.

14 years ago I would have prefered to have been using the VIC to do the
things I wanted to do and to have been doing so all these years. But
that's just not the job I've been given in this matter, due to the greed
of the computer software industry.

If you want to be greedy Phil, to take that job away from me, then give me
the tool under the conditions of which I want it.

I am of the alpha and the omega, the great consumer, the first to consider,
the last to receive. And as the Consumer, I am always right in having the
right to become a producer. The Tool Phil, hand it over, all of it,
including the genuine "we all are in this together" heart felt support.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From tdelaney at avaya.com  Sun Feb 17 18:43:38 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 18 Feb 2002 10:43:38 +1100
Subject: Have to try
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A18D@natasha.auslabs.avaya.com>

From: Rick Hamilton [mailto:rhamilt at pacbell.net]

> OK - I know I am going to get fried here but -
> I am out of options. I'm a programmer (Java,
> VB and C++) and am taking a class on Operating
> systems. The nut teaching this course is asking
> us to do a ton on programming in Python. OK -
> So I am trying to do a (what should be) simple
> thing. I have a string stored in a variable. I
> want to search that string for the occurrence of
> a sub string. I have tried 
>
> Find(s,sub)
> Rfind
> MyString.find(sub)
> 
> Nothing - I have not yet figured out the appropriate
> way to treat a string and am heading for a very weak
> book that we have as a reference. Anyone will to toss
> a fellow programmer a helping had I'm willing to get
> flamed to get this over with.

Python Documentation
http://www.python.org/doc/


and in particular, The Python Tutorial
http://www.python.org/doc/current/tut/tut.html

Tim Delaney



From l0819m0v0smfm001 at sneakemail.com  Sat Feb  9 20:42:33 2002
From: l0819m0v0smfm001 at sneakemail.com (Joshua Macy)
Date: Sun, 10 Feb 2002 01:42:33 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com>
Message-ID: <3C6596B6.20008@sneakemail.com>

wooks wrote:


>>One of the basic philosophies of Python design is that it is wrong for 
>>the language to guess when presented with something that's ambiguous. 
>>
> 
> Which seems to be the opposite of the Perl philosophy - we will try to
> do the right thing.
> 
> 
>>That something is "automagical" is usually a swear word in Python.  This 
>>can make Python seem hairier than other languages which cheerfully take 
>>a stab at what the designers think you may have wanted, but the Python 
>>belief is that those languages are leading you down the garden path, and 
>>sooner rather than later you'll be bitten by the subtlety that the other 
>>language tried to gloss over.
>>
>>
> 
> Hmmm. The good thing about the Perl approch is that (if you read the
> right resource) you get an explanation of the process via it tries to
> do the right thing, which is very effective at getting the conceptual
> message across.
>  



   Yeah, I came to Python from Perl, too, and I used to think that all 
that context sensitivity was the cat's pyjamas--but after the second 
two-day debugging session caused by Perl guessing wrong (or my guessing 
wrong about what Perl would guess), I wasn't so thrilled.  I actually 
tried Python because I bought the O'Reilly book on Advanced Perl 
Programming, and in the author's intro he said really nice things about 
Python.  I went out and looked at Python online, then bought Programming 
Python, and never looked back.  The Perl book is still sitting unread on 
my shelf....


> 
> I was very specific. Is there a forum for the dumber less demanding
> Python user? I got answers that I consider helpful. Thanks to you and
> Sheila.
> 


   Well, glad to be of help.


   Joshua



From core.lists.python at core-sdi.com  Mon Feb 11 13:43:04 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Mon, 11 Feb 2002 15:43:04 -0300
Subject: IPX
Message-ID: <019701c1b32b$f0b676e0$8642a8c0@corelabs.coresdi.com>

Hi there!

Could anybody send me an example of a client/server pair of python scripts to do

(client side)
send "potato"

(server side)
You've said "potato"
send "poteito"

(server side)
You've said "poteito"
 
...... but using IPX.

Thanks! 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020211/6756f551/attachment.html>

From mcherm at destiny.com  Thu Feb 14 13:55:13 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Thu, 14 Feb 2002 13:55:13 -0500
Subject: preference tree like in jdk 1.4
Message-ID: <3C6C0811.2010900@destiny.com>

> i was thinking of an empty function:
> 
>   def empty(): pass
> 
> and then use it to build complex tree of preferences:
> 
>   pref = empty
>   pref.personalData = empty
>   pref.personalData.firstName = 'John'
>   pref.personalData.lastName = 'Cleese'
>   pref.windowSettings = empty
>   pref.windowSettings.windowA = empty
>   pref.windowSettings.windowA.x = 200
>   # and so on...
> 
> 

> to store or load this pref object, just use the pickle mechanism.
> 
> my question is, is there a simpler, more beautiful, elegant way?


Yes.

Instead of using an empty function, use something designed specifically 
for holding fields: a class!


 >>> class Empty:
	pass
 >>> pref = Empty()
 >>> pref.personalData = Empty()
 >>> pref.personalData.firstName = 'John'
 >>> pref.personalData.lastName = 'Cleese'
 >>> pref.windowSettings = Empty()
 >>> pref.windowSettings.windowA = Empty()
 >>> pref.windowSettings.windowA.x = 200

-- Michael Chermside

(Note: there may be fundamentally better solutions too; I just wanted to 
get away from mis-using functions.)





From thomas at weholt.org  Wed Feb 20 18:01:18 2002
From: thomas at weholt.org (Thomas Weholt)
Date: Wed, 20 Feb 2002 23:01:18 GMT
Subject: XML-RPC over HTTPS/SSL
References: <8rKc8.4$5vh.135400448@news.telia.no>
Message-ID: <2VVc8.7284$%m1.150721@news4.ulv.nextra.no>

Ok, after a few hours of "hard" work I finally got it working, even supports
gzipped xmlrpc-responses and basic user authentication. The code is
available at http://home.online.no/~t-weh/releases/xmlrpcauth.py and is no
more than a modified version of the post I found using google ( address
below ).

I didn't see the replies until now so I haven't had time to check those out
yet. Probably learned something while hacking at the code so it wasn't a
total waste of time ( hopefully ).

Anyway, thanks for your replies.

Best regards,
Thomas Weholt

"Thomas Weholt" <thomas at gatsoft.no> wrote in message
news:8rKc8.4$5vh.135400448 at news.telia.no...
> I got a xmlrpc-server running using M2Crypto, making it encrypted with
SSL.
> I cannot get any xmlrpc-client to connect though, no matter what.
>
> Looked at resources found searching google, for instance
> http://lists.zope.org/pipermail/zope/2001-May/089457.html. but with no
luck.
> Does anybody have any code showing how to connect a client to a
> xml-rpc-server using SSL/HTTPS , preferrably with user authentication
also?
> ( to make it perfect, it would accept gzip-encoding to speed up transfers
> :-) )
>
> I'm using xmlrpclib from pythonware.com.
>
> I keep seeing stuff about SSL/HTTPS-support in standard python, but ... ??
> Do you compile the interpreter with OpenSSL etc ?? My clients and servers
> will most probably run 60% on Linux and 40% windows, and re-compilation
> isn't always feasable for windows-users.
>
> Any clues or hints appreciated
>
> Best regards,
> Thomas Weholt
>
>




From fperez528 at yahoo.com  Fri Feb 22 15:11:43 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Fri, 22 Feb 2002 20:11:43 +0000
Subject: Request for comments: Search and replace script
References: <XFMail.20020222152043.shalehperry@attbi.com> <mailman.1014424238.22909.python-list@python.org>
Message-ID: <a571d2$f73$1@peabody.colorado.edu>

VanL wrote:

> 
> >?please?try?to?keep?a?line?down?to?78?chars
> 
> Fixed, at least for most of the code.??I?don't?know?how?to?deal?with
> long strings, tho:
> For example, self.writeLog('This is a very long line that has this %s
> substitution and will go longer than 78 chars")
> 
use

        self.writeLog('This is a very long line that has this %s'
                      ' substitution and will go longer than 78 chars")

contiguous strings are concatenated.

or for really long stuff, use 

        self.writeLog(
"""This is a very long line that has this %s
This is a very long line that has this %sThis is a very long line that has 
this %sThis is a very long line that has this %sThis is a very long line that 
has this %sThis is a very long line that has this %sThis is a very long line 
that has this %s                      ' substitution and will go longer than 
78 chars"""
)

cheers,

f


From tim at vegeta.ath.cx  Fri Feb 15 04:56:49 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Fri, 15 Feb 2002 09:56:49 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013700446.9798.python-list@python.org>
Message-ID: <slrna6pmvr.4ea.tim@vegeta.ath.cx>

Gustavo Cordova <gcordova at hebmex.com> graced us by uttering:
>> No es importante. Es importante que Ud. continue aprender. Puede
>> aprender Espa?ol, Ingles, Japon?s, Perl, Python, o Ruby.
> 
> "Es importante que continue APRENDIENDO."

Lo siento.  English's flexibility comes at the cost of simplicity.  I
intended "continue to learn", but I guess it's not an option. ;)

> Have a nice Feb/14 :-)

Thx.

> Japon?s es muy interesante, me encant? la caligraf?a.

La gram?tica de japon?s es muy interesante, tambi?n.  Es muy f?cil,
pero muy espresivo.  La dif?cil es la escritura...

> Espa?ol, ciertamente, es pr?cticamente predecible en cuanto a su
> pronunciaci?n, eso lo hace agradable.

Si.

> Por otro lado, tiene chorromil reglas para su uso correcto, porque
> Espa?ol correcto es una modalidad de escritura que pocos aprenden.

Que me des unos ejemplos?

Gracias,
Tim Hammerquist
-- 
It is better to give than to lend, and it costs about the same.


From Norbert.Klamann at klamann-software.de  Tue Feb 26 01:53:47 2002
From: Norbert.Klamann at klamann-software.de (Norbert Klamann)
Date: 25 Feb 2002 22:53:47 -0800
Subject: Making breakpoints functional in IDLE on Windows
References: <3C7AAA43.82FEC0D0@tds.net>
Message-ID: <65c27c6b.0202252253.35174fe8@posting.google.com>

"Edward K. Ream" <edream at tds.net> wrote in message news:<3C7AAA43.82FEC0D0 at tds.net>...

[description of history and fix snipped]

> This works for me on Windows XP and Python 2.1.  I'll let you know if
> any
> problems come up.
> 

Thank you for the good work !

Works fine under Win NT 4 SP 6 too (not really surprising).


Best regards

Norbert Klamann


From jwbaxter at spamcop.net  Fri Feb  1 18:11:00 2002
From: jwbaxter at spamcop.net (John W. Baxter)
Date: Fri, 01 Feb 2002 15:11:00 -0800
Subject: What is a "bus error"?
References: <qj658.1752$HL4.244689@newssrv26.news.prodigy.com> <a32tpp02vve@enews3.newsguy.com> <3C56448B.58833301@engcorp.com> <Pine.LNX.4.43.0202011202560.4970-100000@zeon.well.ox.ac.uk>
Message-ID: <010220021511005074%jwbaxter@spamcop.net>

In article
<Pine.LNX.4.43.0202011202560.4970-100000 at zeon.well.ox.ac.uk>, Kenny
<kccw at well.ox.ac.uk> wrote:

>         However, some of these programs now give a "bus error" message -
> what does this mean?

cf:  "What happens if Guido is hit by a bus?"
    (From early Python history.)

   --John (who only did this because you've gotten real answers already)


From tismer at tismer.com  Fri Feb 22 18:43:58 2002
From: tismer at tismer.com (Christian Tismer)
Date: Sat, 23 Feb 2002 00:43:58 +0100
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com> <slrna7dg48.gtf.grante@isis.visi.com>
Message-ID: <3C76D7BE.8080500@tismer.com>

Grant Edwards wrote:

> In article <m37kp5i5a4.fsf at chvatal.cbbrowne.com>, Christopher Browne wrote:
> 
>>Centuries ago, Nostradamus foresaw when Christian Tismer <tismer at tismer.com> would write:
>>
>>>Dear Stackless Python Community,
>>>
>>>As you might have noticed, Stackless has changed quite
>>>much:
>>>- Platform independence is gone.
>>>
>>I hope not...  I would hope to hear that "platform indepedence has
>>_ARRIVED_"
>>
> 
> TANSTAAFL.  IIUC, platform independance has been sacrificed to make the
> other gains that were listed in the post.


Feeling like being understood is sometimes much cheaper
than taking a therapy because of not being understood.
Currently, I' feeling very good, although taking
the latter, anyway (last but not least since it's free).

BTW., you're right, absolutely. There is no free lunch!

sincerely - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From robin at jessikat.fsnet.co.uk  Wed Feb 27 10:24:31 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 27 Feb 2002 15:24:31 +0000
Subject: PIL vs eps
Message-ID: <6LZfFKAvoPf8EwHo@jessikat.fsnet.co.uk>

I think this is a pure ghostscript problem, but PIL fails to convert eps
to other formats and in fact it seems that gs won't create any output
without a showpage at least somewhere.

Can anyone confirm that my hunch is correct and what if anything is the
best way to detect/correct this problem?
-- 
Robin Becker


From ramb at synopsys.com  Mon Feb 25 13:16:13 2002
From: ramb at synopsys.com (Ram Bhamidipaty)
Date: Mon, 25 Feb 2002 18:16:13 GMT
Subject: Need Help: Can only set gdb breakpoints in python module the first time
Message-ID: <m33czpa0te.fsf@oak.ramandgita.com>

Python is embedded into my application. I have a custom module that
is part of the application - it is not loaded as a shared library - rather
the object files are incorporated into the final binary at link time.

The problem I am having is that I can set breakpoints in my module only
when gdb is first started. Once my app exits I am no longer able to set
breakpoints on functions in my module.

I suspect that python is somehow changing the mapping of the text segment,
but that just a guess.

My question: What is python doing that prevents gdb from setting breakpoints
after the program exits the first time? How can I prevent this?


Thanks for any info.
-Ram


From maxm at mxm.dk  Mon Feb 25 16:40:01 2002
From: maxm at mxm.dk (Max M)
Date: Mon, 25 Feb 2002 21:40:01 +0000
Subject: help: Problem with cgi form
References: <Xns91BE6246576BAAusYourStandingInIt@130.133.1.4>
Message-ID: <3C7AAF31.8040201@mxm.dk>

netvegetable wrote:

> I have a hidden input on a cgi form... 
> 
>     print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE=',last_byte_pos,'>'
> 
> I need the VALUE to be the  variable "last_byte_pos", but unfortunately 
> when I run the script for the first time the script crashes because it's 
> undeclared. 
> 
> What subroutine do I use to check if last_byte_pos is undeclared, and 
> assign it a default value if it is? 

Somehow I have the feeling that you are asking in east and I am 
answering in west, but here goes :-)

last_byte_pos = locals().get('last_byte_pos', 'defaultValue')

But this is a stupid and redundant way to do it!

Usually you just declare it in your program like:

last_byte_pos = 'defaultValue'

And then print it to the form like:

print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % last_byte_pos

Btw. If you are trying to keep state (keeping the variables value via a 
form) it is a good idea to call it the same thing in the form as in your 
script. It is a lot easier to remember :-)

like:

print '<INPUT TYPE=HIDDEN NAME= "last_byte_pos" VALUE='%s'>' % last_byte_pos

Then you can get the also get the variable again from the form via:

import cgi
form = cgi.FieldStorage()
if form.has_key('last_byte_pos'):
     last_byte_pos = form['last_byte_pos'].value
else:
     last_byte_pos = 'defaultValue'

And this last snippet is probably what you meant all along.

regards Max M



From emile at fenx.com  Thu Feb 28 10:19:51 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 07:19:51 -0800
Subject: writing efficiently a numpy array
References: <MPG.16e755abd178b85e989682@news.wanadoo.fr> <a5jege$7r6k2$1@ID-11957.news.dfncis.de> <fc905aec.0202280020.738f55db@posting.google.com>
Message-ID: <a5li4q$8cthc$1@ID-11957.news.dfncis.de>

"Stephan" <stephan.schumacher at andra.fr> wrote in message
news:fc905aec.0202280020.738f55db at posting.google.com...
> > In what area are you looking to be more efficient?  Memory, time, or
> > something else?  Do you need it in ascii form for another program,
or
> > are you saving it to be read back in?
>
> in Time !!!
>
> I need it in ascii file because this file will be the input file of
> another program. The numpy arrays store fields of velocity, pressure,
> saturation... The fields are computed on a mesh with about 300000
> nodes. Then another program read the fields and compute the migration
> of several species.

Take advantage of Numeric's output formatting, using:

<untested>
for ii in range(0,len(a),N):
    file.write(str(a[ii:ii+N])[1:-1]+"\n")

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From Bill.Scherer at VerizonWireless.com  Wed Feb 27 09:02:54 2002
From: Bill.Scherer at VerizonWireless.com (Scherer, Bill)
Date: Wed, 27 Feb 2002 09:02:54 -0500 (EST)
Subject: Rotor in Jython?
Message-ID: <Pine.LNX.4.33.0202270847100.4697-100000@bagend>

Hello - 

I've a need to perform rotor decryption in Jython.  A quick 
search didn't reveal anything to help.  I've begun coding it in 
Python from rotormodule.c. My C knowledge is not strong, so there 
are spots that are surely not right.

What I have now creates a 'rotor' object and performs a 
decryption without error, but the results are incorrect.

I can make the code available.

Any help is appreciated.

Thanks!

-- 
Bill.Scherer at Verizon Wireless
RHCE 807101044903581




From mdk at mdk.com  Wed Feb  6 09:23:32 2002
From: mdk at mdk.com (MDK)
Date: Wed, 6 Feb 2002 09:23:32 -0500
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <mailman.1013003651.8981.python-list@python.org>
Message-ID: <a3re97$1aa8h9$1@ID-98166.news.dfncis.de>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message
news:mailman.1013003651.8981.python-list at python.org...
> ----- Original Message -----
> From: "MDK" <mdk at mdk.com>
>
>
> > I need to convert a list into a list of characters.
> >
> > For example:
> >
> > ("cat",5,['dog',[3,3,1]],"zoo")
>
> That's not a "list", it's a tuple containing strings, ints, and lists;
> further, you have nested lists.  Ouch.
>
> > Would become:
> >
> > ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
> >
> > Any information would be appreciated.
>
> This is doable, but there is no Python builtin I am aware of that can
> do it.  Why do you need to do this?  If we understand what you are trying
> to do, we may be able to steer you away from this mess.
>
>
P.S   Sorry, I used the wrong symbols.  I meant
["cat",5,['dog',[3,3,1]],"zoo"].





From rdsteph at earthlink.net  Sun Feb 10 16:44:38 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 10 Feb 2002 21:44:38 GMT
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <pan.2002.02.10.15.23.55.53798.1174@club-internet.fr> <3C668B73.9E6F5BF8@earthlink.net> <mailman.1013353938.15431.python-list@python.org>
Message-ID: <3C66EAA6.23F9BA36@earthlink.net>

Thanks i will try this ;-)))


Chris Gonnerman wrote:
> 
> ----- Original Message -----
> From: "Ron Stephens" <rdsteph at earthlink.net>
> 
> > Hmmm, thanks for your comments. I just noticed that in my syslogs
> > there was an attempt to open /etc/news/inn.conf that failed. So I
> > just went in and chown 'ed that file to be owned by me rather than
> > root.
> 
> That's not the right solution; /etc/news/inn.conf should not be
> writable by "world".  You need to find out why/what tried to open it.
> 
> > I don't know if that is part of the problem. But in general, I
> > am finding that almost nothing I try to do with respect to running
> > Python scripts using crontab works at first, simply because of all
> > the permission stuff. So I change ownership and rights a lot.
> 
> Bad sign.  Your crontab runs with your user credentials, so anything
> you can do, it can do.  The problem here is usually environment
> variables which are set differently.
> 
> > But I didn't even know that /etc/news/inn.conf was needed, and maybe
> > it isn't!
> 
> Shouldn't be, if you aren't calling inn as an external program.
> 
> > I wonder what else is needed but not owned or allowed by my
> > user? How does one ever get any real work done when unknown
> > permission and file ownership problems lurk under every hidden log
> > ;-))))
> 
> In general anything that works at your command line can be made to
> work in your crontab.  Implement the logging as described in my
> previous post and see what happens.


From pricerbumanto at hotmail.com  Tue Feb 12 17:36:00 2002
From: pricerbumanto at hotmail.com (George Partlow)
Date: 12 Feb 2002 14:36:00 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <Pine.LNX.4.33.0202101938190.17089-100000@ns0.tbye.com> <mailman.1013534182.27264.python-list@python.org>
Message-ID: <e5dfe6d0.0202121435.79da2484@posting.google.com>

"Afonso Fernandez Nogueira" <afonso at writeme.com> wrote in message news:<mailman.1013534182.27264.python-list at python.org>...

> Esperantists always use the -ist- suffix ("Esperantistoj") to refer to 
> themselves though. I've seen several webpages about Python written in 
> Esperanto, so I'm sure there has to be an established way to call 
> Pythonistas in this beautiful language.

How established it is I don't know, but I did find (using Google) an
occurence of "pitonisto" at
<http://www.geocrawler.com/archives/3/3453/2000/5/0/3808756/>.

> 
> Cheers,
> Fonso

Korajn salutojn,
Georgo


From mark at codesourcery.com  Sun Feb 17 21:18:23 2002
From: mark at codesourcery.com (Mark Mitchell)
Date: Sun, 17 Feb 2002 18:18:23 -0800
Subject: Synchronous signals vs. robust code
Message-ID: <28870000.1013998703@warlock.codesourcery.com>

Python provides synchronous signals, even though signals are
asynchronous events at the operating system level.  In many ways,
this is a Good Thing.  For example, in C you can't do much of anything
from within a signal handler since you don't know much about the
current state of the application, but in Python you can do just
about anything.

In particular, you can throw an exception from a Python signal handler,
which makes handling signals simple.  In C, you generally have to
set a flag from within the signal handler, and then poll the flag
elsewhere; in Python you simply have the signal handler thrown an
exception and then provide an appropriate exception handler.

But, I'm having a hard time figuring out how to write signal-safe
code in the presence of synchronous signals.  Consider, for example,
the following:

  pipe = os.pipe()

  try:
    # Do stuff with the pipe.
  finally:
    # Close the ends of the pipe.
    os.close(pipe[0])
    os.close(pipe[1])

This code is supposed to make sure that the file descriptors are
not leaked; when we exit this scope, the descriptors will be closed,
even if an exception occurs.  [Note that I am assuming that os.close
will never thrown an exception when given a valid file descriptor.
That's not strictly speaking true on some operating systems, but
let's pretend that it is true.  For the purposes of this discussion,
the particular functions involved don't matter.]

This code, however, is not signal-safe.  For example, if a signal
occurs between the two calls to os.close, we will end up not closing
the second descriptor.  The equivalent C++ code *would* be signal safe;
an exception cannot be thrown from the signal-handler, so if the program
is still executing we can be sure that both calls to os.close will
occur.

One solution, of course, is to make sure that the signal-handlers do
not throw exceptions.  But that's not a reasonable solution in library
code; the library can't go around deciding what signal handlers should
be doing.

Another solution is to install an alternate signal handler around the
entire try-block.  This alternate signal handler would remember the
signal; we could then regenerate it at the end of the try-block.  This,
however, is not very robust; what about multiple signals occurring?  And
if the signal that occurs is important, we're ignoring it until the end
of the try-block, which isn't polite.

Another solution would be to provide a module that provides access to
sigprocmask.  Then, one could explicitly block signals during parts
of the code where these problems could occur.  For example:

  pipe = os.pipe()

  try:
    # Do stuff with the pipe.
  finally:
    # Block signals.
    sigprocmask(...)
    # Close the ends of the pipe.
    os.close(pipe[0])
    os.close(pipe[1])
    # Unblock signals.
    sigprocmask(...)

This does not work either; there are still at least two race conditions.
A signal that arrives after we have entered the finally clause, but
before we have called sigprocmask, would still result in a failure to
close the descriptors.  Similarly, a signal that occurs after the call
to os.pipe, but before the try-block is entered would result in us
skipping the finally clause.  (And moving the try earlier is not valid;
we can't clean up the descriptors unless we know that the pipe has been
created.)

Here is another try:

  # Block signals.
  sigprocmask(...)

  try:
    pipe = os.pipe()

    try:
      # Do stuff with the pipe.
    finally:
      # Close the ends of the pipe.
      os.close(pipe[0])
      os.close(pipe[1])
  finally:
    # Unblock signals.
    sigprocmask(...)

This variant is robust; we are now guaranteed that if the pipe is
created, the descriptors will be closed.

However, it is still unsatisfactory in that it results in the signal
being delayed for an arbitrary amount of time; if a time-critical
signal happens during the try-block, we will not process it until
much later.

Another solution is to require that the code be run in a thread other
than the main thread; since only the main thread receives signals,
the signal-safety problem does not occur.  However, it seems excessive
to require threads simply to write signal-safe code!  And if you have
code that depends on, say, getting SIGPIPE, then you have to have some
way of communicating the signal from the main thread to the thread
running the code above.

What do Python programmers that need to write truly robust code do
about this problem?  Is there a solution that I have missed?

If not, would the Powers That Be be amenable to making changes to the
language to support signal-safety?  One possible fix would be
except/finally clauses that blocks signals on entry.  For example,
if there were a finally_block_signals clause, you could do:

  # Block signals.
  sigprocmask(...)

  try:
    pipe = os.pipe()
  except:
    # Unblock signals.
    sigprocmask(...)
    # Reraise the exception.
    raise

  try:
    # Unblock signals.
    sigprocmask(...)
    # Do stuff with the pipe.
  finally_block_signals:
    # Close the pipe ends.
    os.close(pipe[0])
    os.close(pipe[1])
    # Unblock signals.
    sigprocmask(...) # There needs to be a way of getting the signal
                     # mask on entry to the finally clause.

This is not particularly pretty, but it would work.

Thoughts?

-- 
Mark Mitchell                mark at codesourcery.com
CodeSourcery, LLC            http://www.codesourcery.com



From picodello at yahoo.it  Fri Feb  8 06:09:13 2002
From: picodello at yahoo.it (Piero)
Date: 8 Feb 2002 03:09:13 -0800
Subject: Problems using calldll library
Message-ID: <84323230.0202080309.382bf7f4@posting.google.com>

I need to interface with a windows DLL that is declared in C like:

getword(&ncard,&nb,&adr,tabl,&status)

where the type of parameters are :

ncard=16-bit integer
nb=16-bit integer
adr=32-bit integer
tabl= array of 16-bit integer
status=16-bit integer

where ncard,nb,addr are input parameters
and   tabl,status are output parameters

I try to implement this function, but don't work.
I suppose the problem is in the input parameters format.
Something can help me ?

	def getword(self,ncard,nb,adr,tabl):
		getword =calldll.get_proc_address(self.applicomdll,"getword")
		p = calldll.membuf(4)
		px = p.address()
		pncard = calldll.membuf(2)
		pncard.write(str(ncard))
		pxcard = pncard.address()
		pnb = calldll.membuf(2)
		pnb.write(str(nb))
		pxnb = pnb.address()
		padr = calldll.membuf(4)
		padr.write(str(adr))
		pxadr = padr.address()
		tabll = calldll.membuf(2)
		ptabll = tabll.address()
		ret= calldll.call_foreign_function (getword, 'iilii', 'l',
(pxcard,pxnb,pxadr,ptabll,px))
		res = p.read(0, 2)
		print struct.unpack('bb', res)[0]
		ress = tabll.read(0, 2)
		print struct.unpack('bb', ress)[0]


Thank you in advance

Piero


From usenet at thinkspot.net  Sat Feb 23 15:40:52 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 12:40:52 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com>
Message-ID: <a582kl.3vsjht5.1@kserver.org>

[posted and mailed]

On 23 Feb 2002 11:52:10 -0800, bobnotbob at byu.edu (Bob Roberts) wrote in
comp.lang.python in article
<c4e6b17d.0202231152.5f87765d at posting.google.com>:

> What would be a good way to replace every one or more spaces (" ") in
> a string with just one space?  Or replace any number of newlines with
> just one?

This function, which I'm currently using in an email filter, replaces
all '\n' characters with a space, and then replaces all occurrences of
more than one space with a single space.

def remove_new_lines(searchstring):
    """ Returns a string without newlines or double spaces. """
    newline = '\n'

	# Replace newlines with spaces
    newstring = searchstring.replace(newline, ' ')

	# Replace multiple spaces with single spaces 
    while newstring.find('  ') > -1:
        newstring = newstring.replace('  ', ' ') 
    return newstring

You could also do this with regular expressions, but this method only
requires regular string operations.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From thomas.heller at ion-tof.com  Thu Feb  7 15:50:09 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 7 Feb 2002 21:50:09 +0100
Subject: __class__ for 2.2 types
References: <2ec1bc1c.0202071240.44185531@posting.google.com>
Message-ID: <a3upaq$1arn1e$1@ID-59885.news.dfncis.de>

"Larry" <wrbt at email.com> wrote in message news:2ec1bc1c.0202071240.44185531 at posting.google.com...
> hi, rookie here, I was hoping someone could explain to me why this happens:
>
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
> >>> [].__class__
> <type 'list'>
> >>> int.__class__
> <type 'type'>
>
> why doesn't the int one return <type 'int'>?

int is the *type* of an integer, not an integer itself.

See:

>>> list.__class__
<type 'type'>
>>> 2.__class__
  File "<string>", line 1
    2.__class__
              ^
SyntaxError: invalid syntax
>>> 2 . __class__
<type 'int'>
>>>

Thomas




From martin at v.loewis.de  Mon Feb 11 16:58:39 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 11 Feb 2002 22:58:39 +0100
Subject: Is it possible to call a function from a Tkinter widget.bind and include args???
References: <a4995n$gbg$1@news5.svr.pol.co.uk>
Message-ID: <m3u1snd6k0.fsf@mira.informatik.hu-berlin.de>

"G. Willoughby" <sab at NOSPAM.freeuk.com> writes:

> Is it possible to call a function from a Tkinter widget.bind and include
> args???
> i.e. i want to do something like this:
> 
> def printButtonAndCoords(event, button):
>     print "you rolled over %s and coords are %d:%d" % button, event.x,
> eventy
> 
> Button1.bind("<Enter>", printButtonAndCoords("button1"))
> Button2.bind("<Enter>", printButtonAndCoords("button2"))
> 
> is this possible???

The common idiom is

Button1.bind("<Enter>", lambda ev:printButtonAndCoords(ev,"button1"))
Button2.bind("<Enter>", lambda ev:printButtonAndCoords(ev,"button2"))

Notice that event.widget will give you the widget that was the source
of the event, so you may not need additional parameters at all.

Regards,
Martin


From dalke at dalkescientific.com  Sat Feb 16 17:46:10 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Sat, 16 Feb 2002 15:46:10 -0700
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <ad052e5c.0202161240.73499031@posting.google.com>
Message-ID: <a4mnjv$uj9$1@slb7.atl.mindspring.net>

Dan Bishop:
>Everyone should learn BASIC.  It's the perfect example of how *not* to
>design a programming language :-)

Errm, I've used plenty of worse programming languages.  Most of
the sort of embedded command languages which jest grew.  By
comparison, BASIC is clean and elegant.

But then again, I spent almost a decade programming in BASIC in
its various incarnations, so my brain's probably rotted away :)

                    Andrew
                    dalke at dalkescientific.com





From emile at fenx.com  Thu Feb 21 19:39:09 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 21 Feb 2002 16:39:09 -0800
Subject: strings and sort()
References: <mailman.1014334596.2619.python-list@python.org>
Message-ID: <a544al$4kput$1@ID-11957.news.dfncis.de>

"Delaney, Timothy"  replies to:
> > I like Python, but the biggest problem I have with the language
right
> > now is what low code density it has at times.
>
> Huh? Low code density? Huh?
>

See the thread on lambdas  ;-)

Feeling-low-and-dense-ly y'rs,


--

Emile van Sebille
emile at fenx.com

---------



From db3l at fitlinxx.com  Fri Feb  1 20:32:22 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Feb 2002 20:32:22 -0500
Subject: comparing DateTime instances
References: <mailman.1012602663.18405.python-list@python.org>
Message-ID: <un0yssm6x.fsf@ctwd0143.fitlinxx.com>

"Mark McEahern" <mark at mceahern.com> writes:

> The documentation for mx.DateTime's cmp seems to assume the reader
> understands __cmp__.  This is all we find there:
> 
> "cmp(obj1,obj2,accuracy=0.0)
> Compares two DateTime[Delta] objects.
> If accuracy is given, then equality will result in case the absolute
> difference between the two values is less than or equal to accuracy."

I think the documentation more assumes that the reader is familiar
with the builtin cmp() function (and not necessarily the __cmp__
method, although the former is really just direct access to the
latter), which seems fairly reasonable to me.  And the documentation
for cmp() does detail the <0, 0, >0 result code of the function.

But adding some qualification (even if only to note it mirrors cmp())
certainly doesn't seem a bad thing.

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


From printers at sendme.cz  Sat Feb 23 05:21:02 2002
From: printers at sendme.cz (A)
Date: Sat, 23 Feb 2002 11:21:02 +0100
Subject: Is it possible in Python too?
Message-ID: <3C777B1E.12123.87FDB7@localhost>


Hi,
Is possible to set up a proxy name (address) in MS Explorer 
browser from a running  Python program?
Thank you for your help.
Ladislav






From msoulier at nortelnetworks.com_.nospam  Wed Feb 27 13:34:23 2002
From: msoulier at nortelnetworks.com_.nospam (Michael P. Soulier)
Date: 27 Feb 2002 18:34:23 GMT
Subject: full closures
Message-ID: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com>

    Hello. 

    While I know about bound methods in Python, I'm wondering if a full
closure can be implemented. Can it?

    Example: I want to pass an arbitrary code reference into a closure and
have all of its variables preserved. 

    ie. 

    print "the current count is %d" % counter; counter += 1

    Can I pass the above code into a closure, and then simply call it
repeatedly and see counter increment itself? If I pass in a function with a
local variable, that local will be reset each time, and I don't want to use a
global. 

    Thanks,

    Mike

-- 
Michael P. Soulier, QX41, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to Unix


From robin at jessikat.fsnet.co.uk  Sun Feb  3 19:56:26 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Feb 2002 00:56:26 +0000
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
Message-ID: <iJU5VJA6wdX8Ewk4@jessikat.fsnet.co.uk>

In article <922.799T2387T11875593threeseas at earthlink.net>, Timothy Rue
<threeseas at earthlink.net> writes
>In reviewing over many messages in the autocoding and related threads and
>seeing the evidence of contridictions, hypocracy, irrational and
>illogical responses......
>
>Its damn clear that there is a need to remove the sort of people posing to
>be professional who are in essence just good liars, abstraction
>manipulators, from the industry.
>
>Who knows, maybe that stench will be enough to motivate me see that it
>gets done, gets removed.
>
>Maybe I should be more appriciative of the stench if it so motivates me.
>
>Anyone want to add anything to this?
>
all your ideas are belong to you
-- 
Robin Becker


From sab at NOSPAM.freeuk.com  Tue Feb  5 10:56:57 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Tue, 5 Feb 2002 15:56:57 -0000
Subject: Amount of files on a drive?
References: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk>
Message-ID: <a3ov9n$dn9$1@newsg3.svr.pol.co.uk>

i am using this at the minute:
------------------------------------------------------------------
import win32api
import string
import os

def countFiles(arg, dir, files):
    for file in files:
        count=count+1

fileListings=[]
driveList=string.split(win32api.GetLogicalDriveStrings(),'\0')[:-1]

for drive in driveList:
    os.path.walk(drive, countFiles, 0)
------------------------------------------------------------------
but its pretty slow any other suggestion how i could speed it up??

G. Willoughby


"G. Willoughby" <sab at NOSPAM.freeuk.com> wrote in message
news:a3m9d3$mr4$1 at newsg2.svr.pol.co.uk...
> Is there a fast an easy way to get the number of files on a drive??
> ta,
> G. Willoughby
>
>




From skip at pobox.com  Fri Feb  8 13:45:03 2002
From: skip at pobox.com (Skip Montanaro)
Date: Fri, 8 Feb 2002 12:45:03 -0600
Subject: What's the equivalent in Python for Perl's CPAN?
In-Reply-To: <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com>
        <a3v903$kgk$1@panix2.panix.com>
        <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>
Message-ID: <15460.7343.35758.3998@12-248-41-177.client.attbi.com>

    Mike> So, what's the equivalent in Python for Perl's CPAN?

People are working on stuff.  Search the list archives for Ciphon.  There
was a lightning talk presented yesterday about a server called Gideon that's
supposed to serve up package metadata to tools like Ciphon I think.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From m.faassen at vet.uu.nl  Fri Feb  8 19:00:20 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 9 Feb 2002 00:00:20 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu> <a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <a41jmf$cd2$1@newshost.accu.uu.nl> <mailman.1013208443.28991.python-list@python.org>
Message-ID: <a41oqk$cri$1@newshost.accu.uu.nl>

Cliff Wells <logiplexsoftware at earthlink.net> wrote:
> On 8 Feb 2002 22:32:47 GMT
> Martijn Faassen wrote:

>> Cliff Wells <logiplexsoftware at earthlink.net> wrote:
>> > Arguing with irrational oddballs on usenet is today's equivalent of the
>> > Rubik's Cube.
>> 
>> I never managed to solve those either. Is it still equivalent for those
> who
>> do mananage to solve the cube?

> I'm guessing Rue could solve it, if that answers your question ;)

I've now lost track of who is the irrational oddball and who they're 
arguing with. The only thing I still know is that it's happening on 
usenet -- couldn't be anywhere else. :)

Regards,

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


From martin at v.loewis.de  Wed Feb 20 17:33:57 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 23:33:57 +0100
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk>
Message-ID: <m33czvkckq.fsf@mira.informatik.hu-berlin.de>

"Phil Mayers" <p.mayers at ic.ac.uk> writes:

> Any ideas?

You could try _PyLong_FromByteArray. It is not official API, but it
may still work.

Regards,
Martin



From core.lists.python at core-sdi.com  Thu Feb  7 12:56:31 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Thu, 7 Feb 2002 14:56:31 -0300
Subject: Fine Tkinter Question
Message-ID: <008401c1b000$c65803b0$8642a8c0@corelabs.coresdi.com>

Hi there!

I have a Tkinter window which opens a 2nd one in which I want to close it
when you hit the <"Return"> key upon the "Close" button. But..... isn't
working.

Let's see the code:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

from Tkinter import *
def show_child ():
    win=Toplevel()
    win.focus_set()
    win.grab_set()
    Quitter=Button(win,text="Close Me",command=win.destroy)
    Quitter.pack()
    Quitter.focus_set()
    win.bind("<Return>",lambda event,myself=win:myself.destroy)

root=Tk()
Button (root,text="Show",command=show_child).pack()
root.mainloop()

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

   Do you know why this is not working?
  And do you know what is the elegantly simple way to make it work?

I know the problem is in the " win.bind("<Return>",lambda
event,myself=win:myself.destroy)" line, but I don't know how to make it
right.


Help me for free !

Tnxs!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020207/7224d9e2/attachment.html>

From nospam at bigfoot.com  Tue Feb  5 13:01:26 2002
From: nospam at bigfoot.com (Gillou)
Date: Tue, 5 Feb 2002 19:01:26 +0100
Subject: REQ : encoding windows cp1252 => iso latin 1
Message-ID: <a3p68d$1a56$1@norfair.nerim.net>

Hi,

This is an emergency ;-)

My customers make copy/paste from M$ word docs to forms translated to XML
(expecting ISO latin 1 charset).
My XML parser (pyexpat) does not accept cp1252 character, and I'm looking
for a function that can translate extra cp1252 characters to the closest ISO
latin 1 encoding.

Thanks in advance for any help.

--Gillou

PS1: I can't use Unicode facilities cause I'm always running python 1.5.2
(old Zope).
PS2: Please reply to glenfant_at_bigfoot.com, the "Freo/Reply-to" headers
are antispam.





From eppstein at ics.uci.edu  Thu Feb 28 20:24:40 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 17:24:40 -0800
Subject: Status of PEP's?
References: <mailman.1014944882.14890.python-list@python.org>
Message-ID: <eppstein-80232B.17244028022002@news.service.uci.edu>

In article <mailman.1014944882.14890.python-list at python.org>,
 James_Althoff at i2.com wrote:

> Another issue that I have is that the syntax does not work outside of a
> for-loop header.  I want to be able to create integer intervals anywhere in
> my code, not just in the header of a for-loop.

[x for 3 <= x < 10]

or, for a slightly more complicated example: yesterday I wanted to iterate 
over powers of two:

[1L << x for 0 <= x < 1000]
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From phd at phd.pp.ru  Sun Feb  3 05:55:38 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sun, 3 Feb 2002 13:55:38 +0300
Subject: curses module
In-Reply-To: <e63q5ukd116tkglrbp92k160bjl2jihvu5@4ax.com>; from rhymes@myself.com on Sun, Feb 03, 2002 at 11:11:44AM +0100
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <e63q5ukd116tkglrbp92k160bjl2jihvu5@4ax.com>
Message-ID: <20020203135538.A12784@phd.pp.ru>

On Sun, Feb 03, 2002 at 11:11:44AM +0100, Lawrence Oluyede wrote:
> You're right, i use Win2k and the binary distribution..

   M$ Ophice GUI is not supposed to run text-mode progrms (though some good
guys don't beleive in it and produced pdcurses :)

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



From tim_one at email.msn.com  Sun Feb  3 17:51:26 2002
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 3 Feb 2002 17:51:26 -0500
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
In-Reply-To: <a3k9am$p0c$1@newsreader.mailgate.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEBJMJAA.tim_one@email.msn.com>

[Tim]
> Oddly enough, the more intelligent of us [Python developers] agree with
> him [Guido], but nobody can explain why <wink>.

[Mark Hadfield]
> Who *are* the more intelligent of you?

Those of us who agree with Guido, of course.

leaving-it-a-secret-whether-i-agree-with-him-ly y'rs  - tim



From wurmy at earthlink.net  Mon Feb 25 15:06:19 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Mon, 25 Feb 2002 20:06:19 GMT
Subject: All-on, all-off button for checkbuttons?
References: <mailman.1014664062.9011.python-list@python.org>
Message-ID: <3C7A9975.8D7299BA@earthlink.net>

Nick Arnett wrote:
> 
> I'm just bringing myself up to speed with Python, so forgive what I think is
> an simple question.  I'm building a UI that has a bunch of checkbuttons,
> which also includes buttons to turn all the checkbuttons on or off.  I see
> how to turn them off individually, but it seems to me that there must be a
> more elegant way.
> 
> The checkbuttons are created in __init__.  I'm thinking that there must be a
> way to filter or iterate through the objects in __init__, and for each one,
> if it's a button, select or deselect as appropriate.  I see how to get the
> namespace as a dictionary, but that hasn't helped.

Without knowing what GUI (e.g. Tkinter, wxPython, Qt, etc) you are using,
I can only give you an example in pseudo-Python...

In __init__, keep a list of checkbox instances around:

  c1 = CheckBox(...)
  c2 = CheckBox(...)
  c3 = CheckBox(...)
  self.checkboxes = [c1, c2, c3]

Then, when you want to do something with all these
checkboxes, simply loop over the list:

  for checkbox in self.checkboxes:
      checkbox.checked = false

Again, this is pseudo-Python, the names and method in your 
program will look different. But you'll get the idea.

Depending on the GUI you're using, there should probably
be a way to find all objects on a form/frame, and loop over
them. That would be another way to do it.

HTH,

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From gbreed at cix.compulink.co.uk  Tue Feb 26 11:35:08 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Tue, 26 Feb 2002 16:35:08 +0000 (UTC)
Subject: Python COM -- What is the name of the file?
References: <mailman.1014737443.16139.python-list@python.org>
Message-ID: <a5gdfs$6b8$1@thorium.cix.co.uk>

Mike Brenner wrote:

> Does anybody know how to get the name of the file an Excel Spreadsheet 
> is in?
> 
> For example, to do a file-save-as in tab-delimited format, 
> it is necessary to change "C:\data\stuff.xls"
> to "C:\data\stuff.txt", but I don't know where to
> get the "C:\data\stuff.xls" from in any given spreadsheet.

Do you have the "Microsoft Excel Visual Basic Reference" installed?  It 
should be in your help topics list.  If not, it's an option when you 
install Office, and it's what you need to answer questions like this.

Anyway, for an object called xl, it looks like these are what you want:

>>> xl.Workbooks[0].Path
u'C:\\Documents and Settings\\grahamb\\My Documents'
>>> xl.Workbooks[0].Name
u'Book1.xls'
>>> xl.Workbooks[0].FullName
u'C:\\Documents and Settings\\grahamb\\My Documents\\Book1.xls'

> It would even be okay, if I could just get the name "stuff" from some 
> part of the COM object.
>
> (And, of course, hopefully the same technique will work in Microsoft 
> Project, etc.)

I think "Workbooks" is Excel specific.


                   Graham


From cliechti at gmx.net  Thu Feb  7 16:33:32 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 7 Feb 2002 22:33:32 +0100
Subject: Serial port Input/Output conflicts
References: <ef77152a.0202070729.3a6a5d2a@posting.google.com>
Message-ID: <Xns91AEE5FF91B05cliechtigmxnet@62.2.16.82>

[posted and mailed]

jwk at mrao.cam.ac.uk (James Keen) wrote in 
news:ef77152a.0202070729.3a6a5d2a at posting.google.com:
...
> The status data gets sent to the port from my Hardware, which I can
> read, but the serial port refuses to send any more commands to my
> hardware until I close and then reopen the port.
> 
> serial = os.open('/dev/ttyS1', FCNTL.O_RDWR|os.O_NDELAY,0)

not much of an initialization... your port could be in processing mode 
where it interprets control characters, flow controls etc...
this initialization is not likely to work with binary data.

http://sourceforge.net/project/showfiles.php?group_id=42303
&release_id=67594

(the url above should be one line, here's the project page: 
http://mspgcc.sourceforge.net/) download pybsl, extract serial* and 
use the serial.Serial class. i have trimmed it to work with binary 
data.

...
>    global serial

codesidered writing a class that represents you hardware? would be 
cleaner that juggling around globals.

chris

-- 
Chris <cliechti at gmx.net>



From nhodgson at bigpond.net.au  Tue Feb 19 18:06:27 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Tue, 19 Feb 2002 23:06:27 GMT
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org> <a4ss56$akj6@news1.gtech.com> <b7uc8.8411$Zp2.790511@atlpnn01.usenetserver.com> <fjs47u0n8bscvgtme1vc7j6jajlt4og2n3@4ax.com>
Message-ID: <TTAc8.6742$Uy2.29406@news-server.bigpond.net.au>

Courageous:
> Speaking of typedef and the C/C++ languages, I've long thought
> that in a statically-typed language, the types should never be
> aliased, and should always be distinct. For example, the following
> shouldn't compile:
>
> typedef float F1;
> typedef float F2;
> ...

   Some program checkers such as PC-Lint can check for inconsistent use of
typedefed types. It is not as good as having the compiler check, but it can
be integrated into a build or check in system to ensure a greater degree of
type-safety than standard C/C++.

http://www.pclint.com/html/strong.htm

   Neil





From dalke at dalkescientific.com  Thu Feb 14 17:58:22 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Thu, 14 Feb 2002 15:58:22 -0700
Subject: '/' not allowed in cookie name?
References: <mailman.1013707165.15610.python-list@python.org>
Message-ID: <a4hffl$8m$1@nntp9.atl.mindspring.net>

Oliver Rutherfurd:
> I'm ... receiving a cookie.CookieError when a server I am
> connecting to sends me a cookie with '/' in the name.
   ...
> After spending a little time reading http://www.faqs.org/rfcs/rfc2109.html
> I am still not any closer to understanding whether '/' is a legal value
> that the cookie module should accept in a name, or whether the server
> is misbehaving.

That RFC defines the format for the cookie syntax, in 4.1

   av-pairs        =       av-pair *(";" av-pair)
   av-pair         =       attr ["=" value]        ; optional value
   attr            =       token
   value           =       word
   word            =       token | quoted-string

The "attr" is the name (before the "="), which you want to be 1/2.
The specification for token is defined in RFC 2068 as

    token          = 1*<any CHAR except CTLs or tspecials>

where

    tspecials      = "(" | ")" | "<" | ">" | "@"
                         | "," | ";" | ":" | "\" | <">
                         | "/" | "[" | "]" | "?" | "="
                         | "{" | "}" | SP | HT

So token cannot contain "/", which means "1/2" is not a valid
cookie name.

As to if some clients and servers let you use illegal cookie
names ... that's a different thing.

                    Andrew
                    dalke at dalkescientific.com






From xeno at eskimo.com  Sun Feb 17 14:28:26 2002
From: xeno at eskimo.com (Xeno Campanoli)
Date: 17 Feb 2002 19:28:26 GMT
Subject: chomp equivalents
Message-ID: <a4p08q$2bi$1@eskinews.eskimo.com>

Is there a chomp equivalent in the "string" object?  rstrip doesn't seem to
take off the newline.

Also, I looked up chomp on the Python website search engine, and it said there
is a chomp.  Am I reading this wrong?  I don't show any chomp procedure in my
python environment.


From stuartn at froggy.com.au  Tue Feb 26 23:08:45 2002
From: stuartn at froggy.com.au (Stuart Nettleton)
Date: Wed, 27 Feb 2002 15:08:45 +1100
Subject: Topo-Processor Module
Message-ID: <004401c1bf44$74414340$560114ac@flow.com.au>

Hello,
I am looking for an efficient module that applies a list of equation definitions to a database table, where each row in the table represents a variable and fields are assumed to be forward time sequential (i.e. like a financial model in Excel).  Does anyone know of such a module?
Kind regards,
Stuart
BTW I wrote a topological processor in Microsoft Access VB but it is very slow!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020227/c55869a0/attachment.html>

From gerson.kurz at t-online.de  Sun Feb 10 03:18:19 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 10 Feb 2002 08:18:19 GMT
Subject: [ANN] pydocgen 0.1 - a windows alternative to pydoc
Message-ID: <3c662b40.10617421@news.t-online.de>

pydocgen.py generates a HTMLHelp file from the docstrings of the
python standard library, offering full-text search and a nice index.
As such, it is an alternative to pydoc, which seems to me to be broken
(at least on my python22 system it fails when asked to scan all files
in /lib recursively.) Note that pydocgen also has problems with some
files in the PSL - they simply cannot be imported standalone. 

Some might also find the "htmlhelpgen.py" source helpful, for
generating HTMLHelp files from a python program. "htmlhelptest.py"
shows you how to use those classes from your own code. 

Project Homepage (very little on it right now):
http://p-nand-q.com/pydocgen.htm

Sourcecode:
http://p-nand-q.com/python/pydocgen.zip

Helpfile (Warning, 2mb download):
http://p-nand-q.com/python/python22.zip




From imbaczek at poczta.fm  Fri Feb 15 14:46:45 2002
From: imbaczek at poczta.fm (Marek Baczek =?iso-8859-2?Q?Baczy=F1ski?=)
Date: Fri, 15 Feb 2002 19:46:45 +0000 (UTC)
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org> <slrna6kkpj.is0.tim@vegeta.ath.cx> <pu6e4a.5ti.ln@wombat.san-francisco.ca.us> <slrna6m7jd.o2v.tim@vegeta.ath.cx>
Message-ID: <slrn.pl.a6qeqo.3vs767d.imbaczek@baczek.net.invalid>

Tim Hammerquist wyprodukowa?[ao]?:

> OTOH, exceptions in Spanish are much less frequent than in English or
> French.  Pronunciation is nearly 100% predictable, and there are more
> regular verbs than most languages.
> 
> (I have no knowledge of German.)

Take a latin alphabet. Take a lot of exceptions. Throw in some weird
letters. Take some English words. Change a lot of spelling. Mix in some
more spooky letters. Make rules. Make exceptions to these rules.  Make
even more exceptions. Shuffle. The resulting language will be German :)

(I have very little knowledge of German and some of English :)

Baczek
-- 
Marek Baczy?ski aka Baczek UIN:57114871 RLU:201791 http://republika.pl/imbaczek
There is no pardon FOR Murphy's Law.


From gerson.kurz at t-online.de  Sat Feb  2 23:08:49 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 03 Feb 2002 04:08:49 GMT
Subject: print statement in lambda expression
References: <NW178.16270$bIE.6837@news1.bloor.is>
Message-ID: <3c5cb680.312156@news.t-online.de>

On Sun, 03 Feb 2002 03:14:53 GMT, "Michael P. Soulier"
<msoulier at mcss.mcmaster.ca_.nospam> wrote:

>    Hello people. 
>
>    I just recently found that I couldn't put a print statement in a lambda
>expression as a binding in Tk. Could someone explain why? 

The reason was explained already by other people, here is a hack that
makes it work nonetheless: use sys.stdout.write; e.g.

import sys

lambda x,y: return sys.stdout.write("...")

if you need multiple args, make a varargs output function:

def out(*x):
    print " ".join(map(str,x))

f = lambda x,y: out(x,y)

f("Hello", 42)




From philh at comuno.freeserve.co.uk  Thu Feb  7 18:14:33 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:14:33 +0000
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org> <Xns91AE6032817F5ahmebahhotmailcom@198.99.146.10>
Message-ID: <slrna662ip.d87.philh@comuno.freeserve.co.uk>

On 7 Feb 2002 17:23:38 GMT, ameoba <ahmebah at hotmail.com> wrote:
>"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
>news:mailman.1013090595.10717.python-list at python.org: 
>
> 
>> How the heck to you "enjoy" an idiot like Rue though?
>
>
>With a light chianti?

But, grilled or roasted?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From fperez528 at yahoo.com  Fri Feb 22 15:14:09 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Fri, 22 Feb 2002 20:14:09 +0000
Subject: Request for comments: Search and replace script
References: <XFMail.20020222152043.shalehperry@attbi.com> <mailman.1014424238.22909.python-list@python.org>
Message-ID: <a571hk$f73$2@peabody.colorado.edu>

VanL wrote:

> Sort?of?a?poor?man's?'more'?for?when?this?is?run?on?a?windows?system.
> 

you can take this

#----------------------------------------------------------------------------
def page_dumb(strng,start=0,screen_lines=25):
    """Very dumb 'pager' in Python, for when nothing else works.

    Only moves forward, same interface as page(), except for pager_cmd and
    mode."""

    ncols = 80
    out_ln = strng.splitlines()[start:]
    out = []
    for ln in out_ln:
        if len(ln) <= ncols:
            out.append(ln)
        else:
            out.extend(chop(ln,ncols))
    screens = chop(out,screen_lines-1)
    #print '\nscreens is:',screens,os.linesep  # dbg
    if len(screens) == 1:
        print os.linesep.join(screens[0])
    else:
        for scr in screens[0:-1]:
            print os.linesep.join(scr)
            ans = raw_input('---Return to continue, q to quit--- ')
            if ans.lower().startswith('q'): return
        print os.linesep.join(screens[-1])


If you want its bigger brother, page() (multi-os, with solid fallbacks) go to 
http://www-hep.colorado.edu/~fperez/ipython/

page() and family are in the genutils.py module (plus lots of other useful 
stuff). It's LGPL, so feel free to use it.

cheers,

f.


From brown.2053 at osu.edu  Tue Feb 26 16:21:20 2002
From: brown.2053 at osu.edu (Locke)
Date: Tue, 26 Feb 2002 16:21:20 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu>
Message-ID: <a5gu4t$si9$1@news.cis.ohio-state.edu>

ok, i figured out what i had to do. instead of 'message = s.recv(msg_len)' i
did:

while len(message) < msg_len: message = message + s.recv(1024)

and that seems to work every time. yippee! thanks for help everyone




From john_coppola_r_s at yahoo.com  Sun Feb 17 22:34:18 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Sun, 17 Feb 2002 19:34:18 -0800 (PST)
Subject: Why Python is like BASIC ...  java "vs" python
In-Reply-To: <roy-FCF055.13060217022002@reader2.panix.com>
Message-ID: <20020218033418.46493.qmail@web11808.mail.yahoo.com>

For all the people following this thread.  I think the
main point I was trying to make (although rather
abstract) is that I think differently depending on
which language I am using.  If I am using BASIC I
don't think like python.  If your are, your not using
basic correctly.  You must use basic like BASIC.  And
not try to do anything more complicated than basic can
describe in a clear way, otherwise you begin to use
the language outside the bounds of which it was
intended.  For instance, it's not very efficient to
write xml parsers in pure python.  Is this making
sense?

When I program in C++, I'm not thinking like java. 
C++ is far more low level and requires more detail. C
and C++ are efficiency languages.  Good programming
practices in C although could be useful in C++, but
largely are not since structs and functions have been
supplanted by real Object Orientation (well almost
real).  When I program in C i'm not thinking like C++.

Believe it or not, when I program in Jython, I don't
think like Python.  WHY?  When you do advanced
programming, you will see the differences.

So on basis of that, I feel very strongly that
although there are similarities,  I seldom consider
"print XYZ" to be thinking.  When one tackles advanced
problems, you see the differences in languages, you
see why they are useful, and you see why they are not,
and these differences for me out weigh the
similarities.

Sincerely,
John Coppola



__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From stojek at part-gmbh.de  Fri Feb 15 08:53:35 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Fri, 15 Feb 2002 13:53:35 GMT
Subject: help!: optimization of search problem (example code)
References: <mailman.1013722043.12712.python-list@python.org>
Message-ID: <3c6d12ba.4348562@news.easynews.net>

>
>
>In Python, access to local variables in functions
>is faster than access to global variables.

Thanks, I didn't know that.
Marcus


From sholden at holdenweb.com  Mon Feb  4 18:47:57 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 18:47:57 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com>
Message-ID: <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7x8za8dgle.fsf at ruckus.brouhaha.com...
> "Tim Peters" <tim.one at home.com> writes:
[Paul misunderstands the bugfix process, and wants to produce 1.5.3 and
1.2.1]
[2.1.1 / 2.1.2 confusion]
>
> > If you have messier problems, and you're determined to write code that
works
> > unchanged back through the first Python ever released, then sure --
slicing
> > divmod is clear and works.  ...
>
> No.  I'm taking a practical view of this.  I use HTML tables in my
> webpages even though there's still a few ancient browsers out there
> that don't display them.  But if 25% of browsers didn't support them,
> I'd avoid using them when I didn't need them.  I do avoid making my
> pages depend on Javascript, for basically that reason.
>
"When I didn't need them": surely "must run in old browsers" is not a
partial requirement. Do you really specify your output format as "must
remain compatible with Netscape Navigator 2.0 unless it's really important
not to". No wonder you have problems with the meaning of "backward
compatibility" <0.9 wink>.

> I see publishing Python programs as like publishing web pages.  If I
> have to tell 2% of my users to upgrade those ridiculous old browsers
> to view my pages, I don't mind.  If I have to tell 25% of them to
> upgrade, I'm causing hassles for a significant population and I
> shouldn't do that without a good reason.  If I have to tell 80% of
> them to upgrade, I'm probably going to lose a lot of them.
>
> The question is, what percentage of users are using what Python version?
>
Well, if you want to take a count, add me in to the user population of
1.5.2, 2.0, 2.1.2 and 2.2. And I think there are a few beta copies lying
around from pre-2.2. <grin>

> I bet less than 10% of users have 2.2 right now.  I'm curious about
> the ratio of 2.1 users to 1.5.2 users.  Do you have any idea what it
> is?  I don't know how I'd measure it.  There's no User-agent header
> collected at python.org every time someone runs a Python script.
>
> I go back and forth about whether to code for 1.5.2.  I do like 2.1
> features like +=, and I noticed last night that the "continue"
> statement works differently.  If I had reason to believe that less
> than (say) 25% of today's installations are 1.5.2, then I wouldn't
> bother with 1.5.2; but I think the percentage may be higher than that.

Well, the major problem for most would-be Python software authors right now
is that Red Hat are still shipping 1.5.2 without any real attempt to conceal
it from a user wanting to install some later version from the RPMs. Building
their system so it's hard to upgrade Python without breaking Anaconda was
dumb. Defensible, but dumb. And I think it hurts Python more than the
benefit of being able to say "Red Hat built Python into their system
maintenance software."

I appreciate you are keen to distribute "nifty Python programs". If so, you
might just have to accept the bloat of building installer versions, untilt
someone comes up with a way for installers to determine whether Python is
currently available in an appropriate version or not.

These engineering issues are what will stand in the way of Python's
increasing popularity, which is why I prefer to address them when I see them
raised rather than ignore them. If you insist that Python isn't backwards
compatible, I have to think what you must think about VB.NET, where they
really *have* broken backward compatibility, and in a bug way. Unless VB
application authors are really careful (and many VB programmers aren't
familiar with the concept of software engineering) we're going to see some
really badly broken applications there.

By comparison, Python is a hotbed of sanity. So start working on 1.5.3 right
now. If you ever produce it I'd like to think it would get a prominent
placed on the website. But remember: with very minor exceptions, the rule
for bugfix releases is "no new functionality", so don't expect to sneak
augmented assignments in there, for example!

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/







From jacek.generowicz at cern.ch  Fri Feb 15 03:19:05 2002
From: jacek.generowicz at cern.ch (Jacek Generowicz)
Date: 15 Feb 2002 09:19:05 +0100
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org> <3C6C3805.37BD87FB@ccvcorp.com>
Message-ID: <tyfk7tfduo6.fsf@pb-s-513-1-26-6-1.cern.ch>

Jeff Shannon <jeff at ccvcorp.com> writes:

> Gustavo Cordova wrote:
> 
> > >
> > > I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> > >
> > > When I do a readline the I get back a string.
> > >
> > > How do I convert this string to a dictionary?
> > >
> > > Thanks.
> > >
> >
> > You eval() it?
> >
> > >>> d = { "one":1, "two":2, "three":3 }
> > >>> print d
> > {'three': 3, 'two': 2, 'one': 1}
> > >>> S = str(d)
> > >>> print S
> > {'three': 3, 'two': 2, 'one': 1}
> > >>> dd = eval(S)
> > >>> print dd
> > {'one': 1, 'three': 3, 'two': 2}
> > >>> d is dd
> > 0
> > >>>
> 
> This works, but is generally a Bad Idea(tm).  The problem
> with eval/exec on arbitrary strings (and any string you read
> in from a file is arbitrary) is that it's hard to be
> positive that what you're eval/exec-ing is what you expect.
> And if it's *not*, then many, many very bad things can
> happen.  (Imagine someone "accidentally" replacing that
> textfile with one that contains the line "import
> os;os.system('rm -s /')" -- suddenly your entire filesystem
> is blank....)

Writing python scripts works but is generally a Bad Idea(tm).

Imagine somone "accidentally" replacing the textfile containing your
script with one that contains the line "import os;os.system('rm -s
/')" -- suddenly your entire filesystem is blank....)

Sic probo.

:-)


From QnickQm at alum.mit.edu  Wed Feb 27 23:46:31 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Thu, 28 Feb 2002 04:46:31 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net>
Message-ID: <slrna7rdna.5m1.QnickQm@localhost.localdomain>

In article <3c7c1a91.150875561 at menuhin.netfront.net>, A. Jones wrote:
> It's my second day Pythoning and I've written this code to encypher
> text (or files, ostensibly) into two pieces, neither of which can be
> used to determine the message.
> 
> Any problems pointed out, clunky lines spotted, or security problems
> would be most appreciated.  Also, if you've seen something like this,
> if you'd tell me, please.
 [...]

Okay; A code review follows.  Not bad for two days of Pythoning!

BTW, I would strongly disrecommend using this cipher for real purposes;
it is not as secure as you think.  See my closing comments below.

===================================================================
 
> #encyphering tool 
> plaintext = "Isn't it awfully nice to have a penis; isn't it
> frightfully good to have it on!" #The variable makes it obvious.

#Actually, this won't work: you need to use triple quotes for a string
#to span multiple lines.  Try it like this:

plaintext = '''I would have used the "Bruces' Philosophers Song" myself,
but examples are a matter of taste.  BTW, the final words are
'a dong'.'''

> blarg = [] #To hold translation of ASCII to num
> for b in plaintext:
> #    print b #Was here for development purposes.
>     blarg.append(ord(b)) #Turn it into a list of ASCII numbers
>                          #nums make sorting easier&append/ord makes a
> nice list

#Naming your variables 'blarg', 'blurb', and so forth can make your
#code harder to read.

plaintext = map(ord, plaintext)

> plaintext = "Text purged."

#It won't do any good to 'purge' the plaintext from memory at this point;
#if you're worried about an attacker getting it from RAM, they've already
#had an opportunity to have done so.  If you're worried about them reading
#from swap, your program may have already swapped out.

> blurb = {} #To hold 1:1 value/offset references
> b = 0 #reuse variable so's to destroy evidence/make my life easier

Rather than keeping an index, you might want to use the 'range' or
'xrange' method.

> for z in blarg:
>     if z in blurb: #duplicate key fixer
>             if type(blurb[z]) == "<type 'list'>": #Stops nesting
> strings

#It's bad style to rely in the string value of a type.  Intstead, you
#might want to import types, then say 
#          'if type(blurb[z]) == types.ListType:'.  
#If you're in Python 2.2, you can say, 
#          'if type(blurb[z]) == list:'

>                 blurb[z].append(b)
>             else:
>                 temp = blurb[z] #Used to create first list entry
>                 temp.append(b)
>                 blurb[z] = temp

#The two cases above do exactly the same thing; there's no need for the
# type check.

>     else: blurb[z] = [b] #sets dictionary with values as keys, offsets
> as objects
>     b = b+1 #Next step in the offset counter... clunky.

# Map from each character to an in-order list of its positions in
# the original map.
posMap = {}
for idx in xrange(len(plaintext)):
   character = plaintext[idx]
   if character in posMap:
      posMap[character].append(idx)
   else:
      posMap[character] = [idx]

> valuelist = blurb.keys() #extract keys of dictionary
> valuelist.sort() #to be randomized later(date)

# These lines are reasonable, but see below for thoughts on security.
valuelist = posMap.keys()
valuelist.sort()

> finval = [] #to hold final value list
> finoff = [] #to hold fial offset list
> for r in valuelist:
>     z = 0
>     while z < len(blurb[r]): #Writes number of values equal to number
> of offsets
>         finval.append(r)
>         z = z+1
>     for q in blurb[r]: #Decomposes my clunky nesting lists/dicts
>         finoff.append(q)
> 
> blurb = {}  #Destroy the evidence!
> valuelist = []

# If we use a nested loop, we can eliminate the index variable.

finval = []
finoff = []

for val in valuelist:
    for pos in posMap[val]:
        finval.append(val)
        finoff.append(pos)

# It's too late to destroy the evidence, but if you'd like, del works
# just fine.
del posMap, valuelist, plaintext

============================================================
To summarize, and introduce a function:

def encipher(plaintext):
    plaintext = map(ord, plaintext)

    posMap = {}
    for idx in xrange(len(plaintext)):
        character = plaintext[idx]
        if posMap.has_hey(character):
            posMap[character].append(idx)
        else:
            posMap[character] = [idx]

    valuelist = posMap.keys()
    valuelist.sort()

    finval = []
    finoff = []

    for val in valuelist:
        for pos in posMap[val]:
            finval.append(val)
            finoff.append(pos)

    return (finval, finoff)

> 
> --
> #decyphering tool
> text = [] #finval
> key = [] #finoff
> numdict = {}
> z = 0
> for r in key: #This here is the heart of the decryption process.
> Simply coalates the lists.
>     numdict[r] = text[z]
>     z = z + 1
> readlist = numdict.keys()
> readlist.sort() #puts the keys/offsets in order
> plaintext = ""
> for r in readlist:
>     plaintext = plaintext + chr(numdict[r])
> print plaintext #prints plaintext

#This is my first attempt, an the one that changes your approach least:

def decipher(text, key):
    numdict = {}

    # It's easiest to use range when you want a list of indices
    for idx in xrange(len(key)):
       numdict[key[idx]] = text[idx]

    readlist = numdict.keys()
    readlist.sort()

    # Using "+" to build strings can be very slow; it needs to build
    # a new string every time you use it.
 
    plaintext = [ chr(numdict[r]) for r in readlist ]

    return "".join(plaintext)


# This is my second appempt...
def decipher(text, key):
    result = [None]*len(text)
    for val,pos in zip(text,key):
        result[pos] = val
    result = map(chr, result)
    return "".join(result)

# This is my third:
def decipher(text, key):
    temp = zip(key, text)
    temp.sort()
    return "".join(map(lambda x: chr(x[1]), 
                       temp))

============================================================

And now, some notes on security.  (I'm not trying to be harsh here,
and I know this was probably not written for serious use, but it's
important to understand that cryptography is a lot harder than it
looks, and I wouldn't want you to accidentally leak anything private
through trusting inadequate security.)

First of all, this cipher is a bit suspicious.  It has the undesirable
property that the frequency counts of the characters in the ciphertext
are the same as in the plaintext.  That is to say, if I know you're
sending either "attack at dawn" or "run for your lives", it's easy for
me to tell which one you sent.

Secondly, this cipher doesn't buy you a lot: The key is as long as the
plaintext.  If you have a secure place to store the key, you would
presumably use that place to store the plaintext as well.

Third, the scrambling step is more crucial than you think.  Without
scrambling, if I see the key-portion of a message, I'll have a pretty
good idea which parts of the message go where: indices for low letters
earlier than indices for high ones.  But with scrambling, the quality
of your cipher is limited by the power of your random number
generator. 

For a deeper introduction to cryptography, you may want to get a copy
of Schneier's _Applied Cryptography_, or read the FAQs from sci.crypt.

============================================================

Well, I hope this that didn't dampen your entusiasm for Python!
You're off to a good start (esp. for two days!), and I hope we'll hear
more of you in the future!

Welcome-to-comp-lang-python-ly Yrs,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From joshm at taconic.net  Mon Feb 25 22:36:22 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 25 Feb 2002 22:36:22 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu> <mailman.1014687041.32731.python-list@python.org> <a5eqvu$79a$1@news.cis.ohio-state.edu>
Message-ID: <3c7b0051_1@corp.newsgroups.com>

No, what he's saying is that the sender and the receiver operate
independently.  Think of the socket connection as a hose from the sender to
you, ending in a huge bucket.  You aren't allowed to carry the huge bucket
back to your own code, it is attached to the hose.  Instead, you have to use
your own bucket to scoop stuff out of the huge one, and then carry it off.
The value you pass to recv() is the size of your own bucket.  When you call
recv(), you are going to the huge bucket, scooping out as much as is in
there (up to the size of your own bucket), and going back to your own
program.  If you didn't get all that you needed in that one trip, you have
to go back.

The other half of it is that the sender of the stuff is pushing it through
the hose at its own pace.  If it is very busy doing other things, it might
not put the stuff through the hose very quickly.  So you might get to the
huge bucket looking for stuff, but there isn't very much stuff there.  So
you get what is there, and return.  It might not even be enough to fill your
own bucket.  (There might not even be anything for you on a given trip to
the big bucket.)

So the purpose of sending you the octet count (and I believe that octet ==
byte) is so that you'll know whether to expect more stuff to come out of the
hose or not.  You have to keep track of how much stuff you've gotten so far,
to know whether you have to go back and get more.  You handle this by
calling len() on the returned string, to see how much stuff actually got
scooped into your bucket.  You keep a running total of how much you've
gotten.

Does that help?

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From sab at NOSPAM.freeuk.com  Thu Feb  7 14:42:06 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Thu, 7 Feb 2002 19:42:06 -0000
Subject: Amount of files on a drive?
References: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk> <a3ov9n$dn9$1@newsg3.svr.pol.co.uk> <Xns91ACD5444241CRASXnewsDFE1@130.133.1.4> <slrna65a5i.1m1.rmunn@rmunn.dyndns.org>
Message-ID: <a3ul7p$66j$1@newsg2.svr.pol.co.uk>

Yep it was trimmed down code and yes i am using a global variable!  :)  I
thought of explaining this when i first posted but i was sure everyone would
assume it was a global.  he he...

G. Willoughby

"Robin Munn" <rmunn at pobox.com> wrote in message
news:slrna65a5i.1m1.rmunn at rmunn.dyndns.org...
> On 6 Feb 2002 04:57:54 GMT, Philip Swartzleonard <starx at pacbell.net>
wrote:
> >G. Willoughby || Tue 05 Feb 2002 07:56:57a:
> >
> >> i am using this at the minute:
> >> ------------------------------------------------------------------
> >> import win32api
> >> import string
> >> import os
> >>
> >> def countFiles(arg, dir, files):
> >>     for file in files:
> >>         count=count+1
> >>
> >> fileListings=[]
> >> driveList=string.split(win32api.GetLogicalDriveStrings(),'\0')[:-1]
> >>
> >> for drive in driveList:
> >>     os.path.walk(drive, countFiles, 0)
> >> ------------------------------------------------------------------
> >> but its pretty slow any other suggestion how i could speed it up??
> >
> >Well, first determine how long it actually takes (time it), and see how
it
> >compares to say, selecting everything on the drive and hitting properties
> >in the context menu to get a count (i.e. see how good windows itself is
at
> >it). It took my 800mhz athlon about 35 seconds to tell me that i have
6.15
> >gigs in 101,064 files in 3,930 folders on my C drive. Counting on that
> >scale can get slow.
>
> Also, the bottleneck here is going to be the drive's access time, not
> the processor speed. The access time is usually measured in milliseconds
> and is the average time for the drive's read/write head to move from one
> location to another on the disk surface. This operation is sometimes
> called a "seek". Since the drive has to perform at least one seek for
> every file, and usually more, the total time spent waiting for the hard
> drive far outweighs the time spent by the processor.
>
> On the other hand, I do see one thing you could do to improve your code.
> Chage the countFiles() function thus:
>
> ------------------------------------------------------------------
> def countFiles(arg, dir, files):
>     count = count + len(files)
> ------------------------------------------------------------------
>
> Also, I presume you're using a global variable to hold your count,
> unlike the trimmed-down example code you showed us. In the code you
> showed us above, the count variable would be a variable local to
> countFiles() and would not hold its value across separate invocations of
> countFiles(). Instead, you would do:
>
> ------------------------------------------------------------------
> count = 0
>
> def countFiles(arg, dir, files):
>     global count
>     count = count + len(files)
> ------------------------------------------------------------------
>
> The "global count" statement inside countFiles() specifies that from
> here on, the name count will refer not to a variable local to
> countFiles() but instead to the global variable of that name.
>
> Gee, this is reminding me of that student blooper (probably an urban
> legend) that went, "The Iliad was not written by Homer, but by another
> man of that name." :)
>
> --
> Robin Munn
> rmunn at pobox.com




From joshm at taconic.net  Sat Feb 16 01:09:10 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sat, 16 Feb 2002 01:09:10 -0500
Subject: Apology
References: <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net>
Message-ID: <3c6df547_2@corp.newsgroups.com>

"Timothy Rue" <threeseas at earthlink.net> wrote
> Ten years???? it was Feburary 1988 when I identified the nine commands.
>
> I never claimed to be a programmer and never wanted to make the VIC but
> rather someone to use it. But I also don't, and never will want to be
> abused by any programmer wanting to take credit and property rights to the
> VIC.

Well, after 14 years then (happy anniversary!) of ranting about it with
NOBODY else even desiring to take credit for it, don't you think you should
stop worrying?  You're only going to be abused by programmers who see it for
what it is.  Irrelevent.

All you've said, all this time, is that everything boils down to nine
things.  Yippee.  My entire genetic makeup can be done in four.  Doesn't
mean I find it useful to attempt to function purely on a cellular level.

> The computer industry has alot of ethical problems in this area of taking
> everything the can from the users, from their ideas, to their debugging
> efforts (feedback), to their money in payment for what came from them, and
> now rental....and always it is never quite right but lacking always... for
> upgrading recycling of consumer entrapment abuse.
>
> That's what evidence ..... well what is, is what is. You know what you
> are.

As opposed to the ethical problems in say the energy futures industry, or
savings and loans, or any of those other trifling scandals.  People of poor
moral character exist in all aspects of society.  I fail to see any evidence
they cluster around the programming profession.  Just because there are
companies out there who aren't the nicest, doesn't mean that PROGRAMMERS as
a group are to blame.

You seem to have this amazingly huge and likely malignant chip on your
shoulder about something, and I don't think it has anything to do with
autocoding.  Perhaps we can help you.  Tell us more about the bad people who
screwed you over so many years ago.  You'll feel better.  We'll feel better.
Programmers learn better by example anyway.  Tell us what they did which
made you so angry, and we'll promise to try not to do that.  We may even
apologize.  Wouldn't that be nice?

> There never will be enough programmers, but there is always exactly the
> perfect number of user to do all the things a user needs to do for
> themselves. Given the right freedom for the users, creating tools.

That's right.  Let's all return to the world of hunter/gatherers, where
nobody takes advantage of others, and we all struggle to survive.  If you
despise the evil technology so much, why do you continue to use it?  That
seems very contradictory.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From edcjones at erols.com  Thu Feb 28 18:48:56 2002
From: edcjones at erols.com (Edward C. Jones)
Date: Thu, 28 Feb 2002 18:48:56 -0500
Subject: Python DB that supports numbers
References: <mailman.1014919129.4109.python-list@python.org>
Message-ID: <3C7EC1E8.7070001@erols.com>

Randall Hopper wrote:
> I'm looking for a very simple Python DB which supports:
> 
>      1) ints and floats as keys and values, 
>      2) performs DB access in the client, and 
>      3) stores the DB on disk.  
> 
> Just a simple DB for quick projects without the setup headache of
> server-based database systems.
> 
> For example, on the following table I just want to: "select * where
> column_1 is 1".  I'm using raw binary files now, but I'd prefer something
> more self-describing like an SQL table.
> 
>      1       10       1.814  177.1   13.61   73.2
>      1       5        1.624  182.5   -9999   -9999
>      1       2        1.259  185.9   13.24   73.9
>      2       10       2.672  181.4   13.59   72.5
>      2       5        2.423  183.2   -9999   -9999
>      2       2        1.902  190.7   13.27   73
>      3       10       3.208  181.9   13.46   72.1
>      3       5        2.787  186.6   -9999   -9999
>      3       2        2.17   194.7   13.16   72.3
> 
> Does anything like this exist in Python?

I would use a Python dictionary which I would store to disk using module 
cPickle.

     table = {}

     def AddRow(table, row):
         key = row[0]
         if not table.has_key(key):
         table[key] = []
         table[key].append(row)

     AddRow(table, [1, 10, 1.814, 177.1, 13.61, 73.2])
     AddRow(table, [1, 5, 1.624, 182.5, -9999, -9999])
     # ...
     AddRow(table, [3, 2, 2.17, 194.7, 13.16, 72.3])

     print table[1]

In the C++ Standard Template Library, there is a thing called a 
Multimap. It is like a Python dictionary except that each key can occur 
more than once. My Python name for this concept is "MultiDict". A 
MultiDict can be implemented in Python as a dictionary whose values are 
lists (like table[1] above). A MultiDict is a natural way to process 
sets of n-tuples.

I have attached MultiDict.py which I use when the approach above is too 
simple but MySQL is overkill.

Thanks,
Ed Jones
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: MultiDict.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20020228/f38c6be1/attachment.ksh>

From phansen at kaval.com  Sun Feb 10 20:39:58 2002
From: phansen at kaval.com (Peter Hansen)
Date: Sun, 10 Feb 2002 20:39:58 -0500
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <mailman.1013381417.32346.python-list@python.org> <3C6703F4.8F25BE12@alcyone.com>
Message-ID: <3C6720EE.28E0B68B@kaval.com>

Erik Max Francis wrote:
> 
> That's what list('abc') would return.  But in actuality, for it to be
> consistent with the rest of string.split's behavior, 'abc'.split('')
> would have to return ['', 'a', 'b', 'c', ''], not ['a', 'b', 'c'].
> 
> Another beautiful example of why string.split('') should raise an error
> instead of trying to guess what you mean.

And another:

>>> ''.join(['ab', 'c'])
'abc'

So what should 'abc'.split('') return again?


From jason at jvoegele.com  Fri Feb 15 08:53:44 2002
From: jason at jvoegele.com (Jason Voegele)
Date: 15 Feb 2002 05:53:44 -0800
Subject: Coming of age...
References: <mailman.1013538140.7691.python-list@python.org> <a4bu3r$ic8$1@newshost.accu.uu.nl>
Message-ID: <91acf731.0202150553.298c1852@posting.google.com>

m.faassen at vet.uu.nl (Martijn Faassen) wrote in message news:<a4bu3r$ic8$1 at newshost.accu.uu.nl>...
> Andrew MacIntyre <andymac at bullseye.apana.org.au> wrote:
> > On 12 Feb 2002, Martijn Faassen wrote:
>  
> >> Another interesting thing the other day is that on comp.object someone
> >> was asking which language they should try to experiment with OO, and
> >> there came lots of suggestions to use Ruby (Python was in there, but Ruby was
> >> mentioned a *lot*). Seems that language has been crossing a barrier
> >> as well recently.
>  
> > My take: Ruby turns out to be a language of refuge for escapees from Perl
> > who can't cope with white-space significance (more of them than not, I
> > expect).
> 
> That doesn't explain the recommendations in comp.object, though. 
> They use smalltalk, Java, C++ yes, but I don't expect most of the people
> there are Perl users at all.

Yes, I can attest to the fact that not all Ruby programmers come from
Perl.  I use Python and Ruby fairly regularly, and my background is
Pascal -> C -> C++ -> Java -> Eiffel -> (Smalltalk, Ruby, and Python).

I've used Perl in the past, but never considered myself a "Perl guy". 
I can also attest to the fact that lots of Ruby programmers don't even
*like* Perl ;-)  The similarities between Ruby and Perl are only
skin-deep.  A lot of people like it for its semantic similarity with
Smalltalk, rather than its syntactic similarity to Perl.

In any event, I suspect that popularity of either Ruby or Python will
be mutually beneficial to the other.  Once one of them "crosses the
line", it won't be so hard for the other to do the same.

Jason Voegele


From jonathan at onegoodidea.com  Tue Feb  5 05:56:13 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Tue, 05 Feb 2002 10:56:13 +0000
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com> <7xheow3hl2.fsf@ruckus.brouhaha.com> <3C5F3A07.45126DAF@ccvcorp.com> <7xg04gisvz.fsf@ruckus.brouhaha.com>
Message-ID: <B8856ACD.4463%jonathan@onegoodidea.com>

On 5/2/2002 2:06, in article 7xg04gisvz.fsf at ruckus.brouhaha.com, "Paul
Rubin" <phr-n2002a at nightsong.com> wrote:

> OK, that takes care of 1.5.2.  I gather that 2.0.* never became very
> widespread.  So I wonder now, whether more people are using 2.1.* than
> 1.5.2, and whether there will still be a significant amount of 2.1
> installations when 3.0 is released.  If it sounds like there won't be,
> then I'll probably just put
> 
>  if 3/2==1: <print upgrade-to-new-script message and exit>
> 
> at the top of my scripts and then code for 2.1.  When 2.1 gets more
> displaced by 2.2 and later, I can switch over to import future division.

Eh? Sorry, what is the test meant to achieve? This test will succeed and the
program will exit in all versions of the interpreter until 3.0 when the
semantics of classic division change. Did you mean that to be a '<>' as a
test for the introduction of Python 3.0?

If you want maximum compatibility then continue to code in forward
compatible 1.5.2. Your scripts will be fine for a long time yet. I wouldn't
bother switching to using new 2.x features unless you need to.

After making such a lot of noise about backwards compatibility, you seem to
be remarkably willing to ditch your 1.5.2 users...


On a related note, what exactly broke RH Anaconda in Python 2.x? Has someone
investigated this. I'd be interested in hearing the actual cause rather than
the speculation that has been going on here.

Did anyone from the Python "inner circle" contact RedHat to ask them? or
discuss solving the problem to allow them to ship a modern version of the
interpreter?

Jonathan



From tdelaney at avaya.com  Tue Feb  5 17:45:38 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 6 Feb 2002 09:45:38 +1100 
Subject: My own list object
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A167@natasha.auslabs.avaya.com>

> I made a file mylist.py
> 
> # mylist.py
> class MyList:
>     def __init__(self):
>         self.list=[]
>     def append(self, item):
>         self.list.append(item)
> 
> >>> import mylist
> >>> x = mylist()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
> TypeError: object of type 'module' is not call
> 
> Why is this happening?  My list object must be a class in its own,
> importable file.

That statement is incorrect. Read up on Python classes and modules.

Basically, a module contains definitions of things, including classes. You
can have as many definitions as you want in a module. You can also have code
which executes when the module is imported.

A module is also a namespace - to get at the definitions within that
namespace, you have to qualify them somehow.

With your example, you import the module mylist (thus binding the name
'mylist' to the module object). You then try to use the mylist module as a
callable - which it is not. However, in the mylist namespace is an object
named 'MyList' - which is your class definition.

So the call you need to make is

    x = mylist.MyList()

You should now got to http://www.python.org/doc/current/tut/tut.html and
work your way through the tutorial. If you have further questions, the FAQ
http://www.python.org/cgi-bin/faqw.py should be able to help you.

Tim Delaney



From thomas.heller at ion-tof.com  Fri Feb  1 04:54:38 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Feb 2002 10:54:38 +0100
Subject: How to change the icon in window header of window developed using python?
References: <5d65cf44.0202010126.7b58df35@posting.google.com>
Message-ID: <a3dol3$17e163$1@ID-59885.news.dfncis.de>

"CodeTube" <codetube at yahoo.com> wrote in message news:5d65cf44.0202010126.7b58df35 at posting.google.com...
> Hello Everyone,
>
> Could some one please tell me how to change the icon (a red one saying
> TK) in the window title bar of a window generated by python usging
> tkinter.
> I have developed an application using python and the gui using
> Tkinter. However i am not able to change the red 'TK' icon to
> something more customized.
>

Take a look at http://hackicon.sf.net/.

HTH,
Thomas




From peter at engcorp.com  Thu Feb 28 19:32:31 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 28 Feb 2002 19:32:31 -0500
Subject: listing threads?
References: <mailman.1014930527.20065.python-list@python.org>
Message-ID: <3C7ECC1F.F4FD6AFD@engcorp.com>

maximilianscherr wrote:
> 
> in the docs theres the information, that i can list all active
> threads with threading.enumerate(), but i always just get the
> Mainthread displayed in the interpreter console.
> 
> im doing somethin like:
[...snip]
> then if i do threading.enumerate() i just get the mainthread int he
> list.
> 
> so how can i now list all (active) threads?

Just as you are trying to do.

Does this help?

>>> class test(t.Thread):
...   def run(self):
...     print 'hi'
...     self.terminated = 0
...     import time
...     while not self.terminated:
...        time.sleep(1)
...     print 'done'
...
>>> te  = test()
>>> te.start()
hi
>>> t.enumerate()
[<test(Thread-2, started)>, <_MainThread(MainThread, started)>]
>>> te.terminated = 1
>>> done

>>> t.enumerate()
[<_MainThread(MainThread, started)>]
>>>


Your problem is that (a) you don't have a run() method in 
your thread class, and that is what gets executed when you
call te.start(), and (b) your thread is therefore terminating
immediately upon creation, and therefore t.enumerate()
is returning only the main thread, because no others are
running.

-Peter


From frederic.giacometti at arakne.com  Tue Feb 26 21:18:09 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Wed, 27 Feb 2002 02:18:09 GMT
Subject: [ANN] JPE 2.0 The Java-Python Extension
Message-ID: <BlXe8.68772$dG.28424949@news1.rdc1.sdca.home.com>

This is the official announcement of the second version of JPE
(CVS release tag jpe_2_0_0).

JPE (The Java-Python Extension) is a production-grade seamless integration
of Python and Java.
JPE lets you call Java objects and classes from Python, and vice-versa,
within the same process space. JPE is reentrant (cross-language callbacks),
and supports multi-threading concurently in Java and Python.

The reference web site for JPE is http://jpe.sourceforge.net


As compared to its first release:

1) The build system of JPE has been completely reengineered,
and is based on a new build library (libplus).
In this new version:

  - Java and python are auto-detected from the command line,
and everything else is derived from there (Java libraries, library
paths...).

  - libplus generated its own makefiles, and runtime environment
(env variables) for the build and test phases; so there is no need
to set-up anything

  - everything is now plateform-independant, and the same scripts
are used on Win32 and posix (no more GNU make / cygwin make required on
windows)

  - JPE has been ported to posix environments.
However, a shared python library is required on these platforms.
You can build your onw, or use the branched version of Python 2.1,
named 'pythonx', which is provided for these purposes.

  - JPE is distributed in source form from CVS. Substantial efforts
have been invested to provide a config and build environment
that makes it straightforward to configure, build from source,
and test multi-component, multi-platform software systems.
Hopefully, that can be one step ahead from the current
autoconf or distutil systems.
Easily extensible to support arbitrary software components and tools,
libplus currently supports C, C++, Python, Java, and swig wrappers.
Libplus generates configuration-specific makefiles that chain software
component dependencies together, and provides runtime software/tool
configuration information.
Litteratly, this is an 'open-source' project :))


2) New capabilities of JPE, for the 2d release:

- Java bean support: Jbean properties are accessed as regular python
attributes

- Cross-language native array access: JPE integrates the bufarray python
extension type; a layer on top of the python buffer interface
that provides type-safe access to contiguous fixed-length C arrays
(actually, 99% of the situations in scientific C or Fortran libraries).
This integration permits to safely access arrays by shared memory
reference between Java, Python, and C; while hidding the native memory
format.


3) Acknowledgements:

Arakn?, and The Molecular Graphics Laboratory at The Scripps Research
Institute.


Benevolently yours,

Fr?d?ric Giacometti
fred at arakne.com




From gh_pythonlist at gmx.de  Mon Feb 11 03:22:20 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 11 Feb 2002 09:22:20 +0100
Subject: subclassing tuple
In-Reply-To: <LNBBLJKPBEHFEDALKOLCOEBLNLAA.tim.one@home.com>
References: <20020211071605.GA2528@lilith.hqd-internal> <LNBBLJKPBEHFEDALKOLCOEBLNLAA.tim.one@home.com>
Message-ID: <20020211082219.GB3126@lilith.hqd-internal>

Le 11/02/02 ? 02:51, Tim Peters ?crivit:
> [explanation why __new__ is needed for subclassed immutable types]

Thanks for this great explanation, Tim.

Guess I'll have to read Guido's Python 2.2 tutorial a few more times.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 8.0 ?C      Wind: 6.4 m/s



From tbryan at python.net  Wed Feb 13 21:21:03 2002
From: tbryan at python.net (Tom Bryan)
Date: Thu, 14 Feb 2002 02:21:03 GMT
Subject: Using tabs instead of spaces in IDLE?
Message-ID: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com>

Is there a way to use tabs exclusively for indentation in IDLE?  Basically, 
I'd like tabwidth to equal identwidth while permitting developers to adjust 
the indentwidth.  I see that AutoIndent.py says

    # indentwidth is the number of characters per logical indent level.
    # tabwidth is the display width of a literal tab character.
    # CAUTION:  telling Tk to use anything other than its default
    # tab setting causes it to use an entirely different tabbing algorithm,
    # treating tab stops as fixed distances from the left margin.
    # Nobody expects this, so for now tabwidth should never be changed.
    usetabs = 1
    indentwidth = 4
    tabwidth = 8    # for IDLE use, must remain 8 until Tk is fixed

But I don't quite understand that explanation.  

The reason I ask is that I'm going to be introducing Jython to a bunch of 
Java developers at work.  I expect to get a lot of complaints about blocks 
being delimited by whitespace (already got one), and I'd like to be ready 
with nice editor configurations so that it's never a practical problem for 
our team.  Our group's current coding style (for C++ and Java) includes 
indentation by tabs.  Every developer sets his editor to always insert a 
tab when the tab key is pressed, but they adjust the visual display of the 
tab size to be whatever they want it to be.  It doesn't look like IDLE is 
friendly to that type of customization.  I know that emacs is.  I'm 
currently investigating the other editors that my team uses, such as 
CodeWright.

Thanks,
---Tom



From wolfson at uchicago.edu  Sun Feb  3 01:29:36 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Sun, 03 Feb 2002 00:29:36 -0600
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com>
Message-ID: <pan.2002.02.03.00.29.35.889879.3558@uchicago.edu>

On Sat, 02 Feb 2002 23:55:38 -0600, Joshua Macy wrote:

> try:
>      from __future__ import division
> except ImportError:
>      pass

This construct is illegal, as __future__ imports must be the first
statements in a module.

##tmp.py:
try:
    from __future__ import division
except ImportError:
    pass


##interpreter:
Python 2.2 (#1, Dec 22 2001, 14:06:34) 
[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-81)] on linux2
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import tmp
Traceback (most recent call last):
  File "<pyshell#0>", line 1, in ?
    import tmp
SyntaxError: from __future__ imports must occur at the beginning of the 
file (tmp.py, line 2) 
>>> 

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From jeff at ccvcorp.com  Mon Feb 18 21:04:02 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 18 Feb 2002 18:04:02 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com>
Message-ID: <3C71B291.B8574221@ccvcorp.com>


Peter Milliken wrote:

>
> There are regular discussions on the usefulness of Python for "larger"
> projects. Many claim that they use it without any problems - I suspect that
> it is from a background of ignorance and lack of metrics ...

................

> Of course, it takes me longer to write
> up front but overall development effort (in my subjective opinion - since I
> have no metrics :-)) is that it is much lower.

So, you're just as guilty of the subjectiveness as those Python-supporters, yet
you feel justified in telling them that they're wrong?  :)

Standard disclaimer: I don't have any metrics either, nor am I familiar with Ada
at all, so I'm not making any specific claims either way.  However, there seems
to be enough people making the claim in Python's favor, that I'm inclined to
believe it until such time as I see real (objective) evidence that it's false.
I freely admit that I've never worked on a large (or even midsized) coding
project, in Python or otherwise, but in my year of using Python I can't recall a
single instance of a grueling debugging session finding a name- or type- error
as the cause.  When I compare this to how much time it took me to get C/C++
programs to compile at all, and *then* verify that they worked *correctly*,
Python is a clear win.  Again, perhaps Ada is easier than C++ (I'm sure that
*you* think so >wink< ), but all of my (limited) personal experience supports
the Pythonic claims.  (And please note, I'm intending this as a friendly
discussion, not a personal attack.  I'm *not* trying to flame you, by any means
-- I know I don't know enough to get away with that.  I just find it interesting
that you are always advocating Ada....)

Jeff Shannon
Technician/Programmer
Credit International





From cfelling at iae.nl  Thu Feb 28 08:21:33 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 14:21:33 +0100
Subject: PEP 263 comments
References: <mailman.1014762060.30054.python-list@python.org> <m3u1s31eah.fsf@mira.informatik.hu-berlin.de> <87g03lhm3p.fsf@tleepslib.sk.tsukuba.ac.jp>
Message-ID: <a5last$23p$1@animus.fel.iae.nl>

Stephen J. Turnbull <stephen at xemacs.org> wrote:
> Hi, I'm Steve Turnbull, I do XEmacs.  Mostly Mule.  Barry asked me to
> step up to bat on this.

... lots of interesting experiences snipped

Thanks for sharing, hope this helps us to do the right thing.

Now that we have your attention, what's your personal opinion on using
something else but ascii (in whatever encoding:) for the code itself
(things like keywords and variable names).

My Western centered view would be that it would make reading code
harder, but that's probably what the rest of the world already
experiences.

I personally use English even for my privat code and I even try to
stick to English in my comments, just in case some foreignor might have
to look at it:)
-- 
groetjes, carel


From sabren at manifestation.com  Sat Feb 23 09:16:49 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 23 Feb 2002 09:16:49 -0500 (EST)
Subject: help: Problem with cgi form
In-Reply-To: <Xns91BEB7AC37F78AusYourStandingInIt@130.133.1.4>
Message-ID: <Pine.LNX.4.44.0202230915050.23509-100000@mercury.sabren.com>

On 23 Feb 2002, Flavian Hardcastle wrote:

> I know when it's undeclared ... whenever a user first
> tries to access the form, it is undeclared. What I want to
> know is how to get the script to check whether this value
> has not been declared yet.

Um. Why don't you just declare it, and assign a default
value?

Failing that:

     locals().get("varname", "defaultvalue")

or:

     is_declared = locals().has_key("varname")

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From philh at comuno.freeserve.co.uk  Tue Feb 12 10:46:49 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Tue, 12 Feb 2002 15:46:49 +0000
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org>
Message-ID: <slrna6ie79.2so.philh@comuno.freeserve.co.uk>

On Mon, 11 Feb 2002 18:37:51 -0600, Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
>
>Indeed.  He uses a nonstandard definition of "language" and applies
>"physics"
>to programming.  He can't program, you know; ten years on this project and
>nothing to show for it.  If I had tried to develop something like this for
>ten years and had nothing to show for it but a website, I'd give up.

At least his code is all neatly lined up!

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From phr-n2002a at nightsong.com  Wed Feb 20 23:21:18 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 20:21:18 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <3C747268.6C756B9D@earthlink.net>
Message-ID: <7xvgcrfosh.fsf@ruckus.brouhaha.com>

Hans Nowak <wurmy at earthlink.net> writes:
> > Why doesn't sort() return the sorted list.  I would like to chain it
> > to other operations:
> >         b=[x for x in a].sort()
> 
> The other replies told you why list.sort() doesn't return
> the sorted list. You can easily roll your own function,
> though:
> 
> >>> def sort2(lst):
> 	z = lst[:]
> 	z.sort()
> 	return z
> ...
> This sort2() function returns a new, sorted list
> without affecting the original one. Don't use this
> when performance is an issue, though... it's not
> very efficient because it copies the original
> list first.

Just write it like this:

def sort3(lst):
  lst.sort()
  return lst

The list comprehension already makes a temporary list.  You don't need
to copy it around again.


From phd at phd.pp.ru  Fri Feb  1 05:34:30 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 1 Feb 2002 13:34:30 +0300
Subject: locale on cygwin
In-Reply-To: <3C5A6BC1.6080009@halfcooked.com>; from andy47@halfcooked.com on Fri, Feb 01, 2002 at 09:19:45PM +1100
References: <Xns91A1A678DEEE3andy47halfcookedcom@203.109.250.24> <20020125140805.GC1252@dothill.com> <3C5A607F.6060107@halfcooked.com> <20020201123241.H10316@phd.pp.ru> <3C5A6BC1.6080009@halfcooked.com>
Message-ID: <20020201133430.M10316@phd.pp.ru>

On Fri, Feb 01, 2002 at 09:19:45PM +1100, Andy Todd wrote:
> Thanks for that, but this is what happens if I don't force it;
> 
> """
> $ echo $LANG
> C
> $ python
> Python 2.1.1+ (#1, Jan  8 2002, 00:37:12)
> [GCC 2.95.4 20011006 (Debian prerelease)] on linux2
> Type "copyright", "credits" or "license" for more information.
>  >>> import locale
>  >>> locale.getdefaultlocale()
> Traceback (most recent call last):
>    File "<stdin>", line1, in ?
>    File "/usr/lib/python2.1/locale.py", line 339, in getdefaultlocale
>      return _parse_localename(localename)
>    File "/usr/lib/python2.1/locale.py", line 272, in _parse_localename
>      raise ValueError, 'unknown locale: %s' % localename
> ValueError: unknown locale:
> """

   Strange. It works perfectly for my. On LNAG=C it returns (None, None),
on LANG=ru_RU it returns ("ru_RU", "ISO-8859-5").

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



From jimmy at retzlaff.com  Fri Feb 15 11:45:42 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Fri, 15 Feb 2002 08:45:42 -0800
Subject: VB code -> Python
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F559E@MAYOR.retzlaff.com>

>    Set oContext = GetObjectContext
>    Err.Clear
>    Set oUser = GetObject("WinNT://" & CStr(Domain) & "/" &
CStr(UserID))
>    If Err = 0 Then
>        For Each oGroup In oUser.Groups
>            If UCase(Groupname) = UCase(oGroup.Name) Then
>                MemberofGroup = True
>                'Exit Function
>                Exit For
>            End If
>        Next
>        oContext.SetComplete
>    Else
>        oContext.SetAbort
>        MemberofGroup = False
>    End If
>
>So my questin is how do I translate the GetObject(...) thing in to
>Python code.

Here is a version in Python, using your same variable names, etc.:

import win32com.client

def MemberofGroup(Domain, UserID, Groupname):
    try:
        oUser = win32com.client.GetObject(r'WinNT://%s/%s' % (Domain,
UserID))
    except:
        return 0

    for oGroup in oUser.Groups():
        if Groupname.upper() == oGroup.Name.upper():
            return 1

print MemberofGroup(Domain='CoolDomain', UserID='Cool Guy',
Groupname='Domain Users')


Jimmy



From fredrik at pythonware.com  Mon Feb 11 17:05:22 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 22:05:22 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <3C67C369.597E3517@optonline.net> <mailman.1013434940.27537.python-list@python.org> <3C67D328.985269AC@optonline.net>
Message-ID: <CeX98.17824$n4.3075925@newsc.telia.net>

Frederick wrote:

> Whoops! He's ... Herr Professor Doktor Gerhard Wuensch.

a quick googling revealed the following quotation:

    "I always try to write music which is accessible to the
    widest possible audience, reasonably well crafted and
    economical, in that it requires a minimum of rehearsal
    time."

sounds like he would have been a great python developer,
had he picked up computing instead of music ;-)

</F>




From jriveramerla at yahoo.com  Fri Feb  1 11:58:23 2002
From: jriveramerla at yahoo.com (Jose Rivera)
Date: 1 Feb 2002 08:58:23 -0800
Subject: Is there a Metakit release for python 2.2?
Message-ID: <11e94203.0202010858.17a6bce7@posting.google.com>

Hi..
I just Installed Python 2.2 and I am getting the followin error:

Traceback (most recent call last):
  File "D:\scripts\Rev_Si_Hay_Archivos_En_BCKS_Sin_Procesar.py", line
2, in ?
    import metakit
  File "D:\TOOLS\Python22\lib\metakit.py", line 12, in ?
    from Mk4py import *
ImportError: Module use of python21.dll conflicts with this version of
Python.

It worked with Python 2.1, and I searched on the WEB and did not found
a comment about a new release.

Thanks for your help.


From teddrain at earthlink.net  Tue Feb 26 15:31:01 2002
From: teddrain at earthlink.net (Ted Drain)
Date: Tue, 26 Feb 2002 20:31:01 GMT
Subject: Building 2.2 with the Intel icc compiler
Message-ID: <9gSe8.18435$ZC3.1526281@newsread2.prod.itd.earthlink.net>

Has anyone built Python 2.2 with the new Intel C/C++ compiler icc?  I
configured Python and then edited the Makefile by hand to use icc and icc
options.  Python builds but the modules all report errors (and then remove
themselves).

This compiler generates code that is roughly 30% faster than gcc.  Check
out:
http://www.open-mag.com/754088105111.htm

Any thoughts?
Ted




From tismer at tismer.com  Tue Feb 26 06:48:17 2002
From: tismer at tismer.com (Christian Tismer)
Date: Tue, 26 Feb 2002 12:48:17 +0100
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org> <m3pu2sancq.fsf@chvatal.cbbrowne.com>
Message-ID: <3C7B7601.70906@tismer.com>

Christopher Browne wrote:

...


> If it gets to the point of supporting billions of threads, it would
> _certainly_ make sense to call it "NanoThreading Python."  Or
> "NanoTech Python."  :-)


I doubt that we can easily have a million of threads, so
even the microthread name is a bit optimistic.
MilliThreading is more accurate. Esspecially since a
typical thread takes 1 MB of memory, and tasklets use
a thousandth of that.

> There have got to be some puns heading in the direction of "seriously
> stacked."  Hmm...


Ok, here what "Stackless" really means:
I didn't earn haystacks of money with it yet.
When that happens, I might rename it :)

Pythonlet -- yes, thousands of little snakelets
squirreling around me -- huh! I need a nap

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From jason at jorendorff.com  Tue Feb  5 20:31:00 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 5 Feb 2002 19:31:00 -0600
Subject: My own list object
In-Reply-To: <a3p01b$186eq8$1@ID-98166.news.dfncis.de>
Message-ID: <HFEKILOLEFEFMKAECNDLIEKFDAAA.jason@jorendorff.com>

MDK writes:
> I am trying to create my own special type of list object.
> 
> However, when I do x.append('yo') it uses Python's append instead 
> of the one from my class.
> 
> If this is not easy can someone at least point me to an example of what I
> want to do?

It is very easy.  But it requires Python 2.2.

class X(list):
    def append(self, something):
        print "Hello, world!"

>>> x = X()
>>> x.append(14)
Hello, world!

## Jason Orendorff    http://www.jorendorff.com/



From martin at v.loewis.de  Tue Feb 26 12:52:03 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 18:52:03 +0100
Subject: Python 2.2 on SunOS5.7 install errors
References: <alex-723057.17122625022002@iruka.swcp.com>
Message-ID: <m3u1s4yvuk.fsf@mira.informatik.hu-berlin.de>

Alex Chavez <alex at newicons.com> writes:

> I'm getting the error msg below when I attempt to install python 2.2 on 
> a sun solaris 7 os5.7 sparc box, using gcc 3.0.3
> 
> 
> *** Error code 1
> make: Fatal error: Command failed for target `Parser/tokenizer.o'
> 
> 
> 
> I hope I'm not duplicating the Jeff Bauer's post. I read the thread but 
> I cant make heads or tails of it.

It's certainly different, but we need to know more details. What was
the command that failed, and did that give any indication how it failed?

Regards,
Martin



From hmartires at cultalg.pt  Mon Feb 25 07:32:40 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 12:32:40 -0000
Subject: Counting bytes
Message-ID: <Q7qe8.593$LI1.7401571@newsserver.ip.pt>

How can i count the number of bytes of:
--> List []
--> array('d',[])

There is any function that return's the number of bytes ?

Tanks in advance





From jason at jorendorff.com  Sun Feb 10 22:51:34 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 21:51:34 -0600
Subject: FixedPoint
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBMEMJKBAA.marklists@mceahern.com>
Message-ID: <HFEKILOLEFEFMKAECNDLEEKBDBAA.jason@jorendorff.com>

Mark McEahern wrote:
> [Jason Orendorff]
> > How about:
> >
> >     def __cmp__(self, other):
> >         try:
> >             xn, yn, p = _norm(self, other)
> >             return cmp(xn, yn)
> >         except:
> >             return cmp(float(self), other)
> 
> I like that!  Would it be better style to trap only the specific error I
> know about?

Yep.  If you know what's in _norm() you can do better than I
can.  :)  Also, you might pull the first return statement out
of the try.

## Jason Orendorff    http://www.jorendorff.com/



From python at dejazzd.com  Thu Feb 28 14:08:22 2002
From: python at dejazzd.com (Glenn Stauffer)
Date: Thu, 28 Feb 2002 14:08:22 -0500
Subject: mime multipart messages
Message-ID: <20020228190909.E03C117E8C@juniper.its.swarthmore.edu>

I'm using the MimeWriter module to attach a file, submitted through an html 
form, to an email message.  When I do this and send mail to a unix machine 
and read the mail with mutt or kmail, the file is attached to the message as 
I would expect.  However, when the message is opened on a Mac or Windows 
machine, the file is mixed into the body of the message.

This is the code I use to construct the message:

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

message = StringIO.StringIO()
writer = MimeWriter.MimeWriter(message)
writer.addheader('Subject', mail_subject)
writer.startmultipartbody('mixed')

part=writer.nextpart()
body = part.startbody('text/plain')
body.write(msgbody)

part=writer.nextpart()
part.addheader('Content-Type','application/mixed')
part.addheader('Content-Transfer-Encoding', 'base64')
part.addheader('Content-Disposition','attachment; filename=%s' % 
form['file'].filename)
body = part.startbody('application/binary')
base64.encode(open(tmpfile,'rb'), body)

writer.lastpart()

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

This is how the file looks when someone opens it with Eudora on a Windows PC:

--130.58.64.30.99.29450.1014921136.814.11340
Content-Type: text/plain


--130.58.64.30.99.29450.1014921136.814.11340
Content-Type: application/binary
Content-Type: application/mixed
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename=README

PT09PT09PT09PT09PT09PT09PT0NClRoZSB4bWxycGNsaWIgbW9kdWxlDQo9PT09PT09PT09PT09
PT09PT09PQ0KDQpUaGlzIGtpdCBjb250YWlucyBhIFB5dGhvbiBpbXBsZW1lbnRhdGlvbiBmb3Ig

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

Anyone have any thoughts on what I'm doing wrong or where to look to figured 
this problem out?

Thanks,

Glenn



From ryan.phillips at csus.edu  Tue Feb 12 12:49:19 2002
From: ryan.phillips at csus.edu (Ryan Phillips)
Date: 12 Feb 2002 17:49:19 GMT
Subject: Help with Installer.
References: <Xns91B3864EE6013maccetipl@149.156.4.11> <Xns91B36755D3F49gmcmhypernetcom@199.171.54.214> <Xns91B3A738AD5C1maccetipl@149.156.4.11>
Message-ID: <a4bkiv$972@dispatch.concentric.net>

Did you set the file's attributes to be executable?
chmod +x some_file_name

-ryan

"Maciej Starzyk" <mac at ceti.pl> wrote in message
news:Xns91B3A738AD5C1maccetipl at 149.156.4.11...
> Gordon McMillan <gmcm at hypernet.com> wrote in
> news:Xns91B36755D3F49gmcmhypernetcom at 199.171.54.214:
>
> > The only way I know of (and boy, I hope I've
> > missed something!) is to install the executable
> > on the user's PATH. It should be able to find
> > itself, at the expense of some hunting and pecking.
> >
> > -- Gordon
> > http://www.mcmillan-inc.com/
> >
>
> Thanks for your reply.
> Unfortunately, I cannot put it in PATH.
> It is executed as a CGI script, and I have no
> write access to any directory in the apache's PATH.
>
> Maciek




From quinn at hurl.ugcs.caltech.edu  Mon Feb 11 18:05:07 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 11 Feb 2002 23:05:07 GMT
Subject: why i love python...
References: <mailman.1013045771.13577.python-list@python.org>
Message-ID: <slrna6gjh0.44n.quinn@hurl.ugcs.caltech.edu>

On Wed, 6 Feb 2002 17:34:52 -0800, Mark McEahern <mark at mceahern.com> wrote:
>python comes with a slew of special methods for objects that you can
>leverage to do some really neat things.  consider __str__, which is called
>on your objects when you <b>print</b> them.  if you don't define __str__,
>that's ok, you'll get something like this:

Of course, just about every object oriented language has this, from haskell's
'show' to CLOS's 'print-object' to eiffel's 'out', so python doesn't have
a monopoly :)

>obviously, i should return something more meaningful than that.  for
>instance, it might be nice to return an xml representation of the object.
>again, with python this is unbelievably easy:

I'd assert that this is a misuse of __str__.  __str__ is supposed to be
something all objects use to present human readable output.  I would only use
it to make XML if a) I was prepared to make *all* objects' __str__s return XML
(already too late for that :) and b) I thought XML was human-readable (way too
late for that).

I'd go with a 'to_xml()'.  __str__ is just overloading the built in function
str().  There's nothing special about it except that the REPL invokes it
automatically.  You wouldn't overload '+' to print strings, would you (sather
notwithstanding)?  Or, you wouldn't overload bit-shift to... err... pay
no attention to those other languages.

>this snippet shows:
>
><ul>
>  <li>how you can use self.__dict__ to access an object's attributes.</li>
>  <li>how you can iterate over the keys of a dictionary (i believe that's
>new with python 2.2).</li>
>  <li>how you can use self.__class__ to access an instance's class; and
>__class__.__name__ to access the name of the class as a string.</li>
></ul>

Yep, python's introspection is nice.

>i think this shows some of the power of python.  particularly, how special
>methods can save you an amazing amount of time.  to me, the natural next

There's nothing special about most __special__ methods except some simple
syntax transformation (str(o) -> o.__str__()).  And if you overload a widely
used, general purpose method to do something different from its usual meaning
you might spend an amazing amount of time debugging why some objects aren't
getting XML-ified properly (they're actually the wrong type, but that's hidden
by the fact that they already have a working __str__ method).

><ul>
>  <li>handle list attributes intelligently.</li>
>  <li>determine whether attributes are themselves subclasses of PrintsXml,
>in which case you can just let them print themselves rather than wrapping
>them in xml tags manually.</li>

A more pythonic way to do the second would be:

for elt in stuff_to_xmlify:
    try:
        return elt.to_xml()
    except AttributeError:
        fallback general-purpose xmlification

>of course, i haven't even touched on the most exciting aspect of all this:
>__repr__.  __repr__ allows me to return a string representation of my object
>that can be used to re-create it.  what i don't know is how that compares to
>pickling, but i'm sure the latter is tastier.  ;-)

I hope this doesn't excite you unduly, but it's not just __repr__: *any*
method allows you to return a string representation of your object! :)
That is, if your object can be recreated from a string, which most can't
easily, and those that can do so inefficiently.

The relationship with pickle is that __repr__ traditionally (but actually
usually doesn't) flattens your object to a string which should be short,
printable, and reconstitutable by the python parser (the first one often
contradicts the other two, hence pickle), while pickle flattens your object to
a string which can be really long, not necessarily printable (containing
binary) and is reconstitutable only by the unpickler.

repr() usually *doesn't* flatten to a string, actually, so the distinction
between __str__ and __repr__ is sort of vague.  The exception is for simple
built-in types, like ints, floats, strings, lists, etc., where the difference
may be subtle but is important (compare str('hello\nthere') with
repr('hello\nthere')).

If the python developers had gotten really anal about repr() doing what it's
purpose supposedly is, then repr(function) would print the text of the
function.  And then if the compiler used repr() instead of marshal() to
save compiled modules to disk, .pyc files would be the same as the .py files
they were supposedly compiled from :)

and-what-fun-that-would-be-ly y'rs


From paul at prescod.net  Tue Feb 12 06:32:07 2002
From: paul at prescod.net (Paul Prescod)
Date: Tue, 12 Feb 2002 03:32:07 -0800
Subject: High school Python video
References: <4747DACE-1BFE-11D6-9BC0-003065A7CFBA@mac.com>
Message-ID: <3C68FD37.D285AA2@prescod.net>

Steven Majewski wrote:
> 
>...
> 
> BTW: Do you happen to know the link given at the end of the Student Video
> ?
> ( Or the name of the high school ? )
> I think it was on opencontent.org but I couldn't find anything there.

http://www.careercenter.arlington.k12.va.us/tvprod/clientprojects.htm

Not only is the video there, there is also a rap by Barry Warsaw though
I can't hear the words on my laptop speakers. Gotta burn the CD. ;)

 Paul Prescod



From grante at visi.com  Mon Feb  4 13:37:16 2002
From: grante at visi.com (Grant Edwards)
Date: Mon, 04 Feb 2002 18:37:16 GMT
Subject: How Can I add serial comms events to Tkinter mainloop?
References: <ORr78.26230$oK1.225674@NewsReader>
Message-ID: <wxA78.18172$Wf1.4913678@ruti.visi.com>

In article <ORr78.26230$oK1.225674 at NewsReader>, JH wrote:

> I am trying to write a simple serial console. I was hoping to
> use Tkinter for the GUI but I cannot see how to add RX chars
> from my serial port to the console while Tk.mainloop is running
> is it me ?

You use Tkinter.tkinter.createfilehandler to add a "handler"
routine to a file.  This works for serial ports as well -- at
least under Unix. If you want to see a demo of how this is done
to handle data from a pipe, you can look at 

 ftp://ftp.visi.com/users/grante/python/readHandlerDemo.py

The relevent stuff is in the "start" method of the execWindow
class.

-- 
Grant Edwards                   grante             Yow!  If elected, Zippy
                                  at               pledges to each and every
                               visi.com            American a 55-year-old
                                                   houseboy...


From trentm at ActiveState.com  Fri Feb 15 19:43:33 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Fri, 15 Feb 2002 16:43:33 -0800
Subject: PEP 282: A Logging System  --  comments please
Message-ID: <20020215164333.A31903@ActiveState.com>

Howdy all,

I would appreciate any comments you might have on this proposal for adding a
logging system to the Python Standard Library. This PEP is still an early
draft so please forward your comments just to me directly for now.

Thanks,
Trent

-----------------------------------------------------------
PEP: 282
Title: A Logging System
Version: $Revision: 1.1 $
Last-Modified: $Date: 2002/02/15 04:09:17 $
Author: trentm at activestate.com (Trent Mick)
Status: Draft
Type: Standards Track
Created: 4-Feb-2002
Python-Version: 2.3
Post-History:


Abstract

    This PEP describes a proposed logging package for Python's
    standard library.

    Basically the system involves the user creating one or more
    logging objects on which methods are called to log debugging
    notes/general information/warnings/errors/etc.  Different logging
    'levels' can be used to distinguish important messages from
    trivial ones.
    
    A registry of named singleton logger objects is maintained so that

        1) different logical logging streams (or 'channels') exist
           (say, one for 'zope.zodb' stuff and another for
           'mywebsite'-specific stuff)
           
        2) one does not have to pass logger object references around.

    The system is configurable at runtime.  This configuration
    mechanism allows one to tune the level and type of logging done
    while not touching the application itself.

    
Motivation

    If a single logging mechanism is enshrined in the standard
    library, 1) logging is more likely to be done 'well', and 2)
    multiple libraries will be able to be integrated into larger
    applications which can be logged reasonably coherently.


Influences

    This proposal was put together after having somewhat studied the
    following logging packages:

        o java.util.logging in JDK 1.4 (a.k.a. JSR047) [1]
        o log4j [2]
          These two systems are *very* similar.
        o the Syslog package from the Protomatter project [3]
        o MAL's mx.Log package [4]

    This proposal will basically look like java.util.logging with a
    smattering of log4j.


Simple Example

    This shows a very simple example of how the logging package can be
    used to generate simple logging output on stdout.
    
        --------- mymodule.py -------------------------------
        import logging
        log = logging.getLogger("MyModule")

        def doit():
            log.debug("doin' stuff")
            # do stuff ...
        -----------------------------------------------------

        --------- myapp.py ----------------------------------
        import mymodule, logging
        log = logging.getLogger("MyApp")

        log.info("start my app")
        try:
            mymodule.doit()
        except Exception, e:
            log.error("There was a problem doin' stuff.")
        log.info("end my app")
        -----------------------------------------------------

    > python myapp.py
    0    [myapp.py:4] INFO  MyApp - start my app
    36   [mymodule.py:5] DEBUG MyModule - doin' stuff
    51   [myapp.py:9] INFO  MyApp - end my app
    ^^   ^^^^^^^^^^^^ ^^^^  ^^^^^   ^^^^^^^^^^
    |    |            |     |       `-- message
    |    |            |     `-- logging name/channel
    |    |            `-- level
    |    `-- location
    `-- time

    NOTE: Not sure exactly what the default format will look like yet.


Control Flow

    [Note: excerpts from Java Logging Overview. [5]]

    Applications make logging calls on *Logger* objects.  Loggers are
    organized in a hierarchical namespace and child Loggers may
    inherit some logging properties from their parents in the
    namespace.
    
    Notes on namespace: Logger names fit into a "dotted name"
    namespace, with dots (periods) indicating sub-namespaces.  The
    namespace of logger objects therefore corresponds to a single tree
    data structure.

       "" is the root of the namespace
       "Zope" would be a child node of the root
       "Zope.ZODB" would be a child node of "Zope"

    These Logger objects allocate *LogRecord* objects which are passed
    to *Handler* objects for publication.  Both Loggers and Handlers
    may use logging *levels* and (optionally) *Filters* to decide if
    they are interested in a particular LogRecord.  When it is
    necessary to publish a LogRecord externally, a Handler can
    (optionally) use a *Formatter* to localize and format the message
    before publishing it to an I/O stream.

    Each Logger keeps track of a set of output Handlers.  By default
    all Loggers also send their output to their parent Logger.  But
    Loggers may also be configured to ignore Handlers higher up the
    tree.

    The APIs are structured so that calls on the Logger APIs can be
    cheap when logging is disabled.  If logging is disabled for a
    given log level, then the Logger can make a cheap comparison test
    and return.  If logging is enabled for a given log level, the
    Logger is still careful to minimize costs before passing the
    LogRecord into the Handlers.  In particular, localization and
    formatting (which are relatively expensive) are deferred until the
    Handler requests them.


Levels
    
    The logging levels, in increasing order of importance, are:

        DEBUG
        INFO
        WARN
        ERROR
        FATAL
        ALL

    This is consistent with log4j and Protomatter's Syslog and not
    with JSR047 which has a few more levels and some different names.

    Implementation-wise: these are just integer constants, to allow
    simple comparison of importance.  See "What Logging Levels?" below
    for a debate on what standard levels should be defined.


Loggers

    Each Logger object keeps track of a log level (or threshold) that
    it is interested in, and discards log requests below that level.

    The *LogManager* maintains a hierarchical namespace of named
    Logger objects.  Generations are denoted with dot-separated names:
    Logger "foo" is the parent of Loggers "foo.bar" and "foo.baz".

    The main logging method is:

        class Logger:
            def log(self, level, msg, *args):
                """Log 'msg % args' at logging level 'level'."""
                ...

    however convenience functions are defined for each logging level:

        def debug(self, msg, *args): ...
        def info(self, msg, *args): ...
        def warn(self, msg, *args): ...
        def error(self, msg, *args): ...
        def fatal(self, msg, *args): ...

    XXX How to defined a nice convenience function for logging an exception?
        mx.Log has something like this, doesn't it?

    XXX What about a .raising() convenience function?  How about:

            def raising(self, exception, level=ERROR): ...

        It would create a log message describing an exception that is
        about to be raised.  I don't like that 'level' is not first
        when it *is* first for .log().


Handlers

    Handlers are responsible for doing something useful with a given
    LogRecord.  The following core Handlers will be implemented:

        - StreamHandler: A handler for writing to a file-like object.
        - FileHandler: A handler for writing to a single file or set
          of rotating files.

    More standard Handlers may be implemented if deemed desirable and
    feasible.  Other interesting candidates:

        - SocketHandler: A handler for writing to remote TCP ports.
        - CreosoteHandler: A handler for writing to UDP packets, for
          low-cost logging.  Jeff Bauer already had such a system [5].
        - MemoryHandler: A handler that buffers log records in memory
          (JSR047).
        - SMTPHandler: Akin to log4j's SMTPAppender.
        - SyslogHandler: Akin to log4j's SyslogAppender.
        - NTEventLogHandler: Akin to log4j's NTEventLogAppender.


Formatters

    A Formatter is responsible for converting a LogRecord to a string
    representation.  A Handler may call its Formatter before writing a
    record.  The following core Formatters will be implemented:

        - Formatter: Provide printf-like formatting, perhaps akin to
          log4j's PatternAppender.
    
    Other possible candidates for implementation:

        - XMLFormatter: Serialize a LogRecord according to a specific
          schema.  Could copy the schema from JSR047's XMLFormatter or
          log4j's XMLAppender.
        - HTMLFormatter: Provide a simple HTML output of log
          information. (See log4j's HTMLAppender.)


Filters

    A Filter can be called by a Logger or Handler to decide if a
    LogRecord should be logged.

    JSR047 and log4j have slightly different filtering interfaces. The
    former is simpler:

        class Filter:
            def isLoggable(self):
                """Return a boolean."""

    The latter is modeled after Linux's ipchains (where Filter's can
    be chained with each filter either 'DENY'ing, 'ACCEPT'ing, or
    being 'NEUTRAL' on each check).  I would probably favor to former
    because it is simpler and I don't immediate see the need for the
    latter.
    
    No filter implementations are currently proposed (other that the
    do nothing base class) because I don't have enough experience to
    know what kinds of filters would be common.  Users can always
    subclass Filter for their own purposes.  Log4j includes a few
    filters that might be interesting.


Configuration

    Note: Configuration for the proposed logging system is currently
    under-specified.

    The main benefit of a logging system like this is that one can
    control how much and what logging output one gets from an
    application without changing that application's source code.

    Log4j and Syslog provide for configuration via an external XML
    file.  Log4j and JSR047 provide for configuration via Java
    properties (similar to -D #define's to a C/C++ compiler).  All
    three provide for configuration via API calls.

    Configuration includes the following:

        - What logging level a logger should be interested in.
        - What handlers should be attached to which loggers.
        - What filters should be attached to which handlers and loggers.
        - Specifying attributes specific to certain Handlers and Filters.
        - Defining the default configuration.
        - XXX Add others. 

    In general each application will have its own requirements for how
    a user may configure logging output.  One application
    (e.g. distutils) may want to control logging levels via
    '-q,--quiet,-v,--verbose' options to setup.py.  Zope may want to
    configure logging via certain environment variables
    (e.g. 'STUPID_LOG_FILE' :).  Komodo may want to configure logging
    via its preferences system.

    This PEP proposes to clearly document the API for configuring each
    of the above listed configurable elements and to define a
    reasonable default configuration.  This PEP does not propose to
    define a general XML or .ini file configuration schema and the
    backend to parse it.
    
    It might, however, be worthwhile to define an abstraction of the
    configuration API to allow the expressiveness of Syslog
    configuration.  Greg Wilson made this argument:

        In Protomatter [Syslog], you configure by saying "give me
        everything that matches these channel+level combinations",
        such as "server.error" and "database.*".  The log4j "configure
        by inheritance" model, on the other hand, is very clever, but
        hard for non-programmers to manage without a GUI that
        essentially reduces it to Protomatter's.


Case Scenarios

    This section presents a few usage scenarios which will be used to
    help decide how best to specify the logging API.

    (1) A short simple script.

        This script does not have many lines.  It does not heavily use
        any third party modules (i.e. the only code doing any logging
        would be the main script).  Only one logging channel is really
        needed and thus, the channel name is unnecessary.  The user
        doesn't want to bother with logging system configuration much.

    (2) Medium sized app with C extension module.
    
        Includes a few Python modules and a main script.  Employs,
        perhaps, a few logging channels.  Includes a C extension
        module which might want to make logging calls as well.

    (3) Distutils.

        A large number of Python packages/modules.  Perhaps (but not
        necessarily) a number of logging channels are used.
        Specifically needs to facilitate the controlling verbosity
        levels via simple command line options to 'setup.py'.

    (4) Large, possibly multi-language, app. E.g. Zope or (my
        experience) Komodo.

        (I don't expect this logging system to deal with any
        cross-language issues but it is something to think about.)
        Many channels are used.  Many developers involved.  People
        providing user support are possibly not the same people who
        developed the application.  Users should be able to generate
        log files (i.e. configure logging) while reproducing a bug to
        send back to developers.


Implementation

    XXX Details to follow consensus that this proposal is a good idea.


What Logging Levels?

    The following are the logging levels defined by the systems I looked at:

    - log4j: DEBUG, INFO, WARN, ERROR, FATAL
    - syslog: DEBUG, INFO, WARNING, ERROR, FATAL
    - JSR047: FINEST, FINER, FINE, CONFIG, INFO, WARNING, SEVERE
    - zLOG (used by Zope):
        TRACE=-300   -- Trace messages
        DEBUG=-200   -- Debugging messages
        BLATHER=-100 -- Somebody shut this app up.
        INFO=0       -- For things like startup and shutdown.
        PROBLEM=100  -- This isn't causing any immediate problems, but
                        deserves attention.
        WARNING=100  -- A wishy-washy alias for PROBLEM.
        ERROR=200    -- This is going to have adverse effects.
        PANIC=300    -- We're dead!
    - mx.Log:
        SYSTEM_DEBUG
        SYSTEM_INFO
        SYSTEM_UNIMPORTANT
        SYSTEM_MESSAGE
        SYSTEM_WARNING
        SYSTEM_IMPORTANT
        SYSTEM_CANCEL
        SYSTEM_ERROR
        SYSTEM_PANIC
        SYSTEM_FATAL

    The current proposal is to copy log4j.  XXX I suppose I could see
    adding zLOG's "TRACE" level, but I am not sure of the usefulness
    of others.


Static Logging Methods (as per Syslog)?

    Both zLOG and Syslog provide module-level logging functions rather
    (or in addition to) logging methods on a created Logger object.
    XXX Is this something that is deemed worth including?

    Pros:
        - It would make the simplest case shorter:

            import logging
            logging.error("Something is wrong")

          instead of

            import logging
            log = logging.getLogger("")
            log.error("Something is wrong")

    Cons:
        - It provides more than one way to do it.
        - It encourages logging without a channel name, because this
          mechanism would likely be implemented by implicitly logging
          on the root (and nameless) logger of the hierarchy.


References

    [1] java.util.logging
        http://java.sun.com/j2se/1.4/docs/guide/util/logging/

    [2] log4j: a Java logging package
        http://jakarta.apache.org/log4j/docs/index.html

    [3] Protomatter's Syslog
        http://protomatter.sourceforge.net/1.1.6/index.html
        http://protomatter.sourceforge.net/1.1.6/javadoc/com/protomatter/syslog/syslog-whitepaper.html

    [4] MAL mentions his mx.Log logging module:
        http://mail.python.org/pipermail/python-dev/2002-February/019767.html 

    [5] Jeff Bauer's Mr. Creosote
        http://starship.python.net/crew/jbauer/creosote/

Copyright

    This document has been placed in the public domain.



Local Variables:
mode: indented-text
indent-tabs-mode: nil
fill-column: 70
End:

-- 
Trent Mick
TrentM at ActiveState.com



From mdk at mdk.com  Thu Feb 14 16:19:58 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 16:19:58 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org>
Message-ID: <a4ha5f$av7o$1@ID-98166.news.dfncis.de>

"Gustavo Cordova" <gcordova at hebmex.com> wrote in message
news:mailman.1013720546.5030.python-list at python.org...
> >
> > I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> >
> > When I do a readline the I get back a string.
> >
> > How do I convert this string to a dictionary?
> >
> > Thanks.
> >
>
> You eval() it?
>
> >>> d = { "one":1, "two":2, "three":3 }
> >>> print d
> {'three': 3, 'two': 2, 'one': 1}
> >>> S = str(d)
> >>> print S
> {'three': 3, 'two': 2, 'one': 1}
> >>> dd = eval(S)
> >>> print dd
> {'one': 1, 'three': 3, 'two': 2}
> >>> d is dd
> 0
> >>>
>
>
> Do I pass??
>
> :-)
>
> -gustavo
>

I never would have figured that one out.  Even after reading the
documentation it still does not appear that the eval() function would do
that.

Thanks for the info.




From cgale1 at cox.net  Wed Feb 13 03:02:25 2002
From: cgale1 at cox.net (Van Gale)
Date: Wed, 13 Feb 2002 08:02:25 GMT
Subject: Global Variables in Modules
References: <20020211111025.63f4c956.jsc-news@bluesine.com>
Message-ID: <l4pa8.729$St3.67467@news2.east.cox.net>

My personal preference for global variables is something like the SOAPConfig
class in SOAP.py (see http://pywebsvcs.sourceforge.net/)

Van





From kragen at pobox.com  Sat Feb  2 22:05:11 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 02 Feb 2002 22:05:11 -0500
Subject: Enhanced Generators - the difference between sources and iterators
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org> <83g04kzoy2.fsf@panacea.canonical.org> <mailman.1012677065.886.python-list@python.org>
Message-ID: <837kpvz2mw.fsf@panacea.canonical.org>

Oren Tirosh <oren-py-l at hishome.net> writes:
> On Fri, Feb 01, 2002 at 07:51:01PM -0500, Kragen Sitaker wrote:
> > Oren Tirosh <oren-py-l at hishome.net> writes:
> > > My implementation of xmap, xfilter and xzip returns a source, not an
> > > iterator.  It can be iterated multiple times, just like the output of map,
> > > filter or zip.
> > 
> > That won't work when passed iterators as arguments, right?
> 
> Sure, but as long as iterators are kept as invisible temporary objects

Um, sure.  iter(open("hello")) doesn't work that way already.
Iterators are not required to be restartable, and their greatest
usefulness is when they *aren't* invisible temporary objects ---
e.g. in a parsing algorithm.

In any case, it isn't clear to me that it's more useful to be able to
say:
    mymap = xmap(f, xs)
    for ii in zs:
	for jj in mymap:
	    do_something(ii, jj)
than to have to say:
    for ii in zs:
	for jj in xmap(f, xs):
	    do_something(ii, jj)

Which is what restartable xmap etc. gives you.  Right?  I think you
can generally make nonrestartable xmap() etc. restartable (assuming
their arguments are) with a lambda:
    mymap = lambda: xmap(f, xs) # assumes nested scopes
    for ii in zs:
        for jj in mymap():
            do_something(ii, jj)
So the upside of restartable xmap() etc. is that it saves you a lambda
and a lambda call when you need it; the downside is that it gives you
a function whose semantics subtly break when it's applied to
nonrestartable iterators.

> > Having "sinks" just gives you another way to do the same thing.  
> 
> You are correct in theory, but not in practice.  In the course of a 
> dataflow it can switch only once from pull to push without the use
> of threads or queues (try it!).  I like generators because they let you 
> go much further without resorting to the use of threads.  (Thanks Guido!).
> ...
> Sure, the two forms are equivalent.  But *using* them is very different.
> One of them exposes a push-mode interface and the other exposes a pull-mode
> interface.  Try to write code that wraps one of them so that it exposes 
> the other type of interface.  You'll soon see what I mean.

That's a serious objection I hadn't thought of --- adding the "accept"
feature to the language wouldn't add any expressive power, but it
would introduce gratuitous incompatibility between code written with
one mechanism and code written with the other, which would be very
hard to bridge.  For example, you could write a general 'uniq'
transformation for iterators that would work in 'pull' mode, and you'd
basically have to cut-and-paste the code and make some changes to get
it to work with a sink in 'push' mode.  That sounds like a really good
idea not to consider adding this feature to Python!

If you want to justify its addition to the language, you need to
demonstrate some case where it adds expressive power --- which means
showing some code, even toy code, that becomes simpler or easier to
understand when you add it.



From martin at v.loewis.de  Wed Feb 27 03:50:18 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 09:50:18 +0100
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3C7C6252.FC779971@cosc.canterbury.ac.nz>
Message-ID: <m37koz2trp.fsf@mira.informatik.hu-berlin.de>

Greg Ewing <greg at cosc.canterbury.ac.nz> writes:

> Python's parser is an interesting beast. I haven't looked at
> the internals of the parser generator, so I can't be sure,
> but it seems that, although it may be using a recursive
> descent algorithm at the production level, *within* a
> production it must use some sort of regular expression 
> matching engine.

It doesn't use recursive-descent; it is a table-driven LL(1)
parser. Notice that regular expressions also define LL(1) languages,
so the entire pgen parser is a "regular expression matching engine"
:-)

> For these reasons, saying just "Python's grammar is
> LL(1)" is oversimplifying things considerably.

If you count the semantic analysis as part of the grammar, I believe
there would be few programming languages that have a context-free
grammar. For example, in Python, it is an error if both a yield
statement and a return statement occur in the same function. Try
explaining that to your friendly LR(k) parser...

Regards,
Martin


From borcis at geneva-link.ch  Sat Feb 23 10:02:54 2002
From: borcis at geneva-link.ch (Boris^2)
Date: Sat, 23 Feb 2002 16:02:54 +0100
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
References: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com> <mailman.1014024089.4625.python-list@python.org> <pan.2002.02.19.00.24.33.166041.3802@uchicago.edu>
Message-ID: <3C77AF1E.7B3B15C5@geneva-link.ch>

Ben Wolfson wrote:
> 
> On Mon, 18 Feb 2002 03:20:33 -0600, Tim Legant wrote:
> 
> > From (ancient) memory, isn't it:
> >
> > 'Twas _brillig_, and the _slithy_ _toves_
> > Did _gyre_ and _gimble_ in the _wabe_.
> > All _mimsy_ were the _borogoves_ (borogroves?)
> > And the _mome_ _raths_ _outgabe_.
> 
> Close: the last word is "outgrabe" (past tense of "outgribe", naturally).

Snicker-snack and vorpal blade :

Vorpal clearly reads as oral and verbal, now french
morphology admits a pun shaped like :

(lambda x : death to x)^4 = death by the oral blade to amoral morals.

It's found in the phrase :

"Mettons ainsi par la lame orale la morale amorale ? mort".

A very interesting phrase, this. Feels to me like Lewis Carroll
heard it first. Does anyone know how much french he mastered ?

What's-the-morality-of-a-Bandersnatch-?-ly yours,

Boris
--
beware the filter(lambda W : W not in "ILLITERATE","BULLSHIT") effect !


From JoeSalmeri at home.com  Fri Feb  1 18:37:13 2002
From: JoeSalmeri at home.com (Joe Salmeri)
Date: Fri, 01 Feb 2002 23:37:13 GMT
Subject: Python 2.2 + Win32All 146 = Python/ASP Failures
References: <vtk68.20338$8d1.7162721@news1.rdc1.md.home.com> <3C59DA37.9070506@skippinet.com.au>
Message-ID: <JEF68.22313$8d1.8213037@news1.rdc1.md.home.com>

Mark,

I grabbed framework.py from win32all-142.exe and replaced it in the
installation as you suggested.  I also deleted the framework.pyc just to
make sure the new version was compiled.  Next I restarted the web server,
just as a sanic check to make sure it saw the changes to the python asp
engine.

Repeating my tests with BROKE.ASP (I tested other asp scripts too) and still
had the exact same problem.

Hopefully this may (or may not) give you more insight into the problem.

If I can help out (testing, trying to narrow down the problem further),
please do not hesitate to ask.

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C59DA37.9070506 at skippinet.com.au...
>
> This ia a bug I am hoping to fix very soon.  If you are really keen,
> then you can grab the framework.py from win32all-142.exe - you can open
> that .exe using WinZip.
>
> Mark.
>
>




From killspam at darwinwars.com  Tue Feb 19 13:21:14 2002
From: killspam at darwinwars.com (Andrew Brown)
Date: Tue, 19 Feb 2002 18:21:14 GMT
Subject: confusion regarding os.path.walk()
References: <mailman.1014141812.1766.python-list@python.org>
Message-ID: <Xns91BABAD1C9BE8andrewdarwinwarscom@194.168.222.9>

> I think perhaps you both are somehow thinking that os.path.walk
> changes it's current directory as it 'walks' the tree: it doesn't.
> 
Tuesday, February 19, 2002, 18:02:08, you wrote:

S> The current working directory stays the same and all relative
S> paths are relative to it. When os.path.walk gives you a directory
S> and a list of files in that directory, unless that directory is
S> your current working directory, you need to either join the paths
S> or change your working directory to access the files.


Thanks for your help. What you describe is what I thought should
happen. But I couldn't figure out how to join the paths. It always
looked right: I'd do debugging runs that simply printed out the
command it was supposed to execute, and I could get results like

'ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm'

(which I think should work) but when I ran this through os.system(),
it failed every time. So did every other variant I could think of
except changing to the directory where the file was and linking
without any paths involved.



From cliechti at gmx.net  Fri Feb 15 14:19:36 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 15 Feb 2002 20:19:36 +0100
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
Message-ID: <Xns91B6CF519E265cliechtigmxnet@62.2.16.82>

Artur Skura <arturs at iidea.pl> wrote in 
news:slrna6qqs9.b35.arturs at aph.waw.pdi.net:
> I have a specific problem: I have to create dialogs on the fly, and
> in those dialog text together with text entries.

Jython 2.1 on java1.4.0 (JIT: null)
>>> import java.awt
>>> f = java.awt.Frame()
>>> f.setLayout(java.awt.FlowLayout())
>>> f.add(java.awt.Label("hello"))
>>> f.add(java.awt.TextField("world"))
>>> f.add(java.awt.Label("something like that?"))
>>> f.pack()
>>> f.show()

does this go in the right direction? the Layouters in java ATW and 
SWING make such things realy easy!

chris

-- 
Chris <cliechti at gmx.net>



From mhammond at skippinet.com.au  Mon Feb  4 07:49:27 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 12:49:27 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com> <3C5B90F3.3020503@skippinet.com.au> <CkN68.6145$F6.3643599033@newssvr14.news.prodigy.com> <3C5E34BA.6030702@skippinet.com.au> <5rt78.7926$Da3.4244083767@newssvr14.news.prodigy.com>
Message-ID: <3C5E83A7.4090404@skippinet.com.au>

Presumably _OnWMNCHitTest(self) is not the direct WndProc handler (as 
the param list is different).  Are you sure the result is getting back 
to the top-level correctly?

Other than that I am out of ideas.

Mark.

Keith Farmer wrote:
> Still isn't working -- it's printing "HTBORDER" as before, so it's
> presumably returning it as well.  Win's still changing the cursor and
> allowing resize as if it were still getting HTBOTTOM, etc.
> 
>  def _OnWMNCHitTest(self, params):
>   wParam = params[2]
>   lParam = params[3]
>   lr = win32gui.DefWindowProc(self._GetAppBarWnd().GetSafeHwnd(),
>       WM_NCHITTEST,
>       wParam,
> #      0,
>       lParam
> #      self._LowHighWordsToLparam(win32api.GetCursorPos())
>      )
>   if (lr == HTBOTTOM or lr == HTBOTTOMLEFT or lr == HTBOTTOMRIGHT):
>    print "HTBORDER"
>    return HTBORDER
>   else:
>    return lr
> 
> --
> ----------
> Keith J. Farmer
> kfarmer at thuban.org
> http://www.thuban.org
> 
> "Mark Hammond" <mhammond at skippinet.com.au> wrote in message
> news:3C5E34BA.6030702 at skippinet.com.au...
> 
> 
>>The major difference I see if the call to DefWindowProc.  You seem to be
>>passing 0 for wParam, and GetCursorPos() for lParam.
>>
>>Try passing wParam and lParam exactly as passed to you, as per the C++
>>
> code.
> 
> 
> 
> 




From kkeller at speakeasy.net  Sat Feb  2 02:04:03 2002
From: kkeller at speakeasy.net (Keith Keller)
Date: Fri, 1 Feb 2002 23:04:03 -0800
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
Message-ID: <313g3a.d22.ln@wombat.san-francisco.ca.us>

In article <LwK68.7172$Xk4.381742 at news1.east.cox.net>,
	"Andrew Replogle" <replogle2 at cox.net> writes:
>    Anyone know of any books out there for python system administration or
> should I just stick to perl for these tasks?

Way to start a holy war.  :)

Did you have something specific in mind?

--keith

-- 
kkeller at speakeasy.net
public key:  http://wombat.san-francisco.ca.us/kkeller/public_key
alt.os.linux.slackware FAQ:  http://wombat.san-francisco.ca.us/perl/fom



From aseem at neurobs.com  Wed Feb 13 11:15:32 2002
From: aseem at neurobs.com (A M)
Date: 13 Feb 2002 08:15:32 -0800
Subject: cgi and mime type
References: <mailman.1013554101.2065.python-list@python.org>
Message-ID: <ed6d1c1a.0202130815.257005e0@posting.google.com>

Hi all,
Thanks for the replies... didnt help as far as my script goes but
learnt something new... thanks again.
A


From markabarclay at mediaone.net  Mon Feb 25 18:54:28 2002
From: markabarclay at mediaone.net (Mark Barclay)
Date: Mon, 25 Feb 2002 18:54:28 -0500
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au>
Message-ID: <3C7ACEB4.B590E892@mediaone.net>

Thanks Mark. I have your book, and it is invaluable. I should have credited you
personally.

Is there a way of coding a COM server that can be running before it is accessed
by a COM client? I'm not sure if this makes sense, but I'd like to have a program
running, that can be connected to, perhaps multiple times, without it having to
create multiple instances of teh server, just have multiple clients using it.

Mark

Mark Hammond wrote:

> dsavitsk wrote:
> > just for the sake of giving credit where it is due, and not giving it where
> > it is not, Mark Hammond is responsible for the excellent COM groundwork.
> > ActiveState is responsible for laying him off.
>
> LOL - thanks for the clarification.  I was going to post something
> similar, but it would not have sounded as good coming from me! :)
>
> Mark.

--
Mark Barclay (markabarclay at mediaone.net)




From deathtospam43423 at altavista.com  Sat Feb 23 02:33:42 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 23 Feb 2002 07:33:42 GMT
Subject: help: Problem with cgi form
References: <mailman.1014420097.13372.python-list@python.org>
Message-ID: <Xns91BEB7AC37F78AusYourStandingInIt@130.133.1.4>

Gustavo Cordova <gcordova at hebmex.com> wrote in 
news:mailman.1014420097.13372.python-list at python.org:

>> 
>> I'm a beginner. How do I do this? It's probly very simple ......
>> 
>> I have a hidden input on a cgi form... 
>> 
>>     print '<INPUT TYPE=HIDDEN NAME= "last_pos" 
>> VALUE=',last_byte_pos,'>'
>> 
>> I need the VALUE to be the  variable "last_byte_pos", but 
>> unfortunately when I run the script for the first time the script
>> crashes because it's 
>> undeclared. 
>> 
>> What subroutine do I use to check if last_byte_pos is undeclared, and 
>> assign it a default value if it is? 
>> 
> 
> How can you not know if "last_byte_pos" is undeclared?
> Have you assigned any values to it previously?
> no?
> 
> then it's undeclared.
> 
> -gus
I know when it's undeclared ... whenever a user first tries to access the 
form, it is undeclared. What I want to know is how to get the script to 
check whether this value has not been declared yet.


-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From mhammond at skippinet.com.au  Fri Feb 15 20:41:23 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 16 Feb 2002 01:41:23 GMT
Subject: impossible to reboot OS in python?
References: <a4k9as$ak8$1@nntp.itservices.ubc.ca>
Message-ID: <3C6DB92F.50107@skippinet.com.au>

Kris Botha wrote:
> I heard it was impossible to induce a reboot in python as the os module does
> not cater for that. Is this true?

Yes.

> I know in DOS you could just set a hex value at the correct address to
> induce a cold or warm reboot, but this won't work for windows machines.

Correct.

> Does it depend on the os installed?

Yes.

For Windows NT/2000, do a google search - I have posted code that does 
this a number of times.

For Win9x, MSDN documents that you can execute:

rundll32.exe shell32.dll,SHExitWindowsEx n

where n is one, or a combination of, the following numbers:

0 - LOGOFF
1 - SHUTDOWN
2 - REBOOT
4 - FORCE
8 - POWEROFF

See other replies for *nix.

Mark.




From phr-n2002a at nightsong.com  Mon Feb 11 09:17:50 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 06:17:50 -0800
Subject: Question on if, while, for
References: <e959e7b.0202110545.3dff102@posting.google.com>
Message-ID: <7xd6zcw19t.fsf@ruckus.brouhaha.com>

c at tech.usd253.org (Fred) writes:
> I have written a simple program that I want to simplify. I used if and
> elif. But I am sure this can be simplified with for or while. Or what
> if I wrote a function to do the testing?
> 
> Please DO NOT re-write this for me I want to figure it out for myself.
> I would just like to know what is the BEST way to do it:
> 
> 1) For loop (nested??)
> 2) While loop
> 3) Function
> 4) Leave it alone it's fine??

5) Try to take a clearer look at the problem you're solving, rather
than getting wrapped up in the programming details.  You're trying to
reduce a fraction to lowest terms.  Never mind programming--how would
you explain that process to a math student?  I don't think you'd do it
the same way your program does it.  Try to make your program more
closely follow the methods you'd learn in a math class for reducing
fractions.  Only after you understand the math should you try to
express it as Python code.

I'll give a hint: you'll want to know how to compute the least common
denominator of two integers a and b.  You might want to look in a book
to find out how to do that.

Also (general advice): if you find yourself writing the almost same
thing over and over in a program (for example testing all those
divisors), that's a sure sign that you want to use a loop or a
subroutine.  As to deciding whether to use a for loop or while--try it
both ways and see which one you like better.  Even experienced
programmers sometimes code something two different ways to see which
one looks better on the screen.

Finally, a good book from the old days: The Elements of Programming
Style, by Kernighan and Plauger, has a bunch of examples (not in
Python) of complicated code and how to simplify it.  It's sort of
out-of-date now and I wouldn't recommend buying it at retail (it's
overpriced like many technical books) but if your local library has
it, you might like to look at it.


From lucyroman at webmail.co.za  Tue Feb  5 22:36:46 2002
From: lucyroman at webmail.co.za (lucyroman at webmail.co.za)
Date: Tue, 05 Feb 2002 19:36:46 -0800
Subject: Want To Drive Traffic & Keep Your Site Up??
Message-ID: <6rpp1h8wirlsamy.i44v3@mc5.law13.hotmail.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020205/e76acbd1/attachment.html>

From jkraska at san.rr.com  Tue Feb 26 00:07:18 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 05:07:18 GMT
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org> <m3pu2sancq.fsf@chvatal.cbbrowne.com> <eppstein-42B711.20584025022002@news.service.uci.edu>
Message-ID: <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547@4ax.com>

>>  - Pulchritudinous Python?
>>  - Busty Python?

Clever Linguistic Integration Tricks without Stacks?

C//



From peter.milliken at gtech.com  Mon Feb  4 16:38:53 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Tue, 5 Feb 2002 08:38:53 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com><Wls68.14178$om6.357057@news1.tin.it><mailman.1012590723.7301.python-list@python.org><a3kb3p$se44@news1.gtech.com> <mailman.1012843209.19773.python-list@python.org>
Message-ID: <a3n03l$tfn6@news1.gtech.com>

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
news:mailman.1012843209.19773.python-list at python.org...
> On Mon, 4 Feb 2002 08:42:58 +1100
> Peter Milliken wrote:
>
> > G'Day Cliff,
>
> Peter =)
>

[snip]

> > There are so many what I will class as "poor" programmers in the
> industry,
> > that this is a necessary technique when you have to "manage" other
> > programmers. So I assume that "continuous ongoing code review" is
similar
> to
> > my idea of "early and often"? :-)
>
> I think so.  My personal experience in working closely with another
> programmer was that we tended to write code appropriate for the other
> person to read and often discussed approaches to coding problems.  After
> discussions, prototype code was reviewed (informally) to see how our ideas
> worked out, if there were shortcomings and the like.  We never had a
> "formal" review, but since we were comfortable working with each other and
> had much mutual respect, we were able to criticize and accept criticism
> freely.
>

Hmmm.... Sounds like we have a small "disconnect" - this sounds like what I
would consider design to me - now I am in the "real" world of commercial
programming, it is becoming more and more obvious that this just doesn't
happen (design documentation I mean :-)). As a programmer for defense
contractors, you did preliminary designs and then detailed designs (with
attendant reviews) so the "coding problems" were all worked out at that
stage i.e. by the times you had sorted out design to the point of pseudo
code (or somewhere of that level) the coding became very straight forward -
and you knew it was all going to hang together! :-) (assumed that
traceability of the requirements was done correctly and no boo-boos were
made! :-)).  Perhaps I am wrong though? Anyway, it sounds like a good,
comfortable working relationship that could only help you both generate
better product.

[snip]

> Of course, the ideal programmer is the self-taught individual who later
> goes to college to raise their level of skill.
>

Whoops, got me! :-) Unfortunately I hate the discipline of "school life" - I
swore when I left Uni that I would never do it again and haven't really been
tempted since. Motification also becomes a factor in the sense that at the
moment I am in a maintenance role (only way I could get out of Defense - you
become "branded" after too many years in the one area.....), so I don't
right new code anymore, I only fix other people's screw-ups - fortunately
our stuff is all written in C so I figure I have job security for life! :-)
My company made the decision to move from Fortran to C sometime around
1995 - I killed myself laughing when I found that decision! :-) If they had
chosen differently I would probably still be stuck writing code for Combat
Systems :-) But then that was a prime example of what this thread has been
talking about - a decision that (IMO :-)) was taken with insufficient
research and thought, there were better languages that would have produced a
much more robust product in half the time (based on documented research that
was available even in those days :-)).

Hope you have a good day,
Peter






From christophe.delord at free.fr  Wed Feb 20 15:48:52 2002
From: christophe.delord at free.fr (Christophe Delord)
Date: Wed, 20 Feb 2002 21:48:52 +0100
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <3C740BB4.3070703@free.fr>

Python help says :
---
Help on built-in function sort:

sort(...)
     L.sort([cmpfunc]) -- sort *IN PLACE*; if given, cmpfunc(x, y) -> 
-1, 0, 1
---

sort returns nothing, it just sorts the list :

 >>> print [].sort()
None

Writting a function to sort lists may not be efficient :

def my_sort(L):
     L2 = L[:]
     L2.sort()
     return L2

X = [...]
for x in my_sort(X):
     ....


Christophe.

Rajarshi Guha wrote:

> Hi,
>   I have a dictionary and would like to loop over the elements in sorted 
> order.
> The dictionary is of the form:
> 
> {'1':'option1', '2':'option2', '3':'option3'}
> 
> However when I use:
> 
> for i in d.keys().sort():
>   do something
> 
> I get:
> 
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: loop over non-sequence
> 
> But I know that d.keys() returns a list to which the sort() function should 
> be applicable.
> 
> When I do:
> 
> k = d.keys()
> k.sort()
> for i in k:
>   do something
> 
> it works fine.
> Why can't I use d.keys().sort() directly? And is there any other more 
> concise (elegent?) way to loop over sorted dictionary keys?
> 
> TIA,
> 
> 



From kdahlhaus at yahoo.com  Wed Feb 20 18:51:24 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 20 Feb 2002 15:51:24 -0800
Subject: Hosting
Message-ID: <283adf56.0202201551.7026d548@posting.google.com>

Does anyone know of an inexpensive web host that allows compilation
and use of Python extensions from  your own directory?


From alex at netWindows.org  Wed Feb 27 16:34:11 2002
From: alex at netWindows.org (Alex Russell)
Date: Wed, 27 Feb 2002 16:34:11 -0500
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net>
Message-ID: <20020227163411.71b4b4d8.alex@netWindows.org>

I'm glad they (apparently) got one right.

I was required to purchase their Java book for a class, and not only was
it poorly laid out, poorly executed, badly typeset, and cheaply made, it's
index was utterly useless. Their C++ book wasn't much better. A friend had
it laying about his place day and I spent some time looking through it,
and after comparing it to Lippman/Lajoie, it's clear that it makes better
tinder than a programming reference.

Stear clear of those two if you're thinking about picking them up on the
strength of their Python book.

-- 
Alex Russell
http://netWindows.org
http://alex.netWindows.org

On Thu, 28 Feb 2002 01:51:45 GMT
Ron Stephens <rdsteph at earthlink.net> wrote:

> Sorry but I can't wait to give a brief "preview review" of the new
> Deitel and Deitel book, "Python How to Program" (I also posted this at
> my website at http://www.awaretek.com/plf.html ).
> ...
> Well, let's see, I bought this 1292 page book from Amazon a week ago for
> $72. This is the most I have ever paid for a computer book. Heck, it may
> well be the most I have ever paid for *any* book ;-))).
> 
> I am leaving on a long business trip to Asia in two days, and I just
> can't wait until I get back to tell you how much I am enjoying this
> book. I have read the first 340 pages, which cover the core language; I
> intend to carefully read the remaining chapters, which cover various
> more advanced topics, as I fly around the world from 
> New York City to Amsterdam, to Singapore, to Taiwan, to Japan, to Korea,
> on to Shanghai, China, and finally back home to New York. I am looking
> forward to the book more than the trip ;-))))
> 
> This book is one of the most enjoyable I have ever read. The pedagogical
> approach that the authors use is superb; they have obviously learned
> their educational craft well; for me, it actually works. The quality of
> the paper is the finest in any Python book yet; the text is large and
> clear. The way actual code is highlighted and separated from the text is
> extremely 
> helpful. Everything is first class.
> 
> I actually sort of expected the book to be a second rate effort in the
> sense that the Deitels have written several best sellers about other
> languages, and I sort of expected them to half-heartedly do a me-too
> book by just applying the same old formula to a Python version.
> 
> But the book is fresh and clearly thought out. Since I have never read
> any other Deitel book, it is possible that those who have done so may
> have a different experience.
> 
> But I think even Deitel veterans will be pleased. More than anything
> else, the book reminds me of a great college textbook, and reading such
> a text is something I haven't done in way too many years. But this
> educational quality of the text brings it to a whole new level of
> sophistication, compared to most computer programming books.
> 
> The book includes, at the end of each chapter,  extensive questions and
> answers, review lists, and summaries. It actually feels like you are
> studying for credit in a college course. The end result is that, after
> carefully reading the chapters covering the core language, I feel like I
> have consolidated my Python knowledge considerably and that I am both
> ready to do some really serious coding (once I get back from Asia ;-)))
> and also to successfully attack the advanced chapters while I am on the
> planes.
> 
> In addition to excellent coverage of the core language, the book covers
> CGI, XML, Databases, XHTML (quite extensive) , Data Structures,
> Multimedia, OpenGL, Tkinter, CSS, Python Server, Networking, Threading,
> Regular Expressions, and Unicode. There are even excellent and extensive
> appendixes reviewing Octal, Hexadecimal, and Binary Number Systems;
> HTML; and Python version 2.2 Additions. Most of these topics are not
> just skimmed over but are covered in great detail. yet the book really
> flows well. The exercises and questions and answers are equal to the
> very best college text books  I have encountered. Even the summaries
> after each chapter are useful; I think I will be reviewing the chapter
> summaries on the plane and later again after a few weeks, in order to
> consolidate my learning even more. It seem like a good use of my time,
> because the summaries are well thought out and very helpful.
> 
> I am excited.
> 
> I will do a detailed and very thorough review of the whole book as soon
> as I return home from my trip. But even based on my reading of only the
> first 340 pages, this book is a winner and I think any 
> Pythonista who is willing to part with the (admittedly huge) sum of $72,
> will be richly rewarded. One could easily get engrossed in this fine
> book and wind up absorbing all 1292 pages in 10 days of blissful mind
> mongering. (Well, what the hell, its only a preview review and I can
> coin an asinine phrase like "mind mongering " if I want to ;-))))
> 
> ...
> By the way, I have studied the code of Hans Nowak's mygale.py well
> enough to be confident in making modifications and additions, and I am
> really looking forward to doing that as soon as I return  home. I intend
> to add more starting point sources for the article searches, to add the
> capability to search for any given topic, and also trying to fine tune
> the searching mechanisms. I even have some other ideas for the code, but
> I'll keep them quiet for now. Quite basic coding of course, the real
> work has already been done by Hans and I really appreciate his
> open-sourcing of this fine program.
> 
> I'm still searching for more tutorials and I'm up to 24 links now. I
> have a lot of other stuff I want to add when I get back in mid March. 
> 
> Ron Stephens
> < a href = "http://www.awaretek.com/plf.html">Python City</a>



From sab at NOSPAM.freeuk.com  Tue Feb 12 17:22:22 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Tue, 12 Feb 2002 22:22:22 -0000
Subject: Is it possible to call a function from a Tkinter widget.bind and include args???
References: <a4995n$gbg$1@news5.svr.pol.co.uk>
Message-ID: <a4c4fs$d0q$1@news5.svr.pol.co.uk>

Thankyou all, your comments are very helpful!   :)   Especially the
GenericCallback class!

G. Willoughby


"G. Willoughby" <sab at NOSPAM.freeuk.com> wrote in message
news:a4995n$gbg$1 at news5.svr.pol.co.uk...
> Is it possible to call a function from a Tkinter widget.bind and include
> args???
> i.e. i want to do something like this:
>
> def printButtonAndCoords(event, button):
>     print "you rolled over %s and coords are %d:%d" % button, event.x,
> eventy
>
> Button1.bind("<Enter>", printButtonAndCoords("button1"))
> Button2.bind("<Enter>", printButtonAndCoords("button2"))
>
> is this possible???
>
> i have tried this: but it seems a bit crap!:
>
> [snip]
> ...
>   self.newFileButton.bind("<Enter>", self.getMouseCoords)
>   self.newFileButton.bind("<Leave>", self.deleteToolTip)
> ...
>  def getMouseCoords(self, event):
>   self.xPos=event.x_root-5
>   self.yPos=event.y_root+5
>   self.showToolTip(str(event.widget), self.xPos, self.yPos)
>
>  def showToolTip(self, id, xPos, yPos):
>   if id==".8387596.7951612": #this looks very breakable!
>    self.toolTip=tkToolTip.toolTip(xPos, yPos, liftAbove=self.master,
> text="Create a new file")
>
>  def deleteToolTip(self, event):
>   self.toolTip.destroyToolTip()
>   del self.toolTip
> [/snip]
>
> help...  :'-(
>
> G. Willoughby
>
>
>
>




From jeff at ccvcorp.com  Fri Feb  8 14:56:01 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 11:56:01 -0800
Subject: High-level languages, large projects, GNOME and the .NET controversy
References: <23891c90.0202070501.78df2371@posting.google.com> <3C63A0A7.5060507@thomas-guettler.de> <91acf731.0202080713.652fa392@posting.google.com>
Message-ID: <3C642D51.FC27092C@ccvcorp.com>


Jason Voegele wrote:

> Thomas Guettler <st-newsgroups at thomas-guettler.de> wrote in message
>
> > Would be nice to have the intellisence (dropdown-list if you type
> > object.) of the Visual Studio. Is there already something like this?
>
> This is really hard to do with dynamic languages because the editor
> has no reliable way of knowing the type of a variable.

Quite a few IDE's do this, including PythonWin and PyCrust.  Of course, (at least for PythonWin), it's a
text-matching search.  When you type 'name.', it will present a list of all examples of what 'name' has already been
modified with, in either the text of the current editing window, or any imported modules.  So, if I'm writing several
classes in a single file, and I type 'self.', it'll provide a list that may contain all the attributes of all the
classes in the file...  but if I'm using a newly defined attribute, that won't show up at all.  (Nor will instance
names list attributes on the class, except where they've been used with that name before.)  I wouldn't be at all
surprised if some of the various commercial IDE's do a better job of this...

But I do believe that this *is* the job of the IDE, not the interpreter itself.  The overhead of tracking
code-completion and call-tips, etc, is not needed when simply running scripts; I'd rather keep the coding-help stuff
separate from the running-programs stuff as much as is practical.

Jeff Shannon
Technician/Programmer
Credit International




From ivano.delpiccolo at degroof.be  Wed Feb 20 06:37:03 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Wed, 20 Feb 2002 12:37:03 +0100
Subject: sio-151
Message-ID: <3c738a50$0$12224$4d4efb8e@news.be.uu.net>

I'm searching an example of use of the windows serial module sio-151
If somebody have a simple example, I will be very happy

Thanks




From danyelf at acm.org  Wed Feb 20 16:45:40 2002
From: danyelf at acm.org (Danyel Fisher)
Date: Wed, 20 Feb 2002 13:45:40 -0800
Subject: preference tree like in jdk 1.4
References: <mailman.1013712805.16911.python-list@python.org> <3C6C1761.6BB16ADA@ndh.net>
Message-ID: <a515fe$su9$1@news.service.uci.edu>

I think the container plan is on the right track.
It is fairly clear how to use setattr to do things like this.

[sketchy, only lightly tested...]

class Container:
    attr_table = {}

    def __getattr__( self, arg ):
        if not self.attr_table.has_key( arg ):
            self.attr_table[ arg ] = Container()
        return self.attr_table[ arg ]

    def __setattr__( self, arg, val ) :
        self.attr_table[ arg ] = val

pref = Container()
pref.personaldata.firstname = 'John'
pref.personaldata.lastname = 'Cleese'
pref.windowsettings.windowA.x = 200

print pref.personaldata.firstname

There's a few disadvantages to this: typos, for example, would become
canonical parts of objects, so perhaps the explicit [if ugly]
pref.personaldata = Container()
is necessary.

Because I really want to say
    pref.personaldata = None
and I don't want to elimintae the possibility of saying
    pref.personaldata = 23
    pref.personaldata.detailed = 234

Danyel

"Stefan Schwarzer" <s.schwarzer at ndh.net> wrote in message
news:3C6C1761.6BB16ADA at ndh.net...
> Hello Michael
>
> Michael Chermside wrote:
> > Instead of using an empty function, use something designed specifically
> > for holding fields: a class!
> >
> >  >>> class Empty:
> >         pass
> >  >>> pref = Empty()
> >  >>> pref.personalData = Empty()
> >  >>> pref.personalData.firstName = 'John'
> >  >>> pref.personalData.lastName = 'Cleese'
> >  >>> pref.windowSettings = Empty()
> >  >>> pref.windowSettings.windowA = Empty()
> >  >>> pref.windowSettings.windowA.x = 200
>
> Actually, you use instances of the class to store the values. I think,
> you meant that anyway, didn't you? :-)
>
> Sometimes classes are used for such a purpose. However, this may cause
> problems if you need more than one "container" (like above). Using
> class instances is more common, I assume.
>
> Stefan




From marklists at mceahern.com  Mon Feb 11 07:25:22 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Mon, 11 Feb 2002 04:25:22 -0800
Subject: Beginner problem.
In-Reply-To: <3c67a6a3$0$12223$4d4efb8e@news.be.uu.net>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOECDKCAA.marklists@mceahern.com>

[DelPiccolo Ivano]
> How can i know if i have parameters in the command line.
> 
> Here's my script, where is the error ?
> 
> import sys
> 
> if sys.argv[??????] == "" :
>     print "Usage : sys.argv[0] logname string_to_search"
> else :
>    ...........
>     ..........
>     ..........

This tells you how many arguments you have:

	len(sys.argv)

Here's a sample that should get you started:

    #! /usr/bin/env python
    # junk.py

    import sys

    path = sys.argv[0]

    if len(sys.argv) == 1:
        print "No arguments were specified for %s." % path
    else:
        for arg in sys.argv:
            print arg

Running it...

  $ junk.py
  No arguments were specified for ./junk.py.
  $ junk.py foo bar
  ./junk.py
  foo
  bar

Cheers,

// mark



From mcherm at destiny.com  Fri Feb  1 15:49:25 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Fri, 01 Feb 2002 15:49:25 -0500
Subject: Code Reviews (was: Defending the Python lanuage...)
Message-ID: <3C5AFF55.2000305@destiny.com>

> Actually, if you are *good* at what you do, then code review, whilst still
> beneficial, is not nearly *that* beneficial. Things like code review (IMO
> :-)) is there for the 25-50% of programmers who just plain shouldn't be in
> the industry - what they generate is just absolute garbage - code review
> will hopefully stop the code from getting into Unit test and beyond the
> point of redemption :-).


I completely disagree. There are lots of reasons why code reviews are 
good for EVERYONE, not just the incompetents, but I'll just list a few:


  * Code reviews spread knowledge of the code around.

You don't have TIME to sit down and read every piece of code that anyone 
at your company writes, but on the other hand, if only the author has 
ever worked with the particular piece of code which she wrote, then what 
happens when she's hit by a bus? Code reviews ensure that SEVERAL people 
are intimately familiar with the design of each piece of code in the 
system. Perhaps she wrote some utility method which you could leverage 
in a different part of the code... without spreading the knowledge 
around you wouldn't know about it.

  * Code reviews are a chance to learn.

Sure, you're an experienced, skilled developer. You know all kinds of 
tricks and patterns and can sling code almost as fast as you can type. 
But the ONLY ways to get like that are (1) fooling yourself, or (2) 
continuous learning. A good programmer is always trying to pick up on a 
new idiom, an as-yet-unfamiliar pattern, or just a good hint on creating 
a readable style. Reading books like "The Pragmatic Programmer" is a 
good start. Attending code reviews can be just as good. It's in that 
code review where you're sitting down with other skilled programmers and 
discussing the quality/style/details of a particular bit of code. If 
you're the local guru, then you're probably mentoring the newer 
programmers in these sessions (good in and of itself!), but even if 
you're the most experienced and skilled developer in the group, you're 
still going to learn some new things from the process (unless you're (1) 
fooling yourself).

  * Code reviews give you a chance to show off.

Everyone likes to write really GOOD code. Something that's clever yet 
readable, highly optimized yet nicely organized, using surprisingly 
appropriate algorithms and perfectly chosen data structures. But it's 
even MORE fun if you can show it off (to other programmers, of course, 
as they're the only ones who can really appreciate it). And code reviews 
provide this opportunity. If you are handed 20 pages of code and after 
preparing for the review by reading it over (you DO read the code ahead 
of time don't you <poke-in-the-ribs>), you are just downright impressed 
-- it's readable and brilliant -- then tell the author how good it is. 
They worked hard to achieve that, and deserve a little recognition.


Sorry for going on about this, but I'm the local self-appointed 
code-review advocate.

-- Michael Chermside

PS: For those XP'ers out there, I include pair programming as a limit 
case of code reviews.






From jdavis at empires.org  Wed Feb 20 23:39:28 2002
From: jdavis at empires.org (Jeff Davis)
Date: Wed, 20 Feb 2002 20:39:28 -0800
Subject: Reloading modules
Message-ID: <a51tt1$msq$1@news1.ucsd.edu>

Is there a way to say that all calls to import should check for a newer 
version, if a newer one exists then completely drop the old one and import 
the new one?

I am aware of the reload() function, but it has enough caveats that it 
always leaves lingering questions. I usually end up just stopping and 
restarting the interpreter, which slows development/debugging. I would 
just like to be sure that any time after I import a module, I am using the 
newest version available at the time of import. I am also unsure as to 
whether modules imported within a reloaded module are also reloaded.

Thanks,
        Jeff


From jkraska at san.rr.com  Mon Feb 25 21:44:19 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 02:44:19 GMT
Subject: DLL/Embeded interpreter wierdness
Message-ID: <qltl7uoh7q01drhb37b48bsmipr7h8kaam@4ax.com>

So. Let me see if I can phrase this right:

If I have an embedded interpreter sitting in a .dll and this
self-same dll wants to be a module for import in scripts
executed by the embedded interpreter, will this work?

I have a .dll called "lramm.dll". Before I embedded an
interpreter in it, it could be loaded (from basic python)
just fine (in otherwords, it has an init, has been tested).

I started some redisign work and embedded the interpreter
in the .dll and then executed a script (with PyRun_SimpleFile())
which contains as its first statement:

"import lramm"

The import fails.

Let me repeat myself. It's lramm.dll _itself_ which has
the "PyRun_SimpleFile()" command in it.

I've checked and rechecked: the .dll is _definitely_ in
the PYTHONPATH.

My tentative conclusion is that if the .dll is already
loaded as a consequence of dynamic linkage with main, then
Python can't "twice load it" through dlload() or whatever
routine it's using. Is that correct, or do I have another
problem?

C//



From david.abrahams at rcn.com  Mon Feb 25 19:32:42 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Mon, 25 Feb 2002 19:32:42 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu>
Message-ID: <a5ekpq$n83$1@bob.news.rcn.net>

I've been interested in systems for automatically wrapping C++ and
generating documentation for a while, so I downloaded the PDT (the parser
toolkit used by SILOON) and tried it on one of my source files. After lots
of hacking, I was able to get the EDG front-end component to parse my source
and generate an intermediate language (.cil) file, but the PDT executable
which is supposed to interpret the .cil and produce a high-level description
of the code crashed with an assertion:

Assertion failed: ptr->assoc_template, file c:\documents and
settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-tau\il_display_tau.
c, line 3047

Since the PDT doesn't come with source, there was no way for me to debug/fix
the problem.
Furthermore, the PDT developers don't seem to want to be reached. Their
contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/) says:

    To contact the developers of PDT, please use the "Comments" section in
the download page.

But of course, there is no "Comments" section on the download page. :=(

-Dave

"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
news:3C7A751D.1080703 at physics.ucsb.edu...
> Hi Craig.
>
> Since your site specifically lists g++ 2.95 and NOT 3.x, am I to infer
> that there will be problems with the new ABI?  -- Or any problems that
> will, in practice or principle, cause conflicts with g++ 3.0?
>
> *Holds breath... and crosses fingers...*
>
> --Craig
>
> rasmussn at lanl.gov wrote:
>
> > On Wednesday, February 20, 2002, at 04:48 PM, David Abrahams wrote:
> >
> >>> As I understand it, the SILOON project uses a commercial parser known
as
> >>> PDT from uoregon.  Have you, or has anyone else had any experience
with
> >>> the SILOON/PDT project?  On the face of it, my guess would be that you
> >>> would be somewhat skeptical of their claims?
> >>
> >>
> >> I haven't had any experience with it. I just found that site. My
> >> skepticism
> >> just dropped by a factor of 8. If they are really using the EDG
> >> front-end as
> >> they claim to, then I am sure the parser is excellent. I do have to
> >> wonder
> >> who is paying for the FE. The issues of dealing with the semantics of
> >> what
> >> they've parsed remain, but these guys at least seem to be doing the
> >> parsing
> >> part right.
> >
> >
> > It is my understanding that an educational or research license to the
EDG
> > front end is fairly easy to obtain.  The University of Oregon has
obtained
> > the right to distribute the EDG front-end as binaries and does so for
> > several
> > platforms in the Program Database Toolkit (PDT).  So for anyone wanting
> > interface level information from C++ or Fortran 90 source files, the PDT
> > is an excellent tool.  We use it in several projects.
> >
> > See http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/ and
> > http://www.acl.lanl.gov/siloon/ for more information.
> >
> > Craig
> >
> >
>
>




From stephen at xemacs.org  Thu Feb 28 06:39:06 2002
From: stephen at xemacs.org (Stephen J. Turnbull)
Date: 28 Feb 2002 20:39:06 +0900
Subject: PEP 263 comments
References: <mailman.1014762060.30054.python-list@python.org> <m3u1s31eah.fsf@mira.informatik.hu-berlin.de>
Message-ID: <87g03lhm3p.fsf@tleepslib.sk.tsukuba.ac.jp>

Hi, I'm Steve Turnbull, I do XEmacs.  Mostly Mule.  Barry asked me to
step up to bat on this.

Background: I've been doing Japanese Emacs ("nemacs") and Mule for 12
years now (it's where I live...).  I've been watching Japanese open
source both as a wannabe hacker and a social scientist (that's my day
job) for about 10 years.

>>>>> "Martin" == Martin v Loewis <martin at v.loewis.de> writes:

    Martin> "Jason Orendorff" <jason at jorendorff.com> writes:

    >> Counter-proposal:
    >> - Comment syntax: none.
    >> - UTF-8 file signature: not supported.
    >> - Python source code encoding: must always be UTF-8.
    >> - Implementation: within the parser, everything's just
    >> ordinary UTF-8 bytes.
    >> - IDLE: always save UTF-8 unless otherwise directed.

    Martin> Do you seriously want to pursue this route?

Yes, I think you do.  I've watched Japanese patches to apps languish,
never to be merged, for 2, 5, 10 years.  I'm sure some go back farther
than that.  All because the Japanese want to use their favorite
encodings internally and in sources, none of which (except
ISO-2022-JP) bother to announce that they are Japanese in any way.

You're really not doing anyone a favor by supporting "my favorite
encoding" anymore.  Make EUC and ISO 2022 users, Shift JIS and Big 5
abusers, Windows 125x <l-word deleted> check their weapons at the
door: as soon as you get inside Python, it's all Unicode.

    Martin> If so, how do you want to deal with backwards
    Martin> compatibility?

You don't.  From now on, anything that goes into the official Python
sources is in UTF-8.  Convert any existing stuff at your leisure.
This is recommended practice for 3rd party projects, too.  People can
do want they want with their own stuff, but they are on notice that if
it screws up it's their problem.

XEmacs actually did this (half-way) three years ago.  I convinced
Steve Baur to convert everything in the XEmacs CVS repository that
wasn't ISO 8859/1 to ISO-2022-JP (basically, start in ASCII, all other
character sets must designate to G0 or G1, and get invoked to GL; at
newlines, return to ASCII by designation; the "JP" part is really a
misnomer, it's fully multilingual).  Presto! no more accidental Mule
corruption in the repository.

NB: UTF-8 is much more tractable than ISO-2022-JP, precisely because
of the ASCII 0x22 issue.  We've had problems with that (since we still
support non-Mule XEmacsen that don't understand ISO 2022 controls).
UTF-8 makes this a non-issue.

We left a lot of stuff that was ISO 8859/1 as is.  But this is a bad
idea post-Euro, and causes the occasional embarrassment as we pick up
more Latin-N, N != 1, users.  The Euro by itself wouldn't be a
problem, nobody uses the generic currency symbol except as a bullet in
lists.  But (and again this is based on my Japanese experience) it's
the other characters in Latin-9 that are the most important characters
in the world---to those who use them: they're part of their names.
But people do occasionally use the accents in composing characters and
libraries, so deprecating Latin-1 in favor of Latin-9 probably is
going to annoy a few people who have always followed the rules.  So
make a clean sweep, now.  In two years, you'll have no regrets.

Oh, and does Python have message catalogs and stuff like that?  Do you
really want people doing multilingual work like translation mucking
about with random coding systems and error-prone coding cookies?
UTF-8 detection is much easier than detecting that an iso-8859-1
cookie should really be iso-8859-15 (a reverse Turing test).

    Martin> Currently, you can put arbitrary bytes in character
    Martin> strings, and people make use of this opportunity (even

I have no sympathy for self-inflicted injuries anymore.  The amount of
effort that has gone into maintaining Japanese patches for Pine and
ghostscript has been extremely painful to watch.  Anyway, Python
itself provides the necessary tools for salvation.

    Martin> though the documentation says this is undefined).

So much for the alleged "backward compatibility" non-issue.  :-)
People are abusing implementation dependencies; Just Say No.

    Martin> Will you reject a source module just because it contains a
    Martin> latin-1 comment?

That depends.  Somebody is going to run it through the converter; it's
just a question of whether it's me, or the submitter.  In the case of
XEmacs, because everybody uses Emacs to develop, it's just not an
issue: somebody commits the change from (eg) EUC-JP to ISO-2022-JP,
and after that Mule does its thing---nobody even notices, unless they
do a diff.  Even at the time we got very few complaints about spurious
diffs.  Now, never.

This is true for Python, too.  I don't care if people want to do their
editing in ISCII or KOI8-R or Windows-1252 even.  _They have the tool
needed to convert, by definition: they're Python users._  Here's where
cookies come in.

GNU Emacs supports your coding system cookies.  XEmacs currently
doesn't, but we will, I already figured out what the change is and
told Barry OK.  And I plan to add cookie-checking to my latin-unity
package (which undoes the Mule screwage that says Latin-1 NO-BREAK
SPACE != Latin-2 NO-BREAK SPACE).  Other editors can do something
similar.

So people who insist on using a national coded character set in their
editor use cookies.  Then the python-dev crew prepares a couple of
trivial scripts which munge sources from UTF-8 to national codeset +
cookie, and back (note you have to strip the cookie on the way back),
for the sake of people whose editor's Python-mode doesn't grok cookies.

I expect that with that kind of support, what is left is just enough
pain to induce lots of people to switch to UTF-8-capable editors, and
just little enough that you can say "well, this really is for
everybody's benefit; we know it's inconvenient in the transition, and
we're doing the best we can to ease it" to the rest, and not be lynched.

This disposes of the "not everyone has a UTF-8 editor" issue.  Also,
in my experience distinguishing UTF-8 from "all other coding systems"
is hardly error-prone at all, except for files with extremely low
non-ASCII content, like under 50 bytes of non-ASCII.  Ben Wing has
already implemented statistical detection (ie, returns a degree of
likelihood, and could -- not implemented yet -- look at statistical
properties of the text) for XEmacs 22.  I imagine I could persuade him
to donate code to Python.

My apologies for the flood.  I've been thinking about exactly this
kind of transition for XEmacs for about 5 years now, this compresses
all of that into a few dozen lines....

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
              Don't ask how you can "do" free software business;
              ask what your business can "do for" free software.


From jkraska at san.rr.com  Tue Feb 19 19:10:44 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 20 Feb 2002 00:10:44 GMT
Subject: Adding static typing to Python
References: <fjs47u0n8bscvgtme1vc7j6jajlt4og2n3@4ax.com> <mailman.1014138876.23048.python-list@python.org>
Message-ID: <3rq57u0aq29nircfbcleugo5lcbs456e7m@4ax.com>

>I agree, for your example; but a common use of typedefs now is
>to create an alias for a type whose name is just too long to be
>allowed.
>
>  typedef list<pair<string, bool> >::iterator ItemIter;

Surely. A different problem. This might suggest that along with
"typedef", one might have a "pseudonym" keyword or what have you.

C//



From steve at ferg.org  Wed Feb 13 13:21:26 2002
From: steve at ferg.org (Stephen Ferg)
Date: 13 Feb 2002 10:21:26 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
Message-ID: <b16e4ef7.0202131021.2b105a66@posting.google.com>

I'm working on something that I call EasyGui.  It is still very much
in the roughing-out stage, and I wouldn't announce it except for the
fact that I've seen a couple of questions posted, to which EasyGui
might be the answer.

Here is a bit of a write-up...

subject : EASYGUI -- Simple GUI functions for Python
version : 0.1  2002-02-01
author  : Stephen Ferg (steve at ferg.org)
location: http://www.ferg.org/easgui/index.html


INTRODUCTION
==============================================================

It is important for Python to support GUI operations that are very
simple and easy to use.  Experienced Pythonistas need support for
quick and dirty GUI features. New Python programmers need GUI
capabilities that don't require any knowledge of tkinter, frames,
widgets, callbacks or lambda.

In short, Python should make sending a message to the screen as simple
as:

	msgbox("Hello, world!")
	
This is what EasyGui provides.  Using EasyGui, all GUI interactions
are invoked by simple function calls.

The purpose of EasyGui is to provide a small set of very simple
functions that can support basic GUI programming needs.  Use it until
you can acquire the skill or the time to use something more powerful. 
When EasyGui runs out of steam for you, then you can move up to
tkinter, wxWindows, etc.

I think that if Python is really going to support CP4E (Computer
Programming for Everybody), then something like EasyGui should become
part of the standard distribution.  But for now, the first order of
business is to polish the rough edges.

EasyGui was developed by me, Steve Ferg.  I hope you will find it
useful.

I'm still a newbie Pythonista myself, so I welcome suggestions for
improvement -- especially if they come in the form of working code
snippets!

-- Steve Ferg (Steve at Ferg.org)


From yhatakka at dlc.nospam.fi  Wed Feb 20 01:24:53 2002
From: yhatakka at dlc.nospam.fi (=?iso-8859-1?Q?Yrj=F6?= Hatakka)
Date: Wed, 20 Feb 2002 08:24:53 +0200
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com> <7xu1scew8m.fsf@ruckus.brouhaha.com>
Message-ID: <slrna76g9l.rag.yhatakka@kenny.evolvit.fi>

On 19 Feb 2002 18:13:29 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:
> "Mike Carifio" <carifio.nospam at nospam.usys.com> writes:
>> 
>> def traverse(list):
>>     if list:
>>         print list[0]  # or do something here
>>         traverse(list[1:])
>>     else:
>>         print "done"
>> 
> 
> That's ok in Scheme but I don't think any current Python implementations
> handle tail recursion "correctly".

What do you mean? I tried above an it seems to work. When does it stop
working?

I used Python 2.2

yrj?


From sandskyfly at hotmail.com  Wed Feb 20 15:28:56 2002
From: sandskyfly at hotmail.com (Sandy Norton)
Date: 20 Feb 2002 12:28:56 -0800
Subject: threads and urllib
Message-ID: <b03e80d.0202201228.76ac8525@posting.google.com>

This simple little program retrieves web pages using threads and
urllib. Now it seems to work fine with many websites, however, more
often than not it just hangs there and the thread (or is it the
socket?) does not time out and die.

Is there anyway to specify a time limit in which the operation can
happen and enforce it in the code?

Any help will be much appreciated.

cheers,

Sandy

--------snip-----------------------------------------------------------

import urllib
import threading
import time

t1 = time.time()

class SerialAgent:
    def run(self):
        self.webpage = urllib.urlopen(self.url).read()
        #print self.webpage

class Agent(SerialAgent, threading.Thread):
    def __init__(self, url):
        self.url = url
        threading.Thread.__init__(self, name=url)
        self.webpage = ''

alist  = []

for url in ['http://www.python.org', 'http://www.zope.org']:
    a = Agent(url)
    print a
    a.start()
    print a
    alist.append(a)

for a in alist:
    a.join(20.0)
    print a, len(a.webpage)


From jnix at cafeconleche.de  Sun Feb 17 20:18:17 2002
From: jnix at cafeconleche.de (Johannes Nix)
Date: 18 Feb 2002 02:18:17 +0100
Subject: strange error with dynamic loading of sound extension - dlopen flags
References: <67heown8yc.fsf@zosma.lokal>
Message-ID: <67ofinpoyu.fsf@zosma.lokal>

The cause of the problem is indeed the environment of dlopen(),
and the solution is as following (thanks to Kai Vehmanen):

-------------
import sys
import DLFCN                            # has to go before ALSA module
sys.setdlopenflags(DLFCN.RTLD_LAZY      # has to go before ALSA module
                   |DLFCN.RTLD_GLOBAL)  # has to go before ALSA module
from pcm import *                       # ALSA module

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

Johannes


From wurmy at earthlink.net  Sun Feb 17 20:59:34 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Mon, 18 Feb 2002 01:59:34 GMT
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no>
Message-ID: <3C706035.6AC11606@earthlink.net>

Magnus Lie Hetland wrote:
> 
> In article <8867.813T2365T9245607threeseas at earthlink.net>, Timothy Rue wrote:
> [...]
> >Certainly everyone does understand in reading and responding to post here,
> >they actually make use of all nine action. It's physics, but they may
> >think it's me and that somehow their reactions will cause a dismissal
> >of.... well physics.
> >
> >Lets see now:
> [...]
> 
> No, no! This isn't simple enough. The *real* truth (and it's physics,
> so you can't contradict me) is that there is only one true action:
> 
> DS - Do Stuff
> 
> Now deal with it and go somewhere else with your false theory.

Do you also have a model with zero actions? This one seems kind of
difficult to me.

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From tim.one at home.com  Sat Feb 16 19:48:40 2002
From: tim.one at home.com (Tim Peters)
Date: Sat, 16 Feb 2002 19:48:40 -0500
Subject: Guido wins FSF Award
In-Reply-To: <7xit8xxecr.fsf@ruckus.brouhaha.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEPJNMAA.tim.one@home.com>

[Paul Rubin]
> Congratulations to Guido for winning the fourth Free Software Foundation
> Award, for inventing and implementing Python as Free Software:
>
>   http://www.fsf.org/press/2002-02-16-FSF-Award.html

Hooray for Guido!  Shows how much the world cares about secure pickles
<wink>.

> And to think I learned of this from a Timothy Rue post.  Sigh.

Not at all.  Guido's acceptance speech was a foundational masterpiece,
revealing the Nine Pythonic Actions underlying all that we do.

> Anyway, hats off to Guido!

The Dutch temperament likely prefers cash donations, or at least tulips.

the-march-to-world-domination-continues-as-planned-ly y'rs  - tim




From netzapper at magicstar.net  Wed Feb 27 23:19:53 2002
From: netzapper at magicstar.net (A. Jones)
Date: Thu, 28 Feb 2002 04:19:53 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <7x4rk244zt.fsf@ruckus.brouhaha.com>
Message-ID: <3c7daf73.254541606@mammoth.usenet-access.com>

On 27 Feb 2002 20:14:46 -0800, Paul Rubin <phr-n2002a at nightsong.com>
wrote:

>netzapper at magicstar.net (A. Jones) writes:
>> It's my second day Pythoning and I've written this code to encypher
>> text (or files, ostensibly) into two pieces, neither of which can be
>> used to determine the message.
>> 
>> Any problems pointed out, clunky lines spotted, or security problems
>> would be most appreciated.  Also, if you've seen something like this,
>> if you'd tell me, please.
>
>What the heck?  There's no encryption function in there anywhere.
>
>If you want convenient low-to-medium security encryption, use the
>rotor module already supplied Python.
>
>If you want higher security, use one of the various crypto modules
>available through Vaults of Parnassus or elsewhere.

I didn't claim it was encryption, only encyphering.  But, you've
answered my major question as to whether or not something similar
existed.

Thank you.
______________________________________________________________________________
Posted Via Binaries.net = SPEED+RETENTION+COMPLETION = http://www.binaries.net


From joshm at taconic.net  Mon Feb 11 20:21:25 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 11 Feb 2002 20:21:25 -0500
Subject: Unwieldy Python considered harmful.
Message-ID: <3c686bc9_4@corp.newsgroups.com>

I can't say it enough.  Refactor, refactor, refactor.

http://www.cnn.com/2002/US/02/11/killer.snake.ap/index.html

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From jbauer at rubic.com  Fri Feb 15 10:02:07 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 15 Feb 2002 09:02:07 -0600
Subject: PythonCE mailing list has moved
Message-ID: <3C6D22EF.A9679B9C@rubic.com>

The PythonCE mailing list was originally
started a few years ago on egroups.com,
which was subsequently acquired by Yahoo.
Since then several list members have complained
about Yahoo's aggressive advertising.  As
promised last month, I have looked into moving
the list to a new home.

Barry Warsaw of Zope PythonLabs (and author
of Mailman) was kind enough to start a new
PythonCE mailing list on python.org. <applause>

  Python CE is an implementation of the
  Python programming language for Windows CE
  based devices such as PocketPC. Python CE
  is free software. Python scripts are small, 
  occupying very little space and convenient
  to download. The scripts don't require separate
  compilation for each target platform and are
  more portable than typical Windows CE applications.

To subscribe to the PythonCE mailing list:

  http://mail.python.org/mailman/listinfo/pythonce

Jeff Bauer
Rubicon Research



From sholden at holdenweb.com  Mon Feb 11 18:27:43 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 18:27:43 -0500
Subject: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254>
Message-ID: <FrY98.156468$KM2.7108035@atlpnn01.usenetserver.com>

"Flavian Hardcastle" <deathtospam43423 at altavista.com> wrote in message
news:Xns91B3637BF6C59AusYourStandingInIt at 210.49.20.254...
> The example at http://www.devshed.com/Server_Side/Python/CGI/page3.html is
> supposed to produce the text in my browser "Hello, Python", but it doesn't
> do a thing.
>
> Instead, using the Xitami personal server, I get the error
>
> >Internal error: your request was unsuccessful
> >Cannot create CGI process - program not found
>
> Here is the text of the example...
>
> #!/usr/bin/python
>
> # Tell the browser how to render the text
> print "Content-Type: text/plain\n\n"
>
> print "Hello, Python!"  # print a test string
>
>
> I had a theory that the first line was the problem. The tut says that it's
> used by Unix and Linux servers to tell the OS to initiate Python. I'm
using
> Win98, btw. I tried changing it to the following......
>
> >#path = c:/Python22/Python.exe
>
> and then directing to the file location on my HD ...
>
> >#path = d:/Docs/Web/cgi-bin/index.py
>
> ... but in both cases the browser merely gave me the option of downloading
> the file, or opening it using Python.
>
> So what am I missing?

If you're using Xitami then the problem almost certainly is your first line.
On Windows you aren't likely to find /usr/bin/python without a stretch of
the imagination.

On Win98 my Xitami CGI scripts all start with

#! C:/Python20/python

This is the classic "shebang" line trick: the "#!" announces to the
execution environment that the containing file should be processed by the
named interpreter. Although Windows in general knows nothing of shebang,
Xitami is prepared to go the extra mile. So I'm guessing if you use

#! C:/Python22/python

all will be well. Xitami's a great little server (with a big heart). Enjoy!

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From chris.gonnerman at newcenturycomputers.net  Thu Feb 14 22:17:03 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Thu, 14 Feb 2002 21:17:03 -0600
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com> <3C6C680D.B35E7719@cosc.canterbury.ac.nz> <jash4a.sm1.ln@127.0.0.1>
Message-ID: <00bb01c1b5cf$3f05ffa0$0101010a@local>

----- Original Message ----- 
From: "Carl Banks" <idot at vt.edu>


> Greg Ewing wrote:
> > Courageous wrote:
> >> 
> >> And while vim indeed does just what I said, when you backspace over
> >> these runs of spaces while in insert mode, it knows how those spaces
> >> where generated and so:
> >> 
> >> "deletes any such spaces inserted through a tab, just as if
> >> it were deleting a single tab."
> > 
> > This sounds ingenious, but what happens when you save the
> > file and load it back? Does it remember this information
> > somehow, or does it get lost?
> 
> The editor remembers this information by saving it in the text file
> itself.

Psychic text editing?  :-)

Actually, when backing over a space which is lined up exactly on a 
tab position, vim assumes it must represent a tab; else it is handled
as a "normal" space.





From aahz at panix.com  Fri Feb 15 22:18:02 2002
From: aahz at panix.com (Aahz Maruch)
Date: 15 Feb 2002 19:18:02 -0800
Subject: thread question
References: <mailman.1013817865.2645.python-list@python.org>
Message-ID: <a4kj1a$p57$1@panix2.panix.com>

In article <mailman.1013817865.2645.python-list at python.org>,
maximilianscherr <MaximilianScherr at T-Online.de> wrote:
>
>if a thread (object) is for example a class, would it be possible to 
>call the class objects of this thread from another thread?
>or how could something similar be done?

Absolutely.  See http://starship.python.net/crew/aahz/
for some examples.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From dutoitc at hotmail.com  Wed Feb 27 04:49:39 2002
From: dutoitc at hotmail.com (=?ISO-8859-1?Q?Dutoit_C=E9dric?=)
Date: 27 Feb 2002 01:49:39 -0800
Subject: PYUT : new Python UML Tool (next : reverse engineering)
Message-ID: <7c19d423.0202270149.58956037@posting.google.com>

We are proud to anounce PyUt, a Python UML tool. (Under GPL)

PyUt is a little UML editor, featuring :
- printing
- import/export dynamic plugins
- Export to XML, JPEG, BMP, PS. 
- Actually runs under Windows and Unix/Linux

Very soon (already coded), PyUt will integer a python
reverse-engineering functionality !!

(next : code generation for python, c++, Java, ...)

Download it on http://sourceforge.net/project/showfiles.php?group_id=39548&release_id=77098

or see screenshots on the home page on => http://pyut.sf.net


C.Dutoit for PyUt team


From rmunn at pobox.com  Tue Feb 19 15:00:27 2002
From: rmunn at pobox.com (Robin Munn)
Date: 19 Feb 2002 20:00:27 GMT
Subject: confusion regarding os.path.walk()
References: <mailman.1014141812.1766.python-list@python.org> <Xns91BABAD1C9BE8andrewdarwinwarscom@194.168.222.9>
Message-ID: <slrna75bg5.hg.rmunn@rmunn.dyndns.org>

On Tue, 19 Feb 2002 18:21:14 GMT, Andrew Brown <killspam at darwinwars.com> wrote:
>
>> I think perhaps you both are somehow thinking that os.path.walk
>> changes it's current directory as it 'walks' the tree: it doesn't.
>> 
>Tuesday, February 19, 2002, 18:02:08, you wrote:
>
>S> The current working directory stays the same and all relative
>S> paths are relative to it. When os.path.walk gives you a directory
>S> and a list of files in that directory, unless that directory is
>S> your current working directory, you need to either join the paths
>S> or change your working directory to access the files.
>
>
>Thanks for your help. What you describe is what I thought should
>happen. But I couldn't figure out how to join the paths. It always
>looked right: I'd do debugging runs that simply printed out the
>command it was supposed to execute, and I could get results like
>
>'ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm'
>
>(which I think should work) but when I ran this through os.system(),
>it failed every time. So did every other variant I could think of
>except changing to the directory where the file was and linking
>without any paths involved.

What was failing -- were you getting on OSError, or was the link
apparently being created without effect? And were you putting the names
in the right order? I can't find your original message, but the way ln
works is (if you're creating a symlink):

ln -s original_file filename_of_symlink

There's another gotcha, too, and this one is *not* obvious: the name by
which you reference original_file should be based on the directory the
symlink will go in. So given your example above, and assuming
burrows.html is the "real" file and Burrows.htm is the symlink you want
to create, then the command you really want to be running is:

ln -s burrows.html ./wombats/habitat/Burrows.htm

This will create the symlink Burrows.htm -> burrows.html in the
directory wombats/habitat. If you did:

ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm

then what you would get would be: in the directory wombats/habitat,
there would be a symlink named Burrows.htm that tried to reference
wombats/habitat/wombats/habitat/burrows.html -- this is almost certainly
not what you wanted.

And as someone else suggested, you should probably be using os.symlink()
instead of os.system() do accomplish this:

$ mkdir foo
$ mkdir foo/bar
$ echo foo > foo/bar/orig
$ ls -l foo/bar
total 2
-rw-r--r--   1 rmunn    users          4 Feb 19 13:53 orig
$ python
Python 2.1.2 (#3, Feb  3 2002, 09:57:02)
[GCC 2.95.1 19990816 (release)] on sunos5
Type "copyright", "credits" or "license" for more information.
>>> import os
>>> os.symlink
<built-in function symlink>
>>> print os.symlink.__doc__
symlink(src, dst) -> None
Create a symbolic link.
>>> os.symlink('orig','foo/bar/symlink')
>>> ^D
$ ls -l foo/bar
total 4
-rw-r--r--   1 rmunn    users          4 Feb 19 13:53 orig
lrwxrwxrwx   1 rmunn    users          4 Feb 19 13:54 symlink -> orig
$ cat foo/bar/orig
foo
$ cat foo/bar/symlink
foo
$ 

Hope this helps.

-- 
Robin Munn
rmunn at pobox.com


From pinard at iro.umontreal.ca  Wed Feb 20 10:27:49 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 20 Feb 2002 10:27:49 -0500
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <m2vgcs5hgf.fsf@desk.crynwr.com>
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <oq8z9o2mx6.fsf@carouge.sram.qc.ca>

[Russell Nelson]

> There are two Python ideoms that bother me:

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

Hello, Russell.  Glad to see you here as well! :-)

For the above, I like writing:

    line = file.readline()
    while line:
        ...
        line = file.readline()

even if I agree that there are cases it would be more awkward to do so,
like for example, when one does not want to conceal the line reading code
into a function.

> I'd really, really like it if you could do one(1) assignment within
> the boolean expression in a while and an if.  Like this:

>     while line = file.readline():

I would rather not see Python accepting this, first because it makes the
code less explicit, but also because it would more easily allow the error
of miswriting `==' as `=' unnoticed by the Python code reader.

> I just think it would be easier to read programs if they could be
> written like this.

Programs may get slightly easier to write, I doubt it would really make
them more legible, however.  My feeling is that the current situation is
better than alternatives, all considered.

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




From emile at fenx.com  Thu Feb 14 12:05:32 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 14 Feb 2002 09:05:32 -0800
Subject: Beginner question please help!
References: <28eed654.0202140843.510948d5@posting.google.com>
Message-ID: <a4guaj$83q2$1@ID-11957.news.dfncis.de>

"Michael Tu" <michael.tu at reachasia.net> wrote in message
news:28eed654.0202140843.510948d5 at posting.google.com...
> Hello,
>
> I am trying to do the following exercise but after spending couple of
> hours without any luck, I think it is time to get some helps. Thank
> you in advance.
>
> Write a compare function that return 1 if x > y, 0 if x == y, and -1
> if x < y.
>

This sounds like homework...

Read up on compare in the documentation.  Pay particular attention to
the cmp() builtin function.  Post your effort if you continue to have
trouble so we can see where you're going with it.

Good luck,

Emile van Sebille
emile at fenx.com




From bkd at graphnet.com  Tue Feb  5 04:46:26 2002
From: bkd at graphnet.com (Bruce Dykes)
Date: Tue, 5 Feb 2002 04:46:26 -0500
Subject: OT: caps lock
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com> <a3hnft$cgu$1@newshost.accu.uu.nl> <3C5CC325.7B3FC988@engcorp.com> <flhp5u4jalahtcm1att17i5vuj2evqoott@4ax.com> <a3j97s$d98$1@newshost.accu.uu.nl> <3C5E6620.8070405@skippinet.com.au>
Message-ID: <02ab01c1ae29$fe144d00$5102020a@graphnet.com>

> > By the way, to my great and enduring happiness I managed to turn it off
> > under XFree86. Even the light doesn't go on when I press it! I've
actually
> > remapped it to be an extra control key, but perhaps escape would be
better
> > for vi use. But I don't really dare getting used to it as I'd screw up
> > as soon as I'd switch to anyone else's keyboard..
>
> I discovered that in Win2k, under "Accessibility Options", there is a
> "toggle keys" option.  Enabling this plays a beep whenever caps or num
> lock is pressed.  This seems a good solution.

Jerry Pournelle's solution is to stuff foam rubber under the key...that way,
it's still there, but you have to press REALLY HARD to activate it.

bkd




From mlh at vier.idi.ntnu.no  Sat Feb  9 06:26:14 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 9 Feb 2002 11:26:14 +0000 (UTC)
Subject: I wish.............
References: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com>  <mailman.1013246955.13570.python-list@python.org>
Message-ID: <slrna6a1ql.jm3.mlh@vier.idi.ntnu.no>

In article <mailman.1013246955.13570.python-list at python.org>, Oleg
Broytmann wrote:
>   Python is the (best) languge. CPython is one (and not the best)
>implementation.

Why is it not the best implementation? Which is better in your
opinion?

>
>Oleg.

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From aahz at panix.com  Wed Feb 27 11:49:59 2002
From: aahz at panix.com (Aahz Maruch)
Date: 27 Feb 2002 08:49:59 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014418936.10783.python-list@python.org> <3C799833.AC08A401@cosc.canterbury.ac.nz> <a5ca53$4ma$1@newsreader.mailgate.org>
Message-ID: <a5j2nn$dst$1@panix2.panix.com>

In article <a5ca53$4ma$1 at newsreader.mailgate.org>,
Mark Hadfield <m.hadfield at niwa.co.nz> wrote:
>
>"Usenet is like a herd of performing elephants with diarrhea--
>massive, difficult to redirect, awe-inspiring, entertaining, and a
>source of mind-boggling amounts of excrement when you least expect
>it."

If you're going to use quotes, please make sure to include attributions
for them.  In this case, that should be Gene Spafford.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From logiplexsoftware at earthlink.net  Thu Feb 14 21:31:09 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 14 Feb 2002 18:31:09 -0800
Subject: Using tabs instead of spaces in IDLE?
In-Reply-To: <3C6C680D.B35E7719@cosc.canterbury.ac.nz>
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com>
	<mailman.1013655704.3770.python-list@python.org>
	<0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com>
	<3C6C680D.B35E7719@cosc.canterbury.ac.nz>
Message-ID: <20020214183109.17a53599.logiplexsoftware@earthlink.net>

On Fri, 15 Feb 2002 14:44:45 +1300
Greg Ewing wrote:

> Courageous wrote:
> > 
> > And while vim indeed does just what I said, when you backspace over
> > these runs of spaces while in insert mode, it knows how those spaces
> > where generated and so:
> > 
> > "deletes any such spaces inserted through a tab, just as if
> > it were deleting a single tab."
> 
> This sounds ingenious, but what happens when you save the
> file and load it back? Does it remember this information
> somehow, or does it get lost?

It doesn't.  It saves the file with spaces.  It's basically emulating tabs
with spaces.  That's the whole point - you get the convenience of tabs with
the uniformity (conventions-wise) of spaces.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From sjmachin at lexicon.net  Sun Feb 24 23:12:18 2002
From: sjmachin at lexicon.net (John Machin)
Date: 24 Feb 2002 20:12:18 -0800
Subject: Correction: Re: string substitutions
References: <mailman.1014504639.21135.python-list@python.org> <c76ff6fc.0202232228.201f829e@posting.google.com> <mailman.1014583431.3100.python-list@python.org>
Message-ID: <c76ff6fc.0202242012.48b17a85@posting.google.com>

Mike Dean <klaatu at evertek.net> wrote in message news:<mailman.1014583431.3100.python-list at python.org>...
> 
> And I had forgotten about the raw strings bit - one question about them
> though - if I'm using raw strings in my RE, is there a way to use a
> newline in an RE, or do I need to resort to regular strings for such a
> situation?
> 

REs were "designed" to be typed into text editors like the venerable
ed, not to be embedded in C or Python or whatever source code. Thus
the RE engine itself, as a user convenience, recognises a two-byte
escape sequence such as "\n" and transforms it into the single byte
0x0A (or maybe more precisely, the local representation of a newline).
This important fact is often not documented well or at all. Then you
have the extra layer of complexity when you embed an RE in program
source code -- the compiler will apply the same or similar
transformations. This leads to backslashorrhea unless the language
kindly gives you a no-escape-processing option like Python's r"...".

So, you don't have a problem at all getting a newline into a raw
string; it's a bit more difficult if you are trying to match a literal
backslash-n. Cut out the following and paste it inside your hat:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>>> import re
>>> txt = r"literal backslash then literal n <\n>; newline <" + "\n" +
">"
>>> txt
'literal backslash then literal n <\\n>; newline <\n>'
>>> # effect of different numbers of backslashes in raw pattern
...
>>> re.sub(r"n", "*", txt)
'literal backslash the* literal * <\\*>; *ewli*e <\n>'
>>> re.sub(r"\n", "*", txt)
'literal backslash then literal n <\\n>; newline <*>'
>>> re.sub(r"\\n", "*", txt)
'literal backslash then literal n <*>; newline <\n>'
>>> re.sub(r"\\\n", "*", txt)
'literal backslash then literal n <\\n>; newline <\n>'
>>> # effect of different numbers of backslashes in cooked pattern
...
>>> re.sub("n", "*", txt)
'literal backslash the* literal * <\\*>; *ewli*e <\n>'
>>> re.sub("\n", "*", txt)
'literal backslash then literal n <\\n>; newline <*>'
>>> re.sub("\\n", "*", txt)
'literal backslash then literal n <\\n>; newline <*>'
>>> re.sub("\\\n", "*", txt)
'literal backslash then literal n <\\n>; newline <*>'
>>> re.sub("\\\\n", "*", txt)
'literal backslash then literal n <*>; newline <\n>'
>>> re.sub("\\\\\n", "*", txt)
'literal backslash then literal n <\\n>; newline <\n>'
>>>

Hope you weren't expecting a one-line answer :-)


From jeff at ccvcorp.com  Tue Feb  5 13:50:27 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 05 Feb 2002 10:50:27 -0800
Subject: EDF scheduler
References: <eTJ78.14206$N31.668446@ozemail.com.au>
Message-ID: <3C602972.19BA110@ccvcorp.com>


Colin Brown wrote:

> Hi
>
> I am trying to implement a simple scheduler in Python where the deadline
> times of (non-prioritised) entries as entered are not necessarily in
> chronological order. From what I see on the Web this is best described as an
> "Earliest Deadline First" scheduler.

Go to the Python Cookbook (http://aspn.activestate.com/ASPN/Cookbook/Python) and
search for "Priority Queue" -- there's a couple of different implementations
there, one of which should suit your needs (or be close enough to show you how
to write your own).

The basic idea is that the queue itself will sort its entries, so that you can
place things in the queue in any order but they will be popped from it
highest-priority-first (however "highest priority" is defined).

Jeff Shannon
Technician/Programmer
Credit International




From sdm7g at Virginia.EDU  Tue Feb 19 16:48:39 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 16:48:39 -0500 (EST)
Subject: execfile question
In-Reply-To: <Pine.OSX.4.43.0202191640240.1674-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <Pine.OSX.4.43.0202191646280.1674-100000@d-128-61-180.bootp.virginia.edu>

PS:
 You can get the behaviour you were expecting if you pass an
unbound anonymous dict to execfile -- the namespace is
garbage that's released after execfile returns:

Python 2.2 (#36, Jan 30 2002, 17:50:25)
[GCC 2.95.2 19991024 (release)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> execfile( 'executable.py', {}, {} )
it's me
init
deleted
>>>



On Tue, 19 Feb 2002, Steven Majewski wrote:

> On Tue, 19 Feb 2002, Kerim Borchaev wrote:
> > And can you explain who holds a reference on C instance in my example
> > when 'execfile' completes?
>
> Python 2.2 (#36, Jan 30 2002, 17:50:25)
> [GCC 2.95.2 19991024 (release)] on darwin
> Type "help", "copyright", "credits" or "license" for more information.
> >>> print execfile.__doc__
> execfile(filename[, globals[, locals]])
>
> Read and execute a Python script from a file.
> The globals and locals are dictionaries, defaulting to the current
> globals and locals.  If only globals is given, locals defaults to it.
>
> >>> execfile( 'executable.py' )
> it's me
> init
> >>> dir()
> ['C', '__builtins__', '__doc__', '__name__', 'c']
> >>> del c
> deleted
> >>>




From jason at jorendorff.com  Mon Feb 25 19:13:13 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 18:13:13 -0600
Subject: raw_input on password
In-Reply-To: <slrna7lf63.4pb.msoulier@pmerd071.ca.nortel.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMECNDFAA.jason@jorendorff.com>

>     Is there a way to get raw_input() to mask the response, so it 
> can be used
> to request a password safely?

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

## Jason Orendorff    http://www.jorendorff.com/



From yannick.leteigner at laposte.net  Wed Feb 13 11:38:56 2002
From: yannick.leteigner at laposte.net (Yannick Le Teigner)
Date: Wed, 13 Feb 2002 17:38:56 +0100
Subject: filtering random values (sieves?)
References: <5e8bd451.0202130827.2df6f50b@posting.google.com>
Message-ID: <3C6A96A0.6000305@laposte.net>

Why don't you put in a list the value you allow to be choosen and make a
random(range(len(list)))?

kevin parks wrote:
 > I am not sure I am even using the right terminology here, but I
 > looking for a way to filter random values. Let's say that I am using
 > random and randrange to generate values from 0-11, but i only want to
 > accept values 1, 3, 5, 7, 11, (that is just off the top of my head,
 > could be anything) how would i filter out values that i don't want?
 > The only way i can think of is just asking the program to choose
 > again, but that may not be the best way since, at least
 > hypothetically, you could keep choosing values that are rejected...
 >
 > Any help would be appreciated.
 >
 > -kevin
 >





From spencer at spencerdoidge.com  Sun Feb  3 20:28:38 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sun, 3 Feb 2002 17:28:38 -0800
Subject: how to pass by reference??
Message-ID: <a3knt6$gr3$1@news.efn.org>

I really tried to RTM first this time, but I failed.
Can someone give me a hint where in the doc html's it explains how to pass
by reference?
Specifically, I want to do the equivalent of this:

char s[10];

void foo(char *bfr)
{
bfr[0] = 'a';
bfr[1] = 'b';
return;
}

void main(void)
{
foo(s);
return;
}




From rony.steelandt at bucodi.com  Mon Feb 18 09:11:27 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 18 Feb 2002 06:11:27 -0800
Subject: win32com and MS Excel
References: <mailman.1013886806.19234.python-list@python.org>
Message-ID: <b34d3835.0202180611.37011c0d@posting.google.com>

Jimmy,

Seems that you are quit good with activex stuff.
There is another thread here called Activex where 2 users has the same problem
(one of them is me :)
Could you have a look at this please 

Just to let you save some time:
I use activestate py 2.1
i generated the makepy file
the active works with severall other languages...

Thanks


"Jimmy Retzlaff" <jimmy at retzlaff.com> wrote in message news:<mailman.1013886806.19234.python-list at python.org>...
> >from win32com.client import Dispatch
> >ll = [['aaa', 333, 'ccc']] * 1000
> >xlApp = Dispatch("Excel.Application")
> >xlApp.Visible = 1
> >xlApp.Workbooks.Add()
> >t = xlApp.Workbooks(1).Sheets(1)
> >t.Range(t.Cells(1,1), t.Cells(len(ll),3)).Value = ll
>  
> >but with 50000 elements you'll have to wait some minutes for
> >Excel to show the data. And with 100000, I get a fatal error. 
> 
> If you haven't already, you can get quite a speed boost by running
> Makepy on the Excel type library. In PythonWin, select "COM Makepy
> utility" on the Tools menu and look for Microsoft Excel. The example
> above with 50,000 takes about 1 second on my 850MHz notebook. The Python
> file that makepy generates can also be quite valuable in figuring out
> Python/Excel nuances. You don't need to do anything in your code with
> that file, its mere existence speeds things up considerably (especially
> if you make a large number of calls into Excel).
> 
> The reason for the error with 100,000 is that Excel only supports 256
> columns and 65536 rows per spreadsheet.
> 
> Jimmy


From marklists at mceahern.com  Tue Feb  5 15:43:09 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 5 Feb 2002 12:43:09 -0800
Subject: What is the command to determine an object's type?
In-Reply-To: <a3pf7k$1a0fij$1@ID-98166.news.dfncis.de>
Message-ID: <NCBBLFCOHHDIKCAFGCFBGEADKBAA.marklists@mceahern.com>

MDK wrote:
> How can I tell if an object is a string, a list, or a dictionary in a
> program using if and ifelse so that I can perform a differnent action
> depending on the object's type?

$ python
>>> import types
>>> l = []
>>> if type(l) == types.ListType:
...     print "yeah baby!"
...
yeah baby!
>>>

// mark



From aliocha_z at caramail.com  Sat Feb 16 13:25:10 2002
From: aliocha_z at caramail.com (Aliocha)
Date: Sat, 16 Feb 2002 19:25:10 +0100
Subject: Apache windows server : python module
Message-ID: <a4m865$80v$1@wanadoo.fr>

Hi,

I have followed all indications to install python module for apache on the
web page :
(http://www.modpython.org/live/mod_python-2.7.6/doc-html/app-wininst.html)
but when I start the server, this are the error lines :

"Syntax error on line 631 of c:/program files/apache
group/apache/conf/httpd:
Invalid command 'PythonHandler', perhaps mis-spelled or defined by a module
included in the server configuration
Note the errors or messages above, and press the <ESC> key to exit.  20..."

Somebody have already met this problem ?

Precisions :
----------------------------------------------------------------------------

The line which doesn't work in httpd.conf :
It's the line "PythonHandler mptest" which redirect to "mptest.py".
This file is in my "C:/Program Files/Apache Group/Apache/htdocs/python" .
this file must print "Hello World"  in my browser.

   <Directory "C:/Program Files/Apache Group/Apache/htdocs/python">
    AddHandler python-program .py
    PythonHandler mptest
    PythonDebug on
   </Directory>

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

Content of mptest.py :

from mod_python import apache

def handler(req):
   req.content_type = "text/plain"
   req.send_http_header()
   req.write("Hello World!")
   return apache.OK

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

Configuration :

I have installed :
- Apache : apache_1.3.22-win32-x86.exe
- the python module files : mod_python-2.7.6.tgz
- The dll of python module :  python2.2-apache1.3.22
- Python executable : Python-2.2.exe





From etoffi at bigfoot.com  Fri Feb 22 04:14:18 2002
From: etoffi at bigfoot.com (e toffi)
Date: 22 Feb 2002 01:14:18 -0800
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com> <nkUc8.19043$kX6.133584@NewsReader>
Message-ID: <655abd4a.0202220114.21447213@posting.google.com>

"Andrew Ward" <winelight at spamcop.net> wrote in message news:<nkUc8.19043$kX6.133584 at NewsReader>...
> >         self.webpage = urllib.urlopen(self.url).read()
> 
> Yes, this can cause Python to hang (thread passim).
> 
> It *seems* that using httplib() instead *might* avoid the hang.
> 
> Andrew ward

it seems i might be having the same sort of problem...

i am writing a special (http) server using SimpleHTTPServer and when i
try to read from do_POST, it hangs...

any suggestions?

--
etoffi


From michael at stroeder.com  Wed Feb 20 12:32:20 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 20 Feb 2002 18:32:20 +0100
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <G5Bc8.1$Az6.364@news.hananet.net> <3C7368A0.5050702@bath.ac.uk> <a4vrdm$j98$1@lise.netcom.no>
Message-ID: <3C73DDA4.3DD72038@stroeder.com>

"Andr? Risnes" wrote:
> 
> "Simon Willison" <cs1spw at bath.ac.uk> wrote in message
> news:3C7368A0.5050702 at bath.ac.uk...
> >
> > It's pretty important to provide some kind of checking mechanism like
> > that to avoid people copying/pasting the URL of the page they are on
> > into an e-mail / instant message and inadvertantly giving their session
> > to someone else.
> 
> That can be avoided by embedding the session ID in a hidden
> field in a form instead of the URL (if forms are used, that is).

A hidden field is not really hidden Especially it's part
of the URL if <form method=GET> is used.

Ciao, Michael.


From thomas at gatsoft.no  Tue Feb 19 07:14:23 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Tue, 19 Feb 2002 12:14:23 GMT
Subject: Dynamic object import ??
Message-ID: <zkrc8.172$JWb.179470336@news.telia.no>

I got a dictionary :

objects = {}

that I want to be able to store object instances in and call with :

result = objects['name of object to use'].result(my_params)

The objects to use are stored in modules in a specific folder, say
'dynamic_objects'.
Maybe the disk-structure could look like this :

/dev/test/dynamic_objects/Test1/Test1.py
/dev/test/dynamic_objects/Test1/__init__.py
/dev/test/dynamic_objects/Test2/Test2.py
/dev/test/dynamic_objects/Test2/__init__.py
/dev/test/dynamic_objects/Test3/Test3.py
/dev/test/dynamic_objects/Test3/__init__.py
/dev/test/dynamic_objects/__init__.py

Each module has a folder with name same as module and each module contains
an object called the same as the module, so Test1.py contains an object
called Test1, which has a method named result taking a specific set of
parameters.

How can I dynamically import and create instances of the objects found in
the Test1, Test2, Test3 modules?
The result should be :

>>> print objects.keys()
>>> ['Test1','Test2','Test3']
>>> print objects['Test1'].result(my_params) # yields some result

I need a way to scan a given folder for modules, import the ones which fits
a certain criteria ( ie. has a proper config-file present etc.), create
instances and map this instance to the name of the module, all in a specific
dictionary.

Any clues appreciated.

NB! If possible, it would be great if the modules are imported in such a
manner that they're reloadable, using reload() in some way.

Best regards,
Thomas Weholt





From SBrunning at trisystems.co.uk  Wed Feb 20 10:17:30 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 20 Feb 2002 15:17:30 -0000
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <31575A892FF6D1118F5800600846864DCBCCC1@intrepid>

> From:	Russell Nelson [SMTP:nelson at crynwr.com]
> There are two Python ideoms that bother me:
> 
>     while 1:
>         line = file.readline()
>         if not line: break
 
As of 2.2, try

for line in file:
    line.whatever()
 
> and:
> 
>     matched = re.match(r'egexp(.*)', line)
>     if matched:
>         print matched.group(1)
> 
> I'd really, really like it if you could do one(1) assignment within
> the boolean expression in a while and an if.
 
Assignments and expressions are different things in Python, and I suspect
that they will remain so - see
<http://www.python.org/cgi-bin/faqw.py?req=show&file=faq06.030.htp>. I
*hope* so, certainly.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From emile at fenx.com  Tue Feb 26 09:16:29 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 06:16:29 -0800
Subject: Anagram
References: <3c4e8efc.1036790726@news.mch.sni.de> <slrna7mijs.lfm.gaul@bifrost.cs.utexas.edu> <a5g3j1$rbq$1@news.hccnet.nl>
Message-ID: <a5g5lv$760cu$1@ID-11957.news.dfncis.de>

From: "Anton Vredegoor"
>  I wrote
> a voluminous script to generate anagrams some time ago, and I am
> looking for options to shorten the algorithm to a few lines of code as
> has been done for the algorithm the OP asked a question about.
>
Hi Anton,

If you have a version of python that support list comprehensions, if you
switch to using them your program would be a lot shorter.  One example:

    def computetresholds(self,sf):
    #tresholds for choosing the next integer
        tresholds = []
        np = self.nperm(sf)
        n = self.sum(sf)
        for x in sf:
            if x<> 0:
                t = (np*x)/n
                tresholds.append(t)
            else:
                tresholds.append(0)
        return tresholds

can be changed to: (untested)

    def computetresholds(self,sf):
        np = self.nperm(sf)
        n = self.sum(sf)
        return [ (np*x)/n for x in sf ]

HTH,


Emile van Sebille
emile at fenx.com

---------



From sholden at holdenweb.com  Fri Feb  8 08:56:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 08:56:04 -0500
Subject: 2001 Turing Award
References: <mailman.1013114112.28909.python-list@python.org> <Xns91AF6B22CC852andy47halfcookedcom@203.109.250.24> <slrna6751a.abe.jajvirta@sirppi.helsinki.fi>
Message-ID: <INQ88.105485$KM2.4022806@atlpnn01.usenetserver.com>

"Jarno J Virtanen" <jajvirta at cc.helsinki.fi> wrote in message
news:slrna6751a.abe.jajvirta at sirppi.helsinki.fi...
> Thu, 7 Feb 2002 23:32:27 +0000 (UTC) Andy Todd wrote:
> > "Mark McEahern" <marklists at mceahern.com> wrote in
> > news:mailman.1013114112.28909.python-list at python.org:
> >
> >> Keith Keller wrote:
> >>> Damn--I thought the award was for the person who best impersonated a
> >>> Turing machine.  :)
> >>
> >> Do you have any nominees?
>
> > I believe there is currently only one Tim-bot.
>
> Given the recent modifications made by the PSU that gave timbot
> abilities to perform like human [1, 2], I'd say that he indeed
> is a strong candidate for this kind of award.
>
> [1] http://mail.python.org/pipermail/python-list/2002-February/085107.html
> [2] http://mail.python.org/pipermail/python-list/2002-February/085042.html

I have the ACM Turing Award Lectures from 1966-1985. It was being
remaindered at a railway station bookstore for $3.00. How are the mighty
fallen :-)

While some of the lectures show a uniquely interesting perspective on our
ever-broadening field, some of them are only fit for use as an insomnia cure
(I was going to say toilet paper, but then I realised they are too valuable
as an insomnia cure to subject to one-time-only usages). This would make Tim
an unlikely recipient -- most of his stuff would be eliminated on the
grounds of it's inherently interesting nature.

I thought the actor that the PSF paid to personate Tim at the Conference was
very believable, and a nice guy. I gave him a business card, which I'm
hoping he passed on to the real Tim for scanning.

only-one-timbot-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From rxg218 at psu.edu  Tue Feb 19 18:13:50 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Tue, 19 Feb 2002 18:13:50 -0500
Subject: calling a function indirectly
Message-ID: <a4v7s6$13tk@r02n01.cac.psu.edu>

Hi,
  is it possible to have a variable contain a function name, and use the 
variable to call the function? 
An example:

def func:
   some code

funcvar = 'func'

Now use funcvar (somehow!) to call func

Any suggestions would be appreciated.
TIA,

-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http://jijo.cjb.net
Department of Chemistry        | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
Gravity brings me down.



From gerson.kurz at t-online.de  Mon Feb 25 07:03:40 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 12:03:40 GMT
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com>
Message-ID: <3c7a278a.13899093@news.isar.de>

Here is a suggestion:
------------------------------------------------
import cPickle
import StringIO
import struct
import os

# your classes
class Client:
    def __init__(self, file):
        self.file = file
    def read(self):
        # read length and load object
        x = os.read(self.file, 4)
        length = struct.unpack('i', x)[0]
        buf = os.read(self.file, length)
        f = StringIO.StringIO(buf)
        return cPickle.load(f)
        
class Server:
    def __init__(self, file):
        self.file = file

    def write(self, obj):
        # write length of serialozed object, write serialized object
        f = StringIO.StringIO()
        cPickle.dump(obj, f, 1)
        val = f.getvalue()
        length = len(val)
        plen = struct.pack("i", length)
        os.write(self.file, plen)
        os.write(self.file,val)

# my classes        
class Client2:
    def __init__(self, file):
        self.file = file
    def read(self):
        # assume size repr is less than 20 bytes
        x = os.read(self.file, 20)
        pos = x.find('.')
        size_expected = int(x[:pos])
        if size_expected > 20:
            # read rest of data
            more_data = os.read(self.file, size_expected - pos )
            return cPickle.loads(x[pos+1:] + more_data)
        else:
            return cPickle.loads(x[pos+1:])
        
class Server2:
    def __init__(self, file):
        self.file = file

    def write(self, obj):
        test = cPickle.dumps(obj)
        # interestingly, this seems to be slower:
        #os.write(self.file,"%d." % len(test) + test)
        os.write(self.file,"%d.%s" % (len(test),test))

# test script
if __name__=='__main__':
    rp, wp = os.pipe()

    import time

    test_string = u"this is a test string that is more than 20 bytes
long"

    c = Client(rp)
    s = Server(wp)
    start = time.clock()
    for i in xrange(3000):
        s.write(test_string)
        assert(test_string==c.read())
    print time.clock() - start

    c = Client2(rp)
    s = Server2(wp)
    start = time.clock()
    for i in xrange(3000):
        s.write(test_string)
        assert(test_string==c.read())
    print time.clock() - start


# results:
0.518367439587 (your)
0.228796436055 (mine)
-------------------------------------------------------

But, your version seems reasonably fast already :). My "Full-Blown
RPCServer" that uses dedicated threads for function calls (so that one
RPC call can issue another RPC call while not blocking other threads
making their RPC calls etc) is *a lot* slower...






From chris.gonnerman at newcenturycomputers.net  Sun Feb 17 22:59:03 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 17 Feb 2002 21:59:03 -0600
Subject: Guido wins FSF Award
References: <Pine.LNX.4.32.0202171452390.1394-100000@aesaeion.com>
Message-ID: <005901c1b830$9d5f6420$0101010a@local>

----- Original Message -----
From: <kosh at aesaeion.com>
>
> I have never been an amiga user and have never made posts to the amiga
> newsgroup unless it was crossposted from the linux advocacy newsgroup
> which would be pretty strange. For irc I am in a relatively limited number
> of places. irc.worldforge.org and irc.openprojects.net #zope

Likely "Kosh" is a fairly frequent pseudonym... and as far as I am
concerned,
Vorlons are welcome here.

"All Vorlons are Kosh"...

Rue is so busy abusing the programmers he is trying to convince, he has
no time for proper geek pursuits such as watching TV.

If your handle isn't, in fact, a reference to B5, my apologies (there I
go again) but I'm sure that many who use that handle are B5 fans.





From shriek at gmx.co.uk  Thu Feb 28 22:47:26 2002
From: shriek at gmx.co.uk (Stephen)
Date: 28 Feb 2002 19:47:26 -0800
Subject: How to have multi-threaded server log to one file ?
References: <mailman.1014873293.20465.python-list@python.org>
Message-ID: <97ae44ee.0202281947.78dea8cf@posting.google.com>

> > [...] how do I get the same Queue object into both MyHandler and
> > MyServer in the first place ? [...] Or is this somewhere that a
> > global variable Queue object should be used ?
> 
> Try making a separate module that says:
> 
> ---------------------------------------------------------------
> # logging.py - used by everyone else that needs to log messages
> 
> import Queue
> 
> # This is the queue object that's used to hold log events.
> # LoggingThread knows about it, and the log() function below
> # knows about it.  No one else is allowed to see it.
> _log_queue = Queue.Queue()
> 
> class LoggingThread:
>     ...
>     def run(self):
>         while 1:
>             message = _log_queue.get()
>             self.logfile.write(message + '\n')  # (or whatever)
> 
> # Kick off the logging thread.
> LoggingThread().start()
> 
> def log(msg):
>     _log_queue.put(msg)
> 
> ---------------------------------------------------------------
> 
> Then your MyHandler and MyServer code can just use the log()
> function, like this:
> 
>   from logging import log
>   ...
>   ...
>   log("Something happened")
>   ...
> 
> ## Jason Orendorff    http://www.jorendorff.com/

(I thought I'd posted a reply to this but it seems to have got lost.)

Thank you Jason, this is such a clear and simple example. It's never
occured to me to create the object within the module (in your example,
the Queue.Queue instance "_log_queue" within the "log" module) and
then access it indirectly through a function (ie. log.log()). I
usually end up passing objects around in my code but the method you
describe looks like an incredibly good way to avoid this as well as
hide complexity.

This type of code (having shared objects instantiated within an
imported module) looks like something I should employ more often.
One of them 'ta-da' moments.

Stephen.


From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 08:35:32 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 07:35:32 -0600
Subject: XP (was Re: Defending the Python lanuage... )
References: <3C62EE4A.5090305@destiny.com> <20020207140755.364d5d77.logiplexsoftware@earthlink.net> <200202081311.g18DBQGo020260@ratthing-b246.strakt.com>  <20020208103714.11a2ae01.logiplexsoftware@earthlink.net>  <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com>
Message-ID: <002201c1b301$10d63ee0$0101010a@local>

----- Original Message ----- 
From: "Laura Creighton" <lac at strakt.com>
> 
> Don't formalise the process too much.  You are not trying to bring 
> order to your establishment by building rules that work top-down.  
> You are trying to foster order that works from the bottom up.

Here is the ideal of XP which I think is being lost on managers 
trying to implement it for buzzword compliance.  They don't really
understand what they are doing, so they will *mandate* all parts
of XP even if they are incompatible with the people doing the work.

XP is only going to work if you adapt it together with your people.
Each side may have to give; if the programmers you are working with
are hardheaded (and which of us aren't), forcing them to pair-program
or do other XP things is *not* going to improve anything.

I've lived through too many "paradigm shifts" already.  I remember
the fiasco of TQM being applied to a scientific laboratory, where
commitment to quality was already so high that the technicians were
insulted.  Managers blindly applying buzzwords like Band-Aids makes
me curse... whether it's TQM, or Java, or XP, *no* "new way" will
work for everyone.

It's just not a one-size-fits-all world.






From phr-n2002a at nightsong.com  Mon Feb  4 17:31:25 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 14:31:25 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org>
Message-ID: <7x8za8dgle.fsf@ruckus.brouhaha.com>

"Tim Peters" <tim.one at home.com> writes:
> > If the official Python maintainers think they might release a 1.5.3
> > someday, then it's fair to say that the 1.5 series is still current.
> > If they're unlikely to ever release another 1.2.x or 0.9.x, then
> > those versions are not current, even if I release a 1.2.x myself.
> 
> You misunderstand the bugfix release process:  "the official Python
> maintainers" don't do bugfix releases, period.  We haven't had the bandwidth
> for it, and it doesn't look like we will.   All bugfix releases have been
> produced by community volunteers (2.0.1 was an exception, but was an
> artificial release just to get a GPL-compatible according-to-FSF license on
> it).  Fred Drake, Barry Warsaw and I have so far been able to put a little
> bit of time into doing the last steps of bugfix releases (packaging and
> file-uploading and announcing), but that's all.
>
> The odds of getting a 1.5.2 bugfix release are, as for any other bugfix
> release, identical to the odds of someone volunteering to produce one.

Let me ask it this way:

  a) If I put together a 1.5.2 bugfix release and submit it, will it go
     up on Python.org with a link on the front page like the current
     1.5.2 link?

  b) If I put together a 1.2.whatever bugfix release and submit it, will
     THAT go up on the front page?

If the answer to those two questions is not the same, then the bugfix
releases are not run entirely by the volunteers.  

> > (I'm still using 2.1 and plan to get around to upgrading to 2.1.1).
> 
> You should move to 2.1.2 next, then (2.1.1 isn't even "current" by your
> definition <wink>).

Yes, I confused 2.1.1 with 2.1.2.  I did mean 2.1.2.  Sorry.

> If you have messier problems, and you're determined to write code that works
> unchanged back through the first Python ever released, then sure -- slicing
> divmod is clear and works.  ...

No.  I'm taking a practical view of this.  I use HTML tables in my
webpages even though there's still a few ancient browsers out there
that don't display them.  But if 25% of browsers didn't support them,
I'd avoid using them when I didn't need them.  I do avoid making my
pages depend on Javascript, for basically that reason.

I see publishing Python programs as like publishing web pages.  If I
have to tell 2% of my users to upgrade those ridiculous old browsers
to view my pages, I don't mind.  If I have to tell 25% of them to
upgrade, I'm causing hassles for a significant population and I
shouldn't do that without a good reason.  If I have to tell 80% of
them to upgrade, I'm probably going to lose a lot of them.

The question is, what percentage of users are using what Python version?

I bet less than 10% of users have 2.2 right now.  I'm curious about
the ratio of 2.1 users to 1.5.2 users.  Do you have any idea what it
is?  I don't know how I'd measure it.  There's no User-agent header
collected at python.org every time someone runs a Python script.

I go back and forth about whether to code for 1.5.2.  I do like 2.1
features like +=, and I noticed last night that the "continue"
statement works differently.  If I had reason to believe that less
than (say) 25% of today's installations are 1.5.2, then I wouldn't
bother with 1.5.2; but I think the percentage may be higher than that.


From cfelling at iae.nl  Wed Feb 20 14:31:48 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 20 Feb 2002 20:31:48 +0100
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <a50tj4$4a7$1@animus.fel.iae.nl>

Russell Nelson <nelson at crynwr.com> wrote:
...
>     if matched = re.match(r'egexp(.*)', line):
>         print matched.group(1)

If you are in need of something like this, you could consider wrapping
it up in an assignable-self-returning var, like:

    >>> class Var:
    ...     def assign(self, value):
    ...         self.value = value
    ...         return self.value
    ... 
    >>> import re
    >>> var = Var()
    >>> if var.assign(re.findall("spam", "spam and spam")):
    >>>     print var.value
    ...
    ['spam', 'spam']

And for further fun you could sugar it up, like:

    >>> class CleverVar(Var):
    ...	    def __getattr__(self, attr):
    ...	        return getattr(self.value, attr)
    ...
    >>> var = CleverVar()
    >>> if var.assign(re.match("(spam)", "spam and spam")):
    ...     print matched.groups()
    ('spam',)
-- 
groetjes, carel


From deathtospam43423 at altavista.com  Mon Feb 18 05:45:20 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 18 Feb 2002 10:45:20 GMT
Subject: Newbie: Problems with File IO
References: <mailman.1013981489.9375.python-list@python.org>
Message-ID: <Xns91B9D83F1CFC1AusYourStandingInIt@130.133.1.4>

I cracked it! 

It was a lot simpler than I thought. Immensely obvious infact! 

In the part of the program which processes the fieldstorage, I inserted the 
lines...


The_Form = cgi.FieldStorage()

if (The_Form.has_key("handle")):
    user = str(The_Form["handle"].value)
elif not (The_Form.has_key("handle")):
    user = " "

So the user gets to decide their handle, and handle will be stored in  the 
string "user".

Then, in the actual html form itself, I inserted ....

print '<input type=text NAME="handle" SIZE="30" MAXLENGTH="30" value = 
"'+user+'">'

So whatever the user puts in the  input box "handle", will come back to 
them when the script re creates the html page.

Simplicity itself heh? I can't imagine why I didn't spot it before! Perhaps 
I did, but for some strange reason I thought the script would confuse 
everyone's handles.

I've already had a chat buddy test it out with me, it works ok. The only 
problem is that the posts are stored in the file "log.txt", and the script 
reproduces the file in it's entirety every chat session. The result being, 
the longer you chat, the bigger and bigger web page gets, until it's a cow 
to download.

For my next feat of mental contortion, I have to figure out a way to limit 
the number of posts in the "log.txt" file that are displayed. Ideally, I 
want the user to be able to specify that amount, so that (s)he can chose 
how many old posts (s)he wishes to view.


"Shaun Laughey" <shaun at laughey.com> wrote in 
news:mailman.1013981489.9375.python-list at python.org:

> If you want to remember them forever (sic) then use cookies.
> http://www.python.org/doc/current/lib/module-Cookie.html
> 
> If you want to remember them from post to post during the same "session"
> then use a session object (a temporary cookie in effect) or encode a
> unique session into the url (not as easy).
> 
> pyweblib.session contains all the functions you need.
> 
> Then you would simply
> 
> User connects
> Do they have a session?
> No - generate one.
> Do they have a username?
> No - ask for it and store in session username
> and
> Call the clean up function periodically to clear out old sessions.
> On your page display the session username.
> 
> Remember to use try: or has_key or else you will get errors if there is
> no username key in the session object.
> 
> For pyweblib see http://www.stroeder.com/pylib/PyWebLib/
> 
> Documentation isn't a strong point but the library should give you a
> help.
> 
> 
> Hope this helps
> Shaun Laughey,
> 
> -----Original Message-----
> From: python-list-admin at python.org [mailto:python-list-admin at python.org]
> On Behalf Of Flavian Hardcastle
> Sent: 17 February 2002 02:25
> To: python-list at python.org
> Subject: Re: Newbie: Problems with File IO
> 
> Chris Liechti <cliechti at gmx.net> wrote in 
> news:Xns91B81F8828230cliechtigmxnet at 62.2.16.82:
> 
>> Flavian Hardcastle <deathtospam43423 at altavista.com> wrote in 
>> news:Xns91B873A49D077AusYourStandingInIt at 210.49.20.254: 
>> 
>>> out_file = open("log.txt", "w")
>> mode 'a' does append (thats the one you want), 'w' starts a new file 
>> for writing, adding a plus 'a+' or 'w+' opens for read/write
>> 
> 
> Thanx! That's got me a bit further. What part of the standard Python
> manual 
> (the html one you get at Python.org) would that be under?
> 
> Also, if you could give me a hint as to how I might get the script to 
> remember people's usernames everytime they post, that'd be good.
> 



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From gh_pythonlist at gmx.de  Sat Feb  9 17:45:45 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Sat, 9 Feb 2002 23:45:45 +0100
Subject: Python MySQL and MAC OS X
In-Reply-To: <3C656ACD.1020907@inter.nl.net>
References: <3C656ACD.1020907@inter.nl.net>
Message-ID: <20020209224542.GA1513@lilith.hqd-internal>

Le 09/02/02 ? 19:30, Peter HJ van der Kamp ?crivit:
> Hello,
> 
> I'm trying to install the Python MySQL module, without success.
> According to the MAC OS X documentation of Python 2.0 I have
> to drop the setup.py file onto the interpreter and pass the
> build argument by setting a Unix style command line (which can
> be set by holding the option key during interpreter startup).

Sounds frightening ;-) Don't you have a shell or terminal window on
MacOS X?

> But this procedure leads to a -43 error, file not found. Anyone any
> idea how to build and install the module?

Ugh. You need a shell. Absolutely.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.1 ?C      Wind: 6.1 m/s



From BPettersen at NAREX.com  Wed Feb 20 15:55:51 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Wed, 20 Feb 2002 13:55:51 -0700
Subject: HTMLifyer for Python code?
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E2019221513D@admin56.narex.com>

I just browsed through the Vaults, and did a quick Google search, but I
couldn't find a tool for converting Python source code to
syntax-highlighted HTML. Surely someone must have written something like
this?

-- bjorn



From jajvirta at cc.helsinki.fi  Fri Feb  8 04:02:34 2002
From: jajvirta at cc.helsinki.fi (Jarno J Virtanen)
Date: 8 Feb 2002 09:02:34 GMT
Subject: 2001 Turing Award
References: <mailman.1013114112.28909.python-list@python.org> <Xns91AF6B22CC852andy47halfcookedcom@203.109.250.24>
Message-ID: <slrna6751a.abe.jajvirta@sirppi.helsinki.fi>

Thu, 7 Feb 2002 23:32:27 +0000 (UTC) Andy Todd wrote:
> "Mark McEahern" <marklists at mceahern.com> wrote in 
> news:mailman.1013114112.28909.python-list at python.org:
> 
>> Keith Keller wrote:
>>> Damn--I thought the award was for the person who best impersonated a
>>> Turing machine.  :) 
>> 
>> Do you have any nominees?

> I believe there is currently only one Tim-bot.

Given the recent modifications made by the PSU that gave timbot 
abilities to perform like human [1, 2], I'd say that he indeed
is a strong candidate for this kind of award.

[1] http://mail.python.org/pipermail/python-list/2002-February/085107.html
[2] http://mail.python.org/pipermail/python-list/2002-February/085042.html


From bkd at graphnet.com  Fri Feb  8 10:51:55 2002
From: bkd at graphnet.com (Bruce Dykes)
Date: Fri, 8 Feb 2002 10:51:55 -0500
Subject: [NEWBIE] List issue aka: "Why doesn't this code work?"
Message-ID: <025d01c1b0b8$a01b4fa0$5102020a@graphnet.com>

Okay, here's my function:

def process_numbers(source_list):
    for number in range(len(source_list)):
        entry = string.split(string.strip(source_list[number]))
        print number, len(entry), entry

Now this function is doing *something*...source_list is a list of lines
retrieved via ftp.retrlines(), and this should break each line into it's own
list, and it looks like that's what's being printed out:

['line1', 'header1', 'header2']
['line2', 'datum1', 'datum2']

Notice what's missing? I'm expecting to see:

0 3 ['line1', 'header1', 'header2']
1 3 ['line2', 'datum1', 'datum2']

What the heck am I doing wrong here?
I thought maybe a I had a double nested list, but I tried substituting entry
with entry[0], but that didn't seem to affect anything, and besides, I
should still get number at least, right?


thanks
bkd




From joshm at taconic.net  Tue Feb  5 00:55:50 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 5 Feb 2002 00:55:50 -0500
Subject: EDF scheduler
References: <eTJ78.14206$N31.668446@ozemail.com.au>
Message-ID: <3c5f7184_3@corp.newsgroups.com>

I had a slightly different problem, but perhaps my solution can be applied
to your problem as well.  (If this comes out a bit obtuse, it is because I
can't really give away what the specific problem was to solve.  NDAs suck.)

In my case, I had a collection of objects to be monitored.  Each one
essentially had separate conditions which caused state changes over time.
So, for example, an object might change from state A to state B after 30
seconds, unless some intervening event caused the state to reset, etc.
Another might change from A to C after 10 seconds.  Now this would be easy
to solve by allocating a thread to each object, but in fact the number of
objects wasn't able to be bounded.  (There could have been thousands of
them.)  And since the conditions which caused state changes could vary from
object to object, you couldn't easily group them into common timers.

My solution was to use a single thread, and essentially poll each object,
allowing each one the opportunity to perform its state changing logic, if
necessary.  This function would return immediately if the object didn't need
to be handled at that time.  More importantly, though, is the fact that each
object *knows* when it needs to be handled next.  (It can compute this since
it knows when it last changed state.)  So this function always returns the
value of *when* this object will need to be handled again.  The minimum of
these return values is how long the single thread can sleep before repeating
the cycle.  (There are a variety of optimizations to this process, but you
hopefully get the idea.)  There does need to be a sufficient granularity to
the timing requirements, to deal with the overhead of polling, etc.)

The main loop then becomes something like this (typing off the top of my
head):

while 1:
    nextTime = reduce(min, [x.PollAndComputeNextTime() for x in
listOfObjects])
    now = time.time()
    sleep(nextTime - now)

God, I love Python.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From lac at strakt.com  Tue Feb 26 06:06:31 2002
From: lac at strakt.com (Laura Creighton)
Date: Tue, 26 Feb 2002 12:06:31 +0100
Subject: 15 minute presentation on python? 
In-Reply-To: Message from "Joshua Muskovitz" <joshm@taconic.net> 
   of "Mon, 25 Feb 2002 17:51:36 EST." <3c7abd92_1@corp.newsgroups.com> 
References: <3c7ab974_4@corp.newsgroups.com> <a5dia2.3vs4847.1@kserver.org>  <3c7abd92_1@corp.newsgroups.com> 
Message-ID: <200202261106.g1QB6VtB014487@ratthing-b246.strakt.com>

15 minutes is not a lot.  Any chance you can get some Unit testing in?
The example in http://diveintopython.org/roman_divein.html is the
right sort of stuff, but too long for 15 minutes.

Just an idea,
Laura Creighton



From steve at spvi.com  Wed Feb 27 07:28:37 2002
From: steve at spvi.com (Steve Spicklemire)
Date: Wed, 27 Feb 2002 07:28:37 -0500
Subject: errno, threads, Numeric etc...  how to debug this?
Message-ID: <8676449E-2B7D-11D6-A2AE-0050E480B13C@spvi.com>

Hi Folks,

I'm trying to build Numeric on MacOSX and while it basically works... it 
seems there are some problems. When I try to run the Demos/NumTut 
examples (using xfree86 and python2.2 from fink) I get random 
"unexpected math error" exceptions.

I checked the source and discovered that the errors are occuring in 
ufuncmodule. I edited ufuncmodule to get more info like so:

[spicklemire:~/Packages] steve% diff -c 
Numeric-20.3-orig/Src/ufuncobject.c Numeric-20.3/Src/ufuncobject.c
*** Numeric-20.3-orig/Src/ufuncobject.c Tue Aug 14 11:18:24 2001
--- Numeric-20.3/Src/ufuncobject.c      Mon Feb 25 15:42:08 2002
***************
*** 470,477 ****
         if (self->check_return) check_array(mps[i]);
       }

!     if (self->check_return && errno != 0) {math_error(); return -1;}
!
       return 0;
   }

--- 470,488 ----
         if (self->check_return) check_array(mps[i]);
       }

!     if (self->check_return && (errno != 0))
!       {
!       if ((errno == EDOM) || (errno == ERANGE))
!         {
!           math_error(); return -1;
!         }
!       else
!         {
!           fprintf(stderr, "Hmm.. error encountered -> %i\n", errno);
!           perror("error in ufunc generic function:");
!         }
!       }
!
       return 0;
   }


Now when I try to run the NumTut demo I see:

[spicklemire:~/Packages/Numeric-20.3/Demo] steve% python -c "import 
NumTut;NumTut.test()"
 >>> view(greece)
 >>> view(1.0-greece)
Hmm.. error encountered -> 35
error in ufunc generic function:: Resource temporarily unavailable
 >>> view(greeceBW)
 >>> view(greece*xgrade)
 >>> view(greece*ygrade)
 >>> negative = 1.0 - greece
 >>> view(greece*xgrade + negative*ygrade)
 >>> view(red)
 >>> view(green)
 >>> view(blue)
 >>> sine = sin(xgrade*6*pi)
 >>> view(green*sine + red*(1.0-sine))
 >>> view(green + red[::-1])
 >>> view(transpose(greece, (1,0,2)))
[spicklemire:~/Packages/Numeric-20.3/Demo] steve% python -c "import 
NumTut;NumTut.test()"
 >>> view(greece)
 >>> view(1.0-greece)
 >>> view(greeceBW)
 >>> view(greece*xgrade)
 >>> view(greece*ygrade)
 >>> negative = 1.0 - greece
 >>> view(greece*xgrade + negative*ygrade)
 >>> view(red)
 >>> view(green)
 >>> view(blue)
 >>> sine = sin(xgrade*6*pi)
 >>> view(green*sine + red*(1.0-sine))
 >>> view(green + red[::-1])
 >>> view(transpose(greece, (1,0,2)))
[spicklemire:~/Packages/Numeric-20.3/Demo] steve% python -c "import 
NumTut;NumTut.test()"
 >>> view(greece)
 >>> view(1.0-greece)
Hmm.. error encountered -> 35
error in ufunc generic function:: Resource temporarily unavailable
 >>> view(greeceBW)
 >>> view(greece*xgrade)
 >>> view(greece*ygrade)
 >>> negative = 1.0 - greece
 >>> view(greece*xgrade + negative*ygrade)
 >>> view(red)
 >>> view(green)
 >>> view(blue)
 >>> sine = sin(xgrade*6*pi)
 >>> view(green*sine + red*(1.0-sine))
 >>> view(green + red[::-1])
 >>> view(transpose(greece, (1,0,2)))

So... ufuncs just multiply, divide and suchlike! What "resources" are 
not available? It's fairly clear that it's thread related, since all the 
Numeric tests pass every time in a non-threaded example, as well as my 
own non-threaded testing to attempt to reproduce what I thought 
originally was a linking problem.

So... how can I track down where errno is actually getting set? Does the 
fact that it happens "randomly" mean that it's not due to some numerical 
computation, but more likely about timing between threads?

Also.. it seems that on MacOSX there is a separate "errno" value for 
each thread... so how could they be confused?

thanks for any insight. ;-)

-steve




From tdelaney at avaya.com  Mon Feb 25 18:39:21 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 26 Feb 2002 10:39:21 +1100
Subject: Reloading modules
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1AB@natasha.auslabs.avaya.com>

> From: Jeff Shannon [mailto:jeff at ccvcorp.com]
> Steve Holden wrote:
> 
> >
> > if file[-1] in "co":
> >     file = file[:-1]
> 
> Given the source of 'file' in this case, this is probably not
> significant, but there's a possible gotcha here if empty strings are a
> possibility.

That's why *I* used a slice. However, the following will work ...

if file[-1:] in 'co':
    file = file[:-1]

which is probably the most consise way. Using a sequence containing strings
is the more general form though, for when you need to check more than a
single character.

Tim Delaney



From dalke at dalkescientific.com  Sat Feb  2 12:40:18 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Sat, 2 Feb 2002 10:40:18 -0700
Subject: Equivalent of Perl chomp?
References: <mailman.1012403361.2370.python-list@python.org> <u5i8b8jtee2r13@corp.supernews.com> <7xbsfag023.fsf@ruckus.brouhaha.com> <3C59D2EE.3937ABBB@ndh.net> <a3f7dh$lq9$1@news.island.liu.se> <isF68.111010$8b.4218974@atlpnn01.usenetserver.com>
Message-ID: <a3h81i$ppa$1@nntp9.atl.mindspring.net>

Steve Holden:
>Perhaps I'm just lazy, but why not process
>
>    for line in file(filename).read().split("\n"):

If there is one "\n" terminated line in the file then this
returns two entries, with the second being "".   This is
because of the "" between the "\n" and the end of the file.

                    Andrew
                    dalke at dalkescientific.com






From jgarpc at netscape.net  Mon Feb 11 22:14:49 2002
From: jgarpc at netscape.net (Jose Garcia)
Date: 11 Feb 2002 19:14:49 -0800
Subject: Pysol 4.8 install?
Message-ID: <e0bf95b7.0202111914.5e0324e3@posting.google.com>

Question...

This may not be related.

How is Pysol 4.8 installed on Windows 98?

I have Python 1.5.2, Python 2.2, tcl 8.3, tk 8.3, tcl 8.4a3, tk 8.3a3,
Pysol-4.80-src.tar, and Pysol-4.80.tar

Do these files have to be in the same directory? 
If so, which one?

In C:\Python22 there is the tcl folder which has tcl 8.3, tk 8.3, and
also in C:\Python22 is the Pysol-4.80.tar folder.

When I click onto pysol_15.pyc to pysol_22.pyc, a DOS window just
pops up momentarily.


Pysol 4.2 and 4.6 was just an .exe file, and that was all that was neede
to install.

How do I run Pysol 4.8?

Thanks...


From justin at iago.org  Tue Feb 19 14:34:21 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 19 Feb 2002 14:34:21 -0500
Subject: Adding static typing to Python
In-Reply-To: <a4s51k$8pd1@news1.gtech.com> ("Peter Milliken"'s message of
 "Tue, 19 Feb 2002 11:07:19 +1100")
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
	<a4s51k$8pd1@news1.gtech.com>
Message-ID: <x7n0y56zb6.fsf@ra.iago.org>

"Peter Milliken" <peter.milliken at gtech.com> writes:

> The loose typing of Python is perceived...

Python has strong typing.

It also has dynamic typing, as opposed to static typing.  This seems
to be the thing that you're objecting to.  

Sorry to nitpick, but if you're going to talk about these sorts of
details, getting the terminology right is a good start.

> So I doubt that the proponents of Python in these instances have any
> idea of the true cost of this language "feature" - they appear to be
> wound up in how quickly they can produce a working prototype (or
> whatever) so I suspect their day to day work involves very small,
> manageable projects where the lost efficiency due to loose typing
> isn't really that noticable for them i.e. they "live with it" :-).

I doubt that you have any data to back this up with, and since most of
your talk was about lack of metrics, I don't see that you have any
ground to stand upon from which to criticize others.

I'm currently working on a medium-sized project implemented mostly in
Python that demands serious reliability.  I've also worked on projects
in languages that enforced static typing.  (I've even worked at one of
the sorts of places well-known for using Ada, though I only have a
little bit of personal experience with that language.)  I am familiar
with the issues here.

> I often wonder whether I wouldn't be better off using my real
> language of choice (Ada) because of it's strong typing - it too has
> interfaces to the standard GUI tools etc. My next personal project
> will not be in Python

If Ada is more able to suit the needs of your project (and one major
concern in that area is often a matter of a programmer's personal
taste) then you should use it.

> When I compile a Ada program cleanly I know that 95% of the bugs are out of
> it - before I even start testing! :-) Of course, it takes me longer to write
> up front but overall development effort (in my subjective opinion - since I
> have no metrics :-)) is that it is much lower.

Not only do you have no metrics for your overall development effort, I
am willing to wager that you don't have any to back up that 95% claim either.

> Strong typing requires that you design before you code

This is absolutely false.  

I have personally seen many counterexamples.

Yes, I read your other message explaining what you mean here.  Your
conclusion is still wrong, though.

> You should try it one day :-)

I have used a number of strongly-typed languages, including Python.

I have also spent plenty of time with statically-typed languages.  

> Yes, sane development practices are essential on such large projects -
> doesn't matter what language is used. Part of my argument is that language
> features are designed and provided as further "encouragement" down the path
> of "correct" practices. Typing is recognised as one of these features......

Right, and what practices are correct or useful also depends on the language.

If I was using C++ or Java (or Ada) and the compiler failed to
complain about an obvious type mismatch, I would be quite upset.
However, the Python language encourages different practices, and often
those are such that static typing would be far more of a detriment
than a boon.

To clarify my own position: I would love it if a Python implementation
could do some reasonable type inference, and/or make use of optional
type declarations either for error-checking or optimization.  However,
completely removing dynamic typing from Python would be a disaster.

-Justin

 




From martin at v.loewis.de  Wed Feb 27 15:55:39 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 21:55:39 +0100
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com> <a5jbra$7ok00$1@ID-11957.news.dfncis.de>
Message-ID: <m3ofiair04.fsf@mira.informatik.hu-berlin.de>

"Emile van Sebille" <emile at fenx.com> writes:

> class Counter:
>     def __init__(self):
>         self.count = 0
>     def __int__(self):
>         self.count += 1
>         return self.count
> 
> counter = Counter()
> print "the current count is %d" % counter
> 
> Hopefully, someone will post something pythonish illustrating closures.
> Maybe this time I'll get it.  ;-)

As Zenin points out, Python uses objects to emulate closures, whereas
Lisp uses closures to emulate objects. So your solution is very
Pythonic.  To allow calling the object over and over again, you could
implement __call__ instead of __init__.

With Python 2.2, the yield statement allows that kind of data flow as
well.

Regards,
Martin


From orutherfurd at wwre.org  Thu Feb 14 12:17:50 2002
From: orutherfurd at wwre.org (Oliver Rutherfurd)
Date: Thu, 14 Feb 2002 12:17:50 -0500
Subject: '/' not allowed in cookie name?
Message-ID: <C4EFDBA92975E84D8838E9DA2656632C5615E8@mserver1.wwre.org>

Hi,

I'm using Steve Purcell's excellent 'WebUnit' module, but receiving a cookie.CookieError when a server I am connecting to sends me a cookie with '/' in the name.

Something like:

Set-Cookie: 1/2="one half"

After spending a little time reading http://www.faqs.org/rfcs/rfc2109.html I am still not any closer to understanding whether '/' is a legal value that the cookie module should accept in a name, or whether the server is misbehaving.

Adding a '/' to _LegalChars on line 252 of Cookie.py (using Python 2.2) seems to make my problem go away, but I don't know whether this is the correct things to do.

Is this a bug in Cookie.py or with the server I am connecting to?

Regards,
-Ollie Rutherfurd
<oliver at rutherfurd dot net>



From sholden at holdenweb.com  Sun Feb 10 13:21:37 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 10 Feb 2002 13:21:37 -0500
Subject: HTML Generators
References: <I5m98.1105$dy4.39461@eagle.america.net>
Message-ID: <HSy98.138288$KM2.5977399@atlpnn01.usenetserver.com>

"Mark Sass" <sassman at nucentrix.net> wrote in message
news:I5m98.1105$dy4.39461 at eagle.america.net...
> Are there any HTML generator modules for Python?  I just want to create
> static html and not use scripts i.e. CGI to do it.  Any available?
>
Loads. Try navigatin to www.google.com and entering something like "python
HTML module" as your search string, and you'll find out about PS (several
flavors), Cheetah, HTMLGen and many others.

after-which-yer-pays-yer-money-and-yer-takes-yer-pick-ly y'rs  - steve

--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From phr-n2002a at nightsong.com  Sat Feb 16 18:03:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 16 Feb 2002 15:03:36 -0800
Subject: "implemented by subclass"
References: <slrna6t9de.nt2.philh@comuno.freeserve.co.uk>
Message-ID: <7xwuxdowqf.fsf@ruckus.brouhaha.com>

philh at comuno.freeserve.co.uk (phil hunt) writes:
> Consider the following python code:
> 
> class Log:
>    def write(self, text): 
>       """(implemented by subclass)"""  
> 
> 
> Is this the most pythonic way to do it? If not, what would be
> better? How about rasing a ShouldBeImplementedBySublass exception?
> 
> How do other people do it?

    def write(self,text):
       raise NotImplementedError


Description of NotImplementedError from "built-in exceptions" section
of library doc:

    exception NotImplementedError

    This exception is derived from RuntimeError. In user defined base
    classes, abstract methods should raise this exception when they
    require derived classes to override the method. New in version 1.5.2.


From Moo at houston.rr.com  Mon Feb 18 17:49:02 2002
From: Moo at houston.rr.com (Avi Homes)
Date: Mon, 18 Feb 2002 22:49:02 GMT
Subject: New to python
Message-ID: <yxfc8.145803$Ar5.5568410@typhoon.austin.rr.com>

Having problems with dictionaries..

I have one setup i.e.

---
Items{ 'IName': "empty", 'IDesc': "empty", 'IPrice': "empty"}
temp=raw_input("Whats the name?")
Items('IName')=temp
--

and it is not liking the 3rd line of code.. I have been a c++ coder for a
few years, just started looking at python yesterday, but this simple problem
is kinda eluding me.  temp should hold the value for whatever the user input
was, correct? this is being done in pythonwin btw. and raw_input should make
it so the user doesn't have to say "this is my input" and have to include
the quotes..

thanks.
Avi




From mwh at python.net  Thu Feb  7 10:25:25 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 15:25:25 GMT
Subject: can't find python.h
References: <QPw88.109812$h31.7614605@e420r-atl1.usenetserver.com>
Message-ID: <usn8d8g7b.fsf@python.net>

"Bill Kreamer" <kreamer at 954access.net> writes:

> I have Python1.5 on Red Hat 7.0 Linux. I also installed python_devel. I am
> trying to link Python with C. According to instructions, I have to put
> #include <python.h> is my C language program. gcc says that python.h does
> not exist, and I did a thorough search and verified that this is true. Have
> I missed something?

It's Python.h.

Cheers,
M.

-- 
  It's a measure of how much I love Python that I moved to VA, where
  if things don't work out Guido will buy a plantation and put us to
  work harvesting peanuts instead.     -- Tim Peters, comp.lang.python


From aahz at panix.com  Mon Feb 11 17:25:56 2002
From: aahz at panix.com (Aahz Maruch)
Date: 11 Feb 2002 14:25:56 -0800
Subject: Google/Python (was Re: Dr. Dobb's Python-URL! - weekly Python news and links (Feb 11))
References: <65D322D62049DED1.100EC0150B20FD6E.E4912BB6F9A2A65D@lp.airnews.net>
Message-ID: <a49gdk$cjc$1@panix2.panix.com>

In article <65D322D62049DED1.100EC0150B20FD6E.E4912BB6F9A2A65D at lp.airnews.net>,
Jason Orendorff <jason at jorendorff.com> wrote:
>
>    The Google Programming Contest accepts submissions in only three
>    languages: C++, Java, and Python.
>        http://www.google.com/programming-contest/

Oh, *interesting*.  They changed the rules.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From sholden at holdenweb.com  Wed Feb 13 13:27:17 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 13:27:17 -0500
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net> <yuqa8oHpvCEW092yn@visi.com>
Message-ID: <_dya8.209628$KM2.8685845@atlpnn01.usenetserver.com>

"Richard Steiner" <rsteiner at visi.com> wrote in message
news:yuqa8oHpvCEW092yn at visi.com...
> Here in alt.folklore.computers,
> claird at starbase.neosoft.com (Cameron Laird) spake unto us, saying:
>
> >In article <130220020751360934%jwbaxter at spamcop.net>,
> >John W. Baxter <jwbaxter at spamcop.net> wrote:
> > .
> >>Did the elves at AT&T pick up the "leading 0 means octal" from
> >>something earlier, or did they invent this stupidity themselves?
> > .
> >OK, folks, how far back *can* we trace this?  My
> >mind associates it with DEC systems, going back
> >to the '60s, but I couldn't find any confirmation
> >of that in a quick search.
>
> I've seen the leading-zero-means-octal convention used rather heavily
> in documentation dated 1967-1968 in a UNIVAC 1106/1108 context.
>
> It seems to have been a fairly well-established convention at that
> point in time, at least in the UNIVAC world.
>
Indeed, this was also true of the UNIVAC 418 which predated the 1100 series.
Ah me, ah my, FastRand drums. One of the engineers told me that they turned
sewer pipes down on a precision lathe to make those huge horizontal drums.
Certainly impressive to look at, but less fun when you had to hire a crane
to deliver one through a fourth floor window.

[drools into beard]

those-were-the-days-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From cbbrowne at acm.org  Wed Feb  6 02:05:05 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Wed, 06 Feb 2002 02:05:05 -0500
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net> <3c60c05c$0$79564$3c090ad1@news.plethora.net>
Message-ID: <m3bsf3hyzi.fsf@chvatal.cbbrowne.com>

The world rejoiced as seebs at plethora.net (Peter Seebach) wrote:
> In article <1372.801T1070T13994028threeseas at earthlink.net>,
> Timothy Rue <threeseas at earthlink.net> wrote:
>>This way I don't have to try and figure out why you have a problem knowing
>>the difference between what is idea and what is natural law physics. For
>>MS seems to think it own reality. It doesn't!
>
> This isn't about ownership, it's about finding the community most naturally
> suited to your project.  The C# people are the community least contaminated
> by existing notions of software engineering.

Oh, my.  Sounds like the Pythonesque "cheese shop" that was completely
uncontaminated by cheese...

If Rue's into Natural Law, then he really should be talking with Doug
Henning and his Yogic Flyers, right?

_Natural Law deputy floats in for visit_
<http://www.rickross.com/reference/tm/tm16.html>

Apparently, having 500 yogic flyers in Parliament (apparently not the
_Canadian_ Parliament, unless Natural Law involves ignoring numbers,
since there aren't 500 seats for parliamentarians) would "guarantee
world peace."
-- 
(concatenate 'string "cbbrowne" "@canada.com")
http://www3.sympatico.ca/cbbrowne/wp.html
Visualize whirled peas.


From mal at lemburg.com  Tue Feb 19 09:39:57 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 19 Feb 2002 15:39:57 +0100
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi> <23891c90.0202190434.147ee6d1@posting.google.com> <7x7kp9zkuw.fsf@ruckus.brouhaha.com>
Message-ID: <3C7263BD.A8C7D1AA@lemburg.com>

Paul Rubin wrote:
> 
> paul at boddie.net (Paul Boddie) writes:
> > You could always make a SWIG wrapper for iODBC - I started one to see
> > how hard it might be - but in the end it's probably cheaper (time is
> > money, so they say) to license mxODBC.
> 
> I hadn't heard of iODBC before, but just found the home page www.iodbc.org
> and it mentions that a Python bridge (whatever that is) already exists.

That's what mxODBC is... :-)

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From mwh at python.net  Mon Feb  4 05:51:06 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 4 Feb 2002 10:51:06 GMT
Subject: Equivalent of Perl chomp?
References: <mailman.1012626605.22240.python-list@python.org>
Message-ID: <upu3lr007.fsf@python.net>

"Tim Peters" <tim.one at home.com> writes:

> A few things going on in this test under the covers:
> 
> + endswith does indeed pay a large price for building a bound
>   method object every time.  It's not only the dict lookup to
>   resolve "endswith", but also malloc and free to build the
>   bound method object and discard it afterwards.

Method objects have a free list, so you're not really paying for a
malloc/free most of the time.  Actually, it looks like this free list
can only grow, so you'll never pay for the free...

Cheers,
M.

-- 
  We've had a lot of problems going from glibc 2.0 to glibc 2.1.
  People claim binary compatibility.  Except for functions they
  don't like.                       -- Peter Van Eynde, comp.lang.lisp


From gcordova at hebmex.com  Wed Feb  6 16:55:37 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 6 Feb 2002 15:55:37 -0600 
Subject: Is there anyway to obtain client's IP address using cgi modul
	e?
Message-ID: <19650FF08A99D511B19E00B0D0F0632301898FFA@SRM201002>

Check the $REMOTE_ADDR header, I believe.

Good luck.

-gus



From chad.maine at broadlink.com  Thu Feb 28 12:50:55 2002
From: chad.maine at broadlink.com (Chad Maine)
Date: Thu, 28 Feb 2002 17:50:55 GMT
Subject: Class functions
References: <NVSe8.151$en5.1429@typhoon.sonic.net>
Message-ID: <36uf8.758$en5.6052@typhoon.sonic.net>

I was able to solve the problem using generators and replacing the StringIO
file object with a simple yield.  Fortunately, we had already moved our
development to Python 2.2 for other reasons.

Thanks to Jason Orendorff for the tip.


"Chad Maine" <chad.maine at broadlink.com> wrote in message
news:NVSe8.151$en5.1429 at typhoon.sonic.net...
> I'd like to access the data updated in real time via a class' function.
> I've got something like this:
>
> the class:
>
> def CLASS:
>     def __init__...
>
>     def __del__...
>
>     def FUNC(self):
>         while 1:
>             self.DATA = StringIO.StringIO(UPDATE)
>             if self.DATA == '': break
>         self.DATA.close()
>
> in my client script:
>
> import theclass
> X = theclass.CLASS
> X.FUNC()
> while 1:
>     try:
>         DATA = self.DATA
>         if not DATA: break
>         print DATA.read()
>     except ValueError:
>         break
>
>
>




From sholden at holdenweb.com  Thu Feb  7 17:36:59 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 17:36:59 -0500
Subject: Defending the Python lanuage...
References: <mailman.1013116393.10397.python-list@python.org>
Message-ID: <3kD88.97035$KM2.3634517@atlpnn01.usenetserver.com>

"Michael Chermside" <mcherm at destiny.com> wrote in message
news:mailman.1013116393.10397.python-list at python.org...
> >> Really? What an unusual idea - I didn't realise we had such a surplus
of
> >> programmers to advocate something like this :-)
> >
> > This is a little more "extreme" than I was referring to as well.  The
idea
> > of two programmers sitting at the same computer doesn't really go over
well
> > with me... I was referring having two programmers sit side-by-side at
> > separate computers, working on different parts of the same program
versus
> > having one programmer work in isolation.  Indeed, I don't know many
> > programmers with the patience to sit and watch someone else type ;)
>
>
> Ermmm.... No.
>
> When XP or Agile Programming advocates talk about "pair programming"
> they really DO mean one keyboard, two programmers.
>
> Peter Milliken writes that he didn't know we had such a surplus of
> programmers -- this is a common misconception. Suppose that, working on
> your own, you get X amount of coding done in a day. If we could spare a
> second programmer to stand there and just help you out -- someone to
> bounce ideas off of, to watch for typos, to suggest things that you had
> forgotten -- then you'd get MORE done... say X * k.
>
[how much more productive is it?]
>
> As for Cliff Wells' contention that he doesn't know many programmers who
> would want to do this, I have to differ. There are clearly some
> programmers who absolutely hate working in pairs, but there are also
> many others who LOVE it.
>
> Now I just wish I could convince my company to consider doing pair
> programming. Because for *ME* personally, k is around 3 or 4.
>
Well, you could resign twice and let them notice the huge productivity loss
...

As a "singleton" (I tend to work as a one-man team) what I'm looking for is
the NetMeeting of the pair programming world. So I can pair-program with a
colleague on the net rather than make mistakes on my own. I really enjoy
pair programming, but get to indulge in it far too rarely.

Although there are sometimes advantages in immediate productivity, even a
modest decline in output is rapidly compensated by reductions in program
defects for XP, particularly since addressing them later is hugely more
expensive. I'm *very* interested in anything that reduces defect rates

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From theller at python.net  Thu Feb 21 06:53:10 2002
From: theller at python.net (Thomas Heller)
Date: Thu, 21 Feb 2002 12:53:10 +0100
Subject: Reloading modules
References: <a51tt1$msq$1@news1.ucsd.edu>
Message-ID: <a52n45$46617$1@ID-59885.news.dfncis.de>

"Jeff Davis" <jdavis at empires.org> wrote in message news:a51tt1$msq$1 at news1.ucsd.edu...
> Is there a way to say that all calls to import should check for a newer
> version, if a newer one exists then completely drop the old one and import
> the new one?
>
> I am aware of the reload() function, but it has enough caveats that it
> always leaves lingering questions. I usually end up just stopping and
> restarting the interpreter, which slows development/debugging. I would
> just like to be sure that any time after I import a module, I am using the
> newest version available at the time of import. I am also unsure as to
> whether modules imported within a reloaded module are also reloaded.
>
I once posted a module which does this:

http://mail.python.org/pipermail/python-list/2001-November/072669.html

HTH,

Thomas




From phr-n2002a at nightsong.com  Mon Feb  4 09:57:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 06:57:34 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com>
Message-ID: <7xzo2pqopt.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> > If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> > it isn't backwards compatible.
> >
> Erm, don't you seem to be using a different version of backwards
> compatibilty from most people? As far as I'm concerned, if I run version
> 1.5.2 of a language, and all my programs continue to run, then the new
> release provides backwards compatibility.

Yes, but they DON'T continue to run, then the new release DOESN'T provide
backwards compatibility.

If I write in 2.2 or later 2.x's, I have to choose between two
alternatives, both bad:

 1) use "/" the traditional way.  My programs will then work under 2.1
    and earlier, but they'll break under 3.0 and later.

 2) Use "from __future__ import division" and use // everywhere I
    want traditional integer division.  That means that in the Python
    versions that most people today are actually using, my programs
    won't even parse, let alone run correctly.

> > I know this, but 2.1.1 was released more recently than 2.2, so the most
> > recent release was in the 2.1 series.
>
> Now you are clearly looking for a fight :-) This is bogus. "Current"
> surely refers to "processes the most recent language
> definition". Tim's right, this is rhetorical gimmickry. How
> appropriate your news server is ruckus.brouhaha.com.

I don't agree with you about what "current" surely refers to.  In my
view, "current" refers to whatever is in widespread use at the present
time.  Secondarily, it refers to what the developers are still
maintaining and shipping, which includes 2.1 and (apparently maybe) 1.5.

> > If the official Python maintainers think they might release a 1.5.3
> > someday, then it's fair to say that the 1.5 series is still current.
> 
> If you ever collected stamps, you'd be aware that stamps can be "obsolete"
> (no longer issued) and they may also be "demonetized" (no longer acceptable
> as postage). In the same way, 1.5.2's survival in various Linux
> distributions does not stop it being obsolete. Fortunately it will only be
> "demonetized" (unusable for programming) when there is no longer a platform
> that will allow you to compile to interpreter sources. The open source
> nature of the language means you can continue using 1.5.2 into the far
> future.

But Tim has said there might be a 1.5.3 release.  That means that the
1.5 series is still being "issued" and is not obsolete.  The 2.1
series is definitely still being issued--a new 2.1 release came out
just a few weeks ago, and I believe Tim himself recommended it as the
most stable Python version currently available.  Why on earth would
anyone develop a new 2.1 bugfix release if they thought everyone
should switch to 2.2?

> > If they're unlikely to ever release another 1.2.x or 0.9.x, then
> > those versions are not current, even if I release a 1.2.x myself.
> >
> Nonsense. Do you understand the meaning of the word "currency"? And stop
> relying on the "official Python maintainers" - one of these is Tim Peters,
> and you continue to disagree with him about this matter...

Tim is the one who said there might be a 1.5.3 someday, which is part
of why I refer to the 1.5 series as still current, just like the 2.1
series.  Of course he said it like it (1.5.3) was just a theoretical
possibility rather than something actually likely to happen, so I
won't hold out for that part of "current".  The other part--that lots
of systems still run it--is valid and compelling.

> Coding in a way that requires 2.2 inconveniences *only* those users who a)
> want to use my code *and* b) won't move forward to 2.2.

Even if they're willing to switch to 2.2, making them do so
inconveniences them.  But why should someone be willing to switch to
2.2 today, unless they need the new features for something?  It has a
lot of major rework and I'd rather let the maintainers get the bugs
out before I run production code under it.  I might play with 2.2 at
home, but if I was installing a new production system today, I'd use
2.1.1.

If, someday, there are just a few throwbacks using <= 2.1 (say less
than 10% of all the users), then it will be reasonable to casually use
2.2 features in programs for general distribution.  Right now, it's
probably more like 90% using <= 2.1.  Publishing a script that
requires 2.2 is like publishing a web page that only works with IE6
(or whatever the latest browser is).  It's simply obnoxious to do that
without a good reason.

> I suppose that our writing these posts in English inconveniences
> many of the group's readers for whom English is not the first
> language. They persist in reading them because the benefits outweigh
> the disadvantages (in most cases, anyway, though I'm beginning to
> have my doubts about this particular thread). So, do you want 2.2,
> or 1.5.2. And please *don't* say you want to be able to program 2.2
> features in 1.5.2.

I don't want to use 2.2 features in 1.5.2.  I also don't always get to
choose what version I run, and I almost never get to choose what
version other people run.  On my personal machine I use 2.1.  On
Ruckus (which isn't mine), 1.5.2 is installed and I use that.  Some
other machines I use have even older versions (1.2 or 1.3 from old
Debian releases).

> I still don't see why you want to be able to use "from __future__" in 1.5.2
> programs. What's the point of importing future behavior when you can if you
> also want to run in environments where such behavior isn't available? 

I don't think I asked to use from __future__ in 1.5.2.  Maybe you
didn't understand my request.  I wanted to put something like

   try: from __backward_compatibility__ import old_division
   except ImportError: pass

in all my current scripts.  This is perfectly fine and won't break
anything in 1.5.2 and 2.1.x.  The import will fail and the caught
exception will turn it into a do-nothing.  The request is that in 3.0,
the import should succeed and make my scripts keep working the same
way the did under older versions.

> > I'd non-rhetorically say that for most users, 1.5.2 or 2.1 is current
> > and 2.2 is bleeding edge.  (I'm still using 2.1 and plan to get around
> > to upgrading to 2.1.1).
> >
> This *is* purely rhetoric, despite your protestations, until you provide
> evidence.

The evidence is that 2.2 has been out (non-beta) for only about 1 month,
while the older versions have been around for years.  Lots more people
are using the older versions.

> So, what about 1.5.2 and 2.1 programs won't run in 2.2? THOSE features are
> the ones that break backwards compatibility (isn't this where we came in?)
> not the new ones that won't run in older versions.

That is a problem too.

> No. It has been PEP'd for a long time that division breaks in 3.0+,
> and you'd better get used to the idea if you want to move forward
> [in a year or two].

I don't agree with the notion of changing the meaning of basic arithmetic
operators in a supposedly mature language, but at least I can understand
the motivation for it.  I don't see at all why some mechanism can't be
provided to turn off the change ("import from past") just like there's
now a mechanism to turn on the change ("import from future").

> > I want my programs to run the same way across the different Python
> > versions that people are using now and will be using in the near future.
> > Am I really the first person to want that?
> >
> No, you're not even the first person to be so wrong-headed in the way you
> decry the chosen path for the language's development <0.8 wink>. Most unlike
> you, if I might make a personal observation. Just the same, I claim that
> Python supports a *remarkably high* degree of backwards compatibility. Very
> few 1.5.2 programs broke in 2.0, and most of those (from memory) were using
> a non-documented "feature" of the sockets library.

The rand module also disappeared, completely gratuitously as far as I
can tell.  That's what broke the Redhat scripts.

> Hey, under 3.0 you'll be able to redefine __div__ for everything so you get
> exactly the behavior you want. Not sure I'd like to maintain that code,
> though ;-)

Is that for real?  Is there a bad performance hit?  If that makes a
way to keep my scripts working without slowing them down too much,
that might be a reasonable solution.


From emile at fenx.com  Mon Feb 25 07:48:18 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 04:48:18 -0800
Subject: Numeric arrays - initialising with binary data
References: <cf0ad9fb.0202250446.3df07a08@posting.google.com>
Message-ID: <a5dc4f$6dr9c$1@ID-11957.news.dfncis.de>

"Jesper Olsen" <jolsen at mailme.dk> wrote in message
news:cf0ad9fb.0202250446.3df07a08 at posting.google.com...
> Hi,
>
> The build in array type has the methods
>
> array.tostring()
> array.fromstring()
>
> which make it easy to initialise arrays with binary data
> read from a file.
>
> The Numeric array, however, apperantly only has the tostring method.
>

Numeric 20.1.0 has fromstring.


--

Emile van Sebille
emile at fenx.com

---------



From theller at python.net  Tue Feb 19 14:01:04 2002
From: theller at python.net (Thomas Heller)
Date: Tue, 19 Feb 2002 20:01:04 +0100
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com> <a4rsa2$2grrk$1@ID-59885.news.dfncis.de> <a4smoa$h20$0@216.39.172.122>
Message-ID: <a4u7ec$36637$1@ID-59885.news.dfncis.de>

"Bengt Richter" <bokr at oz.net> wrote in message news:a4smoa$h20$0 at 216.39.172.122...
> >The source code IS available:
> >http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/distutils/
> >in the misc directory. Guess I'll have to add a note about this
> >to the source... There is a tricky symbolic link somewhere on SF
> >which links parts of the distutils tree into the Python source tree.
>
> That self-update of the bin64 EXEDATA string seems a little tricky too ;-)
> I guess I would rename the old version to a backup name before writing
> the new, not being aware of the larger context. Or should one
> copy it to ../otherdir and run ../otherdir/it that way? Or did I read
> too fast?

CVS gives enough backup for me, and it has worked for me so far.

>
> >
> >> Having some additional hooks created in distutils to execute
> >> post-installation sounds like it could be a useful enhancement.  At
> >> least with that sort of scenario you could write your own shortcut
> >> creating code without modifying the core distutils itself.
> >
> >I'm currently working on an bdist_wininst enhancement which
> >takes an additional command line option (--install-script <script.py>).
> >This script would then be run (with sys.argv set to [script_name, '--installed'])
> >after all the files have been copied, and it would again be run
> >(with sys.argv set to [script_name, '--removed']) just before
> >the deinstaller is about to remove all the files and directories.
> >
> Sounds good. I wonder if something is needed to cater to version upgrading
> and/or multiple version installation. E.g., to give an old version a
> chance to have a script automatically executed '--before_upgrade' and
> '--after_upgrade'? Can the binary installer run something before even
> copying any new files? I.e., the old version interpreter and its
> 'im_about_to_have_a_sibling.py' script? Putting it in the installer
> sequence would protect against user forgetfulness (otherwise s/he could
> obviously do it manually), and an old version could at a minimum output
> some attention-grabbing reminders from the authors of the old version.

Well, I want to keep the C-coded part as simple as possible.
On the other hand, you could play any tricks you want in
the installation script. Detect upgrade yourself!

Thomas




From phr-n2002a at nightsong.com  Wed Feb 20 00:45:56 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 21:45:56 -0800
Subject: Editable strings anyone?
References: <mailman.1014178114.23674.python-list@python.org>
Message-ID: <7xsn7w904r.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> Wow, you're right, this is a mess.  I will put together a patch
> for it.  Quite a few little things are involved, but it's mostly
> just a modernization of the way this module uses the Python C API.
> 
> That's not the kind of thing that's likely to be really
> well-documented anywhere, unfortunately.  :P

If you're going to patch up that module, I hope you can also make it
support multiple segments in the buffer interface, at least with 2
segments, to support editable strings.


From tdickenson at devmail.geminidataloggers.co.uk  Wed Feb 20 03:25:55 2002
From: tdickenson at devmail.geminidataloggers.co.uk (Toby Dickenson)
Date: Wed, 20 Feb 2002 08:25:55 +0000
Subject: print and unicode strings
References: <mailman.1014174755.8438.python-list@python.org> <m38z9o6347.fsf@mira.informatik.hu-berlin.de>
Message-ID: <n3n67ucp307imfivlcb4kn2ooe4t57l75c@4ax.com>

martin at v.loewis.de (Martin v. Loewis) wrote:

>"Jason Orendorff" <jason at jorendorff.com> writes:
>
>> I think IDLE should do this by default.  It's fine for
>> Python to "refuse to guess" encodings, but in the case of Tk
>> it's UTF-8 on all platforms... right?
>
>Well, not exactly. Tkinter supports Unicode strings as primary data
>type for text, with byte strings (e.g. UTF-8) only being the second
>choice. So IDLE should set sys.stdout to a stream that accepts Unicode
>objects as-is.

That still wont fly, because 'print' always passes the to-be-printed
objects through str() before calling file.write()

Here is a patch (now sadly rejected) that I used for a while:

http://sourceforge.net/tracker/?group_id=5470&atid=305470&func=detail&aid=403685



Toby Dickenson
tdickenson at geminidataloggers.com


From deathtospam43423 at altavista.com  Mon Feb 11 16:34:12 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 11 Feb 2002 21:34:12 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org>
Message-ID: <Xns91B352E74A030AusYourStandingInIt@130.133.1.4>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in 
news:mailman.1013350037.30507.python-list at python.org:

> ----- Original Message ----- 
> From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>
> 
> 
>> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
>> news:mailman.1013316023.10782.python-list at python.org: 
>> 
>> > "PERL type CGI's" should be read "scripts".  You can't run Python,
>> > Perl, Ruby,
>> > Lua, etc. from this particular server.
>> > 
>> > Try Xitami.  It is quite compatible with Python.  http://xitami.com
>> > 
>> Thanx. I'm downloading it as we speak. 
> 
> I have used it a little; Steve Holden wrote about it in his book which
> I just finished reading.
>  
>> So with this server, will I be able to just send info direct from my 
>> browser form to a file? If I can do that, then I'm up and running. 
> 
> I'm not sure what you are saying here.  You don't send info "to a 
> file," you send it to a CGI script.  Your script must handle the data 
> from there.
> 
>> [snip]
>> > 
>> > The form looks good, but where is script.cgi?
>> > 
>> In the same directory as the html form, cgi-bin.
> 
> Ah... what I meant was, that you should *post* the source to 
> script.cgi.  Anyway, if your form is HTML it is bad form to have it in 
> the cgi-bin folder.
> 
> GENERALLY (though it's not real safe to say this) CGI scripts can be 
> stored in the cgi-bin folder, or in the htdocs/html/webroot folder; 
> this depends on webserver settings.  The Xitami default allows CGI 
> scripts only in the cgi-bin folder, but you can define that cgi-bin is 
> actually the same as webroot.  Xitami recognizes executable files by 
> extension on Windows, so your script should likely be called script.py 
> (assuming it is Python).
> 
> Gah.  So.  Put script.py in the Xitami cgi-bin folder, and your form in 
> the Xitami webpages folder, and if you are using the default 
> configuration, you should be in good shape.  Start Xitami and then surf 
> to http://localhost/yourform.html (whatever you called it) and when you 
> click the Submit button, the form data is sent to 
> http://localhost/cgi-bin/script.py (which should be in the action= 
> attribute of the form tag, perhaps abbreviated as /cgi-bin/script.py).  
> The script receives the data, does whatever, and then must send a new 
> page to the web browser... this is why I want to see your script.
> 
> 
> 
> 
Thanx. I'm working out how to write a script at the moment, following the 
tutorial at http://www.devshed.com/Server_Side/Python/CGI/ . May I get back 
to you if I have any difficulties? 


-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From mill at pedgr571.sn.umu.se  Mon Feb  4 07:08:14 2002
From: mill at pedgr571.sn.umu.se (Olof Oberg)
Date: Mon, 4 Feb 2002 12:08:14 +0000 (UTC)
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
Message-ID: <slrna5sudd.uef.mill@pedgr571.sn.umu.se>

On Mon, 04 Feb 2002 00:01:09 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
> In reviewing over many messages in the autocoding and related threads and
> seeing the evidence of contridictions, hypocracy, irrational and
> illogical responses......
Good thing you are coming to your senses.

> Its damn clear that there is a need to remove the sort of people posing to
> be professional who are in essence just good liars, abstraction
> manipulators, from the industry.
You are working in the industry?

> Who knows, maybe that stench will be enough to motivate me see that it
> gets done, gets removed.
Take a shower.

> Maybe I should be more appriciative of the stench if it so motivates me.
>
> Anyone want to add anything to this?
Yes. Are you going to remove yourself from these newsgroups?

I think you should be proud of yourself that managed to lure 
(Flying Lure(tm)?) so many people into actually trying to 
understand your ramblings. I mean no one (speak up if I am 
wrong) really understood anything of what you were talking about, 
but still tried to find out if you had something worthwhile to 
say.

 /mill

-- 
#############################################################
# millNO at SPAMludd.luth.se # http://pedgr571.sn.umu.se/~mill #
#############################################################


From kreamer at 954access.net  Thu Feb  7 10:12:53 2002
From: kreamer at 954access.net (Bill Kreamer)
Date: Thu, 7 Feb 2002 10:12:53 -0500
Subject: can't find python.h
Message-ID: <QPw88.109812$h31.7614605@e420r-atl1.usenetserver.com>

I have Python1.5 on Red Hat 7.0 Linux. I also installed python_devel. I am
trying to link Python with C. According to instructions, I have to put
#include <python.h> is my C language program. gcc says that python.h does
not exist, and I did a thorough search and verified that this is true. Have
I missed something?





From tim.one at home.com  Fri Feb  1 23:50:12 2002
From: tim.one at home.com (Tim Peters)
Date: Fri, 1 Feb 2002 23:50:12 -0500
Subject: Windows 2000
In-Reply-To: <01C1AB52.89FFDF80@BRIANMIT>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEFNNJAA.tim.one@home.com>

[Brian Mitchell]
> I tried an install of Python 2.2 on a Windows2000 box today. IDLE
> seemed to function okay but I couldn't close the command mode
> with control-Z.  Is there anything special about using Python under
> Windows2000 as compared to Windows98?  TIA.

Shouldn't be, and AFAIK Ctrl-Z never did anything in any release of IDLE
under any version of Windows except get interpreted by Tk as a command to
"erase backwards".  Perhaps you were using a customized IDLE before, or are
remembering how the DOS-box Python works (where Ctrl-Z is interpreted by
Microsoft's input routines-- not Tk's -- as an "end of file" signal).  You
can use Ctrl-D or Ctrl-Q under IDLE to get out (the former closes just the
shell window, the latter closes all IDLE windows).




From guy at wyrdrune.com  Sun Feb  3 03:25:14 2002
From: guy at wyrdrune.com (Guy Gascoigne - Piggford)
Date: Sun, 3 Feb 2002 00:25:14 -0800
Subject: Expect like syntax in Python
Message-ID: <3c5cf3d6_2@news.nwlink.com>

I'm trying to find a way to write something like expect in a python module,
and I'm having trouble getting the syntax the way that I want.  In expect I
can write something like this:

expect {
    -re "ast login" {
           pass
           }
    -re "ogin" {
           send  user
           }
    -re "sword" {
           send password
           }
}

etc.

The nearest that I've managed to get to in Python is something like this:

        while 1:
            try:
                if self.expect( ( "ast login", 0 ),
                                ( "ogin", self.send_user ),
                                ( "sword", self.send_password ),
                                ( "\$", "done" ) ) == "done":
                    break
            except IndexError:
                skipcount += 1
                if skipcount == 2:
                    raise IndexError

Where send_user and send_password are separately defined functions.  What I
really want to be able to do is to define the action beside the pattern,
calling short functions that are defined separately makes it harder to read
and trickier to follow exactly what is happening.  I've also played around
with just returning an index, but that also gets ugly as the list of
possible patterns gets larger.

Am I missing something?  I'm fairly new to python but I don't see a way to
get what I want.  Does anyone have any suggestions on how to handle this
more cleanly?

--
Guy

--
Guy Gascoigne - Piggford - www.wyrdrune.com
"Sometimes verbal ingenuity is not enough" - B. Banzai





From spencer at spencerdoidge.com  Thu Feb  7 15:42:37 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Thu, 7 Feb 2002 12:42:37 -0800
Subject: import/reload problem
Message-ID: <a3uol3$hmr$1@news.efn.org>

Running pythonwin.exe on my Win2k, in the edit window, I make a change to my
script, then check syntax. When all is well, I click on "Import/Reload." The
program indicates success, but the changes don't get implemented. What am I
doing wrong?




From fgeiger at datec.at  Sat Feb 23 06:55:00 2002
From: fgeiger at datec.at (F. GEIGER)
Date: Sat, 23 Feb 2002 12:55:00 +0100
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <a57vtj$8vq$1@rex.ip-plus.net>

a= abs(a)

Cheers Franz

"Tripp Scott" <tripps81 at yahoo.com> schrieb im Newsbeitrag
news:mailman.1014452738.24147.python-list at python.org...
> Suppose I want to make a function that modifies its argument:
>
>   a = -1
>   make_abs(a)
>   print a # = 1
>
> What is the Pythonic way of doing this? Wrap it in a list?
>
>   make_abs([a])
>
> t
>
>




From cliechti at gmx.net  Fri Feb 15 19:52:20 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 16 Feb 2002 01:52:20 +0100
Subject: thread question
References: <mailman.1013817865.2645.python-list@python.org>
Message-ID: <Xns91B713978E478cliechtigmxnet@62.2.16.82>

"maximilianscherr" <MaximilianScherr at T-Online.de> wrote in 
news:mailman.1013817865.2645.python-list at python.org:

> if a thread (object) is for example a class, would it be possible to 
> call the class objects of this thread from another thread?
> or how could something similar be done?

thread live in the same process (same memory) so you can communicate 
through variables/attributes/calls etc.

BUT: complex data types have to be protected from simultaneous access 
from multiple threads at the same time. that's where threading.Lock 
and friends enter the game.

chris

auch noch an einer nachtschicht? :-)

-- 
Chris <cliechti at gmx.net>



From tim.one at comcast.net  Thu Feb 21 15:52:49 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 21 Feb 2002 15:52:49 -0500
Subject: Stupid Lambda Tricks [LONG]
In-Reply-To: <3c74c211.20012718@news.isar.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEFFNOAA.tim.one@comcast.net>

[Gerson Kurz]
> Note: the official URL for the following is
> http://www.p-nand-q.com/lambda.htm
> 
> -----------------------------------------------------
> Stupid Lambda Tricks
> ...

Dear Lord!  The things you could accomplish if you weren't insane <wink>.

it'll-come-back-to-haunt-you-ly y'rs  - tim



From aahz at panix.com  Thu Feb 28 00:59:41 2002
From: aahz at panix.com (Aahz Maruch)
Date: 27 Feb 2002 21:59:41 -0800
Subject: How to have multi-threaded server log to one file ?
References: <97ae44ee.0202261855.15c97708@posting.google.com> <f72dac0d.0202270140.97b925f@posting.google.com> <97ae44ee.0202272013.2094b2b@posting.google.com>
Message-ID: <a5kh0d$hu6$1@panix2.panix.com>

In article <97ae44ee.0202272013.2094b2b at posting.google.com>,
Stephen <shriek at gmx.co.uk> wrote:
>
>I had a look at the effbot queue-example and think I understand the
>Queue class but adapting my program to share the queue instance in the
>first place is making my head spin.  Namely, how do I get the same
>Queue object into both MyHandler and MyServer in the first place ?

Pass it in.  See some examples at http://starship.python.net/crew/aahz/
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From phr-n2002a at nightsong.com  Mon Feb 25 00:59:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 24 Feb 2002 21:59:51 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com> <a5cjc0$cb5$0@216.39.172.122>
Message-ID: <7x7kp2ayp4.fsf@ruckus.brouhaha.com>

bokr at oz.net (Bengt Richter) writes:
> but I didn't get an extra call trying your version:
> 
>  >>> reduce(lambda x,y:(y.f(),0)[1], items, 0)
>  this is a
>  this is b
>  this is c
>  0

Didn't you?  'items' had three elements, but four values got printed.

> you can use the 'and 0' ploy to get a zero length
> list comprehension also:
> 
>  >>> [ 0 for x in items if x.f() and 0]
>  this is a
>  this is b
>  this is c
>  []

Nice!  That's the best solutions so far.


From tanzer at swing.co.at  Thu Feb  7 03:18:03 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Thu, 07 Feb 2002 09:18:03 +0100
Subject: What is the command to determine an object's type?
In-Reply-To: Your message of "Wed, 06 Feb 2002 20:31:44 +0100."
             <bm036ukfdun2it79e91sllcjf2n909242d@4ax.com>
Message-ID: <E16Yjkp-0007HT-00@tswings.swing.cluster>

Tino Lange <tl_news at nexgo.de> wrote:

> On Tue, 05 Feb 2002 22:38:49 GMT, "Fredrik Lundh"
> <fredrik at pythonware.com> wrote:
>
> Hi Fredrik!
>
> (1) if type(l) == types.ListType:
> >or better:
> (2) if type(l) is types.ListType:
> >or even better:
> (3) if isinstance(l, types.ListType):
>
> Why is "if isinstance(l, types.ListType):" even better than the rest?
> What do you mean by "better"?
>
> I just compared the speed of execution (using Python 2.1 on Linux) and
> found consistently:
> (2) is the fastest, the comes (3), (1) is much slower.

This has nothing to do with speed [*]. (3) will work for subtypes of
list, too -- by using (2) you are breaking polymorphism.

As others already pointed out, even (3) is generally a bad idea (TM).
Using any kind of type checking reduces the reusability of your code.
While (3) is definitely better than (1) or (2), it still breaks
signature based polymorphism.

[*] As Qualtinger said in `Da W??de auf sana Maschin`: `I woa? zwoa n?d
    wohin, oba dafur bin i schn?lla durt` <wink>

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From tick at tut.by  Thu Feb 14 10:38:34 2002
From: tick at tut.by (tick)
Date: 14 Feb 2002 07:38:34 -0800
Subject: ftp binary mode
Message-ID: <717c3e15.0202140738.111f9a71@posting.google.com>

Hi,

I want to use python function ftplib.retrbinary() for transfering the
text data to a file.  After transfering every 512 bytes <NL>
separates.

How to increase this limit?

Best regards


From just at xs4all.nl  Fri Feb 15 03:31:19 2002
From: just at xs4all.nl (Just van Rossum)
Date: Fri, 15 Feb 2002 09:31:19 +0100
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net> <mailman.1013718563.21631.python-list@python.org> <5e8bd451.0202142238.580f15c1@posting.google.com> <7xzo2bnt1m.fsf@ruckus.brouhaha.com>
Message-ID: <just-26D236.09311915022002@news1.xs4all.nl>

In article <7xzo2bnt1m.fsf at ruckus.brouhaha.com>,
 Paul Rubin <phr-n2002a at nightsong.com> wrote:

> kp87 at lycos.com (kevin parks) writes:
> > How would you sort something based on 2 criteria. Say you have a list
> > of lists (i hate tuples). first thing you want is everything sorted by
> > the first value and then everything that has the same value for [0]
> > then sorted by [1]
> 
> # try this:
> 
> yourlist =  [['i3', 24.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5], \
>              ['i8', 82.25, 0.75, 101, 7, 6, 5, 0, 0, 0, 0, 0.02, 2], \
>              ['i9', 9, 0, 255], \
>              ['i8', 83.0, 1.0, 101, 1, 1, 5, 0, 0, 0, 0, 0.02, 2]]
> 
> def compare_lists(x,y):
>   a = cmp(x[0],y[0])
>   if a: return a
>   return cmp(x[1],y[1])
> 
> yourlist.sort(compare_lists)

Nah, just do this:

  yourlist.sort()

Just


From mwh at python.net  Tue Feb  5 10:19:01 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 15:19:01 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de>
Message-ID: <uheowosxs.fsf@python.net>

Simon Budig <simon at ping-pong.budig> writes:

> Michael Hudson wrote:
> > Simon Budig <simon at ping-pong.budig> writes:
> [...]
> >> BTW: The stupid fixed width limit of SF messed the patch up.
> >> Maybe creating an attachment is more useful...
> > 
> > Well, anyone can see roughly what it does.  If it gets accepted (or
> > more likely, noone complains for a couple of weeks), I'll check it in
> > from your patch.
> 
> Great! If you do this, please make sure, that a similiar patch goes
> into python 2.2.

Well, it can't go into 2.2.  If you mean 2.2.1, that's a little
greyer; there might be people out there who rely on the current
behaviour.

Cheers,
M.

-- 
  I sense much distrust in you.  Distrust leads to cynicism, cynicism
  leads to bitterness, bitterness leads to the Awareness Of True
  Reality which is referred to by those-who-lack-enlightenment as
  "paranoia".  I approve.    -- David P. Murphy, alt.sysadmin.recovery


From jason at jorendorff.com  Mon Feb 25 19:52:13 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 18:52:13 -0600
Subject: Dictionary type access to list of instances
In-Reply-To: <3C7A5AD4.9090404@wi.mit.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLEEDEDFAA.jason@jorendorff.com>

Brian Kelley wrote:
> In this usage, getattr will raise an Attribute error if property doesn't 
> exist which might not be what you want.
> 
> Here is an attempt using list comprehensions and the built-in list.index 
> function.  I have created a NotEqual class so that getattr can return a 
> default value is never equal to any other instance.  This is because 
> None might be a valid value for a property.
> 
> class NotEqual: [...]
> 
> def get(inputList, property, value, default=NotEqual()):
>      try:
>          properties = [getattr(x, property, default) for x in inputList]
>          return inputList[properties.index(value)]
>      [...]

Here are some alternatives.  (Each must be wrapped in a for loop.)

  (1)   if hasattr(x, property) and getattr(x, property) == value:
             return x

  (2)   try:
            if getattr(x, property) == value:
                return x
        except AttributeError:
            pass  # x doesn't have that property; skip it

  (3)   if getattr(x, property, not value) == value:
             return x

I'm fond of the last one.  No builtin value will have
"value == not value", and user-defined types will typically
avoid it too.

## Jason Orendorff    http://www.jorendorff.com/



From max at alcyone.com  Wed Feb 20 13:16:14 2002
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 20 Feb 2002 10:16:14 -0800
Subject: Adding static typing to Python
References: <mailman.1014147452.26214.python-list@python.org> <a50g0n$gt4$1@thorium.cix.co.uk>
Message-ID: <3C73E7EE.D0F6DE21@alcyone.com>

gbreed at cix.compulink.co.uk wrote:

> Do you have a reference for this terminology?  I haven't seen it
> outside
> of comp.lang.python.  The only definition I have to hand contradicts
> it:
> 
> <http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=strong+typing>
> 
> "Strict enforcement of type rules with no exceptions. All types are
> known
> at compile time, i.e. are statically bound. With variables that can
> store
> values of more than one type, incorrect type usage can be detected at
> run
> time."

There seems to be some consensus (we had exactly this discussion in
#python on IRC) that Foldoc is inaccurate here (and, unfortunately, is
no longer updated).

Strong typing simply means that objects have an intrinsic notion of type
and that operations on them have different meanings depending on their
types.  Weak typing is the reverse; objects (or variables) do not have
any intrinsic type and the operations that are performed determine the
type.  Obviously there are varying degrees of strong typing; C++ is more
strongly typed than C, for instance, but both are strongly typed. 
Python is strongly typed; Perl would be a good example of a weakly-typed
language (e.g., 2 + 2 and 2 . 2 mean different things because "2"
doesn't have any intrinsic type).

In static typing, however, variables must be declared with a given type
and only objects of that type can be placed in those variables.  Dynamic
typing makes no such restrictions; a variable is a variable and can
refer to any type object, and can even be substituted with a different
type of object.

When strong typing is coupled with dynamic typing, typically it's
referred to as manifest typing.  Python and Common Lisp are examples of
this.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From maug at poczta.onet.pl  Mon Feb 25 17:03:07 2002
From: maug at poczta.onet.pl (Marek Augustyn)
Date: Mon, 25 Feb 2002 23:03:07 +0100
Subject: urllib and proxy
Message-ID: <a5ecas$qiq$1@zeus.polsl.gliwice.pl>

Hello,

Is is possible to disable proxy for some urllib.urlopen() call?
Anxiety of urllib to transparently use proxy server seems to cease only
after disabling proxy use at operating system level (IE settings).

I'm using Windows/Python 2.2.

Thanks in advance,
August




From ruediger.maehl_nospam at web.de  Wed Feb  6 10:06:38 2002
From: ruediger.maehl_nospam at web.de (Ruediger Maehl)
Date: Wed, 6 Feb 2002 16:06:38 +0100
Subject: working with files with the same extenstion
References: <3C5EAF50.80004@rmb.co.za>
Message-ID: <newscache$ab94rg$smg$1@www-neu.dzsh.de>

"Ulrich Pfisterer" <ulrich.pfisterer at rmb.co.za> schrieb im Newsbeitrag
news:3C5EAF50.80004 at rmb.co.za...
> Can anyone please tell me how I would go about doing the following.
> I would like to write a script that would take in a command line
> parameter "*.txt". The script would then loop through every file with a
> "txt" extension and something with it.
>
> I am brand new to python (I started using it 2 days ago) so I dont know
> if there is a module that does already this, or whether I have to do
> all the wildcard matching etc myself.
>
> Any help would be much appreciated

Welcome to Python.

You may prefer perhaps a more old-fashioned style like:

# -----
# file_ext.py
import sys
import os

directory = sys.argv[1]
extension = sys.argv[2]

for fn in os.listdir(directory):
    if extension.lower() == os.path.splitext(fn)[1].lower():
        print fn
# -----

And you would call that using the following command in a
DOS box:

python file_ext.py c:\temp .txt

This program loops through _every_ file in the directory,
but also gives you the possibilty to distinguish files and
directories (which may also have extensions).

HTH,

R?diger




From tjreedy at home.com  Wed Feb 13 11:34:39 2002
From: tjreedy at home.com (Terry Reedy)
Date: Wed, 13 Feb 2002 16:34:39 GMT
Subject: Defending the Python lanuage...
References: <mailman.1013116393.10397.python-list@python.org> <q09c4a.ag3.ln@127.0.0.1>
Message-ID: <zAwa8.41853$X64.14949448@news1.rdc2.pa.home.com>

With regard to team programming and 'k':

The team of authors of Extreme Programming Applied put it this way:
pair programmers should work like a car rally pair: one drives while
the other navigates.  More concretely, one types while the other plans
ahead (rather than sit and do nothing).  The difference from rally
pairs is that programming partners trade roles and even partners.  I
think they would claim that K > 2, at least over a product life
cycles, but it would be interesting to see some data.

Terry J. Reedy





From brueckd at tbye.com  Tue Feb 26 18:28:19 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 26 Feb 2002 15:28:19 -0800 (PST)
Subject: Need advice for obfuscating Python code
In-Reply-To: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
Message-ID: <Pine.LNX.4.33.0202261511020.13167-100000@ns0.tbye.com>

On Tue, 26 Feb 2002, Sandeep Gupta wrote:

> I've written a product in Python and I wanted to find out how best to
> obfuscate the code.  I want the distributed files to be moderately hard to
> reverse engineer.
>
> One option is to compile the code and distribute the .pyo files.  Are there
> any better alternatives?

Decompyle can take bytecode and produce Python source that is really close
to the original, so a good software license might give you better results.
That being said, you can store the bytecode encrypted/compressed in a
bundle like this (untested):

import os, rotor, marshal
key = "ILovePurplePantsDon'tYou?"
archiveName = 'msc21.dll' # Use a name nobody will suspect
dict = {}
for file in filesToArchive: # list of .py/.pyc filenames
  data = open(file, 'rb').read()
  base, ext = os.path.splitext(file)
  if ext == '.py':
    dict[base] = (1, data)
  else:
    dict[base] = (0, data[8:]) # beware the voodoo! (8=size of .pyc hdr)

rot = rotor.newrotor(key)
open(archiveName, 'wb').write(rot.encrypt(marshal.dumps(dict)))

With your archive in place, you can have Python use it for imports:

import imputil, marshal, rotor,sys
imputil.ImportManager().install()
sys.path.insert(0, imputil.BuiltinImporter())

class ArcImp(imputil.Importer):
  def __init__(self, key, arcFile):
    rot = rotor.newrotor(key)
    data = rotor.decrypt(open(arcFile, 'rb').read())
    self.arcDict = marshal.load(data)

  def get_code(self, parent, name, fqname):
    try:
      isSrc, data = self.arcDict[fqname]
      if isSrc:
        return 0, compile(data, fqname, 'exec'), {}
      else:
        return 0, marshal.loads(data), {}
    except KeyError:
      pass # Let default importing happen

sys.path.insert(0, ArcImp(key, arcFile))

That should work. Of course, since the code that decrypts and loads your
code will be visible, you probably want to use a long key (or another form
of encryption) and install the customer importer from a C program that
runs an embedded interpreter (installs the importer and then runs the rest
of your program).

-Dave




From john_coppola_r_s at yahoo.com  Fri Feb 15 21:23:29 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Fri, 15 Feb 2002 18:23:29 -0800 (PST)
Subject: Why Python is like BASIC ...  java "vs" python 
In-Reply-To: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph>
Message-ID: <20020216022329.44582.qmail@web11807.mail.yahoo.com>

OK fellas, I'm gonna break it down for you.  Python is
nothing like BASIC or perl.  Neither is Java like C++.
 I will admit that I used BASIC.  A friend turned me
on to Python.  I never looked back.  You cannot
compare things on basis of how they appear or the
syntax of the language or certain similarities.  You
have to know what problem the language is trying to
solve.  What is its philosophy.

When I look at a string in java I view it in a
completely differently way than I do a Basic string or
a string in C++. Every language is useful at
something.  Cobol is great at file extraction on IBM
mainframes using JCL batch jobs.  Use it for anything
else and it is completely inadequate.  I fully
understand the philosophy of Cobol and know what it
was useful for and the problem the language was
attempting to solve.

You will begin to see that basic and python are
nothing alike even though they are both interpreted. 
In what way is basic interpreted.  In what way is perl
interpreted?  Thats like saying a fish is a whale
because they both swim in water.  Java is not at all
similar to C++.  Different.

Sorry to be so harsh.  Long day..

-john

--- "Elton M. Labajo" <elton at mybizlinks.net> wrote:
> I like the idea of lite and full version :)
> > Reading the first few chapters of _How to Think
> Like a Computer
> > Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/
> for those who haven't
> > heard of this book), I was struck by how Python
> resembles BASIC in some
> > ways that make it a very good language for
> teaching:
> >
> > * Interpreted, so there's immediate feedback.
> Students can just type
> >  'python' at the command prompt and start coding,
> receiving *immediate*
> >  feedback on the effects of what they type.
> >
> > * Non-scary assignment and output statements. In
> fact, Python and
> >  BASIC's output statments look identical. "Print"
> is a natural choice
> >  for ease of understanding. Using sys.stdout.write
> can come later, once
> >  students begin to realize that they don't always
> *want* a newline or
> >  space appended to their output.
> >
> > * No need to declare variables. Use them and they
> are immediately
> >  created for you, ex nihilo.
> >
> > All three of these things make for a very low
> barrier-to-entry for
> > people who've never done any programming before.
> I'm sure there are
> > other similarities as well, that maybe someone
> else will point out. Of
> > course, Python goes beyond BASIC in many ways:
> >
> > * NO LINE NUMBERS! Enough said.
> >
> > * Rich control-flow statements. Of course, this is
> available in more
> >  modern versions of BASIC, but back when I was
> first learning to
> >  program (at age six, using BASIC), all I had was
> conditional GOTO. You
> >  know, IF condition GOTO line-number. Or, for the
> advanced students, IF
> >  condition GOSUB line-number (I never did
> understand the necessity for
> >  that kind of statement until well after I'd
> stopped using BASIC
> >  entirely and moved to more "modern" languages
> like... Pascal).
> >
> > * Batteries included. This is the big one, which
> makes Python more than
> >  just a toy language suitable for learning. The
> richness of the
> >  standard library makes Python incredibly useful,
> and I for one would
> >  like to see it continue to grow. I understand the
> concern that people
> >  developing Python-based programs for distribution
> would like their
> >  users to be able to run a 100k script without
> downloading a 10M
> >  distribution, but I don't want to see that cause
> a reduction in the
> >  standard library. Instead, we could make make two
> Python
> >  distributions, a "full" install containing
> everything and a "lite"
> >  install containing only the interpreter and maybe
> -- *maybe* -- a
> >  couple of essential modules like os and sys. Then
> people developing
> >  Python programs like, say, a pygame-based game,
> could distribute a
> >  single package containing the "Python lite"
> distribution, their own
> >  code, and any other modules their code depended
> on -- all in one
> >  easy-for-the-end-used-to-install package.
> >
> > Anyhow, I hope this helps develop some ideas.
> >
> > --
> > Robin Munn
> > rmunn at pobox.com
> > --
> >
> http://mail.python.org/mailman/listinfo/python-list
> 
> 
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From robin at jessikat.fsnet.co.uk  Sat Feb  2 11:30:05 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 2 Feb 2002 16:30:05 +0000
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
 <313g3a.d22.ln@wombat.san-francisco.ca.us>
 <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk> <37d27e024b.tim@worthy.demon.co.uk>
 <PLPSljAQX$W8Ewpx@jessikat.fsnet.co.uk> <ac5588024b.tim@worthy.demon.co.uk>
Message-ID: <ClZxOKANQBX8Ewis@jessikat.fsnet.co.uk>

In article <ac5588024b.tim at worthy.demon.co.uk>, Tim Howarth
<tim at worthy.demon.co.uk> writes
.....
>> but if it's easy with win32net what's the easy way?
>
>import win32net
>
>userinfo=win32net.NetUserGetInfo('','tim',2)
>userinfo['password']='secret'
>win32net.NetUserSetInfo('','tim',2,userinfo)
>
>
>'' is server name, in this case server it is run on (no need for name)
>2 is data level - what to get/set
....
I'll give that a whirl and see how it works. Thanks. I use the win32com
stuff quite a lot, but it never occurred to me that there might be
something equivalent to unix passwd. The GUI centricity of windows must
be rotting my brain.
-- 
Robin Becker


From gerson.kurz at t-online.de  Sat Feb  9 14:21:29 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sat, 09 Feb 2002 19:21:29 GMT
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4>
Message-ID: <3c657554.20448468@news.t-online.de>

On 9 Feb 2002 18:21:31 GMT, Philip Swartzleonard <starx at pacbell.net>
wrote:

>Anyway, i was just wondering, since this seems so great, is there 
>anything wrong with doing things this way that i'm not seeing?
I've been using code like that quite frequently, so if its bad, I'm in
for it, too.

>Also, is there any real reason to have the message classes inherit when 
>the only thing that seems to identify them is an attribute?
No, it works if you omit the "Message" Inheritance. You could use
plain strings and have the same effect. 



From ol1 at v10a.com  Wed Feb 20 04:29:49 2002
From: ol1 at v10a.com (O Polite)
Date: 20 Feb 2002 10:29:49 +0100
Subject: Recursively traverse linked list -- help!
In-Reply-To: <20020219191050.A3778@ibook.distro.conectiva>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
	<mailman.1014106650.11778.python-list@python.org>
	<7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek>
	<mailman.1014126336.32560.python-list@python.org> <7x7kp9blab.fsf@ruckus.b 
	<20020219191050.A3778@ibook.distro.conectiva>
Message-ID: <1014197390.1203.7.camel@matijek>

On Tue, 2002-02-19 at 23:10, Gustavo Niemeyer wrote:
> Hey!! That's not fair:
> 
> > inserts at start of LinkedList
> > 20000 0.457922101021
> [...]
> > inserts in middle of builtin list
> > 20000 0.787240982056
> 
> You're comparing oranges and bananas here... ;-)
> 

Well actually it is fair, because the insertion time for the linked list
is constant, as long as you have a handle to the insertion point. That's
pretty much the reason for using it. Inserting at the start will be just
as fast/slow as inserting in the middle. I just didn't feel like
rewriting the class so that it would take an arbitrary insertion point.
Paul was right all along.

op





From gcordova at hebmex.com  Thu Feb 14 15:49:41 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Thu, 14 Feb 2002 14:49:41 -0600
Subject: Convert String to Dictionary question
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A79@SRM201002>

> 
> I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> 
> When I do a readline the I get back a string.
> 
> How do I convert this string to a dictionary?
> 
> Thanks.
> 

You eval() it?

>>> d = { "one":1, "two":2, "three":3 }
>>> print d
{'three': 3, 'two': 2, 'one': 1}
>>> S = str(d)
>>> print S
{'three': 3, 'two': 2, 'one': 1}
>>> dd = eval(S)
>>> print dd
{'one': 1, 'three': 3, 'two': 2}
>>> d is dd
0
>>> 


Do I pass??

:-)

-gustavo



From sdm7g at Virginia.EDU  Mon Feb 18 19:19:50 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Mon, 18 Feb 2002 19:19:50 -0500 (EST)
Subject: confusion regarding os.path.walk()
In-Reply-To: <20020218180623.086f6c58.ron.l.johnson@cox.net>
Message-ID: <Pine.OSX.4.43.0202181917190.1287-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, Ron Johnson wrote:

> Hello,
>
> In the visit callout function, it looks as though after the top level
> is processed, os.path.isfile() and os.path.isdir() both return 0 when
> passed elements from the names parameter.
>
> Am I doing something wrong?
>

You need to os.path.join() the dirname to the filename and do your
test on that. The filenames are relative to the directory arg,
not to the process current directory.

-- Steve Majewski






From quinn at hurl.ugcs.caltech.edu  Mon Feb  4 17:21:33 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 4 Feb 2002 22:21:33 GMT
Subject: Defending the rodent (was Re: Defending the Python lanuage...)
References: <mailman.1012606383.28987.python-list@python.org> <jacobkm-0302022219430001@192.168.0.120>
Message-ID: <slrna5u2bd.i0j.quinn@hurl.ugcs.caltech.edu>

On Mon, 04 Feb 2002 06:19:25 GMT, Jacob Kaplan-Moss <jacobkm at cats.ucsc.edu>
wrote:
>In article <mailman.1012606383.28987.python-list at python.org>,
>James_Althoff at i2.com wrote:
> 
>> Of course, if Douglas Englebart had had his way, this would all be moot.
>> Englebart -- who invented the computer mouse -- didn't like moving his hand
>> off the keyboard either.  So for his famous NLS system he devised a
>> one-handed "chording" keypad on which he typed (all characters) using his
>> left hand while manipulating the mouse with his right.  Decades later he
>> still maintained that this is *the* way to do it.  :-)
>
>Have you ever tried a chording keyboard?  It's absolutely amazing; I can
>type about twice as fast on a chording keyboard as I can on a QWERTY one
>(although I probably couldn't now -- I haven't used one for years).

Really?  Intuitively it seems to me that they'd be slower.  Even pianists can't
play a vertical figure nearly as quickly as a horizontal one, and I'm not even
a pianist.  I suppose if you used a chording "macro" keyboard you could
potentially go quicker, but *twice* as fast as on qwerty seems to indicate that
maybe you were a poor qwerty touch-typer back then :)

I've seen blind people chunking away at chording braille typewriters, and they
go at quite a clip, but the actual letters come out pretty slowly.

>Jacob (who now wants to get his circa 1990 serial chording keyboard to
>work with his palm for taking notes in class)

I noticed that there are some places online selling them.  Tend to be
expensive, but someday I mean to try one.


My old idea for an ideal "keyboard" was a glove that responds to finger and
wrist twitches.  The you could chord those, stick your hands in a tub of hot
water, put your feet up in the air, and kiss RSI goodbye.


From tim.one at comcast.net  Sat Feb 23 18:29:27 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sat, 23 Feb 2002 18:29:27 -0500
Subject: Object counts
In-Reply-To: <ZQTd8.89849$B61.28524193@news1.rsm1.occa.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOENGNOAA.tim.one@comcast.net>

[Chuck Esterbrook]
> Is there a way to get Python to report the number of instances of all
> classes and types?
> ...

For types, yes, but it's not documented, and if anyone had time to explain
it they'd probably write docs instead <wink>.  You need to recompile Python
with COUNT_ALLOCS #define'd.  Then see the functions get_counts() and
dump_counts() in Objects/object.c, which are only defined when COUNT_ALLOCS
is.  The sys module also grows a new corresponding Python-level function
getcounts() then, and dump_counts() is automatically called when Python
shuts down then.  COUNT_ALLOCS is rarely used (I've never used it, and,
indeed, the only reason I know it exists is that several years ago somebody
submitted a bug report against it that Guido asked me to fix).

A much bigger gun can be loaded by recompiling Python with Py_TRACE_REFS
defined.  Then a list of almost all live objects at any given time can be
obtained by calling the otherwise non-existent sys.getobjects() function.
It's "almost all" because it only includes objects allocated from the heap;
some builtin type objects are statically allocated instead, so don't show up
in the list.

Note:  Py_TRACE_REFS is automatically enabled in a debug-mode build
(Py_DEBUG implies Py_TRACE_REFS).  I don't think many people know that.  So
in a debug mode build, sys.getobjects exists, and you can build just about
any object-use analyzer on top of that in pure Python.

> e.g., a dictionary that would map the names of types/classes to their
> count,

A fine example <wink>.  It's trivial to build such a dict yourself, by
marching over the list returned by sys.getobjects().

Do note that "almost all objects" can be a lot of objects; here on Windows
running a debug-build Python:

C:\Code\python\PCbuild>python_d
Adding parser accelerators ...
Done.
Python 2.3a0 (#29, Feb 13 2002, 18:39:45) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
[7092 refs]
>>> len(sys.getobjects(0))
2459
[7094 refs]
>>>

That is, there are almost 2500 objects alive "immediately" upon startup.
The "0" argument there can be nonzero if you want to clamp the maximum size
of the result list.  An optional second type argument can be specified to
return live objects of only that type:

>>> x = sys.getobjects(0, int) # using 'int' this way requires 2.2
[7155 refs]
>>> len(x)
56
[7155 refs]
>>> x[:10]
[2459, 17, 19, 20, 18, 14, 56, 448, 128, 256]
[7170 refs]
>>> sys.getobjects(10, int)
[2459, 17, 19, 20, 18, 14, 56, 448, 128, 256]
[7170 refs]
>>>

Note that "2459" is still sitting around from the first call I showed in
this session, thanks to the magical _ variable in interactive mode, then
thanks to capturing it in the list x.  After several years, the cause of
such mysteries becomes obvious at first sight <wink>.  Let's get rid of
2459:

>>> del x
[7113 refs]
>>> sys.getobjects(10, int)
[2459, 17, 19, 20, 18, 14, 56, 448, 128, 256]
>>>

It's still there!  Can you guess why?  It's because _ held the result list
from the *preceding* time sys.getobjects(10, int) got called.  Now we'll
nuke it for real:

>>> 3 # get 2459 out of _
3
[7103 refs]
>>> sys.getobjects(10, int)
[17, 19, 20, 18, 14, 56, 448, 128, 256, 512]
[7113 refs]
>>>

Bingo.  You could fritter away many useless hours trying to figure out why
448 is in the list <wink>.




From rick at kaye.to  Fri Feb  1 16:33:39 2002
From: rick at kaye.to (Rick Kaye)
Date: Fri, 01 Feb 2002 21:33:39 GMT
Subject: Urllib conundrum
References: <xr768.14279$B94.2456818@news02.optonline.net> <1wm68.11300$dh.3401976@typhoon.nyroc.rr.com>
Message-ID: <TQD68.40327$B94.6642328@news02.optonline.net>

Thanks. This worked great. I understand how your socket module code works.
However, I don't have a clue why my code using the urllib module doesn't
work. I figure it can't be TOO flawed since it works when the proxy is
installed. Since I'm fairly new to Python and still learning, if anyone
spots the mistake in what I did, I'd appreciate hearing about it.

Thanks again for the solution.


"Darrell" <dgallion1 at yahoo.com> wrote in message
news:1wm68.11300$dh.3401976 at typhoon.nyroc.rr.com...
>
> http://www.dorb.com/darrell/sockets/linksys.py
> Just fixed the getIp method to work with post python1.52
> Didn't check the other methods
>
> --Darrell
>
>
> Rick Kaye wrote:
> >
> > Now comes my problem. Since I no longer needed Cookie Cop II, I removed
> > it. But then my Python program stopped working. Normal browsing (through
> > Internet Explorer) worked fine. And my program could access webpages
that
> > are not password protected (by changing the url). But the program no
> > longer worked to get to a password protected page.
> >
>




From mxp at dynalabs.de  Mon Feb 25 15:09:20 2002
From: mxp at dynalabs.de (Michael Piotrowski)
Date: Mon, 25 Feb 2002 21:09:20 +0100
Subject: Compiling for HPUX and AIX
References: <mailman.1014659269.23795.python-list@python.org>
Message-ID: <x61yf9nx1r.fsf@notus.mii.dynalabs.de>

"czrpb" <nanotech at europa.com> writes:

> I am having the following issues compiling:
>
> HPUX 11.0
> =========

[...]

I just tried building Python 2.2 on HP-UX and had no problems; here's
my build script:

---------------------------------------------------------------------
#! /usr/bin/sh

unset LANG
export CCOPTS='+z'

make distclean

CC=cc ./configure --without-gcc --without-cxx \
 --prefix=/opt/python && make
---------------------------------------------------------------------

HTH

-- 
Michael Piotrowski, M.A.                                  <mxp at dynalabs.de>


From jeff at ccvcorp.com  Tue Feb  5 13:27:58 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 05 Feb 2002 10:27:58 -0800
Subject: Scopes
References: <mailman.1012872608.6113.python-list@python.org>
Message-ID: <3C60242E.75722E32@ccvcorp.com>

dude at deletia.com wrote:

> Hi all,
>
> I've been learning C++ and have been turned onto Python
> but the quesiton i have is
> What is the advantage of pythons nested scope rules over c++ rules.
> If I read correctly
> If func B() has Func A() as its parent, then Func A() has access (can see)
> all of func B()'s variable.
> I am not sure of the advantage here, and would appreciate a little help

Hm, that doesn't sound *quite* right.  Let's start from the beginning, and
talk about Python's "classic" scoping, before we get into nested scopes.

Python (pre-2.2) keeps track of only three scopes--local, global, and
builtin.  (Note that "global" means module-scope, not application-wide scope;
if you import two different modules, each module has its own, independent,
global scope.)  Any time that it is trying to resolve a name, it searches them
in that order.  So, if you have a local name that's the same as a builtin name
(or global name), the local name will be found first, thus effectively
obscuring the builtin/global name.  Once you are out of that local scope, then
the builtin/global name becomes "visible" again.

Thus, you could have something like:

a = 5
def spam():
     a = 1
     print a
spam()  # prints 1
print a   # prints 5

This much seems pretty similar to C/C++ scoping.  The tricky part comes when
you have nested def's or classes (something which C/C++ doesn't allow):

def FunctionFactory(multiplier):
    value = 23
    def func():
        return value * multiplier
    return func
f = FunctionFactory(5)
f()

Now, you might expect that this should print 115, but it doesn't.  Within the
def func(), there is no local variable 'value'; there's also no global
variable 'value' either.  Python (in classic scoping) can't see what's in any
function except the most "current" one, and will thus raise a NameError
because it can't find 'value'.  The traditional workaround has been to modify
the func() definition to pass any needed locals as default parameters:

    def func(value=value):

But that then raises the possibility that someone might later call f(2), and
get 10 instead of 115.  Sometimes that might be desired, but other times it's
not.

Nested scopes is the solution to that quirk.  When nested scopes are enabled
(imported from __future__ in 2.1, or by default in 2.2+), then Python will
search for names not only in the current definition's local scope, but also in
the local namespace of any function/class that that definition is nested
inside of, and any namespace that *that* is nested inside of, etc.  (In
practice, I can't imagine any reason to need more than one level of nesting,
but I believe that nested scopes are recursive anyhow.)  Thus, now func() has
access to the local variables of FunctionFactory() (but not vice versa!) and
has no need for the questionable use of default values.

Hopefully this gives you a bit better of an idea of how namespacing works in
Python.  If you have any further questions, feel free to ask.

Jeff Shannon
Technician/Programmer
Credit International




From peter.milliken at gtech.com  Tue Feb 26 16:30:41 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Wed, 27 Feb 2002 08:30:41 +1100
Subject: advocacy help ( Python vs VB )
References: <mailman.1014748548.28891.python-list@python.org>
Message-ID: <a5gutk$h1e4@news1.gtech.com>

Getting past the 'well I can just go an hire VB programmers.....' is a tough
one, I have seen a similar argument used in a company for moving to C++ :-).
So you need to present a business case - something that most managers *can*
understand :-). Business cases are tough in the sense that bottom line is
what they are interested in, so suggestions for putting one together is:

1. Convince a co-worker about the benefits of Python and *measure* the time
spent learning to produce usable output;

2. Try (somehow :-)) to show that Python development is cheaper than the
equivalent in VB - this would ideally mean that you and a co-worker develop
the same application (assuming you are equivalent in skill! :-)) in Python
and VB gathering metrics (development time primary but also metrics on bugs,
testing etc etc) along the way;

3. Take Andy's suggestion and show how migration to Python could be achieved
whilst still using VB :-)

Assuming that all of these efforts result in a positive case for Python
(anyone measured Python vs VB? :-)) then it shouldn't be too hard from then
on (other than the programmers who don't want to switch, but point 3 should
cater for that situation - although managers can see loads of mine fields in
supporting APs written in two different languages :-)).

I just checked the "Programming Language Levels and Ranges of Source Code
Statements per Function Point" Table in Capers Jones book "Applied Software
Measurement" (2nd edition) and got the following data:

Visual Basic 2   35loc/FP  Level 9.0
Visual Basic 3   32loc/FP  Level 10.0
Perl  27loc/FP  Level 12.0

Unfortunately I couldn't find Python in the table but the figures for Perl
is probably close enough. So assuming that these average figures are correct
and that there is a 1:1 relationship between Perl and Python (I assume
Python would be higher!) then using Python should show approximately a (32 -
27) / 32 * 100 = 16% productivity improvement - hardly earth shattering! :-)
But since Python is better than Perl perhaps we could expect about 25%
???????

Goodluck,
Peter

"Stefan Kuzminski" <SKuzminski at FairIsaac.com> wrote in message
news:mailman.1014748548.28891.python-list at python.org...
> I suppose the fact that I get 20 emails in response to this question is a
> sign, I know you couldn't get 20 positive third party comparisions of VB
to
> Python.  I've tried to sell Python in several organizations over the last
> few years with varying degrees of success.  The *key* sticking point
 other
> than general fear of the unknown ) is enevitably the 'well I can just go
an
> hire VB programmers but nobody knows this Python thing'.  Which of course
is
> absurd since the python learning curve is so small.  But the managers
aren't
> in any position to relate since all they control are resources with
resumes
> that say VB, C++ or Java...  ~sigh~  So my strategy so far has to been to
> build the thing without ever mentioning Python until it's done..
>
> Stefan
>
>
>
> -----Original Message-----
> From: Matthew Sherborne [mailto:miracle at paradise.net.nz]
> Sent: Monday, February 25, 2002 5:21 PM
> To: Stefan Kuzminski; python-list at python.org;
> activepython at listserv.ActiveState.com; python-help at python.org
> Subject: Re: advocacy help ( Python vs VB )
>
>
> Google uses python to script their cluster(s).
>
> http://www.google.com/jobs/openings.html
>
> (Search for python on that page, I counted 10)
>
> Also check out these URLs:
>
> http://home.rmi.net/~lutz/advocacy.html
>
> http://py-howto.sourceforge.net/advocacy/advocacy.html
>
> GBU
> Matthew Sherborne
>
> ----- Original Message -----
> From: "Stefan Kuzminski" <SKuzminski at FairIsaac.com>
> To: <python-list at python.org>; <activepython at listserv.ActiveState.com>;
> <python-help at python.org>
> Sent: Tuesday, February 26, 2002 1:25 PM
> Subject: advocacy help ( Python vs VB )
>
>
> > I need to convince my co workers and erstwhile managers to build a
> > production system in Python.  They're coming from a VB world and are
> fearful
> > to detach from the Redmond mothership.  I need hard facts, not warm and
> > fuzzies about syntax but names of big companies using Python and things
VB
> > just *cannot* do ( threads, regular expressions.. ? ).
> >
> > Thanks,
> > Stefan
> > _______________________________________________
> > ActivePython mailing list
> > ActivePython at listserv.ActiveState.com
> > To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
> >
>




From magnus at thinkware.se  Fri Feb 22 14:27:32 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 20:27:32 +0100
Subject: Newbie: Crazy, but Quick
References: <mailman.1014400057.10182.python-list@python.org>
Message-ID: <3C769BA4.20606@thinkware.se>

Gustavo Cordova wrote:

> Has a python wrapper for SQLite been written yet?

You can use ODBC. But note that SQLite isn't really
a "real" SQL database. It's more comparable with
Gadfly than with PostgreSQL for instance. (But I
guess the same can be said about Access...)



From gtalvola at nameconnector.com  Thu Feb 21 10:04:56 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Thu, 21 Feb 2002 10:04:56 -0500
Subject: HTTP state management without cookies?
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E41240E0@MAILBOX.nameconnector.com>

Paul Rubin wrote:
> Sent: Thursday, February 21, 2002 9:19 AM
> To: python-list at python.org
> Subject: Re: HTTP state management without cookies?
> 
> 
> Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> > I didn't express myself properly.  What I meant is, in a 
> CGI application if
> > you're including the session ID in the query string, then 
> you have to make
> > sure all of your relative links within the site are 
> constructed to include
> > that session ID.  In other words, your links will have to 
> be written like:
> > 
> > '<a href="SomeOtherScript.cgi?_SID_=%s">Link to Some Other 
> Script</a>' %
> > sessionID
> 
> No no, that's what the BASE tag does for you.  It maps
> 
>   <a href=SomeOtherScript.cgi>
> 
> into a request to http://yoursite.com/_SID_=12345/SomeOtherScript.cgi.
> 
> You then use a server side rewrite rule to map the SID path component
> into a query parameter.

I meant that WITHOUT using a rewrite rule, you would have to encode the
session ID unto every URL as a query param.

With a rewrite rule, it's easier -- no need to encode the session ID into
every relative link.  You can either use your method of including a BASE
tag, or you can redirect requests without a session ID so that they include
the session ID (which is what Webware does).  Of course, the drawback is
that you're now tied to a specific Apache feature, so your CGI app is no
longer portable to other web servers.

If your CGI app consists of only a single CGI script, you don't need the
rewrite rule, because you can put the session ID _after_ the name of the CGI
script and use PATH_INFO to get the session ID.  This is how Webware manages
to not need any rewrite rules -- it passes all requests through a single CGI
script WebKit.cgi.


- Geoff



From jeff at ccvcorp.com  Mon Feb  4 16:47:07 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 13:47:07 -0800
Subject: reading ftp log file..
References: <aO078.9013$Xk4.483475@news1.east.cox.net>
Message-ID: <3C5F015B.B512284C@ccvcorp.com>

Andrew Replogle wrote:

> I'm trying to write a python script that watches an ftp log file.
>
> -----------------------------------
> import os
> import sys
> import smtplib
>
> ServerPath = 'c:\\Program Files\\G6 FTP Server\\'
> logExists = os.path.isfile(ServerPath + 'FTP.log')
> line = "230"
> #print logExists
>
> log = open(ServerPath + 'FTP.log')
> lines = log.read()
> for line in lines:
>  print "someone logged in"
>  break
> -------------------------------------

Your for-loop isn't doing what it seems you think it is.  :)

First, you've set line equal to "230", but you never test whether anything is
equal to line--instead, your for-loop immediately replaces what's in line.

Second, you're reading the entire contents of your file into a single string;
thus, in your for-loop, the name "line" is actually referring to a single
character at a time (as Stefan pointed out).

You might want to do something more like this:

target = "230"
lines = log.readlines()
for line in lines:
    if line.find(target) >= 0:
        print "Target found! %s" % line
        break

Note the use of readlines() to create a list of individual lines, rather than a
single long string.  Then, the for-loop looks at one line at a time (and I've
renamed your other variable to avoid your naming conflict).  I call the find()
method on each line.  This will return the index of the character that target
starts at, or -1 if target is not found.  (If the line starts with the target,
it'll return 0; contrary to many expectations, find() doesn't return a usable
true/false value, so you have to explicitly test equality).  If we *do* find the
target, I print out the entire line, and then exit the for-loop.  Note that by
dropping the break statement, this code will print out *all* the lines in the
file that happen to contain the target--which might be preferable depending on
your requirements.

If you have any other questions, just let us know.

Jeff Shannon
Technician/Programmer
Credit International




From aleax at aleax.it  Mon Feb 18 06:49:11 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 11:49:11 GMT
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au> <mailman.1013800706.14636.python-list@python.org> <a4jsk2$q4e$0@216.39.172.122> <mailman.1013807725.17084.python-list@python.org>
Message-ID: <XS5c8.68610$om6.1895071@news1.tin.it>

Dr. David Mertz wrote:
        ...
> I don't necessarily have anything against testing in general.  It is
> possible--at least to a first approximation--to learn something about
> people's skill levels by means of tests.  And I wouldn't even go so far
> as to suppose that a complete novice would get the same result as Alex
> Martelli on the Brainbench exam.  It's not as good as it *should* be, but
> it's not completely random either.

Actually, given that Brainbench's "big idea" is _unsupervised_ tests,
said "complete novice" might well turn out to get exactly the same
results as me -- if he or she paid me enough, or was a friend of mine,
and assuming for the sake of argument that I didn't feel particularly
honor-bound to uphold that certificate's integrity.  The "honor system"
is a nice thing, but if a potential employer trusts a potential
employee enough to assume he or she didn't just hire an expert or have
a friend take the test, why don't they trust him or her enough to
just state out his or her level of expertise?

I guess this specific issue might be finessed by having the
_employer_ assign personnel to supervise the test-taking (this, I
gather, isn't the normal way Brainbench tests are taken, but it
would seem feasible).  This would also let the employer set the
environment to however they expect the prospective employee to
have to "perform under pressure": some people's performance might
be impaired by having to respect a dress-code, others' by not
being allowed to smoke, etc -- if each prospective employee can
arrange his or her own favourite test-taking conditions, this may
be meaningful for a prospective teleworking situation, but likely
not for many other job siuations.

A worse issue I have with most tests (supervised or not) is their
"quiz" nature.  Some people are very effective at taking quizzes (I,
for one), and the correlation between that and the ability to
perform useful work, while not null, isn't necessarily all that
strong either.  For Python, at least, I think a few simple,
well-specified programming tasks (one also requiring design,
one working to a pre-specified design, one about fixing a bug
in a given piece of code, another about adding functionality
to another given piece of code) might well be more meaningful.

Quizes are great for fun -- and some technologies don't really
lend themselves well to directly testing a candidate's performance
on somewhat meaningful tasks -- but Python does, so why not take
advantage of that?


Alex



From jason at jorendorff.com  Sun Feb 17 19:11:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 17 Feb 2002 18:11:28 -0600
Subject: Editable strings anyone?
In-Reply-To: <7xu1sfmz0y.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEEMDDAA.jason@jorendorff.com>

Paul Rubin wrote:
> Next question (what I need to examine the source for):  what 
> happens if you do
> 
>   x = array.array('c','a'*1000000)
>   del x[500000]
>   del x[500000]
>   del x[500000]
> 
> Do a half million characters get copied around for each of the 
> delete operations?

Yes.  There's no gap in arrays.

## Jason Orendorff    http://www.jorendorff.com/




From joshm at taconic.net  Mon Feb  4 15:20:00 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 4 Feb 2002 15:20:00 -0500
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com>
Message-ID: <3c5eea8d_4@corp.newsgroups.com>

Did you chmod the file to be executable?

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From store_li at sina.com  Sun Feb 24 04:41:17 2002
From: store_li at sina.com (Ricky)
Date: Sun, 24 Feb 2002 17:41:17 +0800
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones> <a58885$5acoc$1@ID-12869.news.dfncis.de> <zVPd8.4096$jC3.163217@news1.east.cox.net>
Message-ID: <a5acho$5j86i$1@ID-12869.news.dfncis.de>

I see now. Thank you! But I am still confuse, why I write???? method can't
use filename as a parameter.

Anyway, thanks a lot!
"Jeff Hinrichs" <jlh at cox.net> wrote in message
news:zVPd8.4096$jC3.163217 at news1.east.cox.net...
> fp is a file pointer.
> open the file (for write "w") and you are handed a file pointer
> pass it to the ConfigParser, it writes to the file pointer
> then you close the file.
>
> hth
> "Ricky" <store_li at sina.com> wrote in message
> news:a58885$5acoc$1 at ID-12869.news.dfncis.de...
> > Yes, as you guess, I haven't call .write method, because I was confused
by
> > the usage of write method. It always required a fp object, which I
totally
> > don't what it means. Is it a object create by
> a=ConfigParser.ConfigParser()
> > or a file name or a section name?
> >
> > Please help me! Thank you!
> >
> >
> > "Graham Ashton" <graz at mindless.com> wrote in message
> > news:fUJd8.418$PE2.43612 at stones...
> > > On Fri, 22 Feb 2002 16:36:22 +0000, Ricky wrote:
> > >
> > > > I intend to use ConfigParser module to manage my INI file. I can use
> get
> > > > to retrieve data from my ini file, and use set to change option
value
> in
> > > > Python, just in python, after I closed python, I found that all
> changes
> > > > I made in python lost. There is nothing altered in my ini file!
> > > >
> > > > How can I make my ini file to be changed?
> > >
> > > As you haven't posted any code I can't be sure, but I suspect from
what
> > > you've said that you've not called the write() method on your
> > > ConfigParser object.
> > >
> > > Read the module's docs.
> > >
> > > --
> > > Graham
> >
> >
>
>




From guido at python.org  Tue Feb 12 09:08:45 2002
From: guido at python.org (Guido van Rossum)
Date: Tue, 12 Feb 2002 09:08:45 -0500
Subject: ANN: New Mailing Lists -- conferences-discuss, psf-forms-discuss
Message-ID: <200202121408.g1CE8kH30354@pcp742651pcs.reston01.va.comcast.net>

At the PSF member meeting last week, there was lively discussion about
two (very different) topics: the future of the Python conference, and
the legal terms under which contributions to Python can be
redistributed by the PSF.  It was decided to create two open mailing
lists to continue these valuable discussions, and hopefully come to
concrete recommendations.  I hereby invite everyone who considers
themselves part of the Python community and wants their voice to be
heard to sign up for either list.  A brief description of the lists:

conferences-discuss:
    http://mail.python.org/mailman/listinfo/conferences-discuss

    This list is intended for discussion of the future of the Python
    conference.  Python10 was a big success, but behind the scenes
    there have been a lot of grumblings, and several alternatives have
    been proposed, e.g. a dirt cheap conference like Yet Another Perl
    Conference (held at a university, registration under $100, sleep
    in dorm rooms at $25/night), and/or moving in with the O'Reilly
    Open Source Conference (more expensive but also much larger and
    varied than past Python conferences).  Serious consideration
    should be given to doing both.  On this list we'll try to agree on
    what to do and who will do it.  Because of the lead time for
    organizing a conference, hope is that we'll reach agreement in the
    first few months of 2002 on how, where and when to hold Python11
    in 2003.

psf-forms-discuss at python.org
    http://mail.python.org/mailman/listinfo/psf-forms-discuss

    This is a short-lived discussion list set up to provide a forum
    for agreeing on PSF contributor's forms and procedures.  Its
    charter lasts only until 15-Apr-2002.

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



From trastme at honkong.com  Tue Feb 26 14:12:01 2002
From: trastme at honkong.com (Nikolai)
Date: Tue, 26 Feb 2002 23:12:01 +0400
Subject: =?koi8-r?B?8NLP3tTJ1MUsIM3P1sXULCDc1M8g98HbINvBztMuIPXEwczJ1NggzsXJy8/H?=
	=?koi8-r?B?xMEgzsUg0M/axM7PLg==?=
Message-ID: <147594609.20020226231201@honkong.com>

Hello,


     ???? ????? ???????. ???????? ?? ????????????, ?? ?? ?????????? ??????? ??? ?????? - ??? ????? ??????? ?????? ???? ?
????? ?????! ?????? ??? ????????? ?? ???? ????? ????????? - ????????? ????? ????? - ?????? ?? ??????. ????? ? ??? ??
???????? ???? ????! ??????????? ????????? ??? ?????? ?? ??????? ? ?????? ??????!

     ???????, ???? ?? ?? ?????? ?????????? ??????? ????????????? ???????????? ????????? ????? ??? ???????? ????? US$ ?
????? ?? ?????? ?? ?????, ??? ????? ??? ????????? (?????????? ????? ????? ???????? ?????? ?? ???????, ??????? ?? ?????????
??? ????????? ?????? E-mail(?) ? ?????????? ?? ??? ?????? ??????, ???? ???? ?? ????, ??? ????, ????????, ????? - ? ?????
?????? ????? ?????), ????????? ?????????? ?? ???? ?? ?? ?? ???? ? ???????? (???? ??? ??? ????? ??????) ???????????,
?????????? ?? ?????? ????? 20 US$ (???????, ? ???? ??, ???????? ????????? ????? ?????? ??? ???), ???????? ??? ???? ????
?????? ??????? (?.?. ??? ?? ???? ?? ???????? ?????? ?????????? - ??? ?? ???????? ? ?? ??? ???, ????, ????????? - ???
??????????? ???, ??? ??????? ???????? ??? ???????????? "???????", ??????? ????? ?????? ???, ?? ? ??? ???????????, ???????),
????????? ??? ? ?????????? ???????, ?? ?????? ???????????? ??????? ???? ??????? ?? ????????? ???????? ?????, ????????? ???
??????!

     ????? ????????, ??? ???? ???????? ?????? ????????? ? ???, ??? ???????????? ??? ??????? ????????? ?????????????? ?
????????? ????????????? ????? ????????? 2002 ????!

  ***

  ? ?????? -  ???? ???????? ??????????? ???????????? 



     ??? ????????? ????????????? ?????????. ???? ?? ? ???????, ? ? ??????, ? ??????? ? ??????, ?? ??? ??????, ?????????????
?? ??? ?????????, ? ??????, ?? ????????? ? ????? ????????. ? ????? ?????? ????: "? ?????? ????". ?????? ???????, ????????
??? , ? ?????? ???????, ??????? ???? ?????? ??????????. ? ??????? 5-? ???? ??????? ?? ???? ?? e-mail. ?????? ????? ????????
????? ?????????? 4. ?? ??? ? ???????, ???? ? ???????? ?????, ?????????? ??????, ?????? ???????. ??? ?????? ????? ??
??????????, ????? ???? ?????????, ???? ??? ???? ?? ??????????, ?? ??? ?? ??????? ???? ?????? ? ????. 

     ? ??????????? ???????? ??? ?????????? ??????????, ? ????? ?????, ??? ??? ???? ??????, ????? ???? ??????. ? ?????
?????? ????? ? ?????? ???? ??????? ?????? (??? ??? ???? ????????????? ?????????, ??? ???? ??? ????? ?????, ? ? ?? ??? ??????
????????), ??? ???????????? ? ????? ???????? e-mail ????? ? ????? ?????. ????? ? ??? ?????????, ? ?????? ???? ?????????????
???? ???????. ???????? ????? ???? ?????? ????????? ??????. ?? ??? ??? ????? ??? ??????, ????? ????????? ?????? ?????.
????????? ????? ? ?????? ????? ? ?? ??? ?????????: "??? ????????! ??? ????? ?????????? ???? ?? ???!". ????? ???????? ??
?????????, ?? ? ??? ????? ????????, ? ????? ???????? ??? ?????? e-mail, ???????? ?????????? ?????? ? ??????. 

     ?? ????????? ???? - ?????? ???? ? ????? ? ???????, ??? ??? ?? ????? ????????, ?? ????????? ????????. ?? ????????? ????
? ??????? 3 ??????, ? ??? ?? ?????? ? ?????? ????? ?? ??????????, ????? ????? ???? ?????? ?????????? ?????? (??? ???? ? ???
????). ?? ??? ??????, ?????? ???? ? ????? ???????? 30 ????? ? ?????????? ? ??????? ??????. ? ??????? ???? ?????? ? ???????
29 ??????? ?? ?????????? #1. ????? ?????? ????? ????????? ???? ? ???????, ?????? ?????? ? ??????? ????? ??? ???????, a
?????? ??? ????????? ?? ??? ????. ? ????? ? ????????? ????? 64.000,- USD. ? ??? ?????????? ???? ????????! ?? ??????? ??????
? ????? ???? ????? ????? ? ??? ????????? ?????????. ???? ? ?????? ?? ?? ??????, ??? ??????, ??? ? ??? ??????
?_?_?_?_?_?_?_?_?_? ? ?? ?????????. ??? ??? ????, ???? ??? ????????, ??? ?????? ?????? ?? ???? ?? ????? ?????!

  ?. ??????, ??????.



  ***

  ? ?????? ???????? ? ???????? ????. ? ??? ??????? ???????????? ???????



     ?? ???? ??? ????????? 4 ?????????? (??? ???? - "???????" ????????? ??? ????????? - ??? ????????? ??????) ?? ???? ?????
???????????? (????????????? ???????? ?????????? ? ?????????) ? ??????????? ???? (?? 5 US$ ?? ??????) ? 4-? ?????? ????? -
????? ???????????????? ?? ?????????????? ???????. ?????, ?? ????? ????? ??? ???????? ??? ?????????? ???? ????????
????????????? ?????????.

     ?????? ???? ?? ????? ?????? mail.ru - ????? ????????? ? ??????, ?? ?????? ?? ????? ?????? "???????????" ??????????
??????, ????? ????????????? ?????????? ????? ??? ?? 6 ?????. ????? ????, ? ????????????? ????? ????????? ?????????????
??????? ??????????? ? ???????, ??? - ? ???????????? (?? "????" ????? ? ??? ???????? ?????? ???????, ??? ?? "??" - ? ????)!

     ????? ??? ??? ??????, ?????? ???? ?????????? ??? ???, ?????????? ????????? ????? ?? ???? ????????? - ??:

  1. ???????? ???????, ???????????? ???????? ??? ?????? ?? ?????!

  2. ???????? ???????, ??????????????? ???????? ??? ?????? ?? ?????!

  3. ???????? ???????, ??????? ???????? ??? ?????? ?? ?????!

  4. ??????????? ???? ????????? ? MULTI-LEVEL MARKETING!

  5. ????? ???????????? ???????, ????? ????????? ?????????? ? 20 US$ ???????? ?????? ???? ????? (??? ??????, ??? ?
??????????? ? ?????????)!

  6. ???? ?????????, ??????? ?? ????????, ???????? ?????? ????????!

  7. ??? ????????? ???????? ??????? ???? ?????!

  ***

  ???, ???, ?????????, ????? ????? ??? ??????? (?????????? #0):

  1. ??????????? ????????? ?????? (?? ?????, ??? 2) ????????? WebMoney keeper ?? ????? http://www.webmoney.ru/ ??? ?? ??
??????? ???????? ?????? ? ????????? ? ?????????? ?? ???? ??????? ????????.

  2. ????????? ???? Z ??????? 20 US$ (??? ??????? R ??????? - ???? ? ????? ?????? ??? ????????????????? WebMoney, ? ?????
???????????? ???????????? ???????? ? US$, ? ????? ???????? ????? ?? US$ ?????? ????? ??????? (?.?. ??????? ?????? ?? ??????
R-???????? ?? Z ). ????????? ???????? ???? ???????? - ?? http://www.webmoney.ru/rus/perevods.htm/

  3. ????? ??????????? ????? ? ??? ???????, ???????? ???? ??? ?????? ??????????, ????? ???????? WebMoney ?? ?????? ????????
? ?????? ?? 4-? ????????? ?????????, ????? 4.96 WMZ (0.8% - ???????????? ????? ??????? ?? ????????????? ???????? ?????), ??
?????? ??????????. ?????! ???? ?? ????????? ???? ??????? ????? 20$ ? ?????????? ????? ?? 5$ ?? ?????????? 1-3, ?? ? ???
???????? ???????????? ????? ??? ???????? ?? ?????????? #4!

  * ??????????:

  *? ????????, ?????? ???????? ? ???? ???????? ?????? (WebMoney).?? ????? 1WM = 1 ???. ??? R- ????????, 1WM = 1 ?????? ???
??? Z- ????????.

  *????? ???????? ???? ?????, ?????????, ??? ?? ???????? ??? ??????????. ??? ??? ??????????? ??? ????, ????? ?? ????????? ?
???? ? ?????????? (? ?? ???????, ??? ??????????) ????? ????? ?? ????? ????????? ?????. ??? ????????????? ????? ??? ???
??????????. ???? ? ??? ?? ????? ??????? ????? ?? ???, ?? ?? ??????? ???????????? ????????. ?????? ??? ??? ?????, ? ??????
???????! ???????????, ??????? ? ???? ??????????? ????? ?????????? ? ???? E-mail ?????.

  ?????? ?????????, ???????????? ? ???????? ????? ????? WebMoney: "?????????? #1; E-mail xxxxxxxxxxxxx at xxxxxx.xx"

  ???????:

  ?????????? #    /   . ???????

                   1 / Z341138462178

 2 / Z831800404118

                   3 / Z855678326445

                   4 / Z452925066114

 



  ????????!!!

  ?? ??????????? ???? ??????? ? ????????????? ?????? ? ??????? ?????? "???????? ???????????", "???????? ?? ?????????
??????", ??? "reply" ?? ????, ? ???????? ?? ???????? ?????? ?????? - ? ????????? ?????? ??? ?????? ?? ????? ?????????, ?
????????? ?? ?????? ? ???? ????????? ? ??????? WebMoney.

  4. ? ???????, ??????? ??????, ??????????????? ?????????? #4. ???????? ?????? ?????????? 3 - ?? 4, 2 - ?? 3 ? 1 - ?? 2, ??
????? ????????? ? ??????????????? ??????? ???????. ???????? ? ??????? (??????) ?????? 1 ?? ????? ?????????. ????????? ??? WM
????????????? ?? ???? ? ????????? ?????????????? ?????? ??????? ??????. ???????, ???????? ??? ??? ?? ???? ? ????? ?????? ?
????? ??????. ?????? ??? ????? ???????? ??????! ?????? ?? ????? ????????? ?????????? #1.

  ?????!!!

     ?? ??????? ?????? ?????????, ??????? ????????? ? ???????, ??????? ????????, ????? ?????????? ? ?????? 4, ?????
????????? ??????? ????? ????? ???????. ????? ???????, ??? ??? ?????????, ??? ????? ?????? ???????, ?????? ??? ?????????
????????, ????? ???-?????? ???????? ?? ?? ?????? 4 (? WebMoney ????? ????? ????????? ?? ???? ?? ??????????? ???????).

     ?? ??????? ??????? ????????? ? ??????????!!!

     ??? ?????????? ????? ? ?????? ??????????? ???????? ??????????? ?????????. ?? ??????????, ??? ??? ?? ?????? ??????????
- ??? ???? ?? ????? ?? ??????? ????????? - ???????????????? ? ?????????????? ???????. ??? (????? ???????? ?????????? #4, ??
?? ????? ??????? ? ?????????? ??? ??? ??????) ?????? ?????????????? ? ???????????? ?????? ????? ??????????????, ?.?. ??
???????? ????? ???????????? ? ??????????? ???? ??????. ????? ????, ???????? ?????????? ?????? ?? ?????? ????????? ?? ???. ?
???????, ??? ??????? ?????????? ??? ???? ???????????? ??-?? ????? ?????? ? ????????????? Webmoney.

     ????? ??? 4 ??????????, ?? ????????? ????????? 2 ?????????? ??????????? ??? ????????? ????? ??????? ? ????????
???????? ?????????? ?????, ???????, ???? ?? ?????? ?? ???????? ??????????????, ????????? ???, ??? ???????, ? ?? ?? 20 US$,
??? ? ??? ??????????.

  ??????????? ????????? ???????????? ????????? ???????!!!

     ?????????, ???????? ???????????, ???????????? ???????? ?????? ???????? ??? ????????. ??? ????? ?????, ??? ??? ???? ??
????????? ?????????, ????? ?? ??????, ? ?? ?? ???????? ???? ??????????. ??????? ?????, ????? ?? ?????? ??????? ??? ?????????
? ?? ????????, ?????? ??? ??? ?????? ?????? ????????? ? ???? ???????.

     ??????, ????? ??? ??????? ???????????? ???? ?? ??????, ?? ????????? ????? ?? ????????? ??????????, ??????? ??????
????????? ???? ???????????, ?? ????, ????? ?????????? ?? ??? ?????????? ????! ??? ???? ????? ????? ???????????, ? ????? ????
? ?????, ?????????? ? ?????????????? ??????????! ???? ????????? ??? ???????? ???? ?????, ?? ?????? ????????? ?????????
?????? (? ?????????? ? ?.4 ??? ?????????? #0 ???????????, ???????) ?? ?????? ?????? ???????. ??? ?? ??????? ???? ???????
???????.

     ???? ????? ????????????? ????????? ?????????? ?????? ? ??? ???????? ??????? - ????????? ??, ?? ?????? ?? ?? E-mail, ??
???????? ?? ???????? ?????? ?????? (?? ? ???? ??????? ????? ????????? ????? ????????????), ? ?????? ?? WebMoney
(??????????????? ???????? ?? "????", "?????????" ? "?????????") ?, ??????????, ?????? ???????? ?????????? #1 (????????? ????
- ? ??????? ??????? ??????? - ???????????? ? ?????????????? ?????????? - ??????????? ????????? ??????????), ?? ???????
??????? ???? E-mail ??? ?????. ??? WM ????????????? (????? ??? ?????????? ?? ???????? ??? "?????????" ??????? ?????) -
271302173625.

     ? ??????????? ??????.

     ????? ????? ??????????! ??? ??????? ? ????????!

  ? ????????? ?????????,

  ???????
  

-- 
Best regards,
 Nikolai                          mailto:trastme at honkong.com

From joshm at taconic.net  Mon Feb 18 22:23:00 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 18 Feb 2002 22:23:00 -0500
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <mailman.1014084881.26721.python-list@python.org>
Message-ID: <3c71c2a3_4@corp.newsgroups.com>

> > The Ruebot wrote:
>
> Who, Joshua, it's not right to grace Rue by calling him a "bot".  Only the
> great minds of the Python community deserve that honor!

<smacks forehead>

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Will work for food at this point')




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


From phr-n2002a at nightsong.com  Tue Feb  5 05:28:50 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 02:28:50 -0800
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <7x665cz2xl.fsf@ruckus.brouhaha.com> <3C5FABB1.5010703@thomas-guettler.de>
Message-ID: <7xbsf4cjdp.fsf@ruckus.brouhaha.com>

Thomas Guettler <st-newsgroups at thomas-guettler.de> writes:
> The module random contains a lot of functions. I think getting a
> random 32 random number is needed often.
> 
> randrange(0,2147483647) is not a 32 Bit random number because
>   2**32-1 is 4294967295.

Woops, sorry, I misread 2**32-1 as 2**31-1 which is the largest
32-bit positive integer.  2**32-1 is an overflow if you use 32 bit
ints.  I guess you could try random(-2147483648, 2147483647)
if you want all 32 bits random.  Alternatively you could try
  (random(0,65536) << 16 | random(0,65536))


From henrik.motakef at web.de  Sat Feb 16 08:29:55 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 16 Feb 2002 14:29:55 +0100
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl>
Message-ID: <861yflwo4s.fsf@apo.hmotakef.int>

Piet van Oostrum <piet at cs.uu.nl> writes:

> BASIC a very good language for teaching????
> Well, it depends on what you want to teach, I guess.

Or, in the words of Edsger Dijkstra:

"It is practically impossible to teach good programming style to
students that have had prior exposure to BASIC: as potential
programmers they are mentally mutilated beyond hope of
regeneration."

scnr
Henrik


From =?iso-8859-5?Q?=DF=D5=DD=D3=E3=D8=DD=D8=E1=E2=D0?= at =?iso-8859-5?Q?=DC=D0=D8=DB?=.=?iso-8859-5?Q?=DD=D5=E2?=.=?iso-8859-5?Q?=DC=DA?=  Fri Feb 15 11:08:33 2002
From: =?iso-8859-5?Q?=DF=D5=DD=D3=E3=D8=DD=D8=E1=E2=D0?= at =?iso-8859-5?Q?=DC=D0=D8=DB?=.=?iso-8859-5?Q?=DD=D5=E2?=.=?iso-8859-5?Q?=DC=DA?= (=?iso-8859-5?Q?=B4=D0=DC=F8=D0=DD_=B3=2E?=)
Date: Fri, 15 Feb 2002 17:08:33 +0100
Subject: Tkinter and Greek Symbols
References: <3daa1aa8.0202141718.3500c462@posting.google.com> <m3heoj5fda.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3c6d3278$1@news.mt.net.mk>

> import Tkinter
> 
> l = Tkinter.Label(text = u"\N{GREEK SMALL LETTER ALPHA}")
> l.pack()
> l.tk.mainloop()
> 
> If you happen to know that GREEK SMALL LETTER ALPHA is \u0391, you can
> use that as well. If you have an editor that supports utf-8, you can
> use unicode("<utf-8 encoding of the greek text>", "utf-8") instead
> of the Unicode literal.

Does this mean that Tkinter support unicode?
Does it support the iso10646-1 fonts too?
Can I display Symbols from several alphabets n a single widget?
And how do I get the keyboard input, in unicode od locale specific?



-- 
??????

            Yes, I've heard of "decaf." What's your point?



From usenet at thinkspot.net  Fri Feb 15 22:44:18 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 15 Feb 2002 19:44:18 -0800
Subject: [Python-Dev] PEP 282: A Logging System  --  comments please
References: <20020215164333.A31903@ActiveState.com> <mailman.1013828010.19562.python-list@python.org>
Message-ID: <a4joej.3vve7l5.1@kserver.org>

On Fri, 15 Feb 2002 21:43:33 -0500, Michael McLay <mclay at nist.gov> wrote
in comp.lang.python in article
<mailman.1013828010.19562.python-list at python.org>:

> On Friday 15 February 2002 07:43 pm, Trent Mick wrote:
> > Howdy all,
> >
> > I would appreciate any comments you might have on this proposal for adding
> > a logging system to the Python Standard Library. This PEP is still an early
> > draft so please forward your comments just to me directly for now.
> 
> I scanned the PEP and didn't find a reference to the logging package 
> supporting logging over a network. 

I was curious about whether it would address the issue of persistency or
not.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From nobody at nobody.com  Mon Feb  4 18:11:38 2002
From: nobody at nobody.com (John)
Date: Mon, 4 Feb 2002 17:11:38 -0600
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com>
Message-ID: <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu>

Any will do, e.g., uniform, normal, etc.

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xg04gsvgh.fsf at ruckus.brouhaha.com...
> "John" <nobody at nobody.com> writes:
> > Is there any random number generator in Python, that allows users to
> > specify "skewness"? Thanks!
>
> I'm not too up on that statistics stuff, but I'd think you'd care
> about the shape of the whole distribution and not just the skewness.
> Is there a particular distribution you're interested in?




From brian at sweetapp.com  Thu Feb  7 22:48:35 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Thu, 7 Feb 2002 19:48:35 -0800
Subject: FixedPoint
In-Reply-To: <LNBBLJKPBEHFEDALKOLCAEGHNKAA.tim.one@home.com>
Message-ID: <000501c1b053$7c8c7b00$445d4540@Dell2>

 Tim Peters wrote:
> Magnitude (as opposed to identity -- "is") comparision of a FixedPoint
to
> a non-numeric object makes no sense that I'm comfortable imposing, so
I'm
> happy to let it raise an exception. 

You shouldn't be. There is no more reason for FixedPoint to raise an
exception (aside from your desire not to bother fixing it :-) than for
other other "incompatible" types e.g.

>>> 5 == 5 + 5j
0
>>> 5 == '5'
0

Cheers,
Brian




From donn at u.washington.edu  Wed Feb 20 15:11:45 2002
From: donn at u.washington.edu (Donn Cave)
Date: 20 Feb 2002 20:11:45 GMT
Subject: why does popen2 silently ignore IOError?
References: <mailman.1014229773.31846.python-list@python.org>
Message-ID: <a50vu1$29da$1@nntp6.u.washington.edu>

Quoth Graham Guttocks <graham_guttocks at yahoo.co.nz>:
| Can anyone explain why the popen2 module silently ignores an IOError? 
| I'm trying to use it to send mail instead of os.popen() to avoid 
| invoking /bin/sh.  
|
| If all the command-line arguments are proper, things work fine, but if 
| I open a bogus command-line, popen2 doesn't complain, neither during 
| the write, nor close.  Take the following simple example: 

Where's the IOError?

popen2's behavior here is more or less in line with system() and popen(),
in the general sense that it does not raise an exception in the case of
a non-zero exit from the child process.  I agree that it's a defect, the
kind of thing we hope to get away from when we choose to write Python.
But as a practical matter, process exit statuses aren't as useful as the
error codes you get from system functions, for example.  I personally
prefer to get an exception, with value from stderr output, but it's
debatable whether that would have been a good default for system & popen.
And popen would have to be redesigned anyway, instead of file objects it
would have to return special I/O objects with the ability to monitor their
process.

What does this have to do with failed execs?  By the time an exec fails
like that, it's already in a fork, and the only way it can get the news
back to the caller is in the fork's exit status, and there isn't any way
to make that number say "exec failed", without some ambiguity.  There may
be some way to get that information - for example, if you wanted to go to
the trouble to create an extra pipe for internal use, the fork could write
a message to the pipe indicating exec failure.  Obviously no one has done
this in popen2, though.

popen is actually a bit better than popen2 here, though, inasmuch as when
you close the file object you get a special return value with the exit
status.  To get this value with popen2, you need to use the Popen3 class
and call its wait() member function.

	Donn Cave, donn at u.washington.edu


From michael at rcp.co.uk  Fri Feb  1 04:07:44 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Fri, 1 Feb 2002 09:07:44 +0000 (UTC)
Subject: An improvement to Queue
References: <mailman.1012515655.8263.python-list@python.org>
Message-ID: <Xns91A85CDCAE3FBmichaelrcpcouk@194.238.50.13>

"Tim Peters" <tim.one at home.com> wrote in
news:mailman.1012515655.8263.python-list at python.org: 

> [Michael Abbott]
>> It's the transition from empty to non-empty that's important: I know
>> that the consumer will consume until there's nothing left (that's part
>> of the contract), and the producer should be able to promise not to
>> pester the consumer until it has finished its previous round of work. 
> 
> It *sounds* like you're using Queue to manipulate items at too low a
> level of granularity.  If you created a class to encapsulate the notion
> of "one round of work", then the consumer could pull off "one complete
> round of work" per Queue.get(), and you wouldn't need other synch
> gimmicks (or other complications) to communicate where "one round of
> work's" boundaries lie. 

The problem is that when the consumer gets around to performing its round 
of work, rather than just performing the "jobs" in sequence and in 
isolation, in fact the the consumer is in a position to merge jobs together 
into a single action.

Think of a cursor addressable terminal the wrong end of a serial line (this 
is close to my application): sending characters down the serial line is the 
main limiting resource, and the "consumer" in this case is in the business 
of formatting up an incremental update and sending it on.

However, changes to the display are coming in all the time, too fast to 
display, on a separate thread.  I therefore want to accumulate these 
changes and when there is room on the serial line send the entire batch.  
The consumer should therefore be spending most of its time waiting for the 
serial line to become free, and then formatting up a batch of updates.

Arguably this accumulation should be done in the generating thread, but I 
don't really wish to do it that way in this particular case.  In 
particular, the model I propose makes synchronisation almost trivial, which 
is I think quite an important consideration!


>> However, the question is: what am I going to do with a command that's
>> ready to be executed?  The problem is, the code that's going to obey
>> the command might not be ready for it, 
> 
> If the consumer pulls off a "one round of work" object, then presumably
> it won't *try* another Queue.get() until it is ready for another round
> of work. Queue exists to mediate potentially mismatching flow rates,
> not to supply a brittle way to perform aggregation <wink>.

Brittle?  Surely not.  I guess that's my point: if I have mis-matched flows 
between two threads, and as you point out, I want to perform aggregation of 
the data, I would like the option of performing this aggregation either 
side of the flow.  Having the consumer aggregate is simpler (though, of 
course, has its own problems).

One thing I have learnt in doing a fair amount of (soft) real time data 
redistribution (in C[++]) is to keep event triggers and thread context 
switches under control: they cost, particularly if you are dealing with 
many thousands of events.
    (Unfortunately, Python seems less suited to this approach, given the 2 
orders of magnitude cost we pay up front for using the language, but that's 
another story.)


From kragen at pobox.com  Sun Feb  3 01:20:37 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 03 Feb 2002 01:20:37 -0500
Subject: Naming higher order functions
References: <kuelk63747.fsf@lasipalatsi.fi> <a3bpjt$kle$01$1@news.t-online.com> <a3bqc0$898$1@serv1.iunet.it>
Message-ID: <83zo2rxf0q.fsf@panacea.canonical.org>

"Alex Martelli" <aleax at aleax.it> writes:
> To make it almost-work in older Pythons, the old trick was:
> 
> def makeGetNth(n):
>   def g(x, n=n): return x[n]
>   return g
> 
> ...  The "almost" is because one might then, by
> mistake, call the returned g with two arguments, and the
> second one would take n's place; you have more control now.

You could still write

class makeGetNth:
    def __init__(self, n): self.n = n
    def __call__(self, x): return x[self.n]

which works the same way but avoids the "almost".  Nested scopes are
nicer, of course.



From sdm7g at virginia.edu  Tue Feb 26 16:10:29 2002
From: sdm7g at virginia.edu (Steven Majewski)
Date: Tue, 26 Feb 2002 16:10:29 -0500 (EST)
Subject: subclassing float (or not)
In-Reply-To: <a5gsa4$7b5gj$1@ID-59885.news.dfncis.de>
Message-ID: <Pine.OSX.4.40.0202261604140.560-100000@d-128-54-224.bootp.virginia.edu>


On Tue, 26 Feb 2002, Thomas Heller wrote:

> It seems you always end up with implementing something like
>
> class F(float):
>   # float subclass
>   def __mul__(self, other):
>     return F(float(self) * other, ...)
>
> or
>
> class F:
>   # container class with a float instance variable
>   def __mul__(self, other):
>     return F(self.float * other, ...)
>
> The only *real* difference seems that you can pass float
> subclass instances to round(), log(), and these kind of functions.
> As you already noted, normally they return the wrong results ;-(
>

Well -- the other main advantage of subclassing float is that you
can do:  'isinstance( myFloatType( 1.0 ), float )'  and get true.


The other shortcut it gives you is that coercing to float or
int works without you having to actually write a coerce method:

  int( myFloatType( 1.0 ) )  --> 1


There may be other things lurking there that automatically work
correctly, that wouldn't if you used a container class. If you
subclass float instead, you don't have to wait around to see
what else breaks because you forgot to implement a method to
support it!

-- Steve




From luc.saffre at mail.ee  Thu Feb 21 02:15:03 2002
From: luc.saffre at mail.ee (Luc Saffre)
Date: Thu, 21 Feb 2002 07:15:03 GMT
Subject: current default string encoding
Message-ID: <3c749db4.11089435@news.cis.dfn.de>

Hello,

excerpt from Python 2.2 Library Reference, Section 2.2 Built-in Types:

encode([encoding[,errors]]) 
Return an encoded version of the string. Default encoding is the
current default string encoding. errors may be given to set a
different error handling scheme. The default for errors is 'strict',
meaning that encoding errors raise a ValueError. Other possible values
are 'ignore' and 'replace'. New in version 2.0. 


Question: 
How or where is this current default string encoding defined? Can i
change it during run-time?



From tripps81 at yahoo.com  Tue Feb 19 12:26:00 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Wed, 20 Feb 2002 00:26:00 +0700
Subject: basic books/guides on multithreading programming?
In-Reply-To: <7934d084.0202172009.75324c2a@posting.google.com>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
 <mailman.1013816158.26552.python-list@python.org>
Message-ID: <5.1.0.14.2.20020220002245.0a4ade80@202.95.131.154>

On 18/02/2002 11:09, Andrae Muys wrote:
>"End of Story" is a rather interesting way to summarise the issues of
>Deadlock, Lock-ordering, Livelock, Starvation, Convoying, and
>Lock-Contention to name a few of the problems you may face moving from
>serial to parallel algorthims (although the last two only affect
>perfomance, not correctness).

ok, i understand more or less about deadlock or lock ordering 
(from the database equivalent terms), though have not heard 
about livelock or the others. suppose i decide to be ignorant 
about these issues "for now" and go about building my 
multithreading project. and then suppose these issues cause 
serious problems. would it be difficult to resolve it? would i 
have to rethink/review all the threaded code?

t




From soperpd at yahoo.com  Wed Feb 20 17:07:35 2002
From: soperpd at yahoo.com (Paul Soper)
Date: 20 Feb 2002 14:07:35 -0800
Subject: How to hide Tk root window when using only canned dialogs?
References: <f89f6348.0202200642.7d38ae20@posting.google.com> <a50cvo$2k00$1@norfair.nerim.net>
Message-ID: <f89f6348.0202201407.1e3a054c@posting.google.com>

Thanks!  Works like a charm.


From danielk at aracnet.com  Sat Feb  9 11:45:33 2002
From: danielk at aracnet.com (Daniel Klein)
Date: Sat, 09 Feb 2002 08:45:33 -0800
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <7gka6ukd9bou22gthq0261l319hfnvsdf1@4ax.com>

On 8 Feb 2002 11:10:47 -0800, aahz at panix.com (Aahz Maruch) wrote:

>I know some of y'all still keep roots to the Perl community, so I was
>wondering if there's any kind of definitive nickname for Perl
>programmers, the way we call ourselves Pythonistas?

Masochists.



From mwh at python.net  Mon Feb  4 06:40:51 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 4 Feb 2002 11:40:51 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de>
Message-ID: <uit9dqxp9.fsf@python.net>

Simon Budig <Simon.Budig at unix-ag.org> writes:

> In article <3c5b2c1f at si-nic.hrz.uni-siegen.de>, Simon Budig wrote:
> > Michael Hudson <mwh at python.net> wrote:
> >> Simon Budig <Simon.Budig at unix-ag.org> writes:

[complaint as subject]

> >>> Is there an evil hack to avoid this?
> >> 
> >> Dunno.  Doubt it, somehow.  You'll need to look at the documentation
> >> for readline, though.
> > 
> > Well - I looked at the documentation for the readline-module, but that
> > doesn't say anything about it. Maybe I should read the sources.
> 
> Ok, did that. Here is the Patch to fix this behaviour. It is against
> Modules/readline.c from Python 2.1.2.

Cool.

> ----8<-----------------------------
> --- readline.c.orig	Sat Feb  2 21:44:09 2002
> +++ readline.c	Sat Feb  2 22:01:16 2002
> @@ -346,6 +346,9 @@
>  		   lock released! */
>  		save_tstate = PyThreadState_Swap(NULL);
>  		PyEval_RestoreThread(tstate);
> +		/* Don't use the default filename completion if we
> +		   have a custom completion function... */
> +		rl_attempted_completion_over = 1;
>  		r = PyObject_CallFunction(completer, "si", text, state);
>  		if (r == NULL)
>  			goto error;
> ----8<-----------------------------
> 
> I really would appreciate, if this gets incorporated in the regular
> Python releases. 

First, an awkward question: do you know which versions of readline
support the rl_attempted_completion_over variable?  If it's not been
present since 2.2, some nasty autoconf hackery may be needed (autoconf
can join readline and ncurses on that list we were talking about...).

> Unfortunately I don't have an account at sourceforge to submit this
> the proper way.

Why not?

Cheers,
M.

-- 
  It could be argued that since Suitespot is infinitely terrible,
  that anything else, by very definition of being anything else,
  is infinitely superior.                -- ".", alt.sysadmin.recovery


From emile at fenx.com  Wed Feb 13 11:48:45 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Feb 2002 08:48:45 -0800
Subject: filtering random values (sieves?)
References: <5e8bd451.0202130827.2df6f50b@posting.google.com>
Message-ID: <a4e5na$1efggq$1@ID-11957.news.dfncis.de>

From: "kevin parks" 
> I am not sure I am even using the right terminology here, but I
> looking for a way to filter random values. Let's say that I am using
> random and randrange to generate values from 0-11, but i only want to
> accept values 1, 3, 5, 7, 11, 


Does this help:

>>> import random
>>> random.choice([1,3,5,7,11])
7

Emile van Sebille
emile at fenx.com

---------



From jason at jorendorff.com  Tue Feb 26 17:14:34 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 16:14:34 -0600
Subject: PEP 263 comments
In-Reply-To: <200202250520.g1P5KKD01484@mira.informatik.hu-berlin.de>
Message-ID: <HFEKILOLEFEFMKAECNDLIEHNDFAA.jason@jorendorff.com>

Martin v. Loewis wrote:
> To make some progress on PEP 263, I suggest that some of the open issues
> are resolved as follows:

Counter-proposal:

 - Comment syntax: none.
 - UTF-8 file signature: not supported.
 - Python source code encoding: must always be UTF-8.
 - Implementation: within the parser, everything's just
   ordinary UTF-8 bytes.
 - IDLE: always save UTF-8 unless otherwise directed.

Advantage: simple, universal, easy, similar to what Java does.
No confusion about embedded 0x22 bytes in strings.  Also,
stylistically I prefer not to have a document specify its own
encoding, or for comments to affect the meaning of a source
file.

I think this is better than PEP 263, but PEP 263 is better
than nothing.

## Jason Orendorff    http://www.jorendorff.com/





From tim.one at home.com  Thu Feb  7 23:06:10 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 7 Feb 2002 23:06:10 -0500
Subject: FixedPoint
In-Reply-To: <000501c1b053$7c8c7b00$445d4540@Dell2>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEGKNKAA.tim.one@home.com>

[Tim]
> Magnitude (as opposed to identity -- "is") comparision of a FixedPoint
> to a non-numeric object makes no sense that I'm comfortable imposing, so
> I'm happy to let it raise an exception.

[Brian Quinlan]
> You shouldn't be. There is no more reason for FixedPoint to raise an
> exception (aside from your desire not to bother fixing it :-) than for
> other other "incompatible" types e.g.
>
> >>> 5 == 5 + 5j
> 0
> >>> 5 == '5'
> 0

Note that I said "non-numeric object".  FixedPoint supports a rich set of
auto-coercions from types that "make sense".  I'm not a fan of semantically
senseless comparisons, e.g.

>>> 5 < '5'
1
>>>

Making it easy to supply ridiculous behaviors in subclasses is what OO is
all about <wink>.




From ryan.phillips at csus.edu  Wed Feb 20 19:32:14 2002
From: ryan.phillips at csus.edu (Ryan Phillips)
Date: 21 Feb 2002 00:32:14 GMT
Subject: HTMLifyer for Python code?
References: <mailman.1014238715.7462.python-list@python.org> <Xns91BBEF384B311cliechtigmxnet@62.2.16.82> <mailman.1014244780.2730.python-list@python.org>
Message-ID: <a51f6e$7d7@dispatch.concentric.net>

vim also has a plugin to covert python code to colorized html.

-ryan

"Trent Mick" <trentm at ActiveState.com> wrote in message
news:mailman.1014244780.2730.python-list at python.org...
> > > tool for converting Python source code to syntax-highlighted HTML.
> >
> > the SciTE editor which you can finde here:
> > http://www.scintilla.org/
> >
> > has a save as HTML option in the filemenu. (that editor component is
> > used in PythonWin and wxWindows)
>
> Brian's SilverCity (which he just mentioned in this thread) uses
Scintilla's
> lexers so that above step has already been automated for you there.
>
> Trent
>
>
> --
> Trent Mick
> TrentM at ActiveState.com
>




From andersjm at dancontrol.dk  Tue Feb  5 09:07:32 2002
From: andersjm at dancontrol.dk (Anders J. Munch)
Date: Tue, 5 Feb 2002 15:07:32 +0100
Subject: distutils, C extensions, Borland compiler, wxPython
Message-ID: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk>

Hi,

I have built Python from source on Win2K using cygwin except with a
Borland compiler (BCB4) instead of gcc.  It took a fair amount of
hacks to get the configure script and makefiles to work (s/.o/.obj/ is
just the beginning) but it was worth it.

It's all statically linked.  I kinda like monolithic executables, so
rather than get .dlls to work I have just disabled that part of the
makefile.

It works great.  I've been extending and embedding like crazy for a
while, using Setup.local.

There's just one hitch: distutils.  I'd like to use wxPython, but
wxPython uses distutils with C extensions.

"python setup.py install" fails, unsurprisingly, because it doesn't
know how to build and import the C++ code.  I wonder if the
configure/make process is supposed to store this information
somewhere, but even if it did, it probably wouldn't work.  I presume
distutils relies on the use of .dll/.so's which I'd rather not have to
mess with, and also my python.exe is not a cygwin app. and hence
doesn't understand the cygwin paths from the make process.

Now what I'd like to have is for distutils to just tell me which
.c/.c++ files need to be compiled into python and with which options,
so that I could add these files to Setup.local manually and rebuild
python.exe.

The distutils docs I have seen so far don't seem to get beyond
"press the big green button and everything works automatically".  Are
there any docs on how to tell distutils which compiler to use and how?

TIA for any advice.

the-docs-state-that-"Windows-Python-is-built-in-Microsoft-Visual-C++"-
which-is-of-course-perfectly-true-(except-when-it-isn't)-ly y'rs, 
 Anders


PS: The wxPython setup.py error message that I get:
! Traceback (most recent call last):
!   File "setup.py", line 10, in ?
!     from my_distutils import run_swig, contrib_copy_tree
! File "my_distutils.py", line 125, in ?
!  ccompiler.default_compiler['nt'] = 'my_msvc'
! AttributeError: 'module' object has no attribute 'default_compiler'




From chris.gonnerman at newcenturycomputers.net  Mon Feb  4 09:02:56 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 4 Feb 2002 08:02:56 -0600
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>
Message-ID: <00b401c1ad84$bcd8d5e0$0101010a@local>

----- Original Message -----
From: "Steve Holden" <sholden at holdenweb.com>


> "Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
> [editor stuff]
> >
> > I suppose it might be a brain thing. I use my left hand for writing and
> > drawing while I use my right hand for more coarse-grained tasks, like
> > throwing a ball and such.
>
> Strange. I don't come across many other people who do this. I've always
> thought my brain must have been differently wired for some reason I'll
never
> know.

I am one.  I can't write worth a crap with my right hand, but if you see
me with a hammer in my left hand, RUN AWAY.

> > I wouldn't count myself as ambidextrous though.
> >
> "I'd give my right arm to be ambidextrous".

I tell people "I'm ambidextrous; equally clumsy with both hands."





From phr-n2002a at nightsong.com  Thu Feb 14 01:21:11 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 13 Feb 2002 22:21:11 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com> <7x1yfq2gdf.fsf@ruckus.brouhaha.com> <c76ff6fc.0202130114.6c1ab5ae@posting.google.com>
Message-ID: <7x6650mvmw.fsf@ruckus.brouhaha.com>

sjmachin at lexicon.net (John Machin) writes:
> > Suppose I'm parsing the file and I see a </table> tag and I want to
> > find the matching <table> tag.  It could be pretty far back in the file.
> > That's what I was doing when I encountered this question.
> 
> Paul, I really admire your energy, writing your own HTML parser. I'm a
> lazy old so-and-so, if I had the slightest interest in HTML I'd be
> looking at the HTMLParser and htmllib modules. What was there about
> them that didn't suit your purpose?

I'm not trying to write a general purpose HTML parser; I'm trying to
pull some specific data out of a specific bunch of HTML files that are
several megabytes long and would probably break the htmllib modules.

> ... and if I had the energy to write an HTML parser I probably would
> have used some dumb old technique like reading the file forwards and
> if I met <foo> I'd put it in a stack (or some other data structure) of
> "unclosed" tags (together with the position in the file where I found
> it) and when I met the corresponding </foo> I'd take the appropriate
> "foo" action (which would probably involve the use of the data that
> I'd found between <foo> and </foo>)and then rip "foo" out of the
> pending bag ... using a regex backwards to find the opening tag is so
> innovative that I'm totally gobsmacked.

Basically I was trying to rewrite an Emacs macro as a Python script.
There was no need for such fancy structures.  Anyway, searching
backwards is a perfectly natural thing to want to do, so if there's
no easy way to do it, there's a deficiency in the re module.  (The
underlying regex library takes a direction flag, as mentioned before,
so the re module doesn't fully implement the regex library's API).

> > But searching
> > backwards is a normal thing to want to do in general--for example it's
> > a standard command in any decent text editor.
> 
> ... and in many indecent text editors. However AFAIK the
> implementation is to go back a line at a time and do a forward regex
> search in each line.

That depends on how the editor is implemented, of course.  And that
scheme breaks if the regex is supposed to match something that's
split across multiple lines.

> > Anyway, I just entered a sourceforge bug about it being missing from
> > Python's re module.
> 
> Looks like the effbot gets to be gobsmacked too.

I'm not sure what you mean by that.

> > Thanks
> 
> No, Paul, thank *you* -- you've made my day.

Or that either.


From bokr at oz.net  Wed Feb 27 03:59:04 2002
From: bokr at oz.net (Bengt Richter)
Date: 27 Feb 2002 08:59:04 GMT
Subject: Dictionary type access to list of instances
References: <a5dcgh$b22$1@ctb-nnrp1.saix.net> <3C7A5AD4.9090404@wi.mit.edu> <a5fedu$qe1$1@ctb-nnrp1.saix.net>
Message-ID: <a5i74o$kts$0@216.39.172.122>

On Tue, 26 Feb 2002 09:44:33 +0200, "Grant Beasley" <gbeasley at tsa.ac.za> wrote:
[...]
>
>What I'm ultimately aiming for is a database-like view of a list of
>instances, so that I can use SQL commands to get instances. So in the
>analogy, a list of instances would be a table, with each instance a record,
>and each property a field.
>
>eg.
>class User:
>    def __init__(self, username, accesslevel):
>        self.username = username
>        self.accesslevel = accesslevel
>
>list = [User('Bob', 1), User('Jim', 2), User('Pete', 2)]
>
>Then using my imaginary function ListSQL:
>
>ListSQL("select * from list where accesslevel = 2")
>would return a list consisting of the Jim and Pete instances.
>
>or ListSQL("select username, accesslevel from list where accesslevel = 2")
>would return [('Jim', 2), ('Pete',2)] - i.e. a List of tuples.
>
>Is this an interesting / useful idea? Has it been done before? Any comments?
>
Well, it got me to try out properties and weak references and
modifying a class variable before use of the class ;-).

I'm on thin ice here, so comments from someone who's been using
these features a while would be welcome. ;-)

Below is an interaction from my categorizer.py (see listing at end):
In this case I am just using name strings as objects, but they
could be any objects. The Categorizer class wraps objects you
give to it in a new object which also has a propery (2.2 style),
whose value you can also specify, or the object will belong to
the default category with prop==None. The class keeps track of all
instances in categories [lists] of equal prop values. Actually
it's a dictionary with the properties for keys, and the corresponding
values are lists of weak references to wrapper objects with that key
as property value.

If you update the property of a wrapper object, it's prop value category
membership is automatically updated, and likewise if a wrapper object
is deleted (using weak references seems to work in allowing the ref
count to go to zero and the destructor gets called) ;-).

The access methods to get lists of objects etc are accessible as
methods through any wrapper object instance.

The getCategorizerClass function delivers a class with a separate
{prop_value:[obj_weak_refs]} dictionary, to that several distinct
categories can operate separately. I used a list below (ulist) of
instances to bind the wrapper objects, but they can be bound any
way that keeps them alive until you want to delete them.

I'm not sure this is a practical solution to your problem, but it
was an interesting exercise. I'd be interested in a clean and proper
meta-class solution to generating the Category classes that
categorizer.getCategorizerClass() returns.

Regards,
Bengt Richter

______________________________________________________________

 Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
 Type "help", "copyright", "credits" or "license" for more information.
 >>> import categorizer
 >>> NamePriv = categorizer.getCategorizerClass()
 >>> ulist = [NamePriv('Bob',1), NamePriv('Jim',2), NamePriv('Pete',2)]
 >>> ulist[0].get_objs(1)
 ['Bob']
 >>> ulist[0].get_objs(2)
 ['Jim', 'Pete']
 >>> ulist[1].prop = 1
 >>> ulist[0].get_objs(1)
 ['Bob', 'Jim']
 >>> ulist[0].get_objs(2)
 ['Pete']
 >>> del ulist[0]
 >>> ulist[0].get_objs(1)
 ['Jim']
 >>> ulist[0].get_objs(2)
 ['Pete']
 >>> ulist[0].prop
 1
 >>> ulist[0].obj
 'Jim'
 >>> ulist[0].prop =2
 >>> ulist[0].get_objs(1)
 []
 >>> ulist[0].get_objs(2)
 ['Pete', 'Jim']
 >>> ulist[0].get_obj(2)
 'Pete'

(Note that get_obj gets only the first available, unlike get_objs (plural)).
______________________________________________________________

# categorizer.py
# (strawman discussion example, not production code)
"""Use getCategorizerClass([dict]) to get a Categorizer class"""

import weakref

def getCategorizerClass(d=None):
    """getCategorizerClass([dict]) => Categorizer class"""

    class Categorizer(object):
        """class to wrap object in another with categorizing property"""
        _pd = d or {}  #fresh d instance dict as class variable of form:
                 # { prop_value: [list of weak references to wrapper instances having a given prop value] }
        def __init__(self, v=None, p=None):
            """init Categorizer box instance with obj and property"""
            self.__doc__ = """instance of class Categorizer having property prop"""
            self.obj = v  # the obj in the box
            self.wsr = weakref.ref(self)  # weak self ref to use in dict
            self.prop = p   # triggers set_prop to add to list of objects with p as property

        def __del__(self):
            try:
               self._pd[self._prop].remove(self.wsr) #remove self from current prop's list
               del self.wsr
            except:
                pass

        def get_cat_boxes(self,v=None):
            """find Categorizer box instance(s) whose property is a given value"""
            return filter(None, [y() for y in (self._pd.get(v) or [])])

        def get_objs(self,v=None):
            """get objs from Categorizer box instance(s) whose property is a given value"""
            return [x.obj for x in filter(None, [y() for y in (self._pd.get(v) or [])])]

        def get_obj(self,v=None):
            """get first available obj from Categorizer box instance(s) whose property is a given value"""
            for y in (self._pd.get(v) or []):
                yo = y()                # get obj from weak ref
                if yo: return yo.obj    # return first found only
            return None                 # none found

        def get_prop(self):
            """triggered by box_instance.prop in expression"""
            return self._prop
        def set_prop(self, p):
            """triggered by box_instance.prop on left hand side of '='"""
            try:
               self._pd[self._prop].remove(self.wsr) #remove self from previous prop's list
            except:
                pass
            self._prop = p
            try:
                self._pd[p].append(self.wsr)      # put self on list for prop==p
            except:
                self._pd[p] = [self.wsr]          # ditto

        # tie in property methods
        prop = property(get_prop, set_prop, None, #no deleting the prop
                  "Property 'prop' of this instance")

    return Categorizer
______________________________________________________________



From barry at zope.com  Fri Feb  8 17:38:35 2002
From: barry at zope.com (Barry A. Warsaw)
Date: 08 Feb 2002 17:38:35 -0500
Subject: Parsing mime multipart/alternative emails for the message body
References: <mailman.1011712592.10665.python-list@python.org>
Message-ID: <bazo2j38gk.fsf@anthem.wooz.org>

>>>>> "JPM" == Joao Prado Maia <JMaia at lexgen.com> writes:

    JPM> I'm trying to parse mime based emails to get the message
    JPM> body, but I didn't have much luck on trying to find a way.

You might want to give Python 2.2's email package[*] a try.  It is
specifically designed to do all the things you'd want to do to and on
an email message, including parse a text file into an object model,
etc.

-Barry

[*] Or for older Python's look at the distutils-able mimelib/email
package at:

    http://sourceforge.net/projects/mimelib/

The version in CVS has some much better support for multibyte
character sets.


From threeseas at earthlink.net  Sat Feb  9 14:55:42 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 09 Feb 2002 19:55:42 GMT
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3c655f8a$1_1@corp.newsgroups.com>
Message-ID: <755.805T1832T9425394threeseas@earthlink.net>

On 09-Feb-02 12:52:27 Joshua Muskovitz <joshm at taconic.net> wrote:
>> I kinda like the one on ethics, amoung some of the others.
>> But then group voting is like a conspiracy huh?

>I'll bring it up at the next meeting.  Sorry, no smiley.

The slashdot comments, that help to sort thru and clairify/correct..

http://slashdot.org/article.pl?sid=02/02/08/0541202&mode=thread&tid=156

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From roy at linksheaven.com  Thu Feb 21 18:54:10 2002
From: roy at linksheaven.com (Roy Madden)
Date: Thu, 21 Feb 2002 23:54:10 GMT
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com>
Message-ID: <CMfd8.5037$D6.14223@news.iol.ie>

On Thu, 21 Feb 2002 10:49:52 -0800, "Virginia O" <vodine at direct.ca> wrote:

>Build my database. (I will be using Access,unless convinced otherwise.)

If it was upto me I'd just use a simple delimited text file :)

Roy



-- 
http://www.linksheaven.com


From daniels at mindspring.com  Thu Feb 28 11:22:34 2002
From: daniels at mindspring.com (Alan Daniels)
Date: 28 Feb 2002 08:22:34 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <a5k7ej$d5a$1@animus.fel.iae.nl> <eppstein-F06C24.20302727022002@news.service.uci.edu>
Message-ID: <5d283efe.0202280822.13582440@posting.google.com>

David Eppstein <eppstein at ics.uci.edu> wrote in message news:<eppstein-F06C24.20302727022002 at news.service.uci.edu>...
> In article <a5k7ej$d5a$1 at animus.fel.iae.nl>,
>  Carel Fellinger <cfelling at iae.nl> wrote:
> 
> > So please explain why you think it natural to loop over a number.
> 
> Maybe it's as natural as Perl's ability to concatenate numbers?

Perl does not concatenate numbers, it concantenates *scalars*. There
is a difference. It's a standard Perlism that a scalar can be either a
numerical value or a string, depending on how it's used. Python's is not
as loosely typed (thank Guido), and just because Perl is full of
do-what-you-think-I-mean implicit conversions is no justification for
having similar idioms in Python.

I'd also like to put in my (unrequested) two cents and go on record as
disliking the "for i in 5" syntax. Not just as a matter of taste, but
because it provides unintuitive behavior, as pointed out previously.
The two lines...
    for i in 5: print i
    for i in 5, 6: print i
...would behave differently under the proposal.

A question that isn't directed at anyone in particular: Am I missing
something? I'm trying to understand this better, I really am, but is
it really all that terrible to type range or xrange?

I'll admit my bias, though. I didn't like syntax such as having "for
line in file" transparently iterate over lines when I first saw it
introduced, and I don't like it now. Hell, I've never even liked that
iterating over a string gives you its characters, and that's been in
the language as long as I can remember. All the wonderful improvements
in 2.2 made it worth the switch for me (generators are awesome), but
I'm a strong believer in the "explicit is better than implicit" rule.
I don't understand this desire so many have to give every built-in
data type an implicit iterator.


From philh at comuno.freeserve.co.uk  Sat Feb 16 13:32:14 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sat, 16 Feb 2002 18:32:14 +0000
Subject: "implemented by subclass"
Message-ID: <slrna6t9de.nt2.philh@comuno.freeserve.co.uk>

Consider the following python code:


class Log:
   def write(self, text): 
      """(implemented by subclass)"""  


Is this the most pythonic way to do it? If not, what would be
better? How about rasing a ShouldBeImplementedBySublass exception?

How do other people do it?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From mhammond at skippinet.com.au  Thu Feb  7 07:35:39 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 07 Feb 2002 12:35:39 GMT
Subject: Python method reference kills garbage collection?
References: <a3tmuv$qm1$1@news.ost.eltele.no>
Message-ID: <3C6274EA.7040006@skippinet.com.au>

Troels Walsted Hansen wrote:
> Consider the following program. What is so special about references to the
> object's own methods that cause the object to never be garbage collected? 

Just answered that.

 > Is this a cyclic reference problem,

Yes.  self.f is a bound method.  A bound method has a reference to a 
class and an instance.  The instance in this case is 'self', so you have 
a cycle.

If you really want to do this, store a reference to the unbound method.  Eg:

 >>> class A:
... 	def __init__(self):
... 		self.f = self.__class__.func
... 	def __del__(self):
... 		print "dieing"
... 	def func(self):
... 		print "func"
... 		
 >>> a=A()
 >>> a.f
<unbound method A.func>
 >>> a.func
<bound method A.func of <__main__.A instance at 0x011C9E08>>

To create a bound method from the unbound method (generally as a temp, 
just before calling it), use the new module:

 >>> import new
 >>> new.instancemethod(a.f, a, A)
<bound method A.func of <__main__.A instance at 0x011C9E08>>

And it will self-destruct OK:
 >>> a=None
 >>>

Ahh - but it didn't die!  Oh yeah - the bound method we just created 
will be stored in "_" (as this was entered interactively.  So enter any 
new expression to replace "_"

 >>> 1
dieing
1
 >>>

Yay :)

Mark.



From mdk at mdk.com  Fri Feb 15 08:13:10 2002
From: mdk at mdk.com (MDK)
Date: Fri, 15 Feb 2002 08:13:10 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org>
Message-ID: <a4j1h7$kker$1@ID-98166.news.dfncis.de>

Okay, I am conviced that I should use cPickle over evel().

However, this brings me to another question.  Athough I use Python 2.2 I
need to be sure that everything I do is compatable with 1.5.  Is there a
chart somewhere that will show me, for example, that cPickle will work with
both versions?

Thanks for all the replies.




From timr at probo.com  Mon Feb 25 02:26:33 2002
From: timr at probo.com (Tim Roberts)
Date: Sun, 24 Feb 2002 23:26:33 -0800
Subject: Puzzled: Y am I ending up with extra bytes?
References: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4>
Message-ID: <9ipj7ucf1tvktalhjecg44jn0172855k7m@4ax.com>

"A.Newby" <deathtospam43423 at altavista.com> wrote:

>Why is this happening? I read large chunks of data from a text file, 
>according to byte locations specified on another file, and for some reason, 
>this function (below), spits out a few extra bytes.
>
>Here's the code, as entered into the Python shell...... 
>
>    	index = map(string.rstrip, open('D:\cgi-bin\indx.txt').readlines())

Danger, Will Robinson!  Although this works by accident, it would NOT have
worked if your file had been "ndx.txt" instead of "indx.txt", beause the
"\n" would have been interpreted as a linefeed (0x0A).

When you need to refer to a file name in windows, you must either:
1. Use forward slashes:   open('D:/cgi-bin/indx.txt')
2. Use double backslashes:  open('D:\\cgi-bin\\indx.txt')
3. Use a "backslash suppressed" string:  open(r'D:\cgi-bin\indx.txt')
   (note the r' prefix )

Personally, I recommend option 1.  All Windows APIs accept forward slashes
in filenames; only the command shells require backslashes.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From opengeometry at yahoo.ca  Tue Feb 12 04:45:12 2002
From: opengeometry at yahoo.ca (William Park)
Date: Tue, 12 Feb 2002 04:45:12 -0500
Subject: Shell command
In-Reply-To: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net>; from ivano.delpiccolo@degroof.be on Tue, Feb 12, 2002 at 10:35:21AM +0100
References: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net>
Message-ID: <20020212044512.A9351@node0.opengeometry.ca>

On Tue, Feb 12, 2002 at 10:35:21AM +0100, DelPiccolo Ivano wrote:
> How can I do to launch a command in my python script ?
> example : I want to launch the command : rm test.txt

os.system('rm test.txt')

-- 
William Park, Open Geometry Consulting, <opengeometry at yahoo.ca>
8 CPU cluster, NAS, (Slackware) Linux, Python, LaTeX, Vim, Mutt, Tin



From cliechti at gmx.net  Fri Feb 15 22:51:39 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 16 Feb 2002 04:51:39 +0100
Subject: win extension supports RS232 binary
References: <a4kc8e$ei5$1@news.efn.org>
Message-ID: <Xns91B731FE76397cliechtigmxnet@62.2.16.82>

"Spencer Doidge" <spencer at spencerdoidge.com> wrote in
news:a4kc8e$ei5$1 at news.efn.org: 
> Starting with Isaac Barona's SerialPort extension built on the win32
> extensions, I have a receiver working now on binary data. The 
> original SerialPort handles ASCII, and you can use strlen() to 
> determine the number of bytes received.

my serial lib: http://pyserial.sourceforge.net does binary too (and it 
runs on win32, linux, jython). i had the same problems with some 
modules that only were suitable for ASCII so i made up my own lib (and 
license issues too).

> My update on Barona's module returns a tuple giving the number of 
> bytes received and a list containing those bytes.

transfering binary data in strings is more common in python. that's 
the way it works with files, sockets, struct, zipfile and many (most?) 
other things that deal with such data.

the file-like api that is more useful because you can then just switch 
sources from serial to socket to file etc. e.g. for debugging or 
extended functionality.

> In the spirit of the GNU license agreement, this source is up for
> grabs, and here and now is where I notify any interested parties 
> about that. I would be glad to email the script to anyone 
> interested. Later, when I have things polished up a bit, I suppose I 
> could post it at my web site, but I'm not sure how useful that would 
> be without periodic announcements that they are there.

> You probably have figured out that I'm a newbie at this open source
> stuff. Any suggestions about how to handle this sort of thing?

having a download on some page is good for many things. just get some 
references to your page so that google does index it and others can 
find it. 

chris

-- 
Chris <cliechti at gmx.net>



From mwh at python.net  Thu Feb 28 05:20:08 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 28 Feb 2002 10:20:08 GMT
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3C7C6252.FC779971@cosc.canterbury.ac.nz> <m37koz2trp.fsf@mira.informatik.hu-berlin.de> <3C7D8C76.1918CE52@cosc.canterbury.ac.nz>
Message-ID: <u4rk1ykff.fsf@python.net>

Greg Ewing <greg at cosc.canterbury.ac.nz> writes:

> "Martin v. Loewis" wrote:
> > 
> > Notice that regular expressions also define LL(1) languages,
> > so the entire pgen parser is a "regular expression matching engine"
> > :-)
> 
> Well, yes. But what I was trying to say is that the
> Python parser generator must be doing some sort of
> NFA-to-DFA processing on the productions in order
> to generate its tables.

Yes, I think that's right.

It might be time to bring out the quote from Parser/pgen.c:

        /* This algorithm is from a book written before
           the invention of structured programming... */

I think about all that can be meaningfully said about Python's parser
is that it works... if it was written today, I'd expect it to look
completely different.

Cheers,
M.

-- 
  I think my standards have lowered enough that now I think ``good
  design'' is when the page doesn't irritate the living fuck out of 
  me.                        -- http://www.jwz.org/gruntle/design.html


From sjmachin at lexicon.net  Sun Feb 10 21:35:13 2002
From: sjmachin at lexicon.net (John Machin)
Date: 10 Feb 2002 18:35:13 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <mailman.1013381417.32346.python-list@python.org> <3C6703F4.8F25BE12@alcyone.com>
Message-ID: <c76ff6fc.0202101835.7fcec233@posting.google.com>

Erik Max Francis <max at alcyone.com> wrote in message news:<3C6703F4.8F25BE12 at alcyone.com>...
> Neil Schemenauer wrote:
> 
> > Bengt Richter wrote:
> >
> > > Wouldn't it make sense to return list('abc') ?
> > 
> > It would also make sense to return ['a', 'b', 'c'].  Since it's not
> > obvious what you want Python raises an error.
> 
> That's what list('abc') would return.  But in actuality, for it to be
> consistent with the rest of string.split's behavior, 'abc'.split('')
> would have to return ['', 'a', 'b', 'c', ''], not ['a', 'b', 'c'].
> 
> Another beautiful example of why string.split('') should raise an error
> instead of trying to guess what you mean.

How about this for consistency:

>>> " a b c ".split(" ")
['', 'a', 'b', 'c', '']
>>> " a b c ".split()
['a', 'b', 'c']


From bnaranapatti at yahoo.com  Tue Feb 19 19:34:35 2002
From: bnaranapatti at yahoo.com (balaji)
Date: 19 Feb 2002 16:34:35 -0800
Subject: [Jython/Java] Which books, which IDEs, which other tools?
References: <mailman.1011995381.15187.python-list@python.org>
Message-ID: <2a8442d4.0202191634.672d6142@posting.google.com>

The Java Programmer tried those few lines and a few more !!
And Was I surprised and excited !!

Thanks

Balaji


James_Althoff at i2.com wrote in message news:<mailman.1011995381.15187.python-list at python.org>...
> [Jim Althoff]
> > I know what you mean.  I've seen many examples of Java programmers being
> > unwilling to use Jython because they felt it would make their resumes/CVs
> > less marketable.  (Interestingly, the ones who have been open-minded and
> > willing to learn Jython often later feel that the ease of using Java APIs
> > in conjunction with Jython actually *improves* their Java knowledge.)
> 
> [Dave/brueckd at tbye.com]
> >Yeah, the save-to-file-then-compile-now-run-it process discourages
> >experimentation and penalizes mistakes. Jython makes it nearly
> >effortless to go poking around in different packages and classes to
> >see what's there.
> 
> Absolutely.  One of my favorites is to demo this to a Java programmer and
> watch the startled reaction:
> 
> C:\_Dev\pnp\build>jpython
> Jython 2.1 on java1.3.0 (JIT: null)
> Type "copyright", "credits" or "license" for more information.
> >>> from javax.swing import JFrame
> >>> frame = JFrame('Hello, world!')
> >>> frame.setVisible(1)
> >>>
> 
> ... and then add buttons with handlers to the frame interactively!  :-)
> 
> Jim


From oren-py-l at hishome.net  Sun Feb  3 08:02:17 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Sun, 3 Feb 2002 08:02:17 -0500
Subject: Enhanced Generators - reiterating over a file
In-Reply-To: <837kpvz2mw.fsf@panacea.canonical.org>
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org> <83g04kzoy2.fsf@panacea.canonical.org> <mailman.1012677065.886.python-list@python.org> <837kpvz2mw.fsf@panacea.canonical.org>
Message-ID: <20020203130217.GB9924@hishome.net>

On Sat, Feb 02, 2002 at 10:05:11PM -0500, Kragen Sitaker wrote:
> > > That won't work when passed iterators as arguments, right?
> > 
> > Sure, but as long as iterators are kept as invisible temporary objects
> 
> Um, sure.  iter(open("hello")) doesn't work that way already.

This is a real life example of how useful reiterable x functions can be
and how to solve the problem of reiterating over a file.

The following function takes a vector of samples and yields a stream of samples 
with their values normalized to the range +-1:

def normalized(vector):
    max_value = 0.0
    for sample in vector:
        max_value = max(max_value, abs(sample))
    for sample in vector:
        yield sample/max_value

Note that this must be a two-pass operation: you cannot yield the first
normalized sample before you find the maximum value.

My data is in a text file with one decimal sample per line.  This code reads
it and writes the normalized samples to another file in the same format.

vector = map(float, file('samples.dat'))
outfile = file('normalized.dat','w')
for sample in normalized(vector):
    print >>outfile, sample
outfile.close()

This works in Python 2.2.  The only problem is that it reads the entire file
into memory.  What if the file is too big to read into memory or I just don't 
want to stress virtual memory unnecessarily?

Just add the magic x! change map->xmap, file->xfile and everything works 
exactly the same but without any temporary lists.

xfile is a lazy file object: the object just stores the filename.  Only when 
iter(xfile('filename')) is called the returned iterator object opens a 
temporary file descriptor to walk through the file.

xmap must be the truly lazy version that returns an iterable object, not the 
half-eager half-lazy version that returns an iterator.  This is because the 
function normalized() has to scan the source twice.

An xfile object really simulates a container - you can use iter() to get 
multiple independent iterators of the same container.  It should also appeal 
to the fans of a certain TV show :-)

A Python file object is not really a container: it can be argued that a file 
object already *is* a kind of iterator.  It is a temporary object used to walk 
through a container.  The real container in this case is the actual file on 
the disk.  An xfile object represents a file on the disk.

   x-ly yours,

	Oren




From marklists at mceahern.com  Wed Feb  6 23:42:59 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Wed, 6 Feb 2002 20:42:59 -0800
Subject: money class?
In-Reply-To: <000001c1af86$d050f6d0$445d4540@Dell2>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEFBKBAA.marklists@mceahern.com>

Brian Quinlan wrote:
> Are you looking for any class that happens to be named "Money" or are
> you looking for some particular functionality?

I did a little more searching and for now, at least, it seems like Tim
Peters' FixedPoint.py is more than adequate:


ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoin
t.py.Z

Cheers,

// mark




From stojek at part-gmbh.de  Thu Feb 14 08:41:43 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Thu, 14 Feb 2002 13:41:43 GMT
Subject: searching a list of tuples
References: <3c6aabeb.6302984@news.easynews.net>
Message-ID: <3c6bbe79.12166593@news.easynews.net>

just what I need. 
Tank you.


From sholden at holdenweb.com  Mon Feb 11 14:20:07 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 14:20:07 -0500
Subject: Zope cookbook?
References: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au> <120220020143076768%lee@leeHYPHENphillips.org.invalid>
Message-ID: <wPU98.153435$KM2.6924559@atlpnn01.usenetserver.com>

"Lee Phillips" <lee at leeHYPHENphillips.org.invalid> wrote in message
news:120220020143076768%lee at leeHYPHENphillips.org.invalid...
> You might be happier with webware, or one of the other frameworks
> (skunkweb, ...,).

You can at least get a Zope to try out from www.freezope.org.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jjl at pobox.com  Wed Feb 27 18:04:32 2002
From: jjl at pobox.com (John J. Lee)
Date: Wed, 27 Feb 2002 23:04:32 +0000
Subject: 15 minute presentation on python?
In-Reply-To: <Pine.SOL.4.30.0202272224160.15823-100000@mimosa.csv.warwick.ac.uk>
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de> <3C7ADE12.C6D895A0@engcorp.com> <fa7108b0.0202271256.57e27d82@posting.google.com> <Pine.SOL.4.30.0202272224160.15823-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <Pine.SOL.4.30.0202272303570.16406-100000@mimosa.csv.warwick.ac.uk>

On Wed, 27 Feb 2002, John J. Lee wrote:

> On 27 Feb 2002, Chris wrote:
>
> > Peter Hansen <peter at engcorp.com> wrote in message news:<3C7ADE12.C6D895A0 at engcorp.com>...
> [...]
> > I agree--if you have Internet access, do something that grabs
> > something from a web page.  I also did the above (based on an example
> > in Learning Python to grab the temp for a city from a weather site)
> > and it wowed me.
> [...]
>
> The 'wow' isn't so much the fact of being able to do it, as the length of
> the xxxlib modules that implement the various internet protocols.

Ok, changed my mind: readability is what I really mean.


John



From miah at picsel.com  Wed Feb 27 04:52:53 2002
From: miah at picsel.com (Miah Gregory)
Date: Wed, 27 Feb 2002 09:52:53 GMT
Subject: cross platform (linux/windows) problem
Message-ID: <f5c74c0f4b.miah@picsel.com>

Hi,

Looking through the google newsgroup archive, I've noticed that other
people have had similar problems to mine, but I couldn't find a solution
which fitted me well enough.

I'm trying to write a script, which will run on windows and linux. It's job
is to loop continuously, running an executable passed in as a command line
parameter each time, until that executable exits with a non-zero exit code.
Some exit codes, however, should not stop the script, so I need to know
the specific exit code, and not just whether it was non-zero or not.

The other requirement is that the stdout/stderr from the executable being
run should not be displayed normally, but I need a way to actually display
that output when a suitable exit code is returned.

Simple example which works for linux:

import popen2

testapp = ""
for arg in sys.argv[1:len(sys.argv)]:
    testapp = testapp + " " + arg

while 1:
    p = popen2.Popen3(testapp)
    p.wait
    
    retval   = p.poll()
    exitcode = (retval >> 8) & 0xFF
    signal   = retval & 0xFF
    
    if (exitcode != 0):
        if (exitcode != 2):
            print p.fromchild.read()
            sys.exit(1)
    else:
        sys.exit(0)


This however doesn't work for windows, as it seems the popen call isn't
implemented there.

I've also noticed that the return code from os.system under win2k seems
to be reversed, ie. the exitcode is returned as the signal in the above
script.

This is ok, though, as I can do an explicit check to get around that:

if sys.platform == "win32":
    exitcode = signal
    signal   = 0

Does anyone have any suggestions as to how, if at all, I can make the above
work on both platforms?

Thanks in advance.

-- 
Miah Gregory


From cbbrowne at acm.org  Fri Feb 22 16:18:59 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Fri, 22 Feb 2002 16:18:59 -0500
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org>
Message-ID: <m37kp5i5a4.fsf@chvatal.cbbrowne.com>

Centuries ago, Nostradamus foresaw when Christian Tismer <tismer at tismer.com> would write:
> Dear Stackless Python Community,
>
> As you might have noticed, Stackless has changed quite
> much:
> - Platform independence is gone.

I hope not...  I would hope to hear that "platform indepedence has
_ARRIVED_"
-- 
(reverse (concatenate 'string "ac.notelrac.teneerf@" "454aa"))
http://www3.sympatico.ca/cbbrowne/advocacy.html
REALITY is an illusion that stays put.


From robin at jessikat.fsnet.co.uk  Sat Feb  9 20:34:17 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 10 Feb 2002 01:34:17 +0000
Subject: cgi authoriszation
Message-ID: <q4Sr5IAZ4cZ8EwGa@jessikat.fsnet.co.uk>

Is there any way to get the cgi module to do authorization. I can make scripts
demand authorization by including headers as follows

       print 'Status: 401 Authentification required'
       print 'WWW-Authenticate: Basic realm="dingo"'
       print 'Content-type: text/html'
       print
       print '<html><head></head><body>Authorisation Needed</body></html>

but I don't seem to see either 'authorization' or 'http_authorization' in the
client's response. Is apache filtering out the credentials? 
-- 
Robin Becker


From mclay at nist.gov  Fri Feb 15 21:43:33 2002
From: mclay at nist.gov (Michael McLay)
Date: Fri, 15 Feb 2002 21:43:33 -0500
Subject: [Python-Dev] PEP 282: A Logging System  --  comments please
In-Reply-To: <20020215164333.A31903@ActiveState.com>
References: <20020215164333.A31903@ActiveState.com>
Message-ID: <200202160247.g1G2lkJ5029529@email.nist.gov>

On Friday 15 February 2002 07:43 pm, Trent Mick wrote:
> Howdy all,
>
> I would appreciate any comments you might have on this proposal for adding
> a logging system to the Python Standard Library. This PEP is still an early
> draft so please forward your comments just to me directly for now.

I scanned the PEP and didn't find a reference to the logging package 
supporting logging over a network. 

> Influences
>
>     This proposal was put together after having somewhat studied the
>     following logging packages:
>
>         o java.util.logging in JDK 1.4 (a.k.a. JSR047) [1]
>         o log4j [2]
>           These two systems are *very* similar.
>         o the Syslog package from the Protomatter project [3]
>         o MAL's mx.Log package [4]
>
>     This proposal will basically look like java.util.logging with a
>     smattering of log4j.

Marshal Rose submitted RFC3195[1] to the IETF for a syslog protocol.  The 
specification is defined as a profile on top of the BEEP framework.  The 
format of the messages are encoded in XML. Here is an example of an "entry" 
element.  

     C: <entry facility='8' severity='6' 
      C:   hostname='pipeworks'
      C:   timestamp='Oct 31 23:59:59'
      C:  >&lt;.....eeeek!</entry>

[1] http://www.beepcore.org/beepcore/docs/rfc3195.html.



From quinn at regurgitate.ugcs.caltech.edu  Thu Feb 21 14:00:11 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 21 Feb 2002 19:00:11 GMT
Subject: giant stdlib tuples (was: Re: Help with code)
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com> <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu> <uadu3m5dz.fsf@python.net>
Message-ID: <slrna7agtq.1vk.quinn@regurgitate.ugcs.caltech.edu>

On Thu, 21 Feb 2002 11:35:37 GMT, Michael Hudson <mwh at python.net> wrote:
>quinn at schilling.ugcs.caltech.edu (Quinn Dunkan) writes:
>> It's always seemed to me that things like time, pwd, grp, and stat
>> are pretty non-pythonic.  You can't even really say it's because of
>> the thin wrapper thing because writing and using a StructType or
>> something wouldn't be much harder than using tuples.
>> 
>> Of course, I've learned to carry around my own little wrappers, but
>> I've always wondered why they were that way in the first place.
>
>Used Python 2.2 yet?

Aha!  I hadn't seen this one advertised anywhere, and they disguise themselves
as tuples.  pwd and grp still don't return structs, but it's a good start.

Deserves more advertisement, I think:

The days of 'os.stat(fn)[stat.ST_BLAH]' are over.

I still wonder about those wacky C structure names, though.  Did they think
people would forget what kind of structure they were asking about after they
typed the '.'?


From amanjit.gill at gmx.de  Tue Feb 19 22:10:53 2002
From: amanjit.gill at gmx.de (Amanjit Gill)
Date: Wed, 20 Feb 2002 04:10:53 +0100
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <a4v443$lr0$03$1@news.t-online.com>

"Craig Maloney" <cmaloney at physics.ucsb.edu> schrieb im Newsbeitrag
news:3C7283E4.4050604 at physics.ucsb.edu...
>
> I'm doing some scientific computing and have
> amassed a moderate set of classes.  Due to the
[..]

Take a look at CERN's root framework, which uses CINT at the backend, a
C/C++ Interpreter running something like a virtual machine, which can
directly call DLLs too (which run natively, of course)..

You could use CINT alone, too.
Start scripting in C / C++ :-)

http://root.cern.ch/
http://root.cern.ch/root/Cint.html

If you use NT, you might like to check out Cygwins utilities for bash.





From dooberwah at yahoo.com  Mon Feb 25 17:42:43 2002
From: dooberwah at yahoo.com (Ben Jacobs)
Date: Mon, 25 Feb 2002 17:42:43 -0500
Subject: Website to post code for peer review
Message-ID: <20020225174243.A1563@yahoo.com>

Hello everyone, 
	I'm new to python and haven't totally found my way all around the
community yet. I looking for a good web site where I can submit the code I
write to be looked at and reviewed by other programmers. What I have in mind
is something like http://www.perlmonks.org, but for python related stuff
(please don't flame me because I like perl as well as python).
	PerlMonks is a large web site where perl hackers can hang our, swap
code, and chat about perl or programming in general. It allows people to
reply and vote on code you write. People can gain experience points and
different levels depending on how many votes their code gets.
	comp.lang.python offers some of these of thing, but from my brief
stay here I've felt that it's mostly a place to get your programming
questions answered. I've looked at the http://www.python.org "community"
page, but none of the sites on the list seem to have what I'm looking for.
>From what I've seen, http://starship.python.net is a just a place for python
programmers to host their web pages. It lacks the kind of organization I'm
interested in. If you know of any sites that provide these kinds of
services, please post their urls. Thanks.

-- 
-Ben Jacobs
http://dooberwah.perlmonk.org
"one thing i can tell you is you got to be free."



From jim at jim.jim  Mon Feb 18 14:13:08 2002
From: jim at jim.jim (jimh)
Date: Mon, 18 Feb 2002 11:13:08 -0800
Subject: Python calling Java
Message-ID: <a4rjq4$4n8$1@web1.cup.hp.com>

We have a bunch of Python (cgi) scripts producing html behind an Apache web
server.  We also have a set of Java libraries that I want to access from
these Python scripts, but I haven't been able to figure out how to do this.
I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
a one-line 'print "hi"' script, and this is, of course, not acceptable for a
web server.

Any ideas?





From francis.meyvis at sonycom.com  Thu Feb 28 10:09:12 2002
From: francis.meyvis at sonycom.com (Francis Meyvis)
Date: Thu, 28 Feb 2002 15:09:12 GMT
Subject: HTTP chunk Transfer-Encoding
References: <4Wpf8.13$zd.865@nreader2.kpnqwest.net> <Byqf8.135$5vh.170870272@news.telia.no>
Message-ID: <sKrf8.24$kf3.922@nreader1.kpnqwest.net>

Found myself the answer in Zope & RFC 2616

in do_GET(self) when sending the headers add

SimpleHTTPRequestHandler.send_header(self, "Transfer-Endcoding", "chunked")

when sending the data (imagine it is in variable d)

self.wfile.write("%x\r\n%s\r\n" % (len(d), d))
self.wfile.write("0\r\n\r\n" )

---
Kind regards, best wishes,
francis meyvis
Thomas Weholt <thomas at gatsoft.no> wrote in message
news:Byqf8.135$5vh.170870272 at news.telia.no...
> Medusa has some code for that. I'm looking into implementing that myself
but
> don't have the time right now. Please post any code or findings you can
> share.
>
> Best regards,
> Thomas Weholt
>
>
> "Francis Meyvis" <francis.meyvis at sonycom.com> wrote in message
> news:4Wpf8.13$zd.865 at nreader2.kpnqwest.net...
> > Hoi,
> >
> > Is there some python source that sends chunked packets (RFC 2616) in
> > the SimpleHTTPServer or BaseHTTPServer?
> > Or do you know where to find snippet of code that can split up
> > file to send them in chunk format?
> >
> > Thx
> >
> > --
> > ---
> > Kind regards, best wishes,
> > francis meyvis
> >
> >
>
>




From roy at linksheaven.com  Thu Feb 21 19:02:10 2002
From: roy at linksheaven.com (Roy Madden)
Date: Fri, 22 Feb 2002 00:02:10 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56>
Message-ID: <6Ufd8.5038$D6.14223@news.iol.ie>

On 19 Feb 2002 21:31:00 GMT, Lutz Schroeer
<Lutz.Schroeer at kybernetik-manufaktur.de> wrote:

>I'm creating a website which makes intensive use of cgi scripts and there 
>is the possibility for the visitor to register as a user for having the 
>ability to customize the site appearance, contents and other stuff.
>
>Jumping from one script to the other I need to remember at least the user's 
>name. The W3C recommends using cookies (RFC 2109). Unfortunately many 
>people don't like cookies and filter them using WebWasher or a personal 
>firewall. 
>
I'm going to try and dig up some stats for you, but from experience and
research the population of those who do not accept cookies is very very
small (albeit tend to be technically proficient and vocal online). I once
worked for an online bank with operations in Germany, and we considered
this issue *very* carefully before choosing cookies for our user
interface.

A slightly larger population do not 'like' them, but of the 3 methods of
maintaining session state they are the most secure option (waiting for the
flames to start :) ) - if security is an issue for you, and the 'like'
issue is a matter of user education.


Roy



-- 
http://www.linksheaven.com


From nhodgson at bigpond.net.au  Mon Feb 25 19:18:17 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Tue, 26 Feb 2002 00:18:17 GMT
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <dvAe8.10654$RV3.41715@news-server.bigpond.net.au>

Moray Taylor:

> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.

   By Class/function list do you mean a list of classes and functions in the
current file displayed either as a menu or as a list in a pane next to the
text? PythonWin has that but it is only on Windows. This would not be that
hard to add to one of the open source editors. Boa Constructor, a big
complex unfinished free project, has a source browser as does the commercial
Wing IDE.

http://boa-constructor.sourceforge.net/
http://archaeopteryx.com/wingide

   Neil




From djc at object-craft.com.au  Fri Feb  8 04:08:10 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 08 Feb 2002 20:08:10 +1100
Subject: Saving Browser State without Cookies
References: <D408B6AC060DB24790622E1A54204DB00823AA89@xchange.rightnowtech.com> <mailman.1013113454.24778.python-list@python.org>
Message-ID: <m3sn8ccpdx.fsf@ferret.object-craft.com.au>

>>>>> "Tim" == Tim Churches <tchur at optushome.com.au> writes:

Tim> I believe that Object Craft plan to host some live demos of
Tim> Albatross apps on their Web server in the not too distant future,
Tim> as well as provide some more sample apps as starting points for
Tim> would-be Albatross developers.

Tim is correct, in the next release we are going to place some/all of
the samples up on our web server.  We are also going to change all of
the URLs in the documentation so you can try things out without
needing to download and install Albatross.

In time for the next release we also plan to set up some mailing
lists.

- Dave

-- 
http://www.object-craft.com.au


From martin at v.loewis.de  Sat Feb  2 06:12:16 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 02 Feb 2002 12:12:16 +0100
Subject: Xmi parser
References: <3c5bb698$0$4609$626a54ce@news.free.fr>
Message-ID: <m31yg49lyn.fsf@mira.informatik.hu-berlin.de>

"Bruno Li?nard" <lienard.bruno at free.fr> writes:

> I am looking for a xmi parser. Does anybody here have already
> written one ?

Given that XMI is an XML application, any XML parser should be able to
parse XMI. I recommend xml.parsers.expat (or the SAX and DOM wrappers
around it).

Regards,
Martin



From shalehperry at attbi.com  Wed Feb 27 19:13:25 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 27 Feb 2002 16:13:25 -0800 (PST)
Subject: Regular expression problem
In-Reply-To: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com>
Message-ID: <XFMail.20020227161325.shalehperry@attbi.com>

On 27-Feb-2002 Asheesh Laroia wrote:
> I have some SGML input (PageMaker 6.5 tagged text), and I want to be able
> to recognize (and delete) a tag.  That tag looks like:
> 
>       <@Trap Body text:>
> 
> It may also look like <@Trap Body text: useless-data>.
> 
> So, I tried the regular expression r"<@.?>".  That doesn't match the
> above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
> using, and why doesn't this work?
> 

There is a really nice SGML parser in python's library.  Perhaps it will help
you.  The name is 'sgmllib'.



From wookiz at hotmail.com  Sun Feb 10 03:22:57 2002
From: wookiz at hotmail.com (wooks)
Date: 10 Feb 2002 00:22:57 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu>
Message-ID: <8852d33c.0202100022.70a05925@posting.google.com>

quinn at hork.ugcs.caltech.edu (Quinn Dunkan) wrote in message news:<slrna6bord.9tv.quinn at hork.ugcs.caltech.edu>...
> 
> No, you're being too CS-PhD-y.  "poly-what?"  Yes, python is polymorphic, but
> there's no need to get all fancy-pants about it.  

You are joking right. Just in case you're not. Well no polymorphism
isn't part of my standard vocabulary. Being able to relate a Python
concept to an OO concept helps clarify the raison d'etre of said
python concept (assuming it is not obvious). I then have the option of
going off to look for an idiots guide to polymorphism to further help
my understanding.

>I suspect that neither the 2
> classes subclassed UserDict because their author didn't feel like it.  UserDict
> defines a few methods for you, and can be convenient, but you don't hafta use
> it.  I make dict-likes all the time, but rarely use UserDict, because it
> insists I have an appropriate 'data' attribute.
> 

Maybe, but we're talking about a book that aims to educate  not ones
personal coding prejudices. I still think my guess is right and
explains why userdict was not even mentioned much less used.

> The magic methods are methods that are magically called by non-method-looking
> syntax.  So len(o) actually calls o.__len__() (for historical reasons, IMO),
> and x[y] actually calls x.__getitem__(y) (since 'x[y]' is easier to type), and
> x[y] = z calls x.__setitem__(y, z) (since it's more natural looking to people
> used to infix algol-esques).
> 
> They don't have anything to do with polymorphism per se, but of course they
> make it easier, as any time you define a general-purpose widely-used method
> for your object (i.e. x[y] aka __getitem__ is widely-used so if you define it
> your object can work with any __getitem__-using code).
> 

So they are just syntatical sugar. That description would immediately
explain their raison d'etre. Shame it wasn't used. Anyway now you are
saying that my understanding of polymorphism isn't what it was so it
looks like I need to go and reread. I think I ought to understand it
as it is a fundamental OO concept which I would like to be able to
recognise when I come across it.


From logstx at bellatlantic.net  Wed Feb 20 00:26:21 2002
From: logstx at bellatlantic.net (logistix)
Date: Wed, 20 Feb 2002 05:26:21 GMT
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org>
Message-ID: <1sGc8.3599$9H5.1708@nwrddc01.gnilink.net>

>
> Then I wrote my own hashtable and my own string type and managed
> to get the C++ down to about 8x faster than the Python.
> It solves their WORD.LST in <1 second on my PC.
>
> I solved the other puzzles too...  They still didn't hire me.  :)

They were probably asking "Hasn't this guy heard of STL?" ;-)




From phr-n2002a at nightsong.com  Mon Feb  4 04:05:15 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 01:05:15 -0800
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com> <7xbsf7dqj9.fsf@ruckus.brouhaha.com> <B883FC7B.4309%jonathan@onegoodidea.com>
Message-ID: <7xu1sxzkfo.fsf@ruckus.brouhaha.com>

Jonathan Hogg <jonathan at onegoodidea.com> writes:
> > So much for compatibility then.
> 
> I think I missed the beginning of this thread so I'm confused. Exactly which
> new backwards incompatible feature of the recent Python versions is it that
> is breaking people's old scripts?
> 
> So far I've seen complaints of 'x in somedict' which only affects writing
> new code for the new Python interpreter, future division which is not
> enabled yet by default, and the import __future__ mechanism, which again is
> only meant to be called by new code on the new interpreter.

That's the problem--if I don't import future division, my code will
break in the future, because the // operator isn't supported in the
present.  But if I -DO- import future division, my code will break in
the present, because current (2.1) and older (1.5) interpreters don't
recognize the import statement.  (The most recently released version
of Python is a 2.1 series release, so 2.1 is still current).

I don't see any reasonable way I can write code right now that uses
division that works in both the current 2.1 (which doesn't support
importing future division) and in forthcoming versions (where the
semantics of the / operator will be incompatibly different).

> If you need to run code on 1.5.2 then don't use any of the new features;
> this is a given. This goes just as much for library evolution as core
> language changes.

> Which *old* code of someone's has actually been broken so that it will not
> run correctly on the newer interpreters?

Redhat installation scripts for one--that may be why they still ship 1.5.2.


From syring at email.com  Thu Feb 14 15:18:34 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 14 Feb 2002 21:18:34 +0100
Subject: AttributeError: 'getopt' module has no attribute 'GetoptError'
References: <Xns90D3ECC92390Dgustaflalgonetse@194.213.69.152> <ku7kxp32z3.fsf@lasipalatsi.fi> <Xns90D4988CE4ADEgustaflalgonetse@194.213.69.152> <u6o53mb595981a@corp.supernews.com>
Message-ID: <a4h64c$99rm$1@ID-7529.news.dfncis.de>

"Travis Shirk" <travis at puddy.lan.kerrgulch.net> schrieb
> Gustaf Liljegren <gustafl at algonet.se> wrote:
> > Erno Kuusela <erno-news at erno.iki.fi> wrote:
> 
> >>> I use Python 2.1 on Win98.  Try to use the getopt module, but get
> >>> the error in the subject line. What's wrong?
> 
> >>your test program isn't called "getopt.py" by any chance?
> 
> > Oh. Yes, it is... Thank you.
> 
> I'm having the exact same problem as Gustaf, and
> my program is *not* called getopt.py.  Anyone know what the
> the AttibuteError is all about?

There is not accidentally a file getopt.pyc laying around?

Once bitten, twice shy
Karl M. Syring



From stojek at part-gmbh.de  Fri Feb 22 10:50:46 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Fri, 22 Feb 2002 15:50:46 GMT
Subject: wxlayoutconstraints inside a wxDialog ?
Message-ID: <3c76666e.18420093@news.easynews.net>

Hi,

I'm using wxPython and I'm trying to use
layout constraints in a Dialog.
why doesn't this work?

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

d=CompoDlg(self)
-----
class CompoDlg(wxDialog):
    def __init__(self,parent):
        BORDERPIX=20
        ID_E11=wxNewId()

        win=wxDialog(parent, -1, "Title", wxDefaultPosition,
                               wxSize(350, 200))
        win.SetAutoLayout(true)

        E11 = wxStaticText(win,-1,"E11:",wxDefaultPosition,
                                       wxDefaultSize)
        lc=wxLayoutConstraints()
        lc.top.SameAs(win,wxTop,BORDERPIX)
        lc.left.SameAs(win,wxLeft,BORDERPIX)
        lc.width.AsIs()
        lc.height.AsIs()
        E11.SetConstraints(lc)
--------------------------------

Thanks,
Marcus


From gvdbogae at vub.ac.be  Sat Feb 16 17:23:54 2002
From: gvdbogae at vub.ac.be (Geert-Jan Van den Bogaerde)
Date: 16 Feb 2002 23:23:54 +0100
Subject: FSF award to...
In-Reply-To: <498.812T2576T10686134threeseas@earthlink.net>
References: <498.812T2576T10686134threeseas@earthlink.net>
Message-ID: <1013898235.1573.4.camel@gandalf>

For the curious, it was awarded at FOSDEM (Free Software and Open Source
Developers European Meeting) in Brussels (Belgium) just a few hours ago.
In fact, I just got back from there.

Guido was there to receive the award (though I don't think he's actually
giving a talk at FOSDEM unfortunately).

RMS commented that he personally has not used Python, but has been told
it is a very good language, and that it incorporated many elements from
LISP.

Geert-Jan

On Sat, 2002-02-16 at 23:02, Timothy Rue wrote:
> http://www.fsf.org/press/2002-02-16-FSF-Award.html
> 
> Given what I've seen here.....
> 
> Well I guess Usenet ain't part of .....
> 
> ---
> *3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
>    *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
> Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
> Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
> Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 





From feldspar at ix.netcom.com  Fri Feb 15 15:06:15 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Fri, 15 Feb 2002 16:06:15 -0400
Subject: Basic Question
References: <m3y9huzd3a.fsf@mycomp.tky.hut.fi>
Message-ID: <3C6D6A37.EA8B1893@ix.netcom.com>

Karthikesh Raju wrote:
> 
> Hi,
> 
> Just started, so my question might be stupid. i tried this:
> 
> import random
> 
> def randomList(n):
>     s = [0] * n
>     for i in range(n):
>         s[i] = random.random()
>     return s
> 
> in a file called randomList.py. Started python, imported randomList
> (>> import randomList) and then called this function (>>
> randomList(5)), got this un-understandable error:
> 
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: object of type 'module' is not callable
> 
> But when i when type the function at the python command prompt, it
> works. Couldnt figure out why..
> 
> Thankx in advance,
> 
> Best wishes
> karthik

When you put a function named "functionName()" in a file called
"moduleName.py", the name of that function in another namespace (such as
that of the interpreter) is not "functionName()" but
"moduleName.functionName()".  (There's an exception, if you use the
"from <module> import <names>" statement, but I don't recommend that
until you've got the basics down.)  You can see this if you look at the
code inside your function:  to get a random number, you called
"random.random()" -- i.e., the "random()" function inside the "random"
module.

Your module name and your function name happen to be identical.  When
you typed in your function name, unqualified by a module name, the
interpreter looked for anything in its namespace matching "randomList". 
It found an object that matched that name -- but it wasn't the function
*in* the module, it was the module itself, which had the same name.

When you typed the function at the interpreter prompt, *then* you
established a function with the name randomList in the interpreter's
namespace.  And then, of course, it worked.

If you want to call the function in the module, you'd have to use
"randomList.randomList(n)".

If you want to see what's in your interpreter's namespace at a given
time, you can use the "dir(<name>)" command.  Try it before you import a
module, and after, and try using a name returned by one call to "dir()"
as an argument to the function.  It may help you understand better how
Python namespaces work.

	-jc


From skip at pobox.com  Thu Feb 28 22:21:51 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 28 Feb 2002 21:21:51 -0600
Subject: ASV module, CVS modules
In-Reply-To: <61ff32e3.0202281525.3e9bb102@posting.google.com>
References: <61ff32e3.0202281525.3e9bb102@posting.google.com>
Message-ID: <15486.62415.486257.652769@12-248-41-177.client.attbi.com>

    ron>     file = open(input_file, "rb")
    ron> IOError: invalid argument: rb

    ron> I get a similar message from CSV module. Anyone use either of these
    ron> and can tell me what's wrong?

Sounds to me like your builtin open function is hosed somehow.  Opening
files using "rb" works for me:

    >>> f = open("cities.csv", "rb")
    >>> 

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From bldrake1 at yahoo.com  Thu Feb  7 08:36:38 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 7 Feb 2002 05:36:38 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <fb072181.0202060723.e400eaa@posting.google.com> <ug04da4k1.fsf@python.net>
Message-ID: <fb072181.0202070536.7e1bffcd@posting.google.com>

Michael,
Sorry you missed my point.  Try "Intro to Matrix Computations" by
Stewart, first few chapters for the numerical portion.  As for the
double versus float, that is what I read in Beasley's book, which I
mentioned.  I'm new to Python and obviously don't have the vast
experience you have.  But, I'll keep trying.

Regards,
Barry Drake

Michael Hudson <mwh at python.net> wrote in message news:<ug04da4k1.fsf at python.net>...
> bldrake1 at yahoo.com (Barry Drake) writes:
> 
> > Maybe I'm missing something, but here is what I get:
> 
> Well, I'm missing your point.
> 
> > Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
> > win32
> > Type "copyright", "credits" or "license" for more information.
> > IDLE 0.8 -- press F1 for help
> > >>> import math
> > >>> n = 10
> > >>> nf = 10.0
> > >>> a = 1.0
> > >>> b = 0.5
> > >>> a/n
>  0.10000000000000001
> > >>> b + a/n
>  0.59999999999999998
> > >>> a/nf
>  0.10000000000000001
> > >>> b + a/nf
>  0.59999999999999998
> > >>> 
> > 
> > Python does double precision floats only (see Python Essential
> > Reference 2nd ed. by David Beazley, p.23).
> 
> But you can have arrays (as in, objects of type array.ArrayType) that
> store C single floats.  Which was Jason's point.
> 
> > Python implements IEEE 754: 17 digits of precision with exponent in
> > -308 to 308.
> 
> Oh No It Doesn't.  Python uses (for the objects of type
> types.FloatType) whatever the C compiler that compiled it called
> "double".  This may be IEEE 754 doubles on the platforms you've used,
> but there's certainly no guarantee of that (e.g. Crays, IBM boxes (I
> think)).
> 
> >          float     double
> > python   64 bits   64 bits
> > C        32 bits   64 bits
> 
> "python double" is meaningless.
> 
> > BTW, you might want to check out the Numeric module in the NumPy
> > distribution at http://numpy.sourceforge.net.
> 
> That lets you store floats or doubles too.
> 
> Cheers,
> M.


From ChuckEsterbrook at yahoo.com  Sat Feb 23 16:29:29 2002
From: ChuckEsterbrook at yahoo.com (Chuck Esterbrook)
Date: Sat, 23 Feb 2002 21:29:29 GMT
Subject: Object counts
Message-ID: <ZQTd8.89849$B61.28524193@news1.rsm1.occa.home.com>

Hi,

Is there a way to get Python to report the number of instances of all 
classes and types? e.g., a dictionary that would map the names of 
types/classes to their count, which would be incremented on creation 
and decremented on deletion?

If not, how hard would it be to patch Python for this? Seems like all 
creation and destruction would eventually go through a couple places in 
the Python C implementation.

This would be useful for understanding how your program uses memory 
(kind of like the memory version of profile) and also for hunting down 
memory leaks.

-Chuck


From huaiyu at gauss.almadan.ibm.com  Mon Feb 25 15:28:54 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Mon, 25 Feb 2002 20:28:54 +0000 (UTC)
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <slrna7l7k6.74f.huaiyu@gauss.almadan.ibm.com>

On 25 Feb 2002 08:18:39 -0800, Moray Taylor <mtaylor at lowebroadway.com> wrote:
>
>Can anyone reccommend a good python editor for UNIX/linux/IRIX. 
>
>Before all you Emacs/vi fanatics start ranting at me, I have tried,
>and hate, them both.
>
>I have only two requirements for my editor, syntax coloring, and a
>Class/function list.

This might not help you, but it may be of interest to other Emacs users:

M-x speedbar 

gives you a hierarchical color-coded list of modules, classes, etc.  Use +
and - to expand and collapse the sublists.  You can jump to specific method
in specific class in specific module by hitting Enter.  It works for other
languages, too.

Huaiyu


From starx at pacbell.net  Sat Feb 16 06:00:35 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 11:00:35 GMT
Subject: Windows n' stuff <- Re: IPython 0.2.5
References: <a4ainp$gts$1@peabody.colorado.edu> <fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com> <Xns91B516D3A1E11RASXnewsDFE1@130.133.1.4> <mailman.1013705691.7816.python-list@python.org>
Message-ID: <Xns91B71E975386ARASXnewsDFE1@130.133.1.4>

Cliff Wells || Thu 14 Feb 2002 08:44:09a:

> On 14 Feb 2002 10:14:41 GMT
> Philip Swartzleonard wrote:
> 
>> Then there are those of us who refuse to upgrade to XP because of
>> MS's recent we-own-your-computer
>> you're-a-pirate-until-proven-otherwise 
> paying-
>> customers-come-dead-last registration BS, among other things. (And 
> 
> I wonder how much this is going to affect the size of the Windows
> user-base in the long run... hopefully for the better ;)

=)

>> generally think 98 is the best thing to come out of them in general
>> :). 
> 
> Ick.  If you _must_ use Windoze, by all means, get Win2k.  

Haeh... maybe i will next time i feel like reinstalling, i suppose it's 
been out long enough by now.

I've been thinking more and more lately though, of moving to linux 
completly. I've played with it a couple of times, but never really 
gotten it to the point where i can do everything i want. Part of the 
problem is my video card (voodoo3) seems to suck under it, but i've been 
wanting a geforce about long enough to finally go out and get one. A 
better email client (eudora is wearing on me anyway), a better news 
reader (or get Xnews to work on linux more likely =) are most of what i 
have left to sort out. Assuming gimp works properly with a tablet under 
L, dont' need photoshop... maybe something for the scanner, and a url-
manager. Oh, and something to replace Thumbs+ ... that will be hard to 
do.

Oh... i should stop rambling. Just thinking over loose ends =) 



-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From max at alcyone.com  Sat Feb  2 17:33:58 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Feb 2002 14:33:58 -0800
Subject: ANN: church 1.0 -- Lambda calculus explorer
Message-ID: <3C5C6956.82C5A1AE@alcyone.com>

Here's the second of the Alcyone PyOS modules that has been released:

    http://www.alcyone.com/pyos/church/


Summary

    This module allows simple experimentation with the lambda
    calculus, first developed by Church.  It understands the different
    types of lambda expressions, can extract lists of variables (both
    free and bound) and subterms, and can simplify complicated by
    expression by means of application.


Getting the software

    The software is available in a tarball here:
    http://www.alcyone.com/pyos/church/church-1.0.tar.gz.

    The official URL for this Web site is
    http://www.alcyone.com/pyos/church/.


Notation

    Notations for lambda expressions vary slightly, so it is
    instructive to detail the precise notation used by this module.
    A variable is expressed with a string of alphanumeric characters,
    *e.g.*, 'x' or 'abc1'.  There are three types of lambda
    expressions:

    **variable expression** -- A variable expression is simply an
      expression consisting of a single variable, *e.g.*, 'x'.

    **lambda expression** -- A lambda expression consists of a
      backslash ('\'), one or more variables separated by spaces, a
      dot ('.'), and a final expression.  For example, '\x.M' where
      *x* is a variable and *M* is an expression.  Repeated variables
      are equivalent to recursed lambda expressions, so '\x y.M' is
      equivalent to '\x.\y.M'.

    **application expression** -- An application expression consists
      of two or more adjacent expressions surrounded by parentheses:
      '(M N)', where *M* and *N* are expressions.  Applications are
      left associative, so '(M N P)' is equivalent to '((M N) P)'.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From mcherm at destiny.com  Thu Feb  7 18:52:51 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Thu, 07 Feb 2002 18:52:51 -0500
Subject: How did TimBL's speech at Python 10 go yesterday?
Message-ID: <3C631353.30507@destiny.com>

> 
> For this to work, I think you'd need to have some canonical way of
> handling versions, and enforcing compatibility.  Of course, then you're
> still left with the problems of running a program when some webserver is
> offline, or traffic is extremely heavy, or...
> 
> CPAN-like functionality of automatically making a local copy would seem
> like a much better idea to me.  (Still requires the versioning tools,
> though.)

I agree, and apparently so does TimBL. He was careful to point out that 
you don't actually go out to the web each time to get module, rather 
that the actual code is "cached" locally and there needs to be something 
  fairly smart (and definitely interactive) since the USER needs to be 
able to say what to download, how long to keep it, when (if ever) to 
update, etc.

I agree, it's a good idea, but I think the Python-version-of-CPAN 
approach is more likely to succeed than the build-it-into-import approach.

-- Michael Chermside




From bokr at oz.net  Fri Feb  1 21:28:43 2002
From: bokr at oz.net (Bengt Richter)
Date: 2 Feb 2002 02:28:43 GMT
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu> <B8809EB2.411E%jonathan@onegoodidea.com>
Message-ID: <a3fisr$9j6$0@216.39.172.122>

On Fri, 01 Feb 2002 19:36:18 +0000, Jonathan Hogg <jonathan at onegoodidea.com> wrote:

>On 1/2/2002 18:09, in article a3ell7$29bm$1 at nntp6.u.washington.edu, "Donn
>Cave" <donn at u.washington.edu> wrote:
>
>> Can I ask what operating system that experiment was performed on?
>[...]
>> That's true, but "executable" in this context usually means the binary,
>> i.e., python itself.  Years ago, some UNIX platforms introduced a hack
>> that made setuid effective for script files, but today that is universally
>> recognized to have been misguided.  That's why I'm curious about yours.
>
>Hah! How curious. Of course, you're right. I had forgotten this. Though it
>makes sense from the user's point of view - if being somewhat insecure
>(personally I prefer it in this world where the difference between compiled
>programs and scripts is blurred).
>
>To answer your question, I ran my test on Mac OS X. It seems OS X still
>retains the older semantics for script execution. Unsurprising since it is
>based on an old BSD.
>
>> Perl is subject to the same limitation, but has optional support for
>> setuid.  How can that be?  I'm no Perl whiz, but I assume you install
>> perl itself setuid root, and then it checks its input file for setuid
>> and either sets its ID accordingly or reverts to the invoker's ID.
>> This option is not the default and is rather discouraged as I recall it.
>
>Hmmmm... to return to the original question, in the absence of perl playing
>tricks with setuid, I would suggest a wrapper as being the best bet.
>
>You'll need to write a small program along the lines of:
>
>    #include <stdlib.h>
>
>    int main( int argc, char* argv[] )
>    {
>        char* args[] = { "python", "/my/script.py", NULL };
>        execv( "/usr/bin/python", args );
>    }
>
>Compile this, setuid it and then drop that into the cgi directory instead of
>the Python script. That should do the trick I think. If you can't compile
>stuff on the box or upload executables compiled elsewhere, then you're in
>trouble.
>
>Something to ask the providers is if they are willing to consider using
>Apache's suexec to invoke your CGIs. I have used that in the past to allow
>customers to run things as themselves.
>
They should really be happier doing that than allowing you to
setuid arbitrary binaries as above ;-)

    http://httpd.apache.org/docs/suexec.html

Rule 17 says maybe the above would work if you *don't* setuid it, but I don't think
you need it, unless maybe mod Python could otherwise get inbetween? I haven't set that
up myself.

I do have suEXEC enabled on a test linux box and just tried this:
--
#!/usr/bin/python
print """Content-type: text/html

<html><head><title>cgitest.py</title></head><body><pre>
"""
import os
print 'uid=%d,  gid=%d' % (os.getuid(),os.getgid())
print 'euid=%d,  egid=%d' % (os.geteuid(),os.getegid())
f=os.popen('id')
print f.readlines()
print """
</pre></body></html>
"""
--

Note that I just put it in cgi-bin and did chmod +x on it. Not setuid.
The browser shows this returned from the server for a non-suEXEC'd virtual
host:
--

uid=99,  gid=98
euid=99,  egid=98
['uid=99(nobody) gid=98(nobody) groups=98(nobody)\012']

and for a a virtual host n account with me as user:

uid=1000,  gid=100
euid=1000,  egid=100
['uid=1000(bokr) gid=100(users) groups=100(users),10(wheel)\012']


HTH,
Regards,
Bengt Richter



From tim.one at comcast.net  Thu Feb  7 22:57:04 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 07 Feb 2002 22:57:04 -0500
Subject: FixedPoint
In-Reply-To: <000501c1b053$7c8c7b00$445d4540@Dell2>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGJNKAA.tim.one@comcast.net>

[Tim]
> Magnitude (as opposed to identity -- "is") comparision of a FixedPoint
> to a non-numeric object makes no sense that I'm comfortable imposing, so
> I'm happy to let it raise an exception.

[Brian Quinlan]
> You shouldn't be. There is no more reason for FixedPoint to raise an
> exception (aside from your desire not to bother fixing it :-) than for
> other other "incompatible" types e.g.
>
> >>> 5 == 5 + 5j
> 0
> >>> 5 == '5'
> 0

Note that I said "non-numeric object".  FixedPoint supports a rich set of
auto-coercions from types that "make sense".  I'm not a fan of semantically
senseless comparisons, e.g.

>>> 5 < '5'
1
>>>

Making it easy to supply ridiculous behaviors in subclasses is what OO is
all about <wink>.




From henrik.motakef at web.de  Thu Feb 14 11:22:32 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 14 Feb 2002 17:22:32 +0100
Subject: Recommendations for cross-platform GUI Bindings ?
References: <xBMa8.217$B92.52511@news.xtra.co.nz> <ictn6uomd63papjtut186c67lickq0cjj9@4ax.com>
Message-ID: <864rkk119z.fsf@apo.hmotakef.int>

Courageous <jkraska at san.rr.com> writes:

> For windowing, Tk, WxWindows, and QT are the king of the crop,
> although I'm not sure how well the QT bindings are being maintained.

PyQt is actively developed (e.g. a Qt3 version is available), and has
a active and helpful mailing list. So that's not what should make you
not using it.

However, I doubt it runs on the Mac. But for Unix/Windows, it's definitely
my Toolkit of choice.



From mcfletch at geocities.com  Sat Feb  9 06:56:49 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Sat, 09 Feb 2002 06:56:49 -0500
Subject: The joys of weakrefs... references to instance methods (callbacks)
Message-ID: <3C650E81.1050804@geocities.com>

Sat down to work out a model-view framework for a little application and 
had one of those "oh, darn, didn't think that out very carefully" 
moments.  You see, I was planning on doing a fairly straightforward 
callback-passing mechanism:

     WatchAble.addWatcher( callback )

With callback getting a nicely formatted call, giving both the object 
and an event object from which it could extract all kind of nice "what 
type of event" occured information.

Now, of course, being a good little boy (evidence to the contrary), I 
was going to use weakref.ref objects to hold the callbacks, so that when 
the view objects are destroyed, their callbacks will no longer get 
called (since that causes Nasties(TM)).

It's a strange thing, but the most convenient callback is almost allways 
a bound method of a view object:

     myModelObject.addWatcher( myViewObject.OnModelUpdated )

myViewObject.OnModelUpdated is a _new_ Python object (a bound method), 
with only the reference in the calling stack keeping it alive.  The 
weakref.ref( Which is, of course, where everything goes straight south 
in a handbasket.  myViewObject.OnModelUpdated is actually a new Python 
objectmyViewObject.OnModelUpdated ) call (in addWatcher) will generate 
the reference object.  That reference immediately goes to "dead" status 
as the line of code after addWatcher is executed.

Well, there's an obvious solution, I can create an explicit callback 
object, store it instead of the weakref, and let it do the 
de-referencing to get the method (as long as the instance and function 
are both still available). Basically, I need to wrap weakref.ref like so:

import weakref, types, new

class MethodRef:
     """Weak-reference to a method that takes the component parts of the 
object, rather than the object itself for the "live" definition"""
     def __init__( self, object ):
         self.instance = weakref.ref( object.im_self )
         self.function = weakref.ref( object.im_func )
     def __call__( self ):
         instance = self.instance()
         function = self.function()
         if instance and function:
             return new.instancemethod( function, instance, 
instance.__class__ )
         else:
             return None

def ref( object ):
     if type(object) == types.MethodType:
         return MethodRef( object )
     else:
         return weakref.ref( object )

Yes, it's slightly different semantics than weakref, but it's actually 
useful in the real world ;) .  Anyway, thought I'd share the code so 
that people can object that it's a tool of the devil :) .  If the powers 
that be want to point out any reasons why I shouldn't do this, would 
love to hear it before my house of cards goes *phoosh*.

Enjoy all,
Mike
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From brueckd at tbye.com  Fri Feb  1 15:51:42 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Fri, 1 Feb 2002 12:51:42 -0800 (PST)
Subject: Java vs Python Benchmarks: Java is faster
In-Reply-To: <a3er63$7e3$1@news.island.liu.se>
Message-ID: <Pine.LNX.4.33.0202011248300.21332-100000@ns0.tbye.com>

On Fri, 1 Feb 2002, Anders Dahlberg wrote:

>
> <brueckd at tbye.com> skrev i meddelandet
> news:mailman.1012587303.30347.python-list at python.org...
> > On Fri, 1 Feb 2002, Anders Dahlberg wrote:
> >
> > > could re-run them without it. Maybe the low number of replies > is
> > > because it's not suprising that a semi-compiled-to-native program is >
> > > faster?
> > >
> > > well, but why would you want to compare interpreted java with python?
> >
> > Well for one, it'd be nice to be a little closer to comparing apples to
> > apples (how fast a VM can interpret and execute the bytecodes).
>
> Somewhat valid point (IMO that is)
>
> >Secondly,
> > I was very suprised that, with a JIT, Java was only meagerly faster than
> > Python.
>
> Well, python is written in c or something equivalent right?

Yes, the Python VM is written and C. And what do you think the Java VM
is written in?

> > was curious to know how slow Java was without the JIT. Finally, even
> > though Java and Python are wildly different in many ways, knowing the
> > interpreted vs. JIT'ed speeds might give a little insight into the
> > potential benefits of a Python JIT.
>
> I don't really know if python needs a jit, it would probably not harm
> python, but is it really worth the effort?

Heehee... that's the whole reason I asked my question in the first place.

-Dave




From fperez528 at yahoo.com  Sat Feb  9 12:41:39 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sat, 09 Feb 2002 17:41:39 +0000
Subject: A newbie question about class
References: <a50631f6.0202091635.1dc42d4b@posting.google.com>
Message-ID: <a44fne$33o$1@peabody.colorado.edu>

newgene wrote:

> Hi, group,
>      I have a newbie question about building a class:
> 
> eg.
> 
> class A:
>    def __init__(self,x=None):
>        self.x=x
>        if self.x != None:
>             self.y=x**2
>        else:
>             self.y=x**2

This code has a problem: None**2 gives a TypeError exception.

f


From mwh at python.net  Fri Feb 22 05:58:43 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 22 Feb 2002 10:58:43 GMT
Subject: Python 2.1.2 regression crash on Linux/Sparc
References: <m3it8ql5zr.fsf@abbadon.mcnaught.org>
Message-ID: <uwux5kcfk.fsf@python.net>

Doug McNaught <doug at mcnaught.org> writes:

> I have an UltraSparc running Linux (Red Hat 6.2 + erratea updates).  I
> would love to get Zope running on it.  
> 
> I got Python 2.1.2, and did 'configure --with-threads'.  The make goes
> fine.  When I run the regression test, I get:

Trying compiling mathmodule.c without optimization?

Cheers,
M.

-- 
  I think perhaps we should have electoral collages and construct
  our representatives entirely of little bits of cloth and papier 
  mache.          -- Owen Dunn, ucam.chat, from his review of the year


From kirill_simonov at mail.ru  Sun Feb 24 20:56:34 2002
From: kirill_simonov at mail.ru (Kirill Simonov)
Date: Mon, 25 Feb 2002 03:56:34 +0200
Subject: Tkinter + Entry + cp1251 cyrillic
In-Reply-To: <m3adtz9t8y.fsf@mira.informatik.hu-berlin.de>; from martin@v.loewis.de on Sun, Feb 24, 2002 at 09:30:37AM +0100
References: <26428104.0202231657.639cec52@posting.google.com> <mailman.1014533499.3218.python-list@python.org> <m3adtz9t8y.fsf@mira.informatik.hu-berlin.de>
Message-ID: <20020225035634.B712@abulafia.local>

On Sun, Feb 24, 2002 at 09:30:37AM +0100, Martin v. Loewis wrote:
> Kirill Simonov <kirill_simonov at mail.ru> writes:
> 
> > Try to replace "Python22\Lib\lib-tk\FixTk.py" with the following:
> > 
> > import sys, os
> > 
> > if not os.environ.has_key('TCL_LIBRARY'):
> >     tcl_library = os.path.join(sys.prefix, "tcl", "tclX.Y")
> >     os.environ['TCL_LIBRARY'] = tcl_library
> 
> Can you please explain what this change achieves, compared to the
> FixTk.py that is already there?

_tkinter.pyd is not imported here. Consequently tcl83.dll is not loaded
also.

> In particular, do you mean that "X.Y" should be used literally?

Yes. Tcl also checks the path "$TCL_LIBRARY/../tcl<VERSION>/".

Oh, I just noticed that you had already fixed this in cvs. Many thanks!


Kirill.




From bokr at oz.net  Tue Feb 19 22:42:08 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 03:42:08 GMT
Subject: execfile question
References: <3C724B05.E9167812@attbi.com> <mailman.1014125130.25988.python-list@python.org>
Message-ID: <a4v5ug$i68$0@216.39.172.122>

On Tue, 19 Feb 2002 16:23:19 +0300, Kerim Borchaev <warkid at storm.ru> wrote:

>Mark Barclay answered:
>
>> "del c" will cause your __del__() to be called.
>yes
>
>> So will any other action that causes your instance to be garbage-collected.
>I don't get it.
>And can you explain who holds a reference on C instance in my example
>when 'execfile' completes?
>
I suspect the anonymous directory you passed to execfile. Apparently
dictionaries don't always get cleared automatically. I don't know
if this is a bug. But if you do:

xdict = {}
execfile('executable.py', xdict)
xdict.clear()    # just del xdict doesn't do it

I think your example will give you what you expected.

Using default dicts like this:

execfile('executable.py')

works too.

Regards,
Bengt Richter


>Kerim.
>
>----------------------------------
>My original question:
>
>> Can someone explain why while running this script I don't see C.__del__
>> eventually called?
>>
>> For me the output of it's execution is:
>>
>> >main.py
>> it's me
>> init
>> executed
>> >
>>
>> ##main.py######################
>> execfile('executable.py', {})
>> print 'executed'
>> ########################
>>
>> ##executable.py######################
>> print "it's me"
>> class C:
>>     def __init__(self):
>>         print 'init'
>>     def __del__(self):
>>         print 'deleted'
>>
>> c = C()
>> ########################
>>
>> Thanks in advance.
>
>
>



From brian at sweetapp.com  Mon Feb 18 20:04:17 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 18 Feb 2002 17:04:17 -0800
Subject: Adding static typing to Python
In-Reply-To: <a4s51k$8pd1@news1.gtech.com>
Message-ID: <000801c1b8e1$5b603a70$445d4540@Dell2>

Peter Milliken wrote:
> When I compile a Ada program cleanly I know that 95% of the bugs are
out
> of it - before I even start testing! :-) Of course, it takes me longer

> to write up front but overall development effort (in my subjective
opinion > - since I have no metrics :-)) is that it is much lower.

If 95% of your bugs are of such a trivial nature, then you are either
very sloppy in your code production, you are an absolutely brilliant
designer or you are working on simple problems.

I would say that 99% of my bugs are due to design mistakes (e.g.
inappropriate algorithm selection, poorly designed interface) or
semantically incorrect algorithms (e.g. boundary errors, incorrectly
identified complexity/memory usage).

And finding these errors is substantially more difficult than finding
misnamed variables.

Cheers,
Brian




From dvass at felis.uni-freiburg.de  Wed Feb 20 09:57:23 2002
From: dvass at felis.uni-freiburg.de (=?iso-8859-1?Q?J=F6rg=20W=F6lke?=)
Date: Wed, 20 Feb 2002 15:57:23 +0100
Subject: why does popen2 silently ignore IOError?
References: <mailman.1014167611.14176.python-list@python.org>
Message-ID: <3C73B953.864E9409@felis.uni-freiburg.de>


Graham Guttocks schrieb:
> 
> Greetings,

Hello!

[ snip ]

> As you can see, no exceptions were raised despite the fact that there
> is no such file as '/bin/nonexistent'.  Why is this?

Because popen(3) returns a file descriptor or a Null-pointer?
Try testing for None.
 
> Regards,
> Graham

HTH,HAND
and Greetings J"o!

-- 
.sig not found


From db3l at fitlinxx.com  Tue Feb 26 17:07:24 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 26 Feb 2002 17:07:24 -0500
Subject: DLL/Embeded interpreter wierdness
References: <qltl7uoh7q01drhb37b48bsmipr7h8kaam@4ax.com> <uy9hgvpi2.fsf@ctwd0143.fitlinxx.com> <164m7ug4ramg6hjfiac8mb4inajc59mr9h@4ax.com>
Message-ID: <ur8n7aodf.fsf@ctwd0143.fitlinxx.com>

Courageous <jkraska at san.rr.com> writes:

> I don't really know what you mean here. This module in question is
> actually masquerading as a package, so extended import commands look
> more like...
> 
> from lramm import task

Well, that doesn't really need a package to function, just a working
module.  Since this is a DLL we're talking about, I'm assuming you
just call one of the Py_InitModule calls in your initxxx routine,
right?  Or are you doing something more when you say "masquerading as
a package"?

> ...and so on. So the explicit initialization is required from the embedded
> code, which is otherwise unrestricted Python.

That's fine - in my suggestion, under the embedded environment,
"lramm" will already exist as a module, so this from will simply
reference that module and copy references to the specified objects
into the local namespace.

> I seem to have some sort of knowledge failure here. Outside of the
> usual init[blah] hookups, what does one do to "automatically define
> my module internally"?

Basically just run the same sort of code you currently have in the
init[blah] function, but without waiting for Python to call your
init[blah] function.  You do it before transferring control to the
embedded interpreter to execute the embedded script.

For example, in an embedded app of mine, here's where I initialize
Python:

    #define INSITEM(name,value) \
       intobj = PyInt_FromLong((value) & 0xFF); \
       PyDict_SetItemString(dictobj,name,intobj); \
       Py_XDECREF(intobj)

    (...)

    PyObject *modobj, *dictobj, *intobj;

    Py_Initialize();
    PyEval_InitThreads();
    ecnaPy_RcvThreadState = PyThreadState_New(PyThreadState_Get()->interp);

    PyImport_AddModule("ecna");
    modobj = Py_InitModule("ecna",ecnaPy_Methods);

    /* Add in definitions for the protocol defines into dictionary */
    dictobj = PyModule_GetDict(modobj);
    INSITEM("dSrvHReset",dSrvHReset);
    INSITEM("dSrvInit",dSrvInit);
    INSITEM("dSrvUSetup",dSrvUSetup);
    INSITEM("dSrvUReset",dSrvUReset);

    (...)

    /* Automatically import our own module */
    PyRun_SimpleString("import ecna");

    (...)

    /* Execute script */
    PyRun_SimpleFile(OptScriptFile,OptScriptName);


(I use the embedded interpreter from multiple threads thus the
acquisition of an extra thread state).

This defines "ecna" as a built-in module to the embedded script (and
also automatically imports it on behalf of the executing script).

The script could just as easily have used either:

    import enca

or

    from ecna import dSrvHReset # For example

And the only real difference between this and what I would have used
if this were strictly an extension module is that as an extension
module I wouldn't need to initialize the interpreter.  So its not too
hard to build a single initialization function that is called
post-interpreter initialization in the embedded case, or just directly
from your init[blah] function in the extension case.

Hoping this makes things clearer rather than more muddy...

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


From barry at zope.com  Mon Feb  4 16:47:13 2002
From: barry at zope.com (Barry A. Warsaw)
Date: Mon, 4 Feb 2002 16:47:13 -0500
Subject: FW: Your message to Python-list awaits moderator approval
References: <60FB8BB7F0EFC7409B75EEEC13E20192158CF7@admin56.narex.com>
Message-ID: <15455.353.511698.96738@anthem.wooz.org>

>>>>> "BP" == Bjorn Pettersen <BPettersen at NAREX.com> writes:

    BP> Is this because I'm using Outlook XP?

No, it's probably because you posted from an open relay, so RBL added
a warning header to your message.  Most Mailman lists hold messages
with the warning header for moderator approval.  Yours will likely get
approved when the python-list moderator gets around to it...

-Barry



From andymac at bullseye.apana.org.au  Tue Feb 12 06:17:00 2002
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Tue, 12 Feb 2002 22:17:00 +1100 (EDT)
Subject: Coming of age...
In-Reply-To: <a49sdl$5ao$2@newshost.accu.uu.nl>
Message-ID: <Pine.OS2.4.32.0202122203430.2711-100000@tenring.andymac.org>

On 12 Feb 2002, Martijn Faassen wrote:

> Another interesting thing the other day is that on comp.object someone
> was asking which language they should try to experiment with OO, and
> there came lots of suggestions to use Ruby (Python was in there, but Ruby was
> mentioned a *lot*). Seems that language has been crossing a barrier
> as well recently.

My take: Ruby turns out to be a language of refuge for escapees from Perl
who can't cope with white-space significance (more of them than not, I
expect).

Perl of course has a much larger userbase than Python, let alone Ruby, so
even a small leakage from the Perl community would be a flood to lesser
userbases.

Disclaimer: I've never used either Perl or Ruby; this is my interpretation
of a multitude of comments I've read.

FWIW...

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andymac at bullseye.apana.org.au  | Snail: PO Box 370
        andymac at pcug.org.au            |        Belconnen  ACT  2616
Web:    http://www.andymac.org/        |        Australia




From stojek at part-gmbh.de  Fri Feb 15 09:41:47 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Fri, 15 Feb 2002 14:41:47 GMT
Subject: help!: optimization of search problem (example code)
References: <3c6bcf70.16509609@news.easynews.net> <JGWa8.159434$XZ1.6306345@typhoon.austin.rr.com>
Message-ID: <3c6d134c.4494312@news.easynews.net>

>SciPy has a vector quantization module that does what your looking for and
>it is pretty fast.
>
>C:\temp>python vq_ex.py
>500 points ran in 0.037537 seconds
>
You are comparing two lists with 500 points each. In my example
I perform the search for 500 points out of la but I have to search
for the counterpart in lb which has 20000 points.

Nevertheless i tried scipy. Here are some times:
(same size for both lists)

500 points : 0.33557 sec 
1000 points: 1.321 sec
2000 points: 6.055 sec
20000 points: 967.2 sec

My example-code  needs 197.8 sec ;-)

Anyway, scipy seems to be extremly useful.
Thanks
Marcus


From alien275 at libertysurf.fr  Thu Feb 28 13:51:06 2002
From: alien275 at libertysurf.fr (Riss Nicolas)
Date: Thu, 28 Feb 2002 19:51:06 +0100
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de> <a5j53i$oms$1@wanadoo.fr> <8z9f2rks.fsf@fido.trinitycapital.com> <a5j7fb$fb$1@wanadoo.fr> <a5k352$etg$0@216.39.172.122>
Message-ID: <a5lu00$29p$1@wanadoo.fr>

Has somebody a python program where it is an telnet connection ?
"Bengt Richter" <bokr at oz.net> a ?crit dans le message de news:
a5k352$etg$0 at 216.39.172.122...
> On Wed, 27 Feb 2002 19:14:22 +0100, "Riss Nicolas"
<alien275 at libertysurf.fr> wrote:
>
> >No, the port is correct. The telnet server is cingured to listen of the
port
> >32.
> >Another suggestions ?
> >
> >Thanks
> >"Mike Johnson" <afp_randjohnson at yahoo.com> a ?crit dans le message de
news:
> >8z9f2rks.fsf at fido.trinitycapital.com...
> >> "Riss Nicolas" <alien275 at libertysurf.fr> writes:
> >>
> >> > I jave used it but when y tried to connect it makes an error:
> >> >    File "c:\test.py", line 9, in ?
> >> >     tn = telnetlib.Telnet("192.168.0.9",32)
> >> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 190, in __init__
> >> >     self.open(host, port)
> >> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 219, in open
> >> >     raise socket.error, msg
> >> > socket.error: (10061, 'Connection refused')
> >> >
> >> > Programm:
> >> >
> >> > import getpass
> >> > import sys
> >> > import telnetlib
> >> > password = getpass.getpass()
> >> > tn = telnetlib.Telnet("192.168.0.9",32)
> >> > tn.read_until("Administrator Password : ")
>                     ^^^^^^^^^^^^^^-?      ^-^-??
>                     \_ worth double checking _/
> >> > tn.write(password + "\n")
>                           ^^--need \r\n for default telnet protocol, IIRC
> >> > tn.write("session\n")
>                       ^^--need \r\n ?
> >> >
> >> >
> >> > Where is the mistake ?
> >>
> >> Wrong port. Try:
> >>
> >> tn = telnetlib.Telnet("192.168.0.9", 23)
> >>
>
> I'd log in manually with any normal telnet client
> to double check the prompts.
>
> HTH,
> Regards,
> Bengt Richter
>
>




From jkraska at san.rr.com  Sun Feb  3 21:12:55 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 04 Feb 2002 02:12:55 GMT
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <mailman.1012787645.17462.python-list@python.org>
Message-ID: <csrr5uc3n84jd20mhsen6rpqh328ljqlrs@4ax.com>

>Briefly, you don't.  Or you do.

All good stuff. But to answer directly, the best way to "pass
something by reference" is to have it be an attribute on an object,
and have what its being passed to a method. Those who adhere to
good OO design principles will quickly forget that Python doesn't
really have pass by reference. Or it doesn't. Or it does. :)

C//



From martin at v.loewis.de  Sun Feb 24 03:28:16 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 09:28:16 +0100
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com>
Message-ID: <m3eljb9tcv.fsf@mira.informatik.hu-berlin.de>

redskin at tut.by (Andrey Sidorenko) writes:

> I use Active Python 2.2 on Windows 2000 and have trouble - "Entry"
> class can't display cyrillic letters (cp1251 encoding) correctly in
> entry field. Can anybody help me with this weird problem?

Can you be more specific? I have no problems getting Tkinter to
display Cyrillic characters, using Unicode strings.

Regards,
Martin



From rhymes at myself.com  Thu Feb 14 11:47:56 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Thu, 14 Feb 2002 17:47:56 +0100
Subject: Color database in Tkinter
Message-ID: <veqn6ucitupfrt3jaujtd7rhrq0sl39r2g@4ax.com>

Where i could find the whole database for color names
built in Tkinter? I need a list of colors

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From jeff at ccvcorp.com  Thu Feb  7 15:09:39 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 07 Feb 2002 12:09:39 -0800
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>
Message-ID: <3C62DF02.966C895A@ccvcorp.com>

Peter Hansen wrote:

> ...  If even Jeff is incapable of understanding
> you, and believe me when we had The Vote we concluded he was the *only*
> one with a hope of understanding even a glimmer of what you say, ....

Why does the phrase "damning with faint praise" come to my mind here?  ;)

I suppose I shouldn't even bother trying to point out to him that "simple two-dimensional
bit flipping" ... is really only one-dimensional.... (or that modern physics suggests
that "real physical objects" have more like six (or was it ten?) dimensions...)

Jeff Shannon
Technician/Programmer
Credit International




From gbreed at cix.compulink.co.uk  Thu Feb 14 06:32:11 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Thu, 14 Feb 2002 11:32:11 +0000 (UTC)
Subject: Help! IIS and Python
References: <eqlm6ugbbtsu4i1d61pek0c4g49as7j119@4ax.com>
Message-ID: <a4g77r$ngb$1@thorium.cix.co.uk>

Tim Roberts wrote:

> Since we're on the subject of Python in ASP programming, how do you use 
> the
> <%=xxx%> trick with indentation?  This fails with an indentation error, 
> > for
> example:
> 
> <%
> for i in range(5):
> %>
>   <%="aaa%d" % i %>

<%
formatter = """
  aaa%d
"""

for i in range(5):
  Response.Write(formatter%i)
%>

Why not?


               Graham


From bokr at oz.net  Sun Feb 17 14:34:13 2002
From: bokr at oz.net (Bengt Richter)
Date: 17 Feb 2002 19:34:13 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <mailman.1013934149.9720.python-list@python.org>
Message-ID: <a4p0jl$q23$0@216.39.172.122>

On Sun, 17 Feb 2002 03:21:55 -0500, "Tim Peters" <tim.one at home.com> wrote:
[...]
>
>What programmers expect is:
>
>123456789012345678901
 012345678901234567890 is what I expected ;-)
>    T   T   T   T   T
>a   T   T   T   T   T
>ab  T   T   T   T   T
>abc T   T   T   T   T
>abcd    T   T   T   T   T
>abcde   T   T   T   T   T
>abcdef  T   T   T   T   T
>abcdefg T   T   T   T   T
>

Regards,
Bengt Richter



From matthias_j at lycos.com  Sat Feb  9 14:52:04 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 9 Feb 2002 11:52:04 -0800
Subject: Help with Gordon McMillan Installer
Message-ID: <d7d5ebdf.0202091152.40a6e0af@posting.google.com>

Maybe there is anyone out there to help me in a small problem with the
Installer.

I would like to create a onefile exe but I can't figure out how to
write the right Makespec.file.

I have a single python file (call it /MyApp.py) which also accesses a
data file in d subfolder Data (call it /Data/Data.cfg)

When ever I make a Spec file :python ../Installer/Makespec.py
--onefile MyApp.py

It creats this Spec File:

a = Analysis(['C:\\_Matthias\\_APPS\\Installer\\support\\useUnicode.py',
'C:\\_Matthias\\_APPS\\TEST3\\MyApp.py'],
             pathex=[])
pyz = PYZ(a.pure)
exe = EXE( pyz,
          a.scripts,
          a.binaries,
          name='MyApp.exe',
          debug=0,
          console=1)


I tried different ways to add the Data.cfg file but this is never
included in the ONEFILE.EXE instead it is copied extra in a folder.

One help would be appriciated.

Thanks
Matthias


From c at tech.usd253.org  Mon Feb 11 12:01:48 2002
From: c at tech.usd253.org (Fred)
Date: 11 Feb 2002 09:01:48 -0800
Subject: Question on if, while, for....  Figured it out..
Message-ID: <e959e7b.0202110901.485da0b0@posting.google.com>

OK, I figured it out (one other way at least).

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

num = input("Enter the numerator: ")
den = input("Enter the denominator: ")
test = 0

if num > den:		#Test for improper fraction.
    i = num/den		#This will only return the whole part. 
    j = num%den		#This will return only the remainder.

    if j != 0:
        for i in range(9,0,-1):
            if num%i == 0 and den%i ==0 and test != 1:
                a = j/i
                b = den/i
                print "You entered an improper fraction, simplified it
is: %d %d/%d" % (i,a,b),
                test = 1
    else:
        print "You entered an improper fraction, simplified it is: %d"
% (i)
	
        

else:
    for i in range(9,0,-1):
        if num%i == 0 and den%i ==0 and test != 1:
            a = num/i
            b = den/i
            print "The fraction you entered is %d/%d in lowest terms."
% (a,b),
            test = 1


From dsh8290 at rit.edu  Thu Feb  7 18:11:55 2002
From: dsh8290 at rit.edu (dman)
Date: Thu, 7 Feb 2002 18:11:55 -0500
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
In-Reply-To: <3C630252.6605FA49@geneva-link.ch>
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <e840346c.0202070321.452d405d@posting.google.com> <7xr8nxy0rb.fsf@ruckus.brouhaha.com> <e840346c.0202070724.13440c8d@posting.google.com> <E+9AIXAiRqY8EwKc@jessikat.demon.co.uk> <3C630252.6605FA49@geneva-link.ch>
Message-ID: <20020207231155.GA31848@dman.ddts.net>

On Thu, Feb 07, 2002 at 11:40:18PM +0100, Boris^2 wrote:
| Robin Becker wrote:
| > 
| > I think there are good compilers for lisp which can be a very dynamic
| > language so in principle it's not the dynamic nature of python that
| > prevents native compilation.
| 
| Wouldn't a first step be optional type declarations ?

That was already answered in this thread, and it is "no".  Given the
quoted text, the answer would be "no" -- lisp doesn't have type
declarations, yet it can be fast.  The trick is
    1)  figuring out exactly _what_ causes python to be "slow"
    2)  spending the time to work out an implementation that isn't
        (as) slow, without changing the language at all

The second point is currently the major stumbling block.

-D

-- 

A man of many companions may come to ruin,
but there is a friend that sticks closer than a brother.
        Proverbs 18:24




From theivo at abv.bg  Mon Feb 18 06:23:25 2002
From: theivo at abv.bg (I.C.)
Date: 18 Feb 2002 03:23:25 -0800
Subject: How can I make a screenshot using BitBlt or something else???
References: <6ac1ecad.0202151338.15d3c4b@posting.google.com> <3C6D93F0.8F83FA9E@olen.to>
Message-ID: <6ac1ecad.0202180323.e24dcfe@posting.google.com>

Joonas Paalasmaa <joonas at olen.to> wrote in message news:<3C6D93F0.8F83FA9E at olen.to>...
> "I.C." wrote:
> > 
> > I am new in PythonWin programming so I have a simple question -
> > How can I make a screenshot of the whole screen and then get the colors
> > of some pixels?
> 
> Use Fredrik Lundh's grabscreen utility for PIL.
> http://effbot.org/pil/index.htm

Thank you very much, this seems to work, but do you know how can I
make the
screenshot using win32ui - on the system, where I want to run the
script I won't be able to install the PIL-Module(SysAdmin-Rules).

Thanks


From DavidA at ActiveState.com  Wed Feb 13 12:45:55 2002
From: DavidA at ActiveState.com (David Ascher)
Date: Wed, 13 Feb 2002 09:45:55 -0800
Subject: Compiling Python 2.2 in Microsoft VC7 (.net)
References: <LNBBLJKPBEHFEDALKOLCAEHINLAA.tim.one@comcast.net>
Message-ID: <3C6AA653.6CB4F714@activestate.com>

Tim Peters wrote:
> 
> [Joe Shochet]
> > I'm trying to get Python 2.2 to compile in VC7 (part of Visual Studio
> > .net), but I'm running into all sorts of troubles. Before digging into
> > the problems, has anybody else already done this?
> 
> David Ascher, at ActiveState, has, but it was "on his own time" so I don't
> know whether he'll be able to make time to talk about it.  I don't know of
> anyone else.

Someone more competent (Mark Hammond) fixed it in the tree:

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/python/dist/src/Modules/_hotshot.c?r1=1.14

An early version of the patch (may be the same thing he checked in) is
available at:

http://mail.python.org/pipermail/python-list/2002-February/086932.html

--david



From sjmachin at lexicon.net  Tue Feb 19 21:52:21 2002
From: sjmachin at lexicon.net (John Machin)
Date: 19 Feb 2002 18:52:21 -0800
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com> <m37kpcuu31.fsf@mira.informatik.hu-berlin.de> <7x4rkgw79y.fsf@ruckus.brouhaha.com> <c76ff6fc.0202171321.4bd2e69c@posting.google.com> <3C72C074.306@wi.mit.edu>
Message-ID: <c76ff6fc.0202191852.7df1f4d9@posting.google.com>

Brian Kelley <bkelley at wi.mit.edu> wrote in message news:<3C72C074.306 at wi.mit.edu>...
> John Machin wrote:
> > You'll find one nuisance when working with 'c' arrays; it doesn't
> > support coercing an ordinary string to an array -- you have to do it
> > yourself.
> 
> I encountered the same problem and of course the solution was subclass 
> array.array!  Unfortunately, array.array can't be subclassed.  (Python 
> 2.2)  In fact, I found arraymodule.c very confusing in this respect.
>
[snip] 
> 
> What would it take to make array.array or array.ArrayType an acceptable 
> base type for subclassing?

Not all built-in types can be subclassed. The brochure promises
"subtyping of *selected* built-in types" [my emphasis]. The array type
is probably not classified as built-in anyway.

Quite simple, the array type is not set up to be subclassed.

If one was desperate to subclass the array type, one would need to
read the brochure (PEP 253 (Subtyping Built-in Types)) especially the
section headed "Preparing a type for subtyping", and do what it says
-- I would suggest heavy plagiarism from listobject.c.

One might like to, while one has the lid off arraymodule.c, fix it so
that the original problem is solved at source without the need to
subclass -- for several methods that involve an "other" argument,
simply allow "other" to be an itera{ble|tor} [or maybe just a sequence
if one was worried about speed] that delivers objects that the array
object can contain.

Noting that array.array is short of a few methods (like sort), it
might be worth the effort to refactor array and list so that
list(["x", 42, 0.05]) was just a special case of array i.e.
array.array('O', ["x", 42, 0.05]) and as far as possible all the list
methods and all the care and attention that has gone into list
(sorting, intelligent growth when appending, etc) could apply to
arrays as well. Any volunteers?


From ahmebah at hotmail.com  Sat Feb  9 03:39:09 2002
From: ahmebah at hotmail.com (ameoba)
Date: 9 Feb 2002 08:39:09 GMT
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com>
Message-ID: <Xns91B074EB2FA5ahmebahhotmailcom@198.99.146.10>

joost_jacob at hotmail.com (J.Jacob) wrote in 
news:13285ea2.0202080524.e2e7948 at posting.google.com:


> Windows? Mac?  The pdcurses and WConio modules i found for Windows

Actually, there's a handful of win32 calls that would handle this 
relatively simply; less than a page worth of code (well, until you 
range/error checking and a Python interface).  But since it would require a 
system specific binary, it'd be difficult to distribute.  Perhaps the best 
thing to do would be to talk The Powers That Be into including it in the 
base distribution...


From huaiyu at gauss.almadan.ibm.com  Fri Feb 15 14:59:11 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Fri, 15 Feb 2002 19:59:11 +0000 (UTC)
Subject: import __past__ as a fade-out mechanism?
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <slrna6o1i2.u4q.huaiyu@gauss.almadan.ibm.com> <3C6C6CCC.C5049297@cosc.canterbury.ac.nz>
Message-ID: <slrna6qq4e.15d.huaiyu@gauss.almadan.ibm.com>

On Fri, 15 Feb 2002 15:05:00 +1300, Greg Ewing <greg at cosc.canterbury.ac.nz>
wrote: 
>Huaiyu Zhu wrote:
>> 
>>    from __future__ import zerolead_octal so that 041 will mean 33.
>
>Shouldn't that be from __past__ import zerolead_octal? :-)

Yes, but the timbot had ruled out __past__ in many other threads.  So I just
had to play with __future__.  :-)

Thinking about this a bit more, I think __past__ can be a good mechanism for
fade-out transition just as __future__ is a good mechanism for fade-in
transition, as long as it is made clear that anything in the __past__ is
doomed to disappear sometime in the future.

Here's an example.  Suppose a feature is being changed from A to B in
version 5.  Early birds can import __future__ to use B in version 3, while
old holdouts can import __past__ to hold on to A until version 6:

version     1   2   3   4   5   6   7   8
__future__  .   .   B   B   B   B   B   B
plain       A   A   A  (A)  B   B   B   B
__past__    A   A   A   A   A  (A) (.) (.)

[Parenthesis denotes warning.  Dot denotes unavailable.]

How about that?

Huaiyu


From lac at strakt.com  Mon Feb 11 04:40:28 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 11 Feb 2002 10:40:28 +0100
Subject: Defending the Python lanuage... 
In-Reply-To: Message from Cliff Wells <logiplexsoftware@earthlink.net> 
   of "Fri, 08 Feb 2002 10:37:14 PST." <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> 
References: <3C62EE4A.5090305@destiny.com> <20020207140755.364d5d77.logiplexsoftware@earthlink.net> <200202081311.g18DBQGo020260@ratthing-b246.strakt.com>  <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> 
Message-ID: <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com>

> Cliff Wells:
>
> Laura Creighton wrote:
> 
> > > If if.  No doubt things such as personality are going to be a huge
> factor. 
> > > I guess that sums up my assertion: the type of personality attracted to
> > > programming is not the type to endure watching someone else code.  My
> > > fingers would twitch themselves numb.  To have pair-programming work
> you
> > > would need two programmers who: 
> > > 
> > > 1.  have similar skill levels
> > 
> > This is incorrect.  
> 
> How dare you.   ;)

Grin.  'Cause I know better! <sticks out tongue>

> 
> > It works great with people with radically different
> > skill levels.  Of course, they start to converge, so they do not end
> > up with different skill levels.  thats the point.  For example, people
> > who pair program with me and are not aware of design patterns and
> signature 
> > based polymorphism will get that fixed in a hurry.  But what happens
> > when Alex Martelli and I are pair programming is I learn so much stuff
> > I never even heard about before.
> 
> No doubt.  This is in fact how I have trained people in the past - they sit
> and watch me program with some toolkit and we discuss what I'm doing (I had
> great success teaching someone with no programming experience whatsoever
> how to integrate Python, wxPython and MySQL - in less than 3 months he was
> writing on his own [and damn arrogant too, I might add]).  However, I think
> a key point in his progress a a programmer was when he finally sat at his
> own computer and worked independently of me.  I'm reminded of a Chinese
> proverb I read in a calculus book:
> 
> I hear and I forget
> I see and I remember
> I do and I understand

You are not getting an argument with me about that one.  Here is a
varation.  Make your first prototype alone.  Then pair.  That's how
to get 'make one to throw away' embedded in the creation process.

> Now you know Laura, that your posts are so rare and insightful that I am
> loathe to disagree with you, -but- ;) my point is that while working
> together is extremely beneficial for sharing ideas and information, I don't
> see sharing the same console with another programmer for 90% of the time to
> be as beneficial to productivity.  Sitting side-by-side at separate
> consoles, yes, but not at the same console (except when there are
> interesting discoveries/techniques to be shared or design questions to be
> pondered). 

This is one of the principals of XP which we are stress-testing.  I'll
let you know when we have done this enough to have any firm conclusions.
I came to XP with a background that said ideal is 'everybody has their
own office and we *also* have a shared terminal room.  YOu get/give help
with anybody who is in the terminal room.  We haven't tried that here,
but its only recently that we are enough people that it is a viable idea.
(Of course, we don't have enough machines to do it now....)

> However, this discussion does give me the idea that pair-programming could
> be very beneficial if programmers were rotated on some schedule, so that
> one day they watched someone else and on another someone else watched them,
> especially given a group of several programmers that could be rotated
> through.  This way, programmers would get enough time doing development on
> their own but still have a chance to share ideas with other programmers in
> their group.  Too bad I'm not in a position to give it a try :P

Don't formalise the process too much.  You are not trying to bring order
to your establishment by building rules that work top-down.  You are
trying to foster order that works from the bottom up.  How any two
people best collaborate can vary with the people -- and if you are
on a roll you may want, need, and deserve a few days with each other
and only with each other -- otherwise 10 minutes, or an hour to
kill this one nasty bug may be enough.

The person who is doing the typing is in more control, but the person
who isn't is in a better position to exercise good judgement.  This
usually works better when it swaps in units a lot smaller than 'a
day'.  

> 
> > It is such a joy.
> 
> I'm glad to hear it =)
> 
> > > 2.  like and respect each other
> > > 3.  have compatible programming philosophies
> > 
> > (you are spot on with those, however.)
> > 
> > Cliff, again you are up against 'American management is impossible.'
> > Seriously consider visiting Sweden, for a vacation.  I think you would 
> > like it here.
> 
> You'd better be careful what you ask for - you're starting to tempt me -
> you could end up with some guy with a strange accent and stranger ideas
> hanging around ;)  I expect I could learn quite a bit from you.

And I from you, and I from you.

Have you read 'Engineering and The Minds Eye'?  I just finished it.
Wonderful.  

> Cliff Wells, Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308  (800) 735-0555 x308

Laura Creighton



From python at chajadan.net  Thu Feb 21 22:19:50 2002
From: python at chajadan.net (chajadan)
Date: Thu, 21 Feb 2002 19:19:50 -0800
Subject: execfile question
In-Reply-To: <97297u47not4jc87fh6o1ajual65bhavbv@4ax.com>
References: <mailman.1014117330.24204.python-list@python.org>
Message-ID: <5.1.0.14.0.20020221191737.02924610@63.118.78.184>

I'm not sure if I'm off the mark here, but I believed __del__ guaranteed 
that a particular method got called at the end of it's life.

Mind you, the end of its life, and the end of its scope can be two 
different things. But since I'm not coming from any documented knowledge, I 
know leave my thought with a question marking trialing behind...?

--chajadan

At 09:45 PM 2/20/02 -0800, Tim Roberts wrote:
>Kerim Borchaev <warkid at storm.ru> wrote:
>
> >Can someone explain why while running this script I don't see C.__del__
> >eventually called?
>
>Your object c isn't actually deleted until it gets garbage-collected.  In
>your case, no garbage collection happens until after your program has
>ended.  In that case, there is no sense in taking any further action.
>
>__del__ is not like a C++ destructor.  As I recall, there was a lengthy
>thread on this newsgroup last year on this topic, in which it was concluded
>there is no way for an object to guarantee that a particular method gets
>called at the end of its life.
>--
>- Tim Roberts, timr at probo.com
>   Providenza & Boekelheide, Inc.
>--
>http://mail.python.org/mailman/listinfo/python-list




From jkraska at san.rr.com  Sun Feb 17 17:38:31 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 17 Feb 2002 22:38:31 GMT
Subject: Apology
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <ketv6u49deaj528kctcpj83s8daff0es37@4ax.com> <1896.813T2154T10725160threeseas@earthlink.net>
Message-ID: <6oc07u8ddbq08aaep4sb7771803vmf1euu@4ax.com>

>You are aware that the Python License was changed to be GPL compatable
>after the fact of code?

You are confused. This only applied to _forward_ licensees of the
newer distributions. The old distributions are still covered by the old
license and anyone holding that code can obey the older licensing terms.

One cannot unilaterally change the terms of a contract (which is what a
shrink wrapped license is) unless the contract itself includes an
stipulation which allows it. GPL is not such a license.

Of course, the owners of the software aren't required to distribute the
old code at all and nor are they bound in any way by their own licensing
terms. So, for example, if one releases something under the GPL to the
general public, one can still use one's own work in commercial software,
derive works from it, and ignore the requirements to publish any of this
to the general public.

But what one CANT do is give people rights under the GPL, BSD, or any
other of the classic free licenses and ever be able to expect to take
them back.

It can't be done.

C//



From store_li at sina.com  Mon Feb 25 00:27:27 2002
From: store_li at sina.com (Ricky)
Date: Mon, 25 Feb 2002 13:27:27 +0800
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones> <a58885$5acoc$1@ID-12869.news.dfncis.de> <zVPd8.4096$jC3.163217@news1.east.cox.net> <a5acho$5j86i$1@ID-12869.news.dfncis.de> <slrna7hkcg.i6a.quinn@baht.ugcs.caltech.edu> <just-74F5E1.12532424022002@news1.xs4all.nl>
Message-ID: <a5ci1p$67tml$1@ID-12869.news.dfncis.de>

I agree.
"Just van Rossum" <just at xs4all.nl> wrote in message
news:just-74F5E1.12532424022002 at news1.xs4all.nl...
> In article <slrna7hkcg.i6a.quinn at baht.ugcs.caltech.edu>,
>  quinn at baht.ugcs.caltech.edu (Quinn Dunkan) wrote:
>
> > On Sun, 24 Feb 2002 17:41:17 +0800, Ricky <store_li at sina.com> wrote:
> > >I see now. Thank you! But I am still confuse, why I write???? method
can't
> > >use filename as a parameter.
> >
> > Many files don't have names.
> >
> > E.g. sockets, pipes, StringIOs, etc.
>
> However for methods like ConfigParser's write() it is quite common to
> accept a filename __or__ a file object.
>
> Just




From phd at phd.pp.ru  Fri Feb  8 04:29:53 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 8 Feb 2002 12:29:53 +0300
Subject: web proxy in python?
In-Reply-To: <7xvgd85r6b.fsf@ruckus.brouhaha.com>; from phr-n2002a@nightsong.com on Fri, Feb 08, 2002 at 12:11:24AM -0800
References: <7xvgd85r6b.fsf@ruckus.brouhaha.com>
Message-ID: <20020208122953.D12726@phd.pp.ru>

Hi!

On Fri, Feb 08, 2002 at 12:11:24AM -0800, Paul Rubin wrote:
> Anyone got a nice client side web proxy written in Python?  Something
> along the lines of junkbuster or the j-marshall.com perl proxy, maybe.

   I use proxy3 from http://theory.stanford.edu/~amitp/proxy.html. The
project is stalled, so I slowly update it; you can find my version at
http://phd.pp.ru/Software/Python/#my_proxy (or mirror at
http://phd.by.ru/Software/Python/#my_proxy).

   Other Python proxies are: http://www.softlab.ece.ntua.gr/~ckotso/CTC/,
http://www.zipcon.net/~adamf/adzapper/,
http://www.acooke.org/jara/alfajor/index.html,
http://www.cyberclip.com/webdebug/index.html.

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



From justin at iago.org  Thu Feb 21 02:18:12 2002
From: justin at iago.org (Justin Sheehy)
Date: Thu, 21 Feb 2002 02:18:12 -0500
Subject: A bundle of questions from a Python newbie
In-Reply-To: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
 =?iso-8859-1?q?(Gon=E7alo?= Rodrigues's message of "Wed, 20 Feb 2002
 21:58:50 +0000")
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <x7664r5mmj.fsf@ra.iago.org>

Gon?alo Rodrigues <op73418 at mail.telepac.pt> writes:

> 1. The way I thought that an assignment like
>
> var += 1 
>
> worked was that the interpreter would go to the place pointed by var and
> add 1. In particular, id(var) should remain constant. But this is not
> what happens as a simple test shows it. So what is wrong in my
> perception of what += does? Is this an implemetation issue?

First consider what happens when you do "var = 1":

 - an integer with value 1 is created
 - the name "var" is bound to that integer

In the case you show above, "var += 1" is equivalent to "var = var + 1":

 - an integer with value 1 greater than the value that var is bound to
   is created
 - the name "var" is bound to that integer

A shorter way to say this is that Python's assignments are simply name
bindings and do not modify the values that the names may formerly have
been bound to.  

I should note, however, that the += operator does have the ability to
do what you describe, but only on mutable objects that choose to
enable that behavior.

> 2. What is the idea of having in the loop
>
> for var in range(10):
>     pass
>
> the var variable still available after it? Wasn't it supposed to die (as
> in being out of scope) when the loop ends?

A for loop does not introduce a new scope.  Thus, the variables
created in a for loop continue to exist after the loop finishes.

> And is this a feature, or something that we should not count on?

You can count on it.  Many people use this feature.

> Somehow, I believe that the a new var should be created when the
> loop starts and then destroyed when it ends avoiding any variable
> shadowing along the way.

Aesthetically, that's not a bad idea.  But that's not how it works.  :-)

> 3. This is not a question just a small complaint. I really miss the
> do-until control flow in Python. I always have to waste some extra brain
> cycles to convert it to the while 1, etc. idiom. Hey, we all have a
> right to our complaints, right? ;-)

Sure.  Complain away.  I personally dislike it when languages provide
too many redundant control-flow structures.  More to remember.

> 4. I have read the FAQ about the "self controversy"

There's not much of a controversy among actual users of Python.  I
think that "self" seems to get less flack than the indentation use,
and both are primarily picked on by people that haven't used Python
enough to understand the benefits of the simple explicitness involved.

> I really feel that there must be a better solution. I thought
> something like this could be useful:
>
> a. self doesn't need to be declared.
>
> b. Distinguishing data members from local variables could be achieved by
> instead of using self.whatever use .whatever - a single dot in front of
> the name "whatever".
>
> I don't expect that this solution is adopted (it would break all the
> existing code - ahhh every Nero-like language designer's dream - wreak
> havoc in the citadel of its users and start afresh), what I would like
> to know is how bad/good is this idea.

There is a lot of value to the current model with "self" explicit.  It
is in line with Python's general virtue of explicitness, it leads to
cleanliness in how methods work generally, and provides a simple and
obvious distinction between instance variables and other names.

How good is your idea?  It's fine, if you're designing a new language.
However, for Python, I believe that the current model is better than
yours.  Even if it could be changed at this late date, it shouldn't be.

> Still, What's the rationale behind allowing
>
> class L(list):
>     pass
>
> L.answer = 42
>
> but then not allowing 
>
> L.__dict__['answer'] = 42

What exactly are you trying to accomplish here?

Basically, subclasses of builtin objects like list don't have a normal
dict.  They have a dict-proxy that does everything it needs to.

Why do you think that you want to assign to the class's __dict__?

> 9. What would be better in terms of speed?
>
> for i in range(<whatever>):
>     <whatever> 
>
> Or to code an iterator like

The for loop will almost always be faster for simple cases.  Use
iterators for their useful behavior, not for speed.

> 10. I have a class that works like a list in the sense that supports the
> method __getitem__ and its relatives. But I'm at a loss as to what I
> have to do to support the slice notation.

You could start with __getslice__ and __setslice__.

> 12. This is not exactly about Python, but about programming in general.
> I keep reading about continuations and coroutines and I would like to
> know, in simple terms if possible, what the heck are they and in what
> ways they differ from generators as they exist in Python.

I'll try a quick explanation of continuations.

At any given time in program flow, the current continuation can be
thought of as "the rest of the program", with a hole in it for the
thing currently being executed.  For instance, in the code segment:

f(g(x))

g's continuation is effectively:

f(_)  where "_" is to be filled in when the continuation is run.

Think of your continuation as a function of one argument that will
take your return value as its argument.  It represents everything
that will happen in the program onward from the time that you return.

Unless you have explicit control over them, continuations are just
return points.  If you do have the ability to explicitly handle them,
they are primarily useful for implementing other program control flow
contructs.  

It is rare for continuations to be used directly for something other
than implementation of a control-flow construct.  So, unless you plan
on creating (e.g.) a new microthread system, this simple explanation of
continuations is probably all you'll ever need to know.

-Justin

 




From husam at macrozon.net  Wed Feb 20 07:09:26 2002
From: husam at macrozon.net (husam)
Date: Wed, 20 Feb 2002 13:09:26 +0100
Subject: How do you conver output of "popen('ls')" to a list?
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com> <mailman.1014244477.1738.python-list@python.org>
Message-ID: <3C7391F6.CB6CAE01@macrozon.net>

Neil Schemenauer wrote:

> I'm a bit late coming in on this thread but what's wrong with:
>
>     os.listdir('.')
>
> It's more portable and faster.
>
>   Neil

os.listdir('.') gets you all the files in the '.' dircetory. This want
work if i want to process some selected files. In contrast to this,
"os.popen() allows you to pass arguments like "ls *txt" to retrieve only
*.txt files.



From mhammond at skippinet.com.au  Fri Feb  1 21:09:18 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 02 Feb 2002 02:09:18 GMT
Subject: Python 2.2 + Win32All 146 = Python/ASP Failures
References: <vtk68.20338$8d1.7162721@news1.rdc1.md.home.com> <3C59DA37.9070506@skippinet.com.au> <JEF68.22313$8d1.8213037@news1.rdc1.md.home.com>
Message-ID: <3C5B4A9A.8000809@skippinet.com.au>

Joe Salmeri wrote:
> Mark,
> 
> I grabbed framework.py from win32all-142.exe and replaced it in the
> installation as you suggested.  I also deleted the framework.pyc just to
> make sure the new version was compiled.  Next I restarted the web server,
> just as a sanic check to make sure it saw the changes to the python asp
> engine.

Sorry - my mistake.  You need to go from win32all-140 down to 
win32all-133 (inclusive) for a working version :(  I have mailed you a copy.

Mark.



From quinn at yak.ugcs.caltech.edu  Thu Feb 21 21:00:27 2002
From: quinn at yak.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Feb 2002 02:00:27 GMT
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com> <CMfd8.5037$D6.14223@news.iol.ie> <u7b81629bo9o4a@corp.supernews.com>
Message-ID: <slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu>

On Thu, 21 Feb 2002 17:35:44 -0800, Virginia O <vodine at direct.ca> wrote:
>Thanks, but the database will have to hold sub-groups etc. It will be easier
>for me to enter and organize the items in a database.
>I only gave the scenario as an example. The actual program will hold
>hundreds of items.

Hundreds is just fine for flat text.  If you have a few thousand and you're
modifying it a lot, then flat text will start being clunky.

In that case, shelve should work fine, especially for complicated data
structures since you can dump python types directly, with no clunky converting
to/from some format SQL likes.  And it's much simpler to have everything in
python instead of a whole other database program with its own language, etc.


From pete at shinners.org  Thu Feb  7 23:28:13 2002
From: pete at shinners.org (Pete Shinners)
Date: Fri, 08 Feb 2002 04:28:13 GMT
Subject: ActivePython and wxPython
References: <RiH88.37389$Jq.1973082@news2.calgary.shaw.ca>
Message-ID: <3C6354F5.4070109@shinners.org>

Andrew wrote:
> Sorry if this question is inappropriate for this newsgroup, but I was
> wondering if there are known problems using wxPython with ActivePython?  I
> try to run the first example (from www.wxpython.org) in the wxPython
> tutorial but get a crash when I attempt to create a wxFrame object.  I
> stepped into the code and it's failing on an 'apply' call.

you are probably trying to run the wxpython program from inside 
"PythonWin". you'll quickly find IDLE and PythonWin don't work very well 
when running programs that do their own GUI work.

there's other editors that do a proper job of this. i hear good things 
about the "IDLE Fork", and personally have done ok with scite.

http://idlefork.sourceforge.net/
http://www.scintilla.org/SciTE.html



From ngps at madcap.netmemetic.com  Wed Feb 20 09:51:35 2002
From: ngps at madcap.netmemetic.com (Ng Pheng Siong)
Date: 20 Feb 2002 14:51:35 GMT
Subject: XML-RPC over HTTPS/SSL
References: <8rKc8.4$5vh.135400448@news.telia.no>
Message-ID: <a50d5n$1ee$1@clematis.singnet.com.sg>

According to Thomas Weholt <thomas at gatsoft.no>:
> I got a xmlrpc-server running using M2Crypto, making it encrypted with SSL.
> I cannot get any xmlrpc-client to connect though, no matter what.

Do demo/ssl/xmlrpc_cli.py and demo/ssl/xmlrpc_srv.py work on your
installation?

[In one window.]
ngps at spinnaker:~/prog/m2/demo/ssl$ python xmlrpc_srv.py

[In another window.]
ngps at spinnaker:~/prog/m2/demo/ssl$ python xmlrpc_cli.py
[1, 2, 3]
['SOAP']

[Back in the first window...]
LOOP: SSL accept: before/accept initialization
LOOP: SSL accept: SSLv3 read client hello A
LOOP: SSL accept: SSLv3 write server hello A
LOOP: SSL accept: SSLv3 write certificate A
LOOP: SSL accept: SSLv3 write key exchange A
LOOP: SSL accept: SSLv3 write server done A
LOOP: SSL accept: SSLv3 flush data
LOOP: SSL accept: SSLv3 read client key exchange A
LOOP: SSL accept: SSLv3 read finished A
LOOP: SSL accept: SSLv3 write change cipher spec A
LOOP: SSL accept: SSLv3 write finished A
LOOP: SSL accept: SSLv3 flush data
INFO: SSL accept: SSL negotiation finished successfully
XMLRPC call: Testing (1, 2, 3)
localhost - - [20/Feb/2002 22:37:20] "POST /RPC2 HTTP/1.0" 200 -
LOOP: SSL accept: before/accept initialization
LOOP: SSL accept: SSLv3 read client hello A
LOOP: SSL accept: SSLv3 write server hello A
LOOP: SSL accept: SSLv3 write certificate A
LOOP: SSL accept: SSLv3 write key exchange A
LOOP: SSL accept: SSLv3 write server done A
LOOP: SSL accept: SSLv3 flush data
LOOP: SSL accept: SSLv3 read client key exchange A
LOOP: SSL accept: SSLv3 read finished A
LOOP: SSL accept: SSLv3 write change cipher spec A
LOOP: SSL accept: SSLv3 write finished A
LOOP: SSL accept: SSLv3 flush data
INFO: SSL accept: SSL negotiation finished successfully
XMLRPC call: BringOn ('SOAP',)
localhost - - [20/Feb/2002 22:37:20] "POST /RPC2 HTTP/1.0" 200 -


> xml-rpc-server using SSL/HTTPS , preferrably with user authentication also?
> ( to make it perfect, it would accept gzip-encoding to speed up transfers
> :-) )

M2Crypto's XMLRPC does user authentication. Well, it is using the eff-bot's
xmlrpclib, so it's really that library that does. ;-)

No gzip-encoding, sorry. Shouldn't be too difficult. 


> I keep seeing stuff about SSL/HTTPS-support in standard python, but ... ??
> Do you compile the interpreter with OpenSSL etc ?? My clients and servers
> will most probably run 60% on Linux and 40% windows, and re-compilation
> isn't always feasable for windows-users.

Does Python 2.2's SSL do servers? I am of the impression it doesn't.

Cheers.

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



From QnickQm at alum.mit.edu  Tue Feb  5 11:40:36 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Tue, 05 Feb 2002 16:40:36 GMT
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de>
Message-ID: <slrna602tm.8mg.QnickQm@localhost.localdomain>

In article <a3p190$19v1rc$1 at ID-98166.news.dfncis.de>, MDK wrote:
> "Emile van Sebille" <emile at fenx.com> wrote in message
> news:a3p0ss$19ob7s$1 at ID-11957.news.dfncis.de...
 [...]
>> Call the method append and not __append__?
 [...]
> Yes, I figured that to be the case because it should be transparant to the
> caller.  However, that is not the solution.
> 
> Thanks though.

Really?  This works for me (under Python 2.2)


class newlist(list):
  def append(self, o):
    print "in subtype"

L = newlist( [1,2,3] )
print L
L.append(5)

#output:
#[1, 2, 3]
#in subtype

Yrs,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From dan at dberlin.org  Thu Feb 21 09:01:23 2002
From: dan at dberlin.org (Daniel Berlin)
Date: Thu, 21 Feb 2002 09:01:23 -0500
Subject: Why this behavior? (was: Stupid Lambda Tricks)
In-Reply-To: <js5d8.116$Az6.466@news.hananet.net>
Message-ID: <7D4CA6CA-26D3-11D6-922E-000393575BCC@dberlin.org>

On Thursday, February 21, 2002, at 07:20  AM, Jonathan Gardner wrote:

> < snip excellent lambda tricks >
>>
>> Many list functions return None, so you cannot easily use them in a
>> stacked expression inside lambda. Examples of this are [].sort() and
>> [].reverse(). Instead of writing:
>>
>> list.some_function_that_returns_None()
>> function(list)
>>
>> you can write
>>
>> function((list.some_function_that_returns_None(),list)[1])
>>
>> Here is an example:
>>
>> import random
>>
>> sort = lambda x: (x.sort(),x)[1]
>> reverse  = lambda x: (x.reverse(),x)[1]
>>
>
> I've always wondered:
>
>>>> "abc".upper()
> "ABC"
>>>> [3,1,2].sort()
> < nothing >
>
> Why is that? Is it just easier to implement it this way, or is there 
> some
> sound philosophical reason for this?

Sound reason, IIRC (someone will correct me if i'm wrong).

Upper returns a *new* string with the letters upper-cased. The original 
is not affected.
On the other hand, sort/reverse are *in-place*, and affect the original.
Returning the "new" original would give the impression that the original 
had not been affected.

--Dan




From patio87 at aol.com  Sun Feb 10 21:41:59 2002
From: patio87 at aol.com (Patio87)
Date: 11 Feb 2002 02:41:59 GMT
Subject: defining functions
Message-ID: <20020210214159.13516.00000450@mb-cq.aol.com>

I am pretty new to python, and I dont understand the parameters when defining
functions. Whenever I see a function definition with a argument It seems like
the argument already is a variable, but it goes threw the function and it makes
no sense. I dont know if what I just said made any sence but if you can help me
please reply

def factorial(n):
    if n == 1:
        return 1
    else:
        return n * factorial(n-1)
What the hell does 'n' have assingned to it?



From mcherm at destiny.com  Tue Feb 19 08:56:03 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 19 Feb 2002 08:56:03 -0500
Subject: Adding static typing to Python
Message-ID: <3C725973.90803@destiny.com>

Peter Milliken:
 > When I compile a Ada program cleanly I know that 95% of the bugs
 > are out of it - before I even start testing! :-)

Brian Quinlan:
 > I would say that 99% of my bugs are due to design mistakes [...]
 > or semantically incorrect algorithms[.]

Jason Orendorff:

> I have added a sys.excepthook in the Lib/site.py file
> on the PC where I do most of my work.  It logs uncaught
> exceptions (except in interactive mode).  In a week or two,
> I should have some real numbers to share.
> 
> Obviously this is a crude and subjective measurement,
> but it's better than (ahem) nothing.


Jason:

Yes... it's WAY better. Thanks for running the test, and I eagerly await 
your results!

Furthermore, the experiment might be better if several people joined in. 
  I, for one, will volunteer to run with this hook for, let's say, 2 
weeks (or whatever time period seems right) and contribute my results -- 
if there are other experimental subjects I'll even offer to be the one 
who compiles the results.

Sound good? Send me a copy of the hook you wrote and I'll try it.

-- Michael Chermside




From logiplexsoftware at earthlink.net  Mon Feb 25 12:15:47 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 25 Feb 2002 09:15:47 -0800
Subject: Newbie: Crazy, but Quick
In-Reply-To: <5d283efe.0202230846.1802e6a2@posting.google.com>
References: <u7ag65cbc0q3e@corp.supernews.com>
	<CMfd8.5037$D6.14223@news.iol.ie>
	<u7b81629bo9o4a@corp.supernews.com>
	<5d283efe.0202230846.1802e6a2@posting.google.com>
Message-ID: <20020225091547.290632f2.logiplexsoftware@earthlink.net>

On 23 Feb 2002 08:46:14 -0800
Alan Daniels wrote:

> "Virginia O" <vodine at direct.ca> wrote in message
news:<u7b81629bo9o4a at corp.supernews.com>...
> 
> > If it was upto me I'd just use a simple delimited text file :)
> 
> Or simpler yet, just store the data in one large Python data structure
> (ex: a tuple of dictionaries, a tuple of tuples, etc), and just have
> your program import it as needed.
> 
> You can always move to a true database later if this doesn't scale,
> or if you one day need relational-style joins rather than simple
> flat-file access.

This might not be a bad solution.  Since SQL queries in Python are
typically returned as tuples or lists, it would be fairly trivial to
replace the file save/restore functions with a SQL backend at a later date
if needed without affecting program logic much.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From maxm at mxm.dk  Thu Feb 28 17:52:24 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 28 Feb 2002 22:52:24 +0000
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz> <3C7CFAF8.1080108@mxm.dk> <Xgaf8.10287$pC1.1854469@newssrv26.news.prodigy.com> <3C7E8D47.21BE30B3@ccvcorp.com>
Message-ID: <3C7EB4A8.7080402@mxm.dk>

Jeff Shannon wrote:
> 
> dsavitsk wrote:
> 
> 
>>"Max M" <maxm at mxm.dk> wrote in message news:3C7CFAF8.1080108 at mxm.dk...
>>
>>
>>>btw: if you really want to store it in a list you can get dict-like
>>>behaviour with list speed like this:
>>>
>>is this a python idiom that people often use? cause i've not seen it before,
>>but it is really clever
> 
> No, it's not something that people often use.  Generally, if you want dict-like
> behavior, you just use a dict.

I absolutely agree!

regards Max M



From sorr at rightnow.com  Thu Feb  7 11:21:53 2002
From: sorr at rightnow.com (Orr, Steve)
Date: Thu, 7 Feb 2002 09:21:53 -0700 
Subject: Saving Browser State without Cookies
Message-ID: <D408B6AC060DB24790622E1A54204DB00823AA89@xchange.rightnowtech.com>

You're right. Behind the scenes and transparent to the code, the sessions
feature of PHP4 either uses the hidden variables method or the cookie method
depending on browser support, settings in your code, or server-wide
settings. The point is, state management is core PHP functionality which you
get "for free" and you don't have to cut much code to implement it. 

In this regard I'm looking into Albatross. As a welcome departure from the
norm with frameworks, the number of lines of code is actually exceeded by
the number of lines of documentation. ;-)  

Speaking of verbosity, I got your book. :-) Quite comprehensive. Great
job!!! 

In the book you mention Albatross along with many other Python solutions.
(There's so many it's overwhelming to the relative newbie.) Preferring
brevity and simplicity, Albatross looks like a very well engineered
solution. 

It would be great if someone could share their experience with using
Albatross in a real world Python/Web implementation. 

Anyone???????


AtDhVaAnNkCsE,
Steve Orr
 


-----Original Message-----
From: Steve Holden [mailto:sholden at holdenweb.com]
Sent: Thursday, February 07, 2002 7:52 AM
To: python-list at python.org
Subject: Re: Saving Browser State without Cookies


"Orr, Steve" <sorr at rightnow.com> wrote in message
news:mailman.1012938609.3651.python-list at python.org...
> Python Newbie here.
>
> I need to migrate a web application from PHP to Python and I have a
> requirement to save state info WITHOUT using cookies where much of the
state
> data MUST not be visible to the end users. The PHP application allows
users
> to logon to one of several databases using any of a number of Oracle
> user/passwords. The logon info needs to be saved for queries on other
pages
> in the app. With PHP it's easy to avoid cookies by saving this info using
> its session handling functions. Is there an EASY way to do this in Python
> without having to adopt the Mother of all frameworks?
>
Disclaimer: I have never used PHP.

However, *if* PHP allows you to associate requests with session state, then
*either*:

a) it does use cookies, you just don't know about them, or
b) it puts session references in the URL.

There simply isn't any other available mechanism. So maybe what you're
looking for is an interface that hides the cookie implementation from the
programmer?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py




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



From threeseas at earthlink.net  Tue Feb  5 23:54:21 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:21 GMT
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk>
Message-ID: <3943.801T1063T14354065threeseas@earthlink.net>

On 03-Feb-02 20:41:31 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Mon, 04 Feb 2002 00:00:53 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>>On 30-Jan-02 06:52:48 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>>On 30 Jan 2002 00:48:40 GMT, Peter Seebach <seebs at plethora.net> wrote:
>>
>>>>Anyway, it looks to me like Python is the wrong language for what you want
>>>>to do.  C# is almost certainly the best choice for this; it's designed for
>>>>the kinds of tasks you're looking at,
>>
>>>Is Tim trying to lock people in to the Microsoft monopoly, then?
>>
>>Why do you make up things by taking things out of context?

>Why do you misinterprete a joking question as a personal attack on you?
>You seem to have very fragile self-esteem...

>>Are you naturally dishonest or do you work at it?

>Yes. I mean no. Oops I've been caught lying again.

>>Autocoding will be a nice thing to have, it'll get rid of all the arrogant
>>jerks.

>Then I suggest you create it.

>--
>===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
>Herbivore, a zero-effort email encryption system. Details at:
><http://www.vision25.demon.co.uk/oss/herbivore/intro.html>


You Phil are a simple programmed idiot.

I recall a delphi conference I attended. During the very boring talk the
speaker was giving in going over what all delphi can do to ease the
programmers task of programming, he didn't fail to mention that for the
professional programmers of the audience, there should be no worry about
the ease of delphi programming causing everybody and their brother to
become programmers and undercut the pay scale of those in the room. Be
assured, he said, you won't find carpenters using it.

Phil, I am a carpenter, a set/scenery carpenter. I attended the conference
in order to help me determine if delphi could be used to help me automate
much of the standardized calculating that I was having to do in my job.


I don't give a shit if you have more programming experience then anyone
else in the world Phil. For all that would tell me is that you have more
experience in not doing it right, then anyone else in the world, making
you also the most difficult person in the world to retrain.

Programming is two deminsional, not three.

But even in the three deminsional work I do, when something is done
enough, we figure out jigs and methods in order to do more in less time
with higher quality.

To understand some basic idea of the difference between the two
deminsional world and the three deminsional world, is like colors of
light, red, blue and green. Remove one of them to represent the two
deminsional world and you are greatly constrained by far more than half
of all colors possible in having three primary colors.

The three deminsional world has more variables and as such it is more
difficult to come up with more universal jigs and methods. But programming
is two deminsional and as such the universal jiging or methods should far
better exist than they do. If you want to call near non-existant, existance
at all.

And that's the thing Phil, the jiging of an autocoder tool as what I'm on
about, really does screw you over. Screw My efforts to be nice about
something is is not going to be preceived as positive no matter how I
paint it.

Regardless of all the resistance that may have existed, the automobile
still came along, and then American Auto manufactures found it better to
just start fresh than to try and retrain and retool old production, in
order to compete with the Japanesse, who found a better way, new jiging,
rigging and automating.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From st-newsgroups at thomas-guettler.de  Tue Feb  5 05:34:18 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Tue, 05 Feb 2002 11:34:18 +0100
Subject: Random 32Bit integer was: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <mailman.1012902609.27332.python-list@python.org>
Message-ID: <3C5FB52A.20902@thomas-guettler.de>

Gerhard H?ring wrote:

> Le 05/02/02 ? 10:16, Thomas Guettler ?crivit:
> 
>>I have two python version installed:
>>with the cygwin version "print 2**32-1" works
>>but with the zope version I get OverFlowError.
>>
> 
> Zope uses Python 2.1, while your Cygwin Python is 2.2, right?


yes

[cut]

> Btw. it looks like random.randint/randrange really want ints and won't
> accept longs. A workaround could be to use a construction like:
> 
>     print random.randrange(0, sys.maxint) * 2L


This would give me a long. But I need a 32Bit Integer.
Unfortunately there is no sys.minint.

OK, it is not that important for me. But it would be nice to know
the answer (How to get a radom 32 Bit integer)

thomas





From joshm at taconic.net  Mon Feb 25 17:51:36 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 25 Feb 2002 17:51:36 -0500
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5dia2.3vs4847.1@kserver.org>
Message-ID: <3c7abd92_1@corp.newsgroups.com>

> > If you had to give a 15 minute introduction to Python, what would you
cover?
> > What is a good example of a simple, but useful, text-based app for a
Wintel
> > environment, that has good "wow" quality?
>
> Who is your audience? What is their background and what types of things
> would they like want to do with a computer?

They are the tech faculty for the local tech school.  It is meant to
demonstrate my ability to teach (something).  I have some ideas, but was
looking for a few suggestions.  I expect that I won't be able to determine
the expertise of the audience until the last minute, so I'm trying to come
up with a few different levels of topics, and I'll pick one when it happens.

I think that something demonstrating lists, tuples, and dicts will impress
people who are technical (can program in some language) but haven't seen
python before.  Ideas which specifically show successively more complicated
things by adding a few things to the (evolving) source is really the ticket.

Maybe writing something like wc, then perhaps an indexer (using a dict to
construct a list of references, then producing pretty output).  Hmm...  I
think I answered my own question.

Other suggestions are more than welcome.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From arturs at iidea.pl  Fri Feb 15 14:32:35 2002
From: arturs at iidea.pl (Artur Skura)
Date: Fri, 15 Feb 2002 19:32:35 +0000 (UTC)
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net> <Xns91B6CF519E265cliechtigmxnet@62.2.16.82>
Message-ID: <slrna6qsm2.bob.arturs@aph.waw.pdi.net>

Chris Liechti wrote:
> Artur Skura <arturs at iidea.pl> wrote in 
> news:slrna6qqs9.b35.arturs at aph.waw.pdi.net:
>> I have a specific problem: I have to create dialogs on the fly, and
>> in those dialog text together with text entries.
> 
> Jython 2.1 on java1.4.0 (JIT: null)
>>>> import java.awt
>>>> f = java.awt.Frame()
>>>> f.setLayout(java.awt.FlowLayout())
>>>> f.add(java.awt.Label("hello"))
>>>> f.add(java.awt.TextField("world"))
>>>> f.add(java.awt.Label("something like that?"))
>>>> f.pack()
>>>> f.show()
> 
> does this go in the right direction? the Layouters in java ATW and 
> SWING make such things realy easy!

Mmm, I don't know what's wrong, but after adding longer entries I
got a result like this:

http://7bulls.com/~askura/awt.png

It's not what it's supposed to be, the second part got lost.
What is more, I will need to compose dialogs with much more text
(whole paragraphs). The software is for kids, they will have to
insert correct word into the text entries. 
Seems I will have to calculate everything anyway...

Pozdrowienia,
Artur
-- 
http://www.wolneprogramy.org


From skip at pobox.com  Thu Feb 28 22:19:20 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 28 Feb 2002 21:19:20 -0600
Subject: map-like function on dict values?
In-Reply-To: <a5md0a$an8$1@animus.fel.iae.nl>
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
        <mailman.1014861466.23611.python-list@python.org>
        <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com>
        <mailman.1014929989.16553.python-list@python.org>
        <a5md0a$an8$1@animus.fel.iae.nl>
Message-ID: <15486.62264.311324.231880@12-248-41-177.client.attbi.com>

    >> which relies on 2.2's new "for k in dict" capability to completely
    >> avoid list creation.

    Carel> Ah a new contestent, double fun, better algoritme better timing
    Carel> (linux too), but the real iter still wins, although by a very
    Carel> slight margin:)

    >> iter 6.69
    >> iteritems 28.36
    >> mapvalue 8.12

    Carel> iter 10.95
    Carel> iteriteritems 10.43
    Carel> iteritems 52.15
    Carel> mapvalue 12.88

Not on my system:

iter 6.73
iteritems 28.71
iteriteritems 7.28
mapvalue 8.19

:-)

I'm using 2.3a0 (aka CVS), and compile with

    -O3 -minline-all-stringops -fomit-frame-pointer

Skip



From chris.gonnerman at newcenturycomputers.net  Tue Feb 19 22:55:49 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 19 Feb 2002 21:55:49 -0600
Subject: cursor control
References: <3c731781.5352364@nntp.surry.net>
Message-ID: <00e701c1b9c2$82753d40$0101010a@local>

----- Original Message ----- 
From: "Charles W. Strickland" <cstrickland at surry.net>


> Is it possible to control cursor positioning in python without using
> TKinter?  I wanted to write a program that required it, but could not
> find anything anywhere to explain how to do it.  There were a couple
> references awhile back, but I never did see a definitive answer.  I
> finally wrote the program in C - what a mess!

What does "cursor control" mean in reference to Tkinter?  Sounds like
you are writing Win32 console programs... if you are, you need:

    http://newcenturycomputers.net/projects/wconio.html

my Windows Console I/O module (or you might visit http://effbot.org to
get /F's console module; I never had much luck with it though.)

If this is not what you are doing, please tell us in more detail; in
particular, what platform (Linux/Unix, Windows, Mac OS or OSX, etc.)
and what version of Python (1.5.2, 2.0, 2.1.x, 2.2, etc.)





From sholden at holdenweb.com  Mon Feb 18 10:25:28 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 18 Feb 2002 10:25:28 -0500
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org> <a4ss56$akj6@news1.gtech.com>
Message-ID: <b7uc8.8411$Zp2.790511@atlpnn01.usenetserver.com>

"Peter Milliken" <peter.milliken at gtech.com> wrote [after a "big lunch"]...
>
> "Jason Orendorff" <jason at jorendorff.com> wrote in message
> news:mailman.1014098489.15429.python-list at python.org...
> > Milliken, Peter wrote:
> > > Strong typing requires that you design before you code (Extreme
> > > Programmers beware! :-)) - therefore by the time a clean compile
> > > is achieved 95% of the bugs *are* out :-) That is one of the
> > > primary advantages to strong typing, it *forces* you to do the
> > > design up front otherwise you will never get a clean compile.
> >
> > Hmm.  Perhaps you can explain this in more detail.
> >
>
> Statement is true when a number of conditions are satisfied :-) One of
which
> is that we are talking a non-trivial program!
>
> Strongly typed languages encourage you to override the base types so that
> you aren't forced one day (as an example) to go through all of the source
> looking for places where floats are used and change them to double
floats -
> happened on one rather large project I worked :-) They changed the
compiler
> and it defined float as being 32-bit, the previous version defined float
as
> 64-bit - caused no end of heartache as you can imagine :-)
>
I don't see anything about strong typing that encourages you to define
artificial types in this way. I understand that many programmers use thisp,
thatp and theotherp to point to different types of value despite the fact
that all three lue types are pointers, which will thereforee have the same
storage requirements. But this won't force anyone to declare an "fpthingie"
type to protect them against such a mangling of backward compatibility.

Insane compiler vendors, unfortunately, can't be legislated against.

[don't add temeratures to volumes]
>
> Of course, a strongly typed checked language won't guarantee that such
> features are used! This is always up to the programmer (I once saw a
> programmer mangle his Forth code so badly it looked like his previous
> language - Fortran! :-)). But if they are used then you are forced to make
> sure that you stick with the "rules" as you define them via your type
> declarations. So our argument here is presumptive of the features being
> used - you could argue the same thing about any language feature i.e. I
> could argue that the OO features of Python offer no advantage - based upon
> never using them! :-)
>
> > Are you arguing that compilers for strongly typed languages tend to
> > reject ill-designed or thrown-together code?  I hadn't noticed.
> >
>
> Assuming the above, strongly type checked compilers/languages will indeed
> reject ill-designed and thrown together code - it might take a while, but
> eventually the programmer will back himself into a hole :-). You usually
> have some form of "override" even so, so you could litter your code with
> coercions (I have seen this too! :-)) but this is relatively easy to
> discover, so the culprit is soon exposed.
>
My experience is more typically that inexperienced programmers tend to paint
themselves into a corner by over-zealous typeing, and must resort to
inappropriate coercions as a corrective.

> > Surely you're not arguing that extra-productive programming tools
> > are bad because they encourage the programmer to waste his or her
> > own time by skipping the design step.
>
> Sorry too many "negatives" here for me to work out what you are driving
at -
> I just came back from a "big" lunch, so the old brain isn't functioning on
> all cylinders too well :-)
>
Post in haste, repent at leisure? Jason simply finds it difficult to believe
that you think high productivity is counter-productive.
> >
> > Maybe this is true in some sense - since I enjoy coding Python so
> > much, perhaps I'm overeager to get started?  But if so, then the
> > solution is sane development practices and a touch of
> > self-discipline.  Which will likely come in handy on any ten-year
> > project.
> >
> >
>
> Yes, sane development practices are essential on such large projects -
> doesn't matter what language is used. Part of my argument is that language
> features are designed and provided as further "encouragement" down the
path
> of "correct" practices. Typing is recognised as one of these
features......
>
There are languages that make it difficult to write "bad" programs. I would
cite Algol 68 and C++ as probably the epitomes of such (language) design
style. Then there are languages that make it easy to write good programs. In
my experience Icon and Python are probably the best exemplars of this style.
The debate about which of these styles is preferable will probably go on for
ever.

or-at-least-until-the-heat-death-of-the-universe-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py







From christophertavares at earthlink.net  Sat Feb 16 14:23:53 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Sat, 16 Feb 2002 19:23:53 GMT
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <86n0yaw8u4.fsf@apo.hmotakef.int> <3C6E95FA.F95C914B@earthlink.net>
Message-ID: <dlyb8.3101$tu6.303907@newsread2.prod.itd.earthlink.net>

"Hans Nowak" <wurmy at earthlink.net> wrote in message
news:3C6E95FA.F95C914B at earthlink.net...
> Henrik Motakef wrote:
> >
> > Hans Nowak <wurmy at earthlink.net> writes:
> >
[ ...snip... ]
>
> That is probably my fault. Here's a short summary. I want to store a
> multiline string, containing newlines, in SQL Server, using an
> updategram in XML. I post the record. I retrieve the record back.
> Newlines are gone from the string.
>
> I get the records back as XML too. In raw, unparsed XML (just a string)
> the newlines are still there. As soon as I parse the XML, they are
> somehow removed. See my other post to Martin von Loewis.
>
> I don't know much about XML; if whitespace is considered insignificant,
> like in HTML, then it might be removed when parsing.
>

Whitespace in XML is a little odd. It depends on the DTD/Schema wether it's
significant or not. The normal solution is for the XML parser to have a
"preserveWhitespace" setting, which will guarantee that you get everything
(newlines and all). However, I took a spin through the minidom docs and
couldn't find anything. Hmm... guess that's why it's a "mini" DOM. Have you
tried using either SAX or 4DOM instead?

> > However, one thing that might possibly be related is that XML parsers
> > are required to convert all of the funky end-of-line markers around
> > (i.e. "\r\n", "\n", "\r") to a single newline "\n". Scince you Use SQL
> > Server, I assume that you are also running Windows, where the
> > "official" newline is "\r\n". Could that be related to your problem?
>

This is probably not a problem - MSXML is fairly flexible about such things,
and I'm sure that SQL server is either using that parser, or using a good
chunk of code from it.

 > It was part of a previous problem; the XML supposedly only contained \n,
> where \r\n was desirable. While investigating it, I stumbled upon a
> different, more serious problem-- no newlines at all.
>
> On a side note, any operating system with an Internet connection can talk
> to the SQL Server database using SQLXML. All you need is some HTTP
> functions (like Python has in httplib) to talk to it. No drivers needed.
> All communication with the server is done through XML. Pretty fly for
> a M$ product. ;-)
>

Yeah, MS does get some stuff right once in a while. The real question is -
what if you want to STOP somebody from talking to it over http?

-Chris





From peter.milliken at gtech.com  Sun Feb 24 21:07:19 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 25 Feb 2002 13:07:19 +1100
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <cb4ba455.0202210525.6687556f@posting.google.com>
Message-ID: <a5c6bs$gld2@news1.gtech.com>

"Bill Tate" <tatebll at aol.com> wrote in message
news:cb4ba455.0202210525.6687556f at posting.google.com...
> "Peter Milliken" <peter.milliken at gtech.com> wrote in message
> > Personally, I have used Python for several small, personal projects and
the
> > odd "script" that searches text files - even here, the only perceived
> > advantage is the library support, I often wonder whether I wouldn't be
> > better off using my real language of choice (Ada) because of it's strong
> > typing - it too has interfaces to the standard GUI tools etc. My next
> > personal project will not be in Python, it is a "cute" language and
allows
> > some nice facilities i.e. creating a script at run-time and then running
it
> > can be very powerful in the correct circumstances - so I will probably
use
> > an Ada-Python API for these situations and get the best of both worlds
:-).
> > When I compile a Ada program cleanly I know that 95% of the bugs are out
of
> > it - before I even start testing! :-) Of course, it takes me longer to
write
> > up front but overall development effort (in my subjective opinion -
since I
> > have no metrics :-)) is that it is much lower.
>
> Peter,
> Perhaps you should also use adjectives like "toy-like," "child-like,"
> "foolish," "trivial," and "not a real language" while your at it.  You
> offer an opinion without supporting metrics and effectively relegate a
> language to the realm of triviality.  Should I presume, that you also
> lump the python developer community within the realm of "caution:
> children at play?"  Perhaps you could spare us the "cute" adjectives.
> Your remarks smack of intellectual superiority.
>

No I shouldn't use adjectives like that because I don't believe they are
true. Python is a language that is growing (in leaps and bounds) so I don't
necessarily see why a description of "cute" doesn't apply. The mere fact
that it is growing and changing shows that people recognise there are
deficiences (or perhaps lack of features would be a preferable term to
you?). So "cute" seemed a not unreasonable adjective at the time since in my
mind it had certain connotations - clearly the image that sprung to your
mind didn't match that description - I can't help that at all.

> You're right to assert there are risks and costs associated with using
> a dynamically typed language.  However, are you of the opinion that
> this is a community that doesn't understand what those risks and costs
> are?  Perhaps this is a community that understands, better than others
> I suspect, how/when to make practical tradeoffs and how to apply the
> necessary tradecraft to know when to use python and when not to?
>
> Bill

No, I have had some very productive dialog with individuals that have shown
they have a very good grasp of the risks involved and have chosen Python
despite (or even because of) those features. I don't necessarily agree with
some of their stands/decisions just as they may not agree with mine but then
we all come from completely different backgrounds with different experience
sets and above all, requirements :-). So I respect each and every one of
those decisions. All I am guilty of is expressing some musings aloud about
where I and how I personally intend to trial the use of Python next. I
believe I have tried to explain my choices (in probably a fairly lose style
:-)). Criticism should only be levelled when the party has experience in the
appropriates aspects of the items under discussion. This thread has shown to
me that many people do not have a good grasp of where I am coming from
because they have never used languages with the feature set that I am
referring to - so that is OK, I attempt to make my remarks in that
environment. That is the background of my use of the term "ignorance" -
hardly a good term to use, I admit freely since I am just as guilty of that
as anyone. I certainly haven't tried all available languages!. I have used
C/C++, Python and Ada - so I feel some confidence in comparing the relative
merits. Others have backgrounds in languages like Java and they have shared
their experiences. Python is a scripting language that has some excellent
language features - these features are growing almost daily. It definitely
has a place in my "toolbox", all I am saying is that at the moment, I don't
see it as being my "primary" tool because of what I perceive as deficiences.
These "deficiences" are absolute strengths though in certain situations.

Python appears to attract individuals with a certain approach to
programming, these type of discussions have allowed me to see some of the
background for why they have chosen Python as their development language.
This has provided me with a picture of the "typical" Python programmer that
may be widely wrong! :-) But it does allow me to take this picture into
allowance in future such discussions. Hopefully I will be more diplomatic in
future discussions :-)

Peter






From syring at email.com  Sun Feb  3 05:23:40 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 11:23:40 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <a3j1a4$18e7ov$1@ID-7529.news.dfncis.de> <f73q5uo64b3uhk6hbvb9u4834bvkaf8dr4@4ax.com>
Message-ID: <a3j37a$18b1pm$1@ID-7529.news.dfncis.de>

"Lawrence Oluyede" <rhymes at myself.com> schrieb
> On Sun, 3 Feb 2002 10:52:54 +0100, "Karl M. Syring" <syring at email.com>
> wrote:
>
> >But it shouldn't it build with pdcurses?
>
> What is  pdcurses?

A version, that works on Windows, at least the Lynx builds with it
(http://pdcurses.sourceforge.net/). Someone must have made it work with
Python.

Karl M. Syring




From webmaster at prankcallsunlimited.com  Mon Feb 18 00:41:45 2002
From: webmaster at prankcallsunlimited.com (Bob Pranky)
Date: 17 Feb 2002 21:41:45 -0800
Subject: General Collection of FREE SOUNDS
Message-ID: <638cbd51.0202172141.765ace33@posting.google.com>

Free sounds for all your needs at:  http://www.a1freesoundeffects.com
General collection includes lots of different categories in wav file
format. Sound effects for all needs including professional and
commercial sound effects:
http://www.a1freesoundeffects.com/service.htm

Audio samples and loops for multimedia purposes!


From tim.one at home.com  Fri Feb  1 23:28:11 2002
From: tim.one at home.com (Tim Peters)
Date: Fri, 1 Feb 2002 23:28:11 -0500
Subject: Why has 'dir()' changed in Python 2.2?
In-Reply-To: <yM1W8kKkXktc092yn@bellatlantic.net>                         
Message-ID: <LNBBLJKPBEHFEDALKOLCMEFLNJAA.tim.one@home.com>

[Dr. David Mertz]
> I just noticed a change in Python 2.2... that turns out to break my
> [xml_pickle] module.  It shouldn't be hard to fix (although I *do* have
> a decision to ponder about the best fix)... but I mostly wonder what the
> thoughts behind the change were

The original intent of dir() was to list the names available from a module
object, for convenience at an interactive prompt.  As the years went on,
more gimmicks got added to it in more-or-less random fashion.

For instances of classes, the more-or-less accidental behavior that got
implemented was to list the keys of the instance's dict.

With type/class unification in 2.2, what were instances of builtin types
*became* instances of classes, and then (for example) in the early 2.2 alpha
releases you got

>>> dir([1, 2, 3)]
[]
>>>

Many people complained about that.  It "was logical", since [1, 2, 3] is
indeed an instance of a class now, and doesn't have anything in its
"instance dict".  It *used* to give a listing of list methods (append, sort,
etc), because dir() dealt with instances of *types* in an entirely different
way.

With 2.2 the usefuless of that distinction vanished, and after public
discussion it was decided to make dir(instance_of_class) reveal much more
info than it used to (read the 2.2 dir.__doc__).  People who tried it in the
beta releases appeared to like it, and the complaints stopped regardless.

> (it seems like it will break a lot of code that is moderately
> introspective).

dir() is generally a poor building block for introspection, since what it
does (exactly) has always been unclear.  It's more principled in 2.2 than it
was, but since its primary *intended* use is still for convenience at an
interactive prompt, its behavior is still subject to what people claim is
convenient.  I don't know what you were using it for exactly, but chances
are good you should be looking at object.__dict__.keys() instead (which is
all dir() often did before 2.2).




From gcordova at hebmex.com  Wed Feb 20 18:51:37 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 17:51:37 -0600
Subject: No space left on device
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC8@SRM201002>

> 
> So, I am assuming there is some problem with the FAT32 system 
> when a directory gets lots of files in it.  Since I can make the 100K 
> files if I don't use mktemp, I am also assuming there is some 
> problem with mktemp in this scenario.
> 
> If anyone figures this out, it would be interesting to know what is 
> going on.  Meanwhile, I guess I'll need to redesign my file 
> generation algorithm to split things up.
> 
> thanks for the input.
> 
> sue
> 

The only problem I see is that FAT32 is outright
dead-ugly-inefficient, in a major way, when dealing
with long filename directory entries.

Normally, a directory entry contains the 8+3 filename,
creation / modification date, attributes, and that's all.

So, in order to create and use long filename entries,
there's this hideous hack going to behind the scenes.

A long-filename entry consists of enough directory
entries consecutive to each other, to write the filename
using unicode. Immediately after that, there's a "normal"
short-filename entry, containing it's respective short
filename.

Also, FAT32 doesn't have any kind of directory sorting,
so all filename searches are sequential.

So, there's lots of jumping around when looking for a
filename, looking in both the long-filename entries
and the short-filename ones.

Added to that, mktemp first looks to check if the filename
exists, before opening the file.

So, you have a stat() call, followed by an open() call.
The open() call may be faster than the stat(), because
of OS filename-caching and all that, but still, it's
slow and ugly.

Can you run your program in a linux machine using ext2fs
and reiserfs?? ext2 uses a sequential file lookup mechanism,
only that there's no ugly short- and long-filename entries.
On the other hand, reiserfs uses a b-tree to save it's
directory entries, so searching for files, or opening then,
is quite a bit faster.

Nice stuff, this. :-)

-gustavo



From mlh at vier.idi.ntnu.no  Tue Feb 12 12:20:04 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Tue, 12 Feb 2002 17:20:04 +0000 (UTC)
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <mailman.1013255897.29941.python-list@python.org> <VkV98.2452$yV3.182935@newsread2.prod.itd.earthlink.net>  <mailman.1013502171.782.python-list@python.org>
Message-ID: <slrna6ijm4.cik.mlh@vier.idi.ntnu.no>

In article <mailman.1013502171.782.python-list at python.org>, Mike C.
Fletcher wrote:
>Very nice module.  I just did a re-write of my watcher classes using 
>dispatcher.  Why did I keep the watcher pattern?  Well, it makes 
>creating pre-event processing possible (my one class needs to do some 
>processing to determine conflicts, and until 2.2 object-based-classes 
>(with properties) can live in ZODB, I'm stuck with that).
>
>Would be nice if there was a simple download somewhere (had to re-format 
>the code after Mozilla mangled it with cut-and-paste).  Should put an 
>explicit license in there (as ActiveState's legal policy says that the 
>license included with the contribution applies.
>
>Also looked at the References.py module in anygui.  Didn't spend much 
>time on it, as dispatcher is much closer to what I need.

Take a look at References.py and Events.py together -- Events.py is
the dispatcher part; References.py is just a set of utilities for
references that can be either weak or strong.

> Was one of 
>these modules based on the other (your comments on the ASPN site seem to 
>suggest there's some relationship)?

Yes, the dispatcher in Anygui is based on this and at least two other
similar dispatcher modules (but it is basically a full rewrite). There
was some lengthy discussion on the Anygui devel list about this sort
of dispatcher mechanism a while ago. (If you're interested, I could
probably find the other two modules as well :)

One advantage of the Anygui dispatcher is that it allows you to use
weak references for some handlers and strong references for others.
(Maybe Patrick has also added that now -- I'm not sure.)

>Good job all around. Thanks,
>Mike

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From donod at home.com  Thu Feb 21 00:47:38 2002
From: donod at home.com (Don O'Donnell)
Date: Thu, 21 Feb 2002 05:47:38 GMT
Subject: Adding static typing to Python
References: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com> <mailman.1014092911.31096.python-list@python.org> <gft37u8csm1ukd923nr933iq7meimqu5i7@4ax.com> <a4stbu$ptq$1@peabody.colorado.edu>
Message-ID: <3C748A4F.350BA6F7@home.com>

Fernando P?rez wrote:
> 
> Courageous wrote:
> 
> >
> > One of the things I like about C sharp is their use of attributes,
> > where you can define getters and setters on an attribute, but
> > otherwise the attribute is simply used as if it were a public one.
> > An objects attributes can thereby be used as a published aspect of
> > its interface safely, but no one will be the wiser if you manipulate
> > the getter and setter code.
> 
> Just curious, isn't there something quite similar to what you describe in
> py2.2? I've only glanced at it so I could be wrong...
> 


Yes, there is, and it's not new with C#.  I think something like
attributes with transparent getter and setter methods have been in VB at
least since version 6.

I am just starting to look at the Python implementation.  It is based on
a new built-in type called property.  In 2.2, at the interactive prompt,
type:

>>> print property.__doc__

And read Andrew Kuchling excellent "What's New in Python 2.2", Section
2.4 Attribute Access:

http://www.amk.ca/python/2.2/index.html#SECTION000340000000000000000

Have fun.

Don


From jeff at ccvcorp.com  Thu Feb 14 13:53:41 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 10:53:41 -0800
Subject: filtering random values (sieves?)
References: <5e8bd451.0202130827.2df6f50b@posting.google.com> <a4e5na$1efggq$1@ID-11957.news.dfncis.de> <5e8bd451.0202132254.5e90217b@posting.google.com>
Message-ID: <3C6C07B5.82AA1981@ccvcorp.com>

kevin parks wrote:

> But what i was hoping to do was generate two somewhat large lists
> algorhytmically and use one list to filter the other, which would be
> sort of  like getting the interesection, ....

You may not need this, but here's an easy way to get the intersection of
two lists... (presuming 2.x)

intersection = [x for x in list1 if x in list2]

If you need to work with 1.5.2 as well, then you could use filter/lambda.
I *think* that it would go like this:

intersection = filter(lambda x: x in list2, list1)

But I'm not certain enough about lambda's syntax to want to promise
anything about that.  ;)

Then once you have that, you can easily do the random.choice(intersection)
bit...

Jeff Shannon
Technician/Programmer
Credit International




From matthias_j at lycos.com  Fri Feb 15 01:47:29 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 14 Feb 2002 22:47:29 -0800
Subject: WNetCancelConnection2 -- How Can I cancel all open connection to a Server at once
Message-ID: <d7d5ebdf.0202142247.1f7d98e1@posting.google.com>

Hi to all of you,

Thanks for so much help so far.

My users must cancel all open connection to a remote Server.

I do not know in advance which connection they have made.

How can i force any (ALL) open connection to a specified machine to be canceld.

Thanks a lot

Matthias Janes


From maus at netz.klinik.uni-mainz.de  Tue Feb 26 09:10:25 2002
From: maus at netz.klinik.uni-mainz.de (Christian Maus)
Date: Tue, 26 Feb 2002 15:10:25 +0100
Subject: Socket stream
Message-ID: <a5g5a7$1mt$1@bambi.zdv.Uni-Mainz.DE>

Hello,

I try to control a device via a com server. Setting a command to the device 
works fine, but when I try to read the answer, my loop for reading from the 
socket does not terminate. My loop looks like this:

while 1:
       data = sockobj.recv(1024)
       if not data: break
       print data
sockobj.close()

I get the output:

OV
01,00OV02,00OV0
3,00OV04,00OV0
5,00OV06,01OV07
,00OV08,00OV09
,00OV10,00
OV11,00OV12,00
O
V13,00OV14,00
OV15,00
OA01,00
OA02,00OA03,00
OA04,00OA05,00
OA06,00OA07,00
OA08,00OA09,00OA10,00
OA
11,00OA12,00
O
A13,00OA14,
00OA15,00

and then I'm stuck in the loop. Any idea what the problem is?

Thanks in advance

Christian


From gcordova at hebmex.com  Wed Feb 13 19:08:17 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 13 Feb 2002 18:08:17 -0600
Subject: Container Objects Enforcing Contents
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A6E@SRM201002>

>
> On 13-Feb-2002 jim kraai wrote:
> > Greetings,
> > 
> > I'm looking for suggestions on making container objects which must,
> > may, or must not allow particular kinds of container objects to be
> > added.
> > 
> > Any suggestions on a general approach for this?
> 
> Let's see .......
> 
> a constraints approach:
> 
> [ ... snip ... ]
> 
> You could also try to enforce types by ensuring the type of 
> item matches types
> you know about.  This often leads to a maintenance nightmare though.
> 

I agree, and also I add that you'd be cutting off lots of flexibility
given by Python's OO capabilities.

Let's say that, for a Select() object, you want in it's constructor's
argument, a list of options. The options must be 2-tuples with
it's value and text. That sounds nice and fine, so you validate
that the constructor's argument must be a list, and each element
a 2-tuple.

BUT... what if I want to initialize the select from a database,
so I create an object that behaves "kinda" like a list, returning
as each element a recordset with two elements, containing the
value and text. Why? Because I want to select an ... apartment,
where allocations change daily, so I want to make this as
dynamic as possible:

	class KindaArrayFromDB:
	  def __init__(self, queryString):
	    dbc = GetConnectionFromSomewhere()
	    self.query = dbc.query(queryString)
	    # Everything's nice and dandy.

	  # I'm such a smartypants, I'll redefine __getitem__
	  def __getitem__(self, key):
	    # I'll just return the next recordset and ignore the key.
	    row = self.query.NextRecord()
	    if row is None:
		raise IndexError("Nay! I say: Nay!")
	    return (row["id"], row["descr"])

	  # Just so they say I play by the roolz.
	  def __del__(self):
	    self.query.close()

So here we have our kinda-array. Once the query's initialized,
you can comfortably do a

	for t in KindaArrayFromDB("SELECT ..."):
	  do_something_with_the_results(t)

Kinda neat. But if you're checking your arguments for
compliance with a certain type, your tying it to your
way of thinking how to program, and limiting it's
usefulness.

So... Let's act like adults, shall we?

I believe that programming is an INTENTIONAL process,
where, one does things intentionally. Since you're goint
to thoroughly document your code, anybody who wishes to
use it can read the documents and know how it works and
--more importantly-- how it DOESN'T work.

So, if someone want's to make his programs crash hideously
by using your routines in an incorrect manner, by all means,
let them.

But, if someone want's to be a smartypants and use your
code in creative ways you'd never thought of, don't tie
up your code so that it becomes inflexible.

That one of the things which makes python so fun and so
beatiful, it's practically limitless.

Have a nice evening!!!
-- 
Gustavo C?rdova Avila
*  gcordova at sismex.com <mailto:gcordova at sismex.com> 
* 8351-3861 | 8351-3862



From gcordova at hebmex.com  Fri Feb 22 12:59:14 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 11:59:14 -0600
Subject: is their any limit on the size of the data passed to a progra
	m via popen2?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE3@SRM201002>

> 
>   pdbdir = os.popen('ls /home/data/pdbs') # if this directory 
> contains 
> 10000 files, the code does not work.
> 
> files=pdbdir.readlines()
> pdbdir.close()
> (o,i) = popen2.popen2("/home/programs/profit ")
> 
> i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") 
> # This the 
> reference molecule.
> 
> for PDB in files:
>      i.write(" mobile /home/data/pdbs/" + PDB + "\n") # A new 
> molecule.
>      i.write("fit\n")		# compares reference and new molecules.
> 
> i.close()
> result= o.read()
> o.close()
> 

You read from the input pipe only upon writing all your data,
but maybe the pipe is full?

How about changing the loop with this:


log = StringIO.StringIO()

for PDB in files:
  i.write("mobile /home/data/pdbs/%s\n" % PDB)
  i.write("fit\n")
  log.write(o.read())

i.close()
o.close()


So now your results are in the log object.

-gus



From kalath at lycos.com  Thu Feb  7 06:21:52 2002
From: kalath at lycos.com (Mark Brady)
Date: 7 Feb 2002 03:21:52 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com>
Message-ID: <e840346c.0202070321.452d405d@posting.google.com>

aahz at panix.com (Aahz Maruch) wrote in message news:<a3sk2i$67m$1 at panix2.panix.com>...
> However, they don't AFAIK use Python for their core
> high-performance code, and it appears that this contest is slated for
> that role.
> 
> What does surprise me is that they think Java can do better than Python.

As python lover trapped developing in Java I see this a lot. Sun's
marketing has convinced a lot middle/upper management that Java/J2EE
is the only viable option to C/C++ for large server side systems.
Actually the endorsement and support from IBM/Oracle/HP/etc. for Java
is what seems to have given Java the 'respectability' it currently
enjoys as a C/C++ alternative. On a personal note Java/J2EE while
keeping me employed is driving me insane. Thank god for Pygame
(www.pygame.org). Java feeds the body but Python feeds the soul!

Mark.


From pan-news at asheeshenterprises.com  Thu Feb 28 21:14:16 2002
From: pan-news at asheeshenterprises.com (Asheesh Laroia)
Date: Fri, 01 Mar 2002 02:14:16 GMT
Subject: Regular expression problem
References: <mailman.1014877126.29732.python-list@python.org>
Message-ID: <YtBf8.154108$dh.39975567@typhoon.nyroc.rr.com>

Actually, I think this is the most elegant solution I've seen so far.

Good thinking; I forgot to "Use the Source," as some put it.

Only one problem: the parser still balks on embedded tags, like:

	<@Trap Body text:<P><I><B>>
becomes
	>

It leaves an extra '>' character at the end.  Any suggestions?  I can
write a simple workaround for something like this, but it seems like
it should work "the right way."

Thanks for everything!

-- Asheesh.

On Thu, 28 Feb 2002 01:17:11 -0500, Sean 'Shaleh' Perry wrote:


> On 28-Feb-2002 Asheesh Laroia wrote:
>> I've been trying to use sgmllib, actually, to delete all the other
>> tags.
>> 
>> It just doesn't handle the <@ [...] > condition well.  It refuses to
>> parse it, treating it as text.
>> 
>> 
> The reason is this:
> 
> starttagopen = re.compile('<[>a-zA-Z]') tagfind =
> re.compile(r'[a-zA-Z][-_.a-zA-Z0-9]*')
> 
> near the top of sgmllib.py.
> 
> Changing them in your code will allow the parser to understand the tag.
> However there is another problem which requires more work.  When a tag
> is found the parser tries to run 'start_' + tag.  start_ at Trap() is not a
> valid python name.  You could redefine the function which calls the
> handlers so that it looks for perhaps start_atTrap().  This would allow
> you to use the SGMLParser for all of your parsing needs, but may also be
> overkill for the problem.


From threeseas at earthlink.net  Thu Feb  7 23:27:42 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Fri, 08 Feb 2002 04:27:42 GMT
Subject: The Matrix of >Re: Autocoding evolves from........
References: <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>  <00b901c1af9e$9de1d080$0101010a@local> <mailman.1013103792.9086.python-list@python.org>
Message-ID: <2830.804T805T133451threeseas@earthlink.net>

On 07-Feb-02 12:40:33 Laura Creighton <lac at strakt.com> wrote:
>The only reason that Timothy Rue should leave is because we are absolutely
>of no use to him in the furthering of his project.  If absolutely everybody
>who has no intention of helping him had the discipline to shut up and
>stay shut, he could more accurately judge how helpful posting here is
>to his project.

>Voting people out of a newsgroup, as a concept, disgusts me.

>Laura Creighton

Laura, often value is found in what is not shown, what is absent.

there has been a great deal of very useful evidence generated here.
Evidence that shows programmers can be very illogical, irrational,
cleverly deceptive, etc..

Certainly everyone does understand in reading and responding to post here,
they actually make use of all nine action. It's physics, but they may
think it's me and that somehow their reactions will cause a dismissal
of.... well physics.

Lets see now:

AI - start and stop, change interfaces - Uh start up newsreader..

PK - keep track of where you are - Pick up where you left off on the thread..

OI - get input - read with eyes

IP - input from - internet and monitor

OP - output to - keyboard usenet postings

SF - one step at a time - damn this non-polyphony qwerty keyboard...

IQ - what's the meaning - whats a good non-target meaning I can attach to
                     confuse tim...

ID - identify posters - hey there is one by tim, now I know to be negative..

KE - constraints to apply to meanings and poster named tim rue - must be
        negative and anti-productive


And to think, some think I need to prove these nine action constants.

On the other hand, self observation is a SOB due to inherent subjectivity.


One of the best pieces of evidence, least one I tend to like because it
really suggests the real existance of a matrix in some peoples minds, is
the 3 dimensional reality vs. the 3 deminsional data array...

Truely the Matrix has you, if you can't tell the difference..

Switch = AI

Apoc = PK

Tank = OI

Mouse = IP

Dozer = OP

Neo = SF

Morpheus = IQ

Trinity = ID

Cypher = KE


And the three agents are input, processing and output.

Agent Smith is of course processing....


Since what I've been on about can be made into a movie and become as
popular as it has been.........and due to so many seeing similiar
generalities but different specifics for them....

Nobody is convincing me of any BS, even before the movie was written.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From bosahv at netscapenospam.net  Wed Feb  6 22:34:55 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Thu, 07 Feb 2002 03:34:55 GMT
Subject: Delphi vs Python
Message-ID: <zBm88.333397$pa1.80603034@news3.rdc1.on.home.com>

I'm planning learning strategy and I thought I would ask people who use both
Python and Delphi their opinion.

I've determined that I at least cannot speedily learn both at once. So my
current strategy is to get up to speed in Python first and then maybe do the
slowest bits of code in Delphi.

Since Kylix my head spinning with options but I think Python as a learning
language (This is my first computer language) is turning the tide.

Has anyone used Delphi to make python called DLL's ?

Has anyone seen any Delphi 6 personal edtn. stuff show up to serve python?


Any opinion would be appreciated.

Bo




From lee at leeHYPHENphillips.org.invalid  Tue Feb 12 01:43:07 2002
From: lee at leeHYPHENphillips.org.invalid (Lee Phillips)
Date: Tue, 12 Feb 2002 01:43:07 -0500
Subject: Zope cookbook?
References: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au>
Message-ID: <120220020143076768%lee@leeHYPHENphillips.org.invalid>

You might be happier with webware, or one of the other frameworks
(skunkweb, ...,).


From replogle2 at cox.net  Sat Feb  2 20:57:26 2002
From: replogle2 at cox.net (Andrew Replogle)
Date: Sun, 03 Feb 2002 01:57:26 GMT
Subject: reading ftp log file..
Message-ID: <aO078.9013$Xk4.483475@news1.east.cox.net>

I'm trying to write a python script that watches an ftp log file.

Based on someone logging in and successfully transmitting a file it will use
the smptlib to generate somewhat of a dynamic mail as an "upload
confirmation".  But I'm so new to python and programming in general that
this is proving very difficult.

this is what I have so far..
-----------------------------------
import os
import sys
import smtplib

ServerPath = 'c:\\Program Files\\G6 FTP Server\\'
logExists = os.path.isfile(ServerPath + 'FTP.log')
line = "230"
#print logExists

log = open(ServerPath + 'FTP.log')
lines = log.read()
for line in lines:
 print "someone logged in"
 break
-------------------------------------

the "230" you see is the code the ftp server puts in the log if X user
successfully logs in.. I'm trying to search for that in the file and if it
exists then go on with what the account is & what file they transfered.

it always prints "someone logges in" even if I change 'line'  to a bazillion
zeros..

please let me know where I'm going wrong or what I'm misunderstanding. any
examples along with explinations are greatly appreciated..

thanks.

Andrew




From grante at visi.com  Sun Feb  3 23:27:19 2002
From: grante at visi.com (Grant Edwards)
Date: Mon, 04 Feb 2002 04:27:19 GMT
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org>
Message-ID: <slrna5s3d6.h88.grante@isis.visi.com>

In article <a3knt6$gr3$1 at news.efn.org>, Spencer Doidge wrote:

> I really tried to RTM first this time, but I failed.
> Can someone give me a hint where in the doc html's it explains how to pass
> by reference?

Everything in Python is passed by reference.  Always.

However, some of those "things" can't be changed, so it may or may not be
possible to do what you want.  Strings can't be changed, so strictly
speaking it's not possible to do what you ask below.  Lists can be changed,
so you could modify them in-place as shown in your C code.  Perhaps you
could explain what it is you're trying to accomplish and we could explain
how it is typically done with Python?

> Specifically, I want to do the equivalent of this:
> 
> char s[10];
> 
> void foo(char *bfr)
> {
> bfr[0] = 'a';
> bfr[1] = 'b';
> return;
> }
> 
> void main(void)
> {
> foo(s);
> return;
> }



-- 
Grant Edwards                   grante             Yow!  People humiliating
                                  at               a salami!
                               visi.com            


From jschmitt at ati.com  Thu Feb  7 01:20:15 2002
From: jschmitt at ati.com (John Schmitt)
Date: Wed, 6 Feb 2002 22:20:15 -0800 
Subject: scripted install of Python-2.2 and win32all?
Message-ID: <C17613BF4690D211910C0008C749124204CFE636@ca00exh01.ca.atitech.com>

Hopefully you can get an answer that I can use too.  I wanted to do this too
so I tried this at the Windows command prompt:

>Python-2.2.exe /?

which immediately started the install.  The ActivePython installer did only
a little better:

>ActivePython-2.1.1.msi /?

which immediately pops up an alert warning me that my command line
parameters were incorrect.  No hints found about the installer.

Some InstallShield type installers throw up a dialog box with their
installer options when prompted.  

What does seem to do something is this:

>Python-2.2.exe /s

Don't know if that'll get you the options you want though.

John

-----Original Message-----
From: dsloyer at ieee.org [mailto:dsloyer at ieee.org]
Sent: Wednesday, February 06, 2002 9:53 PM
To: python-list at python.org
Subject: scripted install of Python-2.2 and win32all?


Is it possible for one to install Python-2.2 and win32all without any
user intervention?  I'd like to install these on a set of machines as
part of a script, and would like to do this automatically.  The
installation would take place on Windows XP and Win2K boxes.

I might like to select/de-select a couple of the check boxes, but
otherwise intend to accept the default values throughout.

Thanks!
-- 
http://mail.python.org/mailman/listinfo/python-list
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020206/19a3df3d/attachment.html>

From rony.steelandt at bucodi.com  Sat Feb 16 11:57:15 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 16 Feb 2002 08:57:15 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org>
Message-ID: <b34d3835.0202160857.4111f0be@posting.google.com>

You are definitifly right, i get the same error with 2.1.2 and 1.5.2
Actually i tested it with C++ , VB, Windev and it works.
It only doesn't work with Python.
I think this could be a question for our COM expert Mark Hammond
Can you keep me posted on this,i could use the control too

Thanks

Rony


"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1013487810.369.python-list at python.org>...
> I am developing a headache messing with an OCX I picked up from the
> Internet.  It seems that my forays into ActiveX never pan out; I should
> stick with Linux.
> 
> The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> have to be such a pain to find?)  I can instantiate it:
> 
> >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> >>>
> 
> (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
> 
> >>> o.AboutBox()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py", line
> 29, in AboutBox
>     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
> com_error: (-2147418113, 'Unexpected failure', None, None)
> >>>
> 
> Any method or property, same answer.
> 
> The URL for the control is:
> 
>     http://www.qwerks.com/download/3693/report.zip
> 
> It is supposed to be a general printing control, but nothing works.
> 
> Chris Gonnerman -- chris.gonnerman at newcenturycomputers.net
> http://newcenturycomputers.net


From gjlap at yahoo.com  Sat Feb 23 18:04:21 2002
From: gjlap at yahoo.com (Gil)
Date: 23 Feb 2002 15:04:21 -0800
Subject: pygame puzzler
References: <1eed326b.0202222300.4666822c@posting.google.com> <3C77DDD9.8020507@attbi.com>
Message-ID: <1eed326b.0202231504.407fe39d@posting.google.com>

Pete Shinners <shredwheat at attbi.com> wrote in message news:<3C77DDD9.8020507 at attbi.com>...
> the mixer module in pygame is also optional. that means pygame can build 
> and install without it. pygame comes with a "config.py" script, and if 
> it cannot find the SDL_mixer library, it disables the pygame.mixer 
> module. the first thing to check is the "Setup" file. this is a simple 
> text file created by the config, it stores the compile settings. open 
> the file and first check to see if there are compile flags for the 
> "MIXER" variable. once that seems reasonable, scroll down to the next 
> section which lists the optional modules. make sure the line for "mixer" 
> is not commented out. (uncomment "mixer_music" too)

Setup file OK

[caltrop at localhost pygame-1.4]$ locate mixer.so
/usr/local/lib/libSDL_mixer.so/

> if the Setup file is now looking correct. rerun the "setup.py" which 
> will get your installation up to date. as long as it doesn't error out, 
> i'm certain the mixer module will be installed. if the setup.py runs 

copying build/lib.linux-i586-2.2/pygame/mixer.so ->
/usr/local/lib/python2.2/site-packages/pygame

> cleanly and it still doesn't work, i'm guessing it might have something 
> to do with mismatched python versions?

[caltrop at localhost examples]$ python sound.py
Traceback (most recent call last):
  File "sound.py", line 10, in ?
    import pygame.mixer, pygame.time
ImportError: libSDL_mixer-1.2.so.0: cannot load shared object file: No
such file or directory

Square 1. 

re: versions of python, in #pygame I mentioned that v1.5 went in with
my original install of RedHat7.1, and then I added v2.2 and you(I
think it was you as ShredWheat) told me how to link the newer version
so now I get:

[caltrop at localhost examples]$ python
Python 2.2 (#1, Feb  7 2002, 20:05:27)
[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-81)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>>

I had a similar problem with the font module(ImportError...no such
file) until I moved libSDL_ttf-2.0.so.0 to /usr/local/lib, but this
time libSDL_mixer-1.2.so.0 doesn't have that problem.

Any other suggestions?

-Gil(caltrop)


From emile at fenx.com  Wed Feb 27 14:20:12 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 11:20:12 -0800
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com>
Message-ID: <a5jbra$7ok00$1@ID-11957.news.dfncis.de>

"Michael P. Soulier" <msoulier at nortelnetworks.com_.nospam> wrote in
message news:slrna7q9m0.4pb.msoulier at pmerd071.ca.nortel.com...
>     Hello.
>
>     While I know about bound methods in Python, I'm wondering if a
full
> closure can be implemented. Can it?
>
>     Example: I want to pass an arbitrary code reference into a closure
and
> have all of its variables preserved.
>
>     ie.
>
>     print "the current count is %d" % counter; counter += 1
>
>     Can I pass the above code into a closure, and then simply call it
> repeatedly and see counter increment itself? If I pass in a function
with a
> local variable, that local will be reset each time, and I don't want
to use a
> global.
>

I don't understand closures, but I might try something like:

class Counter:
    def __init__(self):
        self.count = 0
    def __int__(self):
        self.count += 1
        return self.count

counter = Counter()
print "the current count is %d" % counter

Hopefully, someone will post something pythonish illustrating closures.
Maybe this time I'll get it.  ;-)

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From savithari at mindspring.com  Mon Feb  4 22:52:49 2002
From: savithari at mindspring.com (Tarle)
Date: Mon, 04 Feb 2002 22:52:49 -0500
Subject: python and windows short file names
Message-ID: <3C5F5711.990519D@mindspring.com>

Hello All,
                I think this question must have been asked a lot of
times before.  I could not find in the searches I did.

I want to be able to use java to python ( thru jython is fine) to be
able to manipulate the full file name and the long dir structures etc.,

I would like to be able short file names and other useful file i/o's
including ability to delete directory, create directory, rename, move
etc., etc.,

How can I accomplish this without COMMERCIAL SOFTWARE ?

Thanks for your help.

-Narahari.



From tdelaney at avaya.com  Mon Feb 11 18:26:08 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 12 Feb 2002 10:26:08 +1100
Subject: Perhaps I am just dumb
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A177@natasha.auslabs.avaya.com>

> From: Steve Holden [mailto:sholden at holdenweb.com]
> "wooks" <wookiz at hotmail.com> wrote in message
> news:8852d33c.0202110117.374f062 at posting.google.com...
> [Quinn explains polymorphism]
> > ... if I get magic methods right, they are special functions which
> > if present enable the object to apply a unique interpretation of a
> > python operator.
> > So + has standard behaviour but will  adopt whatever is specified in
> > the __add__ function if one is present for the class.
> >
> > Yes?
> 
> Correct

However, also remember that there are *3* add methods ...

__add__
__radd__
__iadd__

I'll leave it up to the documentation to explain the difference :)

Tim Delaney



From robert_kuzelj at yahoo.com  Fri Feb 22 08:17:04 2002
From: robert_kuzelj at yahoo.com (Robert Kuzelj)
Date: 22 Feb 2002 05:17:04 -0800
Subject: pythonisten stammtisch in muenchen - step 2
Message-ID: <eaef2e43.0202220517.7cc555ed@posting.google.com>

ok,

also bisher sinds vier anmeldungen (inklusive
meiner person). das war meine mindestanzahl,
obwohl ich auf mehr gehofft hatte.

also denn, zum zweiten schritt:
wahl des moeglichen wochentages.

die wahl sollte dabei zweigeteilt sein:
a) bevorzugter wochentag
b) wochentag an dem es gar nicht geht

im dritten schritt koennen wir uns dann auf
einen ganz konkreten ersten termin und ort einigen.

ciao robertj


From nospam at bigfoot.com  Wed Feb  6 10:37:30 2002
From: nospam at bigfoot.com (Gillou)
Date: Wed, 6 Feb 2002 16:37:30 +0100
Subject: REQ : encoding windows cp1252 => iso latin 1
References: <a3p68d$1a56$1@norfair.nerim.net> <Pine.SOL.4.30.0202051904310.26540-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <a3ri6f$t2f$1@norfair.nerim.net>

Hi,

Finally, I made my own translator un 100% python.
I suspected pure python to be too slow to convert the amount of text in a
Web server, but it does the job correctly and pretty fast

Regards.

--Gilles

> I realise this is overkill, but:
>
> There is a simple Python wrapper of recode which I have a setup.py for and
> which compiles on win32 with mingw... but though it seems quite stable on
> unix, it crashes on windows, and I haven't got debugging working yet on my
> machine.
>
> So, if you feel like fixing it, you're welcome to it.
>
>
> John
>




From christophe.delord at free.fr  Wed Feb 20 17:33:03 2002
From: christophe.delord at free.fr (Christophe Delord)
Date: Wed, 20 Feb 2002 23:33:03 +0100
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de>
Message-ID: <3C74241F.3060403@free.fr>

Gerson Kurz wrote:

> On Wed, 20 Feb 2002 21:48:52 +0100, Christophe Delord
> <christophe.delord at free.fr> wrote:
> 
> 
>>Writting a function to sort lists may not be efficient :
>>
>>def my_sort(L):
>>    L2 = L[:]
>>    L2.sort()
>>    return L2
>>
>>X = [...]
>>for x in my_sort(X):
>>    ....
>>
> 
> Of course, you can always use quicksort:
> 
> q=lambda x,o=filter,r=lambda o:(lambda x=list(o)+[0],\
> y=len(o)-1:(lambda:map(lambda i,s=x.__setitem__:s(-1,\
> x[i])or s(i,x[y-i])or s(y-i,x[-1]),range((y+1)/2))and\
> 0)() or (lambda:"".join(x[:-1]))())(),s=')]1=<)x(nel'\
> '[]"x",))"o(lambda i,j=x[0]:i"(r,"~"(ecalper.)"q(~<j'\
> ',x))+~==j,x)+q(~>j,x))"(r[(lave':eval(r(s))
> 
> import random
> test = range(20)
> random.shuffle(test)
> print test, "->", q(test)
> 
> 

Not obvious at the first look (lots of eval and reverse) but very funny. 
I ran it by hand and wrote an easier to read qsort :

q=lambda x:(lambda o=lambda s:filter(lambda i:cmp(i,x[0])==s,x):len(x)>1 
and q(o(-1))+o(0)+q(o(1)) or x)()

It uses nested scopes so it needs python 2.2.
For previous versions it may be :

q=lambda x:(lambda o=lambda s,x=x:filter(lambda 
i,x=x:cmp(i,x[0])==s,x):len(x)>1 and q(o(-1))+o(0)+q(o(1)) or x)()

--
Christophe Delord
http://christophe.delord.free.fr



From mpnews at maksoudplaza.com.br  Mon Feb 11 09:52:31 2002
From: mpnews at maksoudplaza.com.br (MPNEWS)
Date: Mon, 11 Feb 2002 11:52:31 -0300
Subject: Vacation Package Promotion for São Paulo - Brazil
Message-ID: <UTC200202111353.OAA01942@bacchus.cwi.nl>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020211/9c1f6a50/attachment.html>

From dsavitsk at e-coli.net  Mon Feb 11 00:14:03 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 05:14:03 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com> <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com>
Message-ID: <vqI98.3303$pC1.669077@newssrv26.news.prodigy.com>

i didn't explain that well at all.  the point that i forgot to mention is
that i want to compare each hex'd number to '00' ... there remains the
problem of comparing  from the left instead of from the right,  maybe i
should have though about this a little more before i posted.

-d


"dsavitsk" <dsavitsk at e-coli.net> wrote in message
news:%hI98.3302$pC1.668524 at newssrv26.news.prodigy.com...
> the truncate part is the problem.  I am building VB Hex strings to send to
a
> VB server.  so
>
> >>> red = hex(56)[2:4]
> >>> green = hex(156)[2:4]
> >>> blue = hex(12)[2:4]
> >>> thestring = '&H' + blue + green + red
>
> but here, since the blue string is only 1 character it throws the rest
off.
>
> -d
>
> "Joshua Muskovitz" <joshm at taconic.net> wrote in message
> news:3c674cbd_2 at corp.newsgroups.com...
> > Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> > >>> ''.join(map(max,zip('00','AA')))
> > 'AA'
> > >>> ''.join(map(max,zip('00','A')))
> > 'A'
> > >>> ''.join(map(max,zip('00','')))
> > ''
> >
> > Damn, I thought I had it.  :-)  Is there an variant of zip which doesn't
> > truncate?  BTW, this does work, provided that the input strings are the
> same
> > length, and even allows for more than two strings!
> >
> > >>> s = ['abte','scrt','sdfw','cxvs']
> > >>> ''.join(map(max,zip(*s)))
> > 'sxvw'
> >
> > --
> > # Joshua Muskovitz
> > # joshm at taconic.net
> > def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
> > lyyrs('Hire me!  I need the work!')
> >
> >
> >
> >
> > -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> > http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> > -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----
>
>




From emile at fenx.com  Wed Feb 27 11:31:40 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 08:31:40 -0800
Subject: Profiling memory usage?
References: <a5iogb$laa@ascc.artsci.wustl.edu>
Message-ID: <a5j1vb$7mq9u$1@ID-11957.news.dfncis.de>

"Michael James Barber" <mjbarber at ascc.artsci.wustl.edu> wrote in message
news:a5iogb$laa at ascc.artsci.wustl.edu...
> I'm currently looking to optimize a couple of my python programs in
terms
> of memory usage.  I think I know what I need to improve, but I suspect
> that "profile before optimizing" applies to memory usage as well as to
> speed.
>
> Is there something analogous to the profile module for describing how
much
> memory is used by e.g. instances of different classes?  I didn't find
> anything using Google or in the Vaults of Parnassus.
>

I also looked for a tool and found none.  One of my projects requires a
large (1+Gb) memory footprint.  In working with a test data set (~10% of
full), I needed to balance speed vs memory.  I started with my test
requiring 272Mb and running in 2.984 seconds.  Through progressive
stages, I managed to take it down to 105Mb and 2.384 seconds.  Since
there weren't good internal profiling tools, I looked at memory usage
using pstat.exe on windows, but you could probably use top in a similar
kludgey output parsing way.  (Now someone can chime in with the *easy*
way to do this ;-)  I then added debugging output that reported memory
usage in what felt like memory consuming areas, checking befores and
afters, refactoring and restructuring, etc, running the test in between.
After six iterations or so I was at 150Mb and 2.73 seconds, and I got to
the final numbers five additional steps later.  The whole process took
about a day.

The kinds of changes I made included:
  -- not holding on to references to large structures when I was done
with them (doh)
  -- separating the values and keys previously held in dicts by
switching the values to point to indexed Numeric structures
  -- switching to a dual pass process, building the structures outside
the run environment, then reading it in later to save having the source
and prepared data in the same process.  With the full data set, the prep
process spikes usage to 1.25Gb, while the run process uses only 850Mb.
  -- forcing commonly used string parts of keys to be interned
  -- writing custom disk storage routines as opposed to using cPickle or
marshall (for both speed and memory footprint considerations)
  -- rewriting certain areas to allow for "missing" data so it didn't
need to pre-allocate memory

This is certainly not the kind of fine-grained introspection I'd like to
have access to, but it certainly worked when I found myself holding the
proverbial "5-pound bag".  ;-)

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From tdelaney at avaya.com  Thu Feb 28 18:21:36 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 1 Mar 2002 10:21:36 +1100 
Subject: Status of PEP's?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1C5@natasha.auslabs.avaya.com>

> I'd also like to put in my (unrequested) two cents and go on record as
> disliking the "for i in 5" syntax. Not just as a matter of taste, but
> because it provides unintuitive behavior, as pointed out previously.
> The two lines...
>     for i in 5: print i
>     for i in 5, 6: print i
> ...would behave differently under the proposal.

They behave differently now. One is a syntax error, the other isn't.

For the record (I've said this before), I like the idea of integers being
iterable (and I'm not a set theory wonk). I never think of it though as

for i in 5:
    print i

but always as

for i in len(seq):
    print i

or something similar, which has obvious semantic logic. However, I could see
myself happily using for i in 5: in a quick script.

One important thing to remember, which I think a few people have missed, is
that an integer will not be an iterator. An integer will be iterable - that
is, to get this behaviour, iter(integer) must be called at some stage. This
is done transparently in a for: loop. Are the following two equivalent
though?

if 1 in obj:
    pass

if 1 in iter(obj):
    pass

i.e. is iter(obj) called implicitly in the first case?

Tim Delaney



From owen at astrono.junkwashington.emu  Mon Feb  4 20:31:24 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Mon, 04 Feb 2002 17:31:24 -0800
Subject: question about subclassing dict
References: <mailman.1012623183.14665.python-list@python.org>
Message-ID: <a3ncld$a36$1@nntp6.u.washington.edu>

In article <mailman.1012623183.14665.python-list at python.org>,
 "Jason Orendorff" <jason at jorendorff.com> wrote:

>Russell E. Owen wrote:
>> I started by defining __iter__. I was then pleased to find that the 
>> methods iterkeys, itervalues and iteritems all worked just fine.
>
>They don't seem to return the objects in the right order (which
>I gather is the whole point):

Thanks for the heads-up. I must have done the testing manually (I 
certainly did the testing somewhere), but probably on a list of only 
three elements, which is hardly a definitive test. I was planning
to override the other methods after asking here if it was necessary.

I've added the missing methods for the iter... methods using your nice 
generator code. I hadn't realized it was quite that trivial. (I first 
implementing iterator classes, but this was just messy enough that I 
decided to qeury the newsgroup before actually using them, hence my 
original posting.)


Laura Creighton pointed out that another ordered dictionary is available 
from <http://newcenturycomputers.net/projects/rbtree.html>. It is based 
on a tree structure and I suspect scales better to large dictionaries. 
In any case, it is well worth a look. Based on it I realized I was 
missing the copy method.


I've put the revised version (including iter..., copy and __str__ 
methods and improved testing code) up as OrderedDict.py at 
<http://www.astro.washington.edu/owen/> in case anybody is interested. 

Thank you both very much for the helpful replies!

-- Russell


From jason at jorendorff.com  Tue Feb 19 23:02:06 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 19 Feb 2002 22:02:06 -0600
Subject: Editable strings anyone?
In-Reply-To: <3C72C074.306@wi.mit.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLKEOADDAA.jason@jorendorff.com>

Brian Kelley wrote:
> What would it take to make array.array or array.ArrayType an acceptable 
> base type for subclassing?

Wow, you're right, this is a mess.  I will put together a patch
for it.  Quite a few little things are involved, but it's mostly
just a modernization of the way this module uses the Python C API.

That's not the kind of thing that's likely to be really
well-documented anywhere, unfortunately.  :P

## Jason Orendorff    http://www.jorendorff.com/




From philh at comuno.freeserve.co.uk  Sat Feb  9 19:31:40 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sun, 10 Feb 2002 00:31:40 +0000
Subject: RFC -- Hygienic Macros in Python
References: <mailman.1013231420.29837.python-list@python.org>
Message-ID: <slrna6bfrc.o7n.philh@comuno.freeserve.co.uk>

On Fri, 8 Feb 2002 23:04:57 -0600, Jason Orendorff <jason at jorendorff.com> wrote:
>phil hunt wrote:
>> Python macros should, if they exist, have a superset of the 
>> capabilities of
>> C macros. Including the ability to be replaced by arbitrary code in the
>> program. Why arbitrary? firstly, because it is conceptually simple;
>> secondly because it doesn't restrict the programmer.
>
>C macros are quite restrictive, precisely *because* they simply
>paste tokens into your program.
>
>> Consider this code:
>>
>> defmacro incj:
>>    j = j + 1
>
>Well, I don't think that's particularly realistic code. 

Nor do I, particularly. I can't offhand think of anything where I'd
particularly want to use macros in Python, except perhaps to define my
own control structures.

But just cos i can't think of it, doesn't mean someone else won't
find it useful.

> Consider
>this instead.
>
>Jason Orendorff wrote:
>>   defmacro synchronized(lock_expr, code_block):
>>       lock = ^lock_expr
>>       lock.acquire()
>>       try:
>>           ^code_block
>>       finally:
>>           lock.release()
>
>Under a non-hygienic macro system, obviously I wouldn't write it
>like that because "lock" might clash with a local name.  Indeed if
>I have nested "synchronized" statements it's guaranteed to clash.
>So what should the code look like in a non-hygienic system?

The way i would do it, if I was implementing it, would be something
like:

defmacro synchronized(lock_expr, code_block):
   local lock
   lock = ^lock_expr
   lock.acquire()
   try:
      ^code_block
   finally:
      lock.release()      

Where the "local lock" declaration means something like: in the
current namespace create a local variable name that doesn't clash with
anything else.

In something like C++ you could just put {...} round it can define
variables within that and they'd be local to it, but you can't do 
that in Python.



-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jkraai at paypal.com  Wed Feb 20 17:43:03 2002
From: jkraai at paypal.com (Jim Kraai)
Date: Wed, 20 Feb 2002 14:43:03 -0800
Subject: Any Interest in EDIFACT Parser/Generator?
Message-ID: <7A9A4362115F384095113F27CBF8BC43F09940@mv1-mail2>

Just curious.

--jim



From phil.rittenhouse at dspfactory.com  Tue Feb 26 17:48:07 2002
From: phil.rittenhouse at dspfactory.com (Phil Rittenhouse)
Date: Tue, 26 Feb 2002 17:48:07 -0500
Subject: Converting C strings to Python problem
References: <BwPe8.21867$X2.226876@nnrp1.uunet.ca>
Message-ID: <GcUe8.21985$X2.229250@nnrp1.uunet.ca>

Hmmm.....

Looks like I missread the doc's.  Only UNKNOWN escape sequences are
represented
by two characters.

My bad.

Phil


"Phil Rittenhouse" <phil.rittenhouse at dspfactory.com> wrote in message
news:BwPe8.21867$X2.226876 at nnrp1.uunet.ca...
> Hi,
>
> I'm trying to pass a path string (eg. "c:\spam\test.exe") from C to
python,
> but python is interpreting the string as "c:\spam<tab>est.exe".
>
> I've run into this problem with converting C strings to Python strings
using
> Py_BuildValue().   When converting a C string with backslashes to a
> python string, the string (including backslashes) is copied exactly.
Sounds
> reasonable, except that, according to the doc's, Python
> uses two character escape sequences to represent things like tabs (\t).
>
> Should the conversion process convert backslashes in C to double
backslashes
> in Python?   What is the recommended method for doing the C string to
python
> string conversion?  Does PyString_FromString() behave differently?
>
> Thanks for your help,
> Phil
> ------------
> Phil Rittenhouse
> phil.rittenhouse at dspfactory.com
> Software Developer
> Dspfactory Inc.
>
>
>
>
>
>
>




From martin.franklin at westerngeco.com  Fri Feb  1 08:52:01 2002
From: martin.franklin at westerngeco.com (Martin Franklin)
Date: Fri, 01 Feb 2002 13:52:01 +0000
Subject: Tkinter widgets: Write options *and* option tags dynamically from 
 dictionary?
References: <3c5a84eb.70228013@news.online.de>
Message-ID: <3C5A9D81.3E99C36F@westerngeco.com>

Johannes Eble wrote:
> 
> Hello all,
> 
> In Tkinter the widgets are created in the form:
> 
> wgt = Widget(master, option_tag1 = option1,  option_tag2 =
> option2,...)
> 
> I want to create widgets dynamically. Therefore I have created a
> dictionary for the options (well, it is a list of lists of lists which
> have this dictionary included, but I will keep it simple here). The
> problem is that not only the options are unknown at compile time, but
> also the *option tags*. For example in Button(root, text='Hello') also
> the *text* option tag should be taken from the keys of the options
> dictionary.
> 
> This is hardcoded:
> 
> from Tkinter import *
> root = Tk()
> bt = Button(root, text='Hello')
> bt.pack(side=TOP)
> root.mainloop()
> 
> When I change it to something like
> 
> ...
> dictOptions = {text: 'Hello',....}
> bt.config(?)
> ...
> 
> ,Python complains that there is no defined symbol text. Also, I
> wouldn't know what to write in bt.config(....)
> 
> On the other hand, I have tried to write
> 
> dictOptions={'text':'Hello'} and compose the string "text='Hello'" on
> the fly, but then I built a button with the text "text='Hello'".
> 
> So, how can I change the options and the option tags dynamically?
> 
> Any help would be appreciated.
> 
> ByeBye
> 
> Johannes

This works......


from Tkinter import *
d={'text' : 'hello'}
r=Tk()
b=Button(r, d)
b.pack()
d={'text' : 'world'}
b.config(d)
r.mainloop()



HTH 

Martin.


From jjl at pobox.com  Mon Feb 18 15:34:07 2002
From: jjl at pobox.com (John J. Lee)
Date: Mon, 18 Feb 2002 20:34:07 +0000
Subject: Python calling Java
In-Reply-To: <a4rjq4$4n8$1@web1.cup.hp.com>
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <Pine.SOL.4.30.0202182029410.13673-100000@mimosa.csv.warwick.ac.uk>

On Mon, 18 Feb 2002, jimh wrote:

> We have a bunch of Python (cgi) scripts producing html behind an Apache web
> server.  We also have a set of Java libraries that I want to access from
> these Python scripts, but I haven't been able to figure out how to do this.
> I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
> a one-line 'print "hi"' script, and this is, of course, not acceptable for a
> web server.
>
> Any ideas?

Fix your Jython installation?!  Seriously, what is the bottleneck here?
Clearly not the execution of the code 'print "hi"'.  I presume you are
talking about entirely server-side execution -- or not?

Google points to JPE, might be worth a look.


John



From phr-n2002a at nightsong.com  Mon Feb  4 12:13:14 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 09:13:14 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org>
Message-ID: <7xu1sx41cl.fsf@ruckus.brouhaha.com>

Justin Sheehy <justin at iago.org> writes:
> You don't seem to understand what "backward-compatible" means.  What
> you are talking about is more related to forward compatibility, and is
> generally far less important than backward compatibility.
>
> Backward compatibility is broken when code that does work in an older
> version does not work in a newer version.

The claim was that "from __future__ import division" and using //
stops your old code from breaking when 3.0 comes out, which fits your
definition of backward compatibility.  I pointed out that it only
works for 2.2 and later.  It doesn't provide backwards compatibility
to the versions of Python that most people actually use today.

> What concrete example do you have of your own working code that was
> correct and working, but broke in 2.2?

I haven't used 2.2 yet.  I do hit some snags trying to run 1.5 scripts
under 2.1.  They're not my scripts, but that makes it even worse.
If it's my own code, I can usually fix those kinds of problems very
easily.  If it's someone else's code, I have to study it and figure
out what it's doing before I can fix it, and still not be sure my fix
is correct.

> (Note that the "correct" bit there rules out socket-argument problems,
> since the code that broke in those cases was not really conforming to
> the admittedly misleading docs.)

Another culprit is the rand module disappeared.  It looks like I can
compatibly replace it with "random", but I'm not sure if there's some
subtle difference that breaks something behind the scenes.

> 2.2 is the only current series, exactly because the Python dev team
> says so.  Older versions may receive bugfix releases, but that does
> not make them current.

Excuse me, the Python team can decide things about the Python
language, but it doesn't get to decide about the English language.
"Current" means what users think it means, not what the Python team
thinks it means.  If Ford is still selling new Mustangs and releases
improved models of it every so often, then the Mustang is a current
model according to anyone who uses the word in the normal way.

> > More importantly, in the real world, a lot of people are still using
> > 2.1 and 1.5, and few are using 2.2.  So coding anything in a way that
> > requires 2.2 inconveniences a lot of users. 
> 
> So, if you don't want to inconvenience those users, don't make your
> code require 2.2.  This seems fairly obvious and easy.

But that means I can't say "from __future__ import division", since
only 2.2 and later support that.  And if I don't say that, then my
scripts will break in 3.0.  How many times do I have to explain this?

> > That might be worth it for serious features like metaclasses or
> > generators, but is silly to do over this division nonsense
> 
> What division nonsense?
> 
> Show me the division-using code that works pre-2.2 but breaks in 2.2.

The breakage will happen in 3.0, not 2.2.


From cliechti at gmx.net  Mon Feb 18 19:33:52 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 19 Feb 2002 01:33:52 +0100
Subject: confusion regarding os.path.walk()
References: <20020218180623.086f6c58.ron.l.johnson@cox.net>
Message-ID: <Xns91BA1077BDD93cliechtigmxnet@62.2.16.82>

Ron Johnson <ron.l.johnson at cox.net> wrote in 
news:20020218180623.086f6c58.ron.l.johnson at cox.net:
> In the visit callout function, it looks as though after the top
> level is processed, os.path.isfile() and os.path.isdir() both return 
> 0 when passed elements from the names parameter.
...
>     for n in names:
>    	    	print n,
>    	    	if os.path.isdir(n):
>          ...

in "names" there are only filenames with no path. to access the files 
you must join the path back to the name:

for n in names:
    	file = os.path.join(path, n)
    	print file,
    	if os.path.isdir(file): 
    	...

-- 
Chris <cliechti at gmx.net>



From michael.amrhein at t-online.de  Thu Feb  7 13:20:31 2002
From: michael.amrhein at t-online.de (Michael Amrhein)
Date: Thu, 7 Feb 2002 19:20:31 +0100
Subject: FixedPoint
References: <mailman.1013070672.23308.python-list@python.org> <Dhx88.95486$KM2.3465594@atlpnn01.usenetserver.com>
Message-ID: <a3ugfs$sto$07$1@news.t-online.com>

"Steve Holden" <sholden at holdenweb.com> schrieb im Newsbeitrag
news:Dhx88.95486$KM2.3465594 at atlpnn01.usenetserver.com...
> "Mark McEahern" <mark at mceahern.com> wrote ...
> > In the FixedPoint module found here:
...
>
> Since the *intent* for None is that there should only ever be one object
of
> type None, the better test would be
>
>     if something is None
>
> which can easily be used without requiring any changes to FixedPoint.
>
...

You're right, Steve, but ...
Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> 1 == None
0
>>>

Seems_it's_supposed_to_be_that_way_ly your's
Michael





From cbrown at met.co.nz  Wed Feb 13 14:24:52 2002
From: cbrown at met.co.nz (Colin Brown)
Date: Thu, 14 Feb 2002 08:24:52 +1300
Subject: UTC time conversion to internal time
References: <49ja8.17872$N31.902613@ozemail.com.au> <m3n0yd3k33.fsf@mira.informatik.hu-berlin.de>
Message-ID: <x6za8.18294$N31.923495@ozemail.com.au>

Thanks very much guys.

The calendar.timegm() works a treat under Windows and keeps my code tidy.

Colin Brown
PyNZ




From hans at inenco.no  Wed Feb 20 10:31:44 2002
From: hans at inenco.no (Hans Kristian Ruud)
Date: Wed, 20 Feb 2002 15:31:44 GMT
Subject: Question on if, while, for
References: <e959e7b.0202110545.3dff102@posting.google.com> <7xd6zcw19t.fsf@ruckus.brouhaha.com>
Message-ID: <3C73C1E0.D05F7AC0@inenco.no>


Paul Rubin wrote:

> c at tech.usd253.org (Fred) writes:
> > I have written a simple program that I want to simplify. I used if and
> > elif. But I am sure this can be simplified with for or while. Or what
> > if I wrote a function to do the testing?
> >
> > Please DO NOT re-write this for me I want to figure it out for myself.
> > I would just like to know what is the BEST way to do it:
> >
> > 1) For loop (nested??)
> > 2) While loop
> > 3) Function
> > 4) Leave it alone it's fine??
>
> 5) Try to take a clearer look at the problem you're solving, rather
> than getting wrapped up in the programming details.  You're trying to
> reduce a fraction to lowest terms.  Never mind programming--how would
> you explain that process to a math student?  I don't think you'd do it
> the same way your program does it.  Try to make your program more
> closely follow the methods you'd learn in a math class for reducing
> fractions.  Only after you understand the math should you try to
> express it as Python code.
>
> I'll give a hint: you'll want to know how to compute the least common
> denominator of two integers a and b.  You might want to look in a book
> to find out how to do that.

And to spoil all the fun, here is a solution

>>> def gcd(a, b):
...  while a > 0:
...   print 'a = %d, b = %d' % (a, b)
...   rest = b % a
...   b = a
...   a = rest
...  return b
...
>>> gcd(4, 6)
a = 4, b = 6
a = 2, b = 4
2

- hans kristian -




From gravel.stephane at hydro.qc.ca  Thu Feb 21 13:10:27 2002
From: gravel.stephane at hydro.qc.ca (Stéphane Gravel)
Date: Thu, 21 Feb 2002 13:10:27 -0500
Subject: PyDoc
Message-ID: <yKad8.2774$p31.98491@wagner.videotron.net>

Hi

I'm trying to generate my documentation with PyDoc but it gives me this
error:
"problem in i:\lib\python\Products\Intranet\__init__.pyc - ImportError:
cannot import name Persistent"

Can anyone help me?

thanks in advance
St?phane





From ponderor at lycos.com  Thu Feb 14 13:51:40 2002
From: ponderor at lycos.com (Dean Goodmanson)
Date: 14 Feb 2002 10:51:40 -0800
Subject: Newbie: Object assignment return value
Message-ID: <e81be8b2.0202141051.157160d6@posting.google.com>

(I think) I'm trying to understand if object assignment has a return value.

Scenario A:
>>> str( spam= 5 )
''
>>> str((spam=5))
  File "<stdin>", line 1
    str((spam=5))
            ^
SyntaxError: invalid syntax
>>> #  continued : Scenario B:
>>> print spam
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
NameError: name 'spam' is not defined

A: Why does including an extra set of paranthesis throw when the original doesn't?

B: What happend to my spam?
Espcially when this works:
>>> str(int())
'0'
>>> str((int()))
'0'
>>>

Wishin'-I-had-something-practical-for-an-example,

Dean


From mdk at mdk.com  Fri Feb 15 10:16:19 2002
From: mdk at mdk.com (MDK)
Date: Fri, 15 Feb 2002 10:16:19 -0500
Subject: Mutex a file?
References: <a4h98k$ao2f$1@ID-98166.news.dfncis.de> <a4ghg4.3vv79at.1@kserver.org>
Message-ID: <a4j8o4$mpqo$1@ID-98166.news.dfncis.de>

"Sheila King" <usenet at thinkspot.net> wrote in message
news:a4ghg4.3vv79at.1 at kserver.org...
> On Thu, 14 Feb 2002 16:12:51 -0500, "MDK" <mdk at mdk.com> wrote in
comp.lang.python in
> article <a4h98k$ao2f$1 at ID-98166.news.dfncis.de>:
>
> > I have a file that is to be read to and written from by many Python
scripts
> > running at the same time.
> >
> > Is there a way to put a Mutex on it so that the scripts wait their turn
> > before tyring to open it?
>
> This is a problem I worked on some time back. On most Unices, it is not
> hard to lock the file (which is what you need to do). I guess there are
> some problems if you are working across certain types of networks. On
> WinNT and Win2000 (and probably XP) there is also a great deal of control
> over files and file access. It's the Win9x where you loose a lot of this
> ability. I'm not sure what platform you are running on...if Windows,
> you might want to look in more detail at the win32all libraries from
> Mark Hammond.
>
> Anyhow, here are some links to the discussion I initiated on this topic
> some time back. Read the threads. There are many useful comments.
>
http://groups.google.com/groups?as_umsgid=t7uqot8n8gjm03r2m03md8kfsjuauchqhu
%404ax.com
>
http://groups.google.com/groups?as_umsgid=djsfotomogdeba70gp1cuogjbt5ii3chrb
%404ax.com
>
http://groups.google.com/groups?as_umsgid=bo2ootg61vmposfvsg8gvrks6eqiuss940
%404ax.com
> http://mail.python.org/pipermail/tutor/2001-August/008366.html
>
>
> Here is the module I wrote, and am using in a production environment
> on a Linux server for file locking. Advice given to me re: the
> Windows component of this modules is...well, that it wouldn't be
> recommended to use in a production environment. But who would run
> a production server on a Win98 machine anyhow? So, for me it let's
> me test my scripts on my home Windows machine and then upload it
> to my Linux server and expect the same results.
>
> FWIW, I've pasted the code I'm currently running below my sig.
>
> --
> Sheila King
> http://www.thinkspot.net/sheila/
>
> "When introducing your puppy to an adult cat,
> restrain the puppy, not the cat." -- Gwen Bailey,
> _The Perfect Puppy: How to Raise a Well-behaved Dog_

<code-snipped>

Well I spent half my morning reading the threads and other things referenced
by the threads.  This is a much more complicated (and interesting) subject
than I thought.

The code that you posted is exactly what I need: simple file locking on a
Windows machine.

Thank you for sharing your hard work.




From sabren at manifestation.com  Thu Feb 21 13:47:11 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 21 Feb 2002 13:47:11 -0500 (EST)
Subject: a question regarding conciseness
In-Reply-To: <3C753E6A.9050201@free.fr>
Message-ID: <Pine.LNX.4.44.0202211346230.29940-100000@mercury.sabren.com>

On Thu, 21 Feb 2002, Christophe Delord wrote:

> The problem with sort is that it sorts the list as a side effect.

You can use copy.copy() or copy.deepcopy() first to
eliminate that side effect.

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From grante at visi.com  Mon Feb 25 14:16:32 2002
From: grante at visi.com (Grant Edwards)
Date: Mon, 25 Feb 2002 19:16:32 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <slrna7l176.3al.quinn@vomit.ugcs.caltech.edu>
Message-ID: <k4we8.3526$N7.997040@ruti.visi.com>

In article <slrna7l176.3al.quinn at vomit.ugcs.caltech.edu>, Quinn Dunkan wrote:
> On Sun, 24 Feb 2002 19:34:34 -0800, David Eppstein <eppstein at ics.uci.edu> wrote:
>>In article <a5cb54$27c2$1 at agate.berkeley.edu>,
>> Daniel Yoo <dyoo at hkn.eecs.berkeley.edu> wrote:
>>
>>> : But what is the functional equvalent of:
>>> 
>>> : for x in items:
>>> :     x.f()
> 
> There really is no functional equivalent of that, because it's not a functional
> concept.  If your function doesn't have any side-effects, calling the function
> only to throw away its value does nothing but suck up CPU time.
> 
> If you're calling the method for its side-effects, I'd write:
> 
> for x in items:
>     x.f()
> 
>>I'd prefer
>>[x.f() for x in items]
>>
>>It's not functional syntax, but so what?
> 
> Sure it is.

Methinks there are two definitions of "functional" at work here:

 1) The traditional CS "no side-effects" definition.
 2) Something that looks like a function  call.

?

-- 
Grant Edwards                   grante             Yow!  Yow! I forgot my
                                  at               PAIL!!
                               visi.com            


From rmunn at pobox.com  Tue Feb 19 11:36:13 2002
From: rmunn at pobox.com (Robin Munn)
Date: 19 Feb 2002 16:36:13 GMT
Subject: newbie here...
References: <mailman.1014111829.2043.python-list@python.org>
Message-ID: <slrna74vh8.et.rmunn@rmunn.dyndns.org>

On Tue, 19 Feb 2002 19:43:16 +1000, Paul Berg <paul at spondo.com> wrote:
>This is a multi-part message in MIME format.

Ah, another victim of Outlook Express's HTML-by-default settings. Please
turn this off, it looks ugly for the rest of us. Go to Tools->Options,
click the Send tab, and change both "Mail Sending Format" and "News
Sending Format" to Plain Text instead of HTML. Thanks!

>Hello all,
>
>Im just new to Python and don't know anything about it. I've been =
>programming in Miva Script if anyone has heard of it and also a little =
>Php.
>
>Could someone send me some script so i can so  i can relate to it. And =
>also things like databases types it can use(if so), Small things like =
>that. Im known to be a quick learner so hopefully will be programming in =
>days.
>
>Paul

I suggest the Python Tutorial, found at:

http://www.python.org/doc/current/tut/tut.html

or some of the tutorials at:

http://www.python.org/doc/Newbies.html

I particularly recommend "How to Think Like a Computer Scientist"; it
will not only teach you Python, but it will also teach you some handy
problem-solving methods that can be applied in any language.

-- 
Robin Munn
rmunn at pobox.com


From syring at email.com  Mon Feb 18 06:08:55 2002
From: syring at email.com (Karl M. Syring)
Date: Mon, 18 Feb 2002 12:08:55 +0100
Subject: win32com and MS Excel
References: <mailman.1013967115.15656.python-list@python.org> <98322540.0202180251.2ddf81f4@posting.google.com>
Message-ID: <a4qnig$29fdu$1@ID-7529.news.dfncis.de>

"Serge Terekhoff" <alife at narod.ru> schrieb
> Dear all!
> 
> on
> from win32com.client import Dispatch
> 
> My Python2.2 replies:
> 
> Traceback (most recent call last):
>   File "F:\Users\Serge\MyPy\to_excel.py", line 1, in ?
>     from win32com.client import Dispatch
> ImportError: No module named win32com.client
> 
> 
> What i did wrong?

Those wrong-doers that did not install win32all will get this reply.

Karl M. Syring



From elton at mybizlinks.net  Fri Feb 15 21:13:19 2002
From: elton at mybizlinks.net (Elton M. Labajo)
Date: Sat, 16 Feb 2002 10:13:19 +0800 (PHT)
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph>

I like the idea of lite and full version :)
> Reading the first few chapters of _How to Think Like a Computer
> Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/ for those who haven't
> heard of this book), I was struck by how Python resembles BASIC in some
> ways that make it a very good language for teaching:
>
> * Interpreted, so there's immediate feedback. Students can just type
>  'python' at the command prompt and start coding, receiving *immediate*
>  feedback on the effects of what they type.
>
> * Non-scary assignment and output statements. In fact, Python and
>  BASIC's output statments look identical. "Print" is a natural choice
>  for ease of understanding. Using sys.stdout.write can come later, once
>  students begin to realize that they don't always *want* a newline or
>  space appended to their output.
>
> * No need to declare variables. Use them and they are immediately
>  created for you, ex nihilo.
>
> All three of these things make for a very low barrier-to-entry for
> people who've never done any programming before. I'm sure there are
> other similarities as well, that maybe someone else will point out. Of
> course, Python goes beyond BASIC in many ways:
>
> * NO LINE NUMBERS! Enough said.
>
> * Rich control-flow statements. Of course, this is available in more
>  modern versions of BASIC, but back when I was first learning to
>  program (at age six, using BASIC), all I had was conditional GOTO. You
>  know, IF condition GOTO line-number. Or, for the advanced students, IF
>  condition GOSUB line-number (I never did understand the necessity for
>  that kind of statement until well after I'd stopped using BASIC
>  entirely and moved to more "modern" languages like... Pascal).
>
> * Batteries included. This is the big one, which makes Python more than
>  just a toy language suitable for learning. The richness of the
>  standard library makes Python incredibly useful, and I for one would
>  like to see it continue to grow. I understand the concern that people
>  developing Python-based programs for distribution would like their
>  users to be able to run a 100k script without downloading a 10M
>  distribution, but I don't want to see that cause a reduction in the
>  standard library. Instead, we could make make two Python
>  distributions, a "full" install containing everything and a "lite"
>  install containing only the interpreter and maybe -- *maybe* -- a
>  couple of essential modules like os and sys. Then people developing
>  Python programs like, say, a pygame-based game, could distribute a
>  single package containing the "Python lite" distribution, their own
>  code, and any other modules their code depended on -- all in one
>  easy-for-the-end-used-to-install package.
>
> Anyhow, I hope this helps develop some ideas.
>
> --
> Robin Munn
> rmunn at pobox.com
> --
> http://mail.python.org/mailman/listinfo/python-list






From phil at river-bank.demon.co.uk  Mon Feb 25 03:21:20 2002
From: phil at river-bank.demon.co.uk (Phil Thompson)
Date: Mon, 25 Feb 2002 08:21:20 +0000
Subject: GUI (PyQt) & thread problem
References: <8b5d764c.0202240451.1c4bda3b@posting.google.com> <slrna7ird3.a0.missive@whave.frontiernet.net>
Message-ID: <3C79F400.F36C946B@river-bank.demon.co.uk>

Lee Harr wrote:
> 
> > Can someone tell me what the problem is with the following program, as it
> > hangs. I'm pretty sure there's some inherent design-flaw involved, but
> > I'm not familiar enough with threads or GUI-programming to know what it
> > could be.
> >
> > I've also tried modyfing the program to use PYSIGNALs instead of passing the
> > caller to the thread, but the results were the same.
> >
> 
> I did a google search for pyqt threading and I am not so sure that
> these things are working together.
> 
> There is a qt QThread class, but it does not seem to be available
> to pyqt yet.

QThread, QMutex, QSemaphore and QWaitCondition are implemented in PyQt
v3.1pre3.

Phil



From pan-news at asheeshenterprises.com  Wed Feb 27 22:32:35 2002
From: pan-news at asheeshenterprises.com (Asheesh Laroia)
Date: Thu, 28 Feb 2002 03:32:35 GMT
Subject: Regular expression problem
References: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com> <3C7D85BC.71550E5F@gmx.de>
Message-ID: <nxhf8.149057$dh.38723170@typhoon.nyroc.rr.com>

This is great, thanks!

Only one problem.  I'm having trouble (I did give it a try) making the
following work:

	<@Trap Body text:>Useful Text

I need to still be able to extract "Useful Text", not delete it.

Thanks again!

-- Asheesh.

On Wed, 27 Feb 2002 20:19:56 -0500, Wolfgang Grafen wrote:

> import re
> 
> rc=re.compile("<@Trap Body text\s*"
>                           "(?:(?P<assigned>=)|(?P<unassigned>>))\s*"
>                           "(?P<rest>.*?)\s*\Z",
>                          re.MULTILINE|re.DOTALL).match
> 
> t1='<@Trap Body text>'
> t2='<@Trap Body text=<FONT "Times">'
> t3="""<@Trap Body text=<FONT "Times"><CCOLOR\n   "Black"><SIZE
> 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n
> 58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
> 0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE
> 0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
> "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
> 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
> 25>>"""
> 
> rc(t1).groups()
> (None, '>', '')
> 
> rc(t2).groups()
> ('=', None, '<FONT "Times">')
> 
> rc(t3).groups()
> ('=', None, '<FONT "Times"><CCOLOR\n   "Black"><SIZE 11><HORIZONTAL
> 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n  58.3><C-POSITION
> 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK 0><CLEADING -0.05\n
> ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER
> 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
> "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
> 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
> 25>>')
> 
> cheers
> 
> wolfgang
> 
> 
> Asheesh Laroia schrieb:
> 
>> I have some SGML input (PageMaker 6.5 tagged text), and I want to be
>> able to recognize (and delete) a tag.  That tag looks like:
>>
>>         <@Trap Body text:>
>>
>> It may also look like <@Trap Body text: useless-data>.
>>
>> So, I tried the regular expression r"<@.?>".  That doesn't match the
>> above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
>> using, and why doesn't this work?
>>
>> Thanks in advance!
>>
>> -- Asheesh Laroia.
>>
>> PS: An example of the tag "in the wild" is the following string:
>>
>> <@Trap Body text=<FONT "Times"><CCOLOR
>>  "Black"><SIZE 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE
>>  70><C+SIZE
>> 58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
>> 0><CLEADING -0.05
>> ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER
>> >0><GALIGNMENT "justify
>> "><GMETHOD "proportional"><G& "ENGLISH"><GPAIRS 4><G% 120><GKNEXT
>> 0><GKWIDOW 1><GKORPHAN 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75
>> 100 150><GSPACE -5 0 25>>


From quinn at regurgitate.ugcs.caltech.edu  Fri Feb 15 21:23:01 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 16 Feb 2002 02:23:01 GMT
Subject: Object copying itself
References: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com>
Message-ID: <slrna6rgk4.9fc.quinn@regurgitate.ugcs.caltech.edu>

On Fri, 15 Feb 2002 18:58:05 +0000, Dale Strickland-Clark
<dale at riverhall.NOTHANKS.co.uk> wrote:
>I have an object which is just data and an __init__ to set up some
>defaults.
>
>Now I need to allow __init__ to get the defaults from another instance
>of itself if passed as an argument.
>
>Most of the values are simple numerics and a few object references but
>there are also a couple of lists. 
>
>I want object references copied (shallow copy) but the lists need to
>be copied entirely (deep copy).
>
>Do I have to do this by hand or is there a neat way of doing it?

Well, if you really have a lot of attributes, you could futz around with
__dict__.

import copy
from types import *
class Foo:
    def __init__(self, default=None):
        if default is None:
            ...
        else:
            for k, v in default.__dict__.items():
                if hasattr(v, '__getitem__') or v in (ListType, TupleTyple):
                    nv = copy.deepcopy(v)
                else:
                    nv = v
                setattr(self, k, v)

I'd be more inclined to just write out a bunch of assignments, though.  Note
that 'self.__dict__.update(other.__dict__)' is a popular pydiom, but that won't
deepcopy the lists.


From brian at sweetapp.com  Sat Feb  2 22:53:23 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sat, 2 Feb 2002 19:53:23 -0800
Subject: print statement in lambda expression
In-Reply-To: <NW178.16270$bIE.6837@news1.bloor.is>
Message-ID: <001001c1ac66$53e8ec60$445d4540@Dell2>

> I just recently found that I couldn't put a print statement in a
> lambda expression as a binding in Tk. Could someone explain why?
> 
>     ie.
> 
>     b1 = Button(frame, text="Message", command=lambda: print
"message")
> 
> If I replace print with sys.stdout.write, it works fine. I don't
> follow why.

'print "message"' is an statement, not an expression, and lambda
expressions must be, well, expressions :-)

All function calls are expressions so sys.stdout.write will work.

Cheers,
Brian




From nelson at crynwr.com  Wed Feb 20 09:57:36 2002
From: nelson at crynwr.com (Russell Nelson)
Date: 20 Feb 2002 09:57:36 -0500
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <m2vgcs5hgf.fsf@desk.crynwr.com>

There are two Python ideoms that bother me:

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

and:

    matched = re.match(r'egexp(.*)', line)
    if matched:
        print matched.group(1)

I'd really, really like it if you could do one(1) assignment within
the boolean expression in a while and an if.  Like this:


    while line = file.readline():

and:

    if matched = re.match(r'egexp(.*)', line):
        print matched.group(1)

I just think it would be easier to read programs if they could be
written like this.  Should I write up a PEP or has this already been
discussed and dismissed?

-- 
-russ nelson              http://russnelson.com | Crypto without a threat
Crynwr sells support for free software  | PGPok | model is like cookies
521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | 


From clj at theWorld.com  Sat Feb  2 22:05:20 2002
From: clj at theWorld.com (Chris Jones)
Date: Sun, 3 Feb 2002 03:05:20 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx>
Message-ID: <tdnzo2r9sen.fsf@shell01.TheWorld.com>

Tim Hammerquist <tim at vegeta.ath.cx> writes:

[...]

  My eyes just can't parse:
  
  if ( some_condition() ) {
    {
      for (i = first_i();i <= 10; i++)
        {
          initialize_some_stuff();
          while ( some_other_condition() )
            {
              do_some_more_stuff()
            }
        }
    }
  
I can, and I notice you've got unbalanced braces (look at your first two
lines). 


From phr-n2002a at nightsong.com  Thu Feb 21 06:22:54 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 03:22:54 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de>
Message-ID: <7x7kp7krjl.fsf@ruckus.brouhaha.com>

"Karl M. Syring" <syring at email.com> writes:
> > Not likely.
> 
> You really should have a look at high precison libraries (e.g.
> http://www.hipilib.de). They will use rational numbers internally.

Yes, but who actually uses those libraries, and for what?


From sjmachin at lexicon.net  Wed Feb 13 16:30:34 2002
From: sjmachin at lexicon.net (John Machin)
Date: 13 Feb 2002 13:30:34 -0800
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net>
Message-ID: <c76ff6fc.0202131330.66bd9d93@posting.google.com>

"John W. Baxter" <jwbaxter at spamcop.net> wrote in message news:<130220020751360934%jwbaxter at spamcop.net>...
> In article <mailman.1013611523.15053.python-list at python.org>, Simon
> Brunning <SBrunning at trisystems.co.uk> wrote:
> 
> > Probably to late to do anything about it now - think of all the code which
> > would break!

Yes, unfortunately there is a lot of octal in the library. Some of it
has the reasonable excuse of copying existing Unix practice or
mechanical translation of existing C code:

stat.py:    return mode & 0170000
stat.py:S_IFDIR  = 0040000

However this sort of thing boggles me:

codecs.py:BOM64_BE = '\000\000\376\377'
codecs.py:#       corresponds to Unicode U+0000FEFF in UCS-4

Why not '\x00\x00\xfe\xff'? There is not a whiff of octal in the whole
Unicode book. Why translate hex into octal?

> 
> Did the elves at AT&T pick up the "leading 0 means octal" from
> something earlier, or did they invent this stupidity themselves?

They were captured by aliens and implanted with concepts like using
octal notation on machines whose word-size was not a multiple of 3,
NUL-terminated strings, signed characters, unprintable characters in
file names, backspace not being honoured during a login attempt, etc.
All the (many) good ideas were their own.

> 
> Either way, we're stuck with it.

Unfortunately.

I've been semi-bitten just in the last week; grabbed a table of data
that included month numbers with leading zeroes and pasted it into a
script -- the 08 and 09 caused a syntax error which puzzled me for a
minute or two (I haven't needed to think about octal much since I
worked on an ICL 1905 (24-bit words)). Fortunately no harm done.

Next step: see if Pychecker has an octal warning feature and if not
pester the maintainer to provide one.

> 
>    --John (could the zero-less Romans have been right after all?)


From gerson.kurz at t-online.de  Mon Feb  4 10:58:16 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 15:58:16 GMT
Subject: Fun with lambda and map
References: <mailman.1012542242.4324.python-list@python.org> <a3mddf$7do$1@news1.xs4all.nl>
Message-ID: <3c5eaf65.22609703@news.t-online.de>

On 4 Feb 2002 16:38:07 GMT, ivo at nospamAmaze.nl wrote:

>I extended the calculus a bit with mul (based on your sig :) and pred, 
>hoping that I could fac in lambda, but I guess I've reached pythons lambda 
>limits.
>
>The statement:
>
>fac = lambda q: if_then_else(is_zero(pred(q)))(n1)(mul(q, fac(pred(q))))
>
>will recurse endlessly, because the else-part will always be evaluated
>(but just not used as a result).
>

Well, you can:

IF_ELSE = lambda x:(map(lambda a:a and 1,[x[0]()])[0] and x[1]()) or
x[2](x[3])
fac = lambda n:IF_ELSE([lambda:not n,lambda:n==n,lambda
n:fac(n-1)*n,n])


for i in range(20):
    print i, "->", fac(i)

If you're in a masochistic mood, you can write this in one line:

fac = lambda n:map(lambda x:(map(lambda a:a and 1,[x[0]()])[0] and
x[1]()) or x[2](x[3]),[[lambda:not n,lambda:n==n,lambda
n:fac(n-1)*n,n]])[0]




From warkid at storm.ru  Tue Feb 19 06:13:17 2002
From: warkid at storm.ru (Kerim Borchaev)
Date: Tue, 19 Feb 2002 14:13:17 +0300
Subject: execfile question
Message-ID: <17592.020219@storm.ru>

Can someone explain why while running this script I don't see C.__del__
eventually called?

For me the output of it's execution is:

>main.py
it's me
init
executed
>

##main.py######################
execfile('executable.py', {})
print 'executed'
########################

##executable.py######################
print "it's me"
class C:
    def __init__(self):
        print 'init'
    def __del__(self):
        print 'deleted'

c = C()
########################

Thanks in advance.





From bkc at Murkworks.com  Wed Feb 27 13:15:17 2002
From: bkc at Murkworks.com (Brad Clements)
Date: Wed, 27 Feb 2002 13:15:17 -0500
Subject: Contributing patches (was Re: urllib and proxy)
References: <a5ecas$qiq$1@zeus.polsl.gliwice.pl> <20020226100136.GA18498@andy.tynant.ftech.net> <mailman.1014733123.3107.python-list@python.org> <un0xwxob7.fsf@python.net>
Message-ID: <3c7d2377$1_6@news.newsgroups.com>

I think the Python keepers are swamped. I've had a patch pending for a
little over a month, made 3 rounds of changes as requested but now it's
stalled.

There are a lot more patches older than mine, so I'll just wait. Don't get
discouraged if your patch sits for a while.


"Michael Hudson" <mwh at python.net> wrote in message
news:un0xwxob7.fsf at python.net...
> Andy Gimblett <gimbo at ftech.net> writes:
>
> > Earlier on today I posted to this list a modified urlopen() which
> > supports proxy specification.  How do I go about suggesting this as a
> > modification to the standard python library?
> >
> > I've seen these:
> >     http://www.python.org/download/cvs.html
> >     http://www.python.org/patches/
> >
> > and they seem to suggest that I can just use my existing SourceForge
> > account and submit a patch to the patch manager, and I presume then
> > someone examines it, decides if it's a good thing or not, and checks
> > it in or rejects it?
>
> Yes, more or less.  It may take a while to get noticed, I'm afraid.
> There's also the quite likely step of being asked to modify your
> patches in some way before they go in.  There can be quite some
> to-and-fro-ing on the tracker.
>
> > Is this how it works?
>
> More or less.
>
> > Just thought I'd ask here before blundering in.
>
> Blunder away!
>
> Cheers,
> M.
>
> --
>   Famous remarks are very seldom quoted correctly.
>                                                     -- Simeon Strunsky




-----=  Posted via Newsfeeds.Com, Uncensored Usenet News  =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
 Check out our new Unlimited Server. No Download or Time Limits!
-----==  Over 80,000 Newsgroups - 19 Different Servers!  ==-----


From jmfbahciv at aol.com  Thu Feb 14 06:22:36 2002
From: jmfbahciv at aol.com (jmfbahciv at aol.com)
Date: Thu, 14 Feb 02 11:22:36 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net>
Message-ID: <a4geh2$hir$12@bob.news.rcn.net>

In article 
<2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345 at lp.airnews.net>,
   claird at starbase.neosoft.com (Cameron Laird) wrote:
>In article <130220020751360934%jwbaxter at spamcop.net>,
>John W. Baxter <jwbaxter at spamcop.net> wrote:
>			.
>			.
>			.
>>Did the elves at AT&T pick up the "leading 0 means octal" from
>>something earlier, or did they invent this stupidity themselves?
>			.
>			.
>			.
>OK, folks, how far back *can* we trace this?  My
>mind associates it with DEC systems, going back
>to the '60s, but I couldn't find any confirmation
>of that in a quick search.

You will never find a DEC standard that ever defined
a number with a leading zero to be octal.  I never
typed up anything that ever had that kind of standard. 

You must be confusing the character zero with the letter O.
In MACRO you could force the assembler to evaluate any
number by using a prefix:  ^D meant decimal; ^O meant octal.

/BAH

Subtract a hundred and four for e-mail.


From sholden at holdenweb.com  Tue Feb 12 10:17:08 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Feb 2002 10:17:08 -0500
Subject: searching backwards in a string
References: <7xk7tjayos.fsf@ruckus.brouhaha.com>
Message-ID: <Jlaa8.174222$KM2.7680225@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xk7tjayos.fsf at ruckus.brouhaha.com...
> I have a string and want to search backwards for a regexp, starting at
> a given index, like if you search backwards in a text editor.  Is
> there a way to do that with re?  The docs aren't clear about what's
> supposed to happen if the ending position to the pat.search method is
> less than the starting position, but I tried it and it doesn't seem
> to do what I want.  Any advice appreciated--thanks.

Warning: IANOREG

If you'd like to find the "last match before a given position" then,
supposing string p to represent the regex matching the pattern you are
looking for, you can try match()'ing

    ".*(%s)" % p

Given the greedy nature of matching, this should give you the rightmost
match in groups(1). Maybe. Anyway, an idea to play with.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From gmcm at hypernet.com  Thu Feb  7 08:56:40 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 07 Feb 2002 13:56:40 GMT
Subject: import subtleties - what's going on?
References: <3c6376a8.15716308@news.muenster.de>
Message-ID: <Xns91AE5B2746D90gmcmhypernetcom@199.171.54.215>

Martin Bless wrote:

[snip]

> I would so much like being to launch Python modules from the
> commandline given their module names instead of those often very long
> and unpythonic filenames.
> 
> The launcher module below _almost_ does the trick - but one issue
> remains unsolved. Modules launched this way will itself not be able to
> import sibling modules from the package they reside in unless that
> package itself is on the path or the import expression is fully
> qualified. Which normally it isn't and it shouldn't be.

Relative imports ("import sibling" where the fully qualified name
is "pkg.sibling") are only triggered when the importing module
belongs to "pkg". Python determines this by examining the fully
qualified name of the importer. 

Which, with your trick, is "__main__".

It's a lousy idea to use a package member as a script. If pkg.mod is
running as __main__ and imports pkg.other, and pkg.other imports
pkg.mod, you'll have 2 copies of pkg.mod and lots of mysterious
errors.

-- Gordon
http://www.mcmillan-inc.com/


From mcherm at destiny.com  Mon Feb 11 09:13:49 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 11 Feb 2002 09:13:49 -0500
Subject: How did TimBL's speech at Python 10 go yesterday?
Message-ID: <3C67D19D.3050207@destiny.com>

>> PS: I intentionally used URL in my comments, because if the URI is a
>> URN instead of a URL then it can't possibly be "downloaded" as there's
>> noplace to connect to. Just trying to keep the terminology straight.
> 
> There's something (unintentionally?) hilarious about this postscript.
> 
> Cheers,
> M.


What can I say? I didn't invent the acronyms... just trying to live with 
'em.

-- Michael Chermside





From tim-dated-1013819586.fbd419 at catseye.net  Thu Feb  7 19:33:03 2002
From: tim-dated-1013819586.fbd419 at catseye.net (Tim Legant)
Date: 07 Feb 2002 18:33:03 -0600
Subject: encapsulation, typing & other questions
In-Reply-To: <3C62E45C.84B8506A@ccvcorp.com>
References: <3C62C5C5.7FEF0B10@daylight.com> <3C62E45C.84B8506A@ccvcorp.com>
Message-ID: <86heosersw.fsf@skitty.catseye.net>

Jeff Shannon <jeff at ccvcorp.com> writes:

> (Note that Python *does* have some level of data hiding, but it's deliberately
> easy to circumvent.  Instead of enforcing secrecy, it simply asks "Please don't
> look at this"... )

And quite frankly, C++, with it's strict type system and much-vaunted
data-hiding tools, can be circumvented with a couple lines of code:

  #define private   public
  #define protected public
  #include <class_whose_members_I_want_to_rape.h>

It's actually more difficult in Python, since you have to remember and
use the mangled name for each access.  It all comes down to
self-discipline.


Tim




From frederic.giacometti at arakne.com  Fri Feb 22 23:06:02 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Sat, 23 Feb 2002 04:06:02 GMT
Subject: Stackless Platform Independence?
References: <mailman.1014420937.15379.python-list@python.org>
Message-ID: <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com>

"Tim Peters" <tim.one at comcast.net> wrote in message
news:mailman.1014420937.15379.python-list at python.org...
> [Christian Tismer]
> > As you might have noticed, Stackless has changed quite
> > much:
> >  - Platform independence is gone.
>
> [Christopher Browne]
> > I hope not...  I would hope to hear that "platform indepedence has
> > _ARRIVED_"
>
> No, he meant what he said.  All the good changes are bought at the cost of
> needing some platform-specific assembler, to transform the platform C
stack
> from an Insurmountable Obstacle into just another piece of manipulable
> state.

Then, one could also deal with the reason of why Christian has had to come
to that point.

I am personally still surprised that stackless has not be considered more
seriously in the python-dev group, especially in regard to the threading
problems and limitations of the current python versions.

Yet, a stackless python would open many possibilities and new application
areas; and would be a quantum improvement on the core engine of a different
substance from all what has been seen since Python 2.0.

Frederic Giacometti





From sab at NOSPAM.freeuk.com  Fri Feb 22 15:03:46 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Fri, 22 Feb 2002 20:03:46 -0000
Subject: re. help needed  :(
Message-ID: <a56838$qb8$1@newsg4.svr.pol.co.uk>

Hi,
    Im trying to replace the single word "lamb" and the word lamb followed
by any punctuation in this string:

x="Mary had a little lamb called lamb, but didn't like the lamb's mood"

i want to replace all of the single words "lamb[punctuation]" but not the
word "lamb's" to achieve this:

x="Mary had a little pig called pig, but didn't like the lamb's mood"

any help with be greatly appreiciated!

ta

G. Willoughby




From phr-n2002a at nightsong.com  Fri Feb  8 22:05:42 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 08 Feb 2002 19:05:42 -0800
Subject: range(seq)?
References: <slrna67foh.dl7.mlh@vier.idi.ntnu.no> <slrna68v2c.qvm.quinn@hurl.ugcs.caltech.edu>
Message-ID: <7xr8nvxsl5.fsf@ruckus.brouhaha.com>

quinn at hurl.ugcs.caltech.edu (Quinn Dunkan) writes:
> I've always thought that sequences should have a keys() method.
> Then we could replace range(len(seq)) with seq.keys().  The
> 'indices' generator would then be spelled [].iterkeys().
> 
> I'm sure this has been suggested and rebuffed, though.

Hey, I kinda like that.  I haven't heard it before, though I haven't
been around that long.


From mlh at vier.idi.ntnu.no  Sun Feb 17 20:38:34 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 18 Feb 2002 01:38:34 +0000 (UTC)
Subject: Directed Acyclic Graph Representation
References: <j9du6uckr4g1am3780qtig50jdhcaf6fcr@4ax.com>
Message-ID: <slrna70moq.som.mlh@vier.idi.ntnu.no>

In article <j9du6uckr4g1am3780qtig50jdhcaf6fcr at 4ax.com>, Tim Roberts wrote:
>Is anybody aware of a reasonable package for representing a directed
>acyclic graph in Python, and for rendering that graph in some meaningful
>way?  I need to build a dependency tree analyzer for, for example,
>determining the necessary link order for a set of object libraries.

At least efficient representation is available through kjbuckets:
http://www.chordate.com/kjbuckets/

If you'd like to store large graphs efficiently, e4graph might be
interesting:
http://e4graph.sourceforge.net/

Not sure how easy it is to use from Python, but it is based on
MetaKit, which has very nice Python mappings:
http://www.equi4.com/metakit/python.html

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From mrsedwards at visi.com  Mon Feb 25 19:44:37 2002
From: mrsedwards at visi.com (Grant's Mother)
Date: Mon, 25 Feb 2002 19:44:37 -0500
Subject: call me
References: <u7fcig3gep8k40@corp.supernews.com> <slrna7fh30.a72.gerhard@lilith.my-fqdn.de> <wgue8.3490$N7.975881@ruti.visi.com>
Message-ID: <3C7ADA75.43CE5F3F@visi.com>

Grant Edwards wrote:
> 
> In article <slrna7fh30.a72.gerhard at lilith.my-fqdn.de>, Gerhard H?ring wrote:
> > Scott Hathaway wrote in comp.lang.python:
> >> Mom,
> >>
> >> Give me a call when you have a chance.
> >
> > Wow, a familiy of diehard Python fans :-)
> 
> I was proud that my mom cut an article about Linus Torvalds out
> of the paper and sent it to me, but I don't ever expect to run
> into her on Usenet.

Grant, dear, you underestimate me!  Just because I don't
post in this newsgroup doesn't mean I'm not here keeping 
an eye on you honey.

> --
> Grant Edwards                   grante             Yow!  My pants just went to
>                                   at               high school in the Carlsbad
>                                visi.com            Caverns!!!

Oh you kids...  And remember to write more often!

Love, Mom

P.S.: Are you still dating that nice girl?  You don't bring her 
by any more.


From pballard at ozemail.com.au  Mon Feb 25 21:11:09 2002
From: pballard at ozemail.com.au (Peter Ballard)
Date: 25 Feb 2002 18:11:09 -0800
Subject: Can I import an arbitary file?
Message-ID: <9d5509fa.0202251811.11837b04@posting.google.com>

G'day all,

I've written a program (an SRAM generator, if you must know) which
has some configuration info (for different SRAM sizes etc.) in a separate
file. This configuration file is python code.

What I want to do is import this file. The problem is, this file could
be any name at all. What I want to do is parse my command line, and
find out the name of my configuration file, and then import the file of
that name. Is that possible?

(Obviously I can read in this arbitrary file and parse it,
but to actually execute it would be much neater).

Or is there any other way I can read in a file of an arbitary name,
and execute its code?

Thanks in advance,

Peter Ballard
(From Australia, the speed skating capital of the world :-)


From weeks at vitus.scs.agilent.com  Thu Feb 14 18:27:13 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Thu, 14 Feb 2002 23:27:13 +0000 (UTC)
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <1013729233.559849@cswreg.cos.agilent.com>

Scirious (Scirious at Scirious.com) wrote:
: I've been learning JAVA for a while, but now I've heard about Python.
: Any coments concerning these two languages?

IMO (this stuff is unprovable):

Python and Java are natural buddies in that they both have big ugly
competitors.  I see Python and Perl in a similar niche: high-level,
untyped-variables, "scripting".  This is a valuable niche, and Python fits
it exceedingly better than Perl.  I also see Java and C++ in a similar
niche: typed-variables, compiled, object-oriented, descendent of C.  This
is also a valuable niche, and Java fits it exceedingly better than C++
(putting aside performance!).

Other similarities:

    All data are implemented as memory addresses (with the exception of a
    few primitive types in Java).

    There is no explicit referencing/dereferencing.

    All arguments to functions are passed-by-value.

    Threads are easy to use.

All of these similarities are language strengths.

Despite my talk of niches, I don't have a clear idea of when I would choose
Java over Python...


Greg


From cfelling at iae.nl  Thu Feb 28 11:36:00 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 17:36:00 +0100
Subject: sorting
References: <3c7e3d12_7@news.teranews.com> <a5ll4a$4ml$1@animus.fel.iae.nl>
Message-ID: <a5lm9g$4vt$1@animus.fel.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:
...
> or to get them all

>    >>> for x in count:
>    ...    print "the number %r occured %d time(s)" % (78, count[78]))

blaim it on stupid cut&paste, but this should read:

    ...    print "the number %r occured %d time(s)" % (x, count[x]))

-- 
groetjes, carel


From sholden at holdenweb.com  Thu Feb  7 09:51:53 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 09:51:53 -0500
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org>
Message-ID: <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com>

"Orr, Steve" <sorr at rightnow.com> wrote in message
news:mailman.1012938609.3651.python-list at python.org...
> Python Newbie here.
>
> I need to migrate a web application from PHP to Python and I have a
> requirement to save state info WITHOUT using cookies where much of the
state
> data MUST not be visible to the end users. The PHP application allows
users
> to logon to one of several databases using any of a number of Oracle
> user/passwords. The logon info needs to be saved for queries on other
pages
> in the app. With PHP it's easy to avoid cookies by saving this info using
> its session handling functions. Is there an EASY way to do this in Python
> without having to adopt the Mother of all frameworks?
>
Disclaimer: I have never used PHP.

However, *if* PHP allows you to associate requests with session state, then
*either*:

a) it does use cookies, you just don't know about them, or
b) it puts session references in the URL.

There simply isn't any other available mechanism. So maybe what you're
looking for is an interface that hides the cookie implementation from the
programmer?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From martin at v.loewis.de  Wed Feb 27 04:12:37 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 10:12:37 +0100
Subject: Python 2.2 Socket Test Failure
References: <Q%Ve8.8688$Dv1.29388@rwcrnsc53>
Message-ID: <m3pu2r1e62.fsf@mira.informatik.hu-berlin.de>

Michael Strayer <mstrayer1 at attbi.com> writes:

> I've built Python 2.2 on a Red Hat 7.2 system and when
> I run make test I get the following error:
> 
> test test_socket crashed -- socket.gaierror: (-2, 'Name or service not 
> known')
[...]
> 
> Could someone explain how to fix the socket crash.  Also how do I
> enable to network resource?

It did not really crash, it just found that the domain name service
does not work properly on your system; perhaps a network
misconfiguration. No need to worry.

Regards,
Martin



From gerson.kurz at t-online.de  Sun Feb 17 01:08:42 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 17 Feb 2002 06:08:42 GMT
Subject: Have to try
References: <mailman.1013916386.10113.python-list@python.org>
Message-ID: <3c6f47dc.3932421@news.t-online.de>

>OK - I know I am going to get fried here but - I am out of options. I'm a
>programmer (Java, VB and C++) ...
> I want to search that string for the occurrence of a substring

If you know C, perhaps you'll find these functions usefull:

def strcmp(a,b):
    "Compare two strings, case-sensitive"
    return a == b

def stricmp(a,b):
    "Compare two strings, not case-sensitive"
    return a.lower() == b.lower()

def strncmp(a,b):
    "Compare the first len(b) characters of a and b, case-sensitive"
    return a[:len(b)] == b

def strnicmp(a,b):
    """Compare the first len(b) characters
    of a and b, not case-sensitive"""
    return a[:len(b)].lower() == b.lower()

def strstr(a,b):
    "Return part of a that starts with substring b"
    x = a.find(b)
    if x >= 0: return a[x:]



From ilen712 at libero.it  Fri Feb  1 19:15:04 2002
From: ilen712 at libero.it (gimar)
Date: Sat, 02 Feb 2002 00:15:04 GMT
Subject: Error opening file ! Help, please
Message-ID: <Xns91A9CA356801ilen712l@193.70.192.103>

Sometimes when I try to create an output file executing:

f=open('\paul','w')

I receive the following error message:

"TypeError: object of type 'list' is not callable".


This has no sense for me. Can you explain me what's happening ?




Thank you


Gpaul




From Lutz.Schroeer at tu-clausthal.de  Wed Feb 20 09:13:43 2002
From: Lutz.Schroeer at tu-clausthal.de (Lutz Schroeer)
Date: 20 Feb 2002 14:13:43 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <G5Bc8.1$Az6.364@news.hananet.net> <3C7368A0.5050702@bath.ac.uk> <KzNc8.17085$Zp2.1759804@atlpnn01.usenetserver.com>
Message-ID: <a50aun$32h$1@ariadne.rz.tu-clausthal.de>

Steve Holden <sholden at holdenweb.com> wrote:

: Indeed, but it needs to be a little more rigorous. That's why a recent
: thread concluded you EITHER pass session id through all URLs OR use a cookie
: to identify the session.

Seems I missed that thread. After thinking it over I came to the conclusion
that using cookies is the most practical way. It's a pity, I hoped there
would have been a different solution, but passing an ID through all my URL
doesn't really attract me ...or maybe... mmh, I really don't know. I think I
try the URL-ID method and if I get completely messed up I use the cookies.

Thanks to all who answered
Lutz


From ponderor at lycos.com  Thu Feb 28 03:03:19 2002
From: ponderor at lycos.com (Dean Goodmanson)
Date: 28 Feb 2002 00:03:19 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers? .>> What's a neophyte?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <3C682A16.1D6C800D@ccvcorp.com> <Xns91B28C2B6EBE0RASXnewsDFE1@130.133.1.4> <6qk7tjm1mg.fsf@abnoba.intevation.de>
Message-ID: <e81be8b2.0202280003.538e6b8f@posting.google.com>

...this seems to be a decent place to interject...

I consider myself a Python Neophyte.  (I prefer "Armchair Pythonista"
but that carries slighty too much negative connotation.)

Would "Neo-Phyt-honista" be an appropriate title?

(-'s added for pronounciation. I'm guessing the -'s need to be removed
or replaced with ` or ''s.)

Happy Friday,

-Dean
http://nomad.freezope.org


From matthias_j at lycos.com  Mon Feb 11 12:36:22 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 11 Feb 2002 09:36:22 -0800
Subject: Help with Gordon McMillan Installer
References: <d7d5ebdf.0202091152.40a6e0af@posting.google.com> <Xns91B167C6A6F08gmcmhypernetcom@199.171.54.216> <d7d5ebdf.0202101122.19931f2a@posting.google.com> <Xns91B25FCC1C7C3gmcmhypernetcom@199.171.54.214>
Message-ID: <d7d5ebdf.0202110936.107b1a33@posting.google.com>

> You might want to think seriously about what benefit --onefile
> is giving you with an app this complex. A --onedir will be much 
> easier to set up, and much faster to start.
> 
> -- Gordon
> http://www.mcmillan-inc.com/

Thank's for anwsering so quickly;

I think you are right - it might be not to much benefit - on the oder
hand I like the idea of having everything cleaned up after the app
finished.

Thank's for this wonderful software - I will be using it proberly on
Win Nt and Linux
Matthias


From syring at email.com  Wed Feb  6 10:40:18 2002
From: syring at email.com (Karl M. Syring)
Date: Wed, 6 Feb 2002 16:40:18 +0100
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <mailman.1013003651.8981.python-list@python.org> <a3re97$1aa8h9$1@ID-98166.news.dfncis.de>
Message-ID: <a3rmm3$1aojpk$1@ID-7529.news.dfncis.de>

"MDK" <mdk at mdk.com> schrieb
> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
message
> news:mailman.1013003651.8981.python-list at python.org...
> > ----- Original Message -----
> > From: "MDK" <mdk at mdk.com>
> >
> >
> > > I need to convert a list into a list of characters.
> > >
> > > For example:
> > >
> > > ("cat",5,['dog',[3,3,1]],"zoo")
> >
> > That's not a "list", it's a tuple containing strings, ints, and lists;
> > further, you have nested lists.  Ouch.
> >
> > > Would become:
> > >
> > > ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
> > >
> > > Any information would be appreciated.
> >
> > This is doable, but there is no Python builtin I am aware of that can
> > do it.  Why do you need to do this?  If we understand what you are
trying
> > to do, we may be able to steer you away from this mess.
> >
> >
> P.S   Sorry, I used the wrong symbols.  I meant
> ["cat",5,['dog',[3,3,1]],"zoo"].

# Does not matter:
import pickle
ob = ["cat",5,['dog',[3,3,1]],"zoo"]
pickle.dumps(ob)

Flat enough?
Karl M. Syring




From jh at comunit.de  Thu Feb 28 07:58:55 2002
From: jh at comunit.de (Janko)
Date: Thu, 28 Feb 2002 13:58:55 +0100
Subject: writing efficiently a numpy array
References: <MPG.16e755abd178b85e989682@news.wanadoo.fr>
Message-ID: <20020228135855.18f768cc.jh@comunit.de>

On Wed, 27 Feb 2002 20:45:02 +0100
stephan.schumacher at andra.fr (Stephan) wrote:

Look for the TableIO module at

http://php.iupui.edu/~mmiller3/python/

or look into scipy, which also has a module in the io subpackage.

__Janko


From spamneggs at mail.utexas.edu  Wed Feb 27 15:10:48 2002
From: spamneggs at mail.utexas.edu (Kareem Badr)
Date: Wed, 27 Feb 2002 14:10:48 -0600
Subject: creating directory heairarchy in a zipfile archive
Message-ID: <3C7D3D48.5B8BC97C@mail.utexas.edu>

Does anyone know how to create file heirarchies within a zipfile object?
I've tried prepending the name of the directory to the archive name of
the file, but that doesn't work. Any tips would be greatly appreciated.
Thanks. 

-Kareem Badr


From starx at pacbell.net  Wed Feb  6 04:28:39 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 09:28:39 GMT
Subject: ANN: HappyDoc 2.0.1
References: <mailman.1012913477.7851.python-list@python.org> <a3pqbh$nru$2@peabody.colorado.edu>
Message-ID: <Xns91ADF0977CADRASXnewsDFE1@130.133.1.4>

Fernando P?rez || Tue 05 Feb 2002 08:36:09a:

> Doug Hellmann wrote:
> 
>> Announcing the latest version of HappyDoc, a Python documentation
>> extraction tool 

And the pesants rejoiced:
Yeeeeey.
[mightn'magic 4 or jonny bravo, take your pick :]

> An honest, non-trolling question: considering the amount of overlap
> that happydoc and pydoc seem to have, and that pydoc is included by
> default with python, what are the chances of merging the two?

Hmm.. i can't really remeber what pydoc's html output looked like (it does 
do that, right? :), but I liked happy's a lot better when i compared them, 
and i think it had a better customization framework...
 
> It seems to make sense to me. Pydoc provides the help() function for
> the prompt and the nice 'pydoc -g' server on localhost:7464 (I have no
> idea if happydoc offers similar services). But it tends to choke on
> tricky imports and produce empty documentation pages way too often for
> my taste. So maybe merging the best of the two into something
> distributed by default would be a good idea.
> 
> I simply think that whatever comes by default with the python distro
> tends to get vastly more use/exposure, so the best tool should go
> there. And since happydoc seems to address at least one of pydoc's
> problems (failing imports in certain cases), it would be in the best
> open source spirit to produce a better tool out of the best features of
> these two, IMHO. 

Well, pydoc provides internal, interactive help, and happy seems to be 
geared toward producing static documents (of which i think it does a better 
job)... but I don't see any moral reason why happydoc's html output system 
and better introspection (and whatever else it does good) couldn't be 
merged into pydoc. :\

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 00:42:08 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 23:42:08 -0600
Subject: Random Module help
References: <20020210215559.13516.00000455@mb-cq.aol.com>
Message-ID: <016b01c1b2be$db0d5920$0101010a@local>

----- Original Message ----- 
From: "Patio87" <patio87 at aol.com>


> I was wondering if any 1 could tell me how to use the random module 
> for a blackjack game i want to make. I am a begginer so try to be 
> easy with the dialouge. Thanks

Hmm.  First, I suppose you want your cards in a list:

    deck = [ (11, "Ace of Spades"),
             (10, "King of Spades"),

and so on.  Note that I am representing the cards as tuples of 
(value, name) where of course you will have to deal with the dual
values of the Aces in your code.

In this case, you might use random.shuffle() to shuffle the
deck:

    random.shuffle(deck)

although if you want to play multiple games, you will need to play
from a *copy* of the deck:

    current_deck =  master_deck[:]

    random.shuffle(current_deck)

something like that.  Draw cards using the .pop() method:

    card = current_deck.pop()

This puts a card tuple into the 'card' variable and removes it from
the current_deck.

Have fun.





From eric.brunel at pragmadev.com  Tue Feb 12 05:03:19 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 12 Feb 2002 10:03:19 +0000
Subject: Is it possible to call a function from a Tkinter widget.bind and include args???
References: <a4995n$gbg$1@news5.svr.pol.co.uk>
Message-ID: <a4alhq$6tp$1@norfair.nerim.net>

Hi,

G. Willoughby wrote:

> Is it possible to call a function from a Tkinter widget.bind and include
> args???
> i.e. i want to do something like this:
> 
> def printButtonAndCoords(event, button):
>     print "you rolled over %s and coords are %d:%d" % button, event.x,
> eventy
> 
> Button1.bind("<Enter>", printButtonAndCoords("button1"))
> Button2.bind("<Enter>", printButtonAndCoords("button2"))
> 
> is this possible???

Since you didn't mention your Python version, maybe the lambda stuff 
provided by Martin and Fredrik won't work (with pre-2.0 versions, the 
lambda may not have the printButtonAndCoords function in its namespace, 
isn't it?).
So here is another solution I used quite a lot. with pre-2.0 Python It 
involves the creation of a callback class:

class GenericCallback:
  def __init__(self, callback, *firstArgs):
    self.__callback = callback
    self.__firstArgs = firstArgs
  def __call__(self, *lastArgs):
    apply(self.__callback, self.__firstArgs + lastArgs)

Then you make your binding like follows:

Button1.bind("<Enter>", GenericCallback(printButtonAndCoords, "button1"))

Your printButtonAndCoords function should then take the string as first 
argument and the Tkinter event as second. The binding will "call" the 
instance of GenericCallback, i.e. call its __call__ method, which will in 
turn call printButtonAndCoords with the arguments provided in the 
constructor, then in the actual call.

The GenericCallback class is also generic enough to have many other uses 
(in fact everywhere you must provide a function).

HTH
 - eric -



From max at alcyone.com  Sat Feb 16 18:35:08 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 16 Feb 2002 15:35:08 -0800
Subject: "implemented by subclass"
References: <slrna6t9de.nt2.philh@comuno.freeserve.co.uk>
Message-ID: <3C6EECAC.A71B3B4B@alcyone.com>

phil hunt wrote:

> Consider the following python code:
> 
> class Log:
>    def write(self, text):
>       """(implemented by subclass)"""
> 
> Is this the most pythonic way to do it? If not, what would be
> better? How about rasing a ShouldBeImplementedBySublass exception?
> 
> How do other people do it?

I use raise NotImplementedError, and also typically put a test in the
constructor:

	class Log:
	    def __init__(self):
	        if self.__class__ is Log:
	            raise NotImplementedError
	        ...

	    def write(self, text):
	        raise NotImplementedError

The one drawback I know of here is that this can have a somewhat
(naively) unexpected effect if you're using mutiple inheritance; in the
case where you are deriving from to classes, one of which implements a
method and the other does not (with the raise NotImplementedError stub
above), it's best to have the new class (derived from both) explicitly
invoke the right method in its parent class, otherwise the lookup
mechanism may find the first method which has a definition, but which is
in fact the stubbed-out NotImplementedError method as above!

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From jeff at ccvcorp.com  Wed Feb  6 12:48:10 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 06 Feb 2002 09:48:10 -0800
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net>
Message-ID: <3C616C5A.4D1BBE1C@ccvcorp.com>


Timothy Rue wrote:


> Programming is two deminsional, not three.

This, sadly, is where you're mistaken, Tim.  Actually, programming is often
4-dimensional, and can sometimes even be higher-dimensional than that.  I
routinely deal with the issues involved in a 3-dimensional database changing over
time (giving a 4th dimension) -- and I consider myself to be a novice.  (Actually,
I think describing the database as being 3-dimensional is underestimating its
complexity, but...)  We won't even get into how often programming is used to solve
problems of fractal dimension...

We understand that you want to make programming easier, so that anyone can do it.
Almost everyone here, professional programmer or no, thinks that that is a great
goal;  if we didn't want basic tasks to be automated and simplified, we'd be using
C and Assembler, not Python.

The problem you're running into, is that you've failed to convince anyone that
*you* understand the real problems involved, enough to be able to solve them.  You
don't need to convince us of the benefits of autocoding; you need to convince us
of *your* competence.  So far, you're failing miserably.

Jeff Shannon
Technician/Programmer
Credit International






From boris at cantal.lip6.fr  Thu Feb 14 06:11:08 2002
From: boris at cantal.lip6.fr (Boris Boutillier)
Date: Thu, 14 Feb 2002 12:11:08 +0100
Subject: Is there a reverse operator to the '%' operator on string ?
Message-ID: <20020214.121107.1335354340.24152@cantal.lip6.fr>

Hi all,

I use a format string and a dictionnary to create a name as follow :
>Format = 'w%(width)dn%(nWord)d_%(fileName)s'
>NameDict = {'width' : 16, 'nWord' : 12, 'fileName' : 'Toto'}

>Name = Format % NameDict   # This gives "w16n12_Toto"

Now  what I'd like to do is revert the '%' operator, I mean I want to get
the dictionary from the name and the Format something like :

>NameDict = Format  op  Name .. = 'w%(width)dn%(nWord)d_%(fileName)s' op
"w16n12_Toto" which will give {'width':16,'nWord':12,'fileName': 'Toto'}

Is there such an operator, function or simple way to do this ? Something
like the 'scanf' C function ?


Boris Boutillier


From magnus at thinkware.se  Wed Feb 20 09:49:11 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 06:49:11 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <5d283efe.0202162142.53e3b402@posting.google.com> <roy-05958C.07303117022002@reader2.panix.com> <ubsenehjm.fsf@online.no> <a4psgi$5jo$3@peabody.colorado.edu>
Message-ID: <258fd9b8.0202200649.7dcfcb9c@posting.google.com>

Fernando P?rez <fperez528 at yahoo.com> wrote in message news:<a4psgi$5jo$3 at peabody.colorado.edu>...
> Syver Enstad wrote:
> 
> > mlh at vier.idi.ntnu.no (Magnus Lie Hetland) writes:
> >> 
> >> Jorg: Du kan l?ne denne av meg :)
> >> 
> > 
> > Begynner det ? bli s? mange norskinger p? comp.lang.python at vi kan
> > begynne ? skrive postingene v?re p? norsk ;-)

2? 3?

> Would you folks mind sticking to English in public discussions? Use private 
> email for other languages, please.

I won't go into details but I can tell you that the
above text was a subtle attempt by the Norse to take
over the news group. Phew, am I glad that you stopped
that! It's a long time since they tried to take over 
the south pole, and now they need something else... ;-)

We'd better watch out for the Norwegian Blue!


From martin at v.loewis.de  Sun Feb 24 17:25:53 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 23:25:53 +0100
Subject: Silly link question: python22_d.dll
References: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com>
Message-ID: <m31yfaqzym.fsf@mira.informatik.hu-berlin.de>

Courageous <jkraska at san.rr.com> writes:

> I have this problem I've been getting around the hard way.
> In VC6.0, when I have a program which talks to Python C/API
> functions, and my program is compiled in Debug configuration,
> it expects to find the python22_d.dll. I don't always want
> or need this, however, and was wondering how's it's possible
> to just accept python22.dll as part of my build?

It is possible, but it requires changes to the Python source code.

In PC/pyconfig.h, modify

#ifdef _DEBUG
#pragma comment(lib,"python23_d.lib")
#else
#pragma comment(lib,"python23.lib")
#endif

appropriately. Alternatively, consider whether not defining _DEBUG in
your Debug configurations may be acceptable.

> Furthermore, why the heck doesn't the binary release include
> python22_d.dll???

For size reasons. You'll not only need to include python22_d.dll, but
also the _d.pyd versions of all modules, together with the
corresponding import libraries.

Regards,
Martin


From gh_pythonlist at gmx.de  Sat Feb  2 21:01:32 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Sun, 3 Feb 2002 03:01:32 +0100
Subject: Windows CE vs PalmOS version
In-Reply-To: <20020203005036.44013.qmail@web20904.mail.yahoo.com>
References: <20020203005036.44013.qmail@web20904.mail.yahoo.com>
Message-ID: <20020203020131.GC1251@lilith.hqd-internal>

Le 02/02/02 ? 16:50, Lance Ellinghaus ?crivit:
> Hello everyone...
> I need to write a program in Python that requires a GUI and to run on a
> hand held device. This would be very easy if it was a desktop
> application, but it is not.
> 
> So the question is:
>     Should I purchase a Windows CE or Palm compatible device and run
> Python on it?
> 
>     What GUI interfaces are available for the Windows CE environment?

I don't know about the maturity of Pippy or the WinCE port at
http://www.murkworks.com/Research/Python/PocketPCPython/Overview - From
what I read there, the WinCE port is still not finished, and I have no
idea if it has bindings to a GUI library, yet. If it only has wrappers
for the WinCE API calls, that's probably not an exceptionally easy way
to program a GUI.

I suggest you consider using a Linux-powered device, like a Sharp Zaurus
(the developer edition comes with Linux, Qtopia/Qt Embedded
preinstalled). The PyQt maintainer recently released a PyQt version for
Qt embedded and the Zaurus is what he uses, so you can also get PyQt
binaries. Plus, you have your app running a real operating system, which
can runs almost everthing that runs on a "normal" x86 Linux.

You can also make this all work on a Compaq iPAQ, but the Zaurus is IMHO
the better option nowadays (reference platform at Trolltech, for
example).

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.8 ?C      Wind: 0.9 m/s



From cfelling at iae.nl  Thu Feb 28 16:19:29 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 22:19:29 +0100
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com> <mailman.1014861466.23611.python-list@python.org> <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com>
Message-ID: <a5m6t1$96p$1@animus.fel.iae.nl>

Huaiyu Zhu <huaiyu at gauss.almadan.ibm.com> wrote:

> PS. The test and result:

Just for kicks I added a true iteritems, and it wins hands down:)
it's lean (no temporary list at all), it is fast and it stays readable.

> import operator
> from time import time
...
   def iteriteritems(f, d):
 	for k, v in d.iteritems():
 		d[k] = f(v)
...
 	timeit(iteriteritems, f, d)
> 	timeit(iteritems, f, d)
> 	timeit(mapvalue, f, d)

> # result
> 	
> 19.8180559874
> 4.17512202263

And my result on my old and crusty machine:

7.69514203072
36.7980870008
9.78200399876
-- 
groetjes, carel


From kirill_simonov at mail.ru  Sun Feb 24 20:21:37 2002
From: kirill_simonov at mail.ru (Kirill Simonov)
Date: Mon, 25 Feb 2002 03:21:37 +0200
Subject: Tkinter + Entry + cp1251 cyrillic
In-Reply-To: <m3eljb9tcv.fsf@mira.informatik.hu-berlin.de>; from martin@v.loewis.de on Sun, Feb 24, 2002 at 09:28:16AM +0100
References: <26428104.0202231657.639cec52@posting.google.com> <m3eljb9tcv.fsf@mira.informatik.hu-berlin.de>
Message-ID: <20020225032137.A712@abulafia.local>

On Sun, Feb 24, 2002 at 09:28:16AM +0100, Martin v. Loewis wrote:
> redskin at tut.by (Andrey Sidorenko) writes:
> 
> > I use Active Python 2.2 on Windows 2000 and have trouble - "Entry"
> > class can't display cyrillic letters (cp1251 encoding) correctly in
> > entry field. Can anybody help me with this weird problem?
> 
> Can you be more specific? I have no problems getting Tkinter to
> display Cyrillic characters, using Unicode strings.

Tk doesn't allow to enter Cyrillic characters in the Entry and Text widgets
if it doesn't load its encodings at startup. But Tk displays Cyrillic letters
correctly if I set the contents of the entry from my program.


Kirill




From threeseas at earthlink.net  Tue Feb  5 23:54:25 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:25 GMT
Subject: Autocoding project proposal.
References: <83it9mghng.fsf@panacea.canonical.org> <B87C2640.3CC2%jonathan@onegoodidea.com> <slrna5d2cq.g57.philh@comuno.freeserve.co.uk> <1341.799T18T11762852threeseas@earthlink.net> <slrna5rpoa.19d.philh@comuno.freeserve.co.uk>
Message-ID: <988.801T1700T14365170threeseas@earthlink.net>

On 03-Feb-02 20:42:34 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Mon, 04 Feb 2002 00:01:03 GMT, Timothy Rue <threeseas at earthlink.net>

>>Where else are you going to take a shot at trying to take down the
>>ultimate cost lower tool. The one that removes the need for putting up
>>with people like you.

>I have no idea what you are trying to say.

>--
>===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
>Herbivore, a zero-effort email encryption system. Details at:
><http://www.vision25.demon.co.uk/oss/herbivore/intro.html>


Good, than you won't know it hit you untill it is to late, for you.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From lac at strakt.com  Tue Feb  5 05:26:31 2002
From: lac at strakt.com (Laura Creighton)
Date: Tue, 05 Feb 2002 11:26:31 +0100
Subject: Get current cursor position in ScrolledText 
In-Reply-To: Message from andrea@proceryon.at (Andrea) 
   of "04 Feb 2002 23:04:46 PST." <4feefc55.0202042304.62cc5cc0@posting.google.com> 
References: <4feefc55.0202042304.62cc5cc0@posting.google.com> 
Message-ID: <200202051026.g15AQVGo013483@ratthing-b246.strakt.com>

Key is a synonym for KeyPress.  There is also a KeyRelease.

Laura Creighton



From idot at vt.edu  Sat Feb 16 00:26:33 2002
From: idot at vt.edu (Carl Banks)
Date: Sat, 16 Feb 2002 00:26:33 -0500
Subject: line break confusion
References: <XFMail.20020215193042.shalehperry@attbi.com> <mailman.1013835745.20583.python-list@python.org>
Message-ID: <9iqk4a.nj.ln@127.0.0.1>

Michael Mell wrote:
> Sean 'Shaleh' Perry wrote:
> 
>> Windows: \r\n
>> Linux (UNIX): \n
>> Mac: \r
>>
>> Each OS's libs transform the sequence '\n' to whatever the right
>> thing is for them.
>
> Thanks for answering such an apparently ignorant question, but the
> answer is not so simple.
> 
> I want to read and write files with line breaks which are not of the
> native platform type.
> 

> Problem #1 is that on a Mac, the values of Mac and Unix are
> /swapped/ and therefore wrong. Writing a file using '\n' results in
> a /Mac/ line break (as reported by BBEdit, the venerable Mac
> editor). Writing a file using '\r' results in a /Unix/ file. The Mac
> seems unable to write Windows \r\n no matter what order is used.


You need to open the file in binary mode:

open(file,"rb")
open(file,"wb")

By default, Python opens the file in text mode, which helpfully
converts your line feeds into whatever the convention of your platform
is.  Helpfully, that is, unless you want to ignore your platform's
conventions and choose your own line delimiters.


> Problem #2 is that when reading a Windows file on Linux or Mac, a
> /single/ line ('\r\n') is interpreted as /two/ lines ('\n\n'). Files
> written out on Linux then have double the correct number of lines.

It can be worked around.  The characters won't be converted in binary
mode, so you can see whether the line ends in '\n' or '\r' and throw
out the superfluous empty '\r' line.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From martin at v.loewis.de  Mon Feb 11 08:33:50 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 11 Feb 2002 14:33:50 +0100
Subject: REQ : encoding windows cp1252 => iso latin 1
References: <a3p68d$1a56$1@norfair.nerim.net>
Message-ID: <m37kpkf8ht.fsf@mira.informatik.hu-berlin.de>

"Gillou" <nospam at bigfoot.com> writes:

> My customers make copy/paste from M$ word docs to forms translated to XML
> (expecting ISO latin 1 charset).
> My XML parser (pyexpat) does not accept cp1252 character, and I'm looking
> for a function that can translate extra cp1252 characters to the closest ISO
> latin 1 encoding.

Depending on the exact version of software you are using, this is a
bug: pyexpat should accept cp1252 as an encoding name. Please report a
bug (including precise test case) to SF.

Regards,
Martin



From cfelling at iae.nl  Thu Feb 14 10:41:14 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 14 Feb 2002 16:41:14 +0100
Subject: Is there a reverse operator to the '%' operator on string ?
References: <20020214.121107.1335354340.24152@cantal.lip6.fr>
Message-ID: <a4glqq$2oj$1@animus.fel.iae.nl>

Boris Boutillier <boris at cantal.lip6.fr> wrote:
...
> Now  what I'd like to do is revert the '%' operator, I mean I want to get
> the dictionary from the name and the Format something like :

>>NameDict = Format  op  Name .. = 'w%(width)dn%(nWord)d_%(fileName)s' op
> "w16n12_Toto" which will give {'width':16,'nWord':12,'fileName': 'Toto'}

> Is there such an operator, function or simple way to do this ? Something
> like the 'scanf' C function ?

What about regex's, like:

>>> import re
>>> matcher = re.compile(r"^w(\d+)n(\d+)_(.*)$").match
>>> matcher("w16n12_Toto").groups()
('16', '12', 'Toto')

-- 
groetjes, carel


From aahz at panix.com  Wed Feb 27 00:28:59 2002
From: aahz at panix.com (Aahz Maruch)
Date: 26 Feb 2002 21:28:59 -0800
Subject: Threads in python?
References: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com>
Message-ID: <a5hqqr$bcl$1@panix2.panix.com>

In article <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5 at 4ax.com>,
Fernando Rodr?guez  <frr at ThouShallNotSpam.EasyJob.NET> wrote:
>
>I'm going to start coding a little utility that needs to retrieve
>webpages and process them.
>
>Does Python support threads? "Learning Python" doesn't mention it...

http://starship.python.net/crew/aahz/
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From threeseas at earthlink.net  Sun Feb 24 20:45:48 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 25 Feb 2002 01:45:48 GMT
Subject: Parse variable into variables >Re: Apology
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com>
Message-ID: <2240.820T2407T12845765threeseas@earthlink.net>

Due to my none computer/programming work load I haven't had the time to
look thru the four O'Reily Python programming books and don't spend
enough time programming with python to remember such things that are bound
to be simple, but here is the question:

I can read a line from a file but then I want to parse the line into a set
of variables. How do I do the parse?

If I have to do this a baby step at a time then so be it.


On a related but different note:

Perhaps this will shead some light on the lack of logic and comprehension
of the following person, not to mention his injection of false claims and
insinuations.... I've asked for help, he didn't have to respond but chose
to in the manner below.

On 18-Feb-02 17:35:47 Joshua Muskovitz <joshm at taconic.net> wrote:
>The Ruebot wrote:
>> If you want to be greedy Phil, to take that job away from me, then give me
>> the tool under the conditions of which I want it.

>So do it already!  You complain that programmers are keeping the keys to the
>kingdom to themselves, and your system will blow those doors open.  If your
>truly amazing design is really able to do that, then why haven't *you*
>empowered yourself and just done it?  Instead, you whine and insult, in
>order to try to convince "us" that we should do it for you.  Why should we?
>Clearly, it doesn't benefit us.  You'll never convince us to give up our
>power.  So instead of harassing us, do it yourself, and then that will
>really show us.  Then you'll have the last laugh.  Otherwise, we'll just
>keep having *all* the laughs.  Especially when we read your posts over
>brandies at the weekly PSU meetings, while we light our cigars with $100
>bills, and chuckle about our complete control over the unsuspecting world.
>Mwahahahahaha.


http://groups.google.com/groups?hl=en&selm=5098.813T2333T10665859threeseas%40earthlink.net&rnum=10



---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From aahz at panix.com  Fri Feb  8 14:13:30 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 11:13:30 -0800
Subject: Is it an iterator? (was Re: FixedPoint)
References: <mailman.1013119093.20033.python-list@python.org> <GJS88.49612$3J1.1396553@news2.tin.it>
Message-ID: <a4180q$gvu$1@panix2.panix.com>

In article <GJS88.49612$3J1.1396553 at news2.tin.it>,
Alex Martelli  <aleax at aleax.it> wrote:
>
>"something is iterable" if and only if you can iterate on it.  So, try 
>iterating on it, and handle the resulting exception should it turn out
>you can't.  I'm not sure exactly what exceptions would be raised by
>various versions of Python -- probably a TypeError, but one never
>knows -- better to experiment on all relevant versions, or maybe, just
>for once, a generic try/except is acceptable.  An almost acceptable
>version might therefore be:
>
>def isIterable(x):
>    try:
>        for i in x: return 1   # iterable
>        else: return 1         # iterable, although empty
>    except:
>        return 0
>
>Unfortunately, if we're in Python 2.2 and x is an iterator, isIterable
>"consumes" x's first item.  This isn't hard to fix:
>
>def isIterable(x):
>    try: x=iter(x)
>    except: pass
>    else: return 1
>    try:
>        for i in x: return 1   # iterable
>        else: return 1         # iterable, although empty
>    except:
>        return 0

This still breaks with files, I believe (don't have 2.2 handy to test).
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From feldspar at ix.netcom.com  Mon Feb  4 18:16:08 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Mon, 04 Feb 2002 19:16:08 -0400
Subject: [Fwd: directly executing a python script in Unix]
Message-ID: <3C5F1638.1D9B527C@ix.netcom.com>

Johannes accidentally replied to me rather than to the newsgroup, so:

Johannes Nix wrote:
> 
> Antaeus Feldspar <feldspar at ix.netcom.com> writes:
> 
> > > "./myscript.py".  However, when I try this, I get an error "Command
> > > not found.".  I get the same error under both Linux and Solaris.
> 
> > Hmmmm.  I tried a few things, none of which reproduced your error:
> >
> > * trying to run the script without execute permission
> > * running without any "if __name__ == '__main__':" code
> > * subtracting/adding spaces between the "#!" token and the path
> 
> I've read once that Linux doesnt finds she-bang command scripts if
> they end with something different from a newline - for example a
> carriage return plut a newline (MS Windows / DOS convention). So it's
> possible to write such a script under DOS, which is visually identical
> but won't run under Linux.
> 
> jnix at zosma:/mnt/newhome/jnix > od -t a s.py
> 0000000   #   !   /   u   s   r   /   b   i   n   /   p   y   t   h   o
> 0000020   n  cr  nl   p   r   i   n   t  sp   1  nl
> 0000033
> jnix at zosma:/mnt/newhome/jnix > chmod u+x s.py
> jnix at zosma:/mnt/newhome/jnix > ./s.py
> bash: ./s.py: No such file or directory
> jnix at zosma:/mnt/newhome/jnix >
> 
> Johannes


From barry at zope.com  Fri Feb  8 19:44:33 2002
From: barry at zope.com (Barry A. Warsaw)
Date: Fri, 8 Feb 2002 19:44:33 -0500
Subject: RELEASED - StandaloneZODB 1.0 final
Message-ID: <15460.28913.823952.784640@anthem.wooz.org>

I'm please to announce the release of StandaloneZODB, the Python
object persistency system also known as the Z Object Database.  ZODB
is the object-oriented database underlying Zope; the StandaloneZODB
project's goal is to provide those same facilities to non-Zope related
Python applications.

Today we are releasing StandaloneZODB 1.0 final.  A brief description
of the changes since release candidate 1 is outlined below.

StandaloneZODB is based on the same code as the ZODB in Zope, albeit
on a separate release branch.  Its inspiration comes from Andrew
Kuchling's StandaloneZODB project on SourceForge.  While there are
still some differences, the Standalone 1.0 release is the first on the
path toward convergence.  Subsequent releases should complete the
merge of Andrew's and Zope Corporation's packages.

The StandaloneZODB release includes the following components:

    - Core ZODB, including the persistence machinery
    - Standard storages such as FileStorage
    - Supporting modules such as ExtensionClass
    - The persistent BTrees modules
    - ZEO
    - Experimental Berkeley storages
    - Some documentation <wink>

See the README file for details on building and installing
StandaloneZODB.  For details on using ZODB, see Andrew's included user
guide.

StandaloneZODB 1.0 is released under the ZPL 2.0.  It should be
compatible with all Python versions from Python 2.1 to Python 2.2.  It
may or may not work with versions earlier than Python 2.1.

Download StandaloneZODB-1.0 from:

    http://www.zope.org/Products/StandaloneZODB

and visit the StandaloneZODB Wiki page at:

    http://www.zope.org/Wikis/ZODB/StandaloneZODB

See also:

    http://www.zope.org/Wikis/ZODB/FrontPage

for more information about our long-range ZODB plans.

Enjoy,
-Barry

Barry A. Warsaw
Zope Corporation, Pythonlabs
barry at zope.com

-------------------- snip snip --------------------
What's new in StandaloneZODB 1.0 final?
Release date: 08-Feb-2002
=======================================

All copyright notices have been updated to reflect the fact that the
ZPL 2.0 covers this release.

Added a cleanroom PersistentList.py implementation, which multiply
inherits from UserDict and Persistent.

Some improvements in setup.py and test.py for sites that don't have
the Berkeley libraries installed.

A new program, zeoup.py was added which simply verifies that a ZEO
server is reachable.  Also, a new program zeopack.py was added which
connects to a ZEO server and packs it.



From fperez at pizero.colorado.edu  Thu Feb 14 13:25:30 2002
From: fperez at pizero.colorado.edu (Fernando =?iso-8859-1?Q?P=E9rez?=)
Date: Thu, 14 Feb 2002 11:25:30 -0700
Subject: MemoryError using NumPy
References: <a4gqd9$68e5$1@ID-11957.news.dfncis.de>
Message-ID: <3C6C011A.7C5A7640@pizero.colorado.edu>

Take a look at weave at http://www.scipy.org/site_content/weave, in
particular the blitz() module may be what you need. It generates c++ for
your numeric calls, which in many cases eliminates the need for
temporaries. That's one of Numpy's weaknesses, that it generates too many
temps when doing things and with large data sets it becomes a hard problem.
weave.blitz() and weave.inline() offer ways out of that.

cheers,
f


From phd at phd.pp.ru  Wed Feb 13 06:52:50 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 13 Feb 2002 14:52:50 +0300
Subject: PocketPC
In-Reply-To: <98322540.0202130340.616e2be7@posting.google.com>; from alife@narod.ru on Wed, Feb 13, 2002 at 03:40:30AM -0800
References: <98322540.0202130340.616e2be7@posting.google.com>
Message-ID: <20020213145250.V20054@phd.pp.ru>

On Wed, Feb 13, 2002 at 03:40:30AM -0800, Serge Terekhoff wrote:
> Are there any Pythons or plans for PocketPC?

   Using Gogle to search for "python pocketpc", found:
http://starship.python.net/crew/skippy/ce/

PS. Why not to ask Russian-speaking Python and Zope User Group?

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



From hamish_lawson at yahoo.co.uk  Tue Feb 26 07:39:00 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 26 Feb 2002 04:39:00 -0800
Subject: Time to rename Stackless?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org>
Message-ID: <915a998f.0202260438.246be39c@posting.google.com>

Christian Tismer wrote:

> Well, it doesn't switch between multiple stacks, it
> repeatedly replaces parts of the same stack.
> (But this would have the same effect, of course).

If it essentiually has that effect, how about Switching Python then as a name?

Hamish Lawson


From martin at v.loewis.de  Wed Feb 20 01:58:05 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 07:58:05 +0100
Subject: Idiots guide to code objects, frame objects and settrace?
References: <VoAc8.7320$k91.367319@wards>
Message-ID: <m3d6z063nm.fsf@mira.informatik.hu-berlin.de>

"Tim Rowe" <tim at eltig.freeserve.co.uk> writes:

> Is there an idiots guide to code objects, frame objects and settrace
> anywhere?

Anybody using those would look like a very smart idiot; so no.  Or,
perhaps, the Python source code would be an acceptable guide to such
people (it certainly is to me).

If you have specific questions, don't hesitate to ask.

Regards,
Martin



From phr-n2002a at nightsong.com  Thu Feb 21 16:36:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 13:36:04 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <7xg03vsidi.fsf@ruckus.brouhaha.com> <e9e6dd1f.0202211235.4e7be9aa@posting.google.com>
Message-ID: <7xzo22o6uz.fsf@ruckus.brouhaha.com>

caljason76 at yahoo.com (Jason) writes:
> > In Python, that's considered immoral--it puts you in danger of
> > forgetting that the sort operation modifies the list that you use it
> > on.  For example, if you say
> > 
> >    a = b.sort()
> > 
> > expecting to get a sorted copy of b without clobbering b, you'd get
> > a rude surprise.
> 
> I still think that the convenience of having sort() return the list
> far outweighs the cost.  People would only get bit once, maybe twice,
> before they learned the behavior of sort.

I tend to agree with you on this, though I'd have preferred that sort
return the list but be called nsort instead of sort, like in Common Lisp.

> I like Python, but the biggest problem I have with the language right
> now is what low code density it has at times.

Agreed on that too.


From jkraska at san.rr.com  Fri Feb 22 23:37:53 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 23 Feb 2002 04:37:53 GMT
Subject: Stackless Platform Independence?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com>
Message-ID: <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com>

>I am personally still surprised that stackless has not be considered more
>seriously in the python-dev group, especially in regard to the threading
>problems and limitations of the current python versions.

Well, keep in mind that stackless threads aren't OS threads at all.

The way I have it, Stackless _was_ considered by Guido and team,
it was just regarded as both too dirivative and too complicated to
make part of core Python. Look out in the real world and try to
find out how many languages have continuation support in them.
Not very many.

I personally make great use of continuations in a simulator core
I wrote. While I grok them, I have the feeling that an average
programmer following in my footsteps would have to read my
comments very very carefully before he could come to understand
how a certain central function call in my simulator actually
returns millions of times.

That's a real mind bender. :-)

C//



From quinn at spew.ugcs.caltech.edu  Wed Feb  6 03:47:42 2002
From: quinn at spew.ugcs.caltech.edu (Quinn Dunkan)
Date: 6 Feb 2002 08:47:42 GMT
Subject: Autocoding project proposal.
References: <83it9mghng.fsf@panacea.canonical.org> <B87C2640.3CC2%jonathan@onegoodidea.com> <slrna5d2cq.g57.philh@comuno.freeserve.co.uk> <1341.799T18T11762852threeseas@earthlink.net> <slrna5rpoa.19d.philh@comuno.freeserve.co.uk> <988.801T1700T14365170threeseas@earthlink.net>
Message-ID: <slrna61rdd.ap.quinn@spew.ugcs.caltech.edu>

On Wed, 06 Feb 2002 04:54:25 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>Good, than you won't know it hit you untill it is to late, for you.

Sounds like the robot army's already on its way.


From cbbrowne at acm.org  Mon Feb 25 23:17:09 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Mon, 25 Feb 2002 23:17:09 -0500
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org>
Message-ID: <m3pu2sancq.fsf@chvatal.cbbrowne.com>

In an attempt to throw the authorities off his trail, Greg Ewing <greg at cosc.canterbury.ac.nz> transmitted:
> Christian Tismer <tismer at tismer.com>:
>
>> It is still stackless in the sense that order of execution is not
>> restricted by the C stack.
>
> Even so, I still think it's a poor name, because it
> focuses on an implementation detail rather than
> telling you what it can do for you.
>
>> We have now tasklets, proclets, and the stack slices are
>> stacklets. :)
>
> Hmmm... maybe it should be called Pythonlet...

If it gets to the point of supporting billions of threads, it would
_certainly_ make sense to call it "NanoThreading Python."  Or
"NanoTech Python."  :-)

There have got to be some puns heading in the direction of "seriously
stacked."  Hmm...

 - Pulchritudinous Python?
 - Busty Python?

There's got to be some more clever way of saying that without
actually, well, saying that...  "Victoria's Python?"  Hmm...
-- 
(concatenate 'string "cbbrowne" "@ntlug.org")
http://www.ntlug.org/~cbbrowne/nonrdbms.html
Walk softly and carry a megawatt laser.


From bkelley at wi.mit.edu  Fri Feb 22 09:44:41 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 22 Feb 2002 09:44:41 -0500
Subject: While loop and time.sleep function
References: <mailman.1014306035.24844.python-list@python.org>
Message-ID: <3C765959.3090609@wi.mit.edu>

Alves, Carlos Alberto - Coelce wrote:

> Hi all,
> I tray to make a simple clock program ( see the code below ).
> Can someone explain to me why code 01 works while code 02 doesn't
> 


You almost never want to use things like time.sleep in Tkinter apps 
since Tkinter is single threaded.  This means while time.sleep is 
sleeping, the gui window will not be updated.

It is much safer to use things like

root.after(milliseconds, function)

to set up timers like you want.  You might want to read Lundh's 
introduction to Tkinter at:

http://www.pythonware.com/library/tkinter/introduction/index.htm


Here is a Clock class that uses the .after timer

import time

class Clock:
     def __init__(self,root):
         self.root = root
         self.lb=Label(root,padx=5,pady=5,
                       fg='blue',font=('Times',20))
         self.lb.pack()
         self.root.after(1000, self.ck)
         self.ck()


     def ck(self):
         self.lb.configure(text=time.asctime()[11:19])
         # set up the next callback
         self.root.after(1000, self.ck)

if __name__ == "__main__":
     from Tkinter import *
     root = Tk()
     root.title('Clock')
     Clock(root)
     mainloop()


Brian Kelley



From BPettersen at NAREX.com  Mon Feb 18 13:11:26 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Mon, 18 Feb 2002 11:11:26 -0700
Subject: How do I view Pages on Palm
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158D1F@admin56.narex.com>

> From: jphammock [mailto:jphammock at email.msn.com] 
> 
> downloaded python but dont know how to use it to view web 
> pages on my palm any help.

You need to import the DWIW (Do-What-I-Want) module, e.g.:

  import DWIW
  DWIW.ShowWebPage("http://www.yahoo.com")

<wink>

-- bjorn

Ps: If you can give us a hint about why you thought you could use Python
to _view_ webpages on the Palm, then maybe we can steer you in a more
productive direction...



From denro at NSPAMearthlink.net  Thu Feb 28 16:23:24 2002
From: denro at NSPAMearthlink.net (Dennis Roark)
Date: Thu, 28 Feb 2002 21:23:24 GMT
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net> <20020227163411.71b4b4d8.alex@netWindows.org>
Message-ID: <Xns91C39C93F933Edenroearthlinknet@207.217.77.23>

Alex Russell <alex at netWindows.org> wrote in
news:20020227163411.71b4b4d8.alex at netWindows.org: 

> I'm glad they (apparently) got one right.
> 
> I was required to purchase their Java book for a class, and not only
> was it poorly laid out, poorly executed, badly typeset, and cheaply
> made, it's index was utterly useless. Their C++ book wasn't much
> better. A friend had it laying about his place day and I spent some
> time looking through it, and after comparing it to Lippman/Lajoie,
> it's clear that it makes better tinder than a programming reference.
> 
> Stear clear of those two if you're thinking about picking them up on
> the strength of their Python book.
> 

I am afraid I have to agree more with Alex's cautionary comments than 
with Ron's strong praise.  I know these authors have the most popular 
college C++ text out there, but it always seemed like a C book with 
Objects added on as an after thought.  I stay with other texts when I 
teach C++.  The Python book appears substantially based on the templates 
of their previous books.  IMHO, not the quality nor the enjoyment of 
Mark Lutz, Programming Python, 2nd ed., or even of the relatively short 
tutorial in Steve Holden's Python Web Programming.  That the D & D book 
came out in Python was more of an indication of Python going mainstream, 
and for that I celebrate it.

-- 
Dennis Roark

denro at earthlink.net
Starting Points:
www.home.earthlink.net/~denro


From mdk at mdk.com  Tue Feb  5 11:08:10 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 11:08:10 -0500
Subject: My own list object
Message-ID: <a3p01b$186eq8$1@ID-98166.news.dfncis.de>

Hello,

I am trying to create my own special type of list object.

However, when I do x.append('yo') it uses Python's append instead of the one
from my class.

How can I get it to use my append method?  I have tried def __append__ and
then called my own class' function but that did not work.  I've played with
__getattr__ but that did not help.

If this is not easy can someone at least point me to an example of what I
want to do?

Thank you.




From dsavitsk at e-coli.net  Wed Feb 27 20:32:14 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 28 Feb 2002 01:32:14 GMT
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org> <3C7D8AC4.3F9A74D0@engcorp.com>
Message-ID: <yMff8.10349$pC1.1875911@newssrv26.news.prodigy.com>

along this line, i'd like to see an advanced python programming on win32.

-d


"Peter Hansen" <peter at engcorp.com> wrote in message
news:3C7D8AC4.3F9A74D0 at engcorp.com...
> Susan Capparelle wrote:
> >
> > We're a publisher of high quality computer books currently seeking
> > authors for a number of Python related books.  A couple of titles or
> > topics would be; 'Enterprise system development with Python,'
> > 'Practical Python' and 'Effective Python.'  If you can recommend
> > anyone with the necessary experience and skills to undertake any of
> > these books, please feel free to contact us.
>
> Although I'm all warm and fuzzy about Python's continually
> expanding domain of influence, I'm a little uncertain what
> advantage more books would bring in the category of
> "general Python programming".  What is it about "Practical
> Python" and "Effective Python" which existing books don't
> cover adequately.
>
> And perhaps more interestingly, what is it about "practical"
> Python which makes it ineffective, and what about "effective"
> Python programs make them impractical?
>
> (Posted and emailed, in case you don't read the newsgroup.)
>
> -Peter




From store_li at sina.com  Mon Feb 25 22:49:26 2002
From: store_li at sina.com (Ricky)
Date: Tue, 26 Feb 2002 11:49:26 +0800
Subject: .exe files
References: <jpae8.66140$X64.23604369@news1.rdc2.pa.home.com>
Message-ID: <a5f0m7$6sd3g$1@ID-12869.news.dfncis.de>

You can use Py2Exe. While if you use it compile Tk program, it will include
in all Tk package in your final relese file. So may be you can't expect your
exe file will be tiny.

"Stephen Smith" <ssmith619 at hotmail.com> wrote in message
news:jpae8.66140$X64.23604369 at news1.rdc2.pa.home.com...
> I'm new to Python, and have created a number of small programs that users
> might find interesting, althought they aren't really geeky and the people
> that use them would probably not have the Python IDLE or anything even
> related to Python installed.  Is there anyway to turn a .py file into a
.exe
> file that can be executable on Windows?  It doesn't call from any other
> modules, it's completely stand alone.  Thank you for any information!
>
>




From peter.milliken at gtech.com  Sun Feb 17 16:37:02 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 18 Feb 2002 08:37:02 +1100
Subject: basic books/guides on multithreading programming?
References: <mailman.1013813246.12449.python-list@python.org>
Message-ID: <a4p7rm$akj3@news1.gtech.com>

G'Day Tripp,

Have a look at Concurrency In Ada by Burns and Wellings - I found it an
excellent book on tasking and related issues, of course it uses features of
the Ada language to explain how various issues can be solved (Ada has
tasking built into the language definition rather than as a library add-on -
guards on entry statements etc etc which are things you can't do with a
"library" :-)) but it explains generic tasking issues quite simply and then
presents the solution in terms of Ada - which is a very readable language.

Peter

"Tripp Scott" <tripps81 at yahoo.com> wrote in message
news:mailman.1013813246.12449.python-list at python.org...
> which ones do you recommend? examples in any language is fine,
> though i prefer python, perl, or ruby. what i would like is
> guidance on what is thread safety, how to achieve it, what are
> the issues in multithreading programming, the global interpreter
> lock, and other basics.
>
> t
>
>




From max at alcyone.com  Sun Feb 24 15:57:50 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Feb 2002 12:57:50 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com> <3C751FEA.24CEFBBC@ford.com> <slrna7i5v6.7kb.scarblac@pino.selwerd.nl>
Message-ID: <3C7953CE.8D430F71@alcyone.com>

Remco Gerlich wrote:

> No, he didn't. He thought the *function* should be false, not the
> function's
> result. An interesting idea, but possibly hard to implement (write
> code to
> decide if a function does nothing). Also, I haven't been able to come
> up
> with any situation where it would be useful yet :-)

It brings up the question of where you'd draw the line, as well.  Would
it just be functions whose body is a pass?  How about ones which just
explicitly return None (which have precisely the same effect when
called)?  What about those that do something local without side effect
and then return None or let control flow out?  Same thing again, but
you're starting to wander into unclear territory.

But, as you say, the biggest problem with the idea is its total lack of
utility :-).

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From usenet at thinkspot.net  Sat Feb 16 09:46:39 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 16 Feb 2002 06:46:39 -0800
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no>
Message-ID: <a4kv8g.3vva8ld.1@kserver.org>

On 16 Feb 2002 13:56:35 +0100, Jan-Fredrik Braseth
<janfbr at sex.ifi.uio.no> wrote in comp.lang.python in article
<xyuvgcxlh4s.fsf at sex.ifi.uio.no>:

>  but after some time I get this error message:
> 
> Traceback (most recent call last):
>   File "./quizbot.py", line 366, in ?
>     foo = QuizBot(name,password,channel)
>   File "./quizbot.py", line 28, in __init__
>     line = self.tn.read_until("\n").split()
>   File "/usr/lib/python2.2/telnetlib.py", line 297, in read_until
>     self.process_rawq()
>   File "/usr/lib/python2.2/telnetlib.py", line 424, in process_rawq
>     self.msg('IAC %d not recognized' % ord(opt))
> UnboundLocalError: local variable 'opt' referenced before assignment
> 
> The main loop of the bot is just a while(1) which reads a line at the
> time from the connection: line =self.tn.read_until("\n").split()
> and it looks like this fails after some time.
> 
> tn is initiated like this:
> 
> tn = telnetlib.Telnet(server,port)
> 
> I really hope someone can help me with this problem

You have an error in your code. Show us the code. I'm guessing, that you
have a variable 'opt' that under some circumstances is not defined
before entering a particular loop or something.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From nospam at bigfoot.com  Tue Feb 12 06:44:53 2002
From: nospam at bigfoot.com (Gillou)
Date: Tue, 12 Feb 2002 12:44:53 +0100
Subject: Shell command
References: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net> <mailman.1013507179.20449.python-list@python.org>
Message-ID: <a4auq9$k00$1@norfair.nerim.net>

Or have a look at the popen2 package if you want to get the output of the
command.

"William Park" <opengeometry at yahoo.ca> a ?crit dans le message news:
mailman.1013507179.20449.python-list at python.org...
> On Tue, Feb 12, 2002 at 10:35:21AM +0100, DelPiccolo Ivano wrote:
> > How can I do to launch a command in my python script ?
> > example : I want to launch the command : rm test.txt
>
> os.system('rm test.txt')
>
> --
> William Park, Open Geometry Consulting, <opengeometry at yahoo.ca>
> 8 CPU cluster, NAS, (Slackware) Linux, Python, LaTeX, Vim, Mutt, Tin
>




From gerson.kurz at t-online.de  Tue Feb 26 23:32:53 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Wed, 27 Feb 2002 04:32:53 GMT
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
Message-ID: <3c7c5eb3.159302234@news.isar.de>

On Tue, 26 Feb 2002 22:08:41 GMT, "Sandeep Gupta"
<sandeep182 at hotmail.com> wrote:

>I've written a product in Python and I wanted to find out how best to
>obfuscate the code.  I want the distributed files to be moderately hard to
>reverse engineer.

Well, "Python Obfuscation" is my middle name ;) For starters, there is

- the "How to write unmaintainable code" FAQ

http://mindprod.com/unmain.html

- and my very own "Obfuscated Code HOWTO"

http://www.p-nand-q.com/obtech.htm

which, alas, is a bit C/C++ specific.

Seriously though, I've been toying around with this idea: Recompile
python22.dll to use redirected file functions in the import module;
redirect the files to a database, so new versions can be managed
remotely. Rather than doing so, you could redirect them to encrypted
files.

I've not looked into this really seriously; maybe the "python import
hook" does the job already. See
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/59867 for an
example that changes imported files on-the-fly. 


From boud at rempt.xs4all.nl  Mon Feb 18 17:08:24 2002
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 18 Feb 2002 22:08:24 GMT
Subject: PyKDE
References: <u72ll0kis0p201@corp.supernews.com>
Message-ID: <a4ru0o$60l$1@news1.xs4all.nl>

Mik Mifflin <dogansmoobs at nospam.prexar.com> wrote:
> I'm trying to learn Qt programming and decided to use my new favorite 
> language (Python, of course).  I'm using SuSE 7.3, and there is an rpm 
> called PyQt installed, but it looks like I need PyKDE as well?  This is the 
> program that I'm trying to run, and the error messages it gives:

> !/usr/bin/python
> from qt import *
> from kdecore import *
> from kdeui import *

> a = KApplication (sys.argv)
> w = QWidget()
> a.setMainWidget (w)
> a.setTopWidget (w)
> w.resize (200, 200)
> w.show()
> a.exec_loop()

>> ./qt1
>   File "./qt1", line 3, in ?
>     from kdecore import *
> ImportError: No module named kdecore

> If I do need PyKDE, does anyone know of an rpm that will work with SuSE?  
> Or am I just better off installing from source?

Yes -- to use KDE modules, you need PyKDE. Now there's some complication,
since KDE is quite complex, and it's only very recently that Jim Bublitz
has succeeded in binding most of KDE to Python. You can find his work at:
http://www.thekompany.com/projects/pyqt. As far as I know, there are no
rpm's of PyKDE available.

This might be a good moment to officially note that my PyKDE tutorial
only works with KDE 1.1.2, and the attendant PyKDE... Which means that
it isn't all that useful. From your code it looks like you're using
my stuff...

But my book on PyQt, from http://www.opendocs.org, has just come back
from the printer, which means that orders will be quickly filled. It
covers PyQt for Qt 2.x and 3.x. PyKDE for KDE 2.x wasn't done when
I was writing the book.

-- 

Boudewijn Rempt  | http://www.valdyas.org 


From Simon.Budig at unix-ag.org  Fri Feb 15 05:38:07 2002
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 15 Feb 2002 11:38:07 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de>   <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de> <u8za6pskh.fsf@python.net> <3c61abe7$1@si-nic.hrz.uni-siegen.de> <uu1sta7s2.fsf@python.net> <upu38f3jt.fsf@python.net>
Message-ID: <3c6ce50f@si-nic.hrz.uni-siegen.de>

Michael Hudson <mwh at python.net> wrote:
> Simon Budig <Simon.Budig at unix-ag.org> writes:
>> When you put it into 2.3(\.[0-9)? I am perfectly happy. I have
>> patched my local 2.1.1 installation and am happy with it...  :-)
> 
> This is done now.

Cool, Thanks!

Bye,
        Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From mwh at python.net  Wed Feb 27 06:18:29 2002
From: mwh at python.net (Michael Hudson)
Date: Wed, 27 Feb 2002 11:18:29 GMT
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com> <usn7oxtpd.fsf@python.net> <71c0a06e.0202261235.2165c300@posting.google.com>
Message-ID: <u8z9fxj96.fsf@python.net>

amrit040 at yahoo.com (Amrit) writes:

> Thanx that fixes the problem. So basically i was trying to read one
> more element than what the input had. This explains the bug but what
> i don't understand is why does this thing alternate ! Shouldn't it
> be consistent.

Guessing, but here's a go:

1) when you first call the function, it get's almost all the way
   through before you try to access the extra element.  This sets an
   exception, but as you don't check for error return, the computation
   completes and returns its result (in particular, not NULL).

2) when you call again you call PyArg_ParseTuple with, amongst other
   things, an "f" parameter.  This winds up calling PyFloat_ToDouble
   (or something like that).  There is no way to tell from the return
   value of this function that an error occured, so PyErr_Occurred
   gets called, and finally notices the error set in 1) above.  So
   PyArg_ParseTuple returns NULL, so your function returns NULL, so a
   traceback gets printed and the threadstate's exception gets
   cleared.  

3) you call the function again, and we're back in the situation of 1)
   again.

I suspect that if you call your function once, and then execute
something like:

>>> float("0.1")

you'll get an exception.

In summary, just try to get this right -- the consequences are
confusing.

Cheers,
M.

-- 
  I sense much distrust in you.  Distrust leads to cynicism, cynicism
  leads to bitterness, bitterness leads to the Awareness Of True
  Reality which is referred to by those-who-lack-enlightenment as
  "paranoia".  I approve.    -- David P. Murphy, alt.sysadmin.recovery


From jonathan at onegoodidea.com  Fri Feb  1 12:59:21 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 01 Feb 2002 17:59:21 +0000
Subject: Equivalent of Perl chomp?
References: <mailman.1012513005.17840.python-list@python.org> <Hxk68.4423$Xk4.215834@news1.east.cox.net> <u5js1gd8eecda8@corp.supernews.com>
Message-ID: <B88087F9.40FE%jonathan@onegoodidea.com>

On 1/2/2002 1:32, in article u5js1gd8eecda8 at corp.supernews.com, "Paul
Watson" <pwatson at mail.com> wrote:

> Perl chomp only removes -all- traliing newlines when in paragraph mode.
> Paragraph mode is when $/ is set to "".  The default mode is not paragraph
> mode, so chomp removes only one (1) newline.

*Bleurgh!*

Jonathan (adding one more thing to his Why I Don't Do Perl list...)



From wookiz at hotmail.com  Sat Feb  9 16:37:00 2002
From: wookiz at hotmail.com (wooks)
Date: 9 Feb 2002 13:37:00 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org>
Message-ID: <8852d33c.0202091336.1822b76a@posting.google.com>

Sheila King <usenet at thinkspot.net> wrote in message news:<a42c8e.3vvbc1l.1 at kserver.org>...
> 
> You might want to try the Tutor mailing list:
> http://mail.python.org/mailman/listinfo/tutor
> 
> I highly recommend it.

I have just subscribed. Thanks for the tip.
>  
> > Bought Python for Win32 book and already been caught out (Part 1 was a
> > breeze part 2 and I shuddered to a grinding halt). For perspective I
> > am a tester who wants to learn Python because of it's excellent unit
> > test framework and it's scripting glue like abilities (not interested
> > in building applications). If vbunit was as easy to figure (or had as
> > good a tutorial) as PyUnit (thank you Mark Pilgrim) I probably never
> > would have come here.  As it is perhaps getting to grips with vbunit
> > might be the lesser evil.
> 
> Ah, well, that's Windows for you. You might want to dabble and get your
> feet wet in the language a bit before diving into the win32 module(s).
> They are more complicated to work with, but that's just because it's
> Windows you're dealing with. Python can't fix that interface.
> 
Never got far enough into the book to look at win32. I couldn't
understand why there were these 2 classes that were variants of a
dictionary class and neither subclassed userdict and then the book
says now I'm using magic methods go look up in the language reference
(I did  - couldn't understand the raison d'etre). I think I have now
sussed answers to both issues - the userdict class (I am guessing) is
a recent addition and the book was at Python 1.5.2. The answer to the
second issue (I think) is that this is how (or one of the ways)
polymorphism is implemented. If I am indeed right on that then a
reference to that concept in the language reference would have saved
me alot of time and pain. I could have gone off and got and got a
conceptual understanding from a non-Python resource. I expect my
progress to be slow from hereon in with like issues cropping up.

> Have you been to this page?
> http://www.python.org/doc/Newbies.html
> Some of the tutorials listed there might be more useful in the "language
> learning stage" than that win32 book. Or not. I'm not sure where you are
> with the whole thing.
> 

I have been referring to a number of tutorials. I am not a newbie
programmer but I think the tutorials aimed at experienced programmers
are really aimed at experienced OO programmers - ignore that at your
peril.

> You might also want to try the win32 mailing list:
> http://mail.python.org/mailman/listinfo/python-win32
> 

OK thanks. I will try that.

> > The benefits of using of Python as a scripting glue like language are
> > widely touted but don't seem (I am probably wrong but it's a 1st
> > impression) to be well catered for. As I said I am not interested in
> > building applications. Is there a forum for the dumber less demanding
> > Python user?
> 
> See above.
> 
> Hope this helps,


From mike at nthwave.net  Sat Feb 16 00:01:18 2002
From: mike at nthwave.net (Michael Mell)
Date: Fri, 15 Feb 2002 21:01:18 -0800
Subject: line break confusion
References: <XFMail.20020215193042.shalehperry@attbi.com>
Message-ID: <3C6DE79D.F794FEEC@nthwave.net>

Sean 'Shaleh' Perry wrote:

> Windows: \r\n
> Linux (UNIX): \n
> Mac: \r
>
> Each OS's libs transform the sequence '\n' to whatever the right thing is for
> them.

Thanks for answering such an apparently ignorant question, but the answer is not
so simple.

I want to read and write files with line breaks which are not of the native
platform type.

Problem #1 is that on a Mac, the values of Mac and Unix are /swapped/ and
therefore wrong. Writing a file using '\n' results in a /Mac/ line break (as
reported by BBEdit, the venerable Mac editor). Writing a file using '\r' results
in a /Unix/ file. The Mac seems unable to write Windows \r\n no matter what
order is used.

Problem #2 is that when reading a Windows file on Linux or Mac, a /single/ line
('\r\n') is interpreted as /two/ lines ('\n\n'). Files written out on Linux then
have double the correct number of lines.

I believe Problem #1 is a bug, and Problem #2 just a fact of life.


--
mike at nthwave.net
llemekim         YahooIM
415.455.8812     voice
419.735.1167     fax





From rdacker at pacbell.net  Mon Feb  4 21:01:38 2002
From: rdacker at pacbell.net (rdack)
Date: 4 Feb 2002 18:01:38 -0800
Subject: [osx] milter problem
Message-ID: <644f6688.0202041801.7b0a5968@posting.google.com>

osx 10.1.2, python2.1, milter-0.3.9
i have installed the milter ok, but when i do: 
python sample.py
i get:
# python sample.py
dyld: python Undefined symbols:
_pthread_sigmask
_sigwait

any ideas?


From jkraska at san.rr.com  Wed Feb  6 01:58:46 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 06 Feb 2002 06:58:46 GMT
Subject: Autocoding project proposal.
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net> <mailman.1012972032.30205.python-list@python.org>
Message-ID: <gbl16u4ug6pau2l393eh50t85o031csmtm@4ax.com>

>Figure it out already!!!!

Has anyone happened to have seen the BCG research? One of their
data ranked the what traits opens source developers value most
in the project leader. The top trait was, at 48%, "makes significant
early contribution to the code."

Basically, you're not much of an open source software developer
unless you walk into the announcement with a prototype that people
can sink their teeth into. Hordes of people with ideas and little
more than 1,000 LOC attention spans soon follow. As always, these
efforts are brought to fruition off of a small group of dynamos
who really know how to make their keyboards ring.

C//




From zeitlin at seth.lpthe.jussieu.fr  Tue Feb 12 11:38:34 2002
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: Tue, 12 Feb 2002 16:38:34 +0000 (UTC)
Subject: wxPython wxListCtrl
References: <1r0a8.14824$Cg5.951049@news1.calgary.shaw.ca>
Message-ID: <slrna6ih8a.1ca.zeitlin@seth.lpthe.jussieu.fr>

On Tue, 12 Feb 2002 03:59:57 GMT, Andrew <anon at anon.com> wrote:
> I have a question about the wxPython wxListCtrl.  I was wondering if there
> is a way to insert a combo box for the items in a column (instead of the
> edit controls that are used).  If there is no easy way, what approach will I
> need to accomplish this?

 There is no easy way, I'm afraid. You might consider using wxGrid instead of
wxListCtrl to have a control over the controls used for editing the cell
contents though. Alternatively, you could try manually positioning a combobox
at the right location and handling all interactions with it but I am not
entirely sure that this can be made to work sufficiently well.

 Regards,
VZ
-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From phr-n2002a at nightsong.com  Wed Feb 20 12:02:58 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 09:02:58 -0800
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk>
Message-ID: <7xadu49jct.fsf@ruckus.brouhaha.com>

"Phil Mayers" <p.mayers at ic.ac.uk> writes:
> I just know someone is going to say "return the octet string and do this:
> 
> intval = reduce(lambda x,y: (x<<8)+ord(y), strval)
> 
> I've tried, and this was too slow (the whole point of the C module being
> to speed this up).
> 
> Any ideas?

Convert to hexadecimal and do intval = int(hexval, 16).
Really, no joke.


From mal at lemburg.com  Mon Feb 18 17:05:08 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 18 Feb 2002 23:05:08 +0100
Subject: ANN: eGenix.com mxCGIPython Version 0.4.0
Message-ID: <3C717A94.2C9ED2C5@lemburg.com>

________________________________________________________________________

ANNOUNCING:
                 eGenix.com mxCGIPython Version 0.4.0

            An easy-to-install one-file Python Interpreter

________________________________________________________________________

WHAT IS IT ?:

mxCGIPython is a project to provide a complete ready-to-run
Python interpreter in one single executable file. The main
intent for supplying such a single file executable is to 
ease deployment of Python for CGI scripting even when the ISP
is not willing or able to install Python for you.

The mxCGIPython archive contains everything needed to build
the one file interpreter. The web page provides a set of links
to contributed binaries for direct use in addition to setup
instructions to get the binaries up and running at your ISP.

See the documentation page for a list of available binary
packages. Binaries for new platforms are always welcome :-)

________________________________________________________________________

WHAT'S NEW ?

Version 0.4.0 was updated for Python 2.1.2.

The project already includes a large collection of Python 1.5.2 
binaries and will hopefully grow an equally large collection of 
Python 2.1.2 binaries soon.

________________________________________________________________________

WHERE CAN I GET IT ?

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

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

________________________________________________________________________

WHAT DOES IT COST ?

The packages comes with a Python 2.0 style license, which means
that you can use them in both commercial and non-commercial settings
without fee or charge.

The binaries are covered by the usual Python licenses.

________________________________________________________________________

WHERE CAN I GET SUPPORT ?

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

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

for details about the eGenix support offerings.

________________________________________________________________________

Enjoy,
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From brian at sweetapp.com  Wed Feb  6 22:23:28 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Wed, 6 Feb 2002 19:23:28 -0800
Subject: money class?
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBKEENKBAA.mark@mceahern.com>
Message-ID: <000001c1af86$d050f6d0$445d4540@Dell2>

Mark McEahern wrote:
> 	Does anybody know of a nice Money class they'd be willing to
share,
> point to?

Are you looking for any class that happens to be named "Money" or are
you looking for some particular functionality?

Cheers,
Brian




From mal at lemburg.com  Sat Feb 16 06:44:54 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 16 Feb 2002 12:44:54 +0100
Subject: ANN: PSF Contribution Documents
References: <mailman.1013769204.16499.python-list@python.org> <a4jpo3$ilt$0@216.39.172.122>
Message-ID: <3C6E4636.D3B3D298@lemburg.com>

[comments on the docs]

Please post comments about the documents to the forms mailing list
only. We will use it's archives later on as reference.

For this reason, follow ups should only be posted there.

I've forwarded your message to the list.

Thanks,
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From usenet at thinkspot.net  Mon Feb 25 17:38:57 2002
From: usenet at thinkspot.net (Sheila King)
Date: Mon, 25 Feb 2002 14:38:57 -0800
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com>
Message-ID: <a5dia2.3vs4847.1@kserver.org>

On Mon, 25 Feb 2002 17:34:02 -0500, "Joshua Muskovitz" <joshm at taconic.net>
wrote in comp.lang.python in article <3c7ab974_4 at corp.newsgroups.com>:

> 
> If you had to give a 15 minute introduction to Python, what would you cover?
> What is a good example of a simple, but useful, text-based app for a Wintel
> environment, that has good "wow" quality?

Who is your audience? What is their background and what types of things
would they like want to do with a computer?

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From mwh at python.net  Tue Feb 12 07:17:09 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 12 Feb 2002 12:17:09 GMT
Subject: Problem building Python 2.2 docs
References: <7xeljrcz1o.fsf@ruckus.brouhaha.com>
Message-ID: <ur8nqhoy9.fsf@python.net>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> I just loaded a fresh Red Hat GNU/Linux 7.2 system on a newly
> formatted hard disc, and then installed Python 2.2 on it from the .tgz
> file.  Python 2.2 itself built and installed just fine.  When I typed
> "make" in the Doc directory though, a bunch of errors resulted, like:
> 
>      *** processing declarations ***
> 
>     Loading /usr/share/latex2html/versions/latin1.pl
> 
>      *** Fatal Error --- but easy to fix ***
> 
>     Cannot have '.' in file-name prefix, else dvips fails on images
>     Change the name from  /home/phr/Python-2.2/Doc/lib/lib.tex  and try again.
> 
>     *** Session transcript and error messages are in /home/phr/Python-2.2/Doc/html/lib/lib.how.
>     *** Exited with status 255.
> 
> I don't remember having this problem with Python 2.1.1 under Red Hat 7.1.
> 
> Is this a known bug, in either Python 2.2 or Red Hat 7.2?

I saw this too.  It's probably a problem with latex2html (ooh! another
one!).  Maybe you need to uprade (or even downgrade) it.  Either way,
it's probably best to ask Fred which version is currently blessed,
either directly, via the python-docs at python.org line or on the
doc-sig.

Cheers,
M.

-- 
8. A programming language is low level when its programs require
   attention to the irrelevant.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From hfoffani at yahoo.com  Tue Feb 12 14:16:36 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Tue, 12 Feb 2002 20:16:36 +0100
Subject: 19 y/o blonde want to know something about Python working with IPX protocol
References: <mailman.1013529085.695.python-list@python.org>
Message-ID: <20020212142721.788$8n@news.newsreader.com>

"Mariano Corsunsky" wrote:
> DISCLAIMER:
>
> Hi!, I tried to get this same question answered before, but
> nobody did. So I'm using the "19 y/o blonde strategy"
> ----
>
> Hi boys!
>
> I want some examples of Python dealing with the IPX protocol.
> I do only need a simple, so so simple example which sends and
> receive strings between 2 computers using IPX.
>
> Could you help me?

I've already answer you before. This was my post.

You want to search google with "IPX" and "library".
It seems to be several free/commercial libraries available.
You'll
need to make a wrapper for Python.

-Hernan





From loewis at informatik.hu-berlin.de  Wed Feb 20 03:58:48 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Feb 2002 09:58:48 +0100
Subject: print and unicode strings
References: <mailman.1014174755.8438.python-list@python.org> <m38z9o6347.fsf@mira.informatik.hu-berlin.de> <n3n67ucp307imfivlcb4kn2ooe4t57l75c@4ax.com>
Message-ID: <j4zo24ttpz.fsf@informatik.hu-berlin.de>

Toby Dickenson <tdickenson at devmail.geminidataloggers.co.uk> writes:

> That still wont fly, because 'print' always passes the to-be-printed
> objects through str() before calling file.write()

In Python 2.2, this is no longer true, due to patch #462849, applied
as fileobject.c 2.127. This is very similar to your patch.

Regards,
Martin


From sabren at manifestation.com  Mon Feb 11 18:47:23 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 11 Feb 2002 18:47:23 -0500 (EST)
Subject: strongbox
Message-ID: <Pine.LNX.4.44.0202111839500.568-100000@mercury.sabren.com>

Hey all,

This is from an idea I'm working on at:

http://www.sixthdev.com/wiki.cgi/strongbox

It concerns a scheme for somewhat strongly typed python,
including getters and setters... There's no code, but I'm
posting it here because I saw Guido's slides from the
conference, and noticed that the development team is
thinking about adding this kind of thing in a future python
version... The strongbox concept offers these things through
metaclasses.

(It also describes a scheme that uses the Observer pattern
to persist objects to various backends (relational, ZODB,
etc))


----

I call it strongbox because it's strongly typed. Python
variables are weakly typed, which means x can be a string
one minute, and an integer the next. That's great most of
the time, but it's a pain for working with relational
databases, because the columns hold only one type of
data. So my zdc applications are full of validation code
that gets repeated all over the place. Data validation is
one of the biggest things slowing me down when I go to write
a web app.


Strongbox changes everything. It starts with a metaclass
called, not surprisingly, Strongbox. When you subclass
Strongbox, you don't get a normal python class, but a
Strongbox class. Strongboxes have built-in getters and
setters. They also let you define strongly-typed slots like
so:


from strongbox import Slot 
from strongbox import Strongbox 
from pytypes import FixedPoint 
 
class Person(Strongbox): 
    fname  = Slot(str, size=15, default="herman") 
    salary = Slot(FixedPoint) 
    favorite = Slot(str, valid=('red','blue','green')) 
    age = Slot(int, valid=lambda x: x > 0) 
 
    def get_canDrink(self): 
        return self.age > 21 
 
>>> fred = Person(fname='fred', age=18) 
>>> fred.canDrink 
0 
>>> fred.age = 25 
>>> fred.canDrink 
1  


The "valid" argument lets you specify rules for valid
data. I plan to support regular expressions, enums, and
lambdas. (If you need a full function, you can just override
get_XXX).


The "pytypes" module mentioned above is also a child of
zdc... I'm taking all the datatypes defined there (Date,
DateTime, IdxDict, FixedPoint) and putting them into their
own module.


Another neat thing about Strongbox is the Observer
pattern. Notice that in the example, fred is just a
freestanding object. No database was needed. Fred doesn't
know anything about databases, but he can be saved to
one. The simplest way is to pickle him and store him in a
blob or a file. (Not 100% sure pickling works with
metaclasses, but we'll find out). But we can also store
fred's slots in a relational database.

To do that, we create a special observer called a Clerk. As
Fred changes, he fires off events notifying anyone who cares
to listen that something has happened. The Clerk listens to
these events, and can therefore decide whether or not he
needs to be saved. This means you don't actually have to
call "save" every time something changes. You *can* do that
if saving is expensive, but you can also just have the Clerk
save fred to the database.

Which database, though? It doesn't matter. Clerk doesn't
know about databases either, it knows about Sources. Source
is just an interface, and there will be different Sources
for different types of storage systems. DBAPI2Source will
map strongboxes to a database like MySQL.


With zdc, the object-relational mapping required all sorts
of work, but with strongbox, it's simple. You've defined
your Slots in python, so there's no need to investigate the
table. Mapping classes to tables can be done with a simple
python dictionary. In fact, we can even generate the SQL
statements for the tables from the class definition.


The best part of all this is that most of it is already
done. I expect Strongbox to be a fraction of the size of
zdc, and it should only take a day or two to implement

----

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From logiplexsoftware at earthlink.net  Fri Feb  1 18:15:12 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 15:15:12 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <a3f6ch$c5$1@news.hccnet.nl>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3f6ch$c5$1@news.hccnet.nl>
Message-ID: <20020201151512.156ebaed.logiplexsoftware@earthlink.net>

On Fri, 01 Feb 2002 23:53:50 +0100
Anton Vredegoor wrote:

> Sometimes this X is: 'It should be ready next friday'. So there's no
> time to invest in changing programming language now, even if it would
> be be a good long term choice. 
> 
> There seems to be no opportunity to introduce a change since the work
> must go on.
> 
> How to handle this kind of resistance? 

I think that reality is the determining factor in this case ;)  The things
we are talking about are long-term benefits of adopting new technologies. 
Just because a technology would be beneficial doesn't mean you have to
adopt it before you go on lunch break today.

However, since you brought it up: _after_ Friday (assuming the release date
was met) and _before_ another specification and release date are set for a
new version, it would be worthwhile to evaluate other options _if_ it is
felt that making such a major change would be beneficial in the long term. 


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 19:37:51 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 18:37:51 -0600
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com>
Message-ID: <00a001c1b35d$865173e0$0101010a@local>

----- Original Message -----
From: "Jeff Shannon" <jeff at ccvcorp.com>
>
> Timothy Rue wrote:
>
> > But then group voting is like a conspiracy huh?
>
> Only for nonstandard definitions of 'conspiracy'.
>
> Conspiracy implies secrecy, and imposing the will of the few upon the
many.
> Voting is nonsecret (even if individual votes are secret, the process is
not)
> and implies determining the will of the many (at the possible expense of
the
> will of the few).

Yes.  I apologized (to the list, not to Rue) because the suggestion is,
well, non-Pythonic.  Tacky.  And I was/am sorry.  Jeff has it right here,
though, that it was certainly no conspiracy; and it is the USENET way to
publicly handle such things.

> Then again, we shouldn't be surprised at you using nonstandard definitions
of
> words.

Indeed.  He uses a nonstandard definition of "language" and applies
"physics"
to programming.  He can't program, you know; ten years on this project and
nothing to show for it.  If I had tried to develop something like this for
ten years and had nothing to show for it but a website, I'd give up.






From phr-n2002a at nightsong.com  Tue Feb 19 04:29:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 01:29:36 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org>
Message-ID: <7x7kp9g6pr.fsf@ruckus.brouhaha.com>

O Polite <ol1 at v10a.com> writes:
> Well, first you'd want to have a way of representing a linked list.
> Here's one straighforward way of doing it, but unless your taking
> Datastructures 101 you'll never be interested in it. The builtin list
> type will always be faster than anything you implement in python, and
> now in python2.2 you can subclass the builtin types as any other class.

That's a misconception.  One of the main reasons for using linked
lists is to be able to insert stuff in the middle taking a constant
amount of time.  If you do that with a built-in list, the time needed
is proportional to the size of the list, which can be very large.


From tim at worthy.demon.co.uk  Sat Feb  2 06:15:50 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Sat, 02 Feb 2002 11:15:50 GMT
Subject: Zipfile - zinfo.extra
Message-ID: <097a74024b.tim@worthy.demon.co.uk>

I've been trying to write Acorn/RISC OS format zip files using the
Python zipfile module.

Filetype information is not stored as an extension but hidden under
normal filesystems and written in the zipinfo's "extra" field in zip
archives.


I've succeeded in creating correctly formatted zips but only by adding
"extra" as an optional parameter to the zipfile.py module's zipfile
class write method.

def write(self, filename, arcname=None, compress_type=None, extra=None)


Anyone know of a better way to do this ?


AFAICT there are the writestr and write methods.

writestr allows writing of a string (which could be anything) and gives
control over the "info" data - but involves more "fiddling" than
"write" as you have to create all (mandatory) zinfo yourself.

write allows simple "copying" of a file into a zip.


What I'd like to do is write then set zinfo.extra to the filetype data.

-- 
___
 |im    ---- ARM Powered ----


From christophertavares at earthlink.net  Sat Feb  2 02:52:21 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Sat, 02 Feb 2002 07:52:21 GMT
Subject: "Private" attributes - a possible idea (maybe PEP-worthy).
References: <mailman.1012520714.2778.python-list@python.org>
Message-ID: <VUM68.1204$3E5.108514@newsread2.prod.itd.earthlink.net>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:mailman.1012520714.2778.python-list at python.org...
[... snip ...]
>
> I'd like to close this thread here (which I created because I thought I'd
> come up with a way to get around the wart) unless someone *can* come up
with
> a way to deal with it.
>
> Tim Delaney
>

How about changing the name mangling rule so that the current number of base
classes is added to the secret string? That way, the first TimKlass's
variables would be __TimKlass_0_Something, and the derived one would be
__TimKlass_1_Something.

This should work in the presence of multiple inheritance, and is fairly
simple. It could possibly break if people are screwing around with their
base classes at run time, at which point I'd say "Don't do that!"

-Chris





From joshm at taconic.net  Mon Feb 18 17:35:47 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 18 Feb 2002 17:35:47 -0500
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net>
Message-ID: <3c717f62_3@corp.newsgroups.com>

The Ruebot wrote:
> If you want to be greedy Phil, to take that job away from me, then give me
> the tool under the conditions of which I want it.

So do it already!  You complain that programmers are keeping the keys to the
kingdom to themselves, and your system will blow those doors open.  If your
truly amazing design is really able to do that, then why haven't *you*
empowered yourself and just done it?  Instead, you whine and insult, in
order to try to convince "us" that we should do it for you.  Why should we?
Clearly, it doesn't benefit us.  You'll never convince us to give up our
power.  So instead of harassing us, do it yourself, and then that will
really show us.  Then you'll have the last laugh.  Otherwise, we'll just
keep having *all* the laughs.  Especially when we read your posts over
brandies at the weekly PSU meetings, while we light our cigars with $100
bills, and chuckle about our complete control over the unsuspecting world.
Mwahahahahaha.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('All hail the PSU and bow down before its might or we just might get
testy')




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


From magnus at thinkware.se  Wed Feb 20 20:01:49 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 17:01:49 -0800
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <258fd9b8.0202201701.5f8e4e42@posting.google.com>

Rajarshi Guha <rxg218 at psu.edu> wrote in message news:<a50umn$1dou at r02n01.cac.psu.edu>...
> Why can't I use d.keys().sort() directly? 

Because .sort() is not a function that returns a sorted
copy of it's input. It sorts the list in-place and returns
none. A sort method returning a copy would use much more
memory.

>And is there any other more 
> concise (elegent?) way to loop over sorted dictionary keys?

Don't be so lazy. ;) If you wish you can do:

def keySort(dict):
    k = dict.keys()
    k.sort()
    return k

and

for k in keySort(myDict):
    whatever...


From sjones16 at rochester.rr.com  Wed Feb 27 23:12:57 2002
From: sjones16 at rochester.rr.com (Seth Jones)
Date: Thu, 28 Feb 2002 04:12:57 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com>
Message-ID: <d7if8.149174$dh.38781704@typhoon.nyroc.rr.com>

> Does Python?  If so, I've never seen it.  I'd like to.
>
Well, there's pygame which wraps around SDL, and python has openGL wrappers.
Guess those "batteries" aren't technically included, though.






From phr-n2002a at nightsong.com  Mon Feb  4 11:52:13 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 08:52:13 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com> <7xzo2pqopt.fsf@ruckus.brouhaha.com> <WZx78.40291$KM2.1247277@atlpnn01.usenetserver.com>
Message-ID: <7xy9i942bm.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> > > > If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> > > > it isn't backwards compatible.
> > > >...
> > Yes, but they DON'T continue to run, then the new release DOESN'T provide
> > backwards compatibility.
> 
> But surely your example above is about asking 1.5.2 or 2.1 to do a "//"
> operation? Anyway, we can omit all further discussion on *that* topic since
> it's clearly not what's foremost in your mind.

The absence of // in 1.5.2 and 2.1 just shows that importing future division
doesn't provide backwards compatibility.  I suggested an alternate way
to provide it, namely giving 3.0 a way to import "past" division.

> Only because you've already decried *the* forward-compatible alternative:
> 
> 3) Use divmod()[0]
> 
> You said "yerch" (or something similar to this), but surely from an
> engineering point of view it gets the pragmatic result you want.

Of course it doesn't.  It discards what are supposed to be among
Python's main virtues--readability, conciseness, and ease of
understanding.  Those are pragmatically nothing to sneeze at.
Just last night I sent some Python code to a Lisp weenie, saying
basically "look how cool Python is".  Can you understand how stupid
it would have looked, if I'd had to say divmod(x,y)[0] every time
I wanted integer division?

> Tim admitted this as a remote possiblity rather than suggesting it was
> probable. As in "there's nothing to stop anyone doing it". Certainly 2.1.2
> is the most stable release, but it doesn't implement the current Python
> language, whose definition is now implemented by the 2.2 release.

What does "nothing to stop anyone doing it" mean?  Maybe Tim can explain
that.  If I make some patches to 1.5.2 and submit it to the maintainers
as 1.5.3, will they release it on the python.org website?  What if
I make patches to 1.2 and submit it as 1.2.1?  (Or 1.2.2 or whatever).

> Indeed. So just write in the forwards-compatible subset of 1.5.2,
> and your problem goes away.

I do try to do that, but if it means avoiding using division and who
knows what else, it gets to be a big pain.  

> > home, but if I was installing a new production system today, I'd use
> > 2.1.1.
> 
> I'd use 2.1.2, but I presume you've just overlooked this January 16 release.
> I must be missing something. 

I thought the Jan 16 release was 2.1.1--I must have gotten confused.  Sorry.

> If I'm not using 2.2 features, I have no need to switch. So, surely,
> if I *want* to use those features then I will switch regardless of
> the inconvenience.

I don't know what the best approach to that is.  I think I'd rather let
2.2 age a little while longer before doing anything important with it.

> Aah, "for general distribution". In that case, you are stuck with either
> remaining in the forwards-compatible 1.5.2 subset, or distributing the
> interpreter with your applications. We can probably agree that while the
> latter is practical for large applications (via py2exe of the McMillan
> installer, or in the same way Zope does), for smaller programs it simply
> carries too much overhead. But I see you wanting the best of both worlds.

I'm not sure what you mean about the best of both worlds.  Python's
"advertising" itself as a mature language to me means there should not
be a useable 1.5.2 subset that keeps shrinking, as long as 1.5.2 usage
is still widespread.  If it's an experimental research language, fine,
keep tweaking and changing it, but don't advertise it as being mature
or production-ready without understanding that that means supporting a
lot of legacy code.  Trying to do both (attract a broad userbase by
claiming maturity, yet also feel free to break existing code by
changing the language at whim) is closer to what I think of as the
poor sense of wanting the best of both worlds.  GCC (a production C
compiler) takes a more responsible approach--it implements language
changes sometimes, but supports legacy programs through compiler flags
like "-traditional".

> > That is a problem too.
>
> But a (relatively) minor one. RedHat have stuck with 1.5.2 for
> rather longer that was strictly necessary for Anaconda support, they
> could have updated it relatively easily but presumably backed off
> for engineering reasons.

I wonder what the demographics really are (what percentage of users
are using what Python version).  If not many are still using 1.5.2
then I can stop worrying about it, but 2.1 is definitely still in the
mainstream.

> > I don't see at all why some mechanism can't be
> > provided to turn off the change ("import from past") just like there's
> > now a mechanism to turn on the change ("import from future").
> >
> Because Guido doesn't feel it will be good for your sanity.

He said that?  The subject has come up before?  I didn't notice
anything about it in the PEP, so I was thinking of entering it as a
request in Sourceforge.


From grante at visi.com  Wed Feb 13 22:30:10 2002
From: grante at visi.com (Grant Edwards)
Date: Thu, 14 Feb 2002 03:30:10 GMT
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com> <1013641215.358571@yasure>
Message-ID: <slrna6mbq1.7ta.grante@isis.visi.com>

In article <1013641215.358571 at yasure>, Donn Cave wrote:
> Quoth grante at visi.com (Grant Edwards):
>| I'd like to capture stdout/stderr from a program run with
>| os.spawnv().  
>|
>| I've searched through c.l.p archives, and the usual answer
>| seems to be "use popen2 et al".  I don't want to use popen
>| since it runs the program as a shell command and I don't want
>| the shell to mess with the strings I'm passing as arguments to
>| the programs.
> 
> Check it out.  Not popen, but as they say, "popen2 et al."

>From the popen2 doc page:

  "the parameter cmd is the shell command to execute"

I guess the use of the phrase "cmd is the shell command" in the
documentation led me to the conclusion that the "cmd" parameter was a shell
command.  ;)

> It will exec the command on its own, no shell, if you give it
> an argument list.  If you give it a string, it will use the
> shell.

Cool, that's news to me. Is that documented anywhere?

-- 
Grant Edwards                   grante             Yow!  Oh my GOD -- the
                                  at               SUN just fell into YANKEE
                               visi.com            STADIUM!!


From phr-n2002a at nightsong.com  Wed Feb 20 18:32:37 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 15:32:37 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014221254.26490.python-list@python.org> <7xu1scdsxh.fsf@ruckus.brouhaha.com> <3C73F4AF.4028FDD7@ccvcorp.com>
Message-ID: <7xit8rwwyy.fsf@ruckus.brouhaha.com>

Jeff Shannon <jeff at ccvcorp.com> writes:
> > So why are augmented assignments not allowed in expressions?
> >
> >   if (foo += 2) < 7:
> >         handle_foo()
> >
> > doesn't seem likely be an =/== confusion.
> 
> Because the benefits of doing this are small compared to the detriment of
> muddying the statement/expression distinction, perhaps?  It would be a weird
> special case, behaving differently from everything else in the language, and
> therefore a wart.

Aha, so it has nothing to do with =/== confusion.  That's what I wanted
to know.  

Next question: why is muddying the statement/expression distinction a
detriment?  As opposed to the alternate view, that the statement/expression
distinction is itself a detriment?


From jason at jorendorff.com  Tue Feb  5 19:38:51 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 5 Feb 2002 18:38:51 -0600
Subject: array precision
In-Reply-To: <d0380df9.0202050933.7481d17b@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEKCDAAA.jason@jorendorff.com>

Hugo Martires writes:
> suposing that n=10 we get: myArray[0] = 0.600000023 ("something 
> like this")
> 
> but if i do: myArray[i] + (1.0/n) it give's me 0.6
> 
> what's the problem of the array, and how can i solve this ?

It's only half of the normal precision.  From the documentation:

    The following type codes are defined:
        Type code   C Type             Minimum size in bytes
        'f'         floating point     4
        'd'         floating point     8

Try using 'd' instead of 'f'.

## Jason Orendorff    http://www.jorendorff.com/



From tdelaney at avaya.com  Tue Feb 19 00:10:21 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 19 Feb 2002 16:10:21 +1100
Subject: help - inserting values into database tables
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A199@natasha.auslabs.avaya.com>

> From: chajadan [mailto:python at chajadan.net]
> 
> It seems I have ISDN on the brain, because I changed the 
> correct "DNS" to 
> the incorrect "DSN" partway through and never went back ;)

Actually, you changed the incorrect "DNS" to the correct "DSN" (Data Source
Name) and never went back :)

Tim Delaney



From sholden at holdenweb.com  Fri Feb 22 07:47:47 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 07:47:47 -0500
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com> <CMfd8.5037$D6.14223@news.iol.ie> <u7b81629bo9o4a@corp.supernews.com> <slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu>
Message-ID: <m5rd8.77889$n73.1507223@atlpnn01.usenetserver.com>

"Quinn Dunkan" <quinn at yak.ugcs.caltech.edu> wrote in message
news:slrna7b9hr.6lj.quinn at yak.ugcs.caltech.edu...
> On Thu, 21 Feb 2002 17:35:44 -0800, Virginia O <vodine at direct.ca> wrote:
> >Thanks, but the database will have to hold sub-groups etc. It will be
easier
> >for me to enter and organize the items in a database.
> >I only gave the scenario as an example. The actual program will hold
> >hundreds of items.
>
> Hundreds is just fine for flat text.  If you have a few thousand and
you're
> modifying it a lot, then flat text will start being clunky.
>
> In that case, shelve should work fine, especially for complicated data
> structures since you can dump python types directly, with no clunky
converting
> to/from some format SQL likes.  And it's much simpler to have everything
in
> python instead of a whole other database program with its own language,
etc.

If you really need relational, Gadfly has the advantage that it's pure
Python, although there are some SQL features (notably NULL values) that it
doesn't implement.

It used to warn about the regex module, but there are now updated
implementations available that don't suffer from that restriction.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From justin at iago.org  Sun Feb  3 18:48:48 2002
From: justin at iago.org (Justin Sheehy)
Date: Sun, 03 Feb 2002 18:48:48 -0500
Subject: Xython - XML-Formed Python
In-Reply-To: <1012751401.4431.13.camel@acropolis> (Gerard Braad's message of
 "03 Feb 2002 16:49:54 +0100")
References: <1012751401.4431.13.camel@acropolis>
Message-ID: <x7y9ia15zz.fsf@ra.iago.org>

Gerard Braad <g_braad at survion.com> writes:

> For a long time now, I had the idea of XML-Formed Python. This way it
> would be possible to send Python code across a network connection. 

This seems like a nonsequitur.

There appear to be a number of ways to send Python code across a
network connection.  What problem are you trying to solve here that is
not solved by any of them?

-Justin

 




From fperez528 at yahoo.com  Fri Feb  8 04:36:01 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Fri, 08 Feb 2002 09:36:01 +0000
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org>
Message-ID: <a40urp$o10$1@peabody.colorado.edu>

Chris Gonnerman wrote:

> Consider this an apology to the list, regarding my recent Survivor-like
> suggestion that we "vote off" Rue.  It was late, I wasn't feeling good, but
> still, it was in bad taste, and for this I'm sorry.

Very gentlemanly of yours. I think the way to deal with people like Rue is 
very simple: silence. What his ilk crave is attention, above anything else. 
So if we simply shut up, no matter how tempting a flamefest with an idiot may 
seem, he'll quickly move away. It will only take a couple of posts by him 
that receive *zero* responses here for us never to hear from him again, I 
would bet.

And that approach doesn't require 'voting off' anyone, which I find a fairly 
repulsive thing to do. Silence from people should come simply as a personal 
decision not to feed needless threads, but not as a 'show of force' of a 
group against an individual, no matter how annoying he may be.

Cheers,

f.


From db3l at fitlinxx.com  Wed Feb 27 17:06:00 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Feb 2002 17:06:00 -0500
Subject: cross platform (linux/windows) problem
References: <f5c74c0f4b.miah@picsel.com>
Message-ID: <u664ibmwn.fsf@ctwd0143.fitlinxx.com>

Miah Gregory <miah at picsel.com> writes:

> Simple example which works for linux:

You may need to have some per-platform support, since I do think you
need to use the popen2.Popen3 class under Unix to get exit codes, but
that isn't supported under Windows.

Note that your sample Linux code might have a problem in that you
don't drain the output from the piped child until you've gotten an
exit code, but if the child generates more output than the system
pipes can buffer, the child may block and never exit until you drain
the data.  Unless you know the output from the child is limited, I'd
suggest draining the data as it runs (even if you just save it for
later use).

Under Windows, you should just be able to use os.popen3 directly.  Do
the same draining of data until the process exits.  Then close the
files you received from os.popen3 - the final one you close (from the
full set of stdin/stdout/stderr) will return the exit code of the
child process.

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


From sanjay at jarna.com  Tue Feb 19 05:26:37 2002
From: sanjay at jarna.com (Sanjay)
Date: 19 Feb 2002 02:26:37 -0800
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com> <3C6D08C1.2000307@skippinet.com.au> <9d4c7023.0202151038.1d3307f@posting.google.com> <3C6DAF39.5080808@skippinet.com.au> <9d4c7023.0202181505.1646f9e6@posting.google.com> <3C71E06E.5080206@skippinet.com.au>
Message-ID: <9d4c7023.0202190226.5c1f0b5d@posting.google.com>

Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C71E06E.5080206 at skippinet.com.au>...
> Sanjay wrote:
> > Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6DAF39.5080808 at skippinet.com.au>...
> > 
> >>Sanjay wrote:
> >>
> >>>What I am trying to do is something that was somewhat discussed on a
> >>>previous thread, but I'm not sure the thought was fully completed. 
> >>>Specifically, I would like to write an exchange server event sink by
> >>>implementing two com interfaces, IExStoreAsyncEvents and
> >>>IExStoreSyncEvents.  The problem I'm having is that when the python
> >>>event listener is loaded as a com+ application, and I register this
> >>>application for events with exchange, I get the error TypeError ...
> >>>IID not supported.  From looking at the previous threads, it seems as
> >>>this problem is caused by not being able to handle arbitrary vtable
> >>>interfaces, such as the interface methods of IExStoreAsyncEvents and
> >>>IExStoreSyncEvents.  If, however,  the two com classes
> >>>IExStoreAsyncEvents and IExStoreSyncEvents were defined in a pyd, then
> >>>pythoncom would have knowledge of these interfaces, and properly call
> >>>my application.
> >>>
> >>>I hope I'm explaining the problem clearly. I am a little confused as
> >>>to how to get around this with pythoncom.
> >>>
> >>You need t6o check out the win32com\servers\test_pycomtest.py.  This 
> >>shows how arbitary vtables must be loaded by your code before it is 
> >>recognized.  This in-turn implies that the interface is described in a 
> >>type library.
> >>
> >>Mark.
> >>
> > 
> > Ok, I looked at test_pycomtest.py and changed my interface to match,
> > i.e. I set _public_methods_ to an empty list, _com_interfaces_ to
> > "[IExStoreSyncEvents]", I created a new _reg_clsid_ value, and
> > implemented the two methods in the interface.  I then registered the
> > class using register.UseCommandLine() with debugging turned on.  Then
> > I told Exchange to notify this object when new events happen.  In the
> > windows Event viewer, I now see the following error when events occur
> > in Exchange and it tries to call my python com object:
> > 
> >     Exception: C0000005
> >     Address: 0x1E12E6A5
> >     Call Stack: 
> >     python22!PyDict_Next + 0xC5
> >     ole32!StringFromCLSID + 0xB91
> >     ...
> >     ...
> > 
> > There doesn't seem to be anything strange in the Python Trace
> > Collector window.  I see a message saying "Object with win32trace
> > dispatcher created (object=None)", followed by 3 calls to
> > QueryInterface with unsupported IIDs (Unregistered, ISTdMarshalInfo,
> > and IExternalConnection) which are all interfaces I should not need
> > implemented here.
> > 
> > Any idea what the problem is?
> 
> It could be IExternalConnection - this is often necessary to support 
> events.  See the various "connect.py" files in the win32com tree for 
> some insights into how events are usually hooked up.
> 
> Do you have sample C++ code that works in this environment.  If so, you 
> should be able to see exactly what is necessary to implement.
> 
> Mark.

There is a sample event handler in C++ that ships with the Exchange
SDK.  Its header file includes the following classes:

    BEGIN_COM_MAP(CExEv)
	COM_INTERFACE_ENTRY(IExEv)
	COM_INTERFACE_ENTRY(IDispatch)
	COM_INTERFACE_ENTRY(IExStoreAsyncEvents)
	COM_INTERFACE_ENTRY(IExStoreSyncEvents)
	COM_INTERFACE_ENTRY(IExStoreSystemEvents)
    END_COM_MAP()

In this example, it implements all 3 event interfaces.  This seems to
handle events fine.  No mention of IExternalConnection, but it is
definitely a dual interface.  I *thought* I had modeled my event
handler after this c++ example and the other python examples.  Thanks
for all your help so far.

Sanjay


From phr-n2002a at nightsong.com  Tue Feb 26 02:06:10 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 23:06:10 -0800
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com> <jg3m7ugd2c3svhi4k9s2qp88jm8f2ib99s@4ax.com>
Message-ID: <7xadtw7me5.fsf@ruckus.brouhaha.com>

Courageous <jkraska at san.rr.com> writes:
> The list part of me wants:
> 
> 	x = 3
> 	let ( x ):
> 		x = 5
> 
> I might call it something other than "let" to make it Pythonic, though.

That's got its attractions, but doesn't handle loop indices so well.

> >2) Add a compiler option similar to Perl's "use strict",
> 
> This has been brought up before. Personally, I favor the idea of a
> directive of some kind, on per module basis, that affects that
> module for that module's data. One of the reasons you're not seeing
> too many people clamour for this kind of thing, however, is that
> misspellings are (so it is claimed, and so my own personal
> experience suggests) actually a relatively scarce cause of Python
> problems.

Besides misspellings, there's a class of bugs where you patch a program
by inserting some code in the middle, not noticing you've clobbered
a variable from far away in the function.  Local scope prevents that.
Use strict makes sure you don't omit a local declaration by accident.

> You can get some of this kind of behavior _now_ by the way. Just implement
> a class which doesn't allow setattr to work on an instance for any variable
> which doesn't likewise exist in it class.

I'm talking about ordinary variables in functions, not class instance
variables.

> *Poof, Guido uses his time machine*
> 
> >   Example 2:
> >     local x(int)
> >     x = "banana"    # implementation-dependent behavior
> 
> I disagree here. If you've gone to the trouble of declaring something,
> to get full use, you need a defined error; further, it would be helfpul
> if this occured prior to execution.

The implementation can define an appropriate error.  I don't mean the
implementation is supposed to go off and do something totally unpredictable.

Also, that particular example is easy for the compiler to flag by
static analysis; but there's also cases like

  x = frob(3)

where frob turns out to return a string if the arg happens to be 3?
In general, the compiler can't figure that out.

> There's some good cause for resistance to this. One of those things which
> keeps Python Pythonic is it's concise expressivity. Declarations foil that.

The declarations are no less concise than assert statements (for error
checking).  For compiler advice, they're like pragmas.


From emile at fenx.com  Sun Feb 24 23:43:34 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 24 Feb 2002 20:43:34 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com>
Message-ID: <a5cfpe$67voi$1@ID-11957.news.dfncis.de>

"Paul Rubin" 
> Both of those build up a new list of the results, instead of
> discarding the values.  If the f function takes an integer and
> computes a 20-megabyte structure, you've got a problem.  

If you know it's coming, throw it away first:

[x.f() and None for x in items]


> This really
> calls for a generator comprehension as discussed a few weeks back.
> 
>    reduce(lambda x,y:(y.f(),0)[1], items, 0)
> 
> almost works, but invokes f an extra time at the end.
> 
> Someone see how to fix it?

Expanding the above to:

filter(None,[x.f() and None for x in items])

returns an empty list vs 0 for the reduce.  


-- 

Emile van Sebille
emile at fenx.com

---------



From greg at cosc.canterbury.ac.nz  Wed Feb 27 20:48:38 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 28 Feb 2002 14:48:38 +1300
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3C7C6252.FC779971@cosc.canterbury.ac.nz> <m37koz2trp.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3C7D8C76.1918CE52@cosc.canterbury.ac.nz>

"Martin v. Loewis" wrote:
> 
> Notice that regular expressions also define LL(1) languages,
> so the entire pgen parser is a "regular expression matching engine"
> :-)

Well, yes. But what I was trying to say is that the
Python parser generator must be doing some sort of
NFA-to-DFA processing on the productions in order
to generate its tables.

If you want to feed the Python grammar into
something which doesn't understand productions with
REs in them, you have to split things like

  arglist: (argument ',')* (argument [',']| 
    '*' test [',' '**' test] | '**' test)

up into several productions. If you do that naively,
you get something like

  arglist: argspart1 argspart2
  argspart1: argument ',' argspart1 | empty
  argspart2: argument opt_comma | '*' test opt_kw | kw
  opt_kw: kw | empty
  kw: '**' test
  opt_comma: ',' | empty

which is ambiguous. To get it into an unambiguous LL(1)
form, some rearrangement is necessary.

I don't think it really makes sense to ask whether
the Python grammar *as written in the Grammar/Grammar 
file* is LL(1). The textbook theory about classification
of grammars all assumes that the grammar is written
using the simple Yacc-like form of productions.

The Python Grammar file is written in a higher-level
language than that, and happens to have been written
in a way that does not admit a trivial translation
into a lower-level form that is LL(1).

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


From aahz at panix.com  Mon Feb 11 09:51:02 2002
From: aahz at panix.com (Aahz Maruch)
Date: 11 Feb 2002 06:51:02 -0800
Subject: XP (was Re: Defending the Python lanuage... )
References: <3C62EE4A.5090305@destiny.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com> <mailman.1013434699.26587.python-list@python.org>
Message-ID: <a48lom$1bn$1@panix2.panix.com>

In article <mailman.1013434699.26587.python-list at python.org>,
Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
>
>I've lived through too many "paradigm shifts" already.  I remember
>the fiasco of TQM being applied to a scientific laboratory, where
>commitment to quality was already so high that the technicians were
>insulted.  Managers blindly applying buzzwords like Band-Aids makes
>me curse... whether it's TQM, or Java, or XP, *no* "new way" will
>work for everyone.
>
>It's just not a one-size-fits-all world.

Which reminds me to tell everyone who hasn't picked up a copy of the
25th anniversary edition of _The Mythical Man-Month_ to go out and get
it immediately.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From aahz at panix.com  Wed Feb 13 19:08:26 2002
From: aahz at panix.com (Aahz Maruch)
Date: 13 Feb 2002 16:08:26 -0800
Subject: Python License Issues
References: <mailman.1013627905.10961.python-list@python.org>
Message-ID: <a4ev5q$eti$1@panix2.panix.com>

In article <mailman.1013627905.10961.python-list at python.org>,
 <christoforou at gaaros.com> wrote:
>
>Some clients demand a certain level of 'protection' from situations    
>of gross misconduct, say python has stolen major chunks of code        
>from other apps and thus exposing them to huge legal risks. We can     
>consider accepting liability on all third party code we use, for       
>such gross misconduct situations but is it technically possible to     
>do so given the PSF license? Ideally we would like to negotiate a      
>single license agreement with these clients (our license agreement)    
>which provides additional liability over and above to that which PSF   
>provides. However given clause 4 it seems that the client *must*       
>directly be bound by the PSF license.                                  
>
>Any comments?

Yeah.  You need to hire an intellectual property lawyer with experience
in Open Source issues.  I don't know how many of those are on the other
side of the pond.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"Twenty million slave girls can't be wrong."  --SJM


From theller at python.net  Thu Feb 28 03:13:12 2002
From: theller at python.net (Thomas Heller)
Date: Thu, 28 Feb 2002 09:13:12 +0100
Subject: distutils source dist question
References: <mailman.1014848148.17200.python-list@python.org>
Message-ID: <a5kos8$7ld35$1@ID-59885.news.dfncis.de>

"Skip Montanaro" <skip at pobox.com>:

> I have a small library consisting of three C source files:
>
>     alf.c
>     alf.h
>     alfmodule.c
>
> The first two files are a standalone library, while the third is an
> extension module wrapping the library.
>
> The naive "python setup.py install" command creates alfmodule.so and
> installs it in site-packages.  I'd like libalf.so or libalf.a to also get
> created and installed in ${prefix}/lib and alf.h to be installed in
> ${prefix}/include.  Can this be done with distutils or do I need to fall
> back to Makefile for this?

This (untested) script

setup(name="alf",
      ext_modules=[Extension("alf",
                             sources=["alfmodule.c"],
                             libraries=["libalf"]),
      libraries=[("libalf", {"sources": ["alfmodule.c"]})],
      headers=["alf.h"])

would probably get most what you want out of distutils,
but there are some problems.
It would build your library and your extension module,
it would install "alf.h" in the alf subdirectory of
${prefix}/include, it would *not* install the libalf
library in the python tree somewhere, and you would
have problems running the sdist command, because the clib
option is somewhat broken (this is even mentioned in the
docs).
All in all, it may be  a start...

Thomas




From sdm7g at Virginia.EDU  Sun Feb  3 18:43:21 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Sun, 3 Feb 2002 18:43:21 -0500 (EST)
Subject: [OSX] milter won't build
In-Reply-To: <644f6688.0202031454.51623317@posting.google.com>
Message-ID: <Pine.OSX.4.43.0202031826200.4385-100000@d-128-61-180.bootp.virginia.edu>


On 3 Feb 2002, rdack wrote:

> trying to build milter module 0.3.9 on OSX 10.1.2 i get this error:
> cc -bundle -undefined suppress build/temp.darwin-5.2-Power
> Macintosh-2.1/miltermodule.o -lmilter -lsm -o
> build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
> /usr/bin/ld: -undefined error must be used when -twolevel_namespace is
> in effect
> error: command 'cc' failed with exit status 1
>
> is it telling me i need  an '-undefined error' option? or that it
> doesn't like the 'undefined suppress' option?

I don't know what the milter module is, but:

OSX gcc/ld doesn't like that combination. You can use either of:
	-twolevel_namespace -undefined error
	-flat_namespace -undefined suppress
But I think the defaults are:
	-undefined error
	-twolevel_namespace
So, you get generate that error with only one of those switches
with the wrong default for the other.

Python modules on OSX is usually built with:
	-bundle -flat_namespace -undefined suppress


> where is the 'cc -bundle' command set up? can i edit it?
> is it possible that it finished and the module is ready to go?

Is milter using distutils ? i.e. is it using a 'setup.py' file to
build and install, or does it use a Makefile ?

What command did you use to build it?

(Distutils should get the build command right, but if you have more
than one version of Python installed, you need to make sure that
you're running the version that you're going to link to. )

> where would the milter module end up after install?
> what will its name be?
> are python modules '.pyo' files? do they end up in the
> /usr/local/lib/python2.1 directory?

If it's a dynamic shared lib, it'll usually be an .so file, and it
should usually install in the site-packages directory, likely in
/usr/local/lib/python[v.n]/ . ( It might also go in plat-darwin*/
or somewhere else in that tree. )

Are you building 2.1.2 ? or just 2.1 ?
You probably ought to stick to 2.1.2 or 2.2 on OSX.




From mwh at python.net  Thu Feb  7 08:07:37 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 13:07:37 GMT
Subject: High-level languages, large projects, GNOME and the .NET controversy
References: <23891c90.0202070501.78df2371@posting.google.com>
Message-ID: <u3d0da15e.fsf@python.net>

paul at boddie.net (Paul Boddie) writes:

> It occurs to me that there are still many people who choose not to use
> languages of a higher level than C/C++ for development, and it would
> be interesting to know what their reasons are. Of course, writing a
> graphical toolkit *purely* in Python is somewhat ambitious, and there
> are clearly performance issues to be considered, but I still can't
> understand why people would want to write *entire* applications (of
> the nature discussed above) in C/C++ these days.
> 
> Can anyone enlighten me?

No, but I share you're bewilderment.

I've given my sigmonster some help for this post :)

Cheers,
M.

-- 
  /* I'd just like to take this moment to point out that C has all
     the expressive power of two dixie cups and a string.
   */                       -- Jamie Zawinski from the xkeycaps source


From dave at pythonapocrypha.com  Tue Feb 26 22:31:08 2002
From: dave at pythonapocrypha.com (Dave Brueck)
Date: 26 Feb 2002 19:31:08 -0800
Subject: Need advice for obfuscating Python code
References: <Pine.LNX.4.33.0202261511020.13167-100000@ns0.tbye.com> <mailman.1014764866.8830.python-list@python.org>
Message-ID: <477762c2.0202261931.4348ff49@posting.google.com>

Richard Jones <rjones at ekit-inc.com> wrote in message news:<mailman.1014764866.8830.python-list at python.org>...
> On Wed, 27 Feb 2002 10:28, brueckd at tbye.com wrote:
> >
> > [snip solution which encodes the module bytecode in some manner]
> >
> > That should work. Of course, since the code that decrypts and loads your
> > code will be visible, you probably want to use a long key (or another form
> > of encryption) and install the customer importer from a C program that
> > runs an embedded interpreter (installs the importer and then runs the rest
> > of your program).
> 
> Or you could bundle the encoded bytecode in a C module, which makes the 
> solution a lot neater.

Except that then you have to recompile your executable each time any
of your Python modules change, whereas with the above method you can
reuse the same executable until you need to change the key - probably
0 times during the course of development.

Like I said in my original reply though, a good license is usually
more effective anyway.

-Dave


From xscottgjunk at yahoo.com  Fri Feb 22 13:23:03 2002
From: xscottgjunk at yahoo.com (Scott Gilbert)
Date: 22 Feb 2002 10:23:03 -0800
Subject: Embedding: How to set globals for PyObject_Call
References: <mailman.1014234635.18641.python-list@python.org> <79b2b0bb.0202210841.457c0b81@posting.google.com> <uit8qlqdm.fsf@python.net> <79b2b0bb.0202211726.7e8ab7f1@posting.google.com> <u664plra0.fsf@python.net>
Message-ID: <79b2b0bb.0202221023.19736775@posting.google.com>

Michael Hudson <mwh at python.net> wrote:
> 
> Oh, I see what's going on.  Another question: what would you expect
> locals() to return?
> 

Well at the least, it should return the same thing it does when you
type locals() at the Python command prompt.  There it is just an alias
for globals().

But for my specific needs (using Python as a high powered
data-structures and expression-evaluator library from C++), I was
planning on having the globals dict be shared among all threads and
having a locals dict per each thread.  I can do this with
PyRun_String(...) because it has me pass in both the globals and
locals.

If I could figure out how to write:

  PyObject* PyObject_CallInScope(
    PyObject* callable, PyObject* args, PyObject* kwds,
    PyObject* globals, PyObject* locals
  );

I'd be happy as a peach.  But so far I've gotten lost trying to
understand what's going on with CodeObjects, ThreadStates,
FrameObjects etc...

> 
> > Well, I hope this is sufficient info if you view this as enough of a
> > problem to fix it.  Thanks for your interest.
> 
> It's probable that Python shouldn't core in these circumstances, at
> least.
> 

:-)

Have a good one and thanks again,
    -Scott


From emile at fenx.com  Sat Feb  2 10:18:55 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 2 Feb 2002 07:18:55 -0800
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <a3h0ap$16taj0$1@ID-11957.news.dfncis.de>

"Fred" <c at tech.usd253.org> wrote in message
news:e959e7b.0202020710.417801c8 at posting.google.com...
> Hello, I am just starting out with Python. I have a very basic book
> that is not helping me much. I have done some simple C programming in
> the past and I think if I could see the below C code done in Python it
> would get me over the beginners hump. I tried to do it in Python but
> got stuck right at the start, Python did not like it when I tried to
> say: for i = 1
>

Take a couple of hours and work through the tutorial.  In there, you'll
find out that a loop in python is spelled 'for i in...' and not 'for
i=...'

HTH,


--

Emile van Sebille
emile at fenx.com

---------



From emile at fenx.com  Wed Feb  6 14:03:02 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 11:03:02 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu> <a3rrii$1ahpac$1@ID-11957.news.dfncis.de>
Message-ID: <a3rv7p$1ac6vd$1@ID-11957.news.dfncis.de>

 "David Eppstein"
> Isn't all that list-copying (ary[1:]) going to make for a very
inefficient
> algorithm?

OK... this compares in timing to yours and doesn't depend on recursion
depth or generators:

def flatten(ary):
    try:
        if ary[0] == ary:rslt.append(ary)
        else:
            flatten(ary[0])
            flatten(ary[1:])
    except:
        if ary: rslt.append(ary)

rslt = []
flatten(("cat",5,['dog',[3,3,1]],"zoo"))
print '\n\n',rslt

I knew it was a stinker when I posted it, but I liked the deceiving
simplicity of the code.  ;-)

Apologies to all...

--

Emile van Sebille
emile at fenx.com

---------



From bokr at oz.net  Tue Feb 19 00:10:34 2002
From: bokr at oz.net (Bengt Richter)
Date: 19 Feb 2002 05:10:34 GMT
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com> <a4rsa2$2grrk$1@ID-59885.news.dfncis.de>
Message-ID: <a4smoa$h20$0@216.39.172.122>

On Mon, 18 Feb 2002 22:38:46 +0100, "Thomas Heller" <theller at python.net> wrote:

>
>"David Bolen" <db3l at fitlinxx.com> wrote in message news:u4rkezl0l.fsf at ctwd0143.fitlinxx.com...
>> The binary portion of the windows specific installer is hardcoded into
>> the distutils package (encoded in the bdist_wininst.py module), and it
>> doesn't appear that source is available (at least I can't seem to find
>> it even in the CVS tree) so changing how it behaves doesn't seem easy.
>> You could replace it outright, but then you'd have to duplicate its
>> existing functionality as well.
>
>The source code IS available:
>http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/distutils/
>in the misc directory. Guess I'll have to add a note about this
>to the source... There is a tricky symbolic link somewhere on SF
>which links parts of the distutils tree into the Python source tree.

That self-update of the bin64 EXEDATA string seems a little tricky too ;-)
I guess I would rename the old version to a backup name before writing
the new, not being aware of the larger context. Or should one
copy it to ../otherdir and run ../otherdir/it that way? Or did I read
too fast?

>
>> Having some additional hooks created in distutils to execute
>> post-installation sounds like it could be a useful enhancement.  At
>> least with that sort of scenario you could write your own shortcut
>> creating code without modifying the core distutils itself.
>
>I'm currently working on an bdist_wininst enhancement which
>takes an additional command line option (--install-script <script.py>).
>This script would then be run (with sys.argv set to [script_name, '--installed'])
>after all the files have been copied, and it would again be run
>(with sys.argv set to [script_name, '--removed']) just before
>the deinstaller is about to remove all the files and directories.
>
Sounds good. I wonder if something is needed to cater to version upgrading
and/or multiple version installation. E.g., to give an old version a
chance to have a script automatically executed '--before_upgrade' and
'--after_upgrade'? Can the binary installer run something before even
copying any new files? I.e., the old version interpreter and its
'im_about_to_have_a_sibling.py' script? Putting it in the installer
sequence would protect against user forgetfulness (otherwise s/he could
obviously do it manually), and an old version could at a minimum output
some attention-grabbing reminders from the authors of the old version.

Just a couple .01USD

Regards,
Bengt Richter



From claird at starbase.neosoft.com  Mon Feb 18 15:45:55 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Feb 2002 14:45:55 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <a4j24g$jao$9@bob.news.rcn.net> <3C6DA497.9AB8D624@ev1.net> <a4ll2i$6bt$10@bob.news.rcn.net>
Message-ID: <809347551A7862CF.C187F8041154543E.7DFBAE965B423080@lp.airnews.net>

In article <a4ll2i$6bt$10 at bob.news.rcn.net>,  <jmfbahciv at aol.com> wrote:
			.
			.
			.
>I don't know what a "trie" is, but I do (make that, did) have 
>the _Hackers_ book.  I don't know where these memoes came from.
			.
			.
			.
A kind of data structure (a tree with certain data) good for storing
string data <URL: http://hissa.nist.gov/dads/HTML/trie.html >.
-- 

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


From trentm at ActiveState.com  Wed Feb 20 16:30:56 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Wed, 20 Feb 2002 13:30:56 -0800
Subject: HTMLifyer for Python code?
In-Reply-To: <60FB8BB7F0EFC7409B75EEEC13E2019221513D@admin56.narex.com>; from BPettersen@NAREX.com on Wed, Feb 20, 2002 at 01:55:51PM -0700
References: <60FB8BB7F0EFC7409B75EEEC13E2019221513D@admin56.narex.com>
Message-ID: <20020220133056.A20550@ActiveState.com>

On Wed, Feb 20, 2002 at 01:55:51PM -0700, Bjorn Pettersen wrote:
> I just browsed through the Vaults, and did a quick Google search, but I
> couldn't find a tool for converting Python source code to
> syntax-highlighted HTML. Surely someone must have written something like
> this?

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52298

Cheers,
Trent

-- 
Trent Mick
TrentM at ActiveState.com



From cfelling at iae.nl  Thu Feb 21 10:07:22 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 21 Feb 2002 16:07:22 +0100
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <a532fa$45a$1@animus.fel.iae.nl>

Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote:
...nice introduction snipped

Welcome aboard.  Le'me start with complimenting you with your
questions.  They show a keen understanding of programming and Python
in particular not often find with people claiming not to do much
programming.  That will make anwsering them much easier:)
I'll anwser the easy ones.

> 1. The way I thought that an assignment like

> var += 1 

> worked was that the interpreter would go to the place pointed by var and
> add 1. In particular, id(var) should remain constant. But this is not
> what happens as a simple test shows it. So what is wrong in my
> perception of what += does? Is this an implemetation issue?

var is indeed `pointing to' something, but that something here isn't
something that can increment itself.  It's a number, and numbers are
immutable objects in Python.

In general augmented assignment works like:

   var <operator>= <expression>   <==>  var = var <operator> <expression>

but for those var's that have redefined that augmented assignment operator
like in:

   class AA:
	def __init__(self, value):  self.value = value
        def __iadd__(self, value):  self.value += value; return self

   a = AA(1);   b = a;   a += 2
   assert a.value == 3  and a is b


> 2. What is the idea of having in the loop

> for var in range(10):
>     pass

> the var variable still available after it? Wasn't it supposed to die (as
> in being out of scope) when the loop ends? And is this a feature, or
> something that we should not count on? Somehow, I believe that the a new
> var should be created when the loop starts and then destroyed when it
> ends avoiding any variable shadowing along the way.

In python the for statement doesn't start a new scope. This comes in
handy when e.g. searching a list, like:

    for line in lines:
	if line.startswith("spam"): break
    # now line is thee first line in lines that starts with "spam".

> 3. This is not a question just a small complaint. I really miss the
> do-until control flow in Python. I always have to waste some extra brain
> cycles to convert it to the while 1, etc. idiom. Hey, we all have a
> right to our complaints, right? ;-)

do-until is not my favorit, but I do miss the more generic
loop-and-a-half construct:)


> 6. Is there a reason why 3.__abs__() does not work? After all 3 is an
> object (in the wider sense of the word). And it does work if we assign
> it to a variable, eg, 

> var = 3
> var.__abs__()

> I know that code like the above (methods applied to number literals) is
> probably rare, but consistence is also a good thing.

It's a parser thing.  `3.' would be parsed as a real.


> 9. What would be better in terms of speed?

> for i in range(<whatever>):
>     <whatever> 

> Or to code an iterator like

time it:)


> 10. I have a class that works like a list in the sense that supports the
> method __getitem__ and its relatives. But I'm at a loss as to what I
> have to do to support the slice notation.

In the __getitem__ method check whether tha argument passed is a slice
object and act accordingly.

> 11. Generators seem to be a very powerful concept. Still,

> a. Why can't a return statement have an expression?

It's deemed wise to make a clear distinction between `yielding' and stopping.

> b. Is there any way to "reset" a generator, e.g. upon a subsequent call
> it starts as if it is being called for the first time?

Not straightforward, but one could concieve a generator that after
each yield checks for some `global' flag whether it should start
afresh. like:

   class ResettableGenerator:
       def gen(self, N):
          self.again = 0
          while 1:
             for n in range(N):
                 yield n
                 if self.again:
                    break
             if not self.again:
                break

> 12. This is not exactly about Python, but about programming in general.
> I keep reading about continuations and coroutines and I would like to
> know, in simple terms if possible, what the heck are they and in what
> ways they differ from generators as they exist in Python.

Continuations give you the possibility to save all/most of a program
state go on whith the program and later on resume from the saved state.

Coroutines give you a form of pseudo parallellism, they are capable
of passing control over to an other coroutine and resume where they
left once control is passed back to them again.  When control is
passed over it's possible to pass/receive information from/to the
other coroutine.
-- 
groetjes, carel


From philh at comuno.freeserve.co.uk  Tue Feb 12 10:53:33 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Tue, 12 Feb 2002 15:53:33 +0000
Subject: Coming of age...
References: <mailman.1013466258.26610.python-list@python.org>
Message-ID: <slrna6iejt.2so.philh@comuno.freeserve.co.uk>

On Mon, 11 Feb 2002 14:20:18 -0800, Paul Prescod <paul at prescod.net> wrote:
>Someone asks in Slashdot (to paraphrase): "What's the best language." Of
>course it's a pretty stupid question.

Of course it's a stupid question, when the answer's obvious:














[spoiler space]


















[more spoiler space]




















[even more space]


















[bear with me, just a bit longer now]





It's Tim Rue's Autocoding!! Except that isn't a language, of course, it's
a, er, well, it's something, whatever.


> In fact the way he asks it is even
>worse, but anyhow, the interesting part is the number of references to
>Python in the followups. There are lots. And they are moderated up,

That's the work of the Propaganda Section of the Python Secret Underground,
no doubt.

>despite being (in at least one case) even more vacuous than the
>question. That means that the moderators are in favour of Python also.

Of course they are. Everyone is! Python is wonderful! Guido I want to
have your babies! (OK I'm need a sex change first)

>This is one more datapoint over a few I've collected in the last few
>weeks that somehow Python has crossed some kind of line from being a
>"like to try it someday" language to an "everybody cool is using it
>these days" language.

Yeah, I think we are getting there. We're not quite there yet IMO.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From seebs at plethora.net  Sun Feb  3 19:47:37 2002
From: seebs at plethora.net (Peter Seebach)
Date: 04 Feb 2002 00:47:37 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net>
Message-ID: <3c5dda29$0$79560$3c090ad1@news.plethora.net>

In article <1173.799T632T11653818threeseas at earthlink.net>,
Timothy Rue <threeseas at earthlink.net> wrote:
>Don't like the idea of autocoding huh?

Haven't been able to perceive whether or not you have an idea.  I have to
say, you're talking to the wrong crowd.  You need people less mired in
conventional ideas of software engineering.  Your best bet is almost certainly
the C# people; they are certainly not mired in conventional ideas of software
engineering.

-s
-- 
   Copyright 2001, all wrongs reversed.  Peter Seebach / seebs at plethora.net
   $ chmod a+x /bin/laden      Please do not feed or harbor the terrorists.
     C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/


From gimbo at ftech.net  Thu Feb 21 09:50:33 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Thu, 21 Feb 2002 14:50:33 +0000
Subject: MySQLdb
In-Reply-To: <XD7d8.203$zZ4.2649220@newsserver.ip.pt>
References: <XD7d8.203$zZ4.2649220@newsserver.ip.pt>
Message-ID: <20020221145033.GC9398@andy.tynant.ftech.net>

> Where i can find an "how to" to learn working with MySQLdb module.
> I have downloaded from sourceforge and install it, but there is no
> much information on how to use it.  Any one knows where i can get
> it?

(Think I just sent this to the wrong address by mistake, but apologies
if it comes through twice...)

http://python.org/topics/database/DatabaseAPI-2.0.html

http://sourceforge.net/docman/?group_id=22307

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From henrik.motakef at web.de  Fri Feb 15 19:48:03 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 16 Feb 2002 01:48:03 +0100
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net>
Message-ID: <86n0yaw8u4.fsf@apo.hmotakef.int>

Hans Nowak <wurmy at earthlink.net> writes:

Hi.

First of all, I have to admit that I'm not sure what your problem
is. :)

However, one thing that might possibly be related is that XML parsers
are required to convert all of the funky end-of-line markers around
(i.e. "\r\n", "\n", "\r") to a single newline "\n". Scince you Use SQL
Server, I assume that you are also running Windows, where the
"official" newline is "\r\n". Could that be related to your problem?

hth
Henrik


From starx at pacbell.net  Wed Feb  6 15:51:16 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 20:51:16 GMT
Subject: get all paths
References: <d5534716.0202061219.60f2af7e@posting.google.com>
Message-ID: <Xns91AD82BFE109RASXnewsDFE1@130.133.1.4>

Ed || Wed 06 Feb 2002 12:19:13p:

> Hi
> I'm trying to create a treeview widget, but i am having problems
> getting alll directories and files from a specified directory
> 
> i.e
> 
> getList("C:/") should return all directories and files from C:/
> onwards whereas
> getList("C:/Python") should return all from that onwards
> 
> any help would be appreciated

Well, for web site maintance, i've written a few fuctions that recruse 
over a tree and apply other scripts to files with certain extentions. 
(i.e. turn .sxhtml into .html by combining with a .msxhtml template :), 
and this is about how i've gone about doing it. It's probably not the 
best way, but you can probably change it to do what you need.

----
import os

def getList( path, ):

    list = []
    os.chdir( path )
    dir = os.listdir( '.' )
    our_dir = os.path.abspath('.')

    for name in dir:
        list += [os.path.abspath(name)]
        if os.path.isdir(name):
            list += getList(os.path.abspath(name))
            os.chdir( our_dir )

    return list
----

and then this bit of code

x = getList ('c:/desk/sproj')
x.sort()
for i in x:
    print i

produces
c:\desk\sproj\MANIFEST
c:\desk\sproj\calls.stats.sx
c:\desk\sproj\error.sx
c:\desk\sproj\gfx
c:\desk\sproj\gfx\default.png
c:\desk\sproj\gfx\tile
c:\desk\sproj\gfx\tile\floor.png
c:\desk\sproj\gfx\tile\wall.png
c:\desk\sproj\gfx\tile\wall2.png
c:\desk\sproj\globals.py
c:\desk\sproj\globals.pyc
c:\desk\sproj\nameparts.sx
c:\desk\sproj\notes.sx
c:\desk\sproj\profile.temp.dat
c:\desk\sproj\setup.py
c:\desk\sproj\sproject.py
c:\desk\sproj\stats.1.sx
c:\desk\sproj\stats.2.sx
c:\desk\sproj\stats_examine.py
c:\desk\sproj\std.stats.sx
c:\desk\sproj\sxrandom.py
c:\desk\sproj\tdrep.py
c:\desk\sproj\tdrep.pyc
c:\desk\sproj\test.py
c:\desk\sproj\tile.py
c:\desk\sproj\tile.pyc
c:\desk\sproj\todo.sx

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From ivano.delpiccolo at degroof.be  Thu Feb 14 03:30:19 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Thu, 14 Feb 2002 09:30:19 +0100
Subject: spawn and standard output
References: <ce0b6758.0202131543.1707e80c@posting.google.com>
Message-ID: <3c6b7581$0$12226$4d4efb8e@news.be.uu.net>

You can also redirect your command to a specific file and read it.
example :
os.system("ls -la>foo.txt")
after you open the file foo.txt and you read it.

The problem of using that method, stdout and stdin are in the same file.
When you read your file you have to do treatment to detect if the
information is coming from stdout or stderr.


"Nigel Moriarty" <nw_moriarty at yahoo.com> wrote in message
news:ce0b6758.0202131543.1707e80c at posting.google.com...
> Hi all,
>
> When I spawn a process from a python script on MSW, the standard
> output doesn't go to the terminal where I started the script.  I would
> like to view this output for debugging.  Where does the output go and
> how can I view it?  Do I need to use popen or can I redirect to a
> file?  Any other suggestions?
>
> Nigel




From jessw at loop.com  Mon Feb  4 02:47:31 2002
From: jessw at loop.com (Jesse W)
Date: Sun, 3 Feb 2002 23:47:31 -0800
Subject: I love obfuscated python!
Message-ID: <3C5DCC13.19141.4B2B23@localhost>

To all Pythonicks-

	Horrey!  I have finally solved the piece of obfuscated python 
posted by Greg Stein in November of 1998.*  I stumbled on it 
searching through Google's archive and decided to try to solve it.  I 
noticed there was a solution, but I have not yet looked at it.
	That code is simply amazing.  Not only does it obfuscate itself, 
but it obfuscates it's obfuscation... (I won't say more because it 
would spoil it for others.  See the above mentioned solution message 
if you are interested)
	I really believe that obfuscated code is a wonderful and 
honorable hackish art form.  It shows off the skill of the creator, and 
rewards the solver with a distinct feeling of accomplishment.  Like 
many hacker traditions, it is a showing of skill, but does not put 
down those who it is shown too(as long as they put in the effort to 
solve it)  As the subject lines says, "I love obfuscated python!"

		Glory and love to Python and all it's works,
				Jesse Weinstein

#The horrifying code itself:
f=lambda 
x=['31312405171810171211313126112','7189749374747628707','96
26701', 
'8873809195908988937382747529513','80797974818230697965292
2711', '25141421122626106','hi there!'], z=globals(),y= lambda x: 
reduce(lambda x,y:x+y, map(lambda x,y=x,z=lambda 
x:int(x[:2]):chr((z(y[x*int(y[-5]):])^z(y[-4:]))+z(y[-4:])),range(int(x[-
2:])))):map(lambda 
z,x=y,y=x,w=z:z(z(w[x(y[2])],x(y[3]))[0],x(y[5]))(x(y[4])),[vars(z[y(x[0]
)])[y(x[1])]]) and x[-1];print f()
#By Greg Stein

* See
http://groups.google.com/groups?oi=djq&selm=an_398922825
for the actual message.



From justin at iago.org  Thu Feb 21 10:56:10 2002
From: justin at iago.org (Justin Sheehy)
Date: Thu, 21 Feb 2002 10:56:10 -0500
Subject: Adding static typing to Python
In-Reply-To: <a52rbc$p1i$1@thorium.cix.co.uk> (gbreed@cix.compulink.co.uk's
 message of "Thu, 21 Feb 2002 13:05:48 +0000 (UTC)")
References: <mailman.1014225213.11001.python-list@python.org>
	<a52rbc$p1i$1@thorium.cix.co.uk>
Message-ID: <x7wux64yn9.fsf@ra.iago.org>

gbreed at cix.compulink.co.uk writes:

> It looks like "strong typing" doesn't have a universally agreed
> definition, so I suppose the best thing is to avoid the term.

Sounds good to me.  I suspect that everything necessary can be
expressed with combinations of "static", "dynamic", and "weak".
("weak" can be used along with "static" to describe languages like C)

> Out of general interest, does anybody know of a dictionary I *can* rely on 
> for these kind of definitions?

Sadly, no.

> If the type was declared only for the initial assignment, for example
>
> def cons(element, stuff: tuple):
>   stuff = list(stuff)
>   stuff.append(element)
>   return tuple(stuff)
>
> would that still be static typing?

Most languages with static typing only require you to declare a given
name's type once.  However, they also tend to enforce that you not
then rebind the name to a value of another type.  So, I'm not sure how
I would describe the behavior shown in the example.  That doesn't mean
that it doesn't make sense, but that I don't know how to classify it.

> Yes, and the example was of an identifier being bound to a different type. 
> I never said any objects were changing type. 

My apologies.  I misread your description and went straight to the example.

-Justin

 




From karsce at usa.net  Sat Feb  2 15:29:52 2002
From: karsce at usa.net (Subramaniam Karthik)
Date: 2 Feb 2002 15:29:52 EST
Subject: Clipboard viewer
Message-ID: <20020202202952.16168.qmail@cpdvg100.netaddress.usa.net>

Hi,
   I am trying to implement a clipboard viewer in python. and i need to know
which application posted the data to the clipboard. any ideas as to how this
can be done? the GetClipboardOwner() just returns the window handle and not
the application name.

Thanks
karthik

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

Subramaniam Karthik

M.S Computer Science            |     1701, N.Tyndall Ave,
Department of Computer Science  |     Apt # R1S
University of Arizona           |     Tucson , Arizona
Tucson, Arizona                 |     Ph: (520) 903-1547
Ph: (520) 626-8785              |

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

____________________________________________________________________
Get free e-mail and a permanent address at http://www.amexmail.com/?A=1



From patio87 at aol.com  Sun Feb 10 21:55:59 2002
From: patio87 at aol.com (Patio87)
Date: 11 Feb 2002 02:55:59 GMT
Subject: Random Module help
Message-ID: <20020210215559.13516.00000455@mb-cq.aol.com>

I was wondering if any 1 could tell me how to use the random module for a
blackjack game i want to make. I am a begginer so try to be easy with the
dialouge. Thanks


From karthik at james.hut.fi  Thu Feb 28 07:59:45 2002
From: karthik at james.hut.fi (Karthikesh Raju)
Date: 28 Feb 2002 14:59:45 +0200
Subject: Numeric Python Installation woes
References: <m3elj6ocvs.fsf@mycomp.tky.hut.fi> <cf0ad9fb.0202280307.2951ec2c@posting.google.com>
Message-ID: <2dg03lvk1q.fsf@james.hut.fi>

Hi,

The second line that had the bigger error message was with python
2.2. Somehow python 2.2 told me that there were some dependency
problems with crypt, readline, and ssl and i thought that these were
specialized packages and ignored them. 

i dont know about this distutils stuff. i did the installation as
superuser, so i guess it should put things where it should. 

For some reasons even after upgrading to python 2.2 python 1.5 is
still there in the bin.

Wishes and thankx in advance
karthik

-- 

-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi		
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------


From calves at coelce.com.br  Sat Feb 23 08:50:04 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Sat, 23 Feb 2002 10:50:04 -0300
Subject: RES: BLT Setup and Installation
Message-ID: <29A97D00F387D411AC7900902770E148040DFAC5@lcoeexc01.coelce.net>

Thanks for your good help. But, unfortunetly, nothing of that can solve my
problem.
I just get it working after dowloading the tcl/tk8.3.4 installer from
activestate site and installed that under C:\Python21\tcl-re directory
(tcl-re could be anything else. I just used this name because I already have
one C:\Python21\tcl directory - wich it was made by standard python21
installation).
By the way, is it means tcl and tk packages that came with python21
installer it isn't enough?!

> -----Mensagem original-----
> De: Howard Lightstone [mailto:howard at eegsoftware.com]
> Enviada em: quinta-feira, 21 de fevereiro de 2002 14:18
> Para: python-list at python.org
> Assunto: Re: BLT Setup and Installation
> 
> 
> "Alves, Carlos Alberto - Coelce" <calves at coelce.com.br> wrote in
> news:mailman.1014305575.22017.python-list at python.org: 
> 
> > Hi all,
> > How can I install and setup the BLT extentions to tcl/tk to 
> use in my
> > python programs.
> > I have downloaded the Pmw and BLT packages (binaries forms) 
> and unzip
> > them into C:\Python21 and C:\Python21\tcl. But when try the 
> following
> > code, it crashs:
> > 
> > from Tkinter import *
> > import Pmw
> > root=Tk()
> > g=Pmw.Blt.Graph(root)
> > Here, the system raise and error and say that 
> "::blt::graph" commands
> > not found, when callig Tk/tcl.
> > 
> > I'm running Python2.1 under Windows98, wich it's installed at
> > C:\Python21 
> 
> I just had to do this myself for Python2.0.1 but it's the same----
> 
> BLT asks for the path where TCL is installed on the machine.  
> 
> I changed that to match the Python tcl path (c:\pythonxx\tcl) and 
> completed the installation.
> 
> Then I copied the BLT 2.4 directory from ----tcl\lib 
>                                     to   ----tcl\tcl8.3
> 
> Copy the blt24.dll from --tcl\bin to \windows\system.
> 
> This worked even when I installed python in a different place.
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020223/858f32e5/attachment.html>

From sholden at holdenweb.com  Fri Feb  8 09:41:30 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 09:41:30 -0500
Subject: Webizing Python
References: <a4032v01prd@enews3.newsguy.com>
Message-ID: <hsR88.105986$KM2.4047777@atlpnn01.usenetserver.com>

"Greg Krohn" <infinitystwin.SPAM at IS.BAD.yahoo.com> wrote in message
news:a4032v01prd at enews3.newsguy.com...
> I don't remember any threads about this, so please forgive me if there has
> been.
>
> In Guido's "The State of the Python Union"
> (http://www.python.org/doc/essays/ppt/python10/py10keynote.ppt) he
mentions
> webizing Python. In particular adding http:/www.python.org/python2.2 to
> sys.path. I would think this would probably be P3K, but does anyone have
any
> info on this yet? It sounds interesting.
>
It's just an idea. It was a reference to Tim Berners-Lee's closing keynote
address, which bore no real relation to Python as it is today, but merely
expressed his desire to incorporate interfaces to the Semantic Web into
Python.

Someone has suggested on this group that the Semantic Web is different from
binary relational information, but if there's no equivalence between
"subject verb object" and "instance relation instance" I'd like to know why
not. I only mention this because I did research into binary relational
knowledge representation way back in the days when knowledge-based systems
were that latest buzz for scoring research grants...

> And seeing as I have very little information, I think it would be a good
> time to start criticizing. I wonder what would happen if something
occurred
> at python.org like hijackers or some honest mistake or accident. I would
be
> putting a lot of faith in the online modules' integrity. I can be
relatively
> certain that a program I run right now will run the same in the morning
(if
> no one touches my computer, of course). This wouldn't be the case if
Python
> were webized. For me, at least.
>
Well, if you've ever downloaded a sourceforge project you've taken the same
risk. When you downloaded Python, did you even bother to check the hash of
your download before you installed it?

Digital signature techniques do make it possible to verify authenticity and
integrity of received information. In fact, you can regard your current
Python installation as a cached copy of the downloadable installer, if you
want to look at things that way. If the sending site has been compromised to
the extent that the hacker has been able to use its private key, naturally
all bets are off.

Obviously the mechanics of importing modules from the web are the least of
the concerns raised by this "proposal" -- and, as Andrew Kuchling has
already pointed out, using a URI to identify modules doesn't imply anything
about physical availability of the information. URI != URL

Before Python *does* support web downloads we will have to put a lot of
solid work in to dependency logging and version control if we are to avoid
the Net-Pythonic equivalent of DLL hell.

wishing-i-could-just-say-"make-it-so"-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py



--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From tatebll at aol.com  Wed Feb  6 07:41:44 2002
From: tatebll at aol.com (Bill Tate)
Date: 6 Feb 2002 04:41:44 -0800
Subject: wxPython book
References: <88bc63c6.0202040505.2a9c3016@posting.google.com>
Message-ID: <cb4ba455.0202060441.54e1d529@posting.google.com>

writeson at earthlink.net (Doug Farrell) wrote in message news:<88bc63c6.0202040505.2a9c3016 at posting.google.com>...
> Hi all,
> 
> I'm busy evangilizing my office co-workers on the wonders of Python
> and wxPython, to which they are responding pretty well. We are
> currently working on a test project that uses both, and the most
> frequent complaint I get is "there's no book on wxPython..... wah wah
> wah". I have seen mention of a book being written, presumably by Robin
> Dunn. Is this true and if so, any idea when it might be coming out?
> 
> Thanks,
> Doug

Doug,
While I would say it would be nice to have a book, suggest you check
out wxDesigner (it still has a way to go in terms of ALL the bells &
whistles but what's there already is pretty damn nice).  I like that I
can use one visual designer for the interface and build the skeleton
of the C++ and Python code from within the same IDE.  It made using
wxWindows/wxPython much more intuitive.  I must say, however, it
really is a bummer when you get those quick responses to questions
from the wxwindows/wxpython discussion list - I much prefer that
"timely and high quality" support you get from commercial vendors :>).
Bill


From martin at v.loewis.de  Wed Feb 20 17:36:18 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 23:36:18 +0100
Subject: Python 2.1.1 on Solaris 7 broken
References: <mailman.1014225333.11561.python-list@python.org>
Message-ID: <m3y9hnixwd.fsf@mira.informatik.hu-berlin.de>

Alex Chavez <alex at laplaza.org> writes:

> ImportError: ld.so.1: python: fatal: libgcc_s.so.1: open failed: No
> such file or directory
[...]
> I'm not sure if the gcc version had anything to do with it but here is
> that info.
> 
> root at elvis cron $ gcc -v
> Reading specs from /usr/local/lib/gcc-lib/sparc-sun-
> solaris2.7/2.8.1/specs
> gcc version 2.8.1

That is likely *not* the compiler you've been using to link Python.
libgcc_s first appears with gcc 3.0, and unless somebody has seriously
broken your gcc installation, gcc 2.8.1 does *not* use a shared
libgcc.

I suggest to remove all gcc installations from your system except for
one, and make that one work, before trying to use it.

Regards,
Martin



From starx at pacbell.net  Tue Feb 12 04:47:13 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 12 Feb 2002 09:47:13 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <eSz98.4759$zb.260473@news1.east.cox.net> <3C682A16.1D6C800D@ccvcorp.com> <Xns91B28C2B6EBE0RASXnewsDFE1@130.133.1.4> <DY_98.1732$v94.371156124@newssvr17.news.prodigy.com>
Message-ID: <Xns91B3122C0BD65RASXnewsDFE1@130.133.1.4>

Steven Rumbalski || Mon 11 Feb 2002 06:19:15p:

> Philip Swartzleonard wrote:
> 
>> Then it becomes 'Guidoista' perhaps, for following a specific leader? ;)
> 
> 'Guidoista': three syllables or four? GWEE-DOY-STAH or GUI-DOH-EE-STAH?
> 

Hm, i was thinking the second, but that may just be because i have a 
tendency to break unknown words into the smallest reasonable sylables 
(probably aqquired learning japanese, or more specificaly, dealing with 
romaji (some thing become obnoxious in this... something like shijichiji 
(not a word AFAIK) looks wretched in a proportional font :)).

Though if you say the second one fast enough, it sounds almost exactly like 
the first....

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From threeseas at earthlink.net  Fri Feb 15 20:45:45 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:45 GMT
Subject: Wow.
References: <MPG.16d23ba6a6f4eb4989681@news.verizon.net>
Message-ID: <934.811T1877T12165131threeseas@earthlink.net>

On 11-Feb-02 20:36:22 A Meowbot <meowbot at meowing.net> wrote:
>In a few weeks of skimming those threads i'd consistently wackyparsed
>the term as "autoconfig." The scary part is that all the flames and
>rebuttals I read made perfect sense in that context.

I'm sure they would make perfect sense in any peer group on any subject
that wasn't wanted.

Why else was there posts to a message on autocoding but not on a prior
post I made here about the same project but with a different theme?

:) When Neo jumped into Agent Smith wouldn't you say Neo Busted Smith?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From osuchw at ecn.ab.ca  Tue Feb  5 10:27:59 2002
From: osuchw at ecn.ab.ca (waldekO)
Date: 5 Feb 2002 07:27:59 -0800
Subject: HappyDoc: Can't get output
References: <fb072181.0202050304.35626d25@posting.google.com>
Message-ID: <c5b3dbee.0202050727.60991126@posting.google.com>

bldrake1 at yahoo.com (Barry Drake) wrote in message news:<fb072181.0202050304.35626d25 at posting.google.com>...
> When I try to run HappyDoc, I get the following:
> 
> commandline:
> c:\python21\happydoclib> python happydocwin.py <anyfile.py>
<snip>
For some reason HappyDoc does not handle windows paths correctly but
if you specify output directory (instead going for default) it works
as expected.
For example
c:\python21\happydoclib> python happydocwin.py -d c:\ <anyfile.py>

waldekO


From ullrich at math.okstate.edu  Mon Feb 11 15:33:58 2002
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 11 Feb 2002 20:33:58 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net>
Message-ID: <3c682a4c.3809135691@news>

On Mon, 11 Feb 2002 10:59:36 GMT, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

>Courageous wrote:
>
>> >Very good example of someone who is afraid of autocoding.
>>
>> You never did answer my question, by the way. If one has
>> become delusional, how does one become aware of it?
>
>for some reason, this whole thread reminds me of this little gem:
>
>    Unskilled and Unaware of It: How Difficulties in
>    Recognizing One's Own Incompetence Lead to
>    Inflated Self-Assessments (Kruger & Dunning, 1999)
>
>    "People tend to hold overly favorable views of their
>    abilities in many social and intellectual domains. The
>    authors suggest that this overestimation occurs, in
>    part, because people who are unskilled in these
>    domains suffer a dual burden: Not only do these
>    people reach erroneous conclusions and make un-
>    fortunate choices, but their incompetence robs
>    them of the metacognitive ability to realize it"
>
>    http://www.apa.org/journals/psp/psp7761121.html

This is just too funny: The other day I couldn't resist pointing
out how much this guy reminded me of that sci.math guy, especially
the bit about how you were just resisting because you had a
vested interest in the status quo. The current funny part is
that exactly the above has been quoted in that other context.

Didn't help.

>of course, Rue is going to argue that this doesn't apply to him, but
>to everyone else.  after all, he KNOWS he's smarter than everyone
>else, and we're just too stupid to realize how stupid we are ;-)
>
></F>
>
>


David C. Ullrich


From grey at despair.dmiyu.org  Fri Feb 15 19:35:38 2002
From: grey at despair.dmiyu.org (Steve Lamb)
Date: Sat, 16 Feb 2002 00:35:38 -0000
Subject: impossible to reboot OS in python?
References: <a4k9as$ak8$1@nntp.itservices.ubc.ca>
Message-ID: <slrna6raav.ji7.grey@teleute.dmiyu.org>

On Fri, 15 Feb 2002 16:33:26 -0800, Kris Botha <3dsphere at home.com> wrote:
> Does it depend on the os installed?

import os
os.system('reboot');

    I'd call that a yes.  :)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
    To email: Don't despair!   |  -- Lenny Nero, Strange Days
-------------------------------+---------------------------------------------


From aahz at panix.com  Mon Feb  4 17:51:53 2002
From: aahz at panix.com (Aahz Maruch)
Date: 4 Feb 2002 14:51:53 -0800
Subject: CPython vs. Jython for Web applications?
References: <c775c8db.0202041126.5177e606@posting.google.com>
Message-ID: <a3n3a9$km7$1@panix2.panix.com>

In article <c775c8db.0202041126.5177e606 at posting.google.com>,
John E. Barham <jbarham at octopuswireless.com> wrote:
>
> - app must run on both Win32 and some free Unix, preferably FreeBSD

The big problem I see is that you're likely to want threading, and my
understanding is that threading doesn't work too well on *BSD.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From bokr at oz.net  Wed Feb 20 16:49:18 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 21:49:18 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl> <xyuvgcujsyl.fsf@sex.ifi.uio.no> <a4shr2$49s$0@216.39.172.122> <a4t9fr$500$1@odysseus.uci.kun.nl> <jwbaxter-020177.08052619022002@corp.supernews.com>
Message-ID: <a515ku$fh1$0@216.39.172.122>

On Tue, 19 Feb 2002 08:05:27 -0800, John Baxter <jwbaxter at spamcop.net> wrote:

>In article <a4t9fr$500$1 at odysseus.uci.kun.nl>,
> "Patrick Vrijlandt" <p.vrijlandt at aig.azn.nl> wrote:
>
>> Although telnetlib seems rather buggy, there is not much request for
>> improvement on c.l.py, so I think it is barely used, or
>> it works fine with most servers.
>
>Well, telnetlib works for the limited purposes we use it for here.  Or 
>does it?  I found the cause (in my code) of fairly frequent false alarms 
>in a monitor I wrote using it, but we still have infrequent false alarms 
>which I haven't yet understood.
>
>  --John
Looking at the rfc, I wonder if the following should be implemented.
It seems to be required for the base level NVT mode with no options:

"""...
         Note that "CR LF" or "CR NUL" is required in both directions
Postel & Reynolds                                              [Page 11]
RFC 854                                                         May 1983
         (in the default ASCII mode), to preserve the symmetry of the
         NVT model.  Even though it may be known in some situations
         (e.g., with remote echo and suppress go ahead options in
         effect) that characters are not being sent to an actual
         printer, nonetheless, for the sake of consistency, the protocol
         requires that a NUL be inserted following a CR not followed by
         a LF in the data stream.  The converse of this is that a NUL
         received in the data stream after a CR (in the absence of
         options negotiations which explicitly specify otherwise) should
         be stripped out prior to applying the NVT to local character
         set mapping.
..."""

On the receive side, dropping nulls in effect does implement it, but the
write side in telnelib.py only doubles IACs.

Also, I didn't see any rfc policy for throwing away ^Q ("\021").
Is that possibly a forgotten expedient test hack?

Regards,
Bengt Richter


From gmuller at worldonline.nl  Fri Feb 22 15:16:28 2002
From: gmuller at worldonline.nl (GerritM)
Date: Fri, 22 Feb 2002 21:16:28 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee> <3C74DD65.3060000@thinkware.se> <m3heoa93gs.fsf@ns.bagley.org> <a53fok$cfg$1@reader1.tiscali.nl> <m38z9m8wqh.fsf@ns.bagley.org>
Message-ID: <a5697i$3or$1@reader1.tiscali.nl>

<usenet+no_spam (" Doug Bagley ") .2002-02-21-2623 @ (the) bagley . (not
.com) org (spam whammy)> schreef in bericht
news:m38z9m8wqh.fsf at ns.bagley.org...
> "GerritM" <gmuller at worldonline.nl> writes:
<...snip...>
> > It would be interesting to have some larger benchmark to compare
language
> > behavior when scaled up in size.
>
> I guess you mean "size of program". I do think that would be interesting.
> But I think that is not easy. Small programs may not be as meaningful,
> but they are certainly more tractable.
>
> Do you have a specific suggestion for a larger program to compare?
>
> cheers,
> doug
> --
> http://www.bagley.org/~doug/contact.shtml

maybe programming contest exercises such as:

http://cristal.inria.fr/ICFP2001/prog-contest/

or

http://contest.uvarov.ru/index.php?level=2&n=1&sn=1&lan=en

can be used. For benchmarking purposes I would remove the strict timing
constraints and strive for the "best" solution,
where "best" will probably be determined by the contributor(s)

For comparison of utilization level still other types of programs are
needed. In this type of comparison I expect even more discussion on the
"rules"; Are you allowed to use the plain language only, or also the
batteries which are included or everything on CPAN like archives; Are
packages which are used included in the line count; etcetera.

regards Gerrit
--
www.extra.research.philips.com/natlab/sysarch





From sjmachin at lexicon.net  Wed Feb 27 06:10:04 2002
From: sjmachin at lexicon.net (John Machin)
Date: 27 Feb 2002 03:10:04 -0800
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com> <c76ff6fc.0202261402.10060a2@posting.google.com>
Message-ID: <c76ff6fc.0202270310.6fe8e6bc@posting.google.com>

sjmachin at lexicon.net (John Machin) wrote in message news:<c76ff6fc.0202261402.10060a2 at posting.google.com>...
> amrit040 at yahoo.com (Amrit) wrote in message news:<71c0a06e.0202260513.4b67841f at posting.google.com>...

>> >       PyArg_Parse (obj[i], "f", &x[i]);
> (1) This is most likely what is causing the exception that you are
> getting -- the first arg to this function should be a *tuple*.

I was wrong -- the first arg can be a single object. However you
should be testing the returned value from PyArg_Parse()...

if (!PyArg_Parse(obj[i], "f", &x[i])) return NULL;


From jason at jorendorff.com  Sat Feb  9 00:04:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 8 Feb 2002 23:04:57 -0600
Subject: RFC -- Hygienic Macros in Python
In-Reply-To: <slrna67k0u.for.philh@comuno.freeserve.co.uk>
Message-ID: <HFEKILOLEFEFMKAECNDLMEFNDBAA.jason@jorendorff.com>

phil hunt wrote:
> Python macros should, if they exist, have a superset of the 
> capabilities of
> C macros. Including the ability to be replaced by arbitrary code in the
> program. Why arbitrary? firstly, because it is conceptually simple;
> secondly because it doesn't restrict the programmer.

C macros are quite restrictive, precisely *because* they simply
paste tokens into your program.

> Consider this code:
>
> defmacro incj:
>    j = j + 1

Well, I don't think that's particularly realistic code.  Consider
this instead.

Jason Orendorff wrote:
>   defmacro synchronized(lock_expr, code_block):
>       lock = ^lock_expr
>       lock.acquire()
>       try:
>           ^code_block
>       finally:
>           lock.release()

Under a non-hygienic macro system, obviously I wouldn't write it
like that because "lock" might clash with a local name.  Indeed if
I have nested "synchronized" statements it's guaranteed to clash.
So what should the code look like in a non-hygienic system?

## Jason Orendorff    http://www.jorendorff.com/



From jason at jorendorff.com  Mon Feb 25 07:47:55 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 06:47:55 -0600
Subject: strings and sort()
In-Reply-To: <e9e6dd1f.0202211235.4e7be9aa@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEPDDEAA.jason@jorendorff.com>

Jason wrote:
> I like Python, but the biggest problem I have with the language right
> now is what low code density it has at times.

Coming from Scheme or Lisp, that is a likely complaint.

Macros let the programmer trim *all* the fat, *all* the
redundancy, from his code.  This means the semantic density
is high.

Typical Lisp style also puts 6 or more tokens on every line,
not counting parens.  This means the visual density is high.

But:
  Flat is better than nested.
  Sparse is better than dense.
  Readability counts.

That's why Python is the way it is.

## Jason Orendorff    http://www.jorendorff.com/



From lee at leeHYPHENphillips.org.invalid  Fri Feb  8 10:21:02 2002
From: lee at leeHYPHENphillips.org.invalid (Lee Phillips)
Date: Fri, 08 Feb 2002 10:21:02 -0500
Subject: boolean values in appleevents - solution
Message-ID: <080220021021023486%lee@leeHYPHENphillips.org.invalid>

   As several people were kind enough to suggest, I joined the
mailinglist at  <http://mail.python.org/mailman/listinfo/pythonmac-sig>
where I indeed got a solution within a day.
   One Vincent Marchetti explains that if you put this in your python
script:

       import aetypes
       AETrue = aetypes.Boolean(1) # or AEFalse  = aetypes.Boolean(0)

then you can use AETrue and AEFalse as the boolean flags. It seems to
work.


From sholden at holdenweb.com  Fri Feb  1 17:57:02 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 17:57:02 -0500
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012597563.30612.python-list@python.org>
Message-ID: <v5F68.110829$8b.4204925@atlpnn01.usenetserver.com>

<brueckd at tbye.com> wrote ...
> On Fri, 1 Feb 2002, Anders Dahlberg wrote:
> > <brueckd at tbye.com> skrev i meddelandet
> > >
[Python/Java execution mechanisms]
> > > was curious to know how slow Java was without the JIT. Finally, even
> > > though Java and Python are wildly different in many ways, knowing the
> > > interpreted vs. JIT'ed speeds might give a little insight into the
> > > potential benefits of a Python JIT.
> >
> > I don't really know if python needs a jit, it would probably not harm
> > python, but is it really worth the effort?
>
> Heehee... that's the whole reason I asked my question in the first place.
>
Without any introduction of static typing into the language, it's extremely
unlikely that a JIT would yield the same benefits as it does for Java,
although recent experiments with compiler techniques have shown interesting
speedups on restricted cases.

The major problem is that with dynamic (latent) typing, the JIT code has to
examine the type of its operands in order to determine which of several
type-based operations to perform. Java can avoid this due to its strong
typing mechanism. But let's not forget: "Java is the COBOL of the
object-oriented world".

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From rasmussn at lanl.gov  Thu Feb 21 12:01:19 2002
From: rasmussn at lanl.gov (rasmussn at lanl.gov)
Date: Thu, 21 Feb 2002 10:01:19 -0700
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <a51cj9$9m1$1@bob.news.rcn.net>
Message-ID: <A06C587C-26EC-11D6-BCCE-0030657EBC3E@lanl.gov>

On Wednesday, February 20, 2002, at 04:48 PM, David Abrahams wrote:

>> As I understand it, the SILOON project uses a commercial parser known as
>> PDT from uoregon.  Have you, or has anyone else had any experience with
>> the SILOON/PDT project?  On the face of it, my guess would be that you
>> would be somewhat skeptical of their claims?
>
> I haven't had any experience with it. I just found that site. My 
> skepticism
> just dropped by a factor of 8. If they are really using the EDG front-end 
> as
> they claim to, then I am sure the parser is excellent. I do have to wonder
> who is paying for the FE. The issues of dealing with the semantics of what
> they've parsed remain, but these guys at least seem to be doing the 
> parsing
> part right.

It is my understanding that an educational or research license to the EDG
front end is fairly easy to obtain.  The University of Oregon has obtained
the right to distribute the EDG front-end as binaries and does so for 
several
platforms in the Program Database Toolkit (PDT).  So for anyone wanting
interface level information from C++ or Fortran 90 source files, the PDT
is an excellent tool.  We use it in several projects.

See http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/ and
http://www.acl.lanl.gov/siloon/ for more information.

Craig




From claird at starbase.neosoft.com  Fri Feb 15 18:51:49 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Feb 2002 17:51:49 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <140220020703593280%jwbaxter@spamcop.net> <a4j24g$jao$9@bob.news.rcn.net> <3C6DA497.9AB8D624@ev1.net>
Message-ID: <D8AB2F5FFDFFBD4F.A40B57999CA5FB13.BD8D4DC731F86177@lp.airnews.net>

In article <3C6DA497.9AB8D624 at ev1.net>,
Charles Richmond  <richmond at ev1.net> wrote:
>jmfbahciv at aol.com wrote:
>> 
>>      [snip...]     [snip...]     [snip...]
>> 
>> I do have two memoes that I found in my office boxes that were
>> addressed to PDP-1 users.  I don't know the people who wrote
>> them (Roland Silver, Edward Fredkin) nor where they were.
>> 
>>      [snip...]     [snip...]     [snip...]
>> 
>> Well, if you can give me a clue about these memoes that I have, I'd
>> appreciate it.
>> 
>Ed Fredkin is semi-famous. He was mentioned prominantly in David
>Levy's book _Hackers_. He also invented some data structures that
>are used today, notably the "trie". 
			.
			.
			.
Oh, yes.  There's a whole cult around Fredkin
<URL: http://www.digitalphilosophy.org/ >.

It's one to which I'm sympathetic, as it happens.
-- 

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


From usenet at thinkspot.net  Sun Feb 10 10:37:49 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sun, 10 Feb 2002 07:37:49 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a43due.3vv9a5r.1@kserver.org> <8852d33c.0202100034.16f1b47b@posting.google.com>
Message-ID: <a4580e.3vsjs5j.1@kserver.org>

On 10 Feb 2002 00:34:45 -0800, wookiz at hotmail.com (wooks) wrote in
comp.lang.python in article
<8852d33c.0202100034.16f1b47b at posting.google.com>:

> Sheila King <usenet at thinkspot.net> wrote in message news:<a43due.3vv9a5r.1 at kserver.org>...

> > I recommend:

> > (2) Core Python Programming by Wesley Chun
> > (not for beginning programmers...aimed at someone with some programming
> > experience...addresses the OO points and many features of the language
> > quite well)
> 
> Yes. Thats the one that all the reviews seem to point to. If I
> remember rightly the copy in the bookshop I went to was rather dated.
> Will investigate further.

I thought there was only one edition of that book??? Mine is copyright
2001. If it based on Python 1.5.2 but has remarks and appendices for
2.0. The only book I know of that assume more than Python 2.0 is the
Python 2.1 Bible (which I also have...and recommend only as a reference
for someone who already knows Python). Most of the recent ones are
either based on 1.5.2 with appendices referring to changes in 2.0, or
they are based on 2.0. That's as recent as you're going to get.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From phd at phd.pp.ru  Fri Feb  1 13:34:11 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 1 Feb 2002 21:34:11 +0300
Subject: Running CGIs under my uid - going slowly insane.
In-Reply-To: <a3ell7$29bm$1@nntp6.u.washington.edu>; from donn@u.washington.edu on Fri, Feb 01, 2002 at 06:09:43PM +0000
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu>
Message-ID: <20020201213411.E18145@phd.pp.ru>

On Fri, Feb 01, 2002 at 06:09:43PM +0000, Donn Cave wrote:
> ...
> | Note that setting an executable's setuid bit will change the "effective" uid
> | of the resulting process, not the uid (see 'os.get*e*uid()' above). This
> | means that the process may use that user's permissions.
> 
> That's true, but "executable" in this context usually means the binary,
> i.e., python itself.  Years ago, some UNIX platforms introduced a hack
> that made setuid effective for script files, but today that is universally
> recognized to have been misguided.  That's why I'm curious about yours.

   Solaris has special mesures that allow setuid scripts to be more secure.

> Perl is subject to the same limitation, but has optional support for
> setuid.  How can that be?  I'm no Perl whiz, but I assume you install
> perl itself setuid root, and then it checks its input file for setuid
> and either sets its ID accordingly or reverts to the invoker's ID.

   This is standard Perl installation - Perl has perl binary and taintperl
setuid binary.

> This option is not the default and is rather discouraged as I recall it.

   In Perl taintperl considered quite secure, much more secure than any
setuid security mesures.

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



From ssmith619 at hotmail.com  Sun Feb 24 13:37:03 2002
From: ssmith619 at hotmail.com (Stephen Smith)
Date: Sun, 24 Feb 2002 18:37:03 GMT
Subject: .exe files
Message-ID: <jpae8.66140$X64.23604369@news1.rdc2.pa.home.com>

I'm new to Python, and have created a number of small programs that users
might find interesting, althought they aren't really geeky and the people
that use them would probably not have the Python IDLE or anything even
related to Python installed.  Is there anyway to turn a .py file into a .exe
file that can be executable on Windows?  It doesn't call from any other
modules, it's completely stand alone.  Thank you for any information!




From fredrik at pythonware.com  Mon Feb 11 05:41:18 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 10:41:18 GMT
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com> <Xns91B074EB2FA5ahmebahhotmailcom@198.99.146.10> <Xgj98.24854$l93.4542796@newsb.telia.net> <13285ea2.0202100707.5fe84495@posting.google.com>
Message-ID: <idN98.25036$l93.4624073@newsb.telia.net>

J.Jacob wrote:
> > the api is relatively clean; writing a curses implementation
> > should be trivial (did most of that once, but lost the code).
>
> The link you give has a link to a "downloads page" and that one gives
> me a 404.

oops.  this link should work better than the one on
that page.

    http://effbot.org/downloads/

> What we really want is something portable in the Python distribution
> itself and i think many more people would love it.  Especially for
> programming projects in schools.  But i am not sure how to ask for it,
> maybe at the sourceforge Python site under "Feature Requests"?

I suggest writing a small, preliminary PEP.

    http://python.sourceforge.net/peps/
    http://python.sourceforge.net/peps/pep-0001.html

(it's probably a good idea to post it here as a "pre-PEP" before
submitting it to the PEP editor).

Try keeping the interface small and simple  (I'd implement a sub-
set of what's provided by the Console module, but that's me.  No
MoveCursorToRowAndColumn methods, please ;-)

</F>




From philh at comuno.freeserve.co.uk  Mon Feb 18 11:03:10 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 18 Feb 2002 16:03:10 +0000
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no> <3C706035.6AC11606@earthlink.net>
Message-ID: <slrna729du.2hb.philh@comuno.freeserve.co.uk>

On Mon, 18 Feb 2002 01:59:34 GMT, Hans Nowak <wurmy at earthlink.net> wrote:
>Magnus Lie Hetland wrote:
>> 
>> In article <8867.813T2365T9245607threeseas at earthlink.net>, Timothy Rue wrote:
>> [...]
>> >Certainly everyone does understand in reading and responding to post here,
>> >they actually make use of all nine action. It's physics, but they may
>> >think it's me and that somehow their reactions will cause a dismissal
>> >of.... well physics.
>> >
>> >Lets see now:
>> [...]
>> 
>> No, no! This isn't simple enough. The *real* truth (and it's physics,
>> so you can't contradict me) is that there is only one true action:
>> 
>> DS - Do Stuff
>> 
>> Now deal with it and go somewhere else with your false theory.
>
>Do you also have a model with zero actions? 

Sure. The contents of Tim's head.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From Mike_ba at web.de  Wed Feb 13 06:17:44 2002
From: Mike_ba at web.de (Michael Bauer)
Date: Wed, 13 Feb 2002 12:17:44 +0100
Subject: Exiting threads silently
References: <a4au9e$u1v$01$1@news.t-online.com> <3C691337.A891C9DA@engcorp.com>
Message-ID: <a4dhr0$8n0$03$1@news.t-online.com>

Peter Hansen wrote:

> The docs (i.e. the SocketServer source) suggest that the default
> error handler will print a traceback and continue when an exception
> is raised.  Try overriding handle_error() on your server class,
> catch SystemExit separately, and re-raise it if you get it.

Ah, i see. Thank you :)

cu Mike


From michael at stroeder.com  Wed Feb 20 17:09:51 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 20 Feb 2002 23:09:51 +0100
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org> <7xr8ng83k0.fsf@ruckus.brouhaha.com>
Message-ID: <3C741EAF.B8E6E14A@stroeder.com>

Paul Rubin wrote:
> 
> <base href=http://foo.bar/cgi-bin/_SID_=36987348732/>

Are you sure all browsers handle this properly?

Ciao, Michael.


From sjmachin at lexicon.net  Mon Feb 25 17:26:22 2002
From: sjmachin at lexicon.net (John Machin)
Date: 25 Feb 2002 14:26:22 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com> <3c787c1f.2523859@news.t-online.de>
Message-ID: <c76ff6fc.0202251426.65fbf1f2@posting.google.com>

gerson.kurz at t-online.de (Gerson Kurz) wrote in message news:<3c787c1f.2523859 at news.t-online.de>...
> On 23 Feb 2002 11:52:10 -0800, bobnotbob at byu.edu (Bob Roberts) wrote:
> 
> >What would be a good way to replace every one or more spaces (" ") in
> >a string with just one space?  Or replace any number of newlines with
> >just one?
> 
> Lets see. There were four solutions mentioned:
> 
> -------------(cut here)---------------
> def test1(newstring):
>     while newstring.find('  ') > -1:
>         newstring = newstring.replace('  ', ' ')
>     return newstring
> 
> def test2(newstring):
>     return " ".join(filter(None,newstring.split(' ')))

This "solution" removes leading and trailing spaces. That may or may
not be what the OP was asking for.

> 
> def test3(newstring):
>     return re.sub(' +', ' ', newstring)

Coded extremely suboptimally.
Try this:
def test3o(newstring, subber=re.compile('  +').sub):
    # Note *two* spaces in pattern!!
    return subber(' ', newstring)

> 
> def test4(newstring):
>     return ' '.join(newstring.split())
> -------------(cut here)---------------
> 
> Note that test4() does also split newlines (and tabs), this is why
> test2() explicitly splits for *blanks only*. Given these, you can test
> which one is fastest

>         func("This  is a    test\n isn't it?")

Only one piece of test data? Try these pieces and then try to explain
the
timings:

[("a" + " " * n) * (100 / (n+1)) for n in range(5)]

Just for fun, try a line with say 10000 spaces and nothing else. Try
to explain why solution 2 which is superficially O(N) goes off the
planet [it does on my platform, YMMV], even worse than solution 1
which is superficially O(N*log(N)) and why solutions 3, 3o & 4 stay
sane.

Then repeat all of the above with Python 2.1 or earlier, for the
benefit of those not yet upgraded.

Then there's the possibility that the results may vary by platform ...

> 
> So, the simple "while"-solution (test1) is actually the fastest if you
> really only want to replace spaces (and not newlines and tabs, also).
> I find that quite interesting because the other three solutions seem
> so much more sophisticated.

Understandable ... "sophisticated" can also mean "deceptive" or
"misleading" according to my dictionary.


From starx at pacbell.net  Sun Feb 10 00:03:22 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 10 Feb 2002 05:03:22 GMT
Subject: signal through popen2
References: <de6c123c.0202092047.424c8091@posting.google.com>
Message-ID: <Xns91B0D62D19FF3RASXnewsDFE1@130.133.1.4>

Lei Chen || Sat 09 Feb 2002 08:47:15p:

> Hi,
> 
> I would like to send a CTRL-C on a process opened by popen2.  Is this
> possible? 
> 

I haven't used popen's before, but i assume you have some method of 
sending a string through it? Then sure, you just have to look up the 
character number that you get from a ctrl-c, i'm not sure what it is. 
But to send, say, a tab, linefeed, and carrage return through (using the 
raw numbers), you would have a string like

"\x09\x0a\x0d"
(if you type this in a shell python will display it as '\t\n\r')

So you would have something like mypipe.write('\xNN'), where NN is the 
hex number of whatever it is ctrl-c does.

But then, the real way of doing what you want to do may be slighly 
different... like closing a socket to produce a proper eof signal... 


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From p.mayers at ic.ac.uk  Tue Feb 26 14:12:29 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Tue, 26 Feb 2002 19:12:29 +0000
Subject: Coredumps with Python 2.1.1-3
References: <20020226.185425.1884661237.5746@ic.ac.uk>
Message-ID: <20020226.191229.159259470.5746@ic.ac.uk>

In article <20020226.185425.1884661237.5746 at ic.ac.uk>, "Phil Mayers"
<p.mayers at ic.ac.uk> wrote:

> I'm using the RPM for Python 2.1.1-3 from the Python website, and have
> just had a *really* weird problem. I was editing some code, and did a

Well, I *still* think it's weird, but an emergency upgrade of the DB
adaptor (don't you just love it when managers say "Too hell with the
users!") seems to have cured it.

Cheers,
Phil


From jason at jorendorff.com  Tue Feb  5 09:08:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 5 Feb 2002 08:08:28 -0600
Subject: Inconsistence syntax in os module?
In-Reply-To: <3C5F731D.4060500@larsshack.org>
Message-ID: <HFEKILOLEFEFMKAECNDLKEIFDAAA.jason@jorendorff.com>

> Rewriting system() and popen() to be consistent with popen2() and 
> friends is trivial, especially if one writes the implementations 
> in python.

I agree that os.system() and os.popen() should accept list arguments
and do the right thing.

In the meantime, os.spawnv() is what you want for os.system().

## Jason Orendorff    http://www.jorendorff.com/



From marknenadov at winningteam.com  Fri Feb 22 15:41:26 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Fri, 22 Feb 2002 15:41:26 -0500
Subject: ncurses
References: <mailman.1014406117.9467.python-list@python.org>
Message-ID: <20020222.154126.2112255763.1938@d57-47-79.home.cgocable.net>

In article <mailman.1014406117.9467.python-list at python.org>, "Charles"
<charles at a3a.com.br> wrote:


> Anybody know where can i find a GOOD tutorial or explantion about ncurse
> (curses) python module ?  i need create a counter in linux console with
> a self refresh. printing the value in same place, overwriting the old
> value. like seconds in a clock.
> Charles
> 

IBM has a fairly good tutorial
(http://www-106.ibm.com/developerworks/linux/library/l-python6.html)


From mhammond at skippinet.com.au  Sat Feb  2 02:09:27 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 02 Feb 2002 07:09:27 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com>
Message-ID: <3C5B90F3.3020503@skippinet.com.au>

Keith Farmer wrote:
> I'm trying to intercept the WM_NCHITTEST message, to selectively disallow
> resize for given window edges. The following code results in
> 
>     (216, 661)
>     43319512
>     Traceback (most recent call last):
>     File "D:\Python\wxAppBar\PyAppBar.py", line 52, in _OnWMNCHitTest
>         self._LowHighWordsToLparam(win32api.GetCursorPos())
>     TypeError: an integer is required

Given your code:

>     def _OnWMNCHitTest(self, params):
>         lr = win32gui.DefWindowProc(self._GetAppBarWnd(),
>                                                         WM_NCHITTEST,
>                                                         0,
> 
> self._LowHighWordsToLparam(win32api.GetCursorPos())

It appears that the error is actually in trying to call 
win32gui.DefWindowProc, rather than anything to do with the:

       self._LowHighWordsToLparam(win32api.GetCursorPos())

code.  Are you sure _GetAppBatWnd() returns an integer?

Mark.



From rjones at ekit-inc.com  Mon Feb 25 19:15:56 2002
From: rjones at ekit-inc.com (Richard Jones)
Date: Tue, 26 Feb 2002 11:15:56 +1100
Subject: raw_input on password
In-Reply-To: <slrna7lf63.4pb.msoulier@pmerd071.ca.nortel.com>
References: <slrna7lf63.4pb.msoulier@pmerd071.ca.nortel.com>
Message-ID: <200202260016.AAA23991@crown.off.ekorp.com>

On Tue, 26 Feb 2002 09:37, Michael P. Soulier wrote:
>     Hey people.
>
>     My apologies if this has been asked before. I didn't see it...
>
>     Is there a way to get raw_input() to mask the response, so it can be
> used to request a password safely?

Use the getpass module:

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


      Richard



From henrik.motakef at web.de  Mon Feb 18 14:35:17 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 18 Feb 2002 20:35:17 +0100
Subject: PyKDE
References: <u72ll0kis0p201@corp.supernews.com>
Message-ID: <86zo26d1my.fsf@apo.hmotakef.int>

Mik Mifflin <dogansmoobs at NOSPAM.prexar.com> writes:

> I'm trying to learn Qt programming and decided to use my new favorite 
> language (Python, of course).  I'm using SuSE 7.3, and there is an rpm 
> called PyQt installed, but it looks like I need PyKDE as well?  This is the 
> program that I'm trying to run, and the error messages it gives:
> 
> !/usr/bin/python
> from qt import *
> from kdecore import *
> from kdeui import *

Well, when you want to use kdecore and kdeui, it seems to be a good
idea to have PyKDE installed :)

> If I do need PyKDE, does anyone know of an rpm that will work with SuSE?  
> Or am I just better off installing from source?

The current PyKDE is in alpha stage (the previous release only deals
with KDE1). You can download it from
http://www.thekompany.com/projects/pykde, but have to compile it yourself.

If you don't want to do this, PyQt is still fine do to GUI programming
- just have a look at the Qt documentation (will surely be installed
with the Qt-Devel-package) to see what's available. Generally,
everything starting with a Q is, while everything starting with a K
isn't.

hth
Henrik


From ramb at synopsys.com  Mon Feb 25 13:46:13 2002
From: ramb at synopsys.com (Ram Bhamidipaty)
Date: Mon, 25 Feb 2002 18:46:13 GMT
Subject: Need Help: Can only set gdb breakpoints in python module the first time
References: <m33czpa0te.fsf@oak.ramandgita.com>
Message-ID: <m3y9hh8kuy.fsf@oak.ramandgita.com>

I found that this same behavior happens with python without my application:

$ gdb /usr/local/bin/python
GNU gdb 5.0
...
This GDB was configured as "i386-redhat-linux"...
(gdb) break PyImport_GetMagicNumber
Breakpoint 1 at 0x8098263: file Python/import.c, line 374.
(gdb) run
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.exit(0)

Program exited normally.


(gdb) run
warning: Cannot insert breakpoint -7:
Cannot access memory at address 0x4002dd24
(gdb) q


In order to place more breakpoints I would need to exit gdb and restart
the whole gdb run.

-Ram





Ram Bhamidipaty <ramb at synopsys.com> writes:

> Python is embedded into my application. I have a custom module that
> is part of the application - it is not loaded as a shared library - rather
> the object files are incorporated into the final binary at link time.
> 
> The problem I am having is that I can set breakpoints in my module only
> when gdb is first started. Once my app exits I am no longer able to set
> breakpoints on functions in my module.
> 
> I suspect that python is somehow changing the mapping of the text segment,
> but that just a guess.
> 
> My question: What is python doing that prevents gdb from setting breakpoints
> after the program exits the first time? How can I prevent this?
> 
> 
> Thanks for any info.
> -Ram


From rmunn at pobox.com  Thu Feb  7 11:21:09 2002
From: rmunn at pobox.com (Robin Munn)
Date: 7 Feb 2002 16:21:09 GMT
Subject: Amount of files on a drive?
References: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk> <a3ov9n$dn9$1@newsg3.svr.pol.co.uk> <Xns91ACD5444241CRASXnewsDFE1@130.133.1.4>
Message-ID: <slrna65a5i.1m1.rmunn@rmunn.dyndns.org>

On 6 Feb 2002 04:57:54 GMT, Philip Swartzleonard <starx at pacbell.net> wrote:
>G. Willoughby || Tue 05 Feb 2002 07:56:57a:
>
>> i am using this at the minute:
>> ------------------------------------------------------------------
>> import win32api
>> import string
>> import os
>> 
>> def countFiles(arg, dir, files):
>>     for file in files:
>>         count=count+1
>> 
>> fileListings=[]
>> driveList=string.split(win32api.GetLogicalDriveStrings(),'\0')[:-1]
>> 
>> for drive in driveList:
>>     os.path.walk(drive, countFiles, 0)
>> ------------------------------------------------------------------
>> but its pretty slow any other suggestion how i could speed it up??
>
>Well, first determine how long it actually takes (time it), and see how it 
>compares to say, selecting everything on the drive and hitting properties 
>in the context menu to get a count (i.e. see how good windows itself is at 
>it). It took my 800mhz athlon about 35 seconds to tell me that i have 6.15 
>gigs in 101,064 files in 3,930 folders on my C drive. Counting on that 
>scale can get slow.

Also, the bottleneck here is going to be the drive's access time, not
the processor speed. The access time is usually measured in milliseconds
and is the average time for the drive's read/write head to move from one
location to another on the disk surface. This operation is sometimes
called a "seek". Since the drive has to perform at least one seek for
every file, and usually more, the total time spent waiting for the hard
drive far outweighs the time spent by the processor.

On the other hand, I do see one thing you could do to improve your code.
Chage the countFiles() function thus:

------------------------------------------------------------------
def countFiles(arg, dir, files):
    count = count + len(files)
------------------------------------------------------------------

Also, I presume you're using a global variable to hold your count,
unlike the trimmed-down example code you showed us. In the code you
showed us above, the count variable would be a variable local to
countFiles() and would not hold its value across separate invocations of
countFiles(). Instead, you would do:

------------------------------------------------------------------
count = 0

def countFiles(arg, dir, files):
    global count
    count = count + len(files)
------------------------------------------------------------------

The "global count" statement inside countFiles() specifies that from
here on, the name count will refer not to a variable local to
countFiles() but instead to the global variable of that name.

Gee, this is reminding me of that student blooper (probably an urban
legend) that went, "The Iliad was not written by Homer, but by another
man of that name." :)

-- 
Robin Munn
rmunn at pobox.com


From empty_one at ameritech.net  Tue Feb 19 15:40:44 2002
From: empty_one at ameritech.net (Empty_One)
Date: Tue, 19 Feb 2002 20:40:44 GMT
Subject: Installing Python
References: <b6dc8.5870$pC1.1234514@newssrv26.news.prodigy.com> <mailman.1014066813.4340.python-list@python.org>
Message-ID: <gLyc8.7047$pC1.1310714@newssrv26.news.prodigy.com>

Gerhard H?ring wrote:

> Le 18/02/02 ? 20:03, Empty_One ?crivit:
>> I just downloaded the python tarball, and installed it on my slack system
>> using sheck install.  All works good, except I cannot start IDLE.
> 
> I'm sure there was an error message when you tried to start IDLE. I've
> lost my crystal ball, so these would be incredibly useful.

Here is the error message:  bash: idle: command not found

I know it didn't install, I was asking how to get it installed.  What 
options do I need to add to ./configure
 
> Chances are that Tkinter wasn't built. You'll need to have Tk installed
> for this, including the header files.

I have the Tk toolkit for Tcl, version 8.3.3 installed.  Do I need more 
installed, and is this the correct version?

>> How can I configure the install script to also install IDLE, since it
>> is included with the base python, and not a seperate download.
> 
> See above, you can also change the paths in Modules/Setup, should that
> be necessary. It usually isn't necessary, unless the Tk header files and
> libraries are in some very strange places.
> 
> Gerhard


-- 
Empty_One
_________
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."--Benjamin Franklin


From hmartires at cultalg.pt  Mon Feb 25 09:29:38 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 14:29:38 -0000
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de>
Message-ID: <vRre8.603$9l1.7321620@newsserver.ip.pt>

Where are the method's ?? (array.tostring() array.fromstring() )
I import the array module, but i don't have them.
I'm using python 1.5

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4bsedlnai.fsf at informatik.hu-berlin.de...
> "Hugo Martires" <hmartires at cultalg.pt> writes:
>
> > How can i count the number of bytes of:
> > --> List []
> > --> array('d',[])
> >
> > There is any function that return's the number of bytes ?
>
> For the list, no. For the array: len(arr.tostring()) will give a rough
> estimate. The actual memory consumption is higher, due to Python and
> malloc overheads.
>
> Regards,
> Martin
>
>




From justin at iago.org  Thu Feb 14 11:33:42 2002
From: justin at iago.org (Justin Sheehy)
Date: Thu, 14 Feb 2002 11:33:42 -0500
Subject: python to lisp translation
In-Reply-To: <7x1yfomva3.fsf@ruckus.brouhaha.com> (Paul Rubin's message of
 "13 Feb 2002 22:28:52 -0800")
References: <3C6A7825.714DF612@setf.de>
	<mailman.1013619443.24325.python-list@python.org>
	<7x1yfomva3.fsf@ruckus.brouhaha.com>
Message-ID: <x7n0ycrpjt.fsf@ra.iago.org>

Paul Rubin <phr-n2002a at nightsong.com> writes:

>> However, it would be a seriously hard thing to do correctly and well.
>
> I've been kicking the idea around for a while.  Do you see any
> specific stumbling blocks?  I can see having to modify the underlying
> Scheme implementation in some minor ways to get things like immutable
> strings, but most of Python maps onto Scheme in a pretty natural way.

There are a lot of issues, including the surprising amount of
interesting work that is done even at the bytecode-interpretation
stage in Python.  Start at ceval.c and keep reading.  :-)  Also,
realize that most Python programs do not only execute Python code.
Major parts of the core standard library are not written in Python.
This could be a huge stumbling block. 

This is only the beginning.  My personal opinion would be that you
might be much better off helping to resurrect John Max Skaller's
"Vyper" project than starting from scratch.  Vyper is a variant of
Python implemented in OCaml, designed for very high performance.  The
project seems to have gone defunct, but it made a very promising
beginning given its ambitious goal.

(OCaml is not Scheme, but it shares many of the features that you
might care about.)

Have fun, and good luck.

-Justin

 






From alans at killergame.com  Fri Feb 15 20:02:45 2002
From: alans at killergame.com (alan scales)
Date: 15 Feb 2002 17:02:45 -0800
Subject: wxPython:  EVT_CLOSE question
Message-ID: <c0a56015.0202151702.1f29d19e@posting.google.com>

i have some controls (in particular wxStyledTextCtrl) - they are NOT
frames or dialogs.  I want to be able to trap when they are closed (or
destroyed)(e.g. to save data). I was hoping to use EVT_CLOSE and
provide a handler.  however it does not seem to work.

  1. is it supposed to work? (it is only documented for
frames/dialogs)
  2. if not, what can be done to detect when a control is closed?

thanks
alan scales


From aahz at panix.com  Sat Feb  2 13:24:04 2002
From: aahz at panix.com (Aahz Maruch)
Date: 2 Feb 2002 10:24:04 -0800
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>
Message-ID: <a3has4$llc$1@panix2.panix.com>

In article <_JU68.2091$3E5.208453 at newsread2.prod.itd.earthlink.net>,
Stede Troisi <stede at computer.org> wrote:
>>
>I was thinking of doing an E-book (because I don't like the high price of
>regular books) on an introduction to programming and artificial intelligence
>in Python.
>
>I was wondering if there would be interest in such a e-book and if it would
>sell?  I would love to here everyone's opinion.

The first question you need to answer is whether you can write such a
book.  If you can, follow Alan Gauld's example:

http://www.freenetpages.co.uk/hp/alan.gauld/tutintro.htm
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From logiplexsoftware at earthlink.net  Fri Feb  8 23:53:30 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 20:53:30 -0800
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <a4290e$1b4hcp$1@ID-11957.news.dfncis.de>
References: <mailman.1013196740.22073.python-list@python.org>
	<a41oi1$a22$1@panix2.panix.com>
	<mailman.1013216835.14351.python-list@python.org>
	<a4290e$1b4hcp$1@ID-11957.news.dfncis.de>
Message-ID: <20020208205330.2e5db24a.logiplexsoftware@earthlink.net>

On Fri, 8 Feb 2002 20:31:04 -0800
Emile van Sebille wrote:

> 
> "Cliff Wells" >
> > What's wrong with Fry's?  As long as you don't listen to the
> salespeople,
> > that is.
> >
> >
> 
> ... and check there's no recycle tag before buying.
> 

I've probably built 6 systems from components bought at Fry's in the last 2
months and had nary a problem with any of them.  Right after posting that
comment, I assembled a Duron system and the damn thing bombs halfway
through my RH 7.2 install.  Of course, that's expected to happen on
occasion when you assemble your own, but the irony wasn't lost on me ;P

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From mdk at mdk.com  Fri Feb  8 14:27:17 2002
From: mdk at mdk.com (MDK)
Date: Fri, 8 Feb 2002 14:27:17 -0500
Subject: What is the most efficient way to do this?
Message-ID: <a418qm$1c055f$1@ID-98166.news.dfncis.de>

What is the most efficient way to remove empty items from this list?

['', '1', '2', '3', '', '456', '789', 'wow']




From cfelling at iae.nl  Thu Feb 21 21:08:00 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 22 Feb 2002 03:08:00 +0100
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com> <3C747AE3.FD16CB33@earthlink.net> <3C7542D1.7918EC87@ccvcorp.com>
Message-ID: <a54960$fhf$1@animus.fel.iae.nl>

Jeff Shannon <jeff at ccvcorp.com> wrote:
...
> IIRC, Slice objects (types.SliceType, etc) are deprecated, and objects wishing
> to use slice notation should use __getslice__() and __setslice__() [and,
> perhaps, __delslice__()] instead.

Nay, it's the other way round, those methods are depricated:)
(aacording to Python pocket reference sec. edition)

-- 
groetjes, carel


From bokr at oz.net  Fri Feb 15 15:55:30 2002
From: bokr at oz.net (Bengt Richter)
Date: 15 Feb 2002 20:55:30 GMT
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au> <a4j5hd$l3j2$1@ID-11957.news.dfncis.de> <mailman.1013800706.14636.python-list@python.org>
Message-ID: <a4jsk2$q4e$0@216.39.172.122>

On Fri, 15 Feb 2002 14:01:49 -0500, "Dr. David Mertz" <mertz at gnosis.cx> wrote:

>|"Trung Hoang" <thoa0025 at mail.usyd.edu.au> wrote in message
>|> i was wondering if i can get certified with python from an
>|> internationally recognised company?
>
>"Emile van Sebille" <emile at fenx.com> wrote previously:
>|I don't know if they're still doing it, but at one point Brainbench.com
>|was working on competency testing, and IIRC, gearing that towards
>|employers and prospective employees. http://www.brainbench.com/
>
>Let me chime in slightly, given the thread.  I happen to have *written*
>the Brainbench Python exam.
>
>Unfortunately, my experience in working with Brainbench so was not
>positive.  Aside from being a real PITA to work with, and not paying me
>nearly enough for my time, I really did not feel that their guidelines
>allowed production of nearly as good a test as was possible.  I have/had
>some background in test design, and it was painful to me to work through
>constraining and poorly conceived Brainbench test creation guidelines.
>Even after creating the exam as best I could within their framework (bad
>creation software also), they still required me to go back and
>rephrase/restructure many question for the worse.  In the end, I felt
>bad about creating something that is seriously flawed as a
>skills-validation tool.
>
>In other words, I don't believe that Brainbench's Python exam is a
>*good* testing tool.  I can't speak specifically for other topics; I did
>not write them, and never took one (except some initial playing around
>in a topic I didn't really know, just to get a feel for the mechanics of
>the exams).
>
>That said, the original question was from someone who wanted a
>certificate.  You can certainly take the Brainbench exam, and get a
>piece of paper out of the process.  There's not necessarily a need to
>forward my criticisms to your potential employer :-).
>
What would your thoughts be re an automated exam as a PSF - licensed free
collaborative project? Maybe even the exam definition itself could somehow
feed off a wiki-like process, if you constrained formats appropriately
for defining test elements?

Regards,
Bengt Richter




From martin at v.loewis.de  Sun Feb 24 03:30:37 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 09:30:37 +0100
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com> <mailman.1014533499.3218.python-list@python.org>
Message-ID: <m3adtz9t8y.fsf@mira.informatik.hu-berlin.de>

Kirill Simonov <kirill_simonov at mail.ru> writes:

> Try to replace "Python22\Lib\lib-tk\FixTk.py" with the following:
> 
> import sys, os
> 
> if not os.environ.has_key('TCL_LIBRARY'):
>     tcl_library = os.path.join(sys.prefix, "tcl", "tclX.Y")
>     os.environ['TCL_LIBRARY'] = tcl_library

Can you please explain what this change achieves, compared to the
FixTk.py that is already there? In particular, do you mean that "X.Y"
should be used literally?

Regards,
Martin



From grante at visi.com  Fri Feb 22 14:01:06 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 22 Feb 2002 19:01:06 GMT
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4> <Nktd8.1458$N7.276578@ruti.visi.com> <4Jud8.1483$N7.287257@ruti.visi.com> <3C768454.A5F778D2@ccvcorp.com>
Message-ID: <Szwd8.1511$N7.299044@ruti.visi.com>

In article <3C768454.A5F778D2 at ccvcorp.com>, Jeff Shannon wrote:

>> I just got bit by the leading-zero thing again about 10 minutes
>> ago when I changed the minor version number of something from 07
>> to 08.  Had to skip from 1.07 to 1.10...
> 
> So try using 010, which should then be converted back to 8 ... (or else keep your
> versioning info as strings.)

[The recent issue was in a C program.]

The major and minor version info is defined in a header file:

#define MAJOR_VERSION 1
#define MINOR_VERSION 08

In some places it's "stringified" and concatenated using
preprocessor tricks, and other places it's used as an integer
literal constant.  Perhaps the design is a bit baroque, but
it's easier to skip from 07 to 10 than it is to change the
usage in multiple places.

-- 
Grant Edwards                   grante             Yow!  Where's SANDY DUNCAN?
                                  at               
                               visi.com            


From kosh at aesaeion.com  Sun Feb 17 16:57:28 2002
From: kosh at aesaeion.com (kosh at aesaeion.com)
Date: Sun, 17 Feb 2002 14:57:28 -0700 (MST)
Subject: Guido wins FSF Award
In-Reply-To: <8867.813T2365T9245607threeseas@earthlink.net>
Message-ID: <Pine.LNX.4.32.0202171452390.1394-100000@aesaeion.com>

On Sun, 17 Feb 2002, Timothy Rue wrote:

> On 17-Feb-02 12:34:12 (Five Fresh) Fish <root at 127.0.0.1> wrote:
> >On Sat, 16 Feb 2002 19:35:23 -0700 (MST), <kosh at aesaeion.com> wrote:
>
> Ah, Kosh has a history of about 88 posts in Amiga newsgroups and up to
> July last year. But May thru August last year he has made 30 post to c.l.p
> until he added his 31st in this thread recently. I've also seen kosh
> explode many time in IRC. But what Kosh says about 3000 pages.... Guess
> kosh hasn't been around much enough to know better then to assume
> documentation consist of about 3/7th's of my usenet posts. As well as his
> putting on blinders to such simplicity as the following:

I have never been an amiga user and have never made posts to the amiga
newsgroup unless it was crossposted from the linux advocacy newsgroup
which would be pretty strange. For irc I am in a relatively limited number
of places. irc.worldforge.org and irc.openprojects.net #zope

Overall I think you have me confused with someone else and missed the
point of the joke entirely.

>
> Certainly everyone does understand in reading and responding to post here,
> they actually make use of all nine action. It's physics, but they may
> think it's me and that somehow their reactions will cause a dismissal
> of.... well physics.

Trying to make you stuff sound like physics does not make it physics or
even related to it. There is no real evidence for what you are claiming
like there is for the theories that govern atmospheric composition and
reactions, nuclear physics, etc.




From cce at clarkevans.com  Tue Feb 12 18:08:29 2002
From: cce at clarkevans.com (Clark C . Evans)
Date: Tue, 12 Feb 2002 18:08:29 -0500
Subject: Re-throwing an exception without loosing the stack trace
In-Reply-To: <15465.39126.250327.239174@beluga.mojam.com>; from skip@pobox.com on Tue, Feb 12, 2002 at 04:36:06PM -0600
References: <20020212174235.A17928@doublegemini.com> <15465.39126.250327.239174@beluga.mojam.com>
Message-ID: <20020212180829.A18166@doublegemini.com>

| when you are finished cleaning up you can just execute the 
| raise statement with no arguments. 

How simple.  *blush*

Yes.  I didn't try using raise without any arguments.  Cool.
Funny, this isn't in Learning Python... and raise all by
itself isn't listed in the syntax in the book.  Hmm.

Thank you _so_ much.

Best,

;) Clark

>>> def x():
...     raise "hi"
...
>>> def y():
...    try:
...       x()
...    except:
...       print "hi"
...       raise
...
>>> def z():
...    y()
...
>>> z()
hi
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in z
  File "<stdin>", line 3, in y
  File "<stdin>", line 2, in x
hi



From mkelly2002NOSPAM at earthlink.net  Wed Feb 20 15:57:52 2002
From: mkelly2002NOSPAM at earthlink.net (Michael Kelly)
Date: Wed, 20 Feb 2002 20:57:52 GMT
Subject: importing dlls with python
References: <e6bacef5.0202192029.2d095cc8@posting.google.com>
Message-ID: <iu287uoqkvdt744mi0rbssso3kvl156h1p@4ax.com>

On 19 Feb 2002 20:29:07 -0800, tkeon0 at netscape.net (Trevor Keon)
wrote:

>Hi, I am new to Python, and was wondering if it was possible to import
>dlls written in Delphi 5?  I have tried a number of methods to make it
>work, and I either get:

[snip]

Hi Trevor.  If you can get by using all Automation Compatible
types in your exported function params and returns then you
can very easily use a COM dll in Python.  If you make the dll
an Automation Library with dual-interface IDispath interface
(the default used by Delphi 5 when making an Automation
Library) then you can make it into a module using Makepy
and just import the module.  To get an instance of the
COM class you just assign the CoClass to a variable as
in:

#assuming you use -o switch in Makeypy to
#create module mymodule.py
#then look in mymodule.py and
#see what the CoClass name is, then:

from mymodule import *

myCoClass = CoClassName()
myResult = myCoClass.SomeMethod(SomeParam)

It's very quick esp. in D5(which is what I use to create most
of my COM and ActiveX stuff.)

The other consideration could be speed if you are
making like 10,000 calls into your dll code per loop
or something like that.


--

"I don't want to belong to any club that would have me as a member."
    -- Groucho Marx


From terabaap at yumpee.org  Mon Feb  4 17:01:35 2002
From: terabaap at yumpee.org (Manoj Plakal)
Date: Mon, 04 Feb 2002 16:01:35 -0600
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org> <7xu1sx41cl.fsf@ruckus.brouhaha.com>
Message-ID: <3C5F04BF.3090806@yumpee.org>

Paul Rubin wrote:

> The claim was that "from __future__ import division" and using //
> stops your old code from breaking when 3.0 comes out, which fits your
> definition of backward compatibility.  I pointed out that it only
> works for 2.2 and later.  It doesn't provide backwards compatibility
> to the versions of Python that most people actually use today.
>

 > [snip]

>  
> But that means I can't say "from __future__ import division", since
> only 2.2 and later support that.  And if I don't say that, then my
> scripts will break in 3.0.  How many times do I have to explain this?
> 


          Correct me if I'm wrong but I think we've all been
          warned that the next major release of Python (3.0,3000,...)
          could break backward compatibility with old versions.

          So if you're distributing Python code, it looks
          like you have some choices:
              - Require use of Python <= 2.1, no matter
                what the latest and greatest version of
                Python provides (perhaps bundle the interpreter
                with your software if people don't want to keep
                an extra copy of Python to run your stuff)

              - Require Python <= 2.1 now, but eventually
                migrate your users to 2.2 or 3.0 by releasing
                newer versions of your software

              - Require Python >= 2.2 now, and hopefully
                use the from future mechanism to avoid
                the above headaches.

          Also, I'm curious, why does your code depend
          so critically on integer division? There
          are some not-so-ugly approximations to
          quotient and division that will work on
          all versions of Python (I believe):
               int(x/y) for integer division
               float(x)/y for "true" division
          There may be problems with rounding
          and negative numbers and what not, but I
          think these will work for the majority of
          positive number calculations you want to do.
          It's the similar situation in C where
          I need to use a cast to do "true" division instead
          of integer division. BTW, I am not
          interested in starting another division
          thread, so if you don't like this
          solution, please ignore it :)

          Manoj



From peter at engcorp.com  Thu Feb 28 19:22:30 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 28 Feb 2002 19:22:30 -0500
Subject: Want to monitor process..
References: <87550ef1.0202272315.15779bbd@posting.google.com> <f2lf8.385$Az6.1854@news.hananet.net> <97ae44ee.0202280901.3153bdfc@posting.google.com>
Message-ID: <3C7EC9C6.D11FB17@engcorp.com>

Stephen wrote:
> 
> Unfortunately, my own applications seem to crash as opposed to
> shutdown normally, so it's probably not going to help me. From the
> library reference ~
> "Note: the functions registered via this module are not called when
> the program is killed by a signal, when a Python fatal internal error
> is detected, or when os._exit() is called. "

Yikes!  "Crash"?  Do you really mean they crash the Python interpreter,
or do you mean they exit with an exception?

If the former, you've got seriously problems, I agree.

If the latter, just wrap the highest level code with a try/finally
or try/except and put code in to ensure the child processes are
properly terminated.  Simple, clean, safe.

My Python programs *never* crash (as in kill the interpreter)
but maybe that's just me... I doubt it.

-Peter


From wolfson at uchicago.edu  Mon Feb 11 21:34:31 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Mon, 11 Feb 2002 20:34:31 -0600
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <a49s16$5ao$1@newshost.accu.uu.nl>
Message-ID: <pan.2002.02.11.20.34.30.302944.2302@uchicago.edu>

On Mon, 11 Feb 2002 19:44:06 -0600, Martijn Faassen wrote:

> I'm *not* a member of the PSU. The Python Secret Underground does not
> exist. I have no idea why people would think it does. Probably because
> they're all incompetent. Yeah..
> 
> There is no conspiracy. Even if such a conspiracy existed, I could tell you
> that there is no reason to

It's a poor conspiracy that censors the messages denying its existence.

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From fbartlet-fixit at optonline.net  Mon Feb 11 08:13:16 2002
From: fbartlet-fixit at optonline.net (Frederick H. Bartlett)
Date: Mon, 11 Feb 2002 13:13:16 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net>
Message-ID: <3C67C369.597E3517@optonline.net>

Years ago, my father -- a professor emeritus of music at the University
of Western Ontario -- coined the acronym ETD: Ego-Talent Discrepancy.

He never published, so he can't get official credit for the observation;
but I try to do what I can to get the coinage into common currency.

Fred

Fredrik Lundh wrote:
> 
> Courageous wrote:
> 
> > >Very good example of someone who is afraid of autocoding.
> >
> > You never did answer my question, by the way. If one has
> > become delusional, how does one become aware of it?
> 
> for some reason, this whole thread reminds me of this little gem:
> 
>     Unskilled and Unaware of It: How Difficulties in
>     Recognizing One's Own Incompetence Lead to
>     Inflated Self-Assessments (Kruger & Dunning, 1999)
> 
>     "People tend to hold overly favorable views of their
>     abilities in many social and intellectual domains. The
>     authors suggest that this overestimation occurs, in
>     part, because people who are unskilled in these
>     domains suffer a dual burden: Not only do these
>     people reach erroneous conclusions and make un-
>     fortunate choices, but their incompetence robs
>     them of the metacognitive ability to realize it"
> 
>     http://www.apa.org/journals/psp/psp7761121.html
> 
> of course, Rue is going to argue that this doesn't apply to him, but
> to everyone else.  after all, he KNOWS he's smarter than everyone
> else, and we're just too stupid to realize how stupid we are ;-)
> 
> </F>


From dwfreeze at yahoo.com  Tue Feb 12 08:50:54 2002
From: dwfreeze at yahoo.com (dwfreeze)
Date: 12 Feb 2002 05:50:54 -0800
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com> <Ry3a8.40197$bP3.337803@NewsReader>
Message-ID: <8f070afd.0202120550.45d8873b@posting.google.com>

"Andrew Ward" <winelight at spamcop.net> wrote in message news:<Ry3a8.40197$bP3.337803 at NewsReader>...
> > Andrew, what version of python are you running?
> 
> I am using 2.2 under Windows XP, but in my case, the read() is of a plain
> old single page of html text, no cookies, no nothing complicated!
> 
> Yes, urlopen() is what I would expect to hang, rather than the read(), so
> I'm very puzzled.
> 
> Andrew Ward

Standard "Mee too" post follows:

>>> sys.version
'2.1.2 (#31, Jan 15 2002, 17:28:11) [MSC 32 bit (Intel)]'
>>> import urllib
>>> f = urllib.urlopen('http://slashdot.org').read()

After I execute the urlopen the IDLE shell is definately "hung".


From BPettersen at NAREX.com  Thu Feb 28 13:53:14 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Thu, 28 Feb 2002 11:53:14 -0700
Subject: Status of PEP's?
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192215148@admin56.narex.com>

> From: Courageous [mailto:jkraska at san.rr.com] 
> 
> >  for i in len(mylist):
> >
> >Creates an iterator over the indices in mylist.
> 
> I just don't think so. It would be better to have something 
> which behaves slightly more like a classic for form. This 
> could almost be achieved by some sort of first class range 
> object thusly. In fact, one might even use slice notation:
> 
> for i in [0:len(mylist):1]:
> 
> ... where the stride is optional of course.

And exactly this notation has been rejected in PEP 204 -- basically
because it's butt-ugly in a Perl'ish sort of way. Besides, this goes
under the section of solving the "interval" problem which PEP 276
*explicitly* says it is not doing (and which I personally don't find
very compelling, especially if PEP 276 is adopted).

-- bjorn



From mhammond at skippinet.com.au  Sun Feb 17 06:12:26 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 17 Feb 2002 11:12:26 GMT
Subject: timer delays and kbhit() in Pythonwin?
References: <a4n85k$ij2$1@news.efn.org>
Message-ID: <3C6F9089.70200@skippinet.com.au>

Spencer Doidge wrote:
> The following script works fine running under Windows in the console, but it
> doesn't work right in Pythonwin.
> The point of this script is that while we are waiting 20 seconds for the
> 'hello' message to be printed, we can intervene with a keystroke and cause
> function doit() to terminate immediately.
> I suppose it doesn't work right in Pythonwin because kbhit() is only
> supported in the console. Is there a way to accomplish this same purpose
> running Pythonwin?

Not really - GUI programs are generally event driven.  You could display 
a dialog with a message and a cancel button.

Mark.



From gerhard at bigfoot.de  Sat Feb 23 11:33:34 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 23 Feb 2002 16:33:34 GMT
Subject: call me
References: <u7fcig3gep8k40@corp.supernews.com>
Message-ID: <slrna7fh30.a72.gerhard@lilith.my-fqdn.de>

Scott Hathaway wrote in comp.lang.python:
> Mom,
> 
> Give me a call when you have a chance.

Wow, a familiy of diehard Python fans :-)

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From martin at v.loewis.de  Wed Feb 13 02:33:29 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 08:33:29 +0100
Subject: Python 2.2 <-> Solaris 2.6 installation errors
References: <mailman.1013548101.10897.python-list@python.org> <i63k6uovaa07umml1b5bs28m88h8ffsira@4ax.com>
Message-ID: <m3sn853kfq.fsf@mira.informatik.hu-berlin.de>

Chris Wysocki <news at chriswysocki.com> writes:

> Any ideas, things to try?

It would help to find out why _built_objects isn't set. In theory, the
build_extension definition in setup.py is supposed to work this way:

- call the super build_extension
- if that fails, the extension will not have been build;
  print a warning, and be done
- if that succeeds, the extension is build, and self._built_objects
  is set to indicate which object files have been created
- the extension is imported. If that fails, the extension and its object
  files will be removed.

Now, for some reason, this does not happen on your system. You invoke
setup.py a second time, and what now happens is:

- call the super build_extension
- this determines that nothing has to be done, and returns without
  building anything, and not setting _built_objects
- importing the module fails, but _built_objects is not set.

It is understandable why it fails crashes the second time. What is
surprising is why it did not manage to remove the faulty readline
extension the first time (when it really did build it). Unfortunately,
Jeff did not report the output from the first make invocation, so this
is somewhat hard to tell.

Regards,
Martin



From hugomartires at hotmail.com  Fri Feb  8 08:16:49 2002
From: hugomartires at hotmail.com (Hugo Martires)
Date: 8 Feb 2002 05:16:49 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <st888.7971$oI4.33917985@newsserver.ip.pt> <a3retd$19cgik$1@ID-11957.news.dfncis.de>
Message-ID: <d0380df9.0202080516.7a76905d@posting.google.com>

"Emile van Sebille" <emile at fenx.com> wrote in message news:<a3retd$19cgik$1 at ID-11957.news.dfncis.de>...
> "Hugo Martires" <hmartires at cultalg.pt> wrote in message
> news:st888.7971$oI4.33917985 at newsserver.ip.pt...
> > but i don?t need all the numbers, i only need 0.6
> 
> See 4.98 in the FAQ at http://python.org/doc/FAQ.html

I have read it but still confused.
Se my point:
--> n=0

myArray = array.array('f', [0.5, 0.5, 0.5, 0.5]

myArray[i] = myArray[i] + (1.0/n)

myArray[0] = 0.600000023 ("something like this")

for i in range ( len(myArray) ):
	if (myArray[i] > 0.0) and (myArray[i]<1.0):
		exit(0)

The problem is that he it never exit, unless i do like this:
for i in range ( len(myArray) ):
	if (myArray[i] > 0.0001) and (myArray[i]<0.9999):
		exit(0)

that is not the better way :(

------------------------------------------------------------------
I have tried with:
    myArray=array.array('f')
    myArray=array.array('d')
and even with a list:  myArray=[]

I know that the problem is in the internal storage of the numbers, but
i canot understand how to solve that.
I just want to make a simple calculation and an if statement.

Tanks for help


From emile at fenx.com  Fri Feb  1 18:53:38 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 15:53:38 -0800
Subject: Windows, RS232, and Python
References: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org> <3c5ada50$0$2204$7586b60c@news.frii.com> <Pine.GSU.4.21.0202011548230.28552-100000@garcia.efn.org>
Message-ID: <a3fa3p$15oq1m$1@ID-11957.news.dfncis.de>

"Spencer Ernest Doidge" <spencer at efn.org> wrote in message
news:Pine.GSU.4.21.0202011548230.28552-100000 at garcia.efn.org...
> Definitely this helps, I think. The author mentions "the win32
extension
> modules." Do you know what exactly he is referring to?
>

http://starship.python.net/crew/mhammond/win32/Downloads.html

--

Emile van Sebille
emile at fenx.com

---------



From donnal at donnal.net  Mon Feb 11 08:52:16 2002
From: donnal at donnal.net (Donnal Walter)
Date: 11 Feb 2002 05:52:16 -0800
Subject: Telnet and SO_LINGER
Message-ID: <918bc22f.0202110552.5296ec87@posting.google.com>

I am trying to use telnetlib.Telnet objects to communicate with two
different servers.

On one of the servers, the Telnet object works just fine off the
shelf, but on the other I keep getting
socket.error: (10054, 'Connection reset by peer').

I have a hunch that my problem may be related to one of the socket
options, specifically SO_LINGER, but from the documentation I cannot
quite figure out the syntax for changing it.

In the following method:
tn.get_socket().setsockopt(level, optname, value) 
what does the 'level' parameter refer to? Also, I presume that
'optname' is SO_LINGER (or 128), but what is the 'value'? Is it a
scalar or a tuple?

Donnal Walter
Arkansas Children's Hospital


From boblancaster at zxmail.com  Mon Feb 25 11:16:21 2002
From: boblancaster at zxmail.com (Bob Lancaster)
Date: 25 Feb 2002 08:16:21 -0800
Subject: Python and Oracle 9i
Message-ID: <b470cb58.0202250816.7839c1c4@posting.google.com>

I am in a bit of a quandry. 
I am writing a python module which converts some data from various
python variables, lists, etc. into SQL, which in turn is inserted to
or selected from a database.

I need to support three combinations.
(1) PostgreSQL and Linux, for academic customers.  No serious problems
here.  In fact, I have learned why folks love PostgreSQL so much.

(2) Oracle 9i and Linux.  I haven't tried this one yet, but it seems
to be at least as easy as the third combination.

(3) Oracle 9i and Windows 2000, for industrial customers.  Trouble!
I have tried two approaches.  
  (a) DCOracle2:  builds, connects, but then does not work properly. 
I see two possible sources of the problem.

The documentation for DCOracle2 says:

"""Oracle installs the C headers to build OCI programs in
$ORACLE_HOME/rdbms/demo, $ORACLE_HOME/network/public, and
$ORACLE_HOME/plsql/public. If these directories do not have the proper
include files, the build will not succeed. Oracle typically only
installs these files when a complete server installation is
performed."""

 
Okay, under my ORACLE_HOME directory, I have rdbms/demo, which has
lots of header files, I have plsql/public, which has a header file,
but my ORACLE_HOME/network dircetory does not have a /public
subdirectory.  I not sure why DCOracle2 would even build without this
subdirectory.  Have the folks at Oracle, in their infinite wisdom,
chosen to place the files elswhere so that I could copy into a
/network/public directory? Or am I screwed?

The other possible problem is that I am using Python 2.1.1.  I rewrote
the install.py to accept 2.1.1 in addition to 2.1, since the Python
home page no longer lists 2.1 as a possible download.  (In fact, they
only list 2.1.2 these days.) Would using 2.1.1 instead of 2.1 make
that big a difference?

(b) I tried using cx_Oracle.  This seemed like the best solution,
until I started doing more complicated stuff.  First, it sometimes
crashes when I do a fetchall().  (Error message 50). I have looked at
newsgroups, etc. and I saw that I am not the only one to ever have
this problem.  Second, although I could insert and update
TIMESTAMP_WITH_TIMEZONE and user-defined types, I could not retrieve
them.  I checked the source code, and cx_Oracle is VERY limited in the
types it handles.  I would hate to have to rewrite that module and
recompile the whole thing, especially since I can't solve the
fetchall() problem.


It appears that I have three options at this point, none of which are
very good.  All of them may require some substantial reworking of my
database, alas.

I.  Somehow get DCOracle2 to work with Oracle 9i.  This seems to be my
best bet, if I can figure out how.

II.  Figure out how to get around the fetchall() problem with
cx_Oracle, and put all my user-defined types into VARCHARs, or
seperate tables, or whatever.

III.  Use Jython and connect with the JDBC and a JDBC:ODBC bridge. 
This seems to be the least satisfying, since I want to be able to put
everything into a single package.


Any advice would be most welcome.

Thanks,
Bob

BTW, I have an exinct email listed, to avoid spam.  If you really need
to contact me, try rlancasteratbruker-axsdotcom, converting the at and
the dot to symbols.


From bc90021 at bc90021redundantdot.net  Mon Feb 18 17:32:57 2002
From: bc90021 at bc90021redundantdot.net (bc90021)
Date: Mon, 18 Feb 2002 22:32:57 GMT
Subject: Newbie question...
References: <3C716B27.3090408@bc90021redundantdot.net> <3C71750E.9020305@bc90021redundantdot.net>
Message-ID: <3C7180F3.9080706@bc90021redundantdot.net>

X-No-Archive: Yes

>> error: failed dependencies:
>>     rpmlib(PartialHardlinkSets) <= 4.0.4-1 is needed by python-2.2-7
> 
> Yes, I really like answering my own posts.  ;)


See?  I did it again.  ;)

> I found the below regarding the above, but it doesn't really tell how to 
> fix it... I'm upgrading my RPM rpm to 4.04, to see what happens.


Upgrading worked.  I had to force the RPM package to be installed, but 
it got rid of that error, so now all I have to do is figure out the 
libcrypt stuff...

bc90021



From phr-n2002a at nightsong.com  Thu Feb 21 07:50:23 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 04:50:23 -0800
Subject: Why this behavior? (was: Stupid Lambda Tricks)
References: <3c74c211.20012718@news.isar.de> <js5d8.116$Az6.466@news.hananet.net>
Message-ID: <7xbsejj8xc.fsf@ruckus.brouhaha.com>

Jonathan Gardner <jgardn at alumni.washington.edu> writes:
> I've always wondered:
> 
> >>> "abc".upper()
> "ABC"
> >>> [3,1,2].sort()
> < nothing >
> 
> Why is that? Is it just easier to implement it this way, or is there some 
> sound philosophical reason for this?

upper() doesn't alter its argument.  sort does.


From hack at socket.net  Thu Feb 21 09:08:07 2002
From: hack at socket.net (Bill Hackmann)
Date: 21 Feb 2002 06:08:07 -0800
Subject: Suggestion for best intro to Python
Message-ID: <c510a321.0202210608.1acc4d29@posting.google.com>

What book/web site/online document/something else would be best as a
source to learn Python for an experienced programmer?

Thanks,
Hackmann


From deathtospam43423 at altavista.com  Sat Feb 16 21:24:44 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 17 Feb 2002 02:24:44 GMT
Subject: Newbie: Problems with File IO
References: <Xns91B873A49D077AusYourStandingInIt@210.49.20.254> <Xns91B81F8828230cliechtigmxnet@62.2.16.82>
Message-ID: <Xns91B8833A74178AusYourStandingInIt@130.133.1.4>

Chris Liechti <cliechti at gmx.net> wrote in 
news:Xns91B81F8828230cliechtigmxnet at 62.2.16.82:

> Flavian Hardcastle <deathtospam43423 at altavista.com> wrote in 
> news:Xns91B873A49D077AusYourStandingInIt at 210.49.20.254:
> 
>> out_file = open("log.txt", "w")
> mode 'a' does append (thats the one you want), 'w' starts a new file 
> for writing, adding a plus 'a+' or 'w+' opens for read/write
> 

Thanx! That's got me a bit further. What part of the standard Python manual 
(the html one you get at Python.org) would that be under?

Also, if you could give me a hint as to how I might get the script to 
remember people's usernames everytime they post, that'd be good.

-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From threeseas at earthlink.net  Sun Feb 17 17:06:23 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 17 Feb 2002 22:06:23 GMT
Subject: Apology
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <ketv6u49deaj528kctcpj83s8daff0es37@4ax.com>
Message-ID: <1896.813T2154T10725160threeseas@earthlink.net>

On 17-Feb-02 13:25:37 Courageous <jkraska at san.rr.com> wrote:
>>                   under the GPL (and I reserve the right to
>>change that to a more secure license should there be ...

>No matter what changes one makes in a licensing scheme, no
>unilateral change in licensing can withdraw rights to current
>licensees (which in your case with the GPL represent the
>entire human race) unless that license has a specific clause
>allowing that. GPL'ing something is a one-way ticket for the
>entire licensed body of work up to that moment in time. For
>all those licensees -- all humans, I mean -- your change in
>licensing terms for that past body of work is irrelevant
>masturbation.

>Any future work you do of course, is up to you. Which is
>indeed begging a question! Given that you've outright
>admitted that you expect OTHER PEOPLE to do the coding, it
>really begs the question why you think YOU could fiddle
>with the licensing at all for something which could not be
>regarded by the sharp eyes of the law as YOURS. You CANNOT
>dictate the rights of a third party to their work. Its
>THEIRS, not yours.


>C//

in the flip flopping world of computers, you really are a Courageous
COWARD. Instead of dealing with real world problems, you find you have
to create problems that don't really exist. Such as the above example
of a problem you have fabricated out of absolutely nothing.

You are aware that the Python License was changed to be GPL compatable
after the fact of code?

Humorously stated to be a bug fix.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jkraska at san.rr.com  Sat Feb  2 11:43:41 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 02 Feb 2002 16:43:41 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012618084.26466.python-list@python.org> <a3gu2l$om7$1@panix2.panix.com>
Message-ID: <m66o5uodgg4qikc2itemvsikjmmspb2a0i@4ax.com>

>Yes.  And nine times out of ten, if you think you disagree with Guido's
>intuition, you'll change your mind later.

This is more or less true, yup. Just consider the vast hordes of
programmers who, upon encountering Python, say "yuck! whitespace
sensitive language!" who, only after trying it for a while, see
the light. I was in that boat several years ago. It kept me from
learning Python for a while. But then. But thennnnnnn. I _tried_
it.

I _still_ don't like yield, however. :) :) :)

And yes, I've tried it. :)

C//



From chris.gonnerman at newcenturycomputers.net  Wed Feb  6 00:43:56 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 5 Feb 2002 23:43:56 -0600
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net> <3c60c05c$0$79564$3c090ad1@news.plethora.net>
Message-ID: <004501c1aed1$5ab39140$0101010a@local>

----- Original Message -----
From: "Peter Seebach" <seebs at plethora.net>


> In article <1372.801T1070T13994028threeseas at earthlink.net>,
> Timothy Rue <threeseas at earthlink.net> wrote:
> >This way I don't have to try and figure out why you have a problem
knowing
> >the difference between what is idea and what is natural law physics. For
> >MS seems to think it own reality. It doesn't!
>
> This isn't about ownership, it's about finding the community most
naturally
> suited to your project.  The C# people are the community least
contaminated
> by existing notions of software engineering.

Does even M$ deserve Rue?  Isn't that cruel and unusual punishment?





From wayne at mishre.com  Mon Feb 11 13:44:28 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Mon, 11 Feb 2002 10:44:28 -0800
Subject: Newbie needs references
Message-ID: <3C68110C.4040700@mishre.com>

> Hi<

Hello, welcome to c.l.py ... be warned, it's addictive.

> I'd like to learn Python, but everything I find on the web is
>     way above my head.
> Does anyone have books or tutorials they would recommend
>     for an illiterate to teach himself Python?

You might want to start off at Python for Beginners[1], there are 
several documents there that would be a good starting place.  What you 
might want to read after that would depend a lot on what you want to do; 
  i.e., GUIs, networking, Internet, etc.

A good starting book would be Learning Python[2], it's a good book, but 
you might grow out of it quickly.  A few other good books are Python 
Programming on Win32[2] and Programming Python[2].  I've also seen some 
good reviews for Core Python Programming[2] as a Beginner/Intermediate 
level book, but I've never read it myself.

[1] http://www.python.org/doc/Newbies.html
[2] http://www.amk.ca/bookstore/python.html

HTH,

-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From mks99 at t-online.de  Sat Feb  2 14:25:59 2002
From: mks99 at t-online.de (=?iso-8859-1?Q?Markus_Sch=F6nhaber?=)
Date: Sat, 2 Feb 2002 20:25:59 +0100
Subject: Why it does NOT work on Linux ?
References: <3C5AAA9E.18034.11AB3F0@localhost>
Message-ID: <000901c1ac1f$781796a0$012aa8c0@sauron>

> I have the following part of program that finds ItemID numbers.
> Here, for example, are two
> 146759 and 146700 .
> This program works well under windows but on Linux it does not
> find any number. Can you please help?
> Thanks.
> Ladislav
>
> ####################
> import re
> Text="""<tr BGCOLOR="#FFFFFF">
>                       <td valign="top" align="left"><a
> href="lead.asp?ItemID=146759">[CN] Oak, Foiled &amp; Antique
> Furniture</a></td>
>                       <td valign="top" align="center">18/12/2001</td>
>                     </tr><tr BGCOLOR="#FFFFFF">
>                     <td valign="top" align="left"><a
> href="lead.asp?ItemID=146700">[CN] Oak, Foiled &amp; Antique
> Furniture</a></td>
>                       <td valign="top" align="center">18/12/2001</td>
>                     </tr>"""
>
> IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> Results=re.findall(IDs,Text)
> print Results

The interesting thing is, that it works at all for you. It definitely
doesn't on my WinXP machine.

1.) Here
> IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> Results=re.findall(IDs,Text)

you call re.findall with a regular expression object as a first parameter
which should be a string. What you want to do is

Results = IDs.findall(Text)

i. e. call the appropriate method on the re object you created.


2.) There are two whitespaces (a space and a newline - the latter may be
inserted by your or my mail agent) between "<a" and "href...". So you should
replace your re with something like this:

IDs = re.compile('<a\s*href="lead.asp\?ItemID=(\d{5,10})', re.MULTILINE)

Since you are using findall, the enclosing ".*" expressions are superfluous.


BTW: Be careful reagarding backslashes in REs since the string gets
interpreted two times.

Regards
  mks





From morton at dennisinter.com  Mon Feb 11 01:43:58 2002
From: morton at dennisinter.com (damien  morton)
Date: 10 Feb 2002 22:43:58 -0800
Subject: subclassing tuple
Message-ID: <4abd9ce7.0202102243.d078f9c@posting.google.com>

Ive been trying to subclass the tuple type, without a whole lot of success

Heres the code Ive been using:

class triple(tuple):
    def __init__(self, x, y, z):
        tuple.__init__(self, (x,y,z))

>>> t = triple(1,2,3)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: tuple() takes at most 1 argument (3 given)

For some reason, subclasses of tuple dont call the proper __init__ code.

Am I missing something here?


From mark at codesourcery.com  Mon Feb 18 13:34:07 2002
From: mark at codesourcery.com (Mark Mitchell)
Date: Mon, 18 Feb 2002 10:34:07 -0800
Subject: Synchronous signals vs. robust code
In-Reply-To: <3cd27uo3p0suf0ds56gmmr9ra9heasfaan@4ax.com>
Message-ID: <15960000.1014057247@warlock.codesourcery.com>


--On Monday, February 18, 2002 05:34:14 PM +0000 Toby Dickenson 
<tdickenson at devmail.geminidataloggers.co.uk> wrote:

> (reply posted, and cc to Mark)

Thanks for CC'ing me; I'm not on the list, so up until now I've
been keeping up with follow-ups by reading the list archives.

> No major damage is done if the exception ends up terminating this
> python process,

Yes.  In almost any environment, one must be willing to accept that
one's process may be summarily killed by the operating system at any
time, or that the power may go off...

> or if the exception is only caught after sufficient
> objects have been destroyed that it has lost all reference counts on
> the crucial file objects.

I'm don't think it's a critical point, but this won't happen here;
Python won't close the pipe when the descriptors are no longer referenced.
There's no object involved and no __del__ method to be invoked; those
file descriptors are just numbers.  So, no, we can't just let the
exception escape far enough and then rely on GC to clean up the
descriptors.

Abstracting away from pipes, the point is that when one writes:

  try:
    f()
  finally:
    g()

one intends that one of two things will happen:

1. f() will not terminate, either because it runs forever, because it
   calls "exec", or because the process is killed, or ...

2. f() will eventually terminate, either by throwing an exception or
   by returning, in which case "g()" will be executed.

In C++, and I believe in Java, and in most other languages with similar
features, those are precisely the two outcomes that can occur.

>> One solution, of course, is to make sure that the signal-handlers do
>> not throw exceptions.  But that's not a reasonable solution in library
>> code; the library can't go around deciding what signal handlers should
>> be doing.
>
> A library working to those standards is already in trouble without
> introducing anything so tricky as a signal. In python almost any
> operation can potentially raise an exception: function calls, slices,
> and even assignments to globals or class attributes may involve memory
> allocation which may fail.

It is true that out-of-memory errors can occur almost anywhere, and
that they are hard to handle.

However, robustness is a matter of degree: some applications are robust
in the presence of erroneous input, some in the presence of
unavailable resources, some in the presence of hardware failure, and
so forth.

So, leaving aside *other* ways in which it may be hard to write robust
Python programs, I'd like to refocus on signals.

Is it possible to write Python programs that are robust in the presence
of signals, in the sense that I originally indicated?

If not, would people consider adding language support to provide this
robustness, so that Python can be used in applications that require
that level of robustness?

(Footnote: at least one interpreted language -- Scheme48 -- handles
signals by making them *more* synchronous than Python; the application
must explicitly poll for signals when it wants to handle them.  That
avoids the signal-safety issues I've raised, at the expense of making
it difficult to respond to signals in a timely fashion.)

-- 
Mark Mitchell                mark at codesourcery.com
CodeSourcery, LLC            http://www.codesourcery.com



From janfbr at sex.ifi.uio.no  Sat Feb 16 12:08:10 2002
From: janfbr at sex.ifi.uio.no (Jan-Fredrik Braseth)
Date: 16 Feb 2002 18:08:10 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org>
Message-ID: <xyuofipl5hh.fsf@sex.ifi.uio.no>

Sheila King <usenet at thinkspot.net> writes:

> On 16 Feb 2002 13:56:35 +0100, Jan-Fredrik Braseth
> <janfbr at sex.ifi.uio.no> wrote in comp.lang.python in article
> <xyuvgcxlh4s.fsf at sex.ifi.uio.no>:
> 
> >  but after some time I get this error message:
> > 
> > Traceback (most recent call last):
> >   File "./quizbot.py", line 366, in ?
> >     foo = QuizBot(name,password,channel)
> >   File "./quizbot.py", line 28, in __init__
> >     line = self.tn.read_until("\n").split()
> >   File "/usr/lib/python2.2/telnetlib.py", line 297, in read_until
> >     self.process_rawq()
> >   File "/usr/lib/python2.2/telnetlib.py", line 424, in process_rawq
> >     self.msg('IAC %d not recognized' % ord(opt))
> > UnboundLocalError: local variable 'opt' referenced before assignment
> > > 
> You have an error in your code. Show us the code. I'm guessing, that you
> have a variable 'opt' that under some circumstances is not defined
> before entering a particular loop or something.

If you look closely at the traceback you see that the error with opt is in the
telnetlib.py, that's a module for python. But it's unlikely that there is a bug
in python itself, isn't it?

BTW: it's python 2.2

-- 
Jan-Fredrik Braseth		www.ifi.uio.no/~janfbr/
2B|!2B


From cliechti at gmx.net  Sun Feb 10 11:36:39 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 10 Feb 2002 17:36:39 +0100
Subject: wxpython and images
References: <6bd9f01b.0202092351.3d73391f@posting.google.com>
Message-ID: <Xns91B1B3AD2B2Bcliechtigmxnet@62.2.16.82>

sameer_ at email.com (sameer) wrote in 
news:6bd9f01b.0202092351.3d73391f at posting.google.com:

> probably something silly i'm missing here, but why is that when i 
run
> the wxpython demo (on win2k), wxImage shows jpg, gif, png, and bmp
> images, but when i run that same file from another directory, only 
bmp
> images work, and wxpython gives a  No image handler found type 17
> error.
> 

somewhere befor the images are loaded you must call:
wxInitAllImageHandlers()

you can do it once, globaly for the entire app.

chris

-- 
Chris <cliechti at gmx.net>



From sholden at holdenweb.com  Wed Feb 20 18:25:25 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 18:25:25 -0500
Subject: HTTP state management without cookies?
References: <BFFEB31BC23CCD44B8924E139E1D87E41240D0@MAILBOX.nameconnector.com>
Message-ID: <004d01c1ba65$e122c4e0$d200000a@holdenweb.com>

----- Original Message -----
From: "Geoffrey Talvola" <gtalvola at nameconnector.com>
To: "'Steve Holden'" <sholden at holdenweb.com>; <python-list at python.org>
Sent: Wednesday, February 20, 2002 11:15 AM
Subject: RE: HTTP state management without cookies?


> Steve Holden wrote:
> > ... That's why a recent
> > thread concluded you EITHER pass session id through all URLs
> > OR use a cookie
> > to identify the session.
>
> Webware's WebKit app server has a different approach that doesn't
require
> cookies and also doesn't require adding the session ID to all URL's.
It has
> an option to use "automatic path sessions".  If you turn on this
option,
> then WebKit will redirect a request without a session ID like:
>
> http://foo.bar/cgi-bin/WebKit.cgi/MyServlet
>
> to:
>
> http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet
>
> Since the session ID comes _before_ the name of the servlet, all
relative
> links work just fine.  WebKit knows how to parse the URL and extract
the
> session ID.
>
> What makes this work is that WebKit passes all requests through a
single CGI
> script (or uses the special mod_webkit Apache module) and uses its own
logic
> to determine how to map the URL to a servlet file.  It would be hard
to make
> this work with plain CGI.  Perhaps you could use some mod_rewrite
magic to
> transform:
>
> http://foo.bar/_SID_=2374820394/myscript.cgi/
>
> into:
>
> http://foo.bar/myscript.cgi?_SID_=2374820394
>
> so the same trick will work for CGI?
>
Geoff:

I appreciate the slightly unusual nature of thios scheme, but I don't
see how it isn't included under the heading "pass session id through all
URLs" as mentioned in my original post.

There's a lot to be gained by ceasing to treat URLs as simple points in
a tree-structured virtual address space, and this highlights some of the
benefits. Thanks for reminding me about the WebKit approach.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From matthias_j at lycos.com  Tue Feb 12 16:32:42 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 12 Feb 2002 13:32:42 -0800
Subject: Just for the gurus -- Remote admin password changing with privileged user ??
Message-ID: <d7d5ebdf.0202121332.4f7d3723@posting.google.com>

Just a small question:

I'm working on a PC (Win NT) loggod on as user Matthias to the DOMAIN
VIENNA:

I would like to execute localy a python script to CHANGE THE PASSWORDS
for
2 Administrators (SuperUser, TechAdmin) which are remote (60 NT
Servers)on 60 different Domains where there is no user account
Matthias.

I thought of using the win32net.NetUserChangePassword to do the job.

             

            BUT THE REAL PROBLEM is that I do not have admin rights on
the other domains.

I could copy the script to all the Domains and execute it logged on as
Admin - but I would realy like to avoid this if possible.  Maybe there
is a way to send an authetification to the domains like
username:TechAdmin + the old passwort - (like a Su command) I'm not
sure..... -- ANY HELP

Would be absolutly create.

THANKS


From jason at jorendorff.com  Mon Feb 11 04:28:53 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 11 Feb 2002 03:28:53 -0600
Subject: Standing Objects ... possible?
In-Reply-To: <7x665fgmmc.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOELBDBAA.jason@jorendorff.com>

Paul Rubin wrote:
> Jason Orendorff wrote:
> > Philip Swartzleonard writes:
> > > What i want is a construct like:
> > > 
> > > object Foo( ancestor ):
> > >     	pass
> > 
> > Yep, it exists.
> 
> This has gone past me somehow.  Can someone explain it?  Thanks.

I will, but I warn you that this isn't understood by many people.
You must use this knowledge only for good.  :)

When you write:

  class X(BasicX, OtherBaseClass):
      name = 'foo'
      bar = 2
      blah = None

what happens as a result is a lot like this:

  X = type('X',
           (BasicX, OtherBaseClass),
           {'name': 'foo', 'bar': 2, 'blah': None})

type is itself a builtin type; when Python calls it like that,
it creates a "type object", which is the same as saying that
it creates the class X.  The arguments, you can see, are
determined by the source code for the class.  They are:

  'X' - the class name
  (BasicX, OtherBaseClass) - a tuple of the base classes
  {...} - a dictionary of the new class's members.

If you define methods within class X, then those functions
go in the member dictionary too.

Here, try it yourself:

>>> X = type('MyType', (), {'n': 5})   # Create a class manually!
>>> X
<class '__main__.MyType'>
>>> X.n
5

Weird, huh?

"Metaclassing" means asking Python to call something else,
instead of type(), to create the class.  There are a few ways
to trigger this.  The easiest is to use __metaclass__.

def make_a_type(name, bases, dict):
    """ My custom function for creating a class. """
    # Do some silly things...
    print "Making a type!"
    print "A message for you:", dict.get('msg')
    dict['made_by'] = 'Jason'  # add my name here
    return type(name, bases, dict)  # type() does the real work

class X:
    __metaclass__ = make_a_type
    msg = 'Hello, world!'

>>> X.made_by
'Jason'

You can use whatever callable __metaclass__ you want.
For example, your __metaclass__ could be an actual class whose
constructor takes three arguments.

class Course(object):
    # This is actually a real, ordinary class.
    # Since we want to use it as a metaclass, its
    # constructor must take 3 arguments (plus 'self').
    def __init__(self, name, bases, dict):
        self.name = name
        self.bases = bases
        self.dict = dict
    def __repr__(self):
        return "<%s>" % self.name

class CS1280:
    __metaclass__ = Course
    professor = 'Cleese'
    title = 'Introduction to Programming with SPAMTRAN'

class CS6196:
    __metaclass__ = Course
    professor = 'van Rossum'
    title = 'Metaprogramming in Python 2.2'

Now you see that this is creating "standing objects" as desired.
Python creates CS1280 and CS6196 by calling Course(); so they
are Course objects.

(You could make the class Course a subclass of type,
if you really want the objects to behave like typical classes
rather than typical objects.)

(There is also a subtler way of getting a metaclass other than
type(); but I won't get into it.  For a more technical overview,
consult http://www.python.org/2.2/descrintro.html#metaclasses .)

## Jason Orendorff    http://www.jorendorff.com/



From sholden at holdenweb.com  Mon Feb  4 08:37:17 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 08:37:17 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com>
Message-ID: <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xsn8hjxzw.fsf at ruckus.brouhaha.com...
> "Tim Peters" <tim.one at home.com> writes:
> > > That's the problem--if I don't import future division, my code will
> > > break in the future, because the // operator isn't supported in the
> > > present.
> >
> > Not so:  introducing "//" was 100% backward-compatible, so isn't
controlled
> > by __future__.  "//" works out of the box in 2.2, no __future__
required.
>
> If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> it isn't backwards compatible.
>
Erm, don't you seem to be using a different version of backwards
compatibilty from most people? As far as I'm concerned, if I run version
1.5.2 of a language, and all my programs continue to run, then the new
release provides backwards compatibility.

> > > (The most recently released version of Python is a 2.1 series release,
> > > so 2.1 is still current).
> >
> > I don't know where you're looking, but 2.2 final was released.
>
> I know this, but 2.1.1 was released more recently than 2.2, so the most
> recent release was in the 2.1 series.
>
Now you are clearly looking for a fight :-) This is bogus. "Current" surely
refers to "processes the most recent language definition". Tim's right, this
is rhetorical gimmickry. How appropriate your news server is
ruckus.brouhaha.com.

> > Bugfix releases for past versions may (or may not) get released from
> > time to time, but if you decide to build a 1.5.3 someday, that won't
> > make 1.5 "current", it will just make 1.5.3 the most recent release
> > of the 1.5 line (as 2.1.2 is the most recent release of the 2.1
> > line, and contains none of the features new in the 2.2 line --
> > active development of 2.1 ended the day 2.1 final was released, so
> > calling it "current" is rhetorical gimmickry).
>
> If the official Python maintainers think they might release a 1.5.3
> someday, then it's fair to say that the 1.5 series is still current.

If you ever collected stamps, you'd be aware that stamps can be "obsolete"
(no longer issued) and they may also be "demonetized" (no longer acceptable
as postage). In the same way, 1.5.2's survival in various Linux
distributions does not stop it being obsolete. Fortunately it will only be
"demonetized" (unusable for programming) when there is no longer a platform
that will allow you to compile to interpreter sources. The open source
nature of the language means you can continue using 1.5.2 into the far
future.

But that isn't good enough for you, you want new language features, but you
want them to be backwards compatible with 1.5.2. In what strange version of
reality does this make them "new"?

> If they're unlikely to ever release another 1.2.x or 0.9.x, then
> those versions are not current, even if I release a 1.2.x myself.
>
Nonsense. Do you understand the meaning of the word "currency"? And stop
relying on the "official Python maintainers" - one of these is Tim Peters,
and you continue to disagree with him about this matter...

> More importantly, in the real world, a lot of people are still using
> 2.1 and 1.5, and few are using 2.2.  So coding anything in a way that
> requires 2.2 inconveniences a lot of users.  That might be worth it
> for serious features like metaclasses or generators, but is silly to
> do over this division nonsense (think of those obnoxious web sites
> that tell you to upgrade your browser before you can view some page).
>
Coding in a way that requires 2.2 inconveniences *only* those users who a)
want to use my code *and* b) won't move forward to 2.2.

I suppose that our writing these posts in English inconveniences many of the
group's readers for whom English is not the first language. They persist in
reading them because the benefits outweigh the disadvantages (in most cases,
anyway, though I'm beginning to have my doubts about this particular
thread). So, do you want 2.2, or 1.5.2. And please *don't* say you want to
be able to program 2.2 features in 1.5.2.

I still don't see why you want to be able to use "from __future__" in 1.5.2
programs. What's the point of importing future behavior when you can if you
also want to run in environments where such behavior isn't available? Surely
this means you're going to have to make your programs larger by
incorporating two different algorithms to solve the same problem. Surely it
would be better to set some version flag based on sys.version, then write

if ver152:
    import oldcrap
    mymod = oldcrap
else:
    import snazzynewstuff as mymod

or something similar.

> I'd non-rhetorically say that for most users, 1.5.2 or 2.1 is current
> and 2.2 is bleeding edge.  (I'm still using 2.1 and plan to get around
> to upgrading to 2.1.1).
>
This *is* purely rhetoric, despite your protestations, until you provide
evidence.

> > > I don't see any reasonable way I can write code right now that uses
> > > division that works in both the current 2.1 (which doesn't support
> > > importing future division) and in forthcoming versions (where the
> > > semantics of the / operator will be incompatibly different).
> >
> > The semantics of "/" didn't change in 2.2, and won't in 2.3, 2.4, 2.5,
...
> > either (read PEP 238).
>
> Yes, but we're talking about backwards compatibility (see the thread
> header).  "Backwards" from 2.2 means versions like 2.1 and earlier.
>
So, what about 1.5.2 and 2.1 programs won't run in 2.2? THOSE features are
the ones that break backwards compatibility (isn't this where we came in?)
not the new ones that won't run in older versions.

> Given that 2.2 can import 3.0 division from the "future", will 3.0
> have a way to import 2.x division from the "past"?  If I can put
>
>    try: from __backward_compatibility__ import use_old_division
>    except ImportError: pass
>
> into all my scripts, and have it maintain the [012].x "/" semantics in
> 3.0 and later while doing nothing in 2.x and earlier, then that at
> least gives me a way to write stable scripts, so I hope 3.0 can
> include that.
>
No. It has been PEP'd for a long time that division breaks in 3.0+, and
you'd better get used to the idea if you want to move forward [in a year or
two].

> > If you need a given static instance of "/" to do
> > radically different things depending on whether you feed it ints or
floats
> > dynamically, you're the first person to claim such a need.
>
> I want my programs to run the same way across the different Python
> versions that people are using now and will be using in the near future.
> Am I really the first person to want that?
>
No, you're not even the first person to be so wrong-headed in the way you
decry the chosen path for the language's development <0.8 wink>. Most unlike
you, if I might make a personal observation. Just the same, I claim that
Python supports a *remarkably high* degree of backwards compatibility. Very
few 1.5.2 programs broke in 2.0, and most of those (from memory) were using
a non-documented "feature" of the sockets library.

> > Else you have a specific flavor of division in mind when you type
> > "/", and it's easy to write it in such a way that it will deliver
> > the same results in 3.0 as in 0.9.8.
>
> By using divmod(x,y)[0] instead of x/y everywhere I want to divide
> ints, per the PEP?  Ugh!

Hey, under 3.0 you'll be able to redefine __div__ for everything so you get
exactly the behavior you want. Not sure I'd like to maintain that code,
though ;-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From chris.gonnerman at newcenturycomputers.net  Tue Feb  5 08:29:43 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 5 Feb 2002 07:29:43 -0600
Subject: OT again... is there an IMAP expert in the house?
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org> <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org> <mailman.1012224450.11377.python-list@python.org> <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu> <mailman.1012892049.28494.python-list@python.org> <Pine.OSF.4.44.0202042332410.184841-100000@goedel2.math.washington.edu>
Message-ID: <001d01c1ae49$421c96a0$0101010a@local>

----- Original Message -----
From: "Eduardo Chappa" <chappa at math.washington.edu>


> *** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:
>
> :) > This is not a problem about the protocol you are using to read your
> :) > folder. IMAP is the way that you communicate to the server. The way
> :) > that the mail is stored in the server is the relevant point here. For
> :) > example, the default for the UW server is unix, and unix (format)
> :) > does not allow concurrent access to a mailbox. If you use the UW
> :) > server, and you use mbx format, then you can have concurrent access
> :) > to your folders. Convert your folders to mbx format and you'll be
> :) > done.
> :)
> :) Tried that, didn't work.  I was using Netscape 4.75 as one client,
> :) Outlook Express as the other, and concurrent deletion (even with
> :) expunges) just didn't work.  Outlook Express never seemed to notice
> :) Netscape doing deletions, and I couldn't get Outlook Express to
> :) properly expunge without closing it.
>
> Just to be sure you did that. What are the first characters of the first
> line of the folder that you want to share?

Gah.  I didn't keep the files around, so I can't tell you for sure.  I did
look at them with vi after conversion and they definitely weren't Unix
mailboxes.

UW's documentation is weak in spots.  Assuming I could get it working, the
IMAP server would have to reside on a Win98 computer, and I have not been
able to find step-by-step instructions.

It would be easier for me if I had time to grok IMAP like I do SMTP and
POP3.
This is my first painful exposure.





From peter at engcorp.com  Sun Feb 24 14:52:49 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 14:52:49 -0500
Subject: FSF award to...
References: <498.812T2576T10686134threeseas@earthlink.net> <mailman.1013898268.3969.python-list@python.org>
Message-ID: <3C794491.25D2A381@engcorp.com>

Geert-Jan Van den Bogaerde wrote:
> 
> For the curious, it was awarded at FOSDEM (Free Software and Open Source
> Developers European Meeting) in Brussels (Belgium) just a few hours ago.
> In fact, I just got back from there.
> 
> Guido was there to receive the award (though I don't think he's actually
> giving a talk at FOSDEM unfortunately).
> 
> RMS commented that he personally has not used Python, but has been told
> it is a very good language, and that it incorporated many elements from
> LISP.

>From the release:
"""
   The selection committee included: Miguel de Icaza, Ian Murdock, 
   Eric Raymond, Peter Salus, Vernor Vinge, and Larry Wall.
   [...]
   Guido van Rossum was chosen from three finalists for the award. 
   [...]
   The prior winners were Larry Wall, Miguel de Icaza, and Brian Paul. 
"""

So will Guido be on the selection committee next year?  ;-)

-Peter


From emile at fenx.com  Mon Feb 25 16:52:13 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 13:52:13 -0800
Subject: On inner classes
References: <mailman.1014665653.16956.python-list@python.org> <e8al7us7vtknrn86cfd5a7rii9b3clt21b@4ax.com>
Message-ID: <a5ebkk$6ovpi$1@ID-11957.news.dfncis.de>

"Gon?alo Rodrigues"
> How in the __init__ of the inner class do i acess the instance
> attributes declared in the __init__ of the outer class? In this case the
> instance attribute example?
>

Pass it in when init'ing the outer_class instance?

class outer_class:
    def __init__(self):
        self.example = "just a test"

    class inner_class:
        def __init__(self, outer_self):
            print outer_self.example
    def doit(self):
        ic = self.inner_class(self)

oc = outer_class()
oc.doit()


HTH,

--

Emile van Sebille
emile at fenx.com





From jjl at pobox.com  Sat Feb  2 16:46:55 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 21:46:55 +0000
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses,
 ...)
In-Reply-To: <mailman.1012678565.11760.python-list@python.org>
References: <mailman.1012678565.11760.python-list@python.org>
Message-ID: <Pine.SOL.4.30.0202022136180.597-100000@mimosa.csv.warwick.ac.uk>

On Sat, 2 Feb 2002, Tim Peters wrote:
[...]
> What the Python developers do get testy about is living with Guido's
> seemingly inconsistent demand that Python code use 4-space indents with no
> hard tabs, C code use 8-space hard-tab indents.  Oddly enough, the more
> intelligent of us agree with him, but nobody can explain why <wink>.

Intelligence here is as judged by preferred indentation style, I presume?

Hypothesis:

Long C type names -->

    tendency to use short variable names -->

    tendency to write too much on one line, or use too many indentation
      levels, on those lines containing no type names -->

    increase indentation to counter this tendency?

Of course, the 8 character tab width is axiomatic.

(I confess I did debate the indentation here, and the positioning of the
arrows)

Deadly serious,


John



From jeff at ccvcorp.com  Thu Feb 28 15:54:05 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 12:54:05 -0800
Subject: Newbie scopes question
References: <vvcq7uso4lujr5i5caodd6j2dv781322hh@4ax.com>
Message-ID: <3C7E98ED.3D1C83DA@ccvcorp.com>


"Gon?alo Rodrigues" wrote:

> As far as i understand, a class statement opens a new scope.
>
> In the following piece of code
>
> class Outer:
>     attribute = "The outer class attribute"
>
>     def __init__(self):
>         try:
>             print attribute
>         except NameError:
>             print "Attribute not defined!"
>
> def Outer_function():
>     local_var = "Local var of the outer function"
>
>     def inner_function():
>         print local_var
>
>     inner_function()
>
> Outer_function()
> a = Outer()
>
> What we get printed is
>
> Local var of the outer function
> Attribute not defined!
>
> This means that function and class scopes behave differently as fas as
> nesting goes. Could someone explain why is this?

Because they do?  :)  Perhaps it's because the problem that nested scopes
intended to solve, never existed with classes.  In the class example, you
can always access Outer.attribute, whether inside the class (or however
deeply nested) or not.  In the function example, prior to nested scopes,
there was *no* way for inner_function() to access local_var, without
having it passed in as an argument.

It makes philosophical sense, too.  Class/instance attributes are distinct
from local variables, and it makes no sense to refer to a class attribute
without specifying the class that you're interested in.  While some
languages (such as C++) will implicitly check unqualified variables to see
if they're class attributes, this is unclear and prone to bugs (explicit
is better than implicit).

Perhaps it'd make more sense, if you think of the class statement as not
opening a new scope, but rather, opening a new namespace.

Jeff Shannon
Technician/Programmer
Credit International




From martin at v.loewis.de  Mon Feb 11 08:49:44 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 11 Feb 2002 14:49:44 +0100
Subject: Bug ? xml.dom.minidom
References: <3C63D2F3.79EB997F@yahoo.com>
Message-ID: <m31yfsf7rb.fsf@mira.informatik.hu-berlin.de>

"A.A" <andtherewaswater at yahoo.com> writes:

> What is happening ???

I can't reproduce this. If you can provide more details, please submit
a bug report to SourceForge. Any chance you have PyXML installed? What
version?

Regards,
Martin



From martin at v.loewis.de  Mon Feb 18 18:44:04 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 19 Feb 2002 00:44:04 +0100
Subject: __class__ for 2.2 types
References: <2ec1bc1c.0202071240.44185531@posting.google.com> <a3upaq$1arn1e$1@ID-59885.news.dfncis.de> <a4otl1$44d$1@news.idiom.com> <a4p0ep$1ptki$1@ID-11957.news.dfncis.de> <G7fc8.189$Zp2.58090@atlpnn01.usenetserver.com>
Message-ID: <m3r8nixsmz.fsf@mira.informatik.hu-berlin.de>

"Steve Holden" <sholden at holdenweb.com> writes:

> Well, syntactically the period acts as a delimiter when there are no spaces,
> but here the spaces themselves are the delimiters. Technically isn't the
> period acting as an operator here, the operation being "find in namespace"
> or some such?

Not really. The period isn't really a delimiter. Instead, it may or
may not be a token of its own: the previous token extends as much as
it can, resulting in "2." being a single token, but "a." being two
tokens. White space normally really delimits a token, as it is not
part of the token, but is no token on its own, either (unless it is
part of a string literal or comment, in which case it does not delimit
a token).

Now, once the input is broken into tokens, it turns out that "." is
not an operator (atleast not more than the ':' in a lambda expression
is an operator, or the while keyword). It appears in the production
'trailer', as in

power: atom trailer* ('**' factor)*
trailer: '(' [arglist] ')' | '[' subscriptlist ']' | '.' NAME

In case you wonder what an atom is:

atom: '(' [testlist] ')' | '[' [listmaker] ']' | '{' [dictmaker] '}' 
      | '`' testlist '`' | NAME | NUMBER | STRING+

Regards,
Martin


From robin at jessikat.fsnet.co.uk  Fri Feb  8 04:26:59 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Feb 2002 09:26:59 +0000
Subject: FixedPoint
References: <000601c1b064$29d3a940$445d4540@Dell2>
 <mailman.1013148194.1285.python-list@python.org>
Message-ID: <19t8ILAjn5Y8Ew8y@jessikat.fsnet.co.uk>

It is perfectly clear that it's OK to ask if f==None for any f. The
whole point about the row over integer division was that it's ok to pass
different types into code and that code should behave reasonably under
all circumstances.

if 5.0==None ==> 0

then making FixedPoint(5.0)==None raise an exception is pretty
pointless.

A typical technique is to have a default argument value of None and use
that correspondence to initialise in some special way.
-- 
Robin Becker


From wookiz at hotmail.com  Tue Feb 12 02:07:19 2002
From: wookiz at hotmail.com (wooks)
Date: 11 Feb 2002 23:07:19 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com> <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu> <8852d33c.0202110117.374f062@posting.google.com> <915a998f.0202110647.2f5a3b01@posting.google.com>
Message-ID: <8852d33c.0202112307.396f4f14@posting.google.com>

So where do I find out what a given built in method is going to do
(source). Are these built ins overriden when you code a magic method.
Is that what happens.

It looks like I can see a practical value in this technique, vis a vis
what I want to do. There is a framework which provides an abstract
layer over a testing tool such that you can specify in a spreadsheet
what you want done without coding a script and the framework gets the
tool to do what you've asked. I want to port this framework to another
tool without changing the way the user communicates with the
framework. So if the user asks for X to be done it happens
polymorhically in whatever tool he wants. Equally if I ask for data to
be retrieved, it happens polymorphically whether the data is in CSV,
XLS or a database. In fact ADO is a good example of polymorphism in
action. Right.

hamish_lawson at yahoo.co.uk (Hamish Lawson) wrote in message news:<915a998f.0202110647.2f5a3b01 at posting.google.com>...
> 
> If an object doesn't have __add__ defined (by itself, its class, or an
> inherited class), then it won't have an interpretation *at all* for +.
> There's no 'standard behaviour' that is defaulted to. This holds true
> for the built-in types too: 4.0 + 3 has an interpretation only because
> the float type has __add__ predefined for it.
> 
> Thus
> 
>     4.0 + 3
> 
> effectively results in a call to
> 
>     (4.0).__add__(3)
> 
> The second form could actually be directly called that way in Python
> 2.2 (prior to 2.2, magic methods for built-in types weren't exposed to
> the programmer); the brackets around the 4.0 are necessary only to
> disambiguate the number from the following dot.
> 
> So there is no standard behaviour for + in the sense of some default
> action that gets performed in the absense of a defined __add__ method.
> However the __add__ method is already defined for most built-in types.
> 
> But the operator symbols provide more than just a prettier notation.
> When the operand types are different the interpreter will inspect the
> operands and will decide which operand to ask to handle the operation.
> 
> Thus 
> 
>     4 + 3.0
> 
> will effectively result in a call to
> 
>     (3.0).__radd__(4)
> 
> rather than
> 
>     (4).__add__(3.0)
> 
> since integers don't know how to add floats.
> 
> 
> Hamish Lawson


From brian at sweetapp.com  Tue Feb  5 14:11:13 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 5 Feb 2002 11:11:13 -0800
Subject: REQ : encoding windows cp1252 => iso latin 1
In-Reply-To: <a3p68d$1a56$1@norfair.nerim.net>
Message-ID: <000a01c1ae78$e0a51bb0$445d4540@Dell2>

Gillou wrote:
> My customers make copy/paste from M$ word docs to forms translated to
XML
> (expecting ISO latin 1 charset).
> My XML parser (pyexpat) does not accept cp1252 character, and I'm
looking
> for a function that can translate extra cp1252 characters to the
closest
> ISO latin 1 encoding.

I am such a nice guy. Here is a completely untested solution:

"""
80 20AC EURO SIGN 
81  UNDEFINED 
82 201A SINGLE LOW-9 QUOTATION MARK 
83 0192 LATIN SMALL LETTER F WITH HOOK 
84 201E DOUBLE LOW-9 QUOTATION MARK 
85 2026 HORIZONTAL ELLIPSIS 
86 2020 DAGGER 
87 2021 DOUBLE DAGGER 
88 02C6 MODIFIER LETTER CIRCUMFLEX ACCENT 
89 2030 PER MILLE SIGN 
8A 0160 LATIN CAPITAL LETTER S WITH CARON 
8B 2039 SINGLE LEFT-POINTING ANGLE QUOTATION MARK 
8C 0152 LATIN CAPITAL LIGATURE OE 
8D  UNDEFINED 
8E 017D LATIN CAPITAL LETTER Z WITH CARON 
8F  UNDEFINED 
90  UNDEFINED 
91 2018 LEFT SINGLE QUOTATION MARK 
92 2019 RIGHT SINGLE QUOTATION MARK 
93 201C LEFT DOUBLE QUOTATION MARK 
94 201D RIGHT DOUBLE QUOTATION MARK 
95 2022 BULLET 
96 2013 EN DASH 
97 2014 EM DASH 
98 02DC SMALL TILDE 
99 2122 TRADE MARK SIGN 
9A 0161 LATIN SMALL LETTER S WITH CARON 
9B 203A SINGLE RIGHT-POINTING ANGLE QUOTATION MARK 
9C 0153 LATIN SMALL LIGATURE OE 
9D  UNDEFINED 
9E 017E LATIN SMALL LETTER Z WITH CARON 
9F 0178 LATIN CAPITAL LETTER Y WITH DIAERESIS 
"""

replacement = {
    # You pick the "closest characters"
    0x80: "e"
    0x82: ","
    ...
}

def cp1252_to_8859_1(str):
    out_str = ''
    for i in str:
        # The only different characters are from 0x80-0x9f
        if (ord(i) >= 0x80) and (ord(i) <= 0x9f):
		# Or throw an exception and ask them not to use 
            # dumb characters
            out_str += replacement[ord(i)]
        else:
            out_str += i

    return out_str






From jacek.generowicz at cern.ch  Fri Feb  8 11:57:53 2002
From: jacek.generowicz at cern.ch (Jacek GENEROWICZ)
Date: 08 Feb 2002 17:57:53 +0100
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1012940770.12866.python-list@python.org>
Message-ID: <tyfr8nv3o8e.fsf@pb-s-513-1-26-6-1.cern.ch>

"Jimmy Retzlaff" <jimmy at retzlaff.com> writes:

> > But about the only hand skill relevant to Python is typing, which I 
> > (and probably most of us) do both-handed...
> 
> Not me. I just had surgery on my dominant right hand last Thursday so
> I'm temporarily non-dexterous. :(
> 
> Since I haven't been able to write much Python since, c.l.py has been a
> welcome diversion!

Maybe a left-hand Dvorak keyboard would be a entertaining/useful
diversion.

http://home1.gte.net/bharrell/kbdtxt.htm
http://www.amputee-coalition.org/inmotion/jul_aug_98/dvorak.html
http://www.pcguide.com/ref/kb/layout/alphaSingle-c.html

etc. etc. . . .


From hankc at nospam.com  Tue Feb 26 18:50:58 2002
From: hankc at nospam.com (HankC)
Date: Tue, 26 Feb 2002 23:50:58 GMT
Subject: Find an Item in a Sorted List
Message-ID: <3c7c1b5b.34558051@news.la.sbcglobal.net>

Greetings:

I'm 2 days new to Python and hope I can get some pointers here...

I have a sorted, delimited text file such as:

Arty,1000
Bobby,2000
Charlie,3000

Knowing a name, I want to return the entire line from the file.  My
approach/problems (any comments appreciated):

# use the filter method
# filter fxn
def Match(line): return line[:toklen] == tok

# find match fxn
def FindMatch:
	fp = open(fname)
	flist = fp.readlines()
	dataline = filter(Match, eodflist)

My problem is that I don't understand how, if it's even possible, to
set tok and toklen values in the FindMatch fxn and pass them to Match.
Is it possible to pass multiple values to the function used by
filter()?

More general...  does filter() have any performance advantage over
just using a for loop?  The datafile list has 10,000+ items.  My
background is in object pascal where I would read the items into a
list then use a binary search to find the target.  I'm hoping there's
an efficient python method to perform a search easily on a sorted
list.

Suppose I want to make tok and toklen global to the module.  Is this
possible?

TIA!


From starx at pacbell.net  Sat Feb 16 05:35:58 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 10:35:58 GMT
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com> <b16e4ef7.0202130950.67f4004d@posting.google.com> <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4> <52e5ab5f.0202141045.5867bfbb@posting.google.com>
Message-ID: <Xns91B71A6CA36C3RASXnewsDFE1@130.133.1.4>

Michael Robin || Thu 14 Feb 2002 10:45:39a:

> Philip Swartzleonard <starx at pacbell.net> wrote in message
> news:<Xns91B4EC515D335RASXnewsDFE1 at 130.133.1.4>... 
>> Stephen Ferg || Wed 13 Feb 2002 09:50:00a:
>> 
>> This doesn't really have anything to do with this thread, but i'd
>> like to make this comment: =PLEASE= don't use the (apparently)
>> standard directory selcetion dialog in windows if you can avoid it.
>> It can be truly aggrivating when you know exactly how to type your
>> way there, but -have- to use the box... and on a reasonably crowded
>> file system, this means 'open node, scroll down, open node, scroll
>> down' repeat-until-insane. 
> 
> You can type the path into the FileOpen dialog at any time, with or
> without mousing first. I would say the opposite, please don't invent
> your own file open dialog unless you have to - you'll lose multiple
> views, property access, drag & drop, filename completion and standard
> keyboard shortcuts - not to mention those who need to hook into
> accessibility features.
> 

Ranting and horrible spelling too... sigh... anyway, i guess we're talking 
about different dialog boxes. The kind i mean have -nothing- but a 
directory tree (explorer's-left-side type), no drag-n-drop, editing, 
properties, or anything. For an example, see what you get from winamp when 
you add or play a directory (from a command specifically named that). This 
isn't the apparently windows-standard one, as it has an addional checkbox, 
but otherwise it's the exact same idea. 

(It's deffenatly not a FileOpen box either, as you could do nothing of the 
sort :)

Also... what's this 'filename completion' you speak of? Are you talking 
about windows? (Though i s'pouse it may have been added past 98, i've 
never known nor can find a key that does this...)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From jonathan at onegoodidea.com  Fri Feb  1 11:32:28 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 01 Feb 2002 16:32:28 +0000
Subject: Naming higher order functions
References: <kuelk63747.fsf@lasipalatsi.fi> <a3bpjt$kle$01$1@news.t-online.com> <3c59901d_2@corp.newsgroups.com>
Message-ID: <B880739C.40E7%jonathan@onegoodidea.com>

On 31/1/2002 18:56, in article 3c59901d_2 at corp.newsgroups.com, "Joshua
Muskovitz" <joshm at taconic.net> wrote:

>> "Erno Kuusela" <erno-news at erno.iki.fi> wrote in message
>> news:kuelk63747.fsf at lasipalatsi.fi...
>>> what would be a good name for the following function?
>>> 
>>> def name_me(n):
>>>     def g(x):
>>>         return x[n]
>>>     return g
> 
> "RefactorMe"


def curry( f, *xs ):
    return lambda *ys: f( *xs+ys )

def get_nth_item( n, x ):
    return x[n]

get_5th_item = curry( get_nth_item, 5 )


(for a suitable definition of "nth" where 0th comes first...)

If you're not an FP-head then you might prefer (also more complete):


def partial_apply( f, args, keys={} ):
    if not callable(f):
        raise TypeError( "'%s' object is not callable" % type(f).__name__ )
    targs = tuple(args)
    tkeys = dict(keys)
    def tempf( *nargs, **nkeys ):
        tkeys.update( nkeys )
        return f( *targs + nargs, **tkeys )
    return tempf

get_5th_item = partial_apply( get_nth_item, (5,) )


Jonathan



From magnus at thinkware.se  Thu Feb 21 19:14:19 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 01:14:19 +0100
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com> <258fd9b8.0202201733.3888e711@posting.google.com> <mailman.1014287322.17283.python-list@python.org> <3C74E31A.1000203@thinkware.se> <K0ed8.53729$n73.986413@atlpnn01.usenetserver.com>
Message-ID: <3C758D5B.8060406@thinkware.se>

Steve Holden wrote:

> This is true to an extent. But certainly Marc Andre's modules do compile on
> cygwin - licensing conditions mean they can't be distributed in pre-compiled
> binary installable form. And MySQLdb also compiles. I'd like to see every
> Windows-capable module run under both: it turns out you don't usually have
> much to do to make the same extension install under Windows and cygwin.


And an advantage with cygwin is that it comes with a C compiler
etc etc. Not something every Windows Python programmer has this.
That certainly makes it more viable to distribute source.

But still, it is something of "a state in the state", and I would
rather see that I didn't have to require that my customers had to
install cygwin to use my products. It's bad enough if they have
to get Python... :-) (I guess I'd really prefer if the next version
of Windows was nothing more than a Window manager and some emulation
layers on top of linux! :-)

Personally, I'd feel very naked without grep, wc, find, gcc etc, but
I do use the windows installers and normal windows versions of Python
when I'm running windows. Actually, since someone in the wxPython list
showed how to enable auto-completion with [tab] in Windows 2000 cmd.exe
I hardly even use bash in windows...

Of course, it would be great if the windows installers for Python
modules said "I found ActivePython 2.1.1 and cygwin Python 2.1.2.
Do you want to install for ActivePython, cygwin, or both?" That is,
until Windows becomes a linux distribution. ;-)



From magnus at thinkware.se  Fri Feb 22 13:03:52 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 19:03:52 +0100
Subject: Indexes and Slices.
References: <mailman.1014391951.30409.python-list@python.org>
Message-ID: <3C768808.5090407@thinkware.se>

>>>>st[12,2] = "hola"
>>>>
> setitem: ((12, 2), 'hola')
> 
> ## THIS was news.
> ## Commas in indexes convert the index into a tuple.
> ## So I can use more complex behaviours, like multi-
> ## dimensional indexing.


No, this is not news. It worked in 2.1 as well at
least, and I frankly don't understand what the
excitement is about.

st[12,2] is short for st[(12,2)]. You've always been
allowed to use tuples of immutable objects as keys for
dicts. Or am I missing something here?

Unless the documentation explicitly says that the
parenthesis can be skipped in this context, I would
be a bit cautious to use that feature. Previously (up
to 1.5.2?) you could (but should not) skip the parenthesis
when you added a tuple with list.append() and in
socket.connect() for instance, but that bug was removed.

Allowing bare tuples in function calls means that you can
get confused about whether the the function takes one tuple
or several parameters. Although inside [] there is really
no risk for confusion. So maybe [1,2,23] won't go away...



From wesc at deirdre.org  Tue Feb 19 13:44:58 2002
From: wesc at deirdre.org (Wesley J. Chun)
Date: 19 Feb 2002 10:44:58 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <mailman.1013843905.25550.python-list@python.org> <5e8bd451.0202170443.7106633a@posting.google.com>
Message-ID: <fcd9edc1.0202191044.4f56a91c@posting.google.com>

kp87 at lycos.com (kevin parks) wrote in message news:<5e8bd451.0202170443.7106633a at posting.google.com>...
> The Wesley Chun book on Python "Core Python" programming has a bit on this.


thx 4 the plug kevin!  :-) let me elaborate a little more tho.

the single chapter on Python MT programming was the 1st published
anything on Python threads, and since the main thrust of the book
is *core* python, i didn't spend adequate time on discussing
threading and concurrency issues.

rather, it is a gentle introduction for either: (1) someone who has
little or no threading experience, wanting a little background and
an intro on how do to it in Python, or (2) someone who already knows
Python and some threading, can skip the intro part and see little
snippets of code that do MT in Python so that they can quickly cut-n-
paste or otherwise adapt it to their application right away.

in addition to a kind-n-gentle intro to threads, the chapter goes
into the thread module (and why you should avoid it), some of the
useful things found in the threading module -- the one you *should*
use, a some info on the global interpreter lock (or Python Virtual
Machine), and how I/O-bound applications take more advantage of the
threading model than CPU-bound ones.

as far as books on threading go...

the java threads book is good to read just for information purposes.
i believe o'reilly also has a Pthreads book.  i got the most info
on locking in my graduate database class using this textbook:

http://dblp.uni-trier.de/db/books/dbtext/bernstein87.html

hope this helps!!

-wesley

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

"Core Python Programming", Prentice Hall PTR, ? 2001
    http://starship.python.net/crew/wesc/cpp/

Silicon Valley-San Francisco Bay Area Python Users Group (BayPIGgies)
    http://deirdre.org/baypiggies

wesley.j.chun :: wesc at deirdre.org
cyberweb.consulting : henderson, nv : cyberweb_consulting at yahoo.com
http://roadkill.com/~wesc/cyberweb/


From cimarron+google at taylors.org  Wed Feb 27 03:14:33 2002
From: cimarron+google at taylors.org (Cimarron Taylor)
Date: 27 Feb 2002 00:14:33 -0800
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
Message-ID: <29e28c51.0202270014.61dbd1d0@posting.google.com>

I have used the following tricks before with Java code, some of them
should be useable with python. 

1. Use a source-to-source translator to convert all your class, method
   and variable names into gibberish.

2. Introduce special "dispatch" classes to obscure the communications
   paths.  E.g instead of calling foo.bar(), call dispatcher.bar(foo) 
   where dispatcher is a dynamically instanciated object.  Have lots
   of dispatchers and periodically switch them around in meaningless ways.

3. Add lots of decoy code.  Put bar() methods in lots of classes which
   are never used.   Include the gibberish versions of a few previous
   versions of your code with logic deliberately reversed.  Intermix 
   meaningless logic with the code which does the actual work.  Bulk up
   your product with a few dozen distributions of other redistributable
   software.

4. Hide some initialization logic outside your program.  Have your
   program step reload classes.  If your program manages external files 
   (e.g. userids or passwords), have it hide some critical initialization 
   logic there.  Encode some of your program state in records in a database
   which requires a separate initialization and/or installation step.

5. Distribute your program in separate pieces. Have your program partially 
   depend on external libraries.

6. Partition your program's operation into separate phases and/or processes.
   Have one process generate code on the fly to be executed by the other.
   Hide important information in otherwise meaningless status messages
   exchanged between the two.  

Most of these things will make your program more difficult to reverse engineer.
Heaven help you if you ever need to fix a bug in it!

Cim


From hfoffani at yahoo.com  Sun Feb  3 10:31:20 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: 3 Feb 2002 07:31:20 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <20020202134649.468$Gg@news.newsreader.com>
Message-ID: <1cef615e.0202030731.4ceea5dd@posting.google.com>

[me]
> I fully agree that when a manager evaluates techs alternatives
> the decision may be affected by non-tech factors. I don't think
> many readers here would deny such claim and c.l.py is a great
> forum to look for answers on TECH's issues of Python.

[orig. poster]
> Wait just a moment here ... what are you saying here? c.l.py is
> a great place to tooks for tech's issues But we don't consider
> managment issues ...
> Well in the ***real*** world where i live all descission are
> taken by managers !

[And I replied again with almost nonsenses]
...

Being too late to make an effective usenet cancel, I'll try to
make my statements more clear.

In my first post, I was addressing your worries on the fact that
"very often wrong questions are asked here."

I claim that your legitimate concerns aren't specific to Python,
and the way this thread turned out is a little proof of it :-)

Therefore you'll find here much more technical info to help you
in your decision processes that on pure software engineering or
business strategies regardless of the high quality of the
postings.

Of course, my personal desires about c.l.py's content were out
of place and I'd never expected to drive it, much less try to
censor you in any way.

Regards,
-Hernan

Hope it's clear now. If so, I wholeheartly recommend reading
usenet with pizza and beer aside!

[posted and mailed to the O.Poster]


From mlh at vier.idi.ntnu.no  Wed Feb 20 05:15:34 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 20 Feb 2002 10:15:34 +0000 (UTC)
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org>
Message-ID: <slrna76tq6.915.mlh@vier.idi.ntnu.no>

In article <mailman.1014175232.10747.python-list at python.org>, Jason
Orendorff wrote:
>Jason wrote:
>> Here is a puzzle that was floating around comp.lang.lisp, and I was
>> wondering what a Python solution would be.
>
>I solved this in Python in a few minutes, then converted it to C++
>for efficiency and found that it ran slower in C++ (because I used
>map<> and string.  Python's stuff is more efficient.)
>
>Then I wrote my own hashtable and my own string type and managed
>to get the C++ down to about 8x faster than the Python.
>It solves their WORD.LST in <1 second on my PC.

That's pretty quick... Anyway, if you ignore the setup-time (reading
in the file and storing it in a useful way; this took 6 seconds),
the search for a solution took 0.9 seconds on my PC -- implemented in
Python, mind you. (Of course ignoring the setup isn't very realisting
in this case, where there is only one solution -- but if you were to
be able to find an addagram of a user-supplied length, for instance,
that would be realistic.)

>I solved the other puzzles too...  They still didn't hire me.  :)
>
>## Jason Orendorff    http://www.jorendorff.com/
>

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From fperez at pizero.colorado.edu  Tue Feb 12 23:36:03 2002
From: fperez at pizero.colorado.edu (Fernando =?iso-8859-1?Q?P=E9rez?=)
Date: Tue, 12 Feb 2002 21:36:03 -0700
Subject: Major league corruption - anyone seen anything like this?
References: <mailman.1013557302.9749.python-list@python.org>
	        <7xzo2ekv5g.fsf@ruckus.brouhaha.com> <mailman.1013572042.18876.python-list@python.org>
Message-ID: <3C69ED33.CE0100DF@pizero.colorado.edu>

Skip Montanaro wrote:
> 
> I realized that might be an issue on my way home.  This is on my development
> machine.  I (user == skip) do have write access to /usr/local/*.  Sorry
> about the missing info.

Still strange. A couple of questions: 

- what filesystem are you using?
- are all the files which get corrupted files that get explicitly open by
your crashing process?

Just yesterday my Mandrake laptop refused to boot because a locked-up
suspend (apm -s) event crapped on /etc/fstab. Not pretty at all. It took
some creative games to bring it back to life. The culprit had been the
pcmcia process trying to modify /etc/fstab to de-register a flash card,
since the suspend calls pcmcia, the file was open for writing when the
lockup happened. I user reiserfs, which logs filesystem meta-data but not
user data, so that didn't save me. I give you this background in case you
can spot some similarities with your situation.

Cheers,

f


From mhammond at skippinet.com.au  Mon Feb 11 21:12:32 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 12 Feb 2002 02:12:32 GMT
Subject: Compiling Python 2.2 in Microsoft VC7 (.net)
References: <a85e8a65.0202111643.5f5fda89@posting.google.com>
Message-ID: <3C687A51.6000400@skippinet.com.au>

Joe Shochet wrote:
> I'm trying to get Python 2.2 to compile in VC7 (part of Visual Studio
> .net), but I'm running into all sorts of troubles. Before digging into
> the problems, has anybody else already done this?
> 
> The main problems seem be related to missing files (largeint.h,
> traceback.c were the first two I saw). Many more followed these
> errors.
> 

I can reproduce this, and am working towards a checkin that fixes it. 
Please try again in a day or so (and feel free to ping me if you still 
have problems.

FWIW, once I get it reviewed I intend checking in the patch below.  All 
seems to work after that.

Mark.

--- timemodule.c	16 Jan 2002 11:04:06 -0000	2.120
+++ timemodule.c	12 Feb 2002 02:04:04 -0000
@@ -46,17 +46,24 @@
  #define altzone _altzone
  #endif /* MS_WIN16 */
  #endif /* MS_WINDOWS */
  #endif /* !__WATCOMC__ || __QNX__ */

  #if defined(MS_WIN32) && !defined(MS_WIN64) && !defined(__BORLANDC__)
  /* Win32 has better clock replacement
     XXX Win64 does not yet, but might when the platform matures. */
+/* NOTE: VC6 used <largeint.h> for these macros.
+   They seem to have vanished with VC7 :( */
+#if _MSC_VER >= 1300 /* MS dropped the largeint.h helpers in VC7 :( */
+# 
define LargeIntegerEqualToZero(X) ( !((X).LowPart | (X).HighPart))
+#else
  #include <largeint.h>
+#endif
+
  #undef HAVE_CLOCK /* We have our own version down below */
  #endif /* MS_WIN32 && !MS_WIN64 */

  #if defined(PYCC_VACPP)
  #include <sys/time.h>
  #endif

  #ifdef __BEOS__
@@ -161,18 +168,24 @@
  		    LargeIntegerEqualToZero(divisor)) {
  	 
		/* Unlikely to happen -
  	 
		   this works on all intel machines at least!
  	 
		   Revert to clock() */
  	 
	return PyFloat_FromDouble(clock());
  		}
  	}
  	QueryPerformanceCounter(&now);
- 
diff = LargeIntegerSubtract(now, ctrStart);
- 
diff = LargeIntegerDivide(diff, divisor, &rem);
+#if _MSC_VER >= 1300 /* MS dropped the largeint.h helpers in VC7 :( */
+ 
diff.QuadPart = now.QuadPart - ctrStart.QuadPart;
+ 
rem.QuadPart = diff.QuadPart % divisor.QuadPart;
+ 
diff.QuadPart /= divisor.QuadPart;
+#else
+ 	diff = LargeIntegerSubtract(now, ctrStart);
+ 	diff = LargeIntegerDivide(diff, divisor, &rem);
+#endif
  	/* XXX - we assume both divide results fit in 32 bits.  This is
  	   true on Intels.  First person who can afford a machine that
  	   doesnt deserves to fix it :-)
  	*/
  	return PyFloat_FromDouble((double)diff.LowPart +
  		              ((double)rem.LowPart / (double)divisor.LowPart));
  }




and:
Index: _hotshot.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Modules/_hotshot.c,v
retrieving revision 1.13
diff -u -8 -r1.13 _hotshot.c
--- _hotshot.c	8 Feb 2002 21:27:50 -0000	1.13
+++ _hotshot.c	12 Feb 2002 02:11:42 -0000
@@ -9,17 +9,16 @@
  #include "structmember.h"

  /*
   * Which timer to use should be made more configurable, but that 
should not
   * be difficult.  This will do for now.
   */
  #ifdef MS_WIN32
  #include <windows.h>
-#include <largeint.h>
  #include <direct.h>    /* for getcwd() */
  typedef __int64 hs_time;
  #define GETTIMEOFDAY(P_HS_TIME) \
  	{ LARGE_INTEGER _temp; \
  	  QueryPerformanceCounter(&_temp); \
  	  *(P_HS_TIME) = _temp.QuadPart; }
  	





From precision at devrandom.net  Fri Feb  8 17:59:12 2002
From: precision at devrandom.net (Uriah Welcome)
Date: 8 Feb 2002 14:59:12 -0800
Subject: SourceForge Python foundry ????
References: <3c6301ac.188163695@news.laplaza.org> <a40hh0$1bdtaj$1@ID-7529.news.dfncis.de> <65c17385.0202081114.14d2a605@posting.google.com> <a41971$1b8tsa$1@ID-7529.news.dfncis.de>
Message-ID: <65c17385.0202081459.4f856aab@posting.google.com>

"Karl M. Syring" <syring at email.com> wrote in message news:<a41971$1b8tsa$1 at ID-7529.news.dfncis.de>...
> "Uriah Welcome" <precision at devrandom.net> schrieb im Newsbeitrag
> news:65c17385.0202081114.14d2a605 at posting.google.com...
> > "Karl M. Syring" <syring at email.com> wrote in message
>  news:<a40hh0$1bdtaj$1 at ID-7529.news.dfncis.de>...
> > > "Mats Wichmann" <mats at laplaza.org> schrieb
> > > > What's up with the Python foundry on SourceForge? It's not listed in
> > > > any of their lists of foundries, although if you plug in "python" in
> > > > the standard path they use for a foundry it comes up fine...
> > >
> > > Perhaps, that is because SF still has a grubby v1.5.2 installed?
> >
> > As soon as debian 'stable' includes a newer version we'll be happy to
> include it.
> 
> Google stalking at work?
> Nonetheless, this is a good message, at least we wont get a prompt saying
> something like "AOL-Linux -- you have got snails!"


Lol.. I always read this list.. I just rarely post. ;)

-U


From pobrien at orbtech.com  Tue Feb 26 14:31:09 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Tue, 26 Feb 2002 19:31:09 GMT
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com> <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com> <258fd9b8.0202200150.1b4e53de@posting.google.com> <k7me8.42891$9_1.1798161575@newssvr14.news.prodigy.com>
Message-ID: <1oRe8.18032$0C1.1579558@newsread1.prod.itd.earthlink.net>

Well, this particular problem still exists. It has nothing to do with
PythonWin or PyCrust. It has to do with the trick used by help, license,
etc. Which is to make an object callable and then hook up that call to the
__repr__ method. This works nicely in the shell in that you can then call
the object without the trailing parens:

>>> help

instead of

>>> help()

The problem is the object is called whenever a representation of itself is
requested, which happens during vars() and other similar functions in the
shell. What you are seeing are the side effects of this trick. If the called
object does something unusual (like exiting the application <wink>) you can
get unexpected (but not entirely unpredictable) behavior.

--
Patrick K. O'Brien
Orbtech


"Keith Farmer" <kfarmer at thuban.org> wrote in message
news:k7me8.42891$9_1.1798161575 at newssvr14.news.prodigy.com...
> Not ActiveState -- regular Win32 install:
>
> Microsoft Windows 2000 [Version 5.00.2195]
> (C) Copyright 1985-2000 Microsoft Corp.
>
> C:\>python
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
>
>
> Here's one symptom I noticed (in the copyright entry .. also in the
license,
> you'll see):
>
>  <class exceptions.AssertionError at 0x00881490>, 'TypeError': <class
> exceptions
> .TypeError at 0x0087E868>, 'copyright': Copyright (c) 2001 Python Software
> Found
> ation.
> All Rights Reserved.
>
> Copyright (c) 2000 BeOpen.com.
> All Rights Reserved.
>
> Copyright (c) 1995-2001 Corporation for National Research Initiatives.
> All Rights Reserved.
>
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved., 'license': Type license() to see the full license
> text, 'c
> lassmethod': <type 'classmethod'>, 'KeyboardInterrupt': <class
> exceptions.Keyboa
> rdInterrupt at 0x0087ED50>, 'UserWarning': <class exceptions.UserWarning
at
> 0x00
>
> What I ended up noticing was that the way the dict gets handled seems to
> execute the value under certain conditions which I didn't bother to
> investigate further.  In PyCrust (before Patrick made his corrections,
which
> I haven't tried out), I was able to cause my program to exit by inserting
a
> reference to the exit function into the namespace and then print it out.
>
> However, the excerpt above doesn't come from PyCrust -- it was from the
> command line interpreter.
>
> I'm running Win2k SP2.
>
> ----------
> Keith J. Farmer
> kfarmer at thuban.org
> http://www.thuban.org





From syver-en+usenet at online.no  Sun Feb 17 19:55:06 2002
From: syver-en+usenet at online.no (Syver Enstad)
Date: Mon, 18 Feb 2002 00:55:06 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <861yflwo4s.fsf@apo.hmotakef.int> <5d283efe.0202162142.53e3b402@posting.google.com> <roy-05958C.07303117022002@reader2.panix.com>
Message-ID: <ubsenehjm.fsf@online.no>

mlh at vier.idi.ntnu.no (Magnus Lie Hetland) writes:
> 
> Jorg: Du kan l?ne denne av meg :)
> 

Begynner det ? bli s? mange norskinger p? comp.lang.python at vi kan
begynne ? skrive postingene v?re p? norsk ;-)

-- 

Vennlig hilsen 

Syver Enstad


From brueckd at tbye.com  Thu Feb  7 10:14:02 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 7 Feb 2002 07:14:02 -0800 (PST)
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users"
 ?
In-Reply-To: <Xns91AE86CF4427Cmichaelrcpcouk@194.238.50.13>
Message-ID: <Pine.LNX.4.33.0202070703190.4847-100000@ns0.tbye.com>

On Thu, 7 Feb 2002, Michael Abbott wrote:

> Michael Hudson <mwh at python.net> wrote in news:uk7tpa574.fsf at python.net:
> > But not over Java, surely?  I guess a JIT or native compilation will
> > help, but...
>
> I think so.  Java byte codes are pretty low level, and don't have to do
> very much, and of course JIT or native compilation can be *very* effective.
> Python byte codes are, on the other hand, much more powerful, and have to
> do quite a lot of run-time work.
[snip]

> My factor of a 100 is a guess; am I wrong?  I believe it's fair.

Wouldn't that mean that simple Jython programs should use 100x less CPU
than CPython ones?

I guess it depends on what you're doing. I spend most of my Python time
writing networking clients/servers, small internally-used utilities, text
processing/parsing tools, and playing with Pygame. As far as CPU usage
goes:

- networking stuff - tends to use 1.5X to 2X more CPU than C, and I've
never seen any Java servers come close to the same performance levels.

- utilities and text processing tools - not entirely CPU bound, but they
usually finish by the time the Java VM has loaded, so who knows (I don't
usually deal with huge volumes of text)

- PyGame - hey, it's built on top of SDL. ;-)

I think the last one is important: maybe JIT-compiled stuff *is* 100x less
CPU-intensive than *pure* Python, but how many programs are pure Python?
A huge amount of time is spent in C libraries (I/O, extension modules).

-Dave




From martin at v.loewis.de  Tue Feb 12 04:20:55 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 12 Feb 2002 10:20:55 +0100
Subject: how to compare two strings?
References: <7xofivazj1.fsf@ruckus.brouhaha.com>
Message-ID: <m3aduf839k.fsf@mira.informatik.hu-berlin.de>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> Is there any built-in function in Python for comparing two strings
> alphabetically?

Can you specify "alphabetically" more precisely? Do you mean
"lexicographically, by character ordinal value" (which is what your
scmp code suggests)? If so, the builtin cmp() function will do;
it compares strings just fine.

Or do you mean "according to the locale's conventions"? That would be
locale.strcoll. Notice that for sorting according to the locale's
conventions, it is more efficient to invoke strxfrm for each string,
and then sort by the transformed results.

Regards,
Martin


From QnickQm at alum.mit.edu  Tue Feb 19 10:44:09 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Tue, 19 Feb 2002 15:44:09 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <slrna74ssf.21s.QnickQm@localhost.localdomain>

In article <24c39b2c.0202181531.187fad4c at posting.google.com>, 
        Alexander Jerusalem wrote:
> Has anyone heard of plans to add static type checking to Python?

Someone else has mentioned an old proposal of Guido's for adding
optional type declarations in order to help the compiler catch errors.

Making a real static type system is harder than you'd think.
Consider this function:

def fn(lst,item):
   for i in lst:
     if i == item:
        return repr(i)

What is the type of fn?  You could call it:

     (list[int], int) -> str

But then you'd lose the polymorphism of the function.  To a first
approximation, you might say:

     (list[A], A) -> str

But keep in mind that not any A will work...

     (list[A], A) -> str  s.t. A has __eq__(A) and __repr__() -> str

And that x==y is sometimes okay even if x.__class__ != y.__class__...

     (list[A], B) -> str  s.t. A has __eq__(B) and __repr__() -> str

But there's nothing that says you can't use a tuple or a string or a UserList 
instead of a list...


     (sequence[A], B) -> str  s.t. A has __eq__(B) and __repr__() -> str

And you aren't using _all_ of the sequence protocol: just iteration...

     (A, B) -> str   s.t. A has __iter()__ -> C 
                     s.t. C has next() -> D (or raises StopIteration)
                     s.t. D has __eq__(B) and repr() -> str

IMNSHO, some kind type inference is the only tractable answer.

And-what-if-repr-doesn't-return-a-string?-ly yrs,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From gh_pythonlist at gmx.de  Mon Feb 11 06:20:42 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 11 Feb 2002 12:20:42 +0100
Subject: Long int to int
In-Reply-To: <dec413074b.tim@worthy.demon.co.uk>
References: <dec413074b.tim@worthy.demon.co.uk>
Message-ID: <20020211112041.GA4618@lilith.hqd-internal>

Le 11/02/02 ? 10:40, Tim Howarth ?crivit:
> I need to convert some long int values to normal ints.
> (4 byte words) [...]

>  |im    ---- ARM Powered ----

Assuming 32-bit hardware, like my Intel x86 and StrongARM processors :-)

0x81234567L is greater than sys.maxint, so you can't convert it to an
int directly. What *should* be possible, is to convert it to an int that
has the same bit pattern as an unsigned int (an unsigned int doesn't
exist in Python, normally).

Could it be that you need this to interface some external library or
file format? Maybe the struct module can help you?

For creating an equivalent bit representation, what about this:

    print struct.unpack("i", struct.pack("I", 0x81234567L))[0]

I currently don't remember enough, CS, but it should be easy to do
without the struct module, too.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.0 ?C      Wind: 6.1 m/s



From hartmut at goebel.noris.de  Tue Feb 26 14:50:32 2002
From: hartmut at goebel.noris.de (Hartmut Goebel)
Date: Tue, 26 Feb 2002 20:50:32 +0100
Subject: ANN: decompyle 2.2 beta 1
Message-ID: <mailman.1014902103.8034.clpa-moderators@python.org>

Announcing:

             decompyle -- A Python byte-code decompiler

                      version 2.2 beta 1


'decompyle' converts Python byte-code back into equivalent Python
source. It accepts byte-code from any Python version starting with 1.5
up to 2.2.

The generated source is very readable: docstrings, lists, tuples and
hashes get pretty-printed.

'decompyle' may also verify the equivalence of the generated source by
by compiling it and comparing both byte-codes.


New Features 
------------

Since Release 0.6.0:

  * Now decompyles byte-code from Python versions 1.5 up to 2.2

  * Now requires Python 2.2 but is able to decompyle older byte-coe,
    too. Prior version of decompyle had to be run with the Python
    version which generated the byte-code.

  * pretty-prints docstrings, hashes, lists and tuples

  * decompyle is now a script and a package

  * Now adds an emacs mode-hint and tab-width to the output files

  * enhanced test suite: more test patterns, .pyc/.pyo included

  * avoids unnecessary 'global' statements

  * many internal changes and code overhouls

  Please refere to the file 'CHANGES' for a list of changes in this
  release.


Where to get it?
----------------

The source archives and instructions for isntalling the package can be
found at:

             http://www.goebel-consult.de/decompyle/


Simple Help Needed!
-------------------

Please help testing 'decompyle'! 

The EXTENDED_ARG token is untested (this is a new token for Python 2.0
which is used only if many items exist within a code object). If you
have byte-code which does or may include this token, please try to
decompyle your application.

It's easy: the script 'test_pythonlib' which is included in the source
           distribution does the job for you. Just read the simple
           instruction there.


Requirements
------------

'decompyle' requires Python 2.2 or later.


Known Bugs/Restrictions
-----------------------

* The EXTENDED_ARG token is untested (this is a new token for Python
  2.0 which is used only if many items exist within a code object).

* Byte-code generated by Python 2.2 differs byte-code generated by a
  prior version even for the same source. This is due the intruduction
  of iterators. Currently 'Decompyle' fails verifying the source if
  the byte-code was generated by an older version of Python.

* Verifying decompyled source with optizimzed byte code (.pyo) when
  running without optimizations (option '-O' not given) fails in most
  cases. Same is true for vis-a-versa. This is due to the fact that
  Python generated different bytecode depending on option '-O'.

+++hartmut
-- 
| hartmut Goebel       | hartmut at goebel.noris.de  // |
| Software Manufactur  |                        \X/  |



From mwh at python.net  Thu Feb  7 05:44:26 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 10:44:26 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de> <u8za6pskh.fsf@python.net> <3c61abe7$1@si-nic.hrz.uni-siegen.de>
Message-ID: <uu1sta7s2.fsf@python.net>

Simon Budig <simon.budig at unix-ag.org> writes:

> In article <u8za6pskh.fsf at python.net>, Michael Hudson wrote:
> > Simon Budig <simon at ping-pong.budig> writes:
> >> In article <uheowosxs.fsf at python.net>, Michael Hudson wrote:
> >> > Well, it can't go into 2.2.  If you mean 2.2.1, that's a little
> >> > greyer; there might be people out there who rely on the current
> >> > behaviour.
> >> 
> >> Hmm. Maybe it is possible to export this variable as an optional
> >> parameter to readline.set_completer with the default on the
> >> current behaviour?
> > 
> > But then that's a new feature...
> > 
> >> I think it is really annoying, that readline in the current
> >> python-module implementation insists on completing filenames when
> >> the completer returns no matches.
> > 
> > Sorry.  But I don't think you can claim it's a pure bug, and so it's
> > not 2.2.1 material.
> 
> Actually I think, that it is a bug, because it makes it impossible
> to write a completer function, that avoids the filename completion.

However, there may be people relying on current behaviour, and the
motto of x.y.z releases is "nothing breaks".

Also, to get it into 2.2.1 you have to convince the person putting
patches into the 22-maint tree, and that currently seems to be me, and
I'm not checking this one in.

> Imagine the rlcompleter: It is an convenient way to have completion
> on the Python commandline. But if you start typing a name that is
> not (yet) in the global dict and type <TAB> to verify this fact,
> readline decides to complete with a filename, which is absolutely
> meaningless in this context - AND YOU CANNOT SWITCH THIS OFF!

It can be useful too; I have actually done

>>> f = open("blabla

and then used tab-completion to finish it off.

> However, I am not familiar with the python guidelines/procedures for
> adding new features and certainly dont think that this is important
> enough to break these guidelines.

I should have mentioned this before:

    http://python.sourceforge.net/peps/pep-0006.html

> I am a little bit astonished, that it seems possible to add this to
> 2.1.x, but not to 2.2.x. Sounds weird to me. 

Huh?  Who said anything about 2.1.x?  (If it was me, that was a typo
:).  It'll go into *2.3*.

> For Gimp development we don't fix some misbehaviours in old
> versions, but are happy to fix them in new versions...
> 
> Oh, well. Maybe I am missing something obvious here...   :-)

Well, that 2.1.x != 2.3, perhaps...

Cheers,
M.

-- 
  incidentally, asking why things are "left out of the language" is
  a good sign that the asker is fairly clueless.
                                        -- Erik Naggum, comp.lang.lisp


From andda715 at student.liu.se  Tue Feb  5 12:02:49 2002
From: andda715 at student.liu.se (Anders Dahlberg)
Date: Tue, 5 Feb 2002 18:02:49 +0100
Subject: CPython vs. Jython for Web applications?
References: <c775c8db.0202041126.5177e606@posting.google.com> <3c5f951f$1_2@news.uni-bielefeld.de>
Message-ID: <a3p3dd$ncn$1@news.island.liu.se>

"Mario Wehbrink" <mario at bio58.uni-bielefeld.de> skrev i meddelandet
news:3c5f951f$1_2 at news.uni-bielefeld.de...
> Am  4 Feb 2002 11:26:27 -0800  schrieb/wrote/a ?crit
> John E. Barham  < jbarham at octopuswireless.com > in comp.lang.python :
>
> > I appreciate that Jython's performance is dependent on the JVM, so
> > advice would be appreciated there as well, especially regarding the
> > performance of JVMs for the free Unices.
>
> I read some days ago in german linux-magazin that the performance of
> Jython is about 80% of CPython when using Microsofts VM. When using one
> of the Java ports of linux (blackdown or IBM) the performance is only
> 20 or 30% of CPython, but i have not done any benchmarks to support that
> statement.

As jdk1.4 is faster (a lot IMO) than microsofts "almost java" vm wouldn't
that make jython faster or as fast as cpython on windows? And I really doubt
the figures on linux... try a recent java version from IBM or SUN - not
some -95 hack...

> Bye
>
> Mario

/Anders
> --
> Mario Wehbrink
> Abteilung 4 Kybernetik
> University of Bielefeld




From jphammock at email.msn.com  Sat Feb 16 10:39:30 2002
From: jphammock at email.msn.com (jphammock)
Date: Sat, 16 Feb 2002 10:39:30 -0500
Subject: How do I view Pages on Palm
Message-ID: <e7fZT$vtBHA.304@cpimsnntpa03>

downloaded python but dont know how to use it to view web pages on my palm
any help.

            Thanks




From mwh at python.net  Fri Feb 22 05:52:43 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 22 Feb 2002 10:52:43 GMT
Subject: Embedding: How to set globals for PyObject_Call
References: <mailman.1014234635.18641.python-list@python.org> <79b2b0bb.0202210841.457c0b81@posting.google.com> <uit8qlqdm.fsf@python.net> <79b2b0bb.0202211726.7e8ab7f1@posting.google.com>
Message-ID: <u664plra0.fsf@python.net>

xscottgjunk at yahoo.com (Scott Gilbert) writes:

> Michael Hudson <mwh at python.net> wrote:
> > xscottgjunk at yahoo.com (Scott Gilbert) writes:
> > 
> > >   // This one breaks
> > >   Var g = VarEval("globals");
> > >   g(); // This causes a segfault
> > 
> > Do we get to see the backtrace?
> > 
> 
> - For "globals", it segfaults in builtin_globals(...) where it tries
> to Py_INCREF the null result from PyEval_GetGlobals().  The dbx stack
> trace is below if you really want it.  :-)
> 
> - For "locals", it segfaults when trying to Py_INCREF the null return
> result from PyEval_GetLocals().  Similar stack trace as below, but
> this is on line 1104 of bltinmodule.c
> 
> - For "vars", it sets up a Python error of "no locals!?" on line 1992
> of bltinmodule.c
> 
> - Finally for "dir", PyObject_Dir [object.c, 2085] is called from
> builtin_dir [bltinmodule.c, 1992] and returns a null, but doesn't
> setup the Python error, so up the call stack PyObject_Call sets up a
> system error of "NULL result without error in PyObject_Call"
> [abstract.c, 2034].

Oh, I see what's going on.  Another question: what would you expect
locals() to return?

> Many other functions work without a hitch.  (For instance, "list" is a
> builtin that can take no args, and I correctly get a new list object
> when I do my equivalent of apply() on that.)

Yeah, but they don't access the frame of the calling (presumed Python)
function.

> Well, I hope this is sufficient info if you view this as enough of a
> problem to fix it.  Thanks for your interest.

It's probable that Python shouldn't core in these circumstances, at
least.

Cheers,
M.

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


From R.Brodie at rl.ac.uk  Thu Feb 21 07:09:25 2002
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Thu, 21 Feb 2002 12:09:25 -0000
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu><gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <mailman.1013991930.24000.python-list@python.org>
Message-ID: <a52o1d$kj6@newton.cc.rl.ac.uk>

"Fran?ois Pinard" <pinard at iro.umontreal.ca> wrote in message
news:mailman.1013991930.24000.python-list at python.org...

> > > [...] a requirement for high precision floats is usually a sign that
> > > the requirer does not understand his problem space.
>
> > Incredibly untrue.
>
> Science people often need more precision.  DOUBLE existed in early FORTRAN.
> C has `float' and `double' as well.  It would be rather surprising if all
> involved people were that dumb! :-)

Well probably not all. However many more scientists use mathematical tools
than have a deep understanding of them. To take the example of fluid
dynamics: some problems are inherently chaotic (e.g. famously, weather
prediction). You will get different results if you ran the simulation at
different precisions. That doesn't mean that the higher precision results
are any more meaningful.

There is a set of problems where single precision isn't good enough
and double precision is. However, that surely is smaller than the set of
problems which are so ill-conditioned that double precision is insufficient.
I strongly suspect the OP was correct in implying most people can't tell
the difference: it's a corollary of Kaa's law.









From tjreedy at home.com  Wed Feb 20 00:48:06 2002
From: tjreedy at home.com (Terry Reedy)
Date: Wed, 20 Feb 2002 05:48:06 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com>
Message-ID: <qMGc8.55539$X64.19862109@news1.rdc2.pa.home.com>

"Alexander Jerusalem" <ajeru at vknn.org> wrote in message
> Well, my hope is that open source projects are more flexible than
> bureaucracies like Sun.

They certainly are.  You and other supporters of static typing are
quite free to design and implement a static typing add-on for Python
in a way you could not do with Java.




From shalehperry at attbi.com  Fri Feb 15 22:30:42 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Fri, 15 Feb 2002 19:30:42 -0800 (PST)
Subject: line break confusion
In-Reply-To: <3C6DC606.5482C478@nthwave.net>
Message-ID: <XFMail.20020215193042.shalehperry@attbi.com>

> 
> In the string module, whitespace = ' \t\n\r\v\f' (5 characters here).
> Yet, when I run this:
>>>> import string
>>>> for c in string.whitespace:
>>>>     print '.'+ c + '.'
> 
> I get 7 results.
> What's going on here?
> 

Windows: \r\n
Linux (UNIX): \n
Mac: \r

Each OS's libs transform the sequence '\n' to whatever the right thing is for
them.



From jhorneman at pobox.com  Thu Feb 28 05:28:07 2002
From: jhorneman at pobox.com (Jurie Horneman)
Date: 28 Feb 2002 02:28:07 -0800
Subject: Trouble writing custom text codec
Message-ID: <9ac02e81.0202280228.5161a43c@posting.google.com>

Why does my custom text codec, which I have based on an existing,
working codec as recommended by the documentation, work with
codecs.open() but not with MyString.decode(), where as the built-in
codecs do?

I've stopped simply calling the decode() method; I'm down to calling
getdecoder() and executing that now. I keep getting "TypeError:
unbound method decode() must be called with Codec instance as first
argument (got class instance instead)" and I have no clue how to get
rid of these. Actually creating instances of the right class and
passing that to the decoding function as the first argument do not
help at all - and how come a) the built-in codecs do not have this
problem and b) codecs.open() works with my codec?

The more I look into it, the more incomprehensible I find the whole
codec system.

Jurie Horneman (very, very frustrated)


From emile at fenx.com  Fri Feb  1 12:27:39 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 09:27:39 -0800
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012578783.8284.python-list@python.org>
Message-ID: <a3ejgh$17vri9$1@ID-11957.news.dfncis.de>

<brueckd at tbye.com> wrote in message
news:mailman.1012578783.8284.python-list at python.org...
> That's why I asked whether or not the JIT was enabled in the tests
and, if
> so, if you could re-run them without it. Maybe the low number of
replies
> is because it's not suprising that a semi-compiled-to-native program
is
> faster?
>

Or that a lot of us already know that there are faster languages, but
that python is fast enough anyway?

--

Emile van Sebille
emile at fenx.com

---------



From tim.one at home.com  Thu Feb 14 20:34:34 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 14 Feb 2002 20:34:34 -0500
Subject: Python License Issues
In-Reply-To: <31575A892FF6D1118F5800600846864DCBCC4E@intrepid>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEFMNMAA.tim.one@home.com>

[Simon Brunning]
> More importantly, good manners. Let's say I'm wrapping up a Python
> program with py2exe, and I'm going to distribute it. Which files would I
> be expected to include as a matter of courtesy to the Python developers
> (and the developers of any additional modules that I use)? Just the
> licence files, or is there anything else?

Each component's license should list its requirements, and if you do just
that much nobody can gripe.

Anything beyond that depends on the specific developers.  I'm generally
mildly annoyed when somebody gives me credit for my code, because I don't
want to endure Even More Email about it -- but then I'm older and grouchier
than anyone else you've ever met <wink>.

If you spontaneously feel compelled to add "Python Rulz!" to your own blurb,
that would be cool.  If you have to fake making it look spontaneous, it
starts to get questionable.

but-then-good-manners-often-are-ly y'rs  - tim




From jhenry19 at ford.com  Thu Feb 21 11:27:22 2002
From: jhenry19 at ford.com (Jeff Henry)
Date: Thu, 21 Feb 2002 11:27:22 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com>
Message-ID: <3C751FEA.24CEFBBC@ford.com>

Roy Smith wrote:
> 
> Roy Smith <roy at panix.com> wrote:
> > PS -- In the lists of false things in python that have been mentioned on
> > this thread, I don't think I've seen anybody mention {}, the empty
> > dictionary.
> 
> Along those same lines, I half-expected the empty function to be false,
> i.e.:
> 
> def foo():
>    pass
> 
> if (foo):

Oops! I think you meant to say -

if (foo()):

>    print "it's true"
> else:
>    print "it's false"
> 
> should have printed "it's false".  Alas, it says it's true.  Seems kind of
> irregular to me, not that I would imagine there's much practical
> consequence one way or the other.


From sholden at holdenweb.com  Thu Feb  7 08:25:22 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 08:25:22 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <slrna5v9io.oqf.scarblac@pino.selwerd.nl>
Message-ID: <Wev88.94094$KM2.3399200@atlpnn01.usenetserver.com>

"Remco Gerlich" <scarblac at pino.selwerd.nl> wrote in message
news:slrna5v9io.oqf.scarblac at pino.selwerd.nl...
> Paul Rubin <phr-n2002a at nightsong.com> wrote in comp.lang.python:
> > "Tim Peters" <tim.one at home.com> writes:
> > > > That's the problem--if I don't import future division, my code will
> > > > break in the future, because the // operator isn't supported in the
> > > > present.
> > >
> > > Not so:  introducing "//" was 100% backward-compatible, so isn't
controlled
> > > by __future__.  "//" works out of the box in 2.2, no __future__
required.
> >
> > If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> > it isn't backwards compatible.
>
> It's a new feature. It's impossible that new features work in old
versions,
> basically by definition. Your idea of backwards compatibility is strange.
>
> However, 1.5.2 programs that use "/" will still work in 2.2 (apart from
> other possible incompatible changes since then).
>
> > > I don't know where you're looking, but 2.2 final was released.
> >
> > I know this, but 2.1.1 was released more recently than 2.2, so the most
> > recent release was in the 2.1 series.
>
> I'm going to release a new version of 0.9 then, tomorrow. Then you'll have
> to claim that's the most recent.
>
> > If the official Python maintainers think they might release a 1.5.3
> > someday, then it's fair to say that the 1.5 series is still current.
> > If they're unlikely to ever release another 1.2.x or 0.9.x, then
> > those versions are not current, even if I release a 1.2.x myself.
>
> The official Python maintainers are working on the next version after 2.2,
> probably 3.0. Other people have released a few bugfix
> releases for people who really need to continue running old versions.
>
As far as I know (and I'm only writing this since the developers are
probably mostly at the Conference Developers Day, whereas  am back home) the
next release is 2.3. The PEP list at http://python.sourceforge.net/peps/ has
no 2.3 release schedule PEP yet, but it does list "PEPs under consideration
for 2.3".

Since 3.0 *will* break backward compatibility, there's going to be a long
slow approach to it, with plenty of whistle stops along the way.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From kdahlhaus at yahoo.com  Fri Feb  1 12:57:29 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 1 Feb 2002 09:57:29 -0800
Subject: finding a module
References: <283adf56.0201250621.65232296@posting.google.com> <3C51F04E.82082244@gol.ge>
Message-ID: <283adf56.0202010957.2cd682bd@posting.google.com>

Thanks, I tried it last night but to no avail.  I am beginning to
think they have disabled dynamic loading of user modules or
something....

Giorgi Lekishvili <gleki at gol.ge> wrote in message news:<3C51F04E.82082244 at gol.ge>...
> Kevin Dahlhausen wrote:
> 
> > I'm having a little trouble finding a c extension.  Here's the
> > situation:
> >
> > I have a web host that allows python cgi.  I've also written a wrapper
> > for a library.  The wrapper is a subdirectory with an __init__.py
> > file.  In the same subdirectory is a swig-generated c wrapper (
> > sqlitec.so ) that is imported by one of the files imported by
> > __init__.py.  The main web application sits in the directory above the
> > module directory:
> >
> >   .....\webapp\main.cgi               (imports sqlitedb)
> >   .....\webapp\sqlitedb\__init__.py   (imports wrapper)
> >   .....\webapp\sqlitedb\wrapper.py    (imports sqlitec)
> >   .....\webapp\sqlitedb\sqlitec.so
> >
> > The code on 'main.cgi' contains an import: 'import sqlitedb'.
> > Python finds the sqlitedb python files, but can not find sqlitec.so.
> > They are runing Python 1.6.
> 
> The SWIG generated module MUST be called sqlitemodule.so... Note the
> difference.
> Try this..
> GRTZ,
> Giorgi
> 
> >
> >
> > How can I allow Python to see the loadable module?   It is
> > possible/common for the web providers to disallow import of native
> > code modules?
> >
> > I've tried adding the following to main.cgi:
> >
> >   import sys
> >   sys.path.append(".")
> >   sys.path.append("./sqlitedb")
> >
> >   import sqlitedb
> >
> > but that did did not fix anything.
> >
> > Any ideas?
> >
> > If this works, we would be very close to having thread-safe
> > python-based web sites w/an sql database but without an sql server.
> > (Sqlite is a file-based database similar to gadfly).


From jason at jorendorff.com  Tue Feb 26 19:47:23 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 18:47:23 -0600
Subject: Find an Item in a Sorted List
In-Reply-To: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <HFEKILOLEFEFMKAECNDLGEIPDFAA.jason@jorendorff.com>

> More general...  does filter() have any performance advantage over
> just using a for loop?  The datafile list has 10,000+ items.  My
> background is in object pascal where I would read the items into a
> list then use a binary search to find the target.  I'm hoping there's
> an efficient python method to perform a search easily on a sorted
> list.

import bisect

Also note the startswith() method of strings.

## Jason Orendorff    http://www.jorendorff.com/







From sjmachin at lexicon.net  Mon Feb 25 06:22:14 2002
From: sjmachin at lexicon.net (John Machin)
Date: 25 Feb 2002 03:22:14 -0800
Subject: Correction: Re: string substitutions
References: <mailman.1014504639.21135.python-list@python.org> <c76ff6fc.0202232228.201f829e@posting.google.com> <mailman.1014583431.3100.python-list@python.org> <c76ff6fc.0202242012.48b17a85@posting.google.com> <mailman.1014611499.10778.python-list@python.org>
Message-ID: <c76ff6fc.0202250322.ca8036b@posting.google.com>

Mike Dean <klaatu at evertek.net> wrote in message news:<mailman.1014611499.10778.python-list at python.org>...
> * John Machin <sjmachin at lexicon.net> [2002-24-02 20:12]:
> > <snipped text about RE engine handling \n natively>
> > This important fact is often not documented well or at all.
> 
> Is there an appropriate place to request that this be added to the next
> incarnation of the documentation?  It seems too useful to be ommitted...
> And does it work for any other special characters like \t?
> 

Sure does.

See http://www.lexicon.net/sjmachin/re_escapes.html

I have sent this to python-docs at python.org plus /F and amk for review.


From timr at probo.com  Sat Feb 16 23:43:09 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 16 Feb 2002 20:43:09 -0800
Subject: Help! IIS and Python
References: <eqlm6ugbbtsu4i1d61pek0c4g49as7j119@4ax.com> <a4g77r$ngb$1@thorium.cix.co.uk>
Message-ID: <oqcu6uc4poo2akj76sg1lh76n8tp7kkdcu@4ax.com>

gbreed at cix.compulink.co.uk wrote:
>
>> Since we're on the subject of Python in ASP programming, how do you use 
>> the <%=xxx%> trick with indentation?  This fails with an indentation 
>> error, for example:
>> 
>> <%
>> for i in range(5):
>> %>
>>   <%="aaa%d" % i %>
>
><%
>formatter = """
>  aaa%d
>"""
>
>for i in range(5):
>  Response.Write(formatter%i)
>%>
>
>Why not?

We're diving deeply into religion and philosophy here.  Partly, it's just
the principle of the thing; I assumed there was a magic method to do this
with Python, but I'm coming to the conclusion that there is not.  The
"Response.Write" call seems extraneous to me; VBScript and JScript forms
can omit it with the <%= trick, as if we could write this:

...
for i in range(5):
%>
  <%= formatter % i %>

Perhaps the issue is that I don't know how the <%= form is actually
implemented in Python/ASP.  If it is just a textual substitition, where the
<%= ... %> is replaced with Response.Write(...) by simple textual
substitution, then I understand why it is not possible.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From gbeasley at tsa.ac.za  Mon Feb 25 07:59:28 2002
From: gbeasley at tsa.ac.za (Grant Beasley)
Date: Mon, 25 Feb 2002 14:59:28 +0200
Subject: Dictionary type access to list of instances
Message-ID: <a5dcgh$b22$1@ctb-nnrp1.saix.net>

Hi

If I have a list of instances, each with a property which I want to use as a
key to access the list in dictionary-like manner, what would be the best way
of doing this?

eg.
func get(list, property, value):
    for x in list:
        if getattr(x, property) == value:
            return x

OR

func get(list, property, value):
    return filter(lambda x: getattr(x, property) == value, list)[0]

Note: Normally property would be fixed, and I probably don't need it as
flexible as this. i.e. get(list, value) would probably satisfy my needs.

I could do it using functions, or even overriding the UserDict class, but
ultimately I'll use a similar algorithm. Any comments or ideas appreciated.

Thanks
Grant Beasley




From sholden at holdenweb.com  Thu Feb  7 16:14:32 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 16:14:32 -0500
Subject: cgi POST question
References: <cc14131d.0202071230.415f2447@posting.google.com>
Message-ID: <M6C88.96519$KM2.3594308@atlpnn01.usenetserver.com>

"Christopher" <chris_mk at hotmail.com> wrote ...
> Hi,
>      I am relatively new to programming in general and Python in
> particular (Two years programming basic stuff and the last six months
> with Python).  For the most part, I use Python purely to process and
> extrapolate data.  I have absolutely no Network Programming
> experience.  Lately, I have been using the urllib module to construct
> cgi queries on the fly (get a list of keywords, etc. and plug them
> into a generic cgi query).  As an example (not real code):
>
> ListOfQueries = []
> cgiQuery =
"http://www.site.com/query.cgi?keyword=MYWORD&defaults=blahblah"
> for x in [MyListOfStuff]:
>         newQuery = cgiQuery.replace('MYWORD', x)
>         ListOfQueries.append(newQuery)
>
>
> I would then use the formatted queries for whatever.  My boss asked me
> to add  the ability to do the same on our intranet server.  I thought
> it would be easy.  I was wrong.  Here is what I think is going on
> (though I am sure I have some details wrong).
>      A query like the one above is what is passed when a method of
> 'GET' is specified (the intranet search specifies a method of 'POST'
> instead, I realized much later).  I carefully constructed the cgi
> query, but it wasn't returning the correct results.  I then took the
> source itselt, modified it to submit using a method of 'GET' and
> tried.  The cgi query that was produced matched what I had constructed
> but still no results.  I spoke with the guy in charge of the intranet
> search engine and he said that the cgi script he wrote is incompatible
> with a 'GET' method (my ignorance shows itself, I thought it didn't
> matter which method was used) and he doesn't have the time or the
> inclination to change it (he was also a little bitter that I would
> even be 'messing' with his 'stuff' even though it was his boss' boss
> that told me to do it).  I looked at the cgi module documentation, but
> it doesn't seem to be able to do what I want.

You are writing client-side programs, and wanting to retrieve data from the
intranet web server, correct? The urllib library *is* the one you want ...
see below.

As far as the intranet search engine guy goes, his behavior betrays a
fragile ego and low self-esteem typical of people who aren't really
comfortable doing what they are doing. The good news is you can ignore hom
from here on in.

It's likely he *could* have written his code to work with *either* POST or
GET, but many people don't know the appropriate techniques. Ironically, in
ASP and Cold Fusion it's actually *easier* to write CGI scripts this way!

>      What I want, by the way, is to know if there is a python module
> that submits information in the 'POST' method.  With urllib, I can
> submit information in the 'GET' method (all it is is a website, after
> all), but I can't figure out how to do the same with the POST method
> (I was thinking it would be an object, with attributes in a dictionary
> and a method to post it to the cgi script in the correct format,
> soemthing like:
>
> myTest = cgimodule.cgiPostObject
> myTest['keyword'] = "MyWord"
> myTest['keyword2'] = "MyOtherWord"
> myTest.cgiSubmit("http://www.site.com/query.cgi", method = 'POST')
>
>
> or something to that effect).  I would write it myself, but I wouldn't
> know how.

You are so close you clearly understand Python quite well. See below ...

>      I did have my boss approve the purchases of a couple networking
> books and "Python Web Programming" by Steve Holden (<= very good so

... and you a clearly a person of discrimination :-) Glad you're enjoying
the book ...

> far, btw) but I am still not up to speed yet (basically, I am learning
> Networking stuff, but reading at home, since I am so busy at work).  I
> really need to figure this out because patience is relatively low
> right now.  Any help would be appreciated.  Thank you very much.  I
> would be happy to answer any additional questions. Talk to you all
> later.
>
Here's an exceprt from the urllib docs:

"""
urlopen (url[, data])
Open a network object denoted by a URL for reading. If the URL does not have
a scheme identifier, or if it has file: as its scheme identifier, this opens
a local file; otherwise it opens a socket to a server somewhere on the
network. If the connection cannot be made, or if the server returns an error
code, the IOError exception is raised. If all went well, a file-like object
is returned. This supports the following methods: read(), readline(),
readlines(), fileno(), close(), info() and geturl().
...
If the url uses the http: scheme identifier, the optional data argument may
be given to specify a POST request (normally the request type is GET). The
data argument must in standard application/x-www-form-urlencoded format; see
the urlencode() function below.
"""

In other words, as you surmised, you can fill a dictionary with the data you
want to POST. If you pass the dictionary as a second argument to
urllib.urlopen(), you do a POST.

I'd write you some sample code, except you sound like you're quite capable
of taking it from here. Let us know how it works out.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jason at jorendorff.com  Wed Feb  6 14:51:48 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 13:51:48 -0600
Subject: What is the command to determine an object's type?
In-Reply-To: <19650FF08A99D511B19E00B0D0F0632301898B8F@SRM201002>
Message-ID: <HFEKILOLEFEFMKAECNDLKEMODAAA.jason@jorendorff.com>

Gustavo Cordova writes:
> NO NO NO NO NO NO!!!
> 
> **WHY*** do you want to break beautiful OO-code with this?
>
> [...isinstance...]
>
> Alas, I don't have an answer to your question [...]

Okay, the "isinstance() considered harmful" point of view
has officially gone too far.

## Jason Orendorff    http://www.jorendorff.com/



From mdk at mdk.com  Thu Feb 14 08:14:09 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 08:14:09 -0500
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de> <m3lmdyw6r3.fsf@mira.informatik.hu-berlin.de>
Message-ID: <a4gd73$2dp9$1@ID-98166.news.dfncis.de>

"Martin v. Loewis" <martin at v.loewis.de> wrote in message
news:m3lmdyw6r3.fsf at mira.informatik.hu-berlin.de...
> "MDK" <mdk at mdk.com> writes:
>
> > When I double click on it in Windows explorer I get a message that
tcl80.dll
> > can not be found.  That file does not exist on my PC.  How then is it
> > running under PythonWin if that file does not exists and how can I make
it
> > run when double-clicked?
>
> One possible explanation is that you have an additional Python
> installation whose _tkinter.pyd is linked with an older Tcl version,
> or that you merely have an older _tkinter.pyd on your system.
>
> Regards,
> Martin
>

How do I find out if this is the case and what do I do about it?




From marklists at mceahern.com  Thu Feb  7 10:58:23 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 07:58:23 -0800
Subject: Autocoding evolves from........
In-Reply-To: <007901c1afdf$f901d820$0101010a@local>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEGIKBAA.marklists@mceahern.com>

Chris Gonnerman:
> Ah, perhaps you are right.  I was in a particularly bad mood at 
> that point.
> 
> How the heck to you "enjoy" an idiot like Rue though?

Compassion, humility, and the delete key (or various other tricks).

// m



From syring at email.com  Thu Feb 14 17:32:15 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 14 Feb 2002 23:32:15 +0100
Subject: More license stuff
References: <u6oe6n32m9pjfb@corp.supernews.com>
Message-ID: <a4hdth$a3ng$1@ID-7529.news.dfncis.de>

"Virginia O" <vodine at direct.ca> schrieb
> I snagged this question from a thread below...
>
> >>OK, let's forget about the legal stuff for the moment.
>
>  More importantly, good manners. Let's say I'm wrapping up a Python
program
> with py2exe, and I'm going to distribute it. Which files would I be
expected
> to include as a matter of courtesy to the Python developers (and the
> develop
ers of any additional modules that I use)? Just the licence files, or
> is there anything else?
>
> Cheers,
> Simon Brunning
> >>>>>>>>>
>
> I also, being a newbie, don't understand what all this licensing jargon
> means in terms of distribution. If I write a program, and sell it, then
> what? I do understand I need to speak with a intellectual properties
lawyer,
> but I'd like to get a handle on things before I get started.

Don't worry now. It takes some time until somebody thinks, your software is
valuable enough to steal it. Generally, you should be more worried about
compensation claims from customers for lost business.

Karl M. Syring
--
C++ is history repeated as tragedy. Java is history repeated as farce.
    - Scott McKay





From martin at v.loewis.de  Tue Feb 26 12:54:59 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 18:54:59 +0100
Subject: Converting C strings to Python problem
References: <BwPe8.21867$X2.226876@nnrp1.uunet.ca>
Message-ID: <m3pu2syvpo.fsf@mira.informatik.hu-berlin.de>

"Phil Rittenhouse" <phil.rittenhouse at dspfactory.com> writes:

> I'm trying to pass a path string (eg. "c:\spam\test.exe") from C to python,
> but python is interpreting the string as "c:\spam<tab>est.exe".
> 
> I've run into this problem with converting C strings to Python strings using
> Py_BuildValue().   

Is this specific string appearing literally in your C source code? If
so, I would not be surprised: \t, in C, indicates a tab
character. Write "c:\\spam\\test.exe" instead.

> Should the conversion process convert backslashes in C to double backslashes
> in Python?   What is the recommended method for doing the C string to python
> string conversion?  Does PyString_FromString() behave differently?

If it is what I think it is, the recommend strategy is to write what
you mean, using the C programming language :-)

Regards,
Martin


From threeseas at earthlink.net  Tue Feb  5 23:54:13 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:13 GMT
Subject: Autocoding project proposal.
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <3062.801T2750T14076324threeseas@earthlink.net>

On 03-Feb-02 23:18:30 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>On Mon, 04 Feb 2002 00:00:49 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>>On 30-Jan-02 23:23:31 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>
>>>This is a discuss group (gnu,misc.discuss anyway). It is not a place to
>>>seek coding help. If you don't want a debate don't post to a discussion
>>>forum. If comp.lang.python is like every other comp.lang group I know then
>>>it is also not a place to seek coding help - other than specific language
>>>problems
>>>(such as how do I convert a string into a number?).
>>
>>>You just don't understand how programmers work. And until you do you
>>>won't find many willing to help you. In fact if someone did think your
>>>ideas were good they would probably just fork your code and do it
>>>themselves without your involvement - since you seem to be going out of
>>>your way to turn
>>> people off your project.
>>
>>
>>Don't like the idea of autocoding huh?

>Actually I have no opinion on it at all. I don't have time to get involved in
>a project at the moment, and thuis I haven't tried to understand your long
>explanations of what it is.

>Since I don't know what it actually is, I don't know if I like or don't
>like the idea.

>If it is something that will help non-programmers better use the
>computer resources they have access to without having to learn how
>to program, then it would actually be something I would be interested
>in - since that is an area I am interested in.

>If it is that old assembler type thing then I'm not intersted since
>that is an area I have no interest in.

>If it is to do with autocoder by Smart Software Company, then I'm also
>not interested, as automating open ended questions is not something I
>personally find interesting.

>I only posted to try and help you better communicate with programmers, and
>if your reply is anything to go by you didn;t bother reading my advice,
>or didn't think it was valid. Since your reply has not generated
>any interest in me (a programmer) for your project - but that really
>doesn't matter, as as I said, I'm busy at the moment.


>Sam Holden


And so is the logic failure of a programmer.


You don't have the time to find out if I'm on to something that would
interest you or not, yet you are arrogant enough to give advice regarding
something you don't know about because you don't have the time to find out
if what I'm on to is something you'd be interested in or not.

Full circle!

My first post had links that would have eliminated your multipul choice
list above. But you don't have time for that, do you?

What advice?

Here some for you. Start thinking logically!

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jeff at ccvcorp.com  Mon Feb 11 15:31:18 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 11 Feb 2002 12:31:18 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net>
Message-ID: <3C682A16.1D6C800D@ccvcorp.com>


Jeff Hinrichs wrote:

> I took it as a unintended play on Sandanista, which is a group of  "freedom
> fighting" rebels in Nicaragua.

cf. also Zapatistas, Peronistas, etc... it's common for followers of a specific
leader to be referred to thusly, just as in English we have Marxists,
Leninists, Maoists, etc...  nothing specific about the "-ista" suffix having
anything to do with "freedom fighters"/"revolutionaries"/"terrorists" ...

(Though using Python *could* perhaps be equated to "freedom fighting" in other
ways...  ;) )

Jeff Shannon
Technician/Programmer
Credit International




From usenet at thinkspot.net  Sun Feb 10 01:43:35 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 09 Feb 2002 22:43:35 -0800
Subject: More email module troubles...<sigh>
Message-ID: <a448mo.3vsj90f.1@kserver.org>

OK, I don't think I'm losing my mind...I'm reading the docs 
for the email module over and over. I've looked at this raw 
email message and it doesn't appear to me to not be correctly 
formed.

I am able to successfully read the following email in from a 
file and instantiate it as an email.Message instance. I can 
walk() the parts. I can retrieve the types of the parts, 
etc...(OK, well the email module only detects a single part, 
which is true...), however, when I try this:

msg.as_string()

I get the following Traceback:

>>> msg.as_string()
Traceback (most recent call last):
  File "<pyshell#84>", line 1, in ?
    msg.as_string()
  File "E:\PYTHON\PYTHON22\lib\email\Message.py", line 59, in as_string
    g(self, unixfrom=unixfrom)
  File "E:\PYTHON\PYTHON22\lib\email\Generator.py", line 83, in __call__
    self._write(msg)
  File "E:\PYTHON\PYTHON22\lib\email\Generator.py", line 104, in _write
    self._dispatch(msg)
  File "E:\PYTHON\PYTHON22\lib\email\Generator.py", line 134, in _dispatch
    meth(msg)
  File "E:\PYTHON\PYTHON22\lib\email\Generator.py", line 240, in _handle_multipart
    for part in msg.get_payload():
  File "E:\PYTHON\PYTHON22\lib\email\Message.py", line 151, in __getitem__
    return self.get(name)
  File "E:\PYTHON\PYTHON22\lib\email\Message.py", line 214, in get
    name = name.lower()
AttributeError: 'int' object has no attribute 'lower'


Umm...I don't think I'm doing anything wrong?

Here is the raw message that I'm testing with, below this sig...

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_


>From DirecTV4Free!!!!@yahoo.com Fri Feb 08 04:45:43 2002
Return-Path: <DirecTV4Free!!!!@yahoo.com>
Delivered-To: xthinkspot-usenet at thinkspot.net
Received: (qmail 3201 invoked from network); 8 Feb 2002 09:45:03 -0000
Received: from unknown (HELO mx.futurequest.net) (130.91.184.200)
  by pt01.futurequest.net (63.151.147.170) with SMTP; 08 Feb 2002 09:45:03 -0000
From: "DirecTV 4 Free" <DirecTV4Free!!!!@yahoo.com>
Date: Fri, 08 Feb 2002 04:45:43
To: usenet at thinkspot.net
Subject: DirecTV 4 Free!!!!
MIME-Version: 1.0
Content-Type: multipart/related;
  boundary="----=_NextPart_MIMSCHNGDM"
Content-Transfer-Encoding: 7bit
Message-ID: PM20004:45:43 AM

This is an HTML email message.  If you see this, your mail client does not support HTML messages.

------=_NextPart_MIMSCHNGDM
Content-Type: text/html;charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

<title>Do you have a DirecTV system</title>
<style>
<!--
DIV.Section1 {
	page: Section1
}
SPAN.GramE {
	
}
H2 {
	FONT-WEIGHT: normal; FONT-SIZE: 12pt; FONT-FAMILY: "Times New Roman"; margin-left:0cm;
margin-right:0cm; margin-top:0cm; margin-bottom:0pt
}
-->
</style>
</head>

<body>

<p>&nbsp;&nbsp;&nbsp;&nbsp; Do you have a DirecTV system? Are you tired of 
paying outrageous prices for what you have? Want more? Are you tired of sending 
your card out to other people and waiting days, possibly paying fees, every time 
DirecTV hits your card? Now you can re-set your own DIRECT TV Access Card 
yourself. You can get all the DIRECT TV channels for free. That's over 300 
channels, including all NFL, NHL, NBA, &amp; Sports, HBO, Pay-per-View, Movies &amp; 
Events, XXX and Local Channels. You can do this by using our simple access card 
activating device. It comes complete with the software you need, and simple 
detailed instructions as well as full support. All you need is a home computer.</p>
<p>Not want to mess around with your card? We provide a service where you can 
send your card to us and we load a PRIVATE Script on your card. The script we 
have does not suffer the shutdowns from DirecTV like others may. The last time 
our script was hit was in Oct. 2001, but if it does happen, we will fix it free 
of charge. Not want to send your card? No problem, we can provide you with a 
brand new card loaded with our software, you just put it in your receiver and 
enjoy the shows.</p>
<p>Not have DirecTV? Paying outrageous amounts for your cable that you only get 
a few channels on? Maybe not have cable available in your area or the &quot;other&quot; 
satellite company isn't providing what you expect? We also have all of the 
satellite equipment available in complete packages. DirecTV with TiVO, Ultimate 
TV, etc. are also available.</p>
<p>Live in an area where high speed broadband internet access is not available? 
We have cutting edge technology that can give you broadband internet access 
while watching your FREE DirecTV programming.</p>
<h2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</h2>
<h2>&nbsp;</h2>
<h2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Want More Information????</h2>
<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Then </p>
<blockquote>
  <blockquote>
    <p class="MsoNormal"><span style="font-size: 12pt">

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span><span
style="font-size: 18pt">
    <a style="color: blue; text-decoration: underline; text-underline: single"
href="mailto:freesat12000 at yahoo.com?subject=I Would Like More Information">
    CLICK HERE</a>.&nbsp; </span></p>
  </blockquote>
</blockquote>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<h2>&nbsp;&nbsp;
<a href="mailto:freesatdelete at yahoo.com?subject=Remove From Your Mailing List Please">
Remove Me&nbsp; From Your Mailing List Please</a></h2>

</body>

</html>
------=_NextPart_MIMSCHNGDM--






From sab at NOSPAM.freeuk.com  Tue Feb  5 11:09:42 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Tue, 5 Feb 2002 16:09:42 -0000
Subject: Updating a progress bar?
Message-ID: <a3p01l$omv$1@newsg4.svr.pol.co.uk>

what is the best way to update a progress bar while searching files. I am
writing a simple file searching program and the way i am updating the
progress bar is to scan the drive for the number of files then extrapolate
the progress from the current file against the number scanned! does this
make sense? is there a quicker way? or another method completely. See my
other post titled "Amount of files of a drive?"

G. Willoughby.




From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 09:46:08 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 08:46:08 -0600
Subject: HELP?
References: <a3jebb$cg8$1@knossos.btinternet.com>
Message-ID: <003801c1acc1$8c177e20$0101010a@local>

----- Original Message -----
From: "Callum Golding" <Rave_Dj at btopenworld.com>


> I'm sorry to bother you, but i was wondering if you could give me a little
> advice. I've recently downloaded Python 2.2. Ive read a begginers guide to
> programming and tried creating a small program, I used the input function.
> Now as soon as I open the program it goes into dos and asks me the
question,
> and when i type in an anwser, dos shuts down. Now its really starting to
bug
> me. If you could give me some advice I would be real grateful. The program
I
> wrote is below; -

Fairly common problem.  You neglected to state your OS but I can easily
assume
Windows; note that many of us use other OS's, so you should always state
that.

You are running the script from Explorer, by double-clicking?  In that mode,
a console application (like python.exe) will open a "dos window" (actually
a Windows Console) and run in it.  When the program terminates, the window
closes (as you have noticed).

Two solutions:  First, run your script from the DOS prompt.  Do Start, Run,
"command", ENTER; or find MS-DOS Prompt under Start, Programs.  If you are
running NT, 2000, or XP, you can change "command" to "cmd" (don't type the
quotes in any case, they are there to tell you that the quoted material
is to be typed in).  Once in the DOS window, change directories to the
folder your script is in, and type

    python myscript.py

If you get a "Bad Command or Filename" message, you need your Python
directory
in your DOS PATH variable.  How to do this varies between 95/98 and
NT/2000/XP,
so you should tell us what you are using.

You could type:

    c:\python22\python myscript.py

if you installed Python 2.2 in the standard place, and that would work.

Second solution:  Add

    raw_input()

as the last line of your script.  Then, you will have to press ENTER to end
the script.

Learn to work at the command prompt.  It's a dying art.





From frr at ThouShallNotSpam.EasyJob.NET  Wed Feb 27 11:06:49 2002
From: frr at ThouShallNotSpam.EasyJob.NET (Fernando Rodríguez)
Date: Wed, 27 Feb 2002 17:06:49 +0100
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]> <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>
Message-ID: <lv0q7usfhlkrriu6bvg630hvih8bopt80j@4ax.com>

On Wed, 27 Feb 2002 16:31:12 +0100, Fernando Rodr?guez
<frr at ThouShallNotSpam.EasyJob.NET> wrote:


>If you want to learn how to program, then I wouldn't recommend using Python.
>Python _is_ a great language for beginners, but there many _great_ (and I mean

there _aren't_ many ....
Damn. O:-)



| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Fernando Rodr?guez   frr at EasyJob.NET 
| http://www.EasyJob.NET/
| Expert resume and cover letter creation system. 
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


From google at cb921.org  Mon Feb 11 03:16:06 2002
From: google at cb921.org (cb921)
Date: 11 Feb 2002 00:16:06 -0800
Subject: mxODBC, stored procedure syntax
References: <e746a5a8.0202080441.4bd2c023@posting.google.com> <a40s9m$aif$1@suaar1aa.prod.compuserve.com> <mailman.1013205451.11448.python-list@python.org>
Message-ID: <e746a5a8.0202110016.4012398f@posting.google.com>

Hi,

"M.-A. Lemburg" <mal at lemburg.com> wrote...
>"Fran?is Lepoutrre" wrote:
>> 
>> As far as i understand it there is currently
>> no support for returning params from
>> storedprocedures at the moment.
>
>There is: let the stored procedure generate a result
>set (or a temporary table) and then fetch the data
>using the standard .fetchxxx() methods.

Thank you.

Excuse my lack of clue for a moment more, please.  Does this
involve any considerations for those creating the stored
procedures?  The stored procedures are already in use by a
number of others, who are using VB to access them.  I am sure
they will not be too happy to change them. :<


again thanks.

c


From mcherm at destiny.com  Mon Feb 11 11:23:43 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 11 Feb 2002 11:23:43 -0500
Subject: The joys of weakrefs... references to instance methods (callbacks)
Message-ID: <3C67F00F.9030203@destiny.com>

Mike Fletcher writes:
 > [complicated workaround to annoying problem with "obvious and
 > pythonic" implementation of common design pattern]

Ouch.

I'm no expert, but I see your point, and can't beat your workaround. 
Thanks for sharing.

-- Michael Chermside




From rony.steelandt at bucodi.com  Mon Feb  4 10:56:48 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 4 Feb 2002 07:56:48 -0800
Subject: A Python console in a Visual Basic application????
References: <5d65cf44.0202032348.79395fea@posting.google.com>
Message-ID: <b34d3835.0202040756.303f145d@posting.google.com>

codetube at yahoo.com (CodeTube) wrote in message news:<5d65cf44.0202032348.79395fea at posting.google.com>...

This book  will answer all this questions:
ISBN 1-56592-621-8 from O'Reilly

Python programming on win32 from Mark Hammond& Andy Robinson

> Hello Everyone,
> 
> I wonder if any one knows some thing about implementing the PythonWin 
> console and other capabilities from with in a Visual Basic 
> application.
> I am writing an application in which I propose to have a console 
> similar to PythonWin and also be able to run python files etc.
> This might be a "big" question, so I would appreciate any clues or 
> leads in that direction.
> I had looked into some examples given by Mark Hammond in his 
> book "Programming Python" but then I am looking for a more tight 
> integration into the VB application.
> 
> Any help would be appreciated...
> 
> Best Regards,
> Code Tube


From phr-n2002a at nightsong.com  Mon Feb 25 22:59:30 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 19:59:30 -0800
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <7xlmdglwpp.fsf@ruckus.brouhaha.com>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> I thought I'd toss out the following two ideas:

Sigh.  "Two ideas" followed by sections labelled 1, 2, and 3.  I
should have labelled the differently.  There are two ideas as stated;
sections 1 and 2 are two parts of the same idea.  Section 3 is the
second idea.  Sorry for confusion.


From rhymes at myself.com  Sat Feb  2 15:07:52 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sat, 02 Feb 2002 21:07:52 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012671184.6510.python-list@python.org>
Message-ID: <kjho5uc7ut6tt12898b62sthsq5hdofvqu@4ax.com>

On Sat, 2 Feb 2002 20:31:16 +0300, Oleg Broytmann <phd at phd.pp.ru>
wrote:

> It *is* in the distribution. It is not a single module, but a package.

>>> import curses
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: No module named curses
>>> help('curses')
no Python documentation found for 'curses'

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From wesc at deirdre.org  Fri Feb  8 13:08:22 2002
From: wesc at deirdre.org (Wesley J. Chun)
Date: 8 Feb 2002 10:08:22 -0800
Subject: ANN: SV-SF Bay Area Python users group (BayPIGgies) mtg 2/13 7:30pm
Message-ID: <mailman.1013205907.14124.clpa-moderators@python.org>

Date:     February 13, 2002
Time:     7:30pm - 9pm
Location: Stanford University
Agenda:   Python 10 Conference Summary
Speaker:  Todd Valentic, SRI International

Many of us were not be able to make it to the Python10 Conference (Feb
4-7 2002). Todd, who presented a paper at the conference, will give us
the lowdown on what happened.

More information including directions at http://deirdre.org/baypiggies



From skip at pobox.com  Wed Feb 13 08:37:08 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 13 Feb 2002 07:37:08 -0600
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <m33d05u2dn.fsf@ferret.object-craft.com.au>
References: <mailman.1013557302.9749.python-list@python.org>
        <7xzo2ekv5g.fsf@ruckus.brouhaha.com>
        <mailman.1013572042.18876.python-list@python.org>
        <3C69ED33.CE0100DF@pizero.colorado.edu>
        <15465.63503.142011.246239@12-248-41-177.client.attbi.com>
        <Pine.LNX.4.21.0202122224100.12912-100000@higgs.colorado.edu>
        <m33d05u2dn.fsf@ferret.object-craft.com.au>
Message-ID: <15466.27652.704528.795698@12-248-41-177.client.attbi.com>

    Dave> Use strace to track system calls.

Thanks, excellent suggestion.  Right after a full backup...

Skip



From quinn at vomit.ugcs.caltech.edu  Sat Feb  9 00:58:53 2002
From: quinn at vomit.ugcs.caltech.edu (Quinn Dunkan)
Date: 9 Feb 2002 05:58:53 GMT
Subject: email module Bug in 2.2
References: <a4164b.3vv7l9p.1@kserver.org>
Message-ID: <slrna69eks.c6s.quinn@vomit.ugcs.caltech.edu>

On Fri, 08 Feb 2002 18:41:14 -0800, Sheila King <usenet at thinkspot.net> wrote:
>OK, here is a bug in the mail module. It has to be????
>
>I took an email that I received (a spam) and saved it to a text 
>file (which I called 'dtv.txt') and then imported it into an email class
>object in an interactive session. 
>
>I've pasted below an interactive session, and below that I've pasted the text 
>file that was used in the session. There seem to be some inconsistencies?

...

>E:\Web\Thinkspotwebsite\dev>python
>Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>Type "help", "copyright", "credits" or "license" for more information.
>>>> import email
>>>> msg = email.message_from_file(open('dtv.txt'))
>>>> msg.get_type()
>'multipart/related'
>>>> msg.get_main_type()
>'multipart'
>>>> msg.is_multipart()
>0
>>>>

>From the libref:
is_multipart()
Return 1 if the message's payload is a list of sub-Message objects, otherwise
return 0. When is_multipart() returns 0, the payload should either be a string
object, or a single Message instance.

So notice that is_multipart() does not check if get_main_type() is 'multipart'.
I don't know much about the MIME hairball, but the msg appears to claim it is
multipart, but doesn't actually have multiple parts.  Message.add_payload only
sets the payload to a list if it's called more than once, not if the msg calls
itself 'multipart'.  I'm not sure if this is correct or not, though I'm
inclined to agree with you that it's not (section 12.2.2.2 implies that you
should only get a non-string, non-list payload for some message/* types), but
is_multipart() is behaving strictly according to documentation.

As far as practicality goes, when recursing over MIME objects, you're going to
have to handle the non-multipart non-string case anyway, because of those
message/* types.  My MIME code looks like:

def frob_mime(msg):
    if msg.is_multipart():
        for m in msg.get_payload():
            frob_mime(m)
    elif hasattr(msg.get_payload(), 'get_payload'):
        from_mime(msg.get_payload())
    else:
        t = msg.get_type()
        ... do stuff


From jason at jorendorff.com  Wed Feb  6 12:22:52 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 11:22:52 -0600
Subject: re difference between python 1.5 and python 2.1.2
In-Reply-To: <a3rk7v$vlb$1@norfair.nerim.net>
Message-ID: <HFEKILOLEFEFMKAECNDLEEMEDAAA.jason@jorendorff.com>

Gillou wrote:
> re.compile(r'((ftp|http)://[\w-]+(?:\.[\w-]+)*(?:/[\w-\.?=]*)*)')
                                                       ^

Put a backslash before the last - in the regular expression.
1.5.2 should not have accepted this.  It's a bug in 1.5.2.

## Jason Orendorff    http://www.jorendorff.com/



From brian at sweetapp.com  Mon Feb  4 14:51:04 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 4 Feb 2002 11:51:04 -0800
Subject: CPython vs. Jython for Web applications?
In-Reply-To: <c775c8db.0202041126.5177e606@posting.google.com>
Message-ID: <000601c1adb5$48b17600$445d4540@Dell2>

John E. Barham wrote:

>   Are there real-world benchmarks comparing, say, Jython & Saxon vs.
> CPython & Pyana/Xalan?

I'm the author of Pyana and I did some Saxon vs. Xalan benchmarks about
2 months ago. Saxon was about 10% faster on average. I should also point
out that Xalan has only been tested on RH6.x and crashes when built on
RH 7.x (at least in my experience) so there are some gcc/libc issues
there. 

The pure Python XSLT implementation from 4suite is supposed to become an
order of magnitude faster with the next release.

Cheers,
Brian




From mdk at mdk.com  Thu Feb  7 10:26:02 2002
From: mdk at mdk.com (MDK)
Date: Thu, 7 Feb 2002 10:26:02 -0500
Subject: Scope question
Message-ID: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de>

I have this:

# myprog.py

from mymod import *

def dosomething(x):
    dosomething(1000)

# End code

I want the dosomething(1000) run from the function in mymod.  However,
instead it is running dosomething(x) in myprog.py.

Note: Changing the names of dosomething() in either module is not an option.
(The names depicted here are not the real names in the program but have been
changed for clarity.)

I know that one way to fix this is to change

from mymod import *

to

import mymod

But I would like to know if, without changing the import line, there is a
way to tell Python to use the dosomething() from mymod.

I tried mymod.dosomething(1000) but it returns an error:
AttributeError: 'builtin_function_or_method' object has no attribute
'dosomething'

Thanks!




From danb_83 at yahoo.com  Sat Feb 16 15:40:49 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 16 Feb 2002 12:40:49 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl>
Message-ID: <ad052e5c.0202161240.73499031@posting.google.com>

Piet van Oostrum <piet at cs.uu.nl> wrote in message news:<wzadu9scs2.fsf at nono.cs.uu.nl>...
> >>>>> rmunn at pobox.com (Robin Munn) (RM) writes:
> 
> RM> Reading the first few chapters of _How to Think Like a Computer
> RM> Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/ for those who haven't
> RM> heard of this book), I was struck by how Python resembles BASIC in some
> RM> ways that make it a very good language for teaching:
> 
> BASIC a very good language for teaching????
> Well, it depends on what you want to teach, I guess.

Everyone should learn BASIC.  It's the perfect example of how *not* to
design a programming language :-)


From h.jehadalwan at student.kun.nl  Fri Feb 22 12:56:01 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Fri, 22 Feb 2002 18:56:01 +0100
Subject: is their any limit on the size of the data passed to a program via popen2?
Message-ID: <3C768631.9050803@student.kun.nl>

hi, I'm trying to pass files to a program ( a biochemistry program 
called "profit") for processing via popen2. I have 10000 files with 
total size of 1500 Mb. The files have to be passed to the program one by 
one. Each file represents a molecule, which has to be compared to a 
reference molecule. The code works only for ~200 files and fails when I 
pass all the files. What should I do?


The code:

  pdbdir = os.popen('ls /home/data/pdbs') # if this directory contains 
10000 files, the code does not work.

files=pdbdir.readlines()
pdbdir.close()
(o,i) = popen2.popen2("/home/programs/profit ")

i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") # This the 
reference molecule.

for PDB in files:
     i.write(" mobile /home/data/pdbs/" + PDB + "\n") # A new molecule.
     i.write("fit\n")		# compares reference and new molecules.

i.close()
result= o.read()
o.close()


Thanks in advance.



From jolsen at mailme.dk  Fri Feb  1 04:41:23 2002
From: jolsen at mailme.dk (Jesper Olsen)
Date: 1 Feb 2002 01:41:23 -0800
Subject: os.fork()
Message-ID: <cf0ad9fb.0202010141.7d9425bd@posting.google.com>

os.fork() is an alternative to thread's

Is there a way for the parent to find out when the child 
has terminated?

fork() is unix specific, so actually I would prefer to use threads.
However, the advantage with fork is that global variables are
not shared between parent and child. 

Cheers
Jesper


From wolfson at uchicago.edu  Wed Feb 20 11:59:05 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Wed, 20 Feb 2002 10:59:05 -0600
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <a50hd2$6vt$1@panix2.panix.com> <a50im3$bf1$1@news.island.liu.se>
Message-ID: <pan.2002.02.20.10.59.02.252838.1869@uchicago.edu>

On Wed, 20 Feb 2002 10:23:22 -0600, Anders Dahlberg wrote:

> 
> "Aahz Maruch" <aahz at panix.com> skrev i meddelandet
> news:a50hd2$6vt$1 at panix2.panix.com...
>> In article <m2wux85hud.fsf at desk.crynwr.com>,
>> Russell Nelson  <nelson at crynwr.com> wrote:
>> >Henrik Motakef <henrik.motakef at web.de> writes:
>> >> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
>> >>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
>> >>>>
>> >>>> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
>> >>>> czignat?  Classifero en bragna!
>> >>> Woas da Deifi wos des hoasn mog.
>> >> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
>> >
>> >Even better: we could switch languages within threads, to *really*
>> >keep the monolinguists confused.
>>
>> B'vakasha
> 
> Ja, varf?r inte?

Leider latinam non bene dicere, sed conari possum...

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From landshark at verticaladdiction.net  Thu Feb 14 10:02:49 2002
From: landshark at verticaladdiction.net (Dave Marotti)
Date: Thu, 14 Feb 2002 15:02:49 -0000
Subject: undefined symbols when embedding python
References: <slrna6mm3i.2aj.landshark@burrito.verticaladdiction.net> <m3y9hwpht2.fsf@mira.informatik.hu-berlin.de>
Message-ID: <slrna6nkf8.2sp.landshark@burrito.verticaladdiction.net>

Hello,

I tried out the separate extension module, and sepearte embbeded application,
and it worked.

I tried out the all-in-one, and it didn't.  It could have something to do with
what SIP created, I'm not sure.  But when I attempt to import my wrapped stuff
in python, it tells me it can't find libwraptestc.so

However, looking at wraptest.py, I see:

------------------------------------------------------------
# Python wrapper code.
#
# Generated by SIP v3.0 on Thu Feb 14 00:21:11 2002
import libwraptestc

libwraptestc.sipInitModule(__name__)

class WrapTest:
    def __init__(self,*args):
        libwraptestc.sipCallCtor(0,self,args)
    def __del__(self):
        __sipDtor__(self)

# Register the classes with the C++ module.
libwraptest.sipRegisterClasses()
-------------------------------------------------------------

So it looks like it has to be a SIP thing.

Any idea how I can avoid the external .so reference here? Is there code I need
to change?  I'm looking at the wraptestcmodule.cpp and trying to figure out if
I should be calling initModule(...) directly in my C++ code to register the
module.  If so, what type is __name__ in python, so I can convert it from C++
to python in the call?

-dave

>Of course, this problem would also go away if you would incorporate
>the extension into the application.


-- 
+-----------------------------+----------------------------------------------+
| Dave Marotti                | Looking for a Visio alternative for *nix?    |
| lndshark ! speakeasy net    | Kivio : http://thekompany.com/projects/kivio |
+-----------------------------+----------------------------------------------+
| I just don't trust anything | I suppose that in a few more hours I will    |
| that bleeds for 5 days and  | sober up. That's such a sad thought. I think |
| doesn't die. - Mr. Garrison | I'll have a few more drinks to prepare myself|
+-----------------------------+----------------------------------------------+



From phr-n2002a at nightsong.com  Sat Feb  9 19:28:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 09 Feb 2002 16:28:17 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a43due.3vv9a5r.1@kserver.org>
Message-ID: <7xg04ai3j2.fsf@ruckus.brouhaha.com>

Sheila King <usenet at thinkspot.net> writes:
> I have a friend I'm working with right now, that we are putting together
> a project in Python and PHP (he's doing all of the PHP part). He has
> only just been learning Python for a few weeks. He might have some good
> suggestions for you, as to how he approached learning Python (as an
> experienced programmer in other languages). If you'd like me to put you
> in touch with him, send me an email.

Topic drift: why are you doing something in Python -and- PHP?
Can't you do it all in Python, or maybe all in PHP?  Why use two languages?
There's some Python template systems that do pretty much what PHP
does.  If something is missing from them, maybe they need improvement.


From afp_randjohnson at yahoo.com  Sun Feb 24 07:19:02 2002
From: afp_randjohnson at yahoo.com (Mike Johnson)
Date: 24 Feb 2002 04:19:02 -0800
Subject: Retrieve a variable from an embedded interpreter
References: <zo1zjioe.fsf@fido.trinitycapital.com>
Message-ID: <sn7r2hu1.fsf@fido.trinitycapital.com>

Talking to myself here. :-)

So I've continued searching the web. (Searching for ".py" is the only
way to go. :-)

Anyhow, I've been playing with the Py_RunString
function. Unfortunately, I now get segfaults. But I think I'm getting
closer...

I was following what this guy did:

http://mail.python.org/pipermail/python-list/1999-November/015502.html

stdout + stderr:

a=hi Assertion failed Aborted

---

code:

---

void cb_script_execute ( void ) {
        PyObject *a;
        PyObject *pdict;
        char *sa;

        Py_Initialize ();

        pdict = PyDict_New();
        PyDict_SetItemString(pdict, "__builtins__", PyEval_GetBuiltins());

        PyRun_SimpleString ( "a = \"hi\"" );
        PyRun_SimpleString ( "print \"a=\" + a" );
        a = PyRun_String ( "a", 0, pdict, pdict );
        sa = PyString_AsString ( a );
        printf ( "sa = %s\n", sa );

        Py_DECREF ( pdict );
        Py_DECREF ( a );
        Py_Finalize ();

}

Thanks,

- Mike Johnson



From jeff at ccvcorp.com  Wed Feb 13 15:33:58 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 13 Feb 2002 12:33:58 -0800
Subject: controlling web browser in windows
References: <1103_1013605186@netnews.cf.ac.uk> <Zeua8.206395$KM2.8473296@atlpnn01.usenetserver.com>
Message-ID: <3C6ACDB6.95186821@ccvcorp.com>


Steve Holden wrote:

> Alan:
>
> There is a COM object in the Windows environment which exposes most of the
> functionality of Internet Exploder. This would proably give you the most
> satisfactory control of the experimental (?) environment, though you would
> have to learn how to "drive" COM objects form Python. Hammond and Robinson's
> "Python Programming on Win32" probably explains most of what you need to
> know about that.
>
> Since I am unable to easily find a reference to the object in question I'll
> leave that to someone who uses it regularly!

When I was working on a similar problem (automating file-uploads via a web
form), I used both the book that Steve mentions (Python Programming on Win32)
and the MS Press book "Programming Internet Explorer 5".  The latter is a bit
heavy on the cheerleading, but it contains *lots* of valuable information on the
IE COM interface.  The only problem is that it's written for VC++ and Visual
Basic... translating to Python can be a nuisance, but isn't *too* hard (but
stick with the VB examples, it's much closer to what you'll want to do).

Of course, I then discovered that much of what I *wanted* to do, was forbidden
from scripts for security reasons, but that's another story.  ;)  (I probably
could've gotten around this by writing my own framework to support their
WebBrowser COM object, but didn't want to put in that much effort...)

Jeff Shannon
Technician/Programmer
Credit International




From aahz at panix.com  Fri Feb 15 22:17:11 2002
From: aahz at panix.com (Aahz Maruch)
Date: 15 Feb 2002 19:17:11 -0800
Subject: basic books/guides on multithreading programming?
References: <mailman.1013813246.12449.python-list@python.org>
Message-ID: <a4kivn$p2g$1@panix2.panix.com>

In article <mailman.1013813246.12449.python-list at python.org>,
Tripp Scott  <tripps81 at yahoo.com> wrote:
>
>which ones do you recommend? examples in any language is fine, though i
>prefer python, perl, or ruby. what i would like is guidance on what is
>thread safety, how to achieve it, what are the issues in multithreading
>programming, the global interpreter lock, and other basics.

http://starship.python.net/crew/aahz/
should give you a start
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From simon at ping-pong.budig  Sat Feb  2 16:09:58 2002
From: simon at ping-pong.budig (Simon Budig)
Date: 2 Feb 2002 22:09:58 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de>
Message-ID: <3c5c55a6@si-nic.hrz.uni-siegen.de>

In article <3c5b2c1f at si-nic.hrz.uni-siegen.de>, Simon Budig wrote:
> Michael Hudson <mwh at python.net> wrote:
>> Simon Budig <Simon.Budig at unix-ag.org> writes:
>>> Something is broken with the completion of readline:
>> 
>> Well, I knew about it.  I don't regard it as a huge problem.
> 
> Hmm. Imagine you are writing an interactive interpreter of a
> custom language to manipulate something non-file'ishes and
> suddenly the commandline starts completing *filenames*. I
> think this is a good showstopper...
> 
>>> Is there an evil hack to avoid this?
>> 
>> Dunno.  Doubt it, somehow.  You'll need to look at the documentation
>> for readline, though.
> 
> Well - I looked at the documentation for the readline-module, but that
> doesn't say anything about it. Maybe I should read the sources.

Ok, did that. Here is the Patch to fix this behaviour. It is against
Modules/readline.c from Python 2.1.2.

----8<-----------------------------
--- readline.c.orig	Sat Feb  2 21:44:09 2002
+++ readline.c	Sat Feb  2 22:01:16 2002
@@ -346,6 +346,9 @@
 		   lock released! */
 		save_tstate = PyThreadState_Swap(NULL);
 		PyEval_RestoreThread(tstate);
+		/* Don't use the default filename completion if we
+		   have a custom completion function... */
+		rl_attempted_completion_over = 1;
 		r = PyObject_CallFunction(completer, "si", text, state);
 		if (r == NULL)
 			goto error;
----8<-----------------------------

I really would appreciate, if this gets incorporated in the regular
Python releases. Unfortunately I don't have an account at sourceforge
to submit this the proper way.

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From jason at jorendorff.com  Sat Feb  2 10:14:16 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 2 Feb 2002 09:14:16 -0600
Subject: Elegant solution needed: Data manipulation
In-Reply-To: <12e3779b.0201311554.2891bb28@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEOJCPAA.jason@jorendorff.com>

Mark wrote:
> What follows is what I came up with from reading the re stuff
> overnight and toying around a bit.  I hope it doesn't fall into the
> Python "anti-idiom" trap.  I'm definitely open to a communal "code
> review".

I accidentally wrote my formula like this:

  add( shoesize, mul ( height, weight ))

without a space between "add" and "(".  So it gave me
wrong answers.  (Anything that manipulates code using
regular expressions will have flaws in some cases.)

When you use regular expressions, use raw strings.
But in some cases you can use the replace
method:

  x = x.replace(name, "data['%s']" % name)

I'm a bit surprised you didn't take Steve's code and
say "Yay, I'm done!".  I'm extra surprised you didn't
reply to any of the responses to your message or
ask any followup questions.  Your approach might be
faster than Steve's (I'm not sure), but it seems a
bit more fragile.

Cheers, and enjoy Python!

## Jason Orendorff    http://www.jorendorff.com/



From tismer at tismer.com  Fri Feb 22 17:04:22 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 22 Feb 2002 23:04:22 +0100
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com>
Message-ID: <3C76C066.9080506@tismer.com>

Christopher Browne wrote:

> Centuries ago, Nostradamus foresaw when Christian Tismer <tismer at tismer.com> would write:
> 
>>Dear Stackless Python Community,
>>
>>As you might have noticed, Stackless has changed quite
>>much:
>>- Platform independence is gone.
>>
> 
> I hope not...  I would hope to hear that "platform indepedence has
> _ARRIVED_"


You can't get everything for nothing.
This time, I paid ease of maintenance
and other plans with my future by
abandoning platform independeance.

I have to say -- this was one of the best
decisions of my life!

If your platform is not supported, well sorry, then
you are a minority anyway. Either you will want to
support my porting efforts with hardware and payware,
or I will stay happy with the 95 percent of all machines
that I got already.

Sounds harsh? Not intended, it's the bare truth

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From mhammond at skippinet.com.au  Thu Feb 14 18:32:42 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 14 Feb 2002 23:32:42 GMT
Subject: Problem freezing Win32 services
References: <a4g3s3$k0s$1@news.ost.eltele.no> <a4g5sl$f89$1@ID-59885.news.dfncis.de>
Message-ID: <3C6C4984.4090405@skippinet.com.au>

Thomas Heller wrote:


> I'm unsure about the state of freeze (at least for COM servers and services).
> Maybe someone else can comment?

I added the service etc support to freeze quite some time ago.

> <shameless plug>
> You could (should ;-) try out the newest py2exe release, which also can
> create service exe-files: http://starship.python.net/crew/theller/py2exe/

And I will hopefully never touch freeze again, thanks to py2exe and the 
McMillan Installer :)

Mark.



From nospam at [127.0.0.1]  Wed Feb 27 10:05:01 2002
From: nospam at [127.0.0.1] (Chris)
Date: Wed, 27 Feb 2002 15:05:01 +0000
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
Message-ID: <UrIWUMDdWPf8EwAj@[127.0.0.1]>

In the olden days I played around with BASIC
just for fun, writing trivial games and making patterns move on the 
screen.

Can I do that sort of thing with Python?

I've at last got a new machine, running XP Home.
Will this be OK?
Or would it be wiser to use my old 200MMX?

Before I start wrestling with Python - and with thinking - and with 
making my brain hurt - is Python the best language for this sort of 
play?
Would Java be better?
Will I be able to make a ball bounce around inside a box etc.?

Please be gentle!  I'm not a rocket scientist - but rather just an 
ordinary bloke who's really quite interested in how computers work and 
who would like to edge sidelongly towards greater technological 
intimacy!
-- 
Chris


From kmj9907 at cs.rit.edu  Fri Feb  8 10:59:38 2002
From: kmj9907 at cs.rit.edu (Keith Jones)
Date: Fri, 8 Feb 2002 10:59:38 -0500
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <a3u9sk$890$1@news.mathworks.com> <3C62D5A5.4B57F487@ccvcorp.com>
Message-ID: <3c63f1e3@buckaroo.cs.rit.edu>

> > from mymod import *
> >
> > def dosomething(x):
> >     import mymod
> >     mymod.dosomething(1000)
>
> Nope,this does not work.  As the O.P. already noted....
>
...
>
> from module import *  does not create a reference to module in the current
> namespace, so there's nothing there to look up attributes from.
>

did you miss the fact that he's importing mymod in the function?  'Course,
based on the example, then, the 'from mymod import *' is unnecessary, I
would think.

--
-Keith Jones
http://www.codeexamples.org





From joshm at taconic.net  Sun Feb 10 23:56:44 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sun, 10 Feb 2002 23:56:44 -0500
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com>
Message-ID: <3c674cbd_2@corp.newsgroups.com>

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>>> ''.join(map(max,zip('00','AA')))
'AA'
>>> ''.join(map(max,zip('00','A')))
'A'
>>> ''.join(map(max,zip('00','')))
''

Damn, I thought I had it.  :-)  Is there an variant of zip which doesn't
truncate?  BTW, this does work, provided that the input strings are the same
length, and even allows for more than two strings!

>>> s = ['abte','scrt','sdfw','cxvs']
>>> ''.join(map(max,zip(*s)))
'sxvw'

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From mwh at python.net  Fri Feb 15 06:29:50 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 15 Feb 2002 11:29:50 GMT
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com> <1013641215.358571@yasure> <a4f29e$bhm$1@animus.fel.iae.nl> <a4h039$2r2q$1@nntp6.u.washington.edu>
Message-ID: <ulmdvf0ab.fsf@python.net>

"Donn Cave" <donn at u.washington.edu> writes:

> I guess I should read the manuals some time.  I wonder if they
> explain why os.popen2 reverses the order of popen2.popen2's returns.

I think (hope!) this one is entirely due to hysterical raisins.

Cheers,
M.

-- 
  I've even been known to get Marmite *near* my mouth -- but never
  actually in it yet.  Vegamite is right out.
 UnicodeError: ASCII unpalatable error: vegamite found, ham expected
                                       -- Tim Peters, comp.lang.python


From calves at coelce.com.br  Mon Feb 18 12:45:16 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Mon, 18 Feb 2002 14:45:16 -0300
Subject: DBgrid in python
Message-ID: <29A97D00F387D411AC7900902770E14803F21CCD@lcoeexc01.coelce.net>

Is there any component in python as DBGrid component in Delphi?! If so,
where can I get it?!

Thanks,

Carlos Alberto
COELCE/DPRON-Departamento de Projetos e Obras Norte
Fone: 677- 2228
e-mail: calves at coelce.com.br
\|||/
(o o)
--ooo0-(_)-0ooo--

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020218/52fa7ffc/attachment.html>

From kp87 at lycos.com  Sun Feb 17 07:35:32 2002
From: kp87 at lycos.com (kevin parks)
Date: 17 Feb 2002 04:35:32 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net>	<mailman.1013718563.21631.python-list@python.org>	<5e8bd451.0202142238.580f15c1@posting.google.com> <mailman.1013783429.16518.python-list@python.org>
Message-ID: <5e8bd451.0202170435.17a9483d@posting.google.com>

Thanks to all for your help. I am learning much by looking at this
code. I am always amazed at how many things are built in to Python.
bisect is one of those things that you always forget is there... I am
going to look at what folks posted a play around some. Though I am
really happy too see those lambdas get explained I was about to pop a
vein trying to get my head 'round those things.

I sort of guess you were goofing on me, but the funny thing is that
those functions did mostly work....

Thanks for your help everyone I am amazed that you could all figure
out what i was getting at. I think that I made a mistake or two in my
examples (i did on the fly) but i will play with the code now to see
that it give the results that I am after.

cheers,

-kevin


From mcherm at destiny.com  Mon Feb 11 15:59:29 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 11 Feb 2002 15:59:29 -0500
Subject: defining functions
Message-ID: <3C6830B1.7050600@destiny.com>

 > I am pretty new to python, and I dont understand the parameters when 
defining
 > functions. Whenever I see a function definition with a argument It 
seems like
 > the argument already is a variable, but it goes threw the function 
and it makes
 > no sense. I dont know if what I just said made any sence but if you 
can help me
 > please reply
 > def factorial(n):
 >     if n == 1:
 >         return 1
 >     else:
 >         return n * factorial(n-1)
 > What the hell does 'n' have assingned to it?

I think that Antaeus Feldspar gave an *excellent* easy-to-understand 
answer to your question, but I wanted to add one more point that might 
help you out.

His explanation told how a method like this would work:

 >   def average_two(first_num, second_num):
 >       sum = first_num + second_num
 >       average = sum / 2.0  # Dividing by 2 would abandon any 
remainder
 >       return average

But your function (factorial) used a special feature called "recursion". 
Recursion is a really useful feature, but it makes it more difficult to 
answer the question "What the hell does 'n' have assigned to it?".

If you think of things the way that Antaeus explained them, you'll 
understand what the factorial() function does when you call it as 
"factorial(1)": It temporarily makes 'n' take on the value of 1, then it 
does the test 'if n == 1', which is TRUE, so it returns 1. But when you 
call "factorial(2)" it gets a lot harder to understand.

YOU could probably work it out in your head: Factorial of 1 is defined 
to be 1, and factorial of two is defined to be 2 * factorial of 1. So 
factorial(2) should be 2*1 or simply 2. Hey... in your head, you could 
probably do higher numbers the same way: Factorial of 3 is defined to be 
3 * factorial(2), which we ALREADY know is 2, so factorial of 3 must be 6.

If I were to ask you the factorial of 5 (and you didn't know it), then 
you'd have to start by working out the factorial of 4 (which is 4 * 
factorial of 3... ie 4 * 6 or 24), and then multiply THAT by 5 (so 5*24 
= 120). And that's exactly what the computer does.

Say you define the factorial function as above, and then type 
"factorial(5)" in at the command line. Well, of course (as Antaeus 
explained quite nicely) 'n' gets assigned the value 5. We run the test 
("if n == 1"), and it's FALSE so we do the later line: "return n * 
factorial(n-1)". The computer calculates n-1 (it's 4), but then needs to 
know what "factorial(4)" is.

So it starts the whole thing over. I "remembers" that it has to multiply 
by 'n' (which is 5) when it's done, and goes off to figure out what 
"factorial(4)" is. To do THAT is has to create a NEW 'n' which is equal 
to 4 and run through the whole calculation again. It keeps doing this 
until it gets to n=1 which (as we already know) is easy, and then it can 
work out all of the answers.

A few pieces of vocabulary. The "remembering" of the old value of n 
while it takes a break to work out "factorial(4)" is called a "stack 
frame". And the whole process is called "recursion".

Hope that helps!

-- Michael Chermside







From jason at jorendorff.com  Sun Feb  3 13:28:23 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 12:28:23 -0600
Subject: Java vs Python Benchmarks: Java is faster
In-Reply-To: <sdqp5u49m52b1u8vq1fe1gtm7gjej7t1fj@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAEBGDAAA.jason@jorendorff.com>

Courageous writes:
> Kragen Sitaker writes:
> >It's true that Java objects all have declared types, but that doesn't
> >save you as much as you'd think;
> 
> While you obviously know what you're talking about and don't need my
> help, I'd just like to add that those folks talking about an optional
> type declaration system for speeding up Python don't seem to have a
> good idea of where the time in Python goes. There are more important
> fish to fry, as I'm sure you realize.

On behalf of all of us that don't know what we're talking about and
do need your help...

I guess I imagine that Python's speed issues are:

 1. Ultra-late binding of, and consequent repetitive lookups of,
    global names, attributes, methods, module members, etc.
    (Of course, unless names like 'int' and 'float' can be bound
    at compile time, optional type declarations are unhelpful.)

    Since many people like this about Python, and many have used it
    in existing code, effort is focused on making lookups cheaper
    (or caching the results) rather than changing the binding
    behavior to require fewer lookups overall.

 2. Nothing ever gets compiled to anything even close to
    machine code.  This is largely because the Python development
    team doesn't have the resources to invest in it.  Clearly
    there's something to gain by moving the huge ceval.c loop/switch
    from runtime to compile time.  No one knows how much.  Clearly
    there's something to lose from the resulting explosion in the
    complexity of the compiler.

 3. Even if you did compile Python to machine code, simple things
    like "x = y + z" would still be packed with object-oriented
    wackiness and much chasing-about of heap pointers.  It doesn't
    compile to simple register code like you want.

 4. Function calls are slow.

 5. Python still has SET_LINENO, which is basically a waste
    of time.  (This is low-hanging fruit, but apparently the
    team is too busy to grab it.)

I believe static typing would help with #1 and #3.

More later.

## Jason Orendorff    http://www.jorendorff.com/



From wolfgang.grafen at gmx.de  Wed Feb 27 20:19:56 2002
From: wolfgang.grafen at gmx.de (Wolfgang Grafen)
Date: Thu, 28 Feb 2002 02:19:56 +0100
Subject: Regular expression problem
References: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com>
Message-ID: <3C7D85BC.71550E5F@gmx.de>

import re

rc=re.compile("<@Trap Body text\s*"
                          "(?:(?P<assigned>=)|(?P<unassigned>>))\s*"
                          "(?P<rest>.*?)\s*\Z",
                         re.MULTILINE|re.DOTALL).match

t1='<@Trap Body text>'
t2='<@Trap Body text=<FONT "Times">'
t3="""<@Trap Body text=<FONT "Times"><CCOLOR\n   "Black"><SIZE 11><HORIZONTAL
100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n  58.3><C-POSITION 33.3><C+POSITION
33.3><P><CBASELINE 0><CNOBREAK 0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST
19.2><G+BEFORE 0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
"ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n  1><GTABS $><GHYPHENATION
2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0 25>>"""

rc(t1).groups()
(None, '>', '')

rc(t2).groups()
('=', None, '<FONT "Times">')

rc(t3).groups()
('=', None, '<FONT "Times"><CCOLOR\n   "Black"><SIZE 11><HORIZONTAL 100><LETTERSPACE
0><CTRACK 127><CSSIZE 70><C+SIZE\n  58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE
0><CNOBREAK 0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE
0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G& "ENGLISH"><GPAIRS
4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n  1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE
75 100 150><GSPACE -5 0 25>>')

cheers

wolfgang


Asheesh Laroia schrieb:

> I have some SGML input (PageMaker 6.5 tagged text), and I want to be able
> to recognize (and delete) a tag.  That tag looks like:
>
>         <@Trap Body text:>
>
> It may also look like <@Trap Body text: useless-data>.
>
> So, I tried the regular expression r"<@.?>".  That doesn't match the
> above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
> using, and why doesn't this work?
>
> Thanks in advance!
>
> -- Asheesh Laroia.
>
> PS: An example of the tag "in the wild" is the following string:
>
> <@Trap Body text=<FONT "Times"><CCOLOR
>  "Black"><SIZE 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE
> 58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK 0><CLEADING -0.05
> ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER 0><GALIGNMENT "justify
> "><GMETHOD "proportional"><G& "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW 1><GKORPHAN
> 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0 25>>



From tjreedy at home.com  Wed Feb 20 11:28:49 2002
From: tjreedy at home.com (Terry Reedy)
Date: Wed, 20 Feb 2002 16:28:49 GMT
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <59Qc8.55794$X64.20056048@news1.rdc2.pa.home.com>

"Russell Nelson" <nelson at crynwr.com> wrote in message
news:m2vgcs5hgf.fsf at desk.crynwr.com...
> There are two Python ideoms that bother me:
>
>     while 1:
>         line = file.readline()
>         if not line: break
>
> and:
>
>     matched = re.match(r'egexp(.*)', line)
>     if matched:
>         print matched.group(1)
>
> I'd really, really like it if you could do one(1) assignment within
> the boolean expression in a while and an if.  Like this:
>
>
>     while line = file.readline():
>
> and:
>
>     if matched = re.match(r'egexp(.*)', line):
>         print matched.group(1)
>
> I just think it would be easier to read programs if they could be
> written like this.  Should I write up a PEP or has this already been
> discussed and dismissed?

Both have been discussed (the former ad nauseum) and dismissed for at
least the present.  You could certainly find some thru groups.google.
Don't know if there is a PEP documenting any of the discussion.

Terry J. Reedy





From mikeb at mitre.org  Fri Feb 22 12:13:34 2002
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 22 Feb 2002 12:13:34 -0500
Subject: Can Python COM execute a Microsoft Office macro?
Message-ID: <3C767C3E.5CFED430@mitre.org>

Is there a way for the Python COM interface to execute a macro in a Microsoft Office product like Project, Word, or Excel?




From janez.jere at void.si  Mon Feb 25 13:02:44 2002
From: janez.jere at void.si (Janez Jere)
Date: Mon, 25 Feb 2002 19:02:44 +0100
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com> <a5d95n$6g35h$1@ID-11957.news.dfncis.de>
Message-ID: <3C7A7C44.9000804@void.si>

> 
> 
> Define slow.  I ran some tests on various sizes, etc, and it seems OK.
> At a certain size, os.write() appears to be blocked, and never returns.
> That's slow... ;-)  Are you hitting something like that?  If not, look
> into the profiler to see where your bottleneck is at.

os.write blocks because pipe buffer is full and there is no reader 
thread to empty the buffer. I just want to learn how to write more 
efficient python code. There is no known bottleneck here, because the 
whole application is far from being complete.

Regards, Janez



From joshm at taconic.net  Sun Feb 10 16:20:06 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sun, 10 Feb 2002 16:20:06 -0500
Subject: A newbie question about class
References: <mailman.1013363302.15955.python-list@python.org>
Message-ID: <3c66e1b6_4@corp.newsgroups.com>

> > Could you explain a little more why "is not" is better than "!="?
>
> The short answer is that I don't think comparing something to None by
> identity will ever raise an error.  Comparing it by magnitude might.
>
> Python allows you to override magnitude comparison methods:  ==, !=, <, >,
> <=, >=.  If you are using code written by somebody else, you don't know
> whether they've accounted for comparison to None in their comparison
> methods.  If you compare to None by identity, you bypass that whole issue
> because comparison by identity ("is" and "is not") happens via id() and
> that's not something you can override (as far as I know).

And yet, unless you do use things like "!= None", bugs in packages like
FixedPoint would propogate.  It's a double edged sword.  Being able to use
"is not None" is a good way to be able to work around a bug like this (if
you don't have access to the source, for example), but I would hesitate to
say that it is "better".

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From sanjay at jarna.com  Fri Feb 15 03:41:36 2002
From: sanjay at jarna.com (Sanjay)
Date: 15 Feb 2002 00:41:36 -0800
Subject: PythonCOM arbritrary vtable based interfaces
Message-ID: <9d4c7023.0202150041.34dc5e6f@posting.google.com>

I saw a thread last september stating that this limitation may soon be
lifted.  Does the latest release of PythonCOM support this feature? 
If not, are there any workarounds?  I also saw the suggestion that
creating your own pyd to wrap the custom interface would be a good
alternative.  Is this still the best way to go?

Thanks,

Sanjay


From emile at fenx.com  Thu Feb 28 14:06:21 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 11:06:21 -0800
Subject: writing efficiently a numpy array
References: <MPG.16e755abd178b85e989682@news.wanadoo.fr>
Message-ID: <a5lvdh$7qhtk$1@ID-11957.news.dfncis.de>

"Stephan" <stephan.schumacher at andra.fr> wrote in message
news:MPG.16e755abd178b85e989682 at news.wanadoo.fr...
>
> Hi,
>
> I want to write a large numpy array on an ascii file. There must be N
> numbers per line (except for the last line...).
>
> Does anyone knows something more efficient than the function below :
>


Take advantage of Numeric's output formatting, using:

<untested>
for ii in range(0,len(a),N):
    file.write(str(a[ii:ii+N])[1:-1]+"\n")

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From dalke at dalkescientific.com  Sat Feb  2 17:47:05 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Sat, 2 Feb 2002 15:47:05 -0700
Subject: Standing Objects ... possible?
References: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>
Message-ID: <a3hq13$2e4$1@slb7.atl.mindspring.net>

Philip Swartzleonard:
>What i want is a construct like:
>
>object Foo( ancestor ):
>    pass
...
>    to be more or less equivelent to:
>
>class _Foo( ancestor ):
>    pass
>Foo = _Foo()

There's no reason you need a new name.  Try this:

  class Spam(ancestor):
    pass
  Spam = Spam()

(where I used the more Python appropriate metasyntactic
variable :)

                Andrew
                dalke at dalkescientific.com






From mbanfi at gmx.net  Wed Feb 27 03:06:06 2002
From: mbanfi at gmx.net (Markus Banfi)
Date: Wed, 27 Feb 2002 08:06:06 GMT
Subject: Python GUI
References: <uZ_e8.24022$dj3.835926@typhoon.austin.rr.com>
Message-ID: <MPG.16e6b1eec8fe2c9989684@news.chello.at>

Hi,

I'm pretty new to Python, too. Starting learning wxPython/BoaConstructur 
now.

Take a look at:

*)
BoaConstructor (http://boa-constructor.sourceforge.net/).
It depends on wxWindows (wxGTK/wxMotif) (http://www.wxwindows.org/) and 
wxPython (http://wxpython.org/).

*)
Glade (http://glade.gnome.org/) with
Glade Python Code Generator (http://glc.sourceforge.net/) and
PyGTK (http://www.daa.com.au/~james/pygtk/) and
gladepyc (http://www.fcoutant.freesurf.fr/gladepyc.html)

*)
Article "Charming Python: A review of Python IDEs" at IBM's DeveloperWorks 
site (http://www-106.ibm.com/developerworks/linux/library/l-
pide/index.html?dwzone=linux)


Best regards,

Markus



In article <uZ_e8.24022$dj3.835926 at typhoon.austin.rr.com>, 
Moo at houston.rr.com says...
> New to python (still.. I made a post last week with my ridiculous error
> about using () for indexing in a dictinoary instead of [], ah well, its the
> easy mistakes that get us. anyway) I was curious if there were some type of
> gui builder or somesuch for python, alla Borland Builder ? This is a
> delightful language,everything seems so easy for it. as a friend pointed out
> "yeah ,but you can do that in c++ as well.." and i replied "yeah, but its
> not this easy! =)"
> thanks guys
> 
> ~Avi
> 
> 
> 


From phr-n2002a at nightsong.com  Mon Feb 25 22:57:10 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 19:57:10 -0800
Subject: Proposal: local variable declarations and type advice
Message-ID: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>

I thought I'd toss out the following two ideas:

1) add a local declaration, similar in syntax to Python's existing global
   declaration, but with meaning more like a local declaration in C or
   Perl.  It declares a variable for the scope of the current indent level:

     x = 2
     if mumble:
        local x
        x = 4       # inner scope
        frob(x)
     print x        # prints 2

   Any local declaration goes out of scope at the end of its indent level:

     # say there's no x in the current scope
     if mumble:
       local x
       x = 4
       frob(x)
     print x     # raises NameError

   Local can also be used in loop indexes:

     x = 3
     for local x in range(10):
        blah(x)
     print x    # still prints 3

     x = 3
     odds = [local x for x in range(10) if x%2 != 0]
     print x    # prints 3

   The list comprehension example should fix the persistent FAQ about
   why list comprehensions leave residue (i.e. variable bindings) around.
   It gives a way to clean up the residue, using a local declaration.
   It's similar to C++'s "for (int i = 0; i < 10; i++)" or perl's
   "foreach my $i { ...}".
   
2) Add a compiler option similar to Perl's "use strict", causing the
   compiler to flag with a warning any use of an undeclared variable
   (variables must be either declared local or declared global).  This
   should get rid of misspelled variable-name bugs, and some other
   bugs as well.  I constantly want something like this.

3) (Separate proposal from above): add optional type advice:

     local x(foo)     # advises Python implementation that x will
                      # only be bound to instances of class foo

   This is called "advice" rather than a "declaration" because it has
   no defined semantics and the compiler is free to ignore it, except
   that if the program acts as the advice says it will, it will work.

   Example 1:
     local x(int)    # advise Python that x is an int
     x = 3 + 2       # guaranteed to set x = 5 and work as you'd expect

   Example 2:
     local x(int)
     x = "banana"    # implementation-dependent behavior

   Some ways Python can handle example 2:

   1a) Compiler does enough static analysis to flag the "banana" assignment
      as a compile time error

   1b) Compiler doesn't flag; interpreter raises runtime TypeError exception.

   2) Compiler doesn't flag; interpreter assigns "banana", i.e. the
      advice is completely ignored and does nothing, the program works
      exactly the same way it does now without the advice

   3) Compiler doesn't flag, interpreter has totally undefined behavior.
      It can work normally, or crash, or delete all your files, etc.
      The idea is that an optimizing compiler can believe the advice and
      generate code based on x always being an int (it can put it in a
      register, perform native machine code arithmetic on it etc).

The idea is that you'd choose between 1a/1b and 3 as a compile time
option--you'd choose 1a/1b for debugging or 3 for speed.  Normally you
wouldn't choose 2--you wouldn't put the type advice into your program
unless you wanted it to mean something--so 2 is really just a special
case of 3 (i.e. it's one possible thing the implementation can do
without warning you).

Python has grown from a tiny acorn to a healthy young tree.  I believe
any changes and additions to its design should be made with the idea
that it may become a mighty oak one day, i.e. that it will have serious
native-code optimizing compilers.   Language design changes should work
toward that goal, or at least not impede it.


From chrishbarker at attbi.com  Fri Feb 15 16:21:31 2002
From: chrishbarker at attbi.com (Chris Barker)
Date: Fri, 15 Feb 2002 13:21:31 -0800
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
Message-ID: <3C6D7BDB.BFD2F5DF@attbi.com>

Artur Skura wrote:

> In other words: I have to put text entries between text labels.
> Now can you suggest an easy, lazy way of doing this (with any
> toolkit) so things don't pile up and everything looks nice?

wxPython's sizers should make this a piece of cake.

-Chris


-- 
Christopher Barker,
Ph.D.                                                           
ChrisHBarker at attbi.net                ---           ---           ---
                                     ---@@       -----@@       -----@@
                                   ------@@@     ------@@@     ------@@@
Oil Spill Modeling                ------   @    ------   @   ------   @
Water Resources Engineering       -------      ---------     --------    
Coastal and Fluvial Hydrodynamics --------------------------------------
------------------------------------------------------------------------


From phr-n2002a at nightsong.com  Mon Feb 11 08:32:05 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 05:32:05 -0800
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com> <69c619074b.tim@worthy.demon.co.uk> <7xlme0fcku.fsf@ruckus.brouhaha.com> <b82222074b.tim@worthy.demon.co.uk>
Message-ID: <7xheoo9may.fsf@ruckus.brouhaha.com>

Tim Howarth <tim at worthy.demon.co.uk> writes:
> > That's the result I get in Python 2.2.  I don't have 2.1 on this system.
> > 1.5.2 raises an exception when it tries to convert 0xFFFFFFFFL to hex.
> 
> Stupid me, yes so it does - I think I might have tried using 8FFFFFFF
> for some reason !

Basically I think the int function should not barf if you give it a
string arg like FFFFFFFF with base 16.  Failing that, maybe there could
be some way of telling it not to barf, because you want an unsigned result.


From tdelaney at avaya.com  Tue Feb 12 18:03:01 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 13 Feb 2002 10:03:01 +1100
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A17E@natasha.auslabs.avaya.com>

> From: brian_l at yahoo.com [mailto:brian_l at yahoo.com]
> "Steve Holden" <sholden at holdenweb.com> wrote in message 
> news:<ZND98.140690$KM2.6229113 at atlpnn01.usenetserver.com>...
> > "Brian Lenihan" <brian_l at yahoo.com> wrote in message
> > news:afe3b146.0202100036.f221c9a at posting.google.com...
> 
> > > I don't throw like a girl right-handed, so maybe I am actually
> > > ambidextrous.
> > >
> > Not to mention gender-biased? :-)
> 
> Of course I'm gender biased.  I love girls.  I'm also sure there are
> plenty of girls who can throw better than I can.  I left out the fact
> that I'm also a switch-hitter because I thought it would only add to
> the confusion.

For those of use who have no interest in (I presume) baseball, what the hell
is a switch-hitter?

Tim Delaney



From seeger at sitewaerts.de  Mon Feb 25 04:30:25 2002
From: seeger at sitewaerts.de (Felix Seeger)
Date: Mon, 25 Feb 2002 10:30:25 +0100
Subject: problem with exception output
Message-ID: <a5d01k$69es1$1@ID-79019.news.dfncis.de>

Hi

I can't get the hole exception if my script (cgi) fails.
I use this to get every exception:

except:
        print '<table width="100%">'
        trace = traceback.extract_tb(sys.exc_traceback)

        print '<tr><td>Exception Type:</td><td>', sys.exc_type, '</td></tr>'
        print '<tr><td>Exception Value:</td><td>', sys.exc_value, '</td></tr>'
        for filename, lineno, function, message in trace:
                print '<tr><td>Filename:</td><td>', filename, '</td></tr>'
                print '<tr><td>Error message:</td><td>', message, '</td></tr>'
                print '<tr><td>Line:</td><td>', lineno, '</td></tr>'
                print '<tr><td>Function:</td><td>', function, '</td></tr>'

        print '<tr><td>Complete Stack:</td></tr>'
        print '<tr><td><hr></td></tr>'
        print '<tr><td colspan="2">', traceback.print_stack(), '</td></tr>'
        print '<tr><td><hr></td></tr>'

        print '<tr><td>Complete Traceback:</td></tr>'
        print '<tr><td><hr></td></tr>'
        print '<tr><td colspan="2">', traceback.print_exc(), '</td></tr>'
        print '<tr><td><hr></td></tr>'
        print '</table>'

Everything works fine but the print_stack() and print_exc() prints None.
On the console I get this:

File "/usr/local/viewcvs-0.9.2/cgi/update.cgi", line 142, in ?
    print '<tr><td colspan="2">', traceback.print_stack(), '</td></tr>'
Traceback (innermost last):
  File "/usr/local/viewcvs-0.9.2/cgi/update.cgi", line 123, in ?
    1/0
ZeroDivisionError: integer division or modulo



What is wrong ? ;)

thanks
have fun
Felix



From gcordova at hebmex.com  Mon Feb 25 18:05:03 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Mon, 25 Feb 2002 17:05:03 -0600
Subject: help: Problem with cgi form
Message-ID: <19650FF08A99D511B19E00B0D0F06323060B03@SRM201002>

> 
> In article <3C7AAF31.8040201 at mxm.dk>, Max M wrote:
> > And then print it to the form like:
> > 
> > print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % 
> last_byte_pos
> 
> Beware that, in general, that is a bug. You are open to Cross Site
> Scripting attacks. Even in the absence of these, if the value happens
> to contain characters significant to HTML (such as '"', '&', '>', etc)
> then your program will go wrong.
> 

Yes.

AS A RULE, always to AT LEAST these substitutions:

def SafeHtmlSubstitutions(txt):
	BasicEntities = [ 
	  ("&","amp"),
	  ("<","lt"),
	  (">","gt"),
	  ('"',"quot") ]
	for ch,ent in BasicEntities:
	    txt = txt.replace(ch,"&%s;" % ent)
	return txt

That way, the text will be opaque to HTML.

Salutations.

-gustavo



From krnylund at msn.com  Mon Feb 25 03:20:42 2002
From: krnylund at msn.com (Kristian Nylund)
Date: 25 Feb 2002 00:20:42 -0800
Subject: GUI (PyQt) & thread problem
References: <8b5d764c.0202240451.1c4bda3b@posting.google.com> <a5c8a2$r5e$1@news.tudelft.nl>
Message-ID: <8b5d764c.0202250020.268f4a81@posting.google.com>

"K. Wojas" <usenet at wojas.vvtp.tudelft.nl> wrote in message news:<a5c8a2$r5e$1 at news.tudelft.nl>...
> Kristian Nylund wrote:
> > Can someone tell me what the problem is with the following program, as it
> > hangs. I'm pretty sure there's some inherent design-flaw involved, but
> > I'm not familiar enough with threads or GUI-programming to know what it
> > could be.
> > 
> > I've also tried modyfing the program to use PYSIGNALs instead of passing
> > the caller to the thread, but the results were the same.
> 
> Never use the Qt toolkit from 2 threads. You can only use the worker thread 
> for non-GUI stuff. The Qt toolkit isn't threadsafe.
> 
> Regards,
> 
> Konrad Wojas

Thank you for your reply. Do you have any pointers on how to implement
this kind of functionality, i.e. when something happens in the worker
thread (not GUI related), the GUI needs to be updated. Is it at all
possible?

Thanks,

Kristian


From peter at engcorp.com  Mon Feb 25 19:33:53 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 25 Feb 2002 19:33:53 -0500
Subject: Threads in python?
References: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com>
Message-ID: <3C7AD7F1.AFCADA42@engcorp.com>

"Fernando Rodr?guez" wrote:
> 
> I'm going to start coding a little utility that needs to retrieve webpages and
> process them.
> 
> However, I worried that the whole app might stall if one server takes too long
> to respond or is offline.
> 
> Does Python support threads? "Learning Python" doesn't mention it...

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

(Yes, Python supports threads.  There may, however, be simpler and/or
more robust approaches to what you are doing, although I personally
would just jump on threads since I grok them better than, for example,
asyncore.)

-Peter


From jason at jorendorff.com  Tue Feb 26 23:21:35 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 22:21:35 -0600
Subject: Parse variable into variables >Re: Apology
In-Reply-To: <1121.822T529T14383507threeseas@earthlink.net>
Message-ID: <HFEKILOLEFEFMKAECNDLOEJGDFAA.jason@jorendorff.com>

Timothy Rue wrote:
> vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = varset.split()
>
> is there a way to add the readline() into this line?
> ..... = testfile.readline().split()

All of this is correct.  You might even want to just do:

  vic['AI'] = testfile.readline().split()

but of course that makes vic['AI'] a list of strings,
not a dictionary.  Which may or may not be okay.

## Jason Orendorff    http://www.jorendorff.com/





From lsmithso at NOhare.SPAM.demon.co.uk  Fri Feb  8 16:17:08 2002
From: lsmithso at NOhare.SPAM.demon.co.uk (Les Smithson)
Date: 08 Feb 2002 21:17:08 +0000
Subject: ANN: pymqi Version 0.2
Message-ID: <a41fsr$bvf$2@newsreader.mailgate.org>

Pymqi is a Python extension for IBM's Messaging & Queueing middleware,
MQSeries. This allows Python scripts to make calls directly to MQI to
connect queues and get/put messages on them etc.

Version 0.2 is available from  http://www.hare.demon.co.uk/pymqi/pymqi.html.

New in 0.2:
. Fully threadable & thread safe.
. Added the MQI verbs MQCONNX, MQPUT1, MQCMIT & MQBACK.
. MQMIError can print string mnemonic for all MQI error codes.
. Built against Python 2.1.2 & MQ  5.2.



From bokr at oz.net  Tue Feb 19 18:22:23 2002
From: bokr at oz.net (Bengt Richter)
Date: 19 Feb 2002 23:22:23 GMT
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com> <a4rsa2$2grrk$1@ID-59885.news.dfncis.de> <a4smoa$h20$0@216.39.172.122> <a4u7ec$36637$1@ID-59885.news.dfncis.de>
Message-ID: <a4umnf$59r$0@216.39.172.122>

On Tue, 19 Feb 2002 20:01:04 +0100, "Thomas Heller" <theller at python.net> wrote:

>
>"Bengt Richter" <bokr at oz.net> wrote in message news:a4smoa$h20$0 at 216.39.172.122...
>> >The source code IS available:
>> >http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/distutils/
>> >in the misc directory. Guess I'll have to add a note about this
>> >to the source... There is a tricky symbolic link somewhere on SF
>> >which links parts of the distutils tree into the Python source tree.
>>
>> That self-update of the bin64 EXEDATA string seems a little tricky too ;-)
>> I guess I would rename the old version to a backup name before writing
>> the new, not being aware of the larger context. Or should one
>> copy it to ../otherdir and run ../otherdir/it that way? Or did I read
>> too fast?
>
>CVS gives enough backup for me, and it has worked for me so far.
>
I'd have guessed so for you, and, ok, I guess anyone having a seriously modified
version would not run it without knowing what it was going to do first ;-))
>>
>> >
>> >> Having some additional hooks created in distutils to execute
>> >> post-installation sounds like it could be a useful enhancement.  At
>> >> least with that sort of scenario you could write your own shortcut
>> >> creating code without modifying the core distutils itself.
>> >
>> >I'm currently working on an bdist_wininst enhancement which
>> >takes an additional command line option (--install-script <script.py>).
>> >This script would then be run (with sys.argv set to [script_name, '--installed'])
>> >after all the files have been copied, and it would again be run
>> >(with sys.argv set to [script_name, '--removed']) just before
>> >the deinstaller is about to remove all the files and directories.
>> >
>> Sounds good. I wonder if something is needed to cater to version upgrading
>> and/or multiple version installation. E.g., to give an old version a
>> chance to have a script automatically executed '--before_upgrade' and
>> '--after_upgrade'? Can the binary installer run something before even
>> copying any new files? I.e., the old version interpreter and its
>> 'im_about_to_have_a_sibling.py' script? Putting it in the installer
>> sequence would protect against user forgetfulness (otherwise s/he could
>> obviously do it manually), and an old version could at a minimum output
>> some attention-grabbing reminders from the authors of the old version.
>
>Well, I want to keep the C-coded part as simple as possible.
>On the other hand, you could play any tricks you want in
>the installation script. Detect upgrade yourself!
>
That would work if there were a reliable convention of loading the new files
completely separately (e.g., no replacing anything in %SystemRoot%\system32)
and also a convention for finding elements of the old version -- in particular,
a place for a pre-upgrade script. The new install script could then
execute the pre-upgrade script and the old version could have its say
about being replaced.

Note that the pre-upgrade file could be modified incrementally (by hand
or potentially sometimes automatically) when special packages were
installed, so the upgrader could be reminded to check whether s/he has
e.g. a new version of PIL or such to go with the new interpreter.

Without complete version storage separation, at the opportunity point
(*after* copying files) you propose, it may be too late to warn
not to copy over something. I hate installation programs that arrogantly
proceed to replace things without warning, but many do. Not to mention
grabbing extension and file type associations and tool bar menus. Grr...

Regards,
Bengt Richter




From phr-n2002a at nightsong.com  Tue Feb 19 21:15:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 18:15:51 -0800
Subject: add-a-gram
References: <e9e6dd1f.0202191634.21da06b9@posting.google.com>
Message-ID: <7xpu30ew4o.fsf@ruckus.brouhaha.com>

caljason76 at yahoo.com (Jason) writes:
> Here is a puzzle that was floating around comp.lang.lisp, and I was
> wondering what a Python solution would be.  I am still learning Python
> and am very curious how Python solutions differ from other languages,
> like Perl or C.  Are there some set of maxims to the Python
> programmer, and are there Python idioms for problem solutions?
> 
> Appended to the message is the problem, you can find the problems and
> word list http://www.itasoftware.com/careers/programmers.php

I did both of those problems (add-a-grams and nine 9's) in Python but
I don't feel it's appropriate to publish the solutions right now.  If
you want to see my solutions I'll send them to you by email.


From martin.franklin at westerngeco.com  Wed Feb 20 10:08:22 2002
From: martin.franklin at westerngeco.com (Martin Franklin)
Date: Wed, 20 Feb 2002 15:08:22 +0000
Subject: Button changing its own grid options
References: <3C73B3A5.60DBCBA9@pop.ntlworld.com>
Message-ID: <a50e3i$104e$1@mail1.wg.waii.com>

a.clarke11 wrote:

> Hi
> Does anyone know if it is possible to allow a Tkinter button, when
> pressed, to change its own grid options?
> I've tried various command callbacks, but I'm having problems
> referencing the buttons own grid info.
> The idea is to get a button to move around, in a grid of other buttons,
> when pressed. Any ideas or even better, code, would be appreciated.
> Thanks in advance
> Tony
> 
> 
from Tkinter import *
root=Tk()
bdict={}
def moveme(b):
    old_row, old_col=bdict[b]
    ## change them 
    row=old_row+1
    col=old_col+1
    b.grid_forget()
    b.grid(row=row, col=col)
    ## update dict to reflect the change
    bdict[b]=(row, col)
for row in range(2):
    for col in range(2):
        b=Button(root)
        b.grid(row=row, col=col)
        b.config(command=lambda b=b: moveme(b))
        bdict[b]=(row, col)
root.mainloop()


From ahimsa at inext.co.za  Sat Feb  9 11:05:35 2002
From: ahimsa at inext.co.za (ahimsa consulting)
Date: Sat, 9 Feb 2002 18:05:35 +0200
Subject: TRYING to like python, but...
References: <E16ZJGR-0004Kj-00@mail.python.org>
Message-ID: <00c101c1b184$674ec720$0100000a@ahimsa>

Hi Jim

I would suggest that you try the 2.2 build. I've used it for a while now and
although I am not an experienced/advanced programmer I have not come across
any of the problems you describe. Also the IDEStudio is a nice GUI if you
wanted one.

HTH

Andrew


> Message: 9
> From: "jimh" <jim at jim.jim>
> Subject: TRYING to like python, but...
> Date: Fri, 8 Feb 2002 13:46:56 -0800
> Organization: SSO-IT, Hewlett-Packard Co.
> To: python-list at python.org
>
> I untarred, configured and built python 2.1.2.  It all seems to be working
> fine.
>
> I wanted to try out the xml.dom stuff, so I looked over the Python Library
> Reference sections on xml.dom and tried one of the examples.  This is what
> happened:
>
>    xml.sax._exceptions.SAXReaderNotAvailable: No parsers found
>
> Well, THAT'S helpful.
>
> Nothing in Setup about xml.dom or xml.sax, so I assume that Setup is ok.
> Besides, I WAS able to import and start using xml.dom.
>
> So I get online and start searching around for this error message.  Find
> some vague references to needing "expat" (whatever that is).  Ok, I
remember
> vaguely something about expat in Setup, so I go back to Setup and see that
> TO USE A STANDARD PYTHON LIBRARY, I need to DOWNLOAD ADDITIONAL code
(expat)
> from SOME PLACE ELSE.
>
> So I download it, compile it, update EXPAT_DIR in Setup to point to it
(just
> like the comments in Setup say).
>
> So I ./configure again, and Make again, and, guess what??? IT STILL
DOESN'T
> WORK!!!  (xmlparse.h: no such file).
>
> so PLEASE PLEASE PLEASE tell me that I don't really have to go through ALL
> OF THIS CONFUSING MESS just to get a STANDARD python library to work.
> (xml.dom IS in the "Python Libary Reference", but there is NO mention of
all
> these additional steps required to actually USE it).
>
> To be honest, if python is really this difficult to use, I am going to try
> something different.
>
> Someone please help me out here.
>
> Jim Haffey





From wolfson at uchicago.edu  Fri Feb 15 11:39:48 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Fri, 15 Feb 2002 10:39:48 -0600
Subject: Is Stackless Python DEAD?
References: <184fbd02.0110262039.74799675@posting.google.com> <j4d72xasx0.fsf@informatik.hu-berlin.de> <uady09pkm.fsf@python.net> <mailman.1009806401.11232.python-list@python.org> <110A0579A87BA846.8CEE70EF24F483BD.740CFC77F4D54FA3@lp.airnews.net> <mailman.1013761643.13888.python-list@python.org>
Message-ID: <pan.2002.02.15.10.39.47.229964.1126@uchicago.edu>

On Fri, 15 Feb 2002 02:27:03 -0600, Christian Tismer wrote:

> The new solution is cool and uncool at the same time.
> All my difficult changes to the core have been abandoned.
> All magic about Stackless' implementation is gone (sigh).
> It is now orthogonal to the Python engine. 

Does this mean that the changes that made Stackless faster than CPython
are gone, too?

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From eric.brunel at pragmadev.com  Mon Feb 11 06:17:53 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 11 Feb 2002 12:17:53 +0100
Subject: Beginner problem.
References: <3c67a6a3$0$12223$4d4efb8e@news.be.uu.net>
Message-ID: <a48927$8r9$1@norfair.nerim.net>

Hi,

"DelPiccolo Ivano" <ivano.delpiccolo at degroof.be> a ?crit dans le message
news: 3c67a6a3$0$12223$4d4efb8e at news.be.uu.net...
> Here's my script, where is the error ?
>
> import sys
>
> if sys.argv[??????] == "" :
>     print "Usage : sys.argv[0] logname string_to_search"
> else :

You're not in a shell script: all arguments are not implicitely set to "".
To test how much arguments are passed, just test the length of the sys.argv
list. E.g.:

import sys
if len(sys.argv) != 3:
  print "Usage: %s logname string_to_search" % sys.argv[0]

It's as simple as that... HTH
 - eric -




From martin at v.loewis.de  Sat Feb 23 18:20:33 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 00:20:33 +0100
Subject: Object counts
References: <ZQTd8.89849$B61.28524193@news1.rsm1.occa.home.com>
Message-ID: <m34rk7pyym.fsf@mira.informatik.hu-berlin.de>

Chuck Esterbrook <ChuckEsterbrook at yahoo.com> writes:

> Is there a way to get Python to report the number of instances of all 
> classes and types? e.g., a dictionary that would map the names of 
> types/classes to their count, which would be incremented on creation 
> and decremented on deletion?

If you compile Python with --with-pydebug, a list of all objects is
maintained at all times. It would be possible to count them on a
per-type basis using this list.

Alternatively, if you are only interested in container objects, you
can use gc.get_objects. From those, you can actually create a list of
all reachable Python objects (including finalizer-reachable
ones). That won't give you objects which are only reachable from C
pointers, or which are garbage due to reference counting bugs.

Regards,
Martin


From ullrich at math.okstate.edu  Mon Feb 25 15:34:17 2002
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 25 Feb 2002 20:34:17 GMT
Subject: Puzzled: Y am I ending up with extra bytes?
References: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4> <9ipj7ucf1tvktalhjecg44jn0172855k7m@4ax.com>
Message-ID: <3c7a9f01.273804295@news>

On Sun, 24 Feb 2002 23:26:33 -0800, Tim Roberts <timr at probo.com>
wrote:

>"A.Newby" <deathtospam43423 at altavista.com> wrote:
>
>>Why is this happening? I read large chunks of data from a text file, 
>>according to byte locations specified on another file, and for some reason, 
>>this function (below), spits out a few extra bytes.
>>
>>Here's the code, as entered into the Python shell...... 
>>
>>    	index = map(string.rstrip, open('D:\cgi-bin\indx.txt').readlines())
>
>Danger, Will Robinson!  Although this works by accident, it would NOT have
>worked if your file had been "ndx.txt" instead of "indx.txt", beause the
>"\n" would have been interpreted as a linefeed (0x0A).
>
>When you need to refer to a file name in windows, you must either:
>1. Use forward slashes:   open('D:/cgi-bin/indx.txt')
>2. Use double backslashes:  open('D:\\cgi-bin\\indx.txt')
>3. Use a "backslash suppressed" string:  open(r'D:\cgi-bin\indx.txt')
>   (note the r' prefix )
>
>Personally, I recommend option 1.  All Windows APIs accept forward slashes
>in filenames; only the command shells require backslashes.

If we were only giving filenames to the system and never getting
filenames from the system right. But if, say, you have a list a 
filenames you got from Windows they're going to have backslashes
in them, and if, say, you try to filter out the ones containing
'/somedirectory/' instead of the ones containing '\somedirectory\'
it's not going to work; seems like at some point we're going to
have to deal with backslashes in strings regardless (so I always
use backslashes, else I have no chance getting things right.)

>--
>- Tim Roberts, timr at probo.com
>  Providenza & Boekelheide, Inc.


David C. Ullrich


From cmaloney at physics.ucsb.edu  Wed Feb 27 11:20:04 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: 27 Feb 2002 08:20:04 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com> <3C7BCCA5.604@physics.ucsb.edu> <a5h9qn$a3h$1@bob.news.rcn.net>
Message-ID: <6bf54244.0202270820.7fa4baf2@posting.google.com>

"David Abrahams" <david.abrahams at rcn.com> wrote in message news:<a5h9qn$a3h$1 at bob.news.rcn.net>...
> "Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
> news:3C7BCCA5.604 at physics.ucsb.edu...
> 
> Another promising but incomplete entry, last time I looked (it still
> couldn't represent the full C++ language in its XML). The mailing list
> archives indicate that's still the case, though things may be changing:
> http://public.kitware.com/pipermail/gccxml/2002-February/000029.html
> 
> > This is what I would have guessed would be the "right" way to go about
> > things (having known nothing about the EDG front-end).  Presumably gcc
> > gets the parsing right if it can compile ;)
> 
> Even the EDG parser has bugs. However, if your code compiles with GCC 3.0,
> gccxml should be able to handle it (note that GCC is up to 3.0.3 and 3.1
> will be released soon, so things have improved further since gccxml was
> developed).
> 
> -Dave

My apologies to the group for my ignorance in compiler issues.
It was unclear to me before that the EDG front end *was* used
in commercial compilers.  *grins sheepishly* I had assumed that
it was a tool constructed for independent reasons (e.g. writing
automatic wrapper generators).

This is what I meant when I said that gccxml was the "right" way 
to do the wrapper generation.  Now that my misconceptions have 
been cleared up, I retract my statement about gccxml being the
"right way" as opposed to the PDT's use of the EDG front end.

As Dave points out, since I'm compiling with gcc anyway, and not,
say KAI or Intel, then I'm at a *slight* advantage using gccxml,
all other things being equal.  But now it seems more like a wash 
to me.

Also -- does anyone have any news or thoughts about gccxml or
something else like it being incorporated into the gcc proper?
It seems that since gcc tries to be a "compiler collection" it might
be nice to include things like automatic wrapper generation for
multi-language support?  Is this on the horizon?

Cheers,
Craig


From tjreedy at home.com  Fri Feb  8 13:26:29 2002
From: tjreedy at home.com (Terry Reedy)
Date: Fri, 08 Feb 2002 18:26:29 GMT
Subject: Smarter header (was Re: Stupid Question)
References: <mailman.1013184443.12158.python-list@python.org>
Message-ID: <pLU88.33090$X64.11050005@news1.rdc2.pa.home.com>

"Andreas Sch?ller" <schoeller at zkm.de> wrote in message
news:mailman.1013184443.12158.python-list at python.org...
>is it possible to delete a key from a Btree() ? I?m using a IOBTree
from
>the ZODB Standalone dist. A simple del tree[key] gives me a
__delitem__
>error ? Reassign works as expected tree[key] = value.

You are much more likely to get a smart answer with a better header,
especially for such a specialized question.

Example: ZODB IOBTree deletion question

Terry J. Reedy







From gaul at spam.utexas.edu  Tue Feb 12 07:03:25 2002
From: gaul at spam.utexas.edu (Andrew Gaul)
Date: Tue, 12 Feb 2002 12:03:25 +0000 (UTC)
Subject: Fun with lambda and map
References: <3c566a21.22129828@news.t-online.de>
Message-ID: <slrna6i14c.sp.gaul@bifrost.cs.utexas.edu>

In article <3c566a21.22129828 at news.t-online.de>, Gerson Kurz wrote:
> primes = lambda o:map(lambda a:filter(None,(map(lambda i:map(lambda x:a.__setitem__(x,0),range(2*i,o,i)),range(2,o)),a)[1])[1:],[range(o)])[0]
> 
> fibonacci = lambda x:map(lambda o:(map(lambda c:map(lambda l:o.__setslice__(l[0],l[1],l[2]),([o[2]+3,o[2]+4,[o[0]]],[0,3,[o[1],reduce(lambda x,o:x+o,o[:2]),o[2]+1]])),range(x)),o)[1],[[1,1,0]+range(x)])[0][3:]
> 
> (Note: These probably need python 2.2 to work because of the lambda
> scope thing)

Here's a shorter implementation that works with both 1.5 and 2.2 that
involves some cleverness:

    fibonacci = lambda z: map(lambda n, f=(lambda f,x,a,b: \
    [x<=0 and b, x>0 and f(f,x-1,b,a+b)][x>0]): f(f,n,0,1), range(z))

It's not as nice as the Haskell version:

    fibonacci n = take n fib
                  where fib = 1 : 1 : [ a+b | (a,b) <- zip fib (tail fib) ]

Enjoy!

-- 
Andrew Gaul
http://gaul.org/


From vodine at direct.ca  Thu Feb 21 20:35:44 2002
From: vodine at direct.ca (Virginia O)
Date: Thu, 21 Feb 2002 17:35:44 -0800
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com> <CMfd8.5037$D6.14223@news.iol.ie>
Message-ID: <u7b81629bo9o4a@corp.supernews.com>

Thanks, but the database will have to hold sub-groups etc. It will be easier
for me to enter and organize the items in a database.
I only gave the scenario as an example. The actual program will hold
hundreds of items.

Virginia


"Roy Madden" <roy at linksheaven.com> wrote in message
news:CMfd8.5037$D6.14223 at news.iol.ie...
On Thu, 21 Feb 2002 10:49:52 -0800, "Virginia O" <vodine at direct.ca> wrote:

>Build my database. (I will be using Access,unless convinced otherwise.)

If it was upto me I'd just use a simple delimited text file :)

Roy



--
http://www.linksheaven.com




From lfini at arcetri.astro.it  Fri Feb  8 16:53:58 2002
From: lfini at arcetri.astro.it (Luca Fini)
Date: Fri, 8 Feb 2002 22:53:58 +0100
Subject: [OT] PPT reader?
In-Reply-To: <mailman.1013181915.301.python-list@python.org>
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <Pine.LNX.4.44.0202082252370.6311-100000@localhost.localdomain>

On Fri, 8 Feb 2002, Skip Montanaro wrote:

> access to PPT sort of out in the cold.  What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT slides
> to other, more accessible, formats?  If we can get a reasonable (but fairly
> short) list of tools, I can probably see they get referenced on the relevant
> page of the Python website.

StarOffice is more or less available for free to individuals and manages 
PPT files pretty well.

						l.f.
-- 
--------------------------------------------------------------------------
       -- )     Luca Fini                            Tel: +39 055 2752 307
 ___    |\      Osservatorio Astrofisico di Arcetri  Fax: +39 055 2752 292
/   |   | |-_   L.go E.Fermi, 5 +-----------------------------------------
(___|___//___)  50125 Firenze  /   WWW: http://www.arcetri.astro.it/~lfini
 (_)      (_)   Italia        / e-mail: lfini at arcetri.astro.it
-----------------------------+--------------------------------------------



From alf at sagittarius.logilab.fr  Mon Feb 25 08:59:59 2002
From: alf at sagittarius.logilab.fr (Alexandre Fayolle)
Date: Mon, 25 Feb 2002 13:59:59 +0000 (UTC)
Subject: Problem with 4Dom xslt processor
References: <8f8ffe67.0202190656.4fb11dda@posting.google.com>
Message-ID: <slrna7kgr0.23j.alf@sagittarius.logilab.fr>

On 19 Feb 2002 06:56:43 -0800, Sunit Joshi <sjoshi at ingr.com> wrote:
>Hello all
>I'm trying to use the 4DOM's xslt processor I keep getting this error
>(below) when trying to append a stylesheet. Any pointers are greatly
>appreciated.

Hi, you didn't tell which version of 4Suite/PyXML you are using. 

If you're using the xslt engine that ships with PyXML 0.7, I'd 
recommend asking your question on the xml-sig (xml-sig at python.org). 

If you're using 4Suite < 0.12a, I'd recommend upgrading and if this 
doesn't solve your problem, asking your question on the 4Suite 
mailing list (4Suite at lists.fourthought.com). 

In both cases, it's better if you can enclose a small xml document
and stylesheet enabling to reproduce the problem

The traffic on c.l.p. is such that I'm not sure the 4Suite 
development team monitors it actively.

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


From sandskyfly at hotmail.com  Wed Feb 20 21:55:02 2002
From: sandskyfly at hotmail.com (Sandy Norton)
Date: 20 Feb 2002 18:55:02 -0800
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com> <a512ap$dep$1@panix2.panix.com>
Message-ID: <b03e80d.0202201855.5f1122b9@posting.google.com>

Sandy Norton:
> >Is there anyway to specify a time limit in which the operation can
> >happen and enforce it in the code?

Aahz Maruch:
> 
> http://www.timo-tasi.org/python/timeoutsocket.py

Thanks Aahz! Tried it out and it works well. (-:

Many thanks to Timothy O'Malley for writing this useful module it in
the first place.

best regards,

Sandy


PS: Appreciate all other responses as well.


From fperez528 at yahoo.com  Tue Feb 26 15:57:59 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 26 Feb 2002 20:57:59 +0000
Subject: ActiveState version of 2.2???
References: <1001ff04.0202170702.6ebaaf55@posting.google.com> <3C704157.B872722F@bellatlantic.net> <1001ff04.0202261911.22131410@posting.google.com>
Message-ID: <a5hljp$blo$1@peabody.colorado.edu>

Bob wrote:

> So there is no ActiveState version now?  Or is it just as well to get
> the regular distro + the win32all stuff?

I did that and it seems to work just fine.

f.


From tiberius at ms28.hinet.net  Thu Feb 21 05:57:09 2002
From: tiberius at ms28.hinet.net (Tiberius Teng)
Date: 21 Feb 2002 02:57:09 -0800
Subject: Interesting for Trekkers and Python Developers ... ST: Bridge Commander
Message-ID: <6c92a8bc.0202210257.702ee9d4@posting.google.com>

http://www.bridgecommander.com/

I think I don't have to describe this game much, just go to their
webpage and check out the demo file :)

It's very interesting that looks like this game uses extensive Python
scripts in its control - AI, ships, weapons, control panels, even
mission progress (story ?!). Check the 'scripts' folder after
installation and you'll find out lots of familiar filenames and folder
structures ;)

And I think it'll be very customizable ... But their .pyc files uses
different magic words in it so I can't just import it from Python
interpreter ...

Anyway just check it out!


From peter at engcorp.com  Sun Feb 10 12:32:41 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 10 Feb 2002 12:32:41 -0500
Subject: box on soap
References: <mailman.1013283562.647.python-list@python.org> <Lli98.24849$l93.4538901@newsb.telia.net>
Message-ID: <3C66AEB9.4A1DF069@engcorp.com>

Fredrik Lundh wrote:
> 
> Mark McEahern wrote:
> > Don Box on the Importance of Being WSDL
> 
> David McCusker on the Importance of Ignoring Don Box:
> 
> http://www.treedragon.com/ged/map/ti/newFeb02.htm#06feb02-box
> 
> (I'll just give you a summary. I won't argue.  David's right
> and Don's wrong ;-)

Thanks for saving us the time Fredrik.  Besides which, clicking
on Mark's link brought up (in Netscape 4.7) a page which
said "Directory Listing Denied:  This Virtual Directory does 
not allow contents to be listed."

Probably caused by some strong typing somewhere. :-)

(Yes, I could have read the text version posted with Mark's
message, but presumably the HTML version was more readable.)

-Peter


From eppstein at ics.uci.edu  Mon Feb 25 23:58:40 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Mon, 25 Feb 2002 20:58:40 -0800
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org> <m3pu2sancq.fsf@chvatal.cbbrowne.com>
Message-ID: <eppstein-42B711.20584025022002@news.service.uci.edu>

In article <m3pu2sancq.fsf at chvatal.cbbrowne.com>,
 Christopher Browne <cbbrowne at acm.org> wrote:

> There have got to be some puns heading in the direction of "seriously
> stacked."  Hmm...
> 
>  - Pulchritudinous Python?
>  - Busty Python?
> 
> There's got to be some more clever way of saying that without
> actually, well, saying that...  "Victoria's Python?"  Hmm...

Huge tracts of ... threads?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From starx at pacbell.net  Sun Feb 10 17:56:38 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 10 Feb 2002 22:56:38 GMT
Subject: importing/reloading in Pythonwin
References: <a46l7r$57e$1@news.efn.org>
Message-ID: <Xns91B198007430ARASXnewsDFE1@130.133.1.4>

Spencer Doidge || Sun 10 Feb 2002 12:33:46p:

> Running Python22's Pythonwin.
> I import all from a script Foo, run some routines, and then edit Foo
> in
> the edit window. I check syntax and save, then click the Import/reload
> button. Returning to the interactive window, I run Foo again, but I
> don't see the expected evidence of the changes I made.
> What am I doing wrong?

When you reload a module, all of the objects in that module get 
recreated and the names in there get refreshed. But any references to 
the old objects are still references to the old objects, which 
persumably hang around till they run out of references and are collected 
by the GC. For example:

x = somemod.x
x is somemod.x -> 1
reload(somemod)
x is somemod.x -> 0

The object formerly known as 'somemod.x' is still there, and x still 
points to it, but the reload has caused 'somemod.x' to point to 
something else.

I don't know if there's ever a way to use reload and get a useful effect 
out of it except -maybe- in the shell if you remeber not to use any of 
the old objects you made again... you may as well just restart the 
program... (Maybe if you can bring the program into a embrionic state 
where nothing has been done yet, reload everything, then continue... but 
it seems like a lot more effort than it's worth :)

Here's a longer version of what happens when you use reload, the 
'reloadtest' module contains:

class Nothing_base(object):
    pass
class Do_nothing(Nothing_base):
    pass


Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help

>>> import reloadtest
>>> from reloadtest import Do_nothing
### ### Equivelent to 'Do_nothing = reloadtest.Do_nothing'
>>> Do_nothing is reloadtest.Do_nothing
1
>>> Do_nothing.__base__ is reloadtest.Nothing_base
1
>>> dn = reloadtest.Do_nothing()
>>> dn.__class__ is Do_nothing
1
>>> dn.__class__ is reloadtest.Do_nothing
1
>>> dn.__class__.__base__ is reloadtest.Nothing_base
1
>>> id(Do_nothing)
11082528
>>> id(reloadtest)
11076864
>>> id(reloadtest.Do_nothing)
11082528
>>> id(reloadtest.Nothing_base)
11083312

>>> reload(reloadtest)
<module 'reloadtest' from 'c:\desk\dv\py\lib\reloadtest.pyc'>

>>> Do_nothing is reloadtest.Do_nothing
0
>>> Do_nothing.__base__ is reloadtest.Nothing_base
0
>>> dn.__class__ is Do_nothing
1
>>> dn.__class__ is reloadtest.Do_nothing
0
>>> dn.__class__.__base__ is reloadtest.Nothing_base
0
>>> id(Do_nothing)
11082528
>>> id(reloadtest)
11076864
>>> id(reloadtest.Do_nothing)
11434928
>>> id(reloadtest.Nothing_base)
11510864
>>> 

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From svbrk at start.no  Fri Feb  1 13:02:21 2002
From: svbrk at start.no (Svein Brekke)
Date: 1 Feb 2002 10:02:21 -0800
Subject: PIL Q: Python interpreter cannot find _imaging.pyd
Message-ID: <21298535.0202011002.4674269@posting.google.com>

I have installed python 2.2 precompiled for Windows2000 Professional
and the PIL version 1.1.2 for python 2.2 (also precompiled).

Then I got this strange message, altough I _think_ my environment
variables are correct. Can anybody explain what is happening?

Environment:
  PATH = c:/usr/python;c:/usr/python/DLLs;c:/usr/python/PIL
  PYTHONPATH = c:/usr/python;c:/usr/python/PIL/PIL

In the python shell:
>>> import Image
>>> i = Image.open('p:/tmp/testx.jpg')
>>> i.resize((0,10),0)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "c:\usr\python\pil\pil\Image.py", line 661, in resize
    self.load()
  File "c:\usr\python\pil\pil\ImageFile.py", line 126, in load
    self.load_prepare()
  File "c:\usr\python\pil\pil\ImageFile.py", line 180, in load_prepare
    self.im = Image.core.new(self.mode, self.size)
  File "c:\usr\python\pil\pil\Image.py", line 37, in __getattr__
    raise ImportError, "The _imaging C module is not installed"
ImportError: The _imaging C module is not installed


I am thankful for any help :-)
Svein Brekke


From robin at jessikat.fsnet.co.uk  Sun Feb 10 16:04:28 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 10 Feb 2002 21:04:28 +0000
Subject: cgi authoriszation
References: <q4Sr5IAZ4cZ8EwGa@jessikat.fsnet.co.uk>
 <mailman.1013356516.24479.python-list@python.org>
Message-ID: <TaDFhIAcBuZ8EwDP@jessikat.fsnet.co.uk>

In article <mailman.1013356516.24479.python-list at python.org>, Oleg
Broytmann <phd at phd.pp.ru> writes
>On Sun, Feb 10, 2002 at 01:34:17AM +0000, Robin Becker wrote:
>> client's response. Is apache filtering out the credentials? 
>
>   Yes. Apache filter out "Authorization" header, but passes "REMOTE_USER"
>environment variable. Though I'm sure it's passed by mod_auth_* modules, so
>you may need to stick to Apache authorization modules, or write your own.
>
>Oleg.
not much use passing the user as no apache authorisation was required so
any password would do. I guess I would need to write my own.
-- 
Robin Becker


From aleax at aleax.it  Mon Feb 18 06:12:34 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 11:12:34 GMT
Subject: Object copying itself
References: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com> <slrna6rgk4.9fc.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <Ck5c8.68519$om6.1892014@news1.tin.it>

Quinn Dunkan wrote:
        ...
>>I want object references copied (shallow copy) but the lists need to
>>be copied entirely (deep copy).
        ...
> Note that 'self.__dict__.update(other.__dict__)' is a popular pydiom, but
> that won't deepcopy the lists.

No, but following that with a specific and explicit deepcopying of
the lists looks like a good, clear way to signal this extremely
peculiar design intent that the OP has asserted.  Specific
assignments if the list-attributes' name are know, else a loop, i.e.:

    self.__dict__.update(other.__dict__)
    self.alist = copy.deepcopy(self.alist)
    self.anotherlist = copy.deepcopy(self.anotherlist)

or (with a couple of 2.2'isms) if e.g. you want to get this from
a mixin class or generic function (useful if you need the same
peculiar functionality for more than one object type!):

    self.__dict__.update(other.__dict__)
    for k, v in self.__dict__.iteritems():
        if isinstance(v, list):
            setattr(self, k, copy.deepcopy(v))


This doesn't work for objects with __slots__ (again, in 2.2), but,
there, it's even easier to "loop over all the slots" after all.

Hmmm, come to think of it, one typically does need to "loop over
all instance-specific attributes, be they slots or else keys
in self.__dict__" -- a good abstraction to hide away in a helper
function such as:

def attributeNames(obj):
    try: return obj.__slots__
    except AttributeError: return obj.__dict__.keys()

(not worth using iterators for the small number of attributes
that typically exist...).  Maybe inspect already has that (I
haven't memorized inspect yet:-)... but compatibility with 2.x
for x<2 suggests coding this small helper out anyway...


Alex




From ivano.delpiccolo at degroof.be  Tue Feb 12 04:35:21 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Tue, 12 Feb 2002 10:35:21 +0100
Subject: Shell command
Message-ID: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net>

How can I do to launch a command in my python script ?
example : I want to launch the command : rm test.txt




From jason at jorendorff.com  Sat Feb  2 23:08:46 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 2 Feb 2002 22:08:46 -0600
Subject: Standing Objects ... possible?
In-Reply-To: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>
Message-ID: <HFEKILOLEFEFMKAECNDLIEABDAAA.jason@jorendorff.com>

Philip Swartzleonard writes:
> What i want is a construct like:
> 
> object Foo( ancestor ):
>     	pass

Yep, it exists.

class MyObject:
    """ This is the class of object I want to create """
    def __init__(self, name, bases, dict):
        # ...do whatever you want here...
        print "NOTE: creating a MyObject with name =", name

# Here I make the ancestor...
ancestor = MyObject('ancestor', (), {})

# Now I can make objects like this:
class Foo(ancestor):
    x = 3
    y = 4
    # ...

print Foo    # voila!

## Jason Orendorff    http://www.jorendorff.com/







From gcordova at hebmex.com  Wed Feb 20 17:56:31 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 16:56:31 -0600
Subject: A bundle of questions from a Python newbie
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC6@SRM201002>

> 
> > 
> > 6. Is there a reason why 3.__abs__() does not work? After 
> all 3 is an
> > object (in the wider sense of the word). And it does work 
> if we assign
> > it to a variable, eg, 
> > 
> > var = 3
> > var.__abs__()
> > 
> > I know that code like the above (methods applied to number 
> literals) is
> > probably rare, but consistence is also a good thing.
> > 
> 
> because not everything is an object (yet)
> 

But, it's because the "3." is being interpreted
as a floating point number. Use a space:

>>> 3 . __abs__()
3
>>>


> > 10. I have a class that works like a list in the sense that supports the
> > method __getitem__ and its relatives. But I'm at a loss as to what I
> > have to do to support the slice notation.
> 
> >>> class Foo:
> ...   def __getslice__(*args):
> ...     print args
> ... 
> >>> f = Foo()
> >>> f[1:1]
> (<__main__.Foo instance at 0x80905e4>, 1, 1)
> 

and ditto for:

...   def __setslice__(*args): print args
...   def __delslice__(*args): print args


-gustavo



From jeff at ccvcorp.com  Fri Feb 22 12:48:04 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 22 Feb 2002 09:48:04 -0800
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4> <Nktd8.1458$N7.276578@ruti.visi.com> <4Jud8.1483$N7.287257@ruti.visi.com>
Message-ID: <3C768454.A5F778D2@ccvcorp.com>


Grant Edwards wrote:

> I just got bit by the leading-zero thing again about 10 minutes
> ago when I changed the minor version number of something from 07
> to 08.  Had to skip from 1.07 to 1.10...

So try using 010, which should then be converted back to 8 ... (or else keep your
versioning info as strings.)

Jeff Shannon
Technician/Programmer
Credit International




From chris.gonnerman at newcenturycomputers.net  Mon Feb 18 21:22:10 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 18 Feb 2002 20:22:10 -0600
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com>
Message-ID: <00bd01c1b8ec$41b8b060$0101010a@local>

Ultimately, adding static typing to Python results in a different
language; it's not Python anymore.

If you don't like Python because the typing isn't strong, use Ada,
or C++, or Pascal... and leave Python for those of us who have 
discovered that a simple, semantically pure language that "just
works" solves a lot more problems than overengineered compilers.





From synthespian at uol.com.br  Sun Feb 10 19:42:32 2002
From: synthespian at uol.com.br (synthespian)
Date: Sun, 10 Feb 2002 21:42:32 -0300
Subject: No explanation for weird behavior in re module!
Message-ID: <a470re$1cl30l$1@ID-78052.news.dfncis.de>

Hi-

	I'm really intrigued by this behavior:

>>> import re
>>> p = re.compile('^(der|die|das(\s\w+))')
>>> m = p.match('die T?r, T?ren')
>>> n = p.match('das Auto, Autos')
>>> m.group(0)
'die'
>>> m.group(1)
'die'
>>> m.group(2)
[nothing!!!!]
>>> n.group(0)
'das Auto'
>>> n.group(1)
'das Auto'
>>> n.group(2)
'Auto'

	I'm using Python2.0 on a Debian potato system. 
	Why didn't m.group(2) produce 'T?r' as the output???
	Python2.0 is supposed to have Unicode support buil-in the re module right?
	Other than the fact that 'T?r' has the '?' unicode charcater, I fail to see any difference!
	I've even tried "import sre", but that didn't do it either...It's too bad this isn't working, because it's a better way to work with regexx than Perl...
	What's going on here? Am I the problem here, not knowing how to make Python understand the umlaut
character (the '?')? Or is it a * bug *??!!!!!

	Please help!
	TIA,
	best regards,

	H





From timr at probo.com  Sat Feb 16 23:46:23 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 16 Feb 2002 20:46:23 -0800
Subject: Directed Acyclic Graph Representation
Message-ID: <j9du6uckr4g1am3780qtig50jdhcaf6fcr@4ax.com>

Is anybody aware of a reasonable package for representing a directed
acyclic graph in Python, and for rendering that graph in some meaningful
way?  I need to build a dependency tree analyzer for, for example,
determining the necessary link order for a set of object libraries.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From usenet at thinkspot.net  Thu Feb 21 19:22:58 2002
From: usenet at thinkspot.net (Sheila King)
Date: Thu, 21 Feb 2002 16:22:58 -0800
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org> <7xbseiga7j.fsf@ruckus.brouhaha.com>
Message-ID: <a536t3.3vsf5uh.1@kserver.org>

On 21 Feb 2002 06:50:56 -0800, Paul Rubin <phr-n2002a at nightsong.com>
wrote in comp.lang.python in article
<7xbseiga7j.fsf at ruckus.brouhaha.com>:

> Usually you don't get nearly as fast a connection through ADSL as you
> get at a well-connected ISP, and also ADSL is pretty unreliable.
> Remember also that ADSL is usually slower in the upload direction,
> which is what you care about if you're running a server.

This is a good point, also. On the speed that you would get for a usual
home DSL connection you would not be able to handle as many visitors to
your site, nor serve your pages as fast, as with a hosted account at a
quality hosting company.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From piet at cs.uu.nl  Sat Feb 23 10:02:45 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 23 Feb 2002 16:02:45 +0100
Subject: Warning about "for line in file:"
References: <just-8C3043.11330718022002@news1.xs4all.nl> <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com> <20020220143307.A2018@glacier.arctrix.com> <mailman.1014256478.16106.python-list@python.org>
Message-ID: <wzzo20dywa.fsf@ucu-106-58.ucu.uu.nl>

>>>>> Aldo Cortesi <aldo at nullcube.com> (AC) writes:

AC> ... but this would leave a discrepancy between the actual
AC> file position and the notional "read" position. Among other
AC> things, file.tell() would then give the wrong information,
AC> and file.write() would write to an unexpected location. 

AC> It looks like the only complete solution (short of changing
AC> the sematics by fiat) would be to keep the file descriptor
AC> offset correct by performing a seek after reading every
AC> line. Then again, this would probably kill any efficiencies
AC> gained by xreadlines...

But seeks would not work with pipes (e.g. stdin) or sockets.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From mhammond at skippinet.com.au  Tue Feb 19 00:17:48 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 19 Feb 2002 05:17:48 GMT
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com> <3C6D08C1.2000307@skippinet.com.au> <9d4c7023.0202151038.1d3307f@posting.google.com> <3C6DAF39.5080808@skippinet.com.au> <9d4c7023.0202181505.1646f9e6@posting.google.com>
Message-ID: <3C71E06E.5080206@skippinet.com.au>

Sanjay wrote:
> Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6DAF39.5080808 at skippinet.com.au>...
> 
>>Sanjay wrote:
>>
>>>What I am trying to do is something that was somewhat discussed on a
>>>previous thread, but I'm not sure the thought was fully completed. 
>>>Specifically, I would like to write an exchange server event sink by
>>>implementing two com interfaces, IExStoreAsyncEvents and
>>>IExStoreSyncEvents.  The problem I'm having is that when the python
>>>event listener is loaded as a com+ application, and I register this
>>>application for events with exchange, I get the error TypeError ...
>>>IID not supported.  From looking at the previous threads, it seems as
>>>this problem is caused by not being able to handle arbitrary vtable
>>>interfaces, such as the interface methods of IExStoreAsyncEvents and
>>>IExStoreSyncEvents.  If, however,  the two com classes
>>>IExStoreAsyncEvents and IExStoreSyncEvents were defined in a pyd, then
>>>pythoncom would have knowledge of these interfaces, and properly call
>>>my application.
>>>
>>>I hope I'm explaining the problem clearly. I am a little confused as
>>>to how to get around this with pythoncom.
>>>
>>You need t6o check out the win32com\servers\test_pycomtest.py.  This 
>>shows how arbitary vtables must be loaded by your code before it is 
>>recognized.  This in-turn implies that the interface is described in a 
>>type library.
>>
>>Mark.
>>
> 
> Ok, I looked at test_pycomtest.py and changed my interface to match,
> i.e. I set _public_methods_ to an empty list, _com_interfaces_ to
> "[IExStoreSyncEvents]", I created a new _reg_clsid_ value, and
> implemented the two methods in the interface.  I then registered the
> class using register.UseCommandLine() with debugging turned on.  Then
> I told Exchange to notify this object when new events happen.  In the
> windows Event viewer, I now see the following error when events occur
> in Exchange and it tries to call my python com object:
> 
>     Exception: C0000005
>     Address: 0x1E12E6A5
>     Call Stack: 
>     python22!PyDict_Next + 0xC5
>     ole32!StringFromCLSID + 0xB91
>     ...
>     ...
> 
> There doesn't seem to be anything strange in the Python Trace
> Collector window.  I see a message saying "Object with win32trace
> dispatcher created (object=None)", followed by 3 calls to
> QueryInterface with unsupported IIDs (Unregistered, ISTdMarshalInfo,
> and IExternalConnection) which are all interfaces I should not need
> implemented here.
> 
> Any idea what the problem is?

It could be IExternalConnection - this is often necessary to support 
events.  See the various "connect.py" files in the win32com tree for 
some insights into how events are usually hooked up.

Do you have sample C++ code that works in this environment.  If so, you 
should be able to see exactly what is necessary to implement.

Mark.



From martin at v.loewis.de  Tue Feb 26 04:05:28 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 10:05:28 +0100
Subject: socket in python 2.2 vs 2.1
References: <mailman.1014206732.25525.python-list@python.org> <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de> <mailman.1014290914.10912.python-list@python.org> <j48z9mq1r0.fsf@informatik.hu-berlin.de> <mailman.1014700245.15051.python-list@python.org>
Message-ID: <m3u1s4fw9z.fsf@mira.informatik.hu-berlin.de>

Ivan Bykov <ivan at kupol.ru> writes:

> IMHO, '255.255.255.255' is a valid IP number. This is number mask, use
> for broadcast packet.

Somebody correct me if I'm wrong, but 255.255.255.255 is *not* a valid
IP address. Formally, it would be a class E address, for experimental
use; but it has nothing to do with broadcast.

> MvL> Just remove the connect call.
> Then program to work not will.

Can you explain why? Are you using .send, or .sendto, later?

Regards,
Martin


From tl_news at nexgo.de  Wed Feb  6 14:31:44 2002
From: tl_news at nexgo.de (Tino Lange)
Date: Wed, 06 Feb 2002 20:31:44 +0100
Subject: What is the command to determine an object's type?
References: <mailman.1012941850.17781.python-list@python.org> <Z9Z78.23594$l93.4286654@newsb.telia.net>
Message-ID: <bm036ukfdun2it79e91sllcjf2n909242d@4ax.com>

On Tue, 05 Feb 2002 22:38:49 GMT, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

Hi Fredrik!

(1) if type(l) == types.ListType:
>or better:
(2) if type(l) is types.ListType: 
>or even better:
(3) if isinstance(l, types.ListType):

Why is "if isinstance(l, types.ListType):" even better than the rest?
What do you mean by "better"?

I just compared the speed of execution (using Python 2.1 on Linux) and
found consistently:
(2) is the fastest, the comes (3), (1) is much slower.

So I would guess (2) is the most effective implementation?

Tino





From s.schwarzer at ndh.net  Thu Feb 14 15:41:12 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 14 Feb 2002 21:41:12 +0100
Subject: Disabling creation of .pyc files?
References: <mailman.1013617184.11819.python-list@python.org>
Message-ID: <3C6C20E8.4390F474@ndh.net>

Hello Dominic

Dominic Sacr? wrote:
> I'm writing a C++ program which embeds Python, and I'm a bit annoyed by the
> fact that for every .py file I'm importing, Python automatically creates a
> corresponding .pyc file, which persists after my program exits.
> 
> Is there a way to disable that feature?

There was a thread on this recently:
http://groups.google.com/groups?hl=de&frame=right&th=5a6c36264f7b6d07&seekm=mailman.1011966458.9451.python-list%40python.org#link1

Stefan


From tom at 8dynamics9.com  Thu Feb  7 10:05:01 2002
From: tom at 8dynamics9.com (Tom F.)
Date: Thu, 07 Feb 2002 15:05:01 GMT
Subject: dependency hell
Message-ID: <xIw88.500$9%6.51706@bgtnsc04-news.ops.worldnet.att.net>

I'm trying to install python2-2.2-2. I apparently need the following files:

libcrypto.so.2, libreadline.so.4, libssl.so.2

As far as I can search, libopenssIO and readline are supposed to supply 
them.
 I can even find libreadline.so.4  specificly on my system (but not the 
others).

I've done many RPM searches and more, to no avail. could someone perhaps 
point me inthe right direction?

-- 
Tom F.
Please remove the "9"
to reply to this message.


From wzdd at lardcave.net  Wed Feb 13 23:22:27 2002
From: wzdd at lardcave.net (Nicholas FitzRoy-Dale)
Date: Thu, 14 Feb 2002 15:22:27 +1100 (EST)
Subject: RELEASED [OOPS - The Python Datastore v0.9]
In-Reply-To: <JmGa8.5516$MR3.311561@wagner.videotron.net>
Message-ID: <Pine.LNX.4.43.0202141511540.604-100000@jollywodger>

At 10:32pm on the 13th of February, Dan P wrote:

> OOPS - The Python Datastore.
>
> OOPS stands for object-oriented persistent storage which should give you an
> idea of its purpose. It was designed to be as easy to learn as python, to
> have an intuitive interface, to be conservative in memory and processor
> usage; basically to be as simple as possible without sacrificing
> scalability.

Hi Dan,

I'm currently working with the ZODB, which is another implementation of a
persistent data store for Python. The ZODB is part of Zope, but a standalone
version is also available.

Have you looked at ZODB? I'm not trying to suggest that OOPS is duplicating
it. I'm just wondering what the differences are. Please excuse me if I
missed something big. :)

ZODB: http://www.zope.org/Products/StandaloneZODB

- Nicholas





From peter at engcorp.com  Thu Feb  7 01:04:58 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 07 Feb 2002 01:04:58 -0500
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net>
Message-ID: <3C62190A.A54E9A57@engcorp.com>

Timothy Rue wrote:
> 
> On 06-Feb-02 12:48:10 Jeff Shannon <jeff at ccvcorp.com> wrote:
> >The problem you're running into, is that you've failed to convince anyone
> >that *you* understand the real problems involved, enough to be able to solve them.
> >[...]
> Since you have taken it upon yourself to be the spokes person for
> programmers, 

Actually, we elected him.  Everything he says goes for *all* of us here.
Keep that in mind as you read the following; Jeff has our complete
and unconditional confidence and support in this...

> I'll have to assume that all of you are so lost that you
> can't tell the difference between three deminsional reality and three
> deminsional data arrays in the two deminsional world of programming.

That's clearly true.  Our spokesman is the best of us, and yet even
he is clearly (so _you_ say... I'm too stupid to tell) out to lunch.

> Seems to me that your two deminsional mindset has trouble comprehending
> what three deminsional reality really is. With that in mind, it's not
> possible for me to convince you of anything beyond your own limited
> two deminsional perceptions.

I think you're absolutely right (although I'm too dumb to be sure, but
I'm going out on a limb).  If even Jeff is incapable of understanding
you, and believe me when we had The Vote we concluded he was the *only*
one with a hope of understanding even a glimmer of what you say, 
then I guess it's really a lost cause for the rest of us.

Now my logic here may be weak, because I am *really* stupid, but
would it not be the case that if *all* of us in this newsgroup are
so pathetically, mind-numbingly obtuse that you cannot convince us 
of anything beyond our own limited two-dimensional perceptions, then
you might as well leave?

Now?

For good?

Please?

(Will you take the hint, Tim?)


From dale at riverhall.NOTHANKS.co.uk  Mon Feb 18 09:04:25 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Mon, 18 Feb 2002 14:04:25 +0000
Subject: Redefining "\n"
References: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>
Message-ID: <rd227u8gsco6iu27d2c7inukqjjghq6r3n@4ax.com>

Dale Strickland-Clark <dale at riverhall.NOTHANKS.co.uk> wrote:

>Is there some way to redefine the "\n" escape sequence?


Thanks for all the replies. Pretty much as I expected.
--
Dale Strickland-Clark
Riverhall Systems Ltd


From logiplexsoftware at earthlink.net  Wed Feb  6 04:06:41 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 6 Feb 2002 01:06:41 -0800
Subject: Autocoding project proposal.
In-Reply-To: <slrna61rdd.ap.quinn@spew.ugcs.caltech.edu>
References: <83it9mghng.fsf@panacea.canonical.org>
	<B87C2640.3CC2%jonathan@onegoodidea.com>
	<slrna5d2cq.g57.philh@comuno.freeserve.co.uk>
	<1341.799T18T11762852threeseas@earthlink.net>
	<slrna5rpoa.19d.philh@comuno.freeserve.co.uk>
	<988.801T1700T14365170threeseas@earthlink.net>
	<slrna61rdd.ap.quinn@spew.ugcs.caltech.edu>
Message-ID: <20020206010641.74beb13b.logiplexsoftware@earthlink.net>

On 6 Feb 2002 08:47:42 GMT
Quinn Dunkan wrote:

> On Wed, 06 Feb 2002 04:54:25 GMT, Timothy Rue <threeseas at earthlink.net>
wrote:
> >Good, than you won't know it hit you untill it is to late, for you.
> 
> Sounds like the robot army's already on its way.


Ruebots!  They won't actually *do* anything, but they'll post senseless
drivel and silly insults until we're real sorry.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From jason at jorendorff.com  Sat Feb  2 10:16:32 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 2 Feb 2002 09:16:32 -0600
Subject: How can I write this C code in Python?
In-Reply-To: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAEOKCPAA.jason@jorendorff.com>

Fred wrote:
> I think if I could see the below C code done in Python it
> would get me over the beginners hump.

for i in range(1, 13):
    for j in range(1, 13):
        print "%4d" % (i*j),
    print

## Jason Orendorff    http://www.jorendorff.com/



From bldrake1 at yahoo.com  Tue Feb  5 18:27:39 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 5 Feb 2002 15:27:39 -0800
Subject: HappyDoc: Can't get output
References: <fb072181.0202050304.35626d25@posting.google.com> <c5b3dbee.0202050727.60991126@posting.google.com>
Message-ID: <fb072181.0202051527.4819c2b0@posting.google.com>

The problems with Win32 paths have been fixed by Doug Hellmann and
release 2.0.1 has been posted at sourceforge.net (please see the
ANNOUNCE by Doug in a later post).  BTW Doug responded within hours of
my original post and I greatly appreciate his efforts.

Thanks for your response waldekO.
Barry

osuchw at ecn.ab.ca (waldekO) wrote in message news:<c5b3dbee.0202050727.60991126 at posting.google.com>...
> bldrake1 at yahoo.com (Barry Drake) wrote in message news:<fb072181.0202050304.35626d25 at posting.google.com>...
> > When I try to run HappyDoc, I get the following:
> > 
> > commandline:
> > c:\python21\happydoclib> python happydocwin.py <anyfile.py>
> <snip>
> For some reason HappyDoc does not handle windows paths correctly but
> if you specify output directory (instead going for default) it works
> as expected.
> For example
> c:\python21\happydoclib> python happydocwin.py -d c:\ <anyfile.py>
> 
> waldekO


From pereira at cis.upenn.edu  Sat Feb 23 18:39:31 2002
From: pereira at cis.upenn.edu (Fernando Pereira)
Date: Sat, 23 Feb 2002 18:39:31 -0500
Subject: Is there a python parser generator that is similar to JavaCC
	from webgain?
References: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com> <m38z9jpz4a.fsf@mira.informatik.hu-berlin.de>
Message-ID: <B89D9263.3864%pereira@cis.upenn.edu>

On 2/23/02 6:17 PM, in article m38z9jpz4a.fsf at mira.informatik.hu-berlin.de,
"Martin v. Loewis" <martin at v.loewis.de> wrote:

> "Shen Wang" <shenwang at ameritech.net> writes:
> 
>> As title
> 
> There are a number of Python parser generators. Since I don't know
> JavaCC (and don't want to spend time learning it), I cannot tell you
> which of those is similar to JavaCC.

JavaCC takes grammar specs (lexical patterns and rules) and produces a
parser (a Java class) using a LL(k)-type of rule compilation (that is,
deterministic recursive descent with lookahead). Rules can have embedded
Java code that not only computes rule output (eg. syntax tree) but also can
affect the parsing in some ways. JavaCC can be quite painful to use because
of the limitations of LL(k), but the resulting code is simple, and the
embedding of Java side-effects into parsing via rules is easy to understand
(as expected for a recursive descent parser).

-- F



From -$P-W$- at verence.demon.co.uk  Mon Feb 18 08:13:01 2002
From: -$P-W$- at verence.demon.co.uk (Paul Wright)
Date: 18 Feb 2002 13:13:01 -0000
Subject: There is more than one way to do it - and for no apparent reason.
References: <3C70ECB1.2090709@mxm.dk>
Message-ID: <a4qukt$4nd$1@verence.demon.co.uk>

In article <3C70ECB1.2090709 at mxm.dk>, Max M  <maxm at mxm.dk> wrote:
>Basically you can say that JScript uses the dictionary notation for 
>getting attributes. As fas I can see this is smarter than the way we do 
>it in Python where we have several notations for the same action:
>
>In an object:
>setattr(myObj, 'name', 'max')
>getattr(myObj, 'name')
>
>In a dict:
>myObj['name'] = 'max'
>myObj['name']
...
>Wouldn't it be a good idea to unify the dict and the object so that both 
>could be called with attribute and dict notation like in JScript?

Not in the language itself, no. Since Python has established a
conceptual difference between dictionary keys and attributes, to change
that now would break existing code. What I mean is that there are
objects which provide a dictionary interface (using "[]") where the
coder does not want to return the attribute of the same name: the
rfc822.Message class, for example.

If you really want it, there's nothing stopping you doing it for your
own classes:

class Thing:

        def __getitem__ (self, key):
                return getattr (self, key)

        def __setitem__ (self, key, value):
                setattr (self, key, value)

probably does what you want. See also
<http://www.python.org/doc/current/ref/sequence-types.html> 

I'm not sure I'd recommend it, though, as using getattr and setattr is
the standard way to do it.

-- 
Paul Wright | http://pobox.com/~pw201 |


From hmartires at cultalg.pt  Mon Feb 25 07:59:18 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 12:59:18 -0000
Subject: Numeric arrays - initialising with binary data
References: <cf0ad9fb.0202250446.3df07a08@posting.google.com>
Message-ID: <Pwqe8.596$xL1.7420993@newsserver.ip.pt>

Where are this method's ??
I import the array module, but i don't have them.
I'm using python 1.5

Tanks

"Jesper Olsen" <jolsen at mailme.dk> wrote in message
news:cf0ad9fb.0202250446.3df07a08 at posting.google.com...
> Hi,
>
> The build in array type has the methods
>
> array.tostring()
> array.fromstring()
>
> which make it easy to initialise arrays with binary data
> read from a file.
>
> The Numeric array, however, apperantly only has the tostring method.
>
> How can I initialise a Numeric array (eg Float32), with binary data?
>
> I can off course do this by using the Numeric C-API, but I am sure
> there must be an easier way.
>
> I am surprised the Numeric array type does not have the same methods
> as the build in type, if the intention is in time to replace this
> module...
>
> Cheers
> Jesper




From bokr at oz.net  Thu Feb 21 02:49:57 2002
From: bokr at oz.net (Bengt Richter)
Date: 21 Feb 2002 07:49:57 GMT
Subject: Adding static typing to Python
References: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com> <mailman.1014092911.31096.python-list@python.org> <gft37u8csm1ukd923nr933iq7meimqu5i7@4ax.com> <a4stbu$ptq$1@peabody.colorado.edu> <3C748A4F.350BA6F7@home.com>
Message-ID: <a528r5$m98$0@216.39.172.122>

On Thu, 21 Feb 2002 05:47:38 GMT, Don O'Donnell <donod at home.com> wrote:

>Fernando P?rez wrote:
>> 
>> Courageous wrote:
>> 
>> >
>> > One of the things I like about C sharp is their use of attributes,
>> > where you can define getters and setters on an attribute, but
>> > otherwise the attribute is simply used as if it were a public one.
>> > An objects attributes can thereby be used as a published aspect of
>> > its interface safely, but no one will be the wiser if you manipulate
>> > the getter and setter code.
>> 
>> Just curious, isn't there something quite similar to what you describe in
>> py2.2? I've only glanced at it so I could be wrong...
>> 
>
>
>Yes, there is, and it's not new with C#.  I think something like
>attributes with transparent getter and setter methods have been in VB at
>least since version 6.
>
IIRC Delphi's properties have worked that way since Delphi 1. I think
Anders Hejlsberg(sp?) probably brought the idea, among others, along with
him when Microsoft made him an offer he couldn't refuse and got him to
leave Borland/Inprise (or whatever it was at the time)(Not to say that
MS wasn't copying Delphi ideas before it got Anders). Delphi had a
lot of stuff first, and was/is the leader in GUI building IDEs. Plus object
pascal is compiled, and the compilation is so fast you literally don't realize
when it happenes for small (more than trivial) programs, unless you have code errors.

>I am just starting to look at the Python implementation.  It is based on
>a new built-in type called property.  In 2.2, at the interactive prompt,
>type:
>
>>>> print property.__doc__
>
ok:
--
Typical use is to define a managed attribute x:
class C(object):
    def getx(self): return self.__x
    def setx(self, value): self.__x = value
    def delx(self): del self.__x
    x = property(getx, setx, delx, "I'm the 'x' property.")
>>>
--
Then compare with an example Delphi property declaration
(snip from help file):
Here is a typical declaration for a property called Count:
--
type
  TYourComponent = class(TComponent)
 private
   FCount: Integer; { used for internal storage }
   procedure SetCount (Value: Integer); { write method }
 public
    property Count: Integer read FCount write SetCount;
 end;
--
{ curly brackets contain comments in object pascal ;-}

Note that you can specify direct access (without getter or setter)
for poperty values for either, neither, or both read and write.
Direct access is like ordinary attribute access. E.g., that way reads
can be direct and writes can make coordinated updates in setters,
as necessary (e.g. preserving shape of a rectangle when one dimension
changes, etc.). If you left out the "write SetCount" above, the
property would be read-only. You can also specify default values.

Delphi also permit array properties (declarations omitted,
snip from help file):
--
An array property is accessed by following the property
identifier with a list of actual parameters enclosed
in square brackets. For example, the statements

if Collection.Objects[0] = nil then Exit;
Canvas.Pixels[10, 20] := clRed;
Params.Values['PATH'] := 'C:\DELPHI\BIN';

correspond to

if Collection.GetObject(0) = nil then Exit;
Canvas.SetPixel(10, 20, clRed);
Params.SetValue('PATH', 'C:\DELPHI\BIN');
--

You could do worse than emulating Delphi's property stuff.

>And read Andrew Kuchling excellent "What's New in Python 2.2", Section
>2.4 Attribute Access:
>
>http://www.amk.ca/python/2.2/index.html#SECTION000340000000000000000
>
>Have fun.
>
Ok ;-)

Regards,
Bengt Richter



From fperez528 at yahoo.com  Sun Feb  3 15:12:38 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 03 Feb 2002 20:12:38 +0000
Subject: Executing User Entered Code?
References: <8efc7149.0202031326.72806bda@posting.google.com>
Message-ID: <a3ku9c$5mq$1@peabody.colorado.edu>

Bob wrote:

> Hi,
> I'm writing a network simulator using Python and Tkinter.  I'd like to
> be able to allow the user to enter the code for the routing algorithm
> in Python and then execute it for each node during the simulation.
> 

You may be interested in looking at my pet project, IPython at 
http://www-hep.colorado.edu/~fperez/ipython/

Even though it started as a python interpreter shell, I later made it a 
trivially embeddable module. With a single call (there's example code for 
that purpose provided in the distribution) you can open an interpreter shell 
at any point in your program execution which knows about its environment.

Note however that there are *many* security issues with this, which may or 
may not be important in your context. IPython does *not* restrict access to 
any library or statement. If you need sandboxing, you'll have to add that 
around it yourself.

Cheers,

f


From ws at mystrobl.de  Wed Feb 13 14:51:32 2002
From: ws at mystrobl.de (Wolfgang Strobl)
Date: Wed, 13 Feb 2002 20:51:32 +0100
Subject: Hidden files, selecting directories, & taxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
Message-ID: <qqgl6uc8pco9b75m706gj22i7aklsp5mqu@4ax.com>

On 12 Feb 2002 16:23:12 -0800, chris_mk at hotmail.com (Christopher) wrote
:

>1)  Is it possible to generate a hidden file (under Windows)?  

http://groups.google.com/groups?selm=0kln0u0k2ivc3ehegq5m9lern9e9uut78r%404ax.com

| From: Wolfgang Strobl <ws at mystrobl.de>
| Newsgroups: comp.lang.python
| Subject: Re: making a file hidden in win32?
| Date: Mon, 03 Dec 2001 20:46:28 +0100
| 
| On Sun, 2 Dec 2001 14:13:46 -0800, "Peoter Veliki"
| <peoter_veliki at hotmail.com> wrote :
| 
| >Is there anyway I can use python to make a file hidden in windows? 
| 
| import win32api
| import win32con
| win32api.SetFileAttributes( "c:/scratch/versteck",
|      win32con.FILE_ATTRIBUTE_HIDDEN )
| 
| 

-- 
Thank you for observing all safety precautions


From marklists at mceahern.com  Sun Feb 10 22:58:37 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 19:58:37 -0800
Subject: Random Module help
In-Reply-To: <20020210215559.13516.00000455@mb-cq.aol.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEBCKCAA.marklists@mceahern.com>

[Patio87]
> I was wondering if any 1 could tell me how to use the random module for a
> blackjack game i want to make. I am a begginer so try to be easy with the
> dialouge. Thanks

What follows is a fully functioning backgammon program:

$ python
>>> import random
>>> random.randint()
>>> random.randint(1,10)
2

<wink>

// mark



From fredrik at pythonware.com  Sun Feb 17 03:32:38 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 17 Feb 2002 08:32:38 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122> <xyuk7tdku0w.fsf@sex.ifi.uio.no> <a4mn4k$gd2$0@216.39.172.122>
Message-ID: <GUJb8.26252$l93.4948148@newsb.telia.net>

Bengt Richter wrote:
>
> >But if there really is something wrong with that, shouldn't someone
> >send a bug report or something to the creators of Python?
>
> No, not to the creators of Python per se. This is not a bug
> in Python itself. Telnetlib is just a piece of the distribution.

the creators of Python gladly accepts bug reports for everything
in the standard distribution:

    http://python.sourceforge.net/sf-faq.html#bugs

</F>




From buzzard at urubu.freeserve.co.uk  Sat Feb  9 19:47:06 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Sun, 10 Feb 2002 00:47:06 -0000
Subject: Animated graphs / wxPython
References: <a43rur$31f$1@newsg1.svr.pol.co.uk> <Xns91B099B307C26RASXnewsDFE1@130.133.1.4>
Message-ID: <a44g23$hv8$1@newsg1.svr.pol.co.uk>

Thanks.  I'll have go at this.  I think my learning curve just got
steeper.:-)

Duncan




From tanzer at swing.co.at  Sun Feb 17 02:52:00 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sun, 17 Feb 2002 08:52:00 +0100
Subject: Convert String to Dictionary question
In-Reply-To: Your message of "Sat, 16 Feb 2002 08:40:24 CST."
             <HFEKILOLEFEFMKAECNDLOEAFDDAA.jason@jorendorff.com>
Message-ID: <E16cM76-0000po-00@tswings.swing.cluster>

"Jason Orendorff" <jason at jorendorff.com> wrote:

> > > It seems to me that security is a compelling reason to choose
> > > pickle() over eval().  Am I wrong?
> >
> > With regard to eval, yes. Check out the two optional arguments to
> > eval, before spreading FUD.
>
> Well, the bottom line is:  If I'm looking for a vulnerability in
> an app, I'm searching for calls to eval(), not pickle.
>
> For example, the idiom you suggested elsewhere in this thread:
>   >>> eval ("""os.system ("echo 'kilroy'")""", {}, {})
>
> is actually exploitable:
>   >>> s = "__import__('os').system('echo kilroy')"
>   >>> eval (s, {}, {})
>   kilroy
>   0

I stand corrected. And I'd like to apologize for being a bit rude.

That one is easily plugged though:

>>> eval (s, {"__builtins__" : {}},{})
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<string>", line 0, in ?
NameError: name '__import__' is not defined

> Whereas there are no known security holes in pickle.

Not true.

To wrap this up, writing secure applications is hard -- and rules of
thumb like `eval is bad, pickle is good` aren't going to make it any
easier.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From phr-n2002a at nightsong.com  Sat Feb  2 21:55:00 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 02 Feb 2002 18:55:00 -0800
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org>
Message-ID: <7x4rkzw9yz.fsf@ruckus.brouhaha.com>

Kragen Sitaker <kragen at pobox.com> writes:
> What do you mean by "stable"?  What's "unstable" about having to put
> 'from __future__ import division' at the top of your module to get
> reasonable division behavior for the next two, three, five, or ten
> years?

I just tried that and it throws an error:

    $ python
    Python 1.5.2 (#1, Mar  3 2001, 01:35:43)  [GCC 2.96 20000731 (Red Hat Linux 7.1 2 on linux-i386
    Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
    >>> from __future__ import division
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    ImportError: No module named __future__
    >>> 

That's with the version of Python currently being shipped with Red Hat
Linux, so any scripts I distribute with that future import will crash
for a lot of people.


From stojek at part-gmbh.de  Fri Feb 22 16:41:28 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Fri, 22 Feb 2002 21:41:28 GMT
Subject: wxlayoutconstraints inside a wxDialog ?
References: <3c76666e.18420093@news.easynews.net> <Xns91BDC97DAC9B9rcamesz@amesz.demon.nl>
Message-ID: <3c76b920.39590390@news.easynews.net>

Hi,
>There is one basic thing wrong with this code: you're not initializing 
>your class correctly. You should call the constructor of the base 
>class, but you're creating a new wxDialog instead.
You are right. I have copied the code from a wx example that
was not used inside a class and I didn't think. 
Thanks.


I did what you said, but still the main effect is there. When the
dialog pops up, all elements are crowded at the upper left corner.
If I resize the dialog or add the line 

self.SetSize((355,205))

everything jumps in place. The Doc says if I don't use 
SetAutoLayout i have to overwrite the OnSize() function
and call Layout() myself. Fine. But I AM using SetAutoLayout
here.


From gbreed at cix.compulink.co.uk  Tue Feb 26 06:08:34 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Tue, 26 Feb 2002 11:08:34 +0000 (UTC)
Subject: Time to rename Stackless?
References: <m3pu2sancq.fsf@chvatal.cbbrowne.com>
Message-ID: <a5fqbi$p5q$1@thorium.cix.co.uk>

> There have got to be some puns heading in the direction of "seriously
> stacked."  Hmm...

Wonder Python!


From tim.one at comcast.net  Sun Feb 24 18:27:49 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sun, 24 Feb 2002 18:27:49 -0500
Subject: Silly link question: python22_d.dll
In-Reply-To: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEPGNOAA.tim.one@comcast.net>

[Courageous]
> ...
> Furthermore, why the heck doesn't the binary release include
> python22_d.dll???

_d files proved to be far more trouble than they were worth when PythonLabs
used to distribute them.  Newbies in massive numbers kept getting confused
by them, while self-identified <wink> experts disagreed about which _d files
were needed and where they should be installed.  The audience for them is
tiny and fractious.

BTW, they were never included in the binary distro, because they're useless
bytes to the vast bulk of Python's Windows users -- the installer download
is already bigger than I'd like.

The offer I made when I stopped distributing them still stands:  if the
handful of people who want this stuff get together and write a PEP, agreeing
on exactly which _d files are needed (e.g., your claim that python22_d.dll
is enough probably won't stand:  there are a dozen other Release-build DLLs
shipped with Python too), where to put them, how to distribute them without
confusing newbies, and documenting the intended use cases clearly enough so
that I don't have to spend most of my life replying to confused email, I'll
be happy to resume shipping _d files pulled from the Windows builds on my
box.

In the meantime, you hit on the intended solution, which only Windows
developers have the chutzpah to complain about (the idea that a Unixish
developer would make a similar complaint used to be <wink> unthinkable):

> I know how to build Python from source, and that's how I do
> it now, but that's a pain sometimes.




From magnus at trhinkware.se  Thu Feb 21 06:47:26 2002
From: magnus at trhinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 12:47:26 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>
Message-ID: <3C74DE4D.2000703@trhinkware.se>

Eduard Smirnov wrote:

  > I'm writing a diploma paper on the difference between scripting
  > (e.g. in Python or Tcl) and system programming (e.g. in C, C++ or
  > Java), and one of the claims I want to make is that you can often
  > develop applications 10x or more faster in a scripting language.

I wouldn't really say that Python is a scripting language though. The
name scripting language suggests a language suited for small snippets.
Python is as much a general language for large scale software
development as C++ is. But Python is surely a rapid development
language.

Bash, awk, Tcl and Perl are in my opinion scripting languages, but
not Python. (Well, Perl probably has the theoretical requirements for
developing large systems, but it's fairly obscure and bolted on as an
afterthought.)

  > I apologue that more applications can be written or re-written in
  > scripting language faster and with more complex functionality then
  > in system programming language. If you have any experience in
  > writing applications in system programming language (e.g. in C,
  > C++) and re-writing it in scripting language (e.g. Python) or vice
  > versa, please compare your work in both implementations,

Actually, I think many comparisions are made like this, and a problem
is that if you first write in C, and then translate, you are likely to
use your C design for Python, and end up with a program that is almost
as big, but much slower. If you go the other way, there is no way you
can use your Python solution directly, since C can't handle the level
of abstraction, and you will probably get a fairer comparision.

  > describe your work/application's functionality, approximate number
  > of lines and time of development of each implementation, skill's
  > level needed. Any more comments about are welcome.

Personally, my Python applications worked well, thank you, so I
never rewrote them in any other language...

There are some links at
http://www.thinkware.se/cgi-bin/thinki.cgi/PythonQuotes
which might guide you do relevant sources. (Glyph Lefkowitz,
Greg Stein and Eric S. Raymond for instance.)

Also, have a look at
http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprt_computer2000.pdf
For this experiment, the productivity gain with Python is roughly
factor three, but this is also

There is a comparision at http://www.bagley.org/~doug/shootout/
but it's really just a pointless play. Don't fall for this one.
It has two fundamental flaws: It only uses small programs, so
it doesn't show you anything about how the language scales. If you
believed in that, do all your coding in C and bash. It also ignores
one of greatest assets of Python, its rich standard library.

So instead of writing the "random generator" as

import random
print random.random()

They reinvent the wheel like this:

import sys

IM = 139968
IA = 3877
IC = 29573

LAST = 42
def gen_random(max):
      global LAST
      LAST = (LAST * IA + IC) % IM
      return( (max * LAST) / IM )

def main():
      N = int(sys.argv[1])
      if N < 1:
          N = 1
      gr = gen_random
      for i in xrange(1,N):
          gr(100.0)
      print "%.9f" % gr(100.0)

main()

In other words, many of the programs are about as long
as C programs since they "level the playfield" by not
using the "included batteries". On the other hand Python
always show more CPU and memory usage since it loads the
interpreter and a lot of modules before it runs the those
twenty lines of code. No wonder that C seems slim and fast
there. If we talked about a REAL application, and compared
50 000 lines of C++ with 10 000 lines of Python code, things
would look very different.








From landshark at verticaladdiction.net  Thu Feb 14 09:36:09 2002
From: landshark at verticaladdiction.net (Dave Marotti)
Date: Thu, 14 Feb 2002 14:36:09 -0000
Subject: undefined symbols when embedding python
References: <slrna6mm3i.2aj.landshark@burrito.verticaladdiction.net> <m3y9hwpht2.fsf@mira.informatik.hu-berlin.de>
Message-ID: <slrna6nitb.2r4.landshark@burrito.verticaladdiction.net>

Hey Martin,

Yes, the extension module is in a shared library.

Most of this is all pretty new to me, so for whatever reason, the approach of
extension library + separate embedded executable seemed more natural to me.

I'll give both ways a shot (extended module + embedded, and embedded with the
wrapper directly in it).  Now that I think of it, I cannot think of any reason
not to have the wrapped classes directly in the embedded app.

Thanks for the tips!

-dave

>I take it that the extension module is in a shared library? (which is
>kind of surprising, because in an embedded application, you'ld
>normally want all extension modules to be linked into the application,
>instead of having them as separate files on disk).
>
>I also take it that you use GCC on Linux? In that case, you need to
>export the symbols from the executable to the shared library, using
>-Wl,--export-dynamic (see ld(1) for details).
>
>Of course, this problem would also go away if you would incorporate
>the extension into the application.


-- 
+-----------------------------+----------------------------------------------+
| Dave Marotti                | Looking for a Visio alternative for *nix?    |
| lndshark ! speakeasy net    | Kivio : http://thekompany.com/projects/kivio |
+-----------------------------+----------------------------------------------+
| I just don't trust anything | I suppose that in a few more hours I will    |
| that bleeds for 5 days and  | sober up. That's such a sad thought. I think |
| doesn't die. - Mr. Garrison | I'll have a few more drinks to prepare myself|
+-----------------------------+----------------------------------------------+



From phr-n2002a at nightsong.com  Wed Feb 20 11:22:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 08:22:34 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014221254.26490.python-list@python.org>
Message-ID: <7xu1scdsxh.fsf@ruckus.brouhaha.com>

"Sean 'Shaleh' Perry" <shalehperry at attbi.com> writes:
> this is a design decision in the language.  By not allowing this style of
> programming we do not have the ever so fun:
> 
> if foo = 2:
>    handle_foo()
> 
> type bugs.  Sure there are ways around it. 

So why are augmented assignments not allowed in expressions?

  if (foo += 2) < 7:
        handle_foo()

doesn't seem likely be an =/== confusion.


From sjmachin at lexicon.net  Mon Feb 25 22:08:09 2002
From: sjmachin at lexicon.net (John Machin)
Date: 25 Feb 2002 19:08:09 -0800
Subject: raw_input on password
References: <slrna7lf63.4pb.msoulier@pmerd071.ca.nortel.com>
Message-ID: <c76ff6fc.0202251908.59773c7f@posting.google.com>

"Michael P. Soulier" <msoulier at nortelnetworks.com_.nospam> wrote in message news:<slrna7lf63.4pb.msoulier at pmerd071.ca.nortel.com>...
> Hey people. 
> 
>     My apologies if this has been asked before. I didn't see it...

Funny how Google has its off days.
> 
>     Is there a way to get raw_input() to mask the response, so it can be used
> to request a password safely?
> 

Not as far as I can tell from reading the manual. There may be a
secret hidden argument which you could discover by reading the source
code. However I doubt it.

Completely off-topic w.r.t. your question, why don't you use the
getpass module?


From owen at astrono.junkwashington.emu  Fri Feb  1 13:09:17 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Fri, 01 Feb 2002 10:09:17 -0800
Subject: Preferences in Tkinter applications -- partial answer
Message-ID: <a3elkd$hmc$1@nntp6.u.washington.edu>

Thanks to reading Welch "Practical Programming in Tcl and Tk" and 
corresponding with Laura Creighton, I have found the following:

Tkinter widgets will live update fonts (by using named fonts) and colors 
(by using tk_setPalett). This was from Welch. Details are available on 
my Tkinter Folklore page 
<http://www.astro.washington.edu/owen/ROTKFolklore.html>.

Other preferences require some sort of callback or signalling mechanism. 
Laura Crieghton had some interesting suggestions including using Patrick 
O'Brien's global signal dispatcher: 
<http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/87056>.
I plan to start with a PrefVar class that includes callbacks. (I could 
simply use Tk variables, and Welch has a nice-looking implementation, 
but callbacks are easy in Python.)


I did come out of this with one question:
- If one creates a named font (a tkFont.Font object), is it possible to 
modify that font's characteristics, starting from just its name? The 
issue is that once you put a tkFont.Font into the option database, you 
can only get the name back out. But I can't see how to modify a named 
font's characteristics with just its name. (Well, it can be done by 
calling explicit tcl code, but it's probably easier to just keep the 
Font objects around.)

-- Russell


From phr-n2002a at nightsong.com  Fri Feb  1 16:33:12 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 01 Feb 2002 13:33:12 -0800
Subject: pickle security
References: <mailman.1012591743.10841.python-list@python.org>
Message-ID: <7xit9g51lz.fsf@ruckus.brouhaha.com>

Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> I'm trying to understand how to safely unpickle untrusted strings.  I'll
> summarize what I think I've learned from Google searches, and someone tell
> me if I'm wrong.

This subject is treated somewhat better in the current version of the
pickle docs than in earlier versions.

> Did I get it right?  Are there any other security issues I need to be aware
> of, or does this cover them?

I continue to feel uncomfortable using unpickle on untrusted data.
For example, it calls eval to handle quoted strings.  While that looks
safe on the surface, there are an awful lot of code paths you have to
examine to make sure it can't nail you.

You may be better off using something like xml-rpc or even marshal.


From sholden at holdenweb.com  Wed Feb  6 13:32:48 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 6 Feb 2002 13:32:48 -0500
Subject: Problem while uploading a binary file using cgi module
References: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU> <u5t6k65m7869d7@corp.supernews.com> <u1o16ugtmp7kn5meao6nm85l0bm1htj3vl@4ax.com>
Message-ID: <9Fe88.87331$KM2.2835653@atlpnn01.usenetserver.com>

"Tim Roberts" <timr at probo.com> wrote in message
news:u1o16ugtmp7kn5meao6nm85l0bm1htj3vl at 4ax.com...
> "DeepBleu" <DeepBleu at DeepBleu.org> wrote:
> >
> >Wang
> >Please send the code indented next time.  You know, you may know it
inside
> >out.  But poor readers like me do not know your intentions and indenting
> >makes it easier to read.
>
> His code WAS indented, using hard tabs.  I'm guessing Microsoft is
> "helping" you here by having Outlook Express skip leading hard tabs.  Next
> time, try using a real news reader.

Given the proliferation of crappy newsreader software, sensible posters
adopt the Internet philosophy of being critical in what they produce and
liberal in what they accept.

So, normally I don't complain when someone posts tab-indented code, but if I
can't read it then someone else is going to have to reply.

Telling people to use some other piece of software is impractical: on a
group like this it isn't usually because they don't know any better, it's
for other pragmatic reasons. Next time, try using real manners <0.75 wink>.

using-outlook-express-and-not-a-bit-proud-of-it-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From mhammond at skippinet.com.au  Mon Feb 25 19:31:48 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 26 Feb 2002 00:31:48 GMT
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au> <3C7ACEB4.B590E892@mediaone.net>
Message-ID: <3C7AD777.7050401@skippinet.com.au>

Mark Barclay wrote:
> Thanks Mark. I have your book, and it is invaluable. I should have credited you
> personally.
> 
> Is there a way of coding a COM server that can be running before it is accessed
> by a COM client? I'm not sure if this makes sense, but I'd like to have a program
> running, that can be connected to, perhaps multiple times, without it having to
> create multiple instances of teh server, just have multiple clients using it.

IIRC, it gets a little messy - you need to implement a LocalServer 
object, and add it to the running object table (ROT).  Searching old 
posts at groups.google.com is probably your best bet.

Mark.



From lac at strakt.com  Wed Feb  6 07:15:05 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 06 Feb 2002 13:15:05 +0100
Subject: Inherit Class Dynamicly 
In-Reply-To: Message from Laura Creighton <lac@strakt.com> 
   of "Tue, 05 Feb 2002 16:06:25 +0100."
Message-ID: <200202061215.g16CF6Go015335@ratthing-b246.strakt.com>

I was really tired when I posted that article yesterday and misunderstood
what you wanted.  Apologies.  I think that this will be more to your liking:

> > Hi, everybody,
> > 
> > from Tkinter import *
> > class NewWidget:
> >         ...
> >         def __init__(self, widget_name = "Button", **dict):
> >                 # What is needed here to make the new born object(self) 
> >                 # behavior like both the class specified by widget_name
> >                 # and NewWidget?
> > 
> > Thanks from
> > 
> > Xiao-Qin
> > 

from __future__ import nested_scopes
import Tkinter

class WrappedWidget:
    def __init__(self, frame, widgetType=Tkinter.Label, **kw):
        self._widget=widgetType(frame, **kw)
        self.bind("<ButtonPress-1>", self.gotPressed)
        
    def __getattr__(self, name):
        if name.startswith("__") and name.endswith("__"):
            raise AttributeError, name
        else:
            return getattr(self._widget, name)

    def gotPressed(self, event):
        print 'I got Pressed!'
        self.configure(bg='red')

###########################  
if __name__ == '__main__':

    root = Tkinter.Tk()
    quit = Tkinter.Button(root, text="QUIT", fg="red", command=root.quit)
    quit.pack(side='left')

    Lab=WrappedWidget(root, fg='green', text='Label (default)')
    Lab.pack(side='left', fill='both', expand = 1)

    mydict={'fg':'yellow', 'text':'Button'}
    But=WrappedWidget(root, Tkinter.Button, **mydict)
    But.pack(side='left', fill='both', expand = 1)
    root.mainloop()
--------------------------------------------

This is the decorator pattern.  Boy is it simple in python.

Note, this relies on the behaviour of classic classes, where you search
all the methods in a class _first_, and then if it is not found, you
use the redefined __getattr__.  In 2.2 in new style classes, the 
__getattr__ is looked up _first_ so this will not work.  I haven't
figured out how to make the decorator pattern in 2.2 not classic
patterns yet.

Sorry to have posted half asleep last night.

Laura Creighton



From opengeometry at yahoo.ca  Thu Feb 21 12:44:10 2002
From: opengeometry at yahoo.ca (William Park)
Date: Thu, 21 Feb 2002 12:44:10 -0500
Subject: Hosting
In-Reply-To: <7xbseiga7j.fsf@ruckus.brouhaha.com>; from phr-n2002a@nightsong.com on Thu, Feb 21, 2002 at 06:50:56AM -0800
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org> <7xbseiga7j.fsf@ruckus.brouhaha.com>
Message-ID: <20020221124410.A1179@node0.opengeometry.ca>

On Thu, Feb 21, 2002 at 06:50:56AM -0800, Paul Rubin wrote:
> William Park <opengeometry at yahoo.ca> writes:
> > On Wed, Feb 20, 2002 at 03:51:24PM -0800, Kevin Dahlhausen wrote:
> > > Does anyone know of an inexpensive web host that allows compilation
> > > and use of Python extensions from  your own directory?
> > 
> > With the money you pay to ISP, why don't you get DSL with static IP, and
> > run your own server?
> 
> Usually you don't get nearly as fast a connection through ADSL as you
> get at a well-connected ISP, and also ADSL is pretty unreliable.
> Remember also that ADSL is usually slower in the upload direction,
> which is what you care about if you're running a server.

All true.  But, OP said "inexpensive ISP" with "Python".  Good luck
finding them.  

Of course, OP has to find local ISP who will give him static-IP and
don't mind him running server.  Here in Toronto, Canada, there are 2
local ISPs who do that for CAN$45/month (US$30).

-- 
William Park, Open Geometry Consulting, <opengeometry at yahoo.ca>
8 CPU cluster, NAS, Hosting, (Slackware) Linux, Python, LaTeX, Vim, Mutt, Tin



From gmuller at worldonline.nl  Mon Feb  4 16:29:28 2002
From: gmuller at worldonline.nl (GerritM)
Date: Mon, 4 Feb 2002 22:29:28 +0100
Subject: apology: Re: changes20020204
References: <a3mtre$mr0$1@reader1.tiscali.nl>
Message-ID: <a3mu61$n61$1@reader1.tiscali.nl>

this message was misdirected, apologies for the noise.

--
www.extra.research.philips.com/natlab/sysarch
"GerritM" <gmuller at worldonline.nl> schreef in bericht
news:a3mtre$mr0$1 at reader1.tiscali.nl...
> 22:23
>
> --
> www.extra.research.philips.com/natlab/sysarch
>
>
>




From tone.mcdonald at ncl.ac.uk  Mon Feb 18 11:19:33 2002
From: tone.mcdonald at ncl.ac.uk (tone)
Date: 18 Feb 2002 16:19:33 GMT
Subject: Any OAI tools available in Python?
Message-ID: <20020218162109858+0000@saturn5.ncl.ac.uk>

Hi all,
We're getting involved in some work using OAI records from the
openarchives initiative at http://www.openarchives.org/

At first glance (ie google, searching this mailing list, the OAI site and
the PyXML distribution) there doesn't seem to be any work being done in
OAI using Python. Lots of Perl, not a lot of python.

Is this really the case?  We use Zope almost exclusively and don't want
to spend any time on the Perl option if we can avoid it at all.

Does anyone know any better?

many thankstone.


From lac at strakt.com  Thu Feb  7 12:40:33 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 07 Feb 2002 18:40:33 +0100
Subject: Autocoding evolves from........ 
In-Reply-To: Message from "Chris Gonnerman" <chris.gonnerman@newcenturycomputers.net> 
   of "Thu, 07 Feb 2002 00:13:46 CST." <00b901c1af9e$9de1d080$0101010a@local> 
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>  <00b901c1af9e$9de1d080$0101010a@local> 
Message-ID: <200202071740.g17HeXGo019034@ratthing-b246.strakt.com>

The only reason that Timothy Rue should leave is because we are absolutely
of no use to him in the furthering of his project.  If absolutely everybody
who has no intention of helping him had the discipline to shut up and
stay shut, he could more accurately judge how helpful posting here is
to his project.

Voting people out of a newsgroup, as a concept, disgusts me.

Laura Creighton



From jkraska at san.rr.com  Tue Feb  5 18:41:52 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 05 Feb 2002 23:41:52 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c6011e3$1_5@news.bluewin.ch>
Message-ID: <tor06u0j1ce99ibgd98uflckcms89apfci@4ax.com>

>3. Given the absence of closure blocks
>local vars clashes should be handled through
>renaming (?) (issues?)

If I understand you properly, I'm considering making the the
macro call itself _generate_ what amounts to a poor man's
closure block. I have a few implementation details for this,
but the trick to understanding it is to understand that no
_first class_ form needs to exist to create the block, only
the backend facility.

C//


From joshm at taconic.net  Tue Feb  5 01:01:13 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 5 Feb 2002 01:01:13 -0500
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <3c5f72c6$1_2@corp.newsgroups.com>

"Courageous" <jkraska at san.rr.com> wrote in message
news:tppu5u48fsmsclucfr5e0qd1musgnqk4vl at 4ax.com...
> I've come up with this concept that I'm calling "lingos": the name
> was chosen to evoke the idea of defining one's own lingo as a subset
> of the language. I'm still very much in the concept definition stage.
[snip]
> So. That, at last, brings me to the point of this message. Ideas?
> Feedback? This has been an interesting experiment so far, I'm just
> wondering how many folks find this sort of thing interesting and if
> anyone has any input.

Looking at what you proposed, I'm not sure I understand what differentiates
your "lingo" from a common function.  A function seems to be everything
you've defined for lingo.  Is there some problem that using function calls
can't solve?
--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From jesusdb at gmx.at  Sat Feb 16 05:35:34 2002
From: jesusdb at gmx.at (Dominik)
Date: 16 Feb 2002 02:35:34 -0800
Subject: XmlprocDriver instance has no attribute 'feed'
Message-ID: <7c4fa2db.0202160235.5efdccac@posting.google.com>

When using the xml.dom.minidom module from the standard python2.2
installation, the an error msg '..XmlprocDriver instance has no
attribute 'feed' appeared.

To solve this problem a new version of PyXML has to be installed. (In
my case version 7)

Good luck.
Dominik


From starx at pacbell.net  Sun Feb  3 14:47:22 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 3 Feb 2002 19:47:22 GMT
Subject: HELP?
References: <a3jebb$cg8$1@knossos.btinternet.com> <mailman.1012747685.1392.python-list@python.org>
Message-ID: <Xns91AA77EC6A6F4RASXnewsDFE1@130.133.1.4>

Chris Gonnerman || Sun 03 Feb 2002 06:46:08a:

> ----- Original Message -----
> From: "Callum Golding" <Rave_Dj at btopenworld.com>
> 
> 
>> I'm sorry to bother you, but i was wondering if you could give me a
>> little advice. I've recently downloaded Python 2.2. Ive read a
>> begginers guide to programming and tried creating a small program, I
>> used the input function. Now as soon as I open the program it goes
>> into dos and asks me the question, and when i type in an anwser, dos
>> shuts down. Now its really starting to bug me. If you could give me
>> some advice I would be real grateful. The program I wrote is below; - 
> 
> Fairly common problem.  You neglected to state your OS but I can easily
> assume
> Windows; note that many of us use other OS's, so you should always
> state that.
> 
> You are running the script from Explorer, by double-clicking?  In that
> mode, a console application (like python.exe) will open a "dos window"
> (actually a Windows Console) and run in it.  When the program
> terminates, the window closes (as you have noticed).
> 
> Two solutions:  First, run your script from the DOS prompt.  Do Start,
> Run, "command", ENTER; or find MS-DOS Prompt under Start, Programs.  If
> you are running NT, 2000, or XP, you can change "command" to "cmd"
> (don't type the quotes in any case, they are there to tell you that the
> quoted material is to be typed in).  Once in the DOS window, change
> directories to the folder your script is in, and type
> 
>     python myscript.py
> 
> If you get a "Bad Command or Filename" message, you need your Python
> directory
> in your DOS PATH variable.  How to do this varies between 95/98 and
> NT/2000/XP,
> so you should tell us what you are using.
> 
> You could type:
> 
>     c:\python22\python myscript.py
> 
> if you installed Python 2.2 in the standard place, and that would work.
> 
> Second solution:  Add
> 
>     raw_input()
> 
> as the last line of your script.  Then, you will have to press ENTER to
> end the script.
> 
> Learn to work at the command prompt.  It's a dying art.

Third solution (more advanced):
create a pystart.bat (or similar) in your python directory, containing
'''
c:\path\to\this\version\of\python.exe %1 %2 %3 %4 %5 %6 %7 %8 %9
pause
'''
and change the command for the default action (open) for python files from 
the type-manger (from some explorer window: view | folder optoins, tab 
'file types') to, well, i have 'C:\programs\python22\pystart.bat "%1" %*'

This gives you a global pause for .py files--- if you generally want to 
read output from console-based scripts for windows, it may be a nicer 
solution. (plus you can turn off the pause, globally, by throwing a 'rem' 
infront of it :)

Just have to be different :}

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From gherman at europemail.com  Fri Feb 15 08:54:06 2002
From: gherman at europemail.com (Dinu Gherman)
Date: Fri, 15 Feb 2002 14:54:06 +0100
Subject: Paper "PDF Editing and Processing" at IPC 10?
Message-ID: <gherman-822C04.14540615022002@news.t-online.com>

Hi!

At IPC 10 there was the following presentation/paper: 

  Jyh-Yuan Deng and Shang-Kang Deng
  Using Python as PDF Editing and Processing Framework

that I'm interested in. I read the paper, but unfortu-
nately the thing seems to have no name, the author 
doesn't reply to emails and the website and/or company 
of one of the authors, www.FlashMix.com, seems not to 
exist.

Hence, I wonder, is this vaporware or has somebody seen 
it in action or even used it?

Regards,

Dinu

.....................................................
"If the Nuremberg laws were applied today, then every 
Post-War American president would have to be hanged."
                                       (Noam Chomsky)


From pj at sgi.com  Mon Feb 11 19:36:02 2002
From: pj at sgi.com (Paul Jackson)
Date: 12 Feb 2002 00:36:02 GMT
Subject: Reading Powerpoint on Linux [was: Slides on the web]
References: <mailman.1013135593.4897.python-list@python.org> <a49mvo$auq6q$1@fido.engr.sgi.com>
Message-ID: <a49o1i$b0j26$1@fido.engr.sgi.com>

Oops - I hadn't finished catching up on this news group.

Thanks to a thread Skip Montanaro started last Friday:

    PPT reader?

this question (Powerpoint reading on Linux) has already
been answered.

Here I was trying to avoid getting hit with Guido's
reputation, and his time machine got me instead.

Nevermind ...

-- 
-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From just at xs4all.nl  Wed Feb 27 16:20:48 2002
From: just at xs4all.nl (Just van Rossum)
Date: Wed, 27 Feb 2002 22:20:48 +0100
Subject: PIL vs eps
References: <mailman.1014835309.1869.python-list@python.org>
Message-ID: <just-B1FF7B.22204827022002@newszilla.xs4all.nl>

In article <mailman.1014835309.1869.python-list at python.org>,
 <brueckd at tbye.com> wrote:

> Forgive me if I'm totally off my rocker, but I think by definition eps
> files should *not* have a showpage and it is technically correct that
> nothing gets rendered unless you have a showpage command. EPS is
> "encapsulated postscript" - postscript that you intend to encapsulate or
> embed in another document, so showpage would muck things up.

The EPS spec says showpage is optional, and that the surrounding ps code 
should redefine showpage as a no-op.

  http://partners.adobe.com/asn/developer/pdfs/tn/5002.EPSF_Spec.pdf

Just


From martin at v.loewis.de  Tue Feb 26 16:19:44 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 22:19:44 +0100
Subject: python stripping LD_LIBRARY_PATH?
References: <3C7BDD3A.9050905@nmt.edu>
Message-ID: <m3sn7ouej3.fsf@mira.informatik.hu-berlin.de>

Jeff Shipman - SysProg <jeff at nmt.edu> writes:

> Why is python stripping out this environment variable?

Most likely, it is not Python which strips this out, but the dynamic
linker (/lib/ld-linux.so.1) will. It does so if the effective and the
real user id are different (likewise for group ids); see
sysdeps/generic/enbl-secure.c of the glibc sources.

Regards,
Martin



From robin at jessikat.fsnet.co.uk  Fri Feb 15 10:53:39 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 15 Feb 2002 15:53:39 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk>
 <3C6C4C37.6020209@skippinet.com.au> <t4oqZKA4$Mb8Ew3k@jessikat.fsnet.co.uk>
 <3C6D07E2.5030005@skippinet.com.au>
Message-ID: <uWWKyJAD8Sb8Ew0N@jessikat.demon.co.uk>

In article <3C6D07E2.5030005 at skippinet.com.au>, Mark Hammond <mhammond at skippinet.com.au> writes
>Robin Becker wrote:
>> sounds right to me, they would also have to fix pythonservice as well.
>
>I don't see why pythonservice needs to be fixed - it just uses the 
>python (if indeed it even needs that registry key.
>
>> Incidentally I notice that the path used by pythonservice is the 8.3 dos
>> mangled version. Inside the Zope startup script they're using the normal
>> long filename version to possibly correct the path. Will that not cause
>> problems? 
>
>I presume you are talking about the sys.path entries?  I have no idea 
>why pythonservice would use the short name - I don't think I have ever 
>seen it do that.  Can't see why it would be a problem though - both 
>names are equivilent.
>
>Mark.
....
well I see the following using procexp indicating two versions of
PyWinTypes21. Also win32api comes form the system's python not from Zope's.

I assume this is unsafe if not downright dangerous.
.......
0xAF0000   0x10000          Common Python 2.1 types for win32                
        2.01.0000.0145   18/01/2002 15:29    C:\WINNT\system32\PyWinTypes21.dll 
0x1E100000 0xAF000          Python Core                                      
        2.01.0017.1010   25/01/2002 13:42    C:\Python\ZOPE_2_5_0\bin\lib\win32\python21.dll 
0x1E200000 0xF000           Python Win32 API extensions                      
        2.01.0000.0145   18/01/2002 15:29    C:\Python\win32\win32api.pyd 
0x1E600000 0x10000          Common Python 2.1 types for win32                
        2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\PyWinTypes21.dll 
0x1E700000 0x8000           Python Win32 Thread and Process extensions       
        2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32process.pyd 
0x1E800000 0x7000           Python Win32 Event extensions                    
        2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32event.pyd 
0x1EA00000 0x7000           Python NT Service extensions                     
        2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32service.pyd 
0x1EB00000 0x8000           Python Service Dispatcher Exe                    
        2.01.0000.0140   22/06/2001 10:47    C:\PythonService.exe
.......
-- 
Robin Becker


From cmaloney at physics.ucsb.edu  Tue Feb 26 12:57:57 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: Tue, 26 Feb 2002 09:57:57 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com>
Message-ID: <3C7BCCA5.604@physics.ucsb.edu>

Craig Maloney wrote:
> Hi David.
> 
> Since I'm not a "computer guy" per se, I was wondering if you could 
> clear up some more of my confusion.  I don't know much about how 
> compilers work, but I would presume that any compiler (eg g++) would
> create some internal representation, similar to the .cil file, after
> parsing but before compiling.  
> 
> Presumably the EDG front end is geared toward producing this information
> in a format more suited to being used by the "wrapper generators" in contrast 
> with a compiler which produces this information only for itself and with 
> the sole goal of compiling an object file?  
> 
> What I'm getting at is that I'm confused as to why people wouldn't have
> used information from the compiler *itself* to automate the process of
> wrapper generation.  Obviously the compiler writer people would have to
> collaborate with the wrapper generator people... but there must be technical
> obstacles?
> 
> Cheers,
> Craig 

Actually, after doing a little more digging, I found this:

http://public.kitware.com/Cable/
http://public.kitware.com/GCC_XML/

Does anyone know anything about this project?

This is what I would have guessed would be the "right" way to go about 
things (having known nothing about the EDG front-end).  Presumably gcc 
gets the parsing right if it can compile ;)

What do people think?

--Craig

> 
> "David Abrahams" <david.abrahams at rcn.com> wrote in message news:<a5ekpq$n83$1 at bob.news.rcn.net>...
> 
>>I've been interested in systems for automatically wrapping C++ and
>>generating documentation for a while, so I downloaded the PDT (the parser
>>toolkit used by SILOON) and tried it on one of my source files. After lots
>>of hacking, I was able to get the EDG front-end component to parse my source
>>and generate an intermediate language (.cil) file, but the PDT executable
>>which is supposed to interpret the .cil and produce a high-level description
>>of the code crashed with an assertion:
>>
>>Assertion failed: ptr->assoc_template, file c:\documents and
>>settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-tau\il_display_tau.
>>c, line 3047
>>
>>Since the PDT doesn't come with source, there was no way for me to debug/fix
>>the problem.
>>Furthermore, the PDT developers don't seem to want to be reached. Their
>>contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/) says:
>>
>>    To contact the developers of PDT, please use the "Comments" section in
>>the download page.
>>
>>But of course, there is no "Comments" section on the download page. :=(
>>
>>-Dave
>>
>>"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
>>news:3C7A751D.1080703 at physics.ucsb.edu...
>>
>>>Hi Craig.
>>>
>>>Since your site specifically lists g++ 2.95 and NOT 3.x, am I to infer
>>>that there will be problems with the new ABI?  -- Or any problems that
>>>will, in practice or principle, cause conflicts with g++ 3.0?
>>>
>>>*Holds breath... and crosses fingers...*
>>>
>>>--Craig
>>>
>>>rasmussn at lanl.gov wrote:
>>>
>>>
>>>>On Wednesday, February 20, 2002, at 04:48 PM, David Abrahams wrote:
>>>>
>>>>
>>>>>>As I understand it, the SILOON project uses a commercial parser known
>>>>>>
>> as
>>
>>>>>>PDT from uoregon.  Have you, or has anyone else had any experience
>>>>>>
>> with
>>
>>>>>>the SILOON/PDT project?  On the face of it, my guess would be that you
>>>>>>would be somewhat skeptical of their claims?
>>>>>>
>>>>>
>>>>>I haven't had any experience with it. I just found that site. My
>>>>>skepticism
>>>>>just dropped by a factor of 8. If they are really using the EDG
>>>>>front-end as
>>>>>they claim to, then I am sure the parser is excellent. I do have to
>>>>>wonder
>>>>>who is paying for the FE. The issues of dealing with the semantics of
>>>>>what
>>>>>they've parsed remain, but these guys at least seem to be doing the
>>>>>parsing
>>>>>part right.
>>>>>
>>>>
>>>>It is my understanding that an educational or research license to the
>>>>
>> EDG
>>
>>>>front end is fairly easy to obtain.  The University of Oregon has
>>>>
>> obtained
>>
>>>>the right to distribute the EDG front-end as binaries and does so for
>>>>several
>>>>platforms in the Program Database Toolkit (PDT).  So for anyone wanting
>>>>interface level information from C++ or Fortran 90 source files, the PDT
>>>>is an excellent tool.  We use it in several projects.
>>>>
>>>>See http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/ and
>>>>http://www.acl.lanl.gov/siloon/ for more information.
>>>>
>>>>Craig
>>>>
>>>>
>>>>
>>>




From Simon.Budig at unix-ag.org  Thu Feb  7 08:33:01 2002
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 7 Feb 2002 14:33:01 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de> <u8za6pskh.fsf@python.net> <3c61abe7$1@si-nic.hrz.uni-siegen.de> <uu1sta7s2.fsf@python.net>
Message-ID: <3c62820d$1@si-nic.hrz.uni-siegen.de>

Michael Hudson wrote:
> Simon Budig <simon.budig at unix-ag.org> writes:
> However, there may be people relying on current behaviour, and the
> motto of x.y.z releases is "nothing breaks".
> 
> Also, to get it into 2.2.1 you have to convince the person putting
> patches into the 22-maint tree, and that currently seems to be me, and
> I'm not checking this one in.

Thanks for the PEP-Link. That made it more clear to me. I am not sure
what made me think that the patch would go into 2.1.x - Maybe it is
just me making the patch against 2.1.1...   :-)

When you put it into 2.3(\.[0-9)? I am perfectly happy. I have patched
my local 2.1.1 installation and am happy with it...  :-)

>> Imagine the rlcompleter: It is an convenient way to have completion
>> on the Python commandline. But if you start typing a name that is
>> not (yet) in the global dict and type <TAB> to verify this fact,
>> readline decides to complete with a filename, which is absolutely
>> meaningless in this context - AND YOU CANNOT SWITCH THIS OFF!
> 
> It can be useful too; I have actually done
> 
>>>> f = open("blabla
> 
> and then used tab-completion to finish it off.

Oh well. Yeah, this can be useful. For the 2.3 release it should be
an optional parameter IMHO. 

Thanks for your patience.

Bye,
	Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From doug at mcnaught.org  Thu Feb 21 19:23:04 2002
From: doug at mcnaught.org (Doug McNaught)
Date: 21 Feb 2002 19:23:04 -0500
Subject: Python 2.1.2 regression crash on Linux/Sparc
Message-ID: <m3it8ql5zr.fsf@abbadon.mcnaught.org>

I have an UltraSparc running Linux (Red Hat 6.2 + erratea updates).  I
would love to get Zope running on it.  

I got Python 2.1.2, and did 'configure --with-threads'.  The make goes
fine.  When I run the regression test, I get:

...
skipping 'fpectl' extension (up-to-date)
skipping 'zlib' extension (up-to-date)
skipping 'linuxaudiodev' extension (up-to-date)
skipping '_tkinter' extension (up-to-date)
running build_scripts
not copying /home/doug/src/Python-2.1.2/Tools/scripts/pydoc (up-to-date)
find ./Lib -name '*.py[co]' -print | xargs rm -f
PYTHONPATH= ./python -tt ./Lib/test/regrtest.py -l
make: *** [test] Segmentation fault (core dumped)
PYTHONPATH= ./python -tt ./Lib/test/regrtest.py -l
make: *** [test] Segmentation fault (core dumped)

Here's a GDB backtrace--any wisdom?  Glibc version is 2.1.3. 

[doug at varsoon Python-2.1.2]$ gdb ./python core
GNU gdb 19991004
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "sparc-redhat-linux"...

warning: "/home/doug/src/Python-2.1.2/core": no core file handler recognizes format, using default
Core was generated by `./python -tt ./Lib/test/regrtest.py -l'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /lib/libpthread.so.0...done.
Reading symbols from /lib/libdl.so.2...done.
Reading symbols from /lib/libutil.so.1...done.
Reading symbols from /lib/libm.so.6...done.
Reading symbols from /lib/libc.so.6...done.
Reading symbols from /lib/ld-linux.so.2...done.
Reading symbols from /home/doug/src/Python-2.1.2/build/lib.linux-sparc64-2.1/math.so...done.
#0  0x701bd534 in math_1 (args=0x701b8470, func=0xabdac <exp>, 
    argsfmt=0x701be0a8 "d:exp")
    at /home/doug/src/Python-2.1.2/Modules/mathmodule.c:75
75		CHECK(x);
(gdb) bt
#0  0x701bd534 in math_1 (args=0x701b8470, func=0xabdac <exp>, 
    argsfmt=0x701be0a8 "d:exp")
    at /home/doug/src/Python-2.1.2/Modules/mathmodule.c:75
#1  0x701bd844 in math_exp (self=0x0, args=0x11156c)
    at /home/doug/src/Python-2.1.2/Modules/mathmodule.c:125
#2  0x22e94 in call_cfunction (func=0xb2d10, arg=0x11156c, kw=0x0)
    at Python/ceval.c:2859
#3  0x21060 in eval_code2 (co=0x13f5b0, globals=0x701bd814, locals=0x1, 
    args=0x1, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, 
    closure=0xb2d10) at Python/ceval.c:1952
#4  0x1d3cc in PyEval_EvalCode (co=0x13f5b0, globals=0x1401dc, locals=0x1401dc)
    at Python/ceval.c:341
#5  0x325c4 in PyImport_ExecCodeModuleEx (name=0xefffec58 "random", 
    co=0x13f5b0, 
    pathname=0xefffe298 "/home/doug/src/Python-2.1.2/Lib/random.pyc")
    at Python/import.c:490
#6  0x32b3c in load_source_module (name=0xefffec58 "random", 
    pathname=0xefffe298 "/home/doug/src/Python-2.1.2/Lib/random.pyc", 
    fp=0x13f5b0) at Python/import.c:754
#7  0x33300 in load_module (name=0xefffec58 "random", fp=0x1302d0, 
    buf=0xefffe778 "/home/doug/src/Python-2.1.2/Lib/random.py", type=1)
    at Python/import.c:1301
#8  0x34110 in import_submodule (mod=0xa0468, subname=0xefffec58 "random", 
    fullname=0xefffec58 "random") at Python/import.c:1829
#9  0x33c3c in load_next (mod=0xa0468, altmod=0xa0468, p_name=0xefffec58, 
    buf=0xefffec58 "random", p_buflen=0xefffec54) at Python/import.c:1685
#10 0x3384c in import_module_ex (name=0x0, globals=0xb944c, locals=0xb944c, 
    fromlist=0xa0468) at Python/import.c:1536
#11 0x339ec in PyImport_ImportModuleEx (name=0x11a144 "random", 
    globals=0xb944c, locals=0xb944c, fromlist=0xa0468) at Python/import.c:1577
#12 0x733e4 in builtin___import__ (self=0x0, args=0xc8af4)
    at Python/bltinmodule.c:31
#13 0x22e94 in call_cfunction (func=0xae840, arg=0xc8af4, kw=0x0)
    at Python/ceval.c:2859
#14 0x22d30 in call_object (func=0xae840, arg=0xc8af4, kw=0x73398)
    at Python/ceval.c:2812
#15 0x22b6c in PyEval_CallObjectWithKeywords (func=0xae840, arg=0xc8af4, 
    kw=0x0) at Python/ceval.c:2745
#16 0x20cd8 in eval_code2 (co=0x16c560, globals=0x0, locals=0xb944c, args=0x0, 
    argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0xae840)
    at Python/ceval.c:1824
#17 0x1d3cc in PyEval_EvalCode (co=0x16c560, globals=0xb944c, locals=0xb944c)
    at Python/ceval.c:341
#18 0x3a6f8 in run_node (n=0xbea28, filename=0x16c560 "", globals=0xb944c, 
    locals=0xb944c, flags=0xeffff60c) at Python/pythonrun.c:1049
#19 0x3a6a4 in run_err_node (n=0xbea28, filename=0x16c560 "", globals=0xb944c, 
    locals=0xb944c, flags=0xeffff60c) at Python/pythonrun.c:1036
#20 0x3a674 in PyRun_FileExFlags (fp=0xae418, 
    filename=0xeffff8cd "./Lib/test/regrtest.py", start=780840, 
    globals=0xb944c, locals=0xb944c, closeit=1, flags=0xeffff60c)
    at Python/pythonrun.c:1027
#21 0x39824 in PyRun_SimpleFileExFlags (fp=0xae418, 
    filename=0xeffff8cd "./Lib/test/regrtest.py", closeit=1, flags=0xeffff60c)
    at Python/pythonrun.c:666
#22 0x392b8 in PyRun_AnyFileExFlags (fp=0xae418, 
    filename=0xeffff8cd "./Lib/test/regrtest.py", closeit=1, flags=0xeffff60c)
    at Python/pythonrun.c:480
#23 0x19ab0 in Py_Main (argc=4, argv=0xeffff7b4) at Modules/main.c:320
#24 0x19408 in main (argc=4, argv=0xeffff7b4) at Modules/python.c:10
(gdb)

-Doug


From webmaster at nospam.linux3dgraphicsprogramming.org.nospam  Thu Feb 28 06:50:39 2002
From: webmaster at nospam.linux3dgraphicsprogramming.org.nospam (nlin)
Date: Thu, 28 Feb 2002 12:50:39 +0100
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org>
Message-ID: <3C7E198C.DF6E7566@nospam.linux3dgraphicsprogramming.org.nospam>

Susan Capparelle wrote:

> We're a publisher of high quality computer books currently seeking
> authors for a number of Python related books.

As a published book author with two well-selling titles to my name, I'd
just
like to warn potential authors about my negative experience with
Manning.
After responding to a post similar to the above, I presented Manning
with a
book idea. My experience with Manning was marked by their lack of
technical
expertise on the subject matter, a failure of upper management to
understand
my book's premise, their lack of faith in the author's competence and
vision,
and their unbalanced reliance on unqualified "reviewers".

I lost months lost reworking my material to try to accommodate their
reviewers
(in the case of my book, some of these were self-proclaimed experts who
knew
little about the topic I was writing about), and to compound the misery,
my
earnest attempts to rework the focus of the book were met only with a
curt
message from upper management saying "Why the change. I thought your
book's
focus was XXX" (where XXX was not even the original focus, but only a
tangential
aspect). I never had the feeling that Manning (or at least the upper
management) was really interested in trying to understand my vision for
the
book or make an earnest attempt to work with me - it was almost as if
the
publisher was the author's adversary instead of ally!

Eventually I gave up on them, went elsewhere, and got my book published.
I was
interested in writing a book, not making an eternal sales pitch.  It was
a
waste of good months working with Manning and attempting to appease
their
reviewers and to convince them of my vision. Manning trusted their
reviewers -
again, many of whom had dubious or inflated qualifications - more than
they
trusted the author - not a very productive situation for the author.

Maybe things have changed at Manning. If anyone has better experiences
with
Manning, feel free to share.

One author's opinion,
N Lin
http://linux3dgraphicsprogramming.org




From usenet at thinkspot.net  Sun Feb 17 00:19:04 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 16 Feb 2002 21:19:04 -0800
Subject: Have to try
References: <004501c1b763$78bca9c0$b278c53f@pacbell.net> <mailman.1013918967.19546.python-list@python.org>
Message-ID: <a4mic9.3vvdvvf.1@kserver.org>

On Sat, 16 Feb 2002 21:13:05 -0700 (MST), <kosh at aesaeion.com> wrote in
comp.lang.python in article
<mailman.1013918967.19546.python-list at python.org>:

> On Sat, 16 Feb 2002, Rick Hamilton wrote:
> 
> > So I am trying to do a (what should be) simple thing. I have a string stored
> > in a variable. I want to search that string for the occurrence of a sub
> > string. I have tried
> > Find(s,sub)
> > Rfind
> > MyString.find(sub)
> >
> 
> I think you might find count to be more appropriate if you just want to
> know if it has some string.
> 

Now that is an interesting thought. I had never considered that.

But find will stop at the first occurrence and thus use less resources
(if that is a concern) than count, since count will search the entire
string, and not just stop after the first occurrence.

Just thinking out loud...

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From lennoncao at 263.net  Mon Feb  4 00:07:31 2002
From: lennoncao at 263.net (lennoncao)
Date: 3 Feb 2002 21:07:31 -0800
Subject: what about python ??
Message-ID: <832e3897.0202032107.25b3e70f@posting.google.com>

pls,tell me what about python and where it can been used!


From phd at phd.pp.ru  Sat Feb  9 04:27:26 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sat, 9 Feb 2002 12:27:26 +0300
Subject: I wish.............
In-Reply-To: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com>; from DeepBleu@DeepBleu.org on Sat, Feb 09, 2002 at 12:16:19AM +0000
References: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com>
Message-ID: <20020209122726.B27446@phd.pp.ru>

On Sat, Feb 09, 2002 at 12:16:19AM +0000, DeepBleu wrote:
> .....that we quit calling Python as CPython.  Something bugs me pesonally
> about the name CPython.  This is the pure original.  It is Python not
> CPython.  There is Python + Java = Jython.  There maybe Perl + Python = P#,
> there maybe Python + .NET = Pythonette, there maybe Fortran +Python =
> Pfeiffer.
> But let us keep hand off the name Python.  It is not CPython.  It is Python.

   Python is the (best) languge. CPython is one (and not the best)
implementation.

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



From logiplexsoftware at earthlink.net  Mon Feb 18 19:51:10 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 18 Feb 2002 16:51:10 -0800
Subject: basic books/guides on multithreading programming?
In-Reply-To: <7xy9hqib0a.fsf@ruckus.brouhaha.com>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
	<mailman.1013816158.26552.python-list@python.org>
	<7934d084.0202172009.75324c2a@posting.google.com>
	<mailman.1014058862.32020.python-list@python.org>
	<7xy9hqib0a.fsf@ruckus.brouhaha.com>
Message-ID: <20020218165110.38312aa8.logiplexsoftware@earthlink.net>

On 18 Feb 2002 16:13:57 -0800
Paul Rubin wrote:

> Cliff Wells <logiplexsoftware at earthlink.net> writes:
> > > Everyone: Does anyone know of an accessable concurrency text?  One
> > > that moves from discussing the basics (data-races, critical sections,
> > > mutural exclusion, lock-order based deadlock, resource based
deadlock,
> > > etc) to the various solutions (mutex/cond-var, semaphore, monitor,
> > > rendezvous) to the various design approaches (master-slave, pipeline,
> > > work-pile, etc).  Possibly even a discussion of various performance
> > > considerations (convoy's, wait-free synchronisation, cache-aliasing,
> > > spin vs. blocking locks, etc).
> > 
> > I would certainly like to see a good discussion of these.  Do you have
a
> > URL for the tutorial you wrote?
> 
> A lot of this stuff isn't much of an issue with Python threads, which
> always run in a single process.

Well, that was sort of my thought too, but on the other hand, I'm also
expecting that that limitation will someday change, and prefer to be
prepared by pretending it doesn't exist now ;)

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From djc at object-craft.com.au  Sat Feb  9 20:20:04 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 10 Feb 2002 12:20:04 +1100
Subject: CSV revisited
References: <slrna62f14.pef.mlh@vier.idi.ntnu.no> <a4474v$9t6$1@bob.news.rcn.net>
Message-ID: <m3ofiyqgjf.fsf@ferret.object-craft.com.au>

>>>>> "Raymond" == Raymond Hettinger <othello at javanet.com> writes:

Raymond> Documentation of key design decisions.  For instance, is the
Raymond> library limited to reading various CSV formats or is
Raymond> write-back in the same style to be supported.  Can the first
Raymond> record, contain optional field names?  Will incomplete
Raymond> records raise an exception?  - Pointers to draft
Raymond> implementations so people can experiment with it.

Raymond> A PEP is an ideal place to collect and document this
Raymond> information.  Also, an addition to the standard library is
Raymond> likely to be considered more than a 'minor feature request'.
Raymond> Another advantage is that it will also help focus the
Raymond> discussion (be prepared, life in the focus is sometimes a
Raymond> little hot).

Raymond> If you want a co-author, I'm game.

I am the author of the fast CSV module at:

        http://www.object-craft.com.au/projects/csv/

One of the problems with CSV is that everyone has their own ideas
about what should be handled and how it should be handled.  If there
was some sort of effort to document the correct behaviour of a CSV
parser then I would be extremely happy to make mine conform to that
document.

If the end result was that Python ended up with a native CSV parser
then the process would have been worth the effort.

- Dave

-- 
http://www.object-craft.com.au


From phr-n2002a at nightsong.com  Fri Feb 22 04:07:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 22 Feb 2002 01:07:51 -0800
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1> <a54qfl$vl4$0@216.39.172.122> <hhu45a.4d1.ln@127.0.0.1>
Message-ID: <7xy9hlubo8.fsf@ruckus.brouhaha.com>

Jim Richardson <warlock at eskimo.com> writes:
> Thanks, this isn't a bad idea (I am not really python guru, so am easily
> overwhelmed sometimes :) I will have to think on it. Speed isn't much of
> an issue, as the data is only 9600 and each packet needs a checksum, and
> usually, only 8-10 packets a second. 
>  Another reply mentioned a "byte" operator, but I can't find that
> anywhere in my python docs. Is it something not in the default setup? 

That was me.  There's no byte operator.  I was using 'byte' as a
variable name to denote an integer variable that held an 8-bit number.
Sorry for the confusion.


From emile at fenx.com  Wed Feb  6 13:04:17 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 10:04:17 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu>
Message-ID: <a3rrii$1ahpac$1@ID-11957.news.dfncis.de>

"David Eppstein"
> Isn't all that list-copying (ary[1:]) going to make for a very
inefficient
> algorithm?

Oh... was that in the spec?  ;-)

And-that's-only-one-of-the-shortcomings-ly y'rs,

--

Emile van Sebille
emile at fenx.com

---------



From tth_1 at yahoo.com  Thu Feb 21 15:24:48 2002
From: tth_1 at yahoo.com (T H)
Date: 21 Feb 2002 12:24:48 -0800
Subject: Record Audio (Automation)
Message-ID: <140415ad.0202211224.40d8b34b@posting.google.com>

I need a simple way to record audio with a scripting language in
Windows.  I
think this can be done by executing sndrec.exe or sndrec32.exe (the
sound recorder that comes with windows that can be executed by going
to Programs->Accessories->Entertainment->Sound Recorder), however I
can't seem to be able to make it record.  I know you can use this
program to PLAY a WAV file by executing
sndrec32.exe /play /close filename.wav

However I was not able to make it record by executing
sndrec32.exe /record /close filename.wav

The command above I tried is just a guess, i have no idea if there is
a /record parameter.  Any idea where I can find a list of command line
parameters for sndrec32.exe?

If anyone knows any other way of automating the task of recording
audio using a scripting language (which executes a free audio
recording program), please let me know.  Any help is much appreciated.


From mcfletch at geocities.com  Sun Feb 17 01:30:21 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Sun, 17 Feb 2002 01:30:21 -0500
Subject: Directed Acyclic Graph Representation
References: <j9du6uckr4g1am3780qtig50jdhcaf6fcr@4ax.com>
Message-ID: <3C6F4DFD.1030701@geocities.com>

Boa Constructor has one for Python classes (but is GPL).  It uses a 
(generation-producing) topological sort to create the logical dependency 
tree, then draw it using wxPython calls.

I'd suggest just using the topological sort and doing your own drawing 
(draw each generation, draw links between each dependent object and its 
dependency).

Topological sort module available at:
	http://members.rogers.com/mcfletch/programming

HTH,
Mike


Tim Roberts wrote:
> Is anybody aware of a reasonable package for representing a directed
> acyclic graph in Python, and for rendering that graph in some meaningful
> way?  I need to build a dependency tree analyzer for, for example,
> determining the necessary link order for a set of object libraries.
> --
> - Tim Roberts, timr at probo.com
>   Providenza & Boekelheide, Inc.
> 


-- 
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From wrbt at email.com  Thu Feb  7 15:40:23 2002
From: wrbt at email.com (Larry)
Date: 7 Feb 2002 12:40:23 -0800
Subject: __class__ for 2.2 types
Message-ID: <2ec1bc1c.0202071240.44185531@posting.google.com>

hi, rookie here, I was hoping someone could explain to me why this happens:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> [].__class__
<type 'list'>
>>> int.__class__
<type 'type'>

why doesn't the int one return <type 'int'>? 

thanks in advance


From jkraska at san.rr.com  Sun Feb 24 21:33:06 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 25 Feb 2002 02:33:06 GMT
Subject: Parse variable into variables >Re: Apology
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <2240.820T2407T12845765threeseas@earthlink.net>
Message-ID: <r29j7uo1ckvhesgc7po0thcevmdg0jeukg@4ax.com>

>I can read a line from a file but then I want to parse the line into a set
>of variables. How do I do the parse?

>>> s = "a,b,c,d"
>>> tple = s.split(",")
>>> tple
['a', 'b', 'c', 'd']
>>> 

For more complex parsing questions, you'll have to ask more precise
questions.

C//



From MaximilianScherr at T-Online.de  Thu Feb 14 07:44:38 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Thu, 14 Feb 2002 12:44:38 -0000
Subject: socket server interrupt?
Message-ID: <a4gbfm+4h68@eGroups.com>

how can i make my socket server interrupt on ctr-c(i know with the 
keyboardinterrupt exception, but how?)?




From phr-n2002a at nightsong.com  Fri Feb 15 16:56:59 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 15 Feb 2002 13:56:59 -0800
Subject: saving complex data structures?
References: <Ocfb8.9438$bS4.618144@weber.videotron.net>
Message-ID: <7xlmdu76j8.fsf@ruckus.brouhaha.com>

"akhar" <akhar at videotron.ca> writes:
> How does one go about saving and restoring complex data structures? like if
> I have made a daatabase in python how do I save the contents for fast
> retrieval, in other ways than text files which do not always take into
> account structure? ( I thought of XML but ... I am wondering if there is a
> binary way?)

See the docs for the pickle/cpickle module.


From quinn at barf.ugcs.caltech.edu  Thu Feb 28 14:27:01 2002
From: quinn at barf.ugcs.caltech.edu (Quinn Dunkan)
Date: 28 Feb 2002 19:27:01 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu> <Vg%c8.4179$XT1.116344@news1.east.cox.net> <a5cnvo$2mdu$1@news.idiom.com> <UTAe8.9921$jC3.362611@news1.east.cox.net> <a5kp0e$2kj4$1@news.idiom.com>
Message-ID: <slrna7t144.7r6.quinn@barf.ugcs.caltech.edu>

On 28 Feb 2002 08:16:14 GMT, Jim Dennis <jimd at vega.starshine.org> wrote:
> Telling people to avoid eval() because it's "dangerous" and might
> lead to "insecure" code is a gross oversimplification.  It is better

In addition, security is usually not an issue for me.  Most of my code runs on
a machine with no passwords anyway (mine).  Most of the rest runs when I invoke
it with inputs I give it, not when some random outsider invokes it with random
outside inputs.  Avoiding eval on security grounds in many scenarios is like
avoiding the shell on security grounds because it evaluates arbitrary text.

It's like web hosting sites that let you write CGI scripts but don't give you a
shell login "because you can run anything with a shell!" never mind that you
can also run anything with CGI, including a shell, just less conveniently.  Or
maybe there's some more substantive reason they do that, I dunno.


From martin at v.loewis.de  Tue Feb 12 04:15:34 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 12 Feb 2002 10:15:34 +0100
Subject: isNumberType? a bug or a feature?
References: <20020212004024.035d005e.xunningy@ece.arizona.edu>
Message-ID: <m3eljr83ih.fsf@mira.informatik.hu-berlin.de>

Xunning Yue <xunningy at ece.arizona.edu> writes:

> Is this a bug or a feature of Python? Can anyone answer this?

It's a known bug. Instances always support the "number protocol" on
the C level, and this is what this function checks.

Regards,
Martin



From jkraska at san.rr.com  Sun Feb  3 00:04:33 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 05:04:33 GMT
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com> <a3hnft$cgu$1@newshost.accu.uu.nl> <3C5CC325.7B3FC988@engcorp.com>
Message-ID: <flhp5u4jalahtcm1att17i5vuj2evqoott@4ax.com>

>Come on, that's *obvious*.  Look at your keyboard: if you took
>away the Caps Lock key, there'd be a big empty space right there!
>Where else would they have put it?  Sheesh...

:)

Some place where I wouldn't accidentally hit it and then spend
the next few seconds wonder why vim is collapsing all my lines and
otherwise having one giant freak fest?

:)

C//



From mlh at vier.idi.ntnu.no  Sat Feb  9 04:55:57 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 9 Feb 2002 09:55:57 +0000 (UTC)
Subject: range(seq)?
References: <mailman.1013206876.19221.python-list@python.org>
Message-ID: <slrna69shc.jm3.mlh@vier.idi.ntnu.no>

In article <mailman.1013206876.19221.python-list at python.org>,
James_Althoff at i2.com wrote:
>
>[Magnus Lie Hetland]
>>   for x in range(seq): ...
>>
>> I.e. let range() take a sequence (or iterable) as an argument, and
>> generate range(len(seq)).
>
>[Jeff Shannon]
>>This seems more logical and intuitive to me, than the previous
>>suggestions.
>
>What Jython users will appreciate is that the above suggestion --
>although it might be handy for Pythonic structures and, possibly,
>might be "more logical and intuitive" than other suggestions --
>doesn't necessarily help with the world of structures one encounters
>outside the arena of 100% Python.

That depends on how well-behaved the container objects are (i.e.
implementing __len__).

> Java/Swing tables, for example, cannot be passed to the Python len
>() function (they are two-dimensional, for starters, and are not defined as
>a sequence of sequences).

OK. But these won't work with the normal "for x in container" either,
right? It seems that when dealing with this sort of object one would
either have to write a wrapper (e.g. for use with adapt()) or accept
that they don't present a Pythonic interface for iteration, e.g. by
using the plain old "for i in range(table.getRowCount())". Sounds
quite fair to me.

> So such an extension to the range() function
>won't make it any easier to deal with (certain) Java structures in Jython.

That's quite true.

>PEP 276, for example, suggests a simple way to deal with index-based
>structures that have been designed outside the sphere of Python practices
>while also making it easier at times to deal with Pythonic designs.

Maybe. I'm not full -1 on PEP 276 myself, but I don find the concept a
bit odd... That one can iterate over a number; Hm...

>
>Jim
>

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From jkraska at san.rr.com  Sat Feb 23 01:40:10 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 23 Feb 2002 06:40:10 GMT
Subject: HELP! Obscure Python Reentrancy Question
References: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com> <ur8ncpxbs.fsf@ctwd0143.fitlinxx.com>
Message-ID: <hdde7ukvfch6jpfovghvm92abaiq26q2i2@4ax.com>

>How have you allocated interpreter state(s) versus your C++ scheduler
>threads (or fibers) and how are you maintaining the GIL?

I haven't considered how and when to grab the GIL, and what it
means from the perspective of two different cooperative threads,
each of which may (or may not!) be executing a Python code
fragment, at some partial state of completion. Let me help:

def f(task):
	do()
	some()
	task.Step(1)
	other()
	stuff()
	task.Succeed()
	
In the above example, do() and some() are ordinary internal or
external python procedure calls. task.Step(1) kicks this codelette
back to the scheduler, which is written in C++. The scheduler is
dealing with other tasks, some in Python, some not, which are
basically doing the same thing. Eventually, the scheduler comes
back to the task, and then other() and stuff() are executed.
task.Succeed() then kicks this task back to the scheduler and
it is deallocated. It is an uncategorical error to end a task
without a success call of some kind.

>When the other thread in 5 resumes and "contains" more Python, does
>that mean that from the perspective of the Python script you are
>"returning" from the call in 4, or does that mean there's some
>independent Python code you are just starting to execute?

Well, each Python code fragment, much of the time, never really
"finishes". Thing of them as complex coroutines. So the scheduler
may be invoking an independent piece of Python code, or it may
awaken older code. Technically, it may reawaken the code which
just went to sleep.

>It seems to me that depending on how you want to link your Python code
>execution with your scheduler thread states, you may want to have an
>interpreter (or at least thread state) per thread,...

I have tens of thousands of threads. They access eachother's data
in some cases. They send messages to eachother, indirectly, through
the simulation queue, sometimes.

Basically what's going on here is that the code is bunched up into
lock steps which only occur at defined points in _simulation time_.
Someone using the API writes "task.Step(20)" and then knows that
the code which follows this won't happen until 20 simulation seconds
have jumped ahead from simulated current time. This is managed by
kicking the current task back to the scheduler and putting an entry
on a priority queue.

Theoretically, at some time in the future, I might want to have
a worker pool of threads, to support running this on SMP machines.
It's okay if Python can't take a lot of advantage of that, but
I do need to make it behave well and not block.

I didn't think that Python was structured to allow multiple interpreters
in a single process?

If it is, how are module imports resolved across interpreters? I need
module globals to behave globally.

C//




From erno-news at erno.iki.fi  Fri Feb  1 02:27:58 2002
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 01 Feb 2002 09:27:58 +0200
Subject: Python on Tandy Model 102?
References: <mailman.1011062662.29105.python-list@python.org> <83n0yxexo0.fsf@panacea.canonical.org> <kud6zs8676.fsf@lasipalatsi.fi> <a39vg5$8ml$0@216.39.172.122>
Message-ID: <ku8zad3blt.fsf@lasipalatsi.fi>

In article <a39vg5$8ml$0 at 216.39.172.122>, bokr at oz.net (Bengt Richter)
writes:

| On 30 Jan 2002 12:49:01 +0200, Erno Kuusela <erno-news at erno.iki.fi> wrote:
|| 
|| i noticed this the other day:
|| 
|| [/usr/local/bin/python2.1]
|| This machine benchmarks at 10526.3 pystones/second
|| [/usr/bin/python2.1]
|| This machine benchmarks at 16129 pystones/second
|| 
|| the debian-provided /usr/bin/python2.1 is dynamically linked against
|| libpython.so whereas the one in /usr/local/bin is not, i can't
|| think of other major differences (both have threading enabled, at least).
|| 
| Maybe compiled with different optimization?

i tried it just to be sure. i did apt-get source python, and did
make -f debian/rules build, and then tested using
cd builddir; LD_LIBRARY_PATH=. ./python /path/to/pystone.py

then i took the orig.tar.gz and extracted it, and ran
./configure; make; ./python /path/to/pystone.py

i verified both were using the same compiler and CFLAGS (gcc 2.95.4,
-O2 -g) with the obvious variation of -fPIC in the debian-patched
version.

the results did not change. 

  -- erno





From srh232 at nyu.edu  Thu Feb 21 12:04:27 2002
From: srh232 at nyu.edu (Shane Hoversten)
Date: Thu, 21 Feb 2002 12:04:27 -0500
Subject: Simple regexp problem
Message-ID: <3C75289B.8050504@nyu.edu>

Hi all,

	This question has gotta be real simple.  I want to remove preceeding 
whitespace from a string.  Seems like I should be able to say (using 
another example to make it easier to see):

re.sub("^1", "", "11111junk")

	and get "junk".  Except it doesn't work.  What I get is:

'1111junk'

	In other words, it replaces only the first "1", and no others.  I've read 
the dox and the howto's and this is such a simple thing.  I can do it 
another way, but I can't figure out why this doesn't work.  Can someone 
help?

Thanks,

Shane



From emile at fenx.com  Fri Feb  8 11:29:17 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 8 Feb 2002 08:29:17 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <st888.7971$oI4.33917985@newsserver.ip.pt> <a3retd$19cgik$1@ID-11957.news.dfncis.de> <d0380df9.0202080516.7a76905d@posting.google.com>
Message-ID: <a40un1$1b5ava$1@ID-11957.news.dfncis.de>

"Hugo Martires" <hugomartires at hotmail.com> wrote in message
news:d0380df9.0202080516.7a76905d at posting.google.com...
[testing floating point problem]
>
> I know that the problem is in the internal storage of the numbers, but
> i canot understand how to solve that.
> I just want to make a simple calculation and an if statement.
>
> Tanks for help

Perhaps one of these may help:

>>> '%.2f' % myArray[1]
'0.60'


class tolerate:
    def __init__(self, val, giveortake):
        self.hi = val + giveortake
        self.lo = val - giveortake
    def __eq__(self, other):
        return self.lo <= other <= self.hi

for a in (.595, .599, .6, .601, .605):

    if a == tolerate(.6, .001):
        print '\n\n',a, "OK"
    else:
        print '\n\n',a, "not OK"

because-round(.6,2)-is-0.59999999999999998-ly y'rs,

--

Emile van Sebille
emile at fenx.com

---------



From gustav at morpheus.demon.co.uk  Sat Feb 16 07:44:41 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Sat, 16 Feb 2002 12:44:41 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <a4jf10$och3$1@ID-59885.news.dfncis.de>
Message-ID: <q0ls6uc1vgoqrf7m2du58u8s75gvhkbprd@4ax.com>

On Fri, 15 Feb 2002 18:03:13 +0100, "Thomas Heller" <theller at python.net>
wrote:

>Have you looked into /F's squeeze? Should exactly do what you want,
>from the description...

Looks pretty close... I'll have to look into it a little. Thanks for the
pointer.

Paul.


From starx at pacbell.net  Sat Feb  9 13:21:31 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 18:21:31 GMT
Subject: Is this..... evil?
Message-ID: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4>

So i've been poking around with ways of doing this messaging thing, and 
i think up this (sorry for the inanity of some of the names, i was sick 
and tired when i worte this :):    	


class Message:
    type = "none"
class FightoMessage(Message):
    type = "fighto"
class SpelloMessage(Message):
    type = "spello"

class Reciever:
    def message( self, message ):
        methname = 'rec_' + message.type
        if hasattr(self, methname):
            if callable(getattr(self, methname)):
                getattr(self, methname)(message)


class F(Reciever):
    def rec_fighto(self, message):
        print "FIGHTO!!"

class S(Reciever):
    def rec_spello(self, message):
        print "SPELLO!!"
    def rec_fighto(self, message):
        print "Damnit i'm a wizard not a whacker!"
    def rec_none(self, message):
        print "Er, what?"


I dont think a better solution i could ever have seen. It does all the 
work of a big ugly switch statment, with none of the uglyness OR the 
hard-codedness. I can define new message types at will and only update 
things that i WANT to recieve them.

Then i realized it would have taken me three days to do this in C++, 
compared to the 15 or so in python, and the former solution would have 
been some kind of ugly hack involving rtti and what have you. I think 
python may have finally risen to the complete and utter top in my mind 
(O)_(O).

Anyway, i was just wondering, since this seems so great, is there 
anything wrong with doing things this way that i'm not seeing?

Also, is there any real reason to have the message classes inherit when 
the only thing that seems to identify them is an attribute?

Thanks %)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From bkelley at wi.mit.edu  Mon Feb 25 10:14:28 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Mon, 25 Feb 2002 10:14:28 -0500
Subject: strings and sort()
References: <mailman.1014641682.25067.python-list@python.org> <7xadtxln11.fsf@ruckus.brouhaha.com>
Message-ID: <3C7A54D4.7000105@wi.mit.edu>

Paul Rubin wrote:

> "Jason Orendorff" <jason at jorendorff.com> writes:
> 
>>But:
>>  Flat is better than nested.
>>  Sparse is better than dense.
>>  Readability counts.
>>
>>That's why Python is the way it is.
>>
> 
> But: dense
>              can
>                    be
>                            more
>           readable
>   than
>                        sparse.
> 
> See?
> 

I first read this as "dense can be more than readable sparse" which, of 
course, nearly splayed coffee over my keyboard.

Brian Kelley



From jeff at ccvcorp.com  Wed Feb 20 16:10:18 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 20 Feb 2002 13:10:18 -0800
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <3C7410BA.BEBEAD8B@ccvcorp.com>


Rajarshi Guha wrote:

> for i in d.keys().sort():
>   do something
>
> I get:
>
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: loop over non-sequence

sort() sorts lists in place, it does not return a value.  (Well, technicially,
it returns None...)  This was, IIRC, a deliberate design decision.  If
mylist.sort() returned the sorted list, then it would be intuitive to expect
that mylist is unchanged, and the sorted list is a copy.  That's not the case,
though.  While sort() could have been specified to return a copy, there are
many times when one would want to sort a list in-place (such as to save memory
with a very large list).  It's easy to make a copy from an in-place sorted
list, but not easy to avoid the copy in other case.

So, in short, what you need to do is:

keys = d.keys()
keys.sort()
for key in keys:
    ....

Yes, it's an extra couple lines of code, but it isn't much more typing, and it
makes it a little more clear just what's happening.

Jeff Shannon
Technician/Programmer
Credit International





From debl2nonospammy at bellatlantic.net  Thu Feb 28 00:45:04 2002
From: debl2nonospammy at bellatlantic.net (David Lees)
Date: Thu, 28 Feb 2002 05:45:04 GMT
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org> <3C7E198C.DF6E7566@nospam.linux3dgraphicsprogramming.org.nospam>
Message-ID: <3C7DC3EA.1827DB65@bellatlantic.net>

They published 'The Quick Python Book', which I have found to be quite
useful and accurate for a beginner like me.

David Lees


nlin wrote:
> 
> Susan Capparelle wrote:
> 
> > We're a publisher of high quality computer books currently seeking
> > authors for a number of Python related books.
> 
> As a published book author with two well-selling titles to my name, I'd
> just
> like to warn potential authors about my negative experience with
> Manning.
> After responding to a post similar to the above, I presented Manning
> with a
> book idea. My experience with Manning was marked by their lack of
> technical
> expertise on the subject matter, a failure of upper management to
> understand
> my book's premise, their lack of faith in the author's competence and
> vision,
> and their unbalanced reliance on unqualified "reviewers".
> 
> I lost months lost reworking my material to try to accommodate their
> reviewers
> (in the case of my book, some of these were self-proclaimed experts who
> knew
> little about the topic I was writing about), and to compound the misery,
> my
> earnest attempts to rework the focus of the book were met only with a
> curt
> message from upper management saying "Why the change. I thought your
> book's
> focus was XXX" (where XXX was not even the original focus, but only a
> tangential
> aspect). I never had the feeling that Manning (or at least the upper
> management) was really interested in trying to understand my vision for
> the
> book or make an earnest attempt to work with me - it was almost as if
> the
> publisher was the author's adversary instead of ally!
> 
> Eventually I gave up on them, went elsewhere, and got my book published.
> I was
> interested in writing a book, not making an eternal sales pitch.  It was
> a
> waste of good months working with Manning and attempting to appease
> their
> reviewers and to convince them of my vision. Manning trusted their
> reviewers -
> again, many of whom had dubious or inflated qualifications - more than
> they
> trusted the author - not a very productive situation for the author.
> 
> Maybe things have changed at Manning. If anyone has better experiences
> with
> Manning, feel free to share.
> 
> One author's opinion,
> N Lin
> http://linux3dgraphicsprogramming.org


From mal at lemburg.com  Fri Feb 15 05:32:10 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 15 Feb 2002 11:32:10 +0100
Subject: ANN: PSF Contribution Documents
Message-ID: <3C6CE3AA.4B8604F4@lemburg.com>

The Python Software Foundation (PSF; http://www.python.org/psf/) 
was created to (eventually) hold the intellectual property (IP)
for the Python programming language.

In order to be able to succeed in this, contributions to the
Python programming language will have to be covered by legally
valid contribution agreements. To make this process as simple
as possible, the PSF members have come up with three different
contribution agreements.

The current versions of those agreements have been posted to
a new mailing list at:

    http://mail.python.org/mailman/listinfo/psf-forms-discuss

If you want to participate in the final discussion of these
documents, please sign up to the list.

Thank you,
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From bokr at oz.net  Mon Feb 25 12:33:00 2002
From: bokr at oz.net (Bengt Richter)
Date: 25 Feb 2002 17:33:00 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com> <a5cjc0$cb5$0@216.39.172.122> <7x7kp2ayp4.fsf@ruckus.brouhaha.com>
Message-ID: <a5dsgc$7dv$0@216.39.172.122>

On 24 Feb 2002 21:59:51 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:

>bokr at oz.net (Bengt Richter) writes:
>> but I didn't get an extra call trying your version:
>> 
>>  >>> reduce(lambda x,y:(y.f(),0)[1], items, 0)
>>  this is a
>>  this is b
>>  this is c
>>  0
>
>Didn't you?  'items' had three elements, but four values got printed.
True, but the zero was the interpretive loop printing the final value of
the reduce expression, which is not evidence of a y.f() call. All those
were logged by "this is ...".

My version of your reduce evaluated to None, so nothing extra printed,
but that was because the last y.f() call returned None, and the 'and 0'
actually didn't get a chance to do anything. I should have used
'y.f() and None or None' to guarantee a final None to prevent the
final interactive printout, if that was the goal.
>
>> you can use the 'and 0' ploy to get a zero length
>> list comprehension also:
>> 
>>  >>> [ 0 for x in items if x.f() and 0]
>>  this is a
>>  this is b
>>  this is c
>>  []
>
>Nice!  That's the best solutions so far.

Actually, I think I like

    reduce(lambda x,y: y.f(), items, 0)

best, since we're ignoring the final result anyway.

Regards,
Bengt Richter



From James_Althoff at i2.com  Thu Feb  7 14:19:17 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 7 Feb 2002 11:19:17 -0800
Subject: 2001 Turing Award
Message-ID: <OF5A92159D.F5FA1C05-ON88256B59.0069ED9D@i2.com>

Some news.

Jim

===========================================

----- Message from Gary Chapman <gary.chapman at mail.utexas.edu> on Thu, 7
Feb 2002 08:41:00 -0600 -----


Subject:  2001 Turing Award
Please feel free to pass on this message:

Dear friends,

I'm happy to pass on the news that the 2001 Turing Award, the highest
award in computer science, has been awarded to Johan Dahl and Kristen
Nygaard of Norway, for their invention of object-oriented
programming, the most widely used programming paradigm in the world
today.

The formal citation of the award is on the Web at
http://www.acm.org/announcements/turing_2001.html. The Association
for Computing Machinery will present the A.M. Turing Award, its most
prestigious technical honor, at the annual ACM Awards Banquet on
April 27, 2002, at the University of Toronto in Canada.

It has been my privilege and honor to serve on the selection
committee for the Turing Award, in a group of five people, and I am
particularly pleased by this year's award. Congratulations, Kristen!

Best to all,

-- Gary

Gary Chapman
LBJ School of Public Affairs
University of Texas at Austin
Austin, TX
gary.chapman at mail.utexas.edu




From rhamilt at pacbell.net  Sat Feb 16 22:30:38 2002
From: rhamilt at pacbell.net (Rick Hamilton)
Date: Sat, 16 Feb 2002 19:30:38 -0800
Subject: Have to try
Message-ID: <004501c1b763$78bca9c0$b278c53f@pacbell.net>

OK - I know I am going to get fried here but - I am out of options. I'm a
programmer (Java, VB and C++) and am taking a class on Operating systems.
The nut teaching this course is asking us to do a ton on programming in
Python. OK -
So I am trying to do a (what should be) simple thing. I have a string stored
in a variable. I want to search that string for the occurrence of a sub
string. I have tried
Find(s,sub)
Rfind
MyString.find(sub)

Nothing - I have not yet figured out the appropriate way to treat a string
and am heading for a very weak book that we have as a reference. Anyone will
to toss a fellow programmer a helping had I'm willing to get flamed to get
this over with.

Thanks

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020216/6d781e23/attachment.html>

From pobrien at orbtech.com  Sun Feb 17 18:39:01 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Sun, 17 Feb 2002 23:39:01 GMT
Subject: Redefining "\n"
References: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>
Message-ID: <paXb8.6535$tu6.627126@newsread2.prod.itd.earthlink.net>

Take a look at the os module, in particular, os.linesep.

--
Patrick K. O'Brien
Orbtech

"Dale Strickland-Clark" <dale at riverhall.NOTHANKS.co.uk> wrote in message
news:nl7v6ugpaa2skti66dlocutacafljsnlh0 at 4ax.com...
> Is there some way to redefine the "\n" escape sequence?
>
> When preparing strings for the Windows clipboard, I need \x0d\x0a for
> a newline which is a bit cumbersome to insert all over the place.
>
> It would be neat if I could make "\n" insert some other sequence.
>
> I know I can do a translation before sending the string to the
> clipboard but this would be nicer.
>
> --
> Dale Strickland-Clark
> Riverhall Systems Ltd
>




From MaximilianScherr at T-Online.de  Fri Feb 15 19:03:06 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Sat, 16 Feb 2002 00:03:06 -0000
Subject: thread question
Message-ID: <a4k7jq+3073@eGroups.com>

if a thread (object) is for example a class, would it be possible to 
call the class objects of this thread from another thread?
or how could something similar be done?




From david.abrahams at rcn.com  Tue Feb 26 16:52:13 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Tue, 26 Feb 2002 16:52:13 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com> <a5gk1u$hlu$1@bob.news.rcn.net> <3c7befb8$1@giga.realtime.net>
Message-ID: <a5gvnl$s7k$1@bob.news.rcn.net>

"Ira D. Baxter" <idbaxter at semdesigns.com> wrote in message
news:3c7befb8$1 at giga.realtime.net...
> > What's in it for the compiler writer people?

> Full employment? More importantly for EDG anyway, a broader market base.

That only counts if someone is willing to pay them for the technology to
make documentation or wrapping tools. My customers all insist on open source
in these areas, so I don't think it's going to make EDG excited... but who
knows, maybe I'll ask them.

> Now, if what you want to do is compiler-like, then throwing
> away the comments, line numbering, number formats,
> and preprocessor instructions is just fine.  And you
> generally only have to focus on one "file" at time.

You can't throw away line numbering if anyone is going to debug the
generated code. If you have the line numbering and the source code, you can
get the comments and the preprocessor instructions. Number formats? I
presume you mean 0xF vs. 15 vs. 1.5E1? I can see how that could be relevant
to a few jobs, but none I want to do in the near term.

> If, however, you want to modify the sources of the program,
> *retaining* all that information, then you can't use
> a conventional compiler technology front end.
> You need to have something that can parse the sources,
> *including* the preprocessor directives.
> One of DMS's unique properties is its ability
> to do that (with some constraints) for C and C++
> code.

What are the constraints? If, like most tools out there which are not
compilers, it doesn't handle the full C++ language (less export), I'm not
interested.

> A second unique property is the ability to parse
> entire suites of files in the same session, so
> that you can build analysis tools that operate
> across file boundaries (i.e., multiple compilation units)
> and make modifications that move code/data
> from one source file to another to accomplish
> refactoring activities.

It sounds neat. The PDT appears to do this by merging multiple parse runs. I
suppose that your approach could be more-efficient, but AFAIK the C++
compilation model is not very friendly to re-use of information across
translation units, due to visibility and ordering dependencies.

> A third unique property of such generalized front ends
> is the ability to *mix* notations from multiple programming
> languages in patterns to enable translation from one langague/dialect
> to another.
>
> Of course, our hope is to be fully employed building/selling
> reengineering tools rather than compilers.

Been watching your company on-and-off for a few years now. It sounds like
fascinating work, but you don't present your message in a way that's
compelling and understandable to me (and I'm a languages guy). I'd be really
interested to see the "killer application" of your technology in action.

-Dave





From andtherewaswater at yahoo.com  Fri Feb  8 08:30:27 2002
From: andtherewaswater at yahoo.com (A.A)
Date: Fri, 08 Feb 2002 17:30:27 +0400
Subject: Bug ? xml.dom.minidom
Message-ID: <3C63D2F3.79EB997F@yahoo.com>

Hello,

In Python 2.2, when I try the following code either on the interpreter
or in a .py file,

----
from xml.dom.minidom import *
----

it returns an error saying xml has no attribute called dom.minidom.

However when you cd into the python 2.2 directory and try the code
again, it works !!!

$ cd /usr/lib/python2.2
$ python2
>>> from xml.dom.minidom import *
>>>

What is happening ???


From justin at iago.org  Thu Feb 28 11:49:35 2002
From: justin at iago.org (Justin Sheehy)
Date: Thu, 28 Feb 2002 11:49:35 -0500
Subject: How to have multi-threaded server log to one file ?
In-Reply-To: <97ae44ee.0202272013.2094b2b@posting.google.com> (shriek@gmx.co.uk's
 message of "27 Feb 2002 20:13:45 -0800")
References: <97ae44ee.0202261855.15c97708@posting.google.com>
	<f72dac0d.0202270140.97b925f@posting.google.com>
	<97ae44ee.0202272013.2094b2b@posting.google.com>
Message-ID: <x7r8n58sbk.fsf@ra.iago.org>

shriek at gmx.co.uk (Stephen) writes:

> I had a look at the effbot queue-example and think I understand the
> Queue class but adapting my program to share the queue instance in the
> first place is making my head spin.  Namely, how do I get the same
> Queue object into both MyHandler and MyServer in the first place ?

Pass the Queue instance to both of those at their instantiation time.

-Justin

 






From core.lists.python at core-sdi.com  Thu Feb  7 12:57:33 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Thu, 7 Feb 2002 14:57:33 -0300
Subject: IPX through Python
Message-ID: <008f01c1b000$eb41a140$8642a8c0@corelabs.coresdi.com>

Hi!

Does enybody knows or have an example of  how to implement IPX using Python?
I need to send some simple packets using IPX.

Tnks!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020207/6ecc0840/attachment.html>

From threeseas at earthlink.net  Sat Feb 16 10:22:06 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 15:22:06 GMT
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <3c6df547_2@corp.newsgroups.com>
Message-ID: <4772.812T386T6443632threeseas@earthlink.net>

On 16-Feb-02 01:09:10 Joshua Muskovitz <joshm at taconic.net> wrote:
>"Timothy Rue" <threeseas at earthlink.net> wrote
>> Ten years???? it was Feburary 1988 when I identified the nine commands.
>>
>> I never claimed to be a programmer and never wanted to make the VIC but
>> rather someone to use it. But I also don't, and never will want to be
>> abused by any programmer wanting to take credit and property rights to the
>> VIC.

>Well, after 14 years then (happy anniversary!) of ranting about it with
>NOBODY else even desiring to take credit for it, don't you think you should
>stop worrying?  You're only going to be abused by programmers who see it for
>what it is.  Irrelevent.

>All you've said, all this time, is that everything boils down to nine
>things.  Yippee.  My entire genetic makeup can be done in four.  Doesn't
>mean I find it useful to attempt to function purely on a cellular level.

You realy do have a comprehension problems huh?

Certainly everyone does understand in reading and responding to post here,
they actually make use of all nine action. It's physics, but they may
think it's me and that somehow their reactions will cause a dismissal
of.... well physics.

Lets see now:

AI - start and stop, change interfaces - Uh start up newsreader..

PK - keep track of where you are - Pick up where you left off on the thread..

OI - get input - read with eyes

IP - input from - internet and monitor

OP - output to - keyboard usenet postings

SF - one step at a time - damn this non-polyphony qwerty keyboard...

IQ - what's the meaning - whats a good non-target meaning I can attach to
                     confuse tim...

ID - identify posters - hey there is one by tim, now I know to be negative..

KE - constraints to apply to meanings and poster named tim rue - must be
        negative and anti-productive


>> The computer industry has alot of ethical problems in this area of taking
>> everything the can from the users, from their ideas, to their debugging
>> efforts (feedback), to their money in payment for what came from them, and
>> now rental....and always it is never quite right but lacking always... for
>> upgrading recycling of consumer entrapment abuse.
>>
>> That's what evidence ..... well what is, is what is. You know what you
>> are.

>As opposed to the ethical problems in say the energy futures industry, or
>savings and loans, or any of those other trifling scandals.  People of poor
>moral character exist in all aspects of society.  I fail to see any evidence
>they cluster around the programming profession.  Just because there are
>companies out there who aren't the nicest, doesn't mean that PROGRAMMERS as
>a group are to blame.

>You seem to have this amazingly huge and likely malignant chip on your
>shoulder about something, and I don't think it has anything to do with
>autocoding.  Perhaps we can help you.  Tell us more about the bad people who
>screwed you over so many years ago.  You'll feel better.  We'll feel better.
>Programmers learn better by example anyway.  Tell us what they did which
>made you so angry, and we'll promise to try not to do that.  We may even
>apologize.  Wouldn't that be nice?

In the 80/20 rule, or even the 90/10, most all other industries have
productive working value of 80% or better, and problems of 20% or much
less. only about 16% of the software industries projects get out on time
and within budget. The hardware industry of computers is better than that
but also highly incompatable.

>> There never will be enough programmers, but there is always exactly the
>> perfect number of user to do all the things a user needs to do for
>> themselves. Given the right freedom for the users, creating tools.

>That's right.  Let's all return to the world of hunter/gatherers, where
>nobody takes advantage of others, and we all struggle to survive.  If you
>despise the evil technology so much, why do you continue to use it?  That
>seems very contradictory.

Better yet to become master over the beast.

There is no contridiction, only your perception of what you want to see
vs. what is.

Perhaps you can provide a reasonable alternative for those who do not want
computers involved in any part of their life? No?!

Guess that only leaves becoming master over the beast then!!!!

you can start by growing up. Perhaps beyond your celluar level to realize
you can apply the nine actions at all levels.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jeff at ccvcorp.com  Fri Feb  1 14:12:50 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 01 Feb 2002 11:12:50 -0800
Subject: Conventions for iterators
References: <a3cu29$grp$1@newsreaderm1.core.theplanet.net>
Message-ID: <3C5AE8B2.BFDCB94E@ccvcorp.com>

jlbec at evilplan.org wrote:

> With "one modification invalidates
> the iterator," you cannot have code like:
>
> for item in list:
>     if matches(item):
>         list.remove(item)

Note that Python's current (non-iterator) behavior does not behave well in
this sort of situation, either.

PythonWin 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2001 Mark Hammond (MarkH at ActiveState.com)
>>> mylist = range(10)
>>> for i in mylist:
...  print i
...  mylist.remove(i)
...
0
2
4
6
8
>>>

As Tim Peters has pointed out, any time you mutate the object that you're
iterating over, you're asking for weirdness.

Jeff Shannon
Technician/Programmer
Credit International




From scarblac at pino.selwerd.nl  Thu Feb 21 02:41:39 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Feb 2002 07:41:39 GMT
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com>
Message-ID: <slrna798vs.6ot.scarblac@pino.selwerd.nl>

Jason <caljason76 at yahoo.com> wrote in comp.lang.python:
> How do I sort the characters in a string, so far I use this:
>         a="qwerasdfzxcv"
>         b=[x for x in a]
>         b.sort()

Yes. Instead of b=[x for x in a] you can use b = list(a).

> Why doesn't sort() return the sorted list.  I would like to chain it
> to other operations:
>         b=[x for x in a].sort()

sort() sorts the list in place, probably because that's usually what you
want. Methods that change the object in place typically don't return the
object, or people might think that the object wasn't changed.

Like

b=[4,3,2,1]
c=b.sort()

People might believe that b wasn't changed. To avoid confusion, you can't do
that.

-- 
Remco Gerlich


From SBrunning at trisystems.co.uk  Thu Feb 14 08:01:52 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 14 Feb 2002 13:01:52 -0000
Subject: Recommendations for cross-platform GUI Bindings ?
Message-ID: <31575A892FF6D1118F5800600846864DCBCC57@intrepid>

> From:	Steven M. Castellotti [SMTP:SteveC at innocent.com]
> Sent:	Thursday, February 14, 2002 10:48 AM
> To:	python-list at python.org
> Subject:	Recommendations for cross-platform GUI Bindings ?
> 
> Greeting all--
> 
> 	I'm writing an Open Source software application that allows children
> to
> create their own video games, and would like to hear anyone's
> recommendations for GUI bindings.
 
Have you considered pygame? (<http://pygame.org/>)

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From threeseas at earthlink.net  Fri Feb 15 20:45:49 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:49 GMT
Subject: Fear of autocoding!
References: <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk>
Message-ID: <1476.811T2040T12515414threeseas@earthlink.net>

On 11-Feb-02 16:01:45 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Mon, 11 Feb 2002 10:59:36 GMT, Fredrik Lundh <fredrik at pythonware.com>
>wrote:
>>    "People tend to hold overly favorable views of their
>>    abilities in many social and intellectual domains. The
>>    authors suggest that this overestimation occurs, in
>>    part, because people who are unskilled in these
>>    domains suffer a dual burden: Not only do these
>>    people reach erroneous conclusions and make un-
>>    fortunate choices, but their incompetence robs
>>    them of the metacognitive ability to realize it"
>>
>>    http://www.apa.org/journals/psp/psp7761121.html
>>
>>of course, Rue is going to argue that this doesn't apply to him, but
>>to everyone else.  after all, he KNOWS he's smarter than everyone
>>else, and we're just too stupid to realize how stupid we are ;-)

>Either that or we're part of the Evil Programmers Conspiracy to make
>programming hard.


Well you certainly don't want to help make it easier!!!!

Which you could have done by not doing anything, in comparison to what you
have done.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From gustav at morpheus.demon.co.uk  Sat Feb 16 07:52:55 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Sat, 16 Feb 2002 12:52:55 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
In-Reply-To: <3C6D8B63.5070909@geocities.com>
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <3C6D8B63.5070909@geocities.com>
Message-ID: <0als6ucp5bet6vah3h71ci5i1iuq49um96@4ax.com>

On Fri, 15 Feb 2002 17:27:47 -0500, "Mike C. Fletcher"
<mcfletch at geocities.com> wrote:

>wxImages can be loaded from strings.  See the wxPython Wiki entry on 
>working with images (particularly the sections on PIL and Numeric):
>
>http://wiki.wxpython.org/index.cgi/WorkingWithImages
>
>You can probably use PIL to get GIF/JPEG/Whatever formatted data into 
>the RAW format for use by wxImage

and Chris Liechti <cliechti at gmx.net> wrote:
>ok just tried this:
[...]
>you can use wxPython to convert the images to strings, save them to a 
>zip using the zipfile module. the image sizes have to be saved 
>separately. i've choosen to stuff them into the filename.

Thanks for both of these pointers, guys. Looks like just what I need.

I'll go and work through all the suggestions I have and come up with
something suitable. I'll report back on what I manage to produce.

Paul.



From gcordova at hebmex.com  Tue Feb 12 09:54:18 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 12 Feb 2002 08:54:18 -0600
Subject: just a moron trying to get python working on win98 and person
	al web  server
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A3C@SRM201002>

"no problemo"

Just installed yesterday the Python2.2 installer and the win32all installer.
Right this moment, I made a tiny python program:

================================
<%@ Language=Python %>
<%
#
# Teoricamente, este es un ASP escrito en Python.
# Esperemos que funcione.
#

class Message:
  def __init__(self, message):
    self.message = message
  def __str__(self):
    return self.message

msg = Message("Hola, estoy escrito en Python.")

Response.write(str(msg))
%>
================================

and saved it into a "default.asp" file. So, I browsed
to "http://localhost/" and bingo, I got a little web
page with "Hola, estoy escrito en Python" on it.

The file MUST have the ".asp" extension for it to work
with the PWS server; second, the first line MUST have
"<%@ Language=Python %>" in it, to inform the ASP framework
the language it's written it.

>From there, you're flying.

Good luck.

-gustavo
-- 
Gustavo C?rdova Avila
*  gcordova at sismex.com <mailto:gcordova at sismex.com> 
* 8351-3861 | 8351-3862



From magnus at thinkware.se  Thu Feb 21 07:07:54 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 13:07:54 +0100
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com> <258fd9b8.0202201733.3888e711@posting.google.com> <mailman.1014287322.17283.python-list@python.org>
Message-ID: <3C74E31A.1000203@thinkware.se>

M.-A. Lemburg wrote:

> The egenix-mx-* packages should compile out of the box
> using cygwin, not sure about the other extensions.


That's great. I guess what I was trying to point out was
that normal windows installs for python modules are unlikely
to work with cygwin. And this will cause some inconveniences
(and confusion if you aren't aware of the problem).






From edcjones at erols.com  Tue Feb 12 14:40:55 2002
From: edcjones at erols.com (Edward C. Jones)
Date: Tue, 12 Feb 2002 14:40:55 -0500
Subject: Reading and writing C structs from Python
Message-ID: <3C696FC7.1090600@erols.com>

Suppose I have a C struct for example

typedef struct {
float x;
int n[4];
} test;

and a pointer to an instance of the struct

test* t = (test*) malloc(sizeof(test));

I would like to wrap this in a python object "py_t" so that (in Python) 
"x = py_t.x" reads "t.x" and "py_t.x = 13.0" writes to "t.x". Ditto for 
"py_t.n[i]". Ideally I would like to do this using a C function 
something like:

PyObject* PyObject_FromCStruct(void* t, char* format)
void* PyObject_AsCStruct(PyObject* pyo, char* format)

The format strings used in module "struct" would be fine.

My guess is that this is a job for the notoriously confusing "buffer 
interface". Does anyone have any sample code for problems like this?

Thanks,
Ed Jones



From jkraska at san.rr.com  Tue Feb  5 18:38:20 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 05 Feb 2002 23:38:20 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <mailman.1012925350.31242.python-list@python.org>
Message-ID: <7kr06u8dctsk6b54qbthbligih0np0v31t@4ax.com>

>Seems like a fun experiment, but I'm not sure where I think it ought
>to end up.  Some languages (such as Scheme) have a clear need for
>user-defined forms.  Python, both in structure and culture, does not
>seem to be crying out for macros.

That may or may not be a case of the chicken in the egg. But indeed,
it's an experiment; too see what form they would appear in, what the
issues are, what works, what doesn't work, and whether or not defining
a domain specific sublanguage provides any use at all. This is a case
of "implement it to find out."

C//



From mpoulin at verinet.com  Fri Feb  1 16:25:44 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Fri, 01 Feb 2002 16:25:44 -0500
Subject: Help with time
References: <a3ecqt$16bk9d$1@ID-98166.news.dfncis.de> <3c5ad0bf$0$2200$7586b60c@news.frii.com> <a3emd2$17tc99$1@ID-98166.news.dfncis.de>
Message-ID: <3c5b22e4$0$2201$7586b60c@news.frii.com>

In article <a3emd2$17tc99$1 at ID-98166.news.dfncis.de>, "MDK" <mdk at mdk.com>
wrote:

> Yes, that was it. Thanks! I never would have thought of that.

Glad I could help.
A friend of mine used to call that sort of
thing an "awwshitical" problem. When you
finally figure it out you slap your forehead
and say "Aww... shit!"


From justin at iago.org  Sun Feb  3 18:45:58 2002
From: justin at iago.org (Justin Sheehy)
Date: Sun, 03 Feb 2002 18:45:58 -0500
Subject: lambda expressions
In-Reply-To: <20020203115247.GA15877@kub.nl> (Bas van Gils's message of
 "Sun, 3 Feb 2002 12:52:48 +0100")
References: <20020203115247.GA15877@kub.nl>
Message-ID: <x73d0i2kp5.fsf@ra.iago.org>

Bas van Gils <bas.vangils at home.nl> writes:

> I've been wondering, lately, about lambda-expressions. Most programming
> languages support 

Well, most lispy languages, and a few others.

They're really useful for talking _about_ programming languages.
However, you asked about using them _in_ programming...

> I wondered why (mostly because I find them hard to read, valueing
> _readable_ code). Collegues at work couldn't give me a satisfying
> answer so ... can anyone explain to me why lambda-expressions are so
> useful?

They are extremely useful in languages designed around their use.  For
instance, Scheme wouldn't be nearly as fun without lambda.  There are
a few differences between Scheme (an example of a language where
lamdba is essential and good) and Python (where lambda is somewhat
rightfully sneered at).

In Scheme, lambda is the usual way to define a function.  In Python,
lambda is a poor cousin to "def".

The structure of Scheme is such that all functions can be described by
lambdas, whereas Python's lambda is crippled by being much more
restricted than def with regard to its body.

I could go on, but the difference basically comes down to the fact
that lambda expressions are a fundamental part of how some languages
describe functions, but in Python they clearly stand out as not
seeming much like the rest of the language as designed.

So, to answer your question: in Python, they are not really all that useful.

The times I tend to use them:

- When I want to pass a simple callback to another function or class.

- When the argument to map or filter is so simple that a lambda actually
  makes my code easier to read (for me, anyway).
 
- In debugging at the REPL.

Even in those cases, I use a named function more often than not.

-Justin

 







From tripps81 at yahoo.com  Thu Feb 14 11:52:16 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Thu, 14 Feb 2002 23:52:16 +0700
Subject: preference tree like in jdk 1.4
Message-ID: <5.1.0.14.2.20020214234713.033fc280@202.95.131.154>

i was looking into the preferences api in the jdk 1.4 
documentation [1] and wonder how this could be done as simple as 
possible in python. in short, preferences is a tree branch of 
options you can store and load, just like ini files or registry settings.

i was thinking of an empty function:

  def empty(): pass

and then use it to build complex tree of preferences:

  pref = empty
  pref.personalData = empty
  pref.personalData.firstName = 'John'
  pref.personalData.lastName = 'Cleese'
  pref.windowSettings = empty
  pref.windowSettings.windowA = empty
  pref.windowSettings.windowA.x = 200
  # and so on...

to store or load this pref object, just use the pickle mechanism.

my question is, is there a simpler, more beautiful, elegant way?

t

[1] 
http://developer.java.sun.com/developer/technicalArticles/releases/preferences/




From jimd at vega.starshine.org  Thu Feb 28 04:25:59 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 28 Feb 2002 09:25:59 GMT
Subject: SSL Server Sockets?
References: <a5ijos$1565$1@news.idiom.com> <a5j0e9$rkg$1@dahlia.singnet.com.sg>
Message-ID: <a5kt37$63j$1@news.idiom.com>

In article <a5j0e9$rkg$1 at dahlia.singnet.com.sg>, Ng Pheng Siong wrote:
>
>According to Jim Dennis <jimd at vega.starshine.org>:
>>  I know everyone recommends m2crypto --- but m2crypto-0.07-snap2
>>  doesn't build on my Debian box.  I get errors like:

>> Generating wrappers for Python
>> _bio.i : Line 30. Syntax error in input.
>> _bio.i : Line 31. Syntax error in input.
>> _bio.i : Line 32. Variable %constant multiply defined (2nd definition
>> ignored).
>> _bio.i : Line 32. Syntax error in input.
>> _bio.i : Line 33. Variable %constant multiply defined (2nd definition
>> ignored).
>> _bio.i : Line 33. Syntax error in input.

> What does 'swig -version' say on your box?

> Only this version is supported:

>    ngps at madcap:~$ swig -version
    
>    SWIG Version 1.3.6u-20010725-2356
>    Copyright (c) 1995-1998
>    University of Utah and the Regents of the University of California
>    Copyright (c) 1998-2001
>    University of Chicago
    
>    Compiled with CC
:r! swig -version


SWIG Version 1.1 (Build 883)
Copyright (c) 1995-98
University of Utah and the Regents of the University of California

Compiled with c++

 Thank you Ng, I think we've found the problem.

$ dpkg -l swig\*
+++-==============-==============-============================================
ii  swig           1.1.p883-4     Generate scripting interfaces to C/C++ code.
ii  swig-doc       1.1.p5-3       Documentation files for SWIG
un  swig-doc-pdf   <none>         (no description available)
un  swig-examples  <none>         (no description available)
un  swig1.3        <none>         (no description available)

# apt-get -f install swig1.3


:r!swig -version


SWIG Version 1.3.10u-20020113-1638
Copyright (c) 1995-1998
University of Utah and the Regents of the University of California
Copyright (c) 1998-2001
University of Chicago

Compiled with CC

 ... still no joy.  



 SWIG Version 1.3.10u-20020113-1638
 Copyright (c) 1995-1998
 University of Utah and the Regents of the University of California
 Copyright (c) 1998-2001
 University of Chicago

 Compiled with CC

  ... still no joy.


_m2crypto_wrap.c: In function `ssl_verify_callback':
_m2crypto_wrap.c:867: too few arguments to function
`SWIG_NewPointerObj'
_m2crypto_wrap.c:868: too few arguments to function
`SWIG_NewPointerObj'
_m2crypto_wrap.c: In function `ssl_info_callback':
_m2crypto_wrap.c:895: too few arguments to function
`SWIG_NewPointerObj'
_m2crypto_wrap.c: In function `smime_read_pkcs7':
_m2crypto_wrap.c:3395: too few arguments to function
`SWIG_NewPointerObj'
_m2crypto_wrap.c:3401: too few arguments to function
`SWIG_NewPointerObj'
make: *** [_m2crypto] Error 1

 ... so I guess and it's still too closely coupled to the 
 swig version so 1.3.10u is too new and 1.1 is (understandably)
 too old.

 Ng, are you planning on updating this to Python2.2 using the 
 latest swig?  Could we find an (international) volunteer to 
 maintain a Debian package?  


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


From gerhard at bigfoot.de  Mon Feb 25 09:19:11 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 25 Feb 2002 14:19:11 GMT
Subject: Threads in python?
References: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com>
Message-ID: <slrna7khuv.1m1.gerhard@lilith.my-fqdn.de>

Fernando Rodr?guez wrote in comp.lang.python:
> 
> I'm going to start coding a little utility that needs to retrieve webpages and
> process them. 

You might find linkchecker a good model for your program.

http://linkchecker.sourceforge.net/

> However, I worried that the whole app might stall if one server takes too long
> to respond or is offline.

You'll need to set a timeout for the sockets then.

http://www.timo-tasi.org/python/timeoutsocket.py

> Does Python support threads? "Learning Python" doesn't mention it... 

Sure Python does support threads. Again, linkchecker uses threads, so
it might be worth looking at its code.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From DeepBleu at DeepBleu.org  Fri Feb  1 15:48:45 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Fri, 1 Feb 2002 14:48:45 -0600
Subject: telnetlib read_until()( question
References: <u5lqt76m0ogfd8@corp.supernews.com> <a3er56$17h5lp$1@ID-11957.news.dfncis.de> <u5lsfu2m34gpa9@corp.supernews.com> <a3esav$17p2n5$1@ID-11957.news.dfncis.de> <u5lvb7fj4qjrcc@corp.supernews.com>
Message-ID: <u5lvr7c7ke0671@corp.supernews.com>

"DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
news:u5lvb7fj4qjrcc at corp.supernews.com...
A small correction:
> Could it be that the script itself running on the web browser is being
> dropped by the firewall while the script is waiting for processing to
finish
> through telnet?  Is this possible as you are suggesting?

This should read as:
> Could it be that the script itself running on the web *server* is being
> dropped by the firewall while the script is waiting for processing to
finish
> through telnet?  Is this possible as you are suggesting?





From ivano.delpiccolo at degroof.be  Thu Feb 14 03:21:26 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Thu, 14 Feb 2002 09:21:26 +0100
Subject: spawn and standard output
References: <ce0b6758.0202131543.1707e80c@posting.google.com>
Message-ID: <3c6b736c$0$12224$4d4efb8e@news.be.uu.net>

If you use popen3, you'll have 3 files with stdin,stdout and stderr.
the 2 last are readable and are 2 open files.
well you don't need to create sp?cific file, popen do it for you.
"Nigel Moriarty" <nw_moriarty at yahoo.com> wrote in message
news:ce0b6758.0202131543.1707e80c at posting.google.com...
> Hi all,
>
> When I spawn a process from a python script on MSW, the standard
> output doesn't go to the terminal where I started the script.  I would
> like to view this output for debugging.  Where does the output go and
> how can I view it?  Do I need to use popen or can I redirect to a
> file?  Any other suggestions?
>
> Nigel




From jnxburxke at xatt.netx  Mon Feb 18 20:09:39 2002
From: jnxburxke at xatt.netx (John Burke)
Date: Tue, 19 Feb 2002 01:09:39 GMT
Subject: Operating system requirements
Message-ID: <3C71A556.3E7BF27E@xatt.netx>

What are the operating system requirements for python to run on a Windows
pc. Thank you.

John


From aleax at aleax.it  Tue Feb 26 02:00:58 2002
From: aleax at aleax.it (Alex Martelli)
Date: Tue, 26 Feb 2002 08:00:58 +0100
Subject: [Python-Help] advocacy help ( Python vs VB )
In-Reply-To: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>
Message-ID: <02022608005803.10662@arthur>

On Tuesday 26 February 2002 01:25, Stefan Kuzminski wrote:
> I need to convince my co workers and erstwhile managers to build a
> production system in Python.  They're coming from a VB world and are
> fearful to detach from the Redmond mothership.  I need hard facts, not
> warm and fuzzies about syntax but names of big companies using Python and
> things VB just *cannot* do ( threads, regular expressions.. ? ).

Matthew Sherborne already gave you "the positive" so I'll focus on the
"against VB" side.

Ah, but WHAT VB are we "selling" against?  Microsoft is in the process
of changing VB *incompatibly* -- you would need to study it again and
rewrite just about all of your code to migrate to old VB6, which MS does
not like any more, to new VB.NET, which means betting your company
on MS's unproven, fascinating but risky .NET strategy.

http://groups.google.com/groups?q=visual+basic+incompatible+group:comp.lang.python+author:alex+author:martelli&hl=en&scoring=r&selm=9f93pa11ioh%40enews1.newsguy.com&rnum=1

for a short summary that I wrote of the backwards-incompatible changes
in the new Visual Basic versus the old one.

In theory the new VB can do anything, e.g. threads, and what it can't do,
you'll find or buy .NET assemblies that can and call them from VB.NET.

But, it IS all theory: *NObody* has yet shipped a real application nor
made money from VB.NET (except the authors of books and teachers
who are making a bundle from the justified panic of VB programmers
as they start to realize the huge extent of the changes and desperately
purchase training despite the parlous state of budgets -- nice timing,
Microsoft).  Are you willing to bet your company and your own personal
future on this blue sky concept?  MS will eventually get it right (they have
the money to finance that), but it generally takes them more than one
release AND patches etc -- will YOU still be around in, say, 3-4 years,
when finally the .NET platform is solid and stable?  Companies that
jumped onto the bandwagon of Windows back in Windows 1.0 bitterly
regretted it -- it took until release 3.0 (some would say 3.1) before
Windows at least established itself, and until Windows 95 before it
became decently productive to program for.

If your plans are on sticking with old VB6, realize that Microsoft itself
is now actively trying to force you all to migrate.  Have you ever been
on such a tug-of-war with a large vendor?  It IS reasonably frequent,
as vendors make their money only if you do migrate (which they call
"upgrade").  Soon you cannot buy new machines with the old software
any more, bugs aren't fixed any more, new functionality never appears
(of course: it all goes into the NEW incompatible system as one of the
inducements to you to migrate!).  It's a very uncomfortable and risky
situation for all involved.

It's a terrible time for users of VB because of this -- the old one is
becoming stranded and actively fought against by the vendor, the
new one is not yet stable and accepted in the market.

Look at Python as your lifeboat in this Titanic disaster!  Python does
not work by one release every 3 years or so -- a "big bang" -- it
does twice yearly releases with smaller changes and care for back
compatibility, PLUS bugfix releases of the stable previous release.

Today you can choose between 2.2 (the newest release, quite stable
but due for a bugfix 2.2.1 any day now) and 2.1.2 (the previous one, 
very stable with all bugs out), and just about every script you wrote
in 2.1.anything keeps working in 2.2 anyway (differently from VB.6
vs VB.NET where just about every program you wrote with the former
breaks in the latter).  And Python is simpler (fewer bugs) AND better
to start with (just about all changes VB6 -> VB.NET are in the
direction of taking VB a bit closer to Python ... MS is thus practically 
confessing that Python is a better language than VB6!-).


Python is free and, for free, comes with usable IDE's, but simpler
ones than the rich environment you get with VB6 Professional and
Enterprise.  But you can also purchase commercial IDE's which take
Python well beyond the VB IDE's abilities: take your pick between
PythonWorks, Wing IDE, BlackAdder ... the very fact of such friendly
but earnest competition makes each IDE better -- MS's lack of
competition makes them relatively stagnant and less keen to serve
you, the customer.  Personally, I don't (yet) use commercial IDE's
because so far I've found myself at ease with the free tools (but I'm
not doing lots of rich GUI's which is one of the main things IDE's
help with -- both VB's and Python's).  If you have some budget for
SW tools, look into the demo versions of the three IDE's above if
you need such things as "painting GUIs".

All of the tools (except a few windows-specialized ones such as
win32all / Pythonwin) are cross-platform.  ActiveState, which you
addressed in your original request, also have windows-specific
tools, such as Visual Python for integration of Python with MS
Visual Studio .NET, as well as cross-platform ones, such as
Komodo based on Mozilla's Gecko, but I didn't look at those so
I don't know what they provide.  Cross-platform means you can
do all your work on WIndows if you want, AND still tomorrow if a
market opportunity appears for you on (say) Linux or BSD or
Solaris, you can easily meet it.  Nice ace to have up your sleeve
FOR FREE -- heads you win, tails you break even (should there
never materialize such an extra market opp for your products).


Give it a try -- you'll never look back!-)

Alex



From timr at probo.com  Wed Feb  6 02:53:04 2002
From: timr at probo.com (Tim Roberts)
Date: Tue, 05 Feb 2002 23:53:04 -0800
Subject: Problem while uploading a binary file using cgi module
References: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU> <u5t6k65m7869d7@corp.supernews.com>
Message-ID: <u1o16ugtmp7kn5meao6nm85l0bm1htj3vl@4ax.com>

"DeepBleu" <DeepBleu at DeepBleu.org> wrote:
>
>Wang
>Please send the code indented next time.  You know, you may know it inside
>out.  But poor readers like me do not know your intentions and indenting
>makes it easier to read.

His code WAS indented, using hard tabs.  I'm guessing Microsoft is
"helping" you here by having Outlook Express skip leading hard tabs.  Next
time, try using a real news reader.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From tim.one at comcast.net  Fri Feb 22 18:26:46 2002
From: tim.one at comcast.net (Tim Peters)
Date: Fri, 22 Feb 2002 18:26:46 -0500
Subject: Stackless Platform Independence?
In-Reply-To: <m37kp5i5a4.fsf@chvatal.cbbrowne.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEKJNOAA.tim.one@comcast.net>

[Christian Tismer]
> As you might have noticed, Stackless has changed quite
> much:
>  - Platform independence is gone.

[Christopher Browne]
> I hope not...  I would hope to hear that "platform indepedence has
> _ARRIVED_"

No, he meant what he said.  All the good changes are bought at the cost of
needing some platform-specific assembler, to transform the platform C stack
from an Insurmountable Obstacle into just another piece of manipulable
state.

the-net-win-is-overwhelming-ly y'rs  - tim




From phr-n2002a at nightsong.com  Mon Feb 25 06:20:31 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 03:20:31 -0800
Subject: Remote debugging ?
References: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com>
Message-ID: <7xadtxajuo.fsf@ruckus.brouhaha.com>

Lothar Scholz <llothar at mailandnews.de> writes:
> Is there any debugger for python that allows me to do remote debugging
> from a windows 2000 client to a python script running on a Linux
> Server.
> 
> It must not be freeware. 

If freeware is unacceptable, you better stop using Python as well.


From gerson.kurz at t-online.de  Tue Feb  5 14:28:09 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Tue, 05 Feb 2002 19:28:09 GMT
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org>
Message-ID: <3c603127.18127765@news.t-online.de>

On Tue, 5 Feb 2002 12:48:29 -0700, "Orr, Steve" <sorr at rightnow.com>
wrote:
> 
>I need to migrate a web application from PHP to Python and I have a
>requirement to save state info WITHOUT using cookies where much of the state
>data MUST not be visible to the end users. The PHP application allows users

I'm not quite sure if that is what you need, but give it a try.

1) Write an outer frame, so that the user cannot see the inner links.
2) Encoded the state in the href links. In the code below, you can
fill a plain python dictionary with any kind of data you want, and
encode/decode it in the generated URL. As long as you keep in mind
(1), the user won't know.

Here is a very simple example:

----------------------------------------------------------------------
#!/usr/bin/python -u

import cgi, sys, os, string, binascii, copy

# this func generates a dictionary ("a") that
# has all the state variables you wanna keep
# and returns an encoded string
def GenerateLinkData(value=None):
    global args
    a = copy.copy(args)
    if value is not None:
        a['value'] = value
    s = str(a)
    return binascii.b2a_hex(s)

# helper function for generating a link to a state 
def GenerateLink(value=None):
    return "/cgi-bin/state.py?cmd=%s" % GenerateLinkData(value)

# decode "cmd" arg
def GetArguments(form):
    result = { 'value':1 }
    if form.has_key('cmd'):
        try:
            new_dict = eval(binascii.a2b_hex(form['cmd'].value))
            for key in new_dict.keys():
                result[key] = new_dict[key]
        except:
            pass
    return result

form = cgi.FieldStorage()
args = GetArguments(form)  

print "Content-type: text/html\n"

value = args['value']

print '<html><body><p>State = %d.' % value
print '<a href="%s">Click me.</a>' % GenerateLink(value+1)
print '</p><body></html>'
----------------------------------------------------------------------

Something like that, only bigger, and with sourcecode:

http://www.p-nand-q.com/cgidocs.htm



From mwh at python.net  Thu Feb 21 07:02:08 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 12:02:08 GMT
Subject: Reloading modules
References: <a51tt1$msq$1@news1.ucsd.edu> <u664rm510.fsf@python.net>
Message-ID: <u1yffm45s.fsf@python.net>

Michael Hudson <mwh at python.net> writes:

> Jeff Davis <jdavis at empires.org> writes:
> 
> > Is there a way to say that all calls to import should check for a newer 
> > version, if a newer one exists then completely drop the old one and import 
> > the new one?
> 
> An import hook could do this.  Wouldn't be trivial to write, but not
> too hard, either.

This "seems to work".  It's probably still a bit fragile, but might be
handy during development, I guess.

import __builtin__, os, time

old_import = __import__

def my_import(*args):
    mod = old_import(*args)
    file = getattr(mod, "__file__", "")
    if file.endswith("c"):
        file = file[:-1]
    if not file.endswith(".py"):
        return mod
    mtime = os.path.getmtime(mod.__file__[:-1])
    itime = getattr(mod, "__importtime__", 0)
    if itime < mtime:
        reload(mod)
    mod.__importtime__ = time.time()
    return mod

__builtin__.__import__ = my_import

Cheers,
M.

-- 
  ARTHUR:  But which is probably incapable of drinking the coffee.
                    -- The Hitch-Hikers Guide to the Galaxy, Episode 6


From fredrik at pythonware.com  Sat Feb  9 18:58:29 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 09 Feb 2002 23:58:29 GMT
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org> <B88AE186.48CD%jonathan@onegoodidea.com> <a43ipn$s37$1@panix2.panix.com> <B88B46C1.48F3%jonathan@onegoodidea.com>
Message-ID: <FIi98.24853$l93.4540249@newsb.telia.net>

Jonathan Hogg wrote:

> A couple of friends and I were swapping timings the other day and we got
> this table for a 'make -j2 all' of Python plus all the Mac modules:
>
>    What         CPU        Time
>    ----------------------------
>    iBook        600Mhz     7:23
>    TiBook       500Mhz     6:27
>    G4       2 x 800Mhz     3:32
>    G4       2 x   1Ghz     2:45

build times???  where are the pystones?

(I run python programs a lot more than I compile the inter-
preter -- and I'm sure I compile the interpreter more often
than the average comp.lang.python reader ;-)

</F>




From sholden at holdenweb.com  Fri Feb  8 09:53:26 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 09:53:26 -0500
Subject: readlines() a file in 2 parts.
References: <87vgd85a7h.fsf@cosette.nampaz.com>
Message-ID: <tDR88.106087$KM2.4033605@atlpnn01.usenetserver.com>

"Jason Price" <jprice at cosette.nampaz.com> wrote in message
news:87vgd85a7h.fsf at cosette.nampaz.com...
> I have several files to read that logically break into 2 parts.  It would
> be much cleaner to write 2 different 'for line in file.readline() :'
loops,
> but the break point is different in each file, so I don't know the
starting
> line.

    for line in file.readline():

will iterate over the characters contained in the next line from file. The
readline() method returns a string, being the next line from the file, and
the for statement loops over the components of that string as a sequence
(i.e. the characters).

>         What I tried to do origionally was:
>
> for line in file.readlines() :

The problem *here* is that the readlines() method reads the whole file in.
Since you haven't assigned that result to anything, you have no way of
retrieveing the back half of the file once you've identified where the front
half ends and executed the break.

>    if condition :
>       break
>    else:
>       do thing 1
>
> for line in file.readlines()
>    do thing 2
>
> I tried both readlines and xreadlines, and neither seem to do what I'd
> like.  Is there a way to do this cleanly, or should I punt, and just make
> the loop more complex?
>
Punt! What you need is something like this:

    contents = file.readlines()

    for i in xrange(len(contents)):
        if contents[i] is the last line of the first part:
            break

    firstpart = contents[:i]
    secondpart = contents[i:]

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From christophe.delord at free.fr  Thu Feb 21 13:37:30 2002
From: christophe.delord at free.fr (Christophe Delord)
Date: Thu, 21 Feb 2002 19:37:30 +0100
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de>
Message-ID: <3C753E6A.9050201@free.fr>

The problem with sort is that it sorts the list as a side effect.

I found other solutions (but not as short as yours :-)

q=lambda L,c=cmp:(lambda f=lambda s:filter(lambda i:c(i,L[0])==s,L):L 
and q(f(-1))+f(0)+q(f(1)))()

or without filter:

q=lambda L,c=cmp:(lambda f=lambda s:[i for i in L if c(i,L[0])==s]:L and 
q(f(-1))+f(0)+q(f(1)))()

or

q=lambda L,c=cmp:L and(lambda m=[],e=[],M=[]:map(lambda 
x:{-1:m,0:e,1:M}[c(x,L[0])].append(x),L)and q(m)+e+q(M))()



Christophe.


Gerson Kurz wrote:

> I found an easier solution:
> 
> sort = lambda x:(x.sort(),x)[1]
> 
> 

--
Christophe Delord
http://christophe.delord.free.fr



From jgardn at alumni.washington.edu  Mon Feb 11 02:30:36 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Mon, 11 Feb 2002 16:30:36 +0900
Subject: Random Module help
References: <20020210215559.13516.00000455@mb-cq.aol.com>
Message-ID: <whK98.289$ab7.1102@news.hananet.net>

Patio87 wrote:

> I was wondering if any 1 could tell me how to use the random module for a
> blackjack game i want to make. I am a begginer so try to be easy with the
> dialouge. Thanks

You'd want a card class first of all that represents a card. I think 
something like this would be sufficient:

class card:
        def __init__(self, number, suit):
                self.number = number
                self.suit = suit

        def __str__(self):
                return "%s %s" % (self.number, self.suit)

To represent a group of cards, you can just use an array. So, to define the 
original stack of 52 cards, you'd do something like:

deck = []
for suit in ('heart', 'spade', 'diamond', 'club'):
        for number in (
                '2', '3', '4', '5', '6', 
                '7', '8', '9', '10', 'J', 
                'Q', 'K', 'A' ):
                deck.append(number, suit)

Now you want to shuffle the cards up. There are several ways to shuffle the 
cards. The way that I think is best is to use random.shuffle() because it 
is easy.

import random
random.shuffle(deck)

Now you have a shuffled deck of cards.

Good luck - the devil is in the details, of course.

Jonathan





From kfarmer at thuban.org  Mon Feb 25 02:46:26 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 25 Feb 2002 07:46:26 GMT
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com> <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com> <2rFc8.2002$ZC3.167000@newsread2.prod.itd.earthlink.net>
Message-ID: <mZle8.42888$1d1.1794957329@newssvr14.news.prodigy.com>

I haven't updated my wxPython installation from the last official release
(where I mentioned it to you).

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org


"Patrick K. O'Brien" <pobrien at orbtech.com> wrote in message
news:2rFc8.2002$ZC3.167000 at newsread2.prod.itd.earthlink.net...
> In what version of PyCrust can you reproduce this behavior? I don't get
the
> error in any version I have on my machine, which happens to be a Win98
> machine.





From david.abrahams at rcn.com  Tue Feb 26 13:33:06 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Tue, 26 Feb 2002 13:33:06 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com>
Message-ID: <a5gk1u$hlu$1@bob.news.rcn.net>

"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
news:6bf54244.0202260920.5920efc6 at posting.google.com...
> Hi David.
>
> Since I'm not a "computer guy" per se, I was wondering if you could
> clear up some more of my confusion.  I don't know much about how
> compilers work, but I would presume that any compiler (eg g++) would
> create some internal representation, similar to the .cil file, after
> parsing but before compiling.

After parsing but before code generation (parsing is just part of
compiling), but yes, most compilers work that way.

> Presumably the EDG front end is geared toward producing this information
> in a format more suited to being used by the "wrapper generators" in
contrast
> with a compiler which produces this information only for itself and with
> the sole goal of compiling an object file?

Not exactly. EDG makes a front-ends which they sell to vendors of full
compiler solutions (Intel, KAI, Comeau, Compaq, ...). As such, their
requirements for a well-defined and accessible intermediate representation
(IR) are higher than other front-end writers.

> What I'm getting at is that I'm confused as to why people wouldn't have
> used information from the compiler *itself* to automate the process of
> wrapper generation.

They did. EDG's front end is part of many of the best C++ compilers.

> Obviously the compiler writer people would have to
> collaborate with the wrapper generator people... but there must be
technical
> obstacles?

What's in it for the compiler writer people?

ask-not-what-your-compiler-can-do-for-you-ly y'rs,
Dave





From marklists at mceahern.com  Tue Feb 12 09:55:46 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 12 Feb 2002 06:55:46 -0800
Subject: searching backwards in a string
In-Reply-To: <7xk7tjayos.fsf@ruckus.brouhaha.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEGGKCAA.marklists@mceahern.com>

[Paul Rubin]
> I have a string and want to search backwards for a regexp, starting at
> a given index, like if you search backwards in a text editor.  Is
> there a way to do that with re?  The docs aren't clear about what's
> supposed to happen if the ending position to the pat.search method is
> less than the starting position, but I tried it and it doesn't seem
> to do what I want.  Any advice appreciated--thanks.

Short answer:  Reverse the string first.

Sample:

    #! /usr/bin/env python
    # findBackwards.py

    import re

    def reverseWord(s):
        l = list(s)
        l.reverse()
        return ''.join(l)

    def findBackwards(s, pattern):
        """Return the location of the first match of pattern in s, searching
backwards."""
        r = reverseWord(s)
        m = re.search(pattern, r)
        if m is None:
            print "%s not found in %s." % (pattern, s)
            return
        # Express location relative to original orientation of word.
        start = len(s) - m.start()
        print "%s found backwards in %s at %d." % (pattern, s, start)

    s = "foobar"
    pattern = "o"
    findBackwards(s, pattern)

    pattern = "z"
    findBackwards(s, pattern)

Running:

    $ findBackwards.py
    o found backwards in foobar at 3.
    z not found in foobar.





From peter at engcorp.com  Mon Feb 25 19:47:57 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 25 Feb 2002 19:47:57 -0500
Subject: Is it possible in Python too?
References: <3C7A6BA1.26999.19EEC27@localhost> <mailman.1014666949.20850.python-list@python.org>
Message-ID: <3C7ADB3D.CA98D513@engcorp.com>

Matthew Sherborne wrote:
> 
> > Is possible to set up a proxy name (address) in MS Explorer
> > browser from a running  Python program?
> > Thank you for your help.
> > Ladislav
>
> Yes but I don't know how.
> 
> Tell me how to do it in another language, and I'll tell you how in python ;)

I just set up a proxy in the Control Panel settings and searched
the registry.  Turns out the key is under HKEY_CURRENT_USER\Software\
Microsoft\Windows\CurrentVersion\Internet Settings.  ProxyServer 
is set to "10.2.3.4:999" (my test value) and, among other things,
ProxyEnable is set to 1.  Probably lots of documentation about this
stuff buried somewhere, if anyone cared.  I don't.

(Oh yeah, the winreg module would presumably handle this all,
if you knew what to do.)

-Peter


From gjlap at yahoo.com  Sat Feb 23 02:00:12 2002
From: gjlap at yahoo.com (Gil)
Date: 22 Feb 2002 23:00:12 -0800
Subject: pygame puzzler
Message-ID: <1eed326b.0202222300.4666822c@posting.google.com>

Hi everyone,

Can somebody help me here? I've installed SDL_mixer-1.2.1 from a
.tar.gz package without complaint except one:
"SMPEG-version>=0.4.3...no"

[caltrop at localhost examples]$ python sound.py
Traceback (most recent call last):
  File "sound.py", line 10, in ?
    import pygame.mixer, pygame.time
ImportError: No module named mixer


I don't get it; my ducks seem to all be in a line.Is that SMPEG thing
significant?

My sound card works and I ran test_linuxaudiodev.py successfully.

Any help greatly appreciated.

-Gil


From dickey at saltmine.radix.net  Sun Feb  3 07:24:07 2002
From: dickey at saltmine.radix.net (Thomas Dickey)
Date: 3 Feb 2002 12:24:07 GMT
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org>
Message-ID: <a3ja57$car$1@news1.Radix.Net>

Gerhard H?ring <gh_pythonlist at gmx.de> wrote:
> Le 02/02/02 ? 16:59, Lawrence Oluyede ?crivit:
>> 
>> Why in the std distribution of Python 2.2 there isn't the curses
>> module?

> If you're using Windows, that explains it ;-) There is indeed no curses
> module included in the Python binary distribution for Windows. IIRC the
> reason is that there is no ncurses library readily available on win32 to
> build the curses support against.

pdcurses would work (I'm too busy with other programs to add on python)

-- 
Thomas E. Dickey <dickey at radix.net> <dickey at herndon4.his.com>
http://dickey.his.com
ftp://dickey.his.com


From just at xs4all.nl  Sat Feb 23 10:11:27 2002
From: just at xs4all.nl (Just van Rossum)
Date: Sat, 23 Feb 2002 16:11:27 +0100
Subject: Adding static typing to Python
References: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu> <a55qob$ebu$1@thorium.cix.co.uk> <slrna7dhbm.5et.quinn@regurgitate.ugcs.caltech.edu> <u7f8mhek950ec6@news.supernews.com> <a588iq$mu2$1@bob.news.rcn.net>
Message-ID: <just-5D0C78.16112723022002@news1.xs4all.nl>

In article <a588iq$mu2$1 at bob.news.rcn.net>,
 "David Abrahams" <david.abrahams at rcn.com> wrote:

> I was privileged to hear a talk by Niklaus Wirth, the inventor of the
> original OO language, SIMULA, at OOPSLA 2001. [...]

  http://java.sun.com/people/jag/SimulaHistory.html
  http://www.acm.org/awards/turing_citations/dahl-nygaard.html

Wirth designed many languages, but not Simula.

Just


From mwh at python.net  Tue Feb  5 09:02:50 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 14:02:50 GMT
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com> <60018f02.0202050547.4836c5ff@posting.google.com>
Message-ID: <uvgdcowgq.fsf@python.net>

laymon at crd.ge.com (Marc Laymon) writes:

> Several people suggested that the script needs to be
> executable.  However, python is in
> /usr/bin/python and the script is executable:
> 
> (laymon) [23] which python
> /usr/bin/python
> (laymon) [24] ls -l retest.py
> -rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*
> 
> If the script is not executable, you get a "Permission denied"
> error, not "Command not found"
> 
> (laymon) [25] chmod -x retest.py
> (laymon) [26] ./retest.py
> ./retest.py: Permission denied.
> (laymon) [27] chmod +x retest.py
> (laymon) [28] ./retest.py
> ./retest.py: Command not found.
> (laymon) [29] ls -l retest.py
> -rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*
> 
> I can duplicate the error message with a perl script by changing
> "#!/usr/bin/perl" to "#!/usr/bin/foo", so it seems that for some
> reason, Unix can't find python, but since /usr/bin is in my path
> and I am explicitly giving the command "usr/bin/python",
> I can't figure out why.  Anybody have any other ideas ?

Have carriage returns snuck into your script?

try

$ head -n1 retest.py | cat -v

if the result is

#!/usr/bin/python^M

then that's your problem.

Cheers,
M.

-- 
  If a train station is a place where a train stops, what's a
  workstation?                            -- unknown (to me, at least)


From jason at jorendorff.com  Sun Feb  3 23:11:58 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 22:11:58 -0600
Subject: parsing an xml document with funky non-ascii characters
In-Reply-To: <be58dbe.0202031714.3ebee2df@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOECKDAAA.jason@jorendorff.com>

andrew writes:
> How do I deal with xml documents with characters like '?'?
>
> I have tried:
> 	- setting encoding="ISO-8859-1 in the xml doc itself

I put this in mine:

<?xml version="1.0" encoding="ISO-8859-1"?>

and it works.  It should work for you too, unless there's
a bug in the XML parser.  Note that the <? must be the first
characters of the file, though.


> 	- escaping the character in the doc: ('\x84')

This isn't a meaningful escape in XML, just 4 ordinary
text characters.  But you could do &#x84; and that might work.

## Jason Orendorff    http://www.jorendorff.com/




From sholden at holdenweb.com  Wed Feb 20 08:19:49 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 08:19:49 -0500
Subject: confusion regarding os.path.walk()
References: <mailman.1014141812.1766.python-list@python.org> <Xns91BABAD1C9BE8andrewdarwinwarscom@194.168.222.9> <slrna75bg5.hg.rmunn@rmunn.dyndns.org> <Xns91BB770F47AD4andrewdarwinwarscom@194.168.222.19>
Message-ID: <vnNc8.17041$Zp2.1755671@atlpnn01.usenetserver.com>

"Andrew Brown" <killspam at darwinwars.com> wrote in message
news:Xns91BB770F47AD4andrewdarwinwarscom at 194.168.222.19...
> rmunn at pobox.com (Robin Munn) wrote in
> news:slrna75bg5.hg.rmunn at rmunn.dyndns.org:
>
> > assuming
> > burrows.html is the "real" file and Burrows.htm is the symlink you
> > want to create, then the command you really want to be running is:
> >
> > ln -s burrows.html ./wombats/habitat/Burrows.htm
> >
> > This will create the symlink Burrows.htm -> burrows.html in the
> > directory wombats/habitat. If you did:
> >
> > ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm
> >
> > then what you would get would be: in the directory wombats/habitat,
> > there would be a symlink named Burrows.htm that tried to reference
> > wombats/habitat/wombats/habitat/burrows.html -- this is almost
> > certainly not what you wanted.
> >
>
> Yes. This is exactly the gotcha that got me. It's not at all obvious, and
> it's not really a python problem either. It seems to arise from the way
> that Unix makes symlinks: I had assumed that it first looks for the file
> to link to, and then makes the link file. But it turns out to be the
> other way around. So, though I had tried all sorts of path manipulations
> to ensure that the link was found relative to the file linked to, I never
> tried chopping the path of the target file, which would have solved my
> problem.
>
> Thanks for your help, and I hope this sometime helps a passing googler.
> Other people will surely fall into the same pit.

They surely will, and you've done a useful piece of learning (as well as
helping future Googlers). Next time, remind yourself: "There's a reason why
symbolic links are called symbolic"!

The path is, as you have discovered, generated by appending strings, not by
traversing the filestore tree.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From kfarmer at thuban.org  Sat Feb  2 01:15:23 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Sat, 02 Feb 2002 06:15:23 GMT
Subject: Pythonwin, DefWindowProc bug?
Message-ID: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com>

I'm trying to intercept the WM_NCHITTEST message, to selectively disallow
resize for given window edges. The following code results in

    (216, 661)
    43319512
    Traceback (most recent call last):
    File "D:\Python\wxAppBar\PyAppBar.py", line 52, in _OnWMNCHitTest
        self._LowHighWordsToLparam(win32api.GetCursorPos())
    TypeError: an integer is required

.. I am wondering where the error lies: my logic (am I creating lParam
incorrectly?), or is win32gui.DefWindowProc not accepting a wide enough
variable? I've got the constants elsewhere in the file, btw.

    def _LowHighWordsToLparam(self, lh):
        print lh
        print lh[1]*0x10000 + lh[0]
        return lh[1]*0x10000 + lh[0]

    def _OnWMNCHitTest(self, params):
        lr = win32gui.DefWindowProc(self._GetAppBarWnd(),
                                                        WM_NCHITTEST,
                                                        0,

self._LowHighWordsToLparam(win32api.GetCursorPos())
                                                        )
        if (lr == HTBOTTOM or lr == HTBOTTOMLEFT or lr == HTBOTTOMRIGHT):
            return HTBORDER
        else:
            return lr

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org




From tjreedy at home.com  Fri Feb  8 12:34:34 2002
From: tjreedy at home.com (Terry Reedy)
Date: Fri, 08 Feb 2002 17:34:34 GMT
Subject: [NEWBIE] List issue aka: "Why doesn't this code work?"
References: <mailman.1013184014.10227.python-list@python.org>
Message-ID: <K_T88.32933$X64.11034857@news1.rdc2.pa.home.com>

[posted and cc'd]
"Bruce Dykes" <bkd at graphnet.com> wrote
> What the heck am I doing wrong here?

Beats me.  Works fine here.  Copying your posted function (and filling
in rest) ...

>>> import string
>>>
>>> lines = [
...   'line1 header1 header2',
...   'line2 datum1  datum2']
>>>
>>> def process_numbers(source_list):
...     for number in range(len(source_list)):
...         entry = string.split(string.strip(source_list[number]))
...         print number, len(entry), entry
...
>>> process_numbers(lines)
0 3 ['line1', 'header1', 'header2']
1 3 ['line2', 'datum1', 'datum2']

gives exactly the output you expected.  So you must have actually run
something else.

Moral: always post what you actually ran (cut to minimal size) and its
output (cut and paste), not what you think you did run, or a
description thereof.

Terry J. Reedy






From mhammond at skippinet.com.au  Sat Feb 23 19:35:57 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 24 Feb 2002 00:35:57 GMT
Subject: cannot start pythonwin
References: <a58k0o$31ur$1@as201.hinet.hr>
Message-ID: <3C7835EB.4040809@skippinet.com.au>

zvonko wrote:
> i've just installed python 2.2 and the pythonwin extensions.
> however, when i try to start pythonwin, i get the following error:
> 
> 
> The application cannot locate win32ui.pyd (1157)
> 
> but the file is in the same directory where the pythonwin is.

This is probably because you installed Python itself for the current 
user, rather than for the entire machine.  In this scenario, win32all 
usually fails to install right at the end.

The workaround is to move python22.dll from your Python directory into 
your SYSTEM32 directory, then reinstall win32all.

Mark.



From dsavitsk at e-coli.net  Wed Feb 27 14:17:11 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Wed, 27 Feb 2002 19:17:11 GMT
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz> <3C7CFAF8.1080108@mxm.dk>
Message-ID: <Xgaf8.10287$pC1.1854469@newssrv26.news.prodigy.com>

"Max M" <maxm at mxm.dk> wrote in message news:3C7CFAF8.1080108 at mxm.dk...
> Beda Kosata wrote:
>

<snip>

> btw: if you really want to store it in a list you can get dict-like
> behaviour with list speed like this:
>
> keys = (
>      NAME, AGE, SEX
>      ) = range(3)
> def pseudoDict():
>      return [None]*len(keys)
>
> items = []
>
> person = pseudoDict()
> person[NAME] = 'Max M'
> person[AGE]  = 36
> person[SEX]  = 'Male'
> items.append(person)
>

<snip>

> regards Max M
>

is this a python idiom that people often use? cause i've not seen it before,
but it is really clever.




From mlh at vier.idi.ntnu.no  Wed Feb 13 12:11:30 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 13 Feb 2002 17:11:30 +0000 (UTC)
Subject: PullDOM docs?
Message-ID: <slrna6l7i2.8hs.mlh@vier.idi.ntnu.no>

I just wondered if perhaps some more documentation could be added to
the standard docs about PullDOM? What's currently available[1] would
certainly not be enough for me to figure out how to use it, while Paul
Prescod's introduction[2] is quite understandable.

It seems to me that for people who want simple XML processing (like
me), but who might at times find minidom a bit inefficient, pulldom is
a nice choice... And it seems it deserves at least a basic explanation
in the docs... (Perhaps I've missed a section elsewhere?)

[1] http://www.python.org/doc/current/lib/module-xml.dom.pulldom.html
[2] http://www.prescod.net/python/pulldom.html

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From spencer at spencerdoidge.com  Fri Feb 15 20:26:29 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Fri, 15 Feb 2002 17:26:29 -0800
Subject: win extension supports RS232 binary
Message-ID: <a4kc8e$ei5$1@news.efn.org>

Starting with Isaac Barona's SerialPort extension built on the win32
extensions, I have a receiver working now on binary data. The original
SerialPort handles ASCII, and you can use strlen() to determine the number
of bytes received. My update on Barona's module returns a tuple giving the
number of bytes received and a list containing those bytes.
In the spirit of the GNU license agreement, this source is up for grabs, and
here and now is where I notify any interested parties about that.
I would be glad to email the script to anyone interested. Later, when I have
things polished up a bit, I suppose I could post it at my web site, but I'm
not sure how useful that would be without periodic announcements that they
are there.
You probably have figured out that I'm a newbie at this open source stuff.
Any suggestions about how to handle this sort of thing?





From tjreedy at home.com  Tue Feb 19 12:15:32 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 19 Feb 2002 17:15:32 GMT
Subject: newbie here...
References: <mailman.1014111829.2043.python-list@python.org>
Message-ID: <UKvc8.54298$X64.19386135@news1.rdc2.pa.home.com>

"Paul Berg" <paul at spondo.com> wrote in message
news:mailman.1014111829.2043.python-list at python.org...

>I'm just new to Python,,,
>Could someone send me some script so i can relate to it.

Start with the tutorial at python.org, which has many snippets of
script.  It's only a couple of hours or so for an experienced
programmer.  There are also pointers to much other stuff on the web.
Most of the standard library modules are written in Python.  Read a
few of these to see examples of usually well-written production code.

> And also things like databases types it can use(if so),

Look at the db module in the library and look for references to
interface wrappers for various database programs.

>I'm known to be a quick learner so hopefully will be programming in
days.

If you are willing to start before learning all the ins and outs of
the language, it may only be a few hours.

Terry J. Reedy






From bergeston at yahoo.fr  Fri Feb 15 07:53:02 2002
From: bergeston at yahoo.fr (Bertrand Geston)
Date: 15 Feb 2002 04:53:02 -0800
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131> <3C6CAFF8.4D3B5092@dotcom.ee>
Message-ID: <df3b00ed.0202150453.49e54903@posting.google.com>

Martin Roos <moro at dotcom.ee> wrote in message news:<3C6CAFF8.4D3B5092 at dotcom.ee>...
> Scirious wrote:
> 
> > Dear friends;
> >
> > I've been learning JAVA for a while, but now I've heard avout Python.
> > Any coments concerning these two languages?
> >
> > TIA;
> > Scirious.
> 
> java is powerful and quite fast with jit,
> python is more fun and slower ....
> 
>  That's it ...
> 
>  ps. java has more api's (till now)
> 
> i like both of them ...
>  java is for serious kinda stuff , python is for testing etc.

I fully agree with that (but never had speed issues with Python), and
I would like to add:
- Java is for larges applications developed by teams, Python is for
your own needs, for prototyping (even new functionnalities in a Java
app with Jython and that's really cool)
- Java is probably easier to maintain, Java works on a kind of
platform which is robust, reliable, provide security facilities, ...
and Python is easy to extend, gives powerfull access to the system
(but, for obvious reasons, sometimes in a platform dependant way).
- Java documentation is better but Python is easier to try.
- ...

I like to see it like that (and I am kindly proud of my trademarked
metaphor :-):
               "Java is Lego, Python is Plasticine"

You know, Java is my wife and Python is my mistress. Don't ask me to
choose :-))

B.


From kccw at well.ox.ac.uk  Fri Feb  1 07:09:44 2002
From: kccw at well.ox.ac.uk (Kenny)
Date: Fri, 1 Feb 2002 12:09:44 +0000
Subject: What is a "bus error"?
In-Reply-To: <3C56448B.58833301@engcorp.com>
References: <qj658.1752$HL4.244689@newssrv26.news.prodigy.com>
 <a32tpp02vve@enews3.newsguy.com> <3C56448B.58833301@engcorp.com>
Message-ID: <Pine.LNX.4.43.0202011202560.4970-100000@zeon.well.ox.ac.uk>

Hello All,

	I recently installed Python 2.2-5 from Fink (Unix Software for
MacOSX) and tried running a few sample scripts/programs.
	However, some of these programs now give a "bus error" message -
what does this mean?

Kenny

*******************************************************************************
                     tout casse, tout passe, tout lasse.
*******************************************************************************





From bugcreator at gmx.de  Wed Feb 13 10:53:28 2002
From: bugcreator at gmx.de (Dominic =?iso-8859-1?q?Sacr=E9?=)
Date: Wed, 13 Feb 2002 16:53:28 +0100
Subject: Disabling creation of .pyc files?
Message-ID: <E16b1mB-0001Ls-00@mail.python.org>

Hi,

I'm writing a C++ program which embeds Python, and I'm a bit annoyed by the 
fact that for every .py file I'm importing, Python automatically creates a 
corresponding .pyc file, which persists after my program exits.

Is there a way to disable that feature?

Thanks,
    Dominic




From aahz at panix.com  Mon Feb 11 17:23:37 2002
From: aahz at panix.com (Aahz Maruch)
Date: 11 Feb 2002 14:23:37 -0800
Subject: Defending the Python lanuage...
References: <3C62EE4A.5090305@destiny.com> <200202081311.g18DBQGo020260@ratthing-b246.strakt.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <mailman.1013420657.25827.python-list@python.org>
Message-ID: <a49g99$bt4$1@panix2.panix.com>

In article <mailman.1013420657.25827.python-list at python.org>,
Laura Creighton  <lac at strakt.com> wrote:
>
>Have you read 'Engineering and The Minds Eye'?  I just finished it.
>Wonderful.  

How about _Things That Make Us Smart_?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From gerson.kurz at t-online.de  Tue Feb  5 12:10:52 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Tue, 05 Feb 2002 17:10:52 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <3c600bb5.8541953@news.t-online.de>

On Tue, 05 Feb 2002 05:36:47 GMT, Courageous <jkraska at san.rr.com>
wrote:

>I am currently experimenting with an implementation of hygienic macros
>in a custom Python parser as part of a larger experiment.

Great, I love macros! They make for excellent code obfuscation ;)

Seriously though, I think its a very good idea. I'm known for rather
odd views on CS, so some would argue when I think its a good idea it
actually isn't (remember I didn't like exceptions?). Anyway, I like
macros, and here are a few reasons for it: 

- they can provide generic solutions that do not require runtime
interpretation (OK, that line is blurred in python). In many cases,
rather than resorting to a C++ "template", you can use plain
oldfashioned macros, and have much more fun, too. [Please, nobody tell
me that the C++ template syntax is cleaner than the macro syntax - one
look at ANY of the STL sources will prove you wrong]. 

- one change in the macro definition can change loads of code, even
code you didn't know exist, giving you POWER over LITTLE HUMANS, HA HA
HA <evil grin>.

- conditional macro redefinition makes for interesting code and has
the potential to scare the maintenance people - a property also known
as "job security". 

2 and 3 are probably nonsense; and 1 shows the main problem: because
all python code is interpreted, you could have generated the code
programmatically in python in the first place and have far more power
at your hand than any "third-hand" preprocessor might give (if I
understand you correct, you don't want to have a preprocessor,
though).

See, I too miss "do-while", but I can easily write a textprocessing
tool in python, that replaces

do:
	function()
while expression.

with

function()
while expression:
	function()

or some more suitable solution - in python itself. (Or, "of curse", in
"the VIC" as we all know since last week) 




From op73418 at mail.telepac.pt  Wed Feb 20 16:58:50 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Wed, 20 Feb 2002 21:58:50 +0000
Subject: A bundle of questions from a Python newbie
Message-ID: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>

Hi all,

A few months ago I started groking at Python. I am not a computer
scientist or a professional programmer and in my activity (mathematics)
I hardly do any programming at all - and if I needed to I would almost
surely use something like Mathematica.

Still, I do some programming in my spare time and besides, since my
early teens I have been fascinated with computer programming. Along the
way I have "learned" several languages like BASIC, C, Java and Ruby, but
never actually got around doing anything useful with them, because of
lack of time, patience and self-discipline, and also because the
languages themselves never helped, forcing me to pay attention to
nitty-gritty boring details. And then came Python. The clean and
uncluttered syntax is just amazing. I actually have fun programming with
Python and that is why in these few months I've already done some COM
programming and dished out some GUI's via wxPython, assembling these in
a program that actually accomplishes something useful. This makes me
believe that Python is very benign and gentle towards newbies like me.
So much so that I am thinking in tackling some more serious projects.

But enough babbling. I have several questions on the rationale of
Python. I don't know if this is the best option, or if I'm breaking some
netiquette rule, but I've decided to bundle them in a single post. My
aim in making these questions is to understand what lies behind the
specific choices that were made in designing Python, to get myself
acquainted with the Pythonic mind-set so to speak.

Here they go:

1. The way I thought that an assignment like

var += 1 

worked was that the interpreter would go to the place pointed by var and
add 1. In particular, id(var) should remain constant. But this is not
what happens as a simple test shows it. So what is wrong in my
perception of what += does? Is this an implemetation issue?

2. What is the idea of having in the loop

for var in range(10):
    pass

the var variable still available after it? Wasn't it supposed to die (as
in being out of scope) when the loop ends? And is this a feature, or
something that we should not count on? Somehow, I believe that the a new
var should be created when the loop starts and then destroyed when it
ends avoiding any variable shadowing along the way.

3. This is not a question just a small complaint. I really miss the
do-until control flow in Python. I always have to waste some extra brain
cycles to convert it to the while 1, etc. idiom. Hey, we all have a
right to our complaints, right? ;-)

4. I have read the FAQ about the "self controversy" - why one has to
declare it as an argument and then use it explicitly in the body of the
method - but I really feel that there must be a better solution. I
thought something like this could be useful:

a. self doesn't need to be declared.

b. Distinguishing data members from local variables could be achieved by
instead of using self.whatever use .whatever - a single dot in front of
the name "whatever".

I don't expect that this solution is adopted (it would break all the
existing code - ahhh every Nero-like language designer's dream - wreak
havoc in the citadel of its users and start afresh), what I would like
to know is how bad/good is this idea.

5. The FAQ explains why adding methods to built-ins is disallowed (btw,
IMHO I find this to be one of the few strong points of Ruby when
compared to Python).

Still, What's the rationale behind allowing

class L(list):
    pass

L.answer = 42

but then not allowing 

L.__dict__['answer'] = 42

6. Is there a reason why 3.__abs__() does not work? After all 3 is an
object (in the wider sense of the word). And it does work if we assign
it to a variable, eg, 

var = 3
var.__abs__()

I know that code like the above (methods applied to number literals) is
probably rare, but consistence is also a good thing.

7. What is the reason why (new-style) classes do not allow assignment to
__bases__?

8. Can the id() of an object change? e.g. via the garbage collector
moving it around? I'm asking this, because in the negative case, mutable
user-defined classes could be made hashable by adding an attribute with
the id() of the object and then defining __hash__ to return this
integer.

P.S: I'm well aware that hashing in a dictionary relies on comparison by
value and the above would be by identity. A single consistent choice is
the best and I have no qualms on the by-value choice, still the
by-identity is sometimes useful.

9. What would be better in terms of speed?

for i in range(<whatever>):
    <whatever> 

Or to code an iterator like

from __future__ import generators

class Range:
    def __init__(self, end, start = 0, step = 1):
        self.end = end
        self.start = start
        self.step = step

    def __iter__(self):
        i = self.start
        while i < self.end:
            yield i
            i += self.step

and then use it in the for loop.

10. I have a class that works like a list in the sense that supports the
method __getitem__ and its relatives. But I'm at a loss as to what I
have to do to support the slice notation.

11. Generators seem to be a very powerful concept. Still,

a. Why can't a return statement have an expression?

b. Is there any way to "reset" a generator, e.g. upon a subsequent call
it starts as if it is being called for the first time?

12. This is not exactly about Python, but about programming in general.
I keep reading about continuations and coroutines and I would like to
know, in simple terms if possible, what the heck are they and in what
ways they differ from generators as they exist in Python.

My deepest-deepest thanks for any comments/answers.

With my best regards,
Gon?alo Rodrigues.



From phr-n2002a at nightsong.com  Thu Feb 21 11:41:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 08:41:34 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org> <7xit8rhebq.fsf@ruckus.brouhaha.com> <slrna7a6sv.ekg.scarblac@pino.selwerd.nl>
Message-ID: <7xvgcqpz29.fsf@ruckus.brouhaha.com>

scarblac at pino.selwerd.nl (Remco Gerlich) writes:
> > Because we've already seen, assignments in expressions are useful.
> > And blurring the statement/expression distinction has proven quite
> > successful in languages like C and Lisp.
> 
> Are you sure? writing 'if (a=3) {' must be the most common bug in C
> programs. I wouldn't call it successful. In fact, Python already caught
> three of those for me this week as syntax errors - that would have been bugs
> in C.

Oh come on, the most common errors in C are probably pointer errors.
Anyway, =/== is a dead horse--didn't we agree that using := for assigments
in expressions totally fixes it?  So there's no reason to keep coming
back to it.

> Also, putting multiple conceptual things on one line isn't good for code
> clarity.

That depends on your style and on the context.

> > Anyway, language attempts to impose "morality" on programmers tend to
> > annoy programmers more than they makes code any cleaner.
> 
> When Python flagged those typos this week, I was *happy*.

If you want Python to catch more typos for you, you're better off
asking for Python to check for variable declarations (I don't mean
static types, I just mean "local x") and flag undeclared variables
like you can with perl's "use strict" option.  That will prevent a lot
more bugs than := is likely to cause.

> > Maybe there's a principle to be found here: "design by FAQ".  If newbies
> > keep arriving and asking over and over again "how do I do X?", maybe
> > there's a good reason for having a way to do X.  We've seen that already
> > for augmented assignments.
> 
> Most newbies tend to come from language X, think Python is great,
> except it isn't quite X - why block indentation when { } would work? Why
> 'self'? Why not assignments in expressions? Etc.

But even oldbies ask about assignments in expressions, augmented
assignment, etc.

> The Python idiom for while (m := re.match('whee',x)): blah() is
> 
> while 1:
>    m = re.match('whee',x)
>    if not m:
>       break
>    blah()
>    
> And to me, an experienced Python programmer who hasn't really used other
> languages in quite a while, that looks *much* clearer than the one liner
> assignment-in-the-while version. It's a common pattern, and the eye
> recognizes it immediately.

And it fills more space on the screen, crowding out other code, so you
can't see as much of the program flow.  I like the short version better.
(Well, I'd probably put the blah() on a separate line).  It's a matter
of style and preference that varies from one programmer to another,
not something the language should impose.  

> But in the end, neither of our subjective opinions matter - Guido's does.
> That's probably why this language is so great.

Guido doesn't write our code; we do.  And we all have our own styles
and preferences that are reflected in the code we write.  I like
Python a lot but I don't think it's perfect, and I don't think anyone
thinks it's perfect.  It changes (mostly for the better) from version
to version, partly based on the shortcomings people find while
programming in it.  If experienced Python programmers regularly lament
the lack of assignment expressions or anything else that's
straightforward to fix, that's probably grounds for considering a fix.


From spamneggs at mail.utexas.edu  Wed Feb 27 15:55:38 2002
From: spamneggs at mail.utexas.edu (Kareem Badr)
Date: Wed, 27 Feb 2002 14:55:38 -0600
Subject: creating directory heairarchy in a zipfile archive
References: <3C7D3D48.5B8BC97C@mail.utexas.edu>
Message-ID: <3C7D47CA.9CFCB38D@mail.utexas.edu>

It seems zipfile handles directories just fune. The zip reader I was
using was the problem. Thanks anyway. 

Kareem Badr wrote:
> 
> Does anyone know how to create file heirarchies within a zipfile object?
> I've tried prepending the name of the directory to the archive name of
> the file, but that doesn't work. Any tips would be greatly appreciated.
> Thanks.
> 
> -Kareem Badr


From jbauer at rubic.com  Tue Feb 12 15:54:32 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 12 Feb 2002 14:54:32 -0600
Subject: Python 2.2 <-> Solaris 2.6 installation errors
Message-ID: <3C698108.4B86C8E1@rubic.com>

I'm having problems installing Python 2.2 onto
my Solaris 2.6 workstation.

I am doing the boilerplate ...

  ./configure
  make
  make install

... but it bombs, possibly because distutils
is looking for something that isn't there?

Earlier versions of Python install fine on
this machine.

-Jeff


skipping 'rgbimg' extension (up-to-date)
skipping 'readline' extension (up-to-date)
WARNING: removing "readline" since importing it failed
Traceback (most recent call last):
  File "./setup.py", line 793, in ?
    main()
  File "./setup.py", line 787, in main
    scripts = ['Tools/scripts/pydoc']
  File "/export/home/jbauer/Python-2.2/Lib/distutils/core.py", 
    line 138, in setup
    dist.run_commands()
  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
    line 893, in run_commands
    self.run_command(cmd)
  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
    line 913, in run_command
    cmd_obj.run()
  File "/export/home/jbauer/Python-2.2/Lib/distutils/command/build.py", 
    line 107, in run
    self.run_command(cmd_name)
  File "/export/home/jbauer/Python-2.2/Lib/distutils/cmd.py", 
    line 330, in run_command
    self.distribution.run_command(command)
  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
    line 913, in run_command
    cmd_obj.run()
  File
"/export/home/jbauer/Python-2.2/Lib/distutils/command/build_ext.py", 
    line 256, in run
    self.build_extensions()
  File "./setup.py", line 147, in build_extensions
    build_ext.build_extensions(self)
  File
"/export/home/jbauer/Python-2.2/Lib/distutils/command/build_ext.py", 
    line 383, in build_extensions
    self.build_extension(ext)
  File "./setup.py", line 179, in build_extension
    for filename in self._built_objects:
  File "/export/home/jbauer/Python-2.2/Lib/distutils/cmd.py", 
    line 107, in __getattr__
    raise AttributeError, attr
AttributeError: _built_objects
make: *** [sharedmods] Error 1



From threeseas at earthlink.net  Sat Feb 16 10:22:08 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 15:22:08 GMT
Subject: Fear of autocoding!
References: <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <1476.811T2040T12515414threeseas@earthlink.net> <slrna6rk5j.2pn.quinn@cruzeiro.ugcs.caltech.edu>
Message-ID: <3422.812T515T6683481threeseas@earthlink.net>

On 15-Feb-02 22:23:33 Quinn Dunkan <quinn at cruzeiro.ugcs.caltech.edu> wrote:
>On Sat, 16 Feb 2002 01:45:49 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>>Which you could have done by not doing anything, in comparison to what you
>>have done.

>Do people laughing at you make it harder for you to write your thing?

>Awww... sorry.  Other people will entertain themselves that way, though, if
>you give them the opportunity.  Obviously you're entertained by it too, or
>else you wouldn't come back and post when you could be writing up your gizmo.

>So everyone wins, and we all keep each other occupied.  Usenet is such a
>friendly place.

>If it takes 14 years to write 1 of 9, then Armageddon for programmers is at
>2114.  By then we'll have already exploded when the unix 32bit epoch rolled
>over.

>Start your clocks:
>http://timecube.com/index15.html


All you have said is that there is no intent on making programming easier
for the masses.

You see, I and the majority of people in this world do not preform the
act of software programming 40 plus hours a week. In fact I understand
that even Programmers don't actually preform even half of that a week. But
the majority of us are busy doing other things, many of which are to do
things that you cannot live without (Food, clothing, shelter, medicine,
education, etc..), unlike programming.

As such, its clear that everyone is capable of putting things together,
from kids putting blocks together to rocket scientist putting things
together to figure out how to put other things together. It is also clear
that the majority of mental focus in this world is not software
programming but in using computer the users are subjected to thinking in
terms of what the programmers dictate.

To be constantly telling users they are to stupid to put things together
is not a good thing to do. The solution is of course to make programming
easy enough to do that the typical user actually has the resources they
need to actually do it. What that means is that programming has got to be
framed within the common action set that everyone is inherently familiar
with, experienced with.

As far a laughing at me goes, isn't that a MS tactic or item on their list
of what they do to those who do not agree with them?

What's the actual list of counter competition practices MS uses?

Right now they are trying to make out that they own the credit for
innovating linux via cause and effect. Saying that even Linux supporters
like .net in their marketing hype of .net with the appearance of being
news rather than marketing hype.

Meaning that from you and your comments you preceive me at the "laugh at
stage" but even you have to use all nine actions in reading and responding
to posts.

Guess that makes you a hypocrite laughing at yourself.

Python allowed me to very easily add internet access to files, in IQ.

Guess maybe that should be taken out now, huh?

That is if you really want to believe that 114 years is how long it's
gonna take. Based of course on programming not getting any easier.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From robin at jessikat.fsnet.co.uk  Wed Feb 27 12:17:14 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 27 Feb 2002 17:17:14 +0000
Subject: PIL vs eps
References: <mailman.1014824145.17587.python-list@python.org>
Message-ID: <JFkzWtAaSRf8Ewgj@jessikat.fsnet.co.uk>

In article <mailman.1014824145.17587.python-list at python.org>, Gustavo
Cordova <gcordova at hebmex.com> writes
>> 
>> I think this is a pure ghostscript problem, but PIL fails to 
>> convert eps to other formats and in fact it seems that gs
>> won't create any output without a showpage at least somewhere.
>> 
>
>Isn't "showpage" the postscript command to render
>all the code that's been processed?
>
>-gustavo
>
yes, but most eps files don't seem to have a show page and thus
ghostscript doesn't seem to render anything. The EPSPlugin seems to pass
the data to ghostscript, but a client reports errors with .eps. When I
do the gs command handomatically I get no file. If I add in a showpage
just before the final restore I at least get the converted image/ppm
etc.
-- 
Robin Becker


From quinn at schilling.ugcs.caltech.edu  Wed Feb 20 13:29:27 2002
From: quinn at schilling.ugcs.caltech.edu (Quinn Dunkan)
Date: 20 Feb 2002 18:29:27 GMT
Subject: giant stdlib tuples (was: Re: Help with code)
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com>
Message-ID: <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu>

On Wed, 20 Feb 2002 08:41:31 -0500, Steve Holden <sholden at holdenweb.com> wrote:
>Besides the underscores another poster mentioned, note also that you have
>replaced a period with a comma. Your line which reads
>
>            self,dst=time.localtime(self.t)
>
>should read
>
>            self.dst=time.localtime(self.t)

Speaking of time.localtime() and giant tuple matches, I always wind up writing
my own wrappers for times and pwents so access them as proper objects.

It's always seemed to me that things like time, pwd, grp, and stat are pretty
non-pythonic.  You can't even really say it's because of the thin wrapper thing
because writing and using a StructType or something wouldn't be much harder
than using tuples.

Of course, I've learned to carry around my own little wrappers, but I've always
wondered why they were that way in the first place.


From kragen at pobox.com  Fri Feb  1 16:10:07 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 01 Feb 2002 16:10:07 -0500
Subject: OT: Reason to love the snake #332
References: <mailman.1012370928.15598.python-list@python.org>
Message-ID: <83it9gzz68.fsf@panacea.canonical.org>

"Bruce Dykes" <bkd at graphnet.com> writes:
> Found on another board:
> > Actual code found in a production application:
> >
> > foreach (@list) {
> >         /=/;
> >         eval( qq( \$$` = "$'" ));
> > }

(Kids, this is not only an obscure way to do it, this is a stupid
thing to do.  Uncle Mark says it could cost you thousands of dollars.
See http://perl.plover.com/varvarname.html for more on how dumb it
is.)

It translates directly into Python:
for _ in alist:
	__ = _.find('=')
	exec '%s = "%s"' % (_[:__], _[__+1:])

> > Now, that's obscure and overlong. You could just do:
> >
> > s/([^=]*?)=(.*)/\$$1="$2"/e for (@list);

The ? is superfluous there.

You can't do that in Python, fortunately.  But you can do
map(lambda _:(lambda __,_=_:globals().update({_[:__]:_[__+1:]}))(_.find('=')),alist)

While longer and more verbose, this version is also more correct: it
will always update the globals instead of some random scope, it
correctly handles strings with embedded newlines and any kind of
quotes.

,_=_ is omittable in Pythons with nested scopes.

If we really want to bite it, we can do this:
import re
map(lambda _:(lambda (x,y),_=_:globals().update({_[:x]:_[y:]}))(re.search('=',_).span()),alist)

> > (Of course, no sane person would, but somebody managed to whip out the
> > first one without any apparent discomfort, either).

Such a person should not be writing production code.



From gerhard at bigfoot.de  Thu Feb 28 12:04:40 2002
From: gerhard at bigfoot.de (Gerhard =?ISO-8859-1?Q?H=E4ring?=)
Date: Thu, 28 Feb 2002 18:04:40 +0100
Subject: Can use python to play MP3 file?
References: <a5llkf$8a75f$1@ID-12869.news.dfncis.de>
Message-ID: <3C7E6328.6050500@bigfoot.de>

Ricky wrote:
> I hear that PHP can online play MP3 file. Is there any method that can make
> me play MP3 by using Python CGI?
> 
> Thanks for any reply!

Do you want to play the MP3 file on the server, like a jukebox. Or do 
you want to stream MP3 files back to the client computer?

(Of course you can do both with Python)

Gerhard



From sholden at holdenweb.com  Wed Feb 20 08:14:02 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 08:14:02 -0500
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <258fd9b8.0202191512.4f4693ff@posting.google.com>
Message-ID: <4iNc8.17011$Zp2.1752237@atlpnn01.usenetserver.com>

"Magnus Lyck?" <magnus at thinkware.se> wrote ...
> ajeru at vknn.org (Alexander Jerusalem) wrote ...
> > The argument that exhaustive testing should be done anyway is
> > certainly true. I nevertheless tend to think that it's better to let
> > the compiler do as much as possible in terms of error checking and
> > then use testing to find the true runtime errors. It just takes much
> > less time than starting a test run every time just to find all my
> > typos. I don't know if there are test suites that let you run all the
> > branches of all your methods and show you all typos at a time. I
> > suspect that I'd have to start the test suite for each and every typo
> > but I'm not knowledgable enough when it comes to Python test suites to
> > be sure of that claim.
>
I'm not sure I'd agree with that, given that Python is fairly good at only
compiling what needs to be recompiled. I can agree it will be something of a
problem for long batch jobs. Before you do such integration testing,
however, your unit tests, which are almost always lightweight, should have
pulled any syntax errors and name typos.

> I'm trying to adopt the XP habit of writing tests first, and then
> only code enough to make the test pass. It's sometimes difficult.
> It requires some diciplin not to get carried away while writing the
> code, and do more than I've written tests for. When I do get carried
> away I end up having to write a whole bunch of boring tests afterwards,
> and that feels like a drag, because then the writing of tests is not
> part of the creative process. And it's difficult to be sure that I
> actually have all the tests I need then.
>
As far as this goes, EITHER you aren't being sufficiently rigorous in
defining your tests, OR you are over-specifying unnecessarily during
implementation. This latter is a particularly troublesome waste of
programmer time, which XP attempts to eliminate with its well-known (?) "You
aren't gonna need it" credo:

    http://c2.com/cgi/wiki?YouArentGonnaNeedIt

In other words, when you find yourself thinking "Wow, this frobnicator could
also uglify with only ten extra lines of code", you can most often save
yourself the ten extra lines of effort until the customer demands
uglification :-0

[...]
regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From tismer at tismer.com  Tue Feb 26 10:39:51 2002
From: tismer at tismer.com (Christian Tismer)
Date: Tue, 26 Feb 2002 16:39:51 +0100
Subject: Time to rename Stackless?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org> <915a998f.0202260438.246be39c@posting.google.com>
Message-ID: <3C7BAC47.6090108@tismer.com>

Hamish Lawson wrote:

> Christian Tismer wrote:
> 
> 
>>Well, it doesn't switch between multiple stacks, it
>>repeatedly replaces parts of the same stack.
>>(But this would have the same effect, of course).
>>
> 
> If it essentiually has that effect, how about Switching Python then as a name?


Not bad!
Should I start a contest?

ciao - chis

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From nospam at bigfoot.com  Fri Feb 15 06:18:50 2002
From: nospam at bigfoot.com (Gillou)
Date: Fri, 15 Feb 2002 12:18:50 +0100
Subject: xmlrpclib RPC calls sometimes hang
References: <mailman.1013724025.19473.python-list@python.org>
Message-ID: <a4iqd6$dqs$1@norfair.nerim.net>

Hi,

timeoutsocket acts as a plug in replacement of the blocking parts of the
socket package.
This makes its use (almost) transparent when programming with network
modules (httplib, urllib, ftplib...). You only need to handle Timeout
exceptions in your app.
Theorically, you don't need to change anything in xmlrpclib if the transport
is based on urllib or httplib.
But I dunno for M2Crypto.
Just try it (and feedback to the author), installing and using timeoutsocket
is easy.

import timeoutsocket
import xmlrpclib

timeoutsocket.setDefaultSocketTimeout(20)
s = xmlrpclib.Server('http://www.somewhere.net/foo')
try:
  data = s.getSomething()
except timeoutsocket.Timeout, e:
  # 8((  No response. Keep zen

HTH

--Gilles

"Simrod Furman" <sfurman at nameconnector.com> a ?crit dans le message news:
mailman.1013724025.19473.python-list at python.org...
> I'm writing a xmlrpclib client in Python 2.1 using xmlrpclib 0.9.9 on
> Windows NT. I'm having a problem however, since the RPC call through
> xmlrpclib is a blocking call, if there are any problems during the
function
> call such as a dropped modem or a problem on the XML-RPC server, the RPC
> call wedges. As far as I can tell, xmlrpclib just hangs inside the RPC
call.
>
> I'm also using SSL from the M2Crypto library as a plug in the 'transport'
> argument to xmlrpc.Server()
>
> I've heard the suggestion to use timeout sockets from here:
>
> http://aspn.activestate.com/ASPN/Mail/Message/python-announce/935327
>
> as a way to add this functionality in, but I think it would require me to
> modify xmlrpclib.
>
> Question 1: Is there a way to get xmlrpclib to timeout and throw an
> exception after a specified amount of time?
>
> Question 2: Has anyone tried xmlrpclib, SSL, and timeout sockets in Python
> 2.2?
>
> --Simrod
>
> Simrod Furman
> Parlance Corporation, home of the NameConnector Service
> http://www.nameconnector.com
>




From philh at comuno.freeserve.co.uk  Sun Feb  3 20:42:34 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 01:42:34 +0000
Subject: Autocoding project proposal.
References: <83sn8rgmc7.fsf@panacea.canonical.org> <B87ACAC8.3B6F%jonathan@onegoodidea.com> <83it9mghng.fsf@panacea.canonical.org> <B87C2640.3CC2%jonathan@onegoodidea.com> <slrna5d2cq.g57.philh@comuno.freeserve.co.uk> <1341.799T18T11762852threeseas@earthlink.net>
Message-ID: <slrna5rpoa.19d.philh@comuno.freeserve.co.uk>

On Mon, 04 Feb 2002 00:01:03 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 29-Jan-02 06:38:02 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>On Tue, 29 Jan 2002 10:13:20 +0000, Jonathan Hogg <jonathan at onegoodidea.com>
>>wrote:
>>>On 28/1/2002 23:49, in article 83it9mghng.fsf at panacea.canonical.org, "Kragen
>>>Sitaker" <kragen at pobox.com> wrote:
>>>
>>>> These are problems, but they are not problems AppleScript solves.
>>>
>>>The only two sensible things I could get out of Timothy's manifesto was the
>>>idea of GUI programs exposing their internal functionality through an
>>>RPC-like mechanism, and that programming languages have a high entry cost
>>>[...]
>>>c.l.py isn't really the place for any of this
>
>>I think c.l.py is an approriate place tob discuss the entry costs of
>>programming and how programming languages can be made easier. Think of
>>why python is the way it is.
>
>Where else are you going to take a shot at trying to take down the
>ultimate cost lower tool. The one that removes the need for putting up
>with people like you.

I have no idea what you are trying to say.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From othello at javanet.com  Sat Feb  9 17:15:21 2002
From: othello at javanet.com (Raymond Hettinger)
Date: Sat, 9 Feb 2002 17:15:21 -0500
Subject: CSV revisited
References: <slrna62f14.pef.mlh@vier.idi.ntnu.no>
Message-ID: <a4474v$9t6$1@bob.news.rcn.net>

I think your idea warrants a PEP.

It is the place to collect:
- the pros and cons of the proposal
- what to include/exclude
- various approaches
- References.  For instance, Kernighan & Pike, The Practice of Programming
  devotes a full chapter to issues in building a production CSV library.
  Also, there should be references to code and documentation for CSV
libraries
  in other languages.
- Documentation of key design decisions.  For instance, is the library
limited
   to reading various CSV formats or is write-back in the same style to be
supported.
   Can the first record, contain optional field names?  Will incomplete
records
   raise an exception?
- Pointers to draft implementations so people can experiment with it.

A PEP is an ideal place to collect and document this information.
Also, an addition to the standard library is likely to be considered more
than a 'minor feature request'.  Another advantage is that it will also help
focus the discussion (be prepared, life in the focus is sometimes a little
hot).

If you want a co-author, I'm game.


Raymond Hettinger
"Magnus Lie Hetland" <mlh at vier.idi.ntnu.no> wrote in message
news:slrna62f14.pef.mlh at vier.idi.ntnu.no...
> A little while ago I started a thread about adding CSV support to the
> standard library. There was (as expected) some controversy about this,
> but I think the idea had enough support that it might be turned into a
> formal proposal... However, I'm not sure wheter this is the sort of
> thing that should be formalised as a PEP... Any opinions in the
> matter? (Not pro/con the proposal itself... ;)
>
> --
> Magnus Lie Hetland                                  The Anygui Project
> http://hetland.org                                  http://anygui.org




From stojek at part-gmbh.de  Wed Feb 13 13:15:19 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Wed, 13 Feb 2002 18:15:19 GMT
Subject: searching a list of tuples
Message-ID: <3c6aabeb.6302984@news.easynews.net>

Hi,

I have a (huge) list of tuples :
list=(float 1,float 2,float 3,int).

The list has been sorted with list.sort()
Now I have to slice the list in a way
that:

  a <= float 1 <= b

As I have to do this very often for different
values of a and b and the list is quite large
(100.000 tuples) I would like to do the slicing
as fast as possible.

Can anybody help me?

Tanks
Marcus


From fredrik at pythonware.com  Mon Feb 11 17:03:02 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 22:03:02 GMT
Subject: Is it possible to call a function from a Tkinter widget.bind and include args???
References: <a4995n$gbg$1@news5.svr.pol.co.uk>
Message-ID: <qcX98.17823$n4.3075841@newsc.telia.net>

G. wrote:

> Button1.bind("<Enter>", printButtonAndCoords("button1"))

this calls the printButtonAndCoords function, and passes the
result to the bind function.

(well, to be precise, it gives a TypeError exception)

try this instead:

    Button1.bind("<Enter>", lambda e: printButtonAndCoords(e, "button1"))
    Button2.bind("<Enter>", lambda e: printButtonAndCoords(e, "button2"))

(the lambda statement creates a callable object taking one
argument, which it passes on to the printButtonAndCoords
function)

you might prefer writing this as

    callback = lambda e: printButtonAndCoords(e, "button1")
    Button1.bind("<Enter>", callback)
    callback = lambda e: printButtonAndCoords(e, "button2")
    Button2.bind("<Enter>", callback)

or even:

    def callback(e):
        printButtonAndCoords(e, "button1")
    Button1.bind("<Enter>", callback)

    def callback(e):
        printButtonAndCoords(e, "button2")
    Button2.bind("<Enter>", callback)

hope this helps!

</F>

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




From jkraska at san.rr.com  Sun Feb  3 16:37:59 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 21:37:59 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org> <a3k9am$p0c$1@newsreader.mailgate.org>
Message-ID: <fmbr5uc0uikf74flscdlilq7bq8q1i3tmr@4ax.com>

>Who *are* the more intelligent of you?

I saw an interesting bit of sociological trivia the other day:

Some researchers took a subject population and divided it into
smaller groups. They than asked each group to rate their abilities
against all the other groups. Interestingly, each group always
rated themselves as more capable than the other.

It would therefor follow that _all_ of Tim's fellows are the
more intelligent of them, yes?

And at that, I'll leave you with a movie quote for the day:

"Have you ever heard of Plato? Aristotle? Socrates?"
"Yes."
"Morons."

C//



From Tom_Good1 at excite.com  Wed Feb 27 19:21:39 2002
From: Tom_Good1 at excite.com (Tom Good)
Date: 27 Feb 2002 16:21:39 -0800
Subject: dirwalk.py generator version of os.path.walk
References: <a5d36e$1daf$1@news.idiom.com>
Message-ID: <ac677656.0202271621.5a134d44@posting.google.com>

jimd at vega.starshine.org (Jim Dennis) wrote in message news:<a5d36e$1daf$1 at news.idiom.com>...
> This function could probably use a bit of polishing,
>  and it certainly could use some enhancement (some options to
>  control if, and how we follow symlinks, to how to handle 
>  exceptions on listdir(), whether to be depth first, and an 
>  option to avoid crossing mount boundaries with os.path.ismount(), 
>  etc).
> 
>  However, it seems to work.  
> 
>  dirwalk() simply takes an optional top level directory/path name
>  as an argument and instantiates a generator which will walk down
>  that tree and return every filename that it can access.  
> 
>  It's late and I need sleep.  So I'm just going to post this in
>  it's rough (and probably buggy) form and let y'all thrash on it
>  a bit.
> 
>  I guess there's some sort of statcache module that might let me
>  cache the stat() tuples.  I guess I'm implicitly incurring a stat()
>  system call for each node by checking islink() and isdir() on it
>  so it seems like I ought to cache that and make it available to 
>  my caller (without forcing them to make an additional stat system
>  call).
> 
>  I hope that something like this (a simple dirwalk() or other 
>  greatly simplified alternative to os.path.walk()) makes it into 
>  Python 2.3 or later.
> 
> #!/usr/bin/env python2.2
> from __future__ import generators 
> import os
> 
> def dirwalk(startdir=None):
> 	if not startdir:
> 		startdir="."
> 	if not os.path.isdir(startdir):
> 		raise ValueError ## Is this the right exception?
> 	stack = [startdir]
> 	while stack:
> 		cwd = stack.pop(0)
> 		try:
> 			current = os.listdir(cwd)
> 		except (OSError):
> 			continue	# Skip it if we don't have access
> 		for each in current:
> 			each = os.path.join(cwd,each)
> 			if os.path.islink(each): 
> 				pass
> 			elif os.path.isdir(each):
> 				stack.append(each)
> 			yield(each)
> 
> if __name__ == "__main__":
> 	# import unittest?
> 	# test suite should consist of:
> 	# 	dirwalk() vs. os.listdir()
> 	# 	dirwalk("/") vs. os.path.walk()
> 	# 	dirwalk("/etc/passwd") (should raise exception)
> 	import sys
> 	for i in sys.argv[1:]:
> 		for j in dirwalk(i):
> 			print j
> 	# should compare this to os.popen("find ....") and
> 	# or to os.path.walk(...)

Hi,

I wrote a different implementation of this general concept at:

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/105873

You don't really need to keep a stack of directories and push/pop
things, because with generators you can recurse instead.


Tom


From threeseas at earthlink.net  Sun Feb 10 13:03:39 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 10 Feb 2002 18:03:39 GMT
Subject: 3rd User Interface >Re: Please stop responding to Rue
References: <a3ucsk$12h8$1@nntp6.u.washington.edu>	<a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com> <5659.806T1220T5243918threeseas@earthlink.net>
Message-ID: <1190.806T1901T8305155threeseas@earthlink.net>

On 10-Feb-02 08:00:30 Timothy Rue <threeseas at earthlink.net> wrote:

[snip]

>Today the Mac might be the best supplier of a system with all three
>user interfaces.

>But then there is Linux and the Hurd. Also with nothing much more than a
>Command Line interface and GUIs, but the potential of adding in the third
>user interface on the GNU applications is tremendious. With a skelton code
>frame, such conversion can happen very quickly across the platform.

In addition to this, or as an elaboration on GNU base, there is
established three types of communications with such a Automation Port
Interface. A signal, a message and an memory range address.

>So, there really is not yet an existing system having all three primary
>User Interfaces openly available. Yet every one know there is value to be
>had in such.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From aries at bitstream.net  Sat Feb 23 04:33:15 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Sat, 23 Feb 2002 03:33:15 -0600
Subject: help: Problem with cgi form
References: <mailman.1014420097.13372.python-list@python.org> <Xns91BEB7AC37F78AusYourStandingInIt@130.133.1.4>
Message-ID: <3c77603c$0$79562$6536d21b@news.bitstream.net>

Flavian Hardcastle <deathtospam43423 at altavista.com> wrote in message
news:Xns91BEB7AC37F78AusYourStandingInIt at 130.133.1.4...
> Gustavo Cordova <gcordova at hebmex.com> wrote in
> news:mailman.1014420097.13372.python-list at python.org:
>
> >>
> >> I'm a beginner. How do I do this? It's probly very simple ......
> >>
> >> I have a hidden input on a cgi form...
> >>
> >>     print '<INPUT TYPE=HIDDEN NAME= "last_pos"
> >> VALUE=',last_byte_pos,'>'
> >>
> >> I need the VALUE to be the  variable "last_byte_pos", but
> >> unfortunately when I run the script for the first time the script
> >> crashes because it's
> >> undeclared.
> >>
> >> What subroutine do I use to check if last_byte_pos is undeclared, and
> >> assign it a default value if it is?
> >>
> >
> > How can you not know if "last_byte_pos" is undeclared?
> > Have you assigned any values to it previously?
> > no?
> >
> > then it's undeclared.
> >
> > -gus
> I know when it's undeclared ... whenever a user first tries to access the
> form, it is undeclared. What I want to know is how to get the script to
> check whether this value has not been declared yet.
>
>
> --
> "You can tell whether a man is clever by his answers. You can tell whether
> a man is wise by his questions."  Naguib Mahfouz
>
> netvegetable at dingoblue.net.au

if value not in last_byte_pos:
    last_byte_pos = value
    print "this is no not know if value is undeclared funny business??"
elif:
    last_byte_pos = value = last_byte_pos = last_byte_pos = value
    print "this is no know if value is declared funny business"
else:
    print "maybe its a design flaw"





From jonathan at onegoodidea.com  Sun Feb  3 05:17:24 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Sun, 03 Feb 2002 10:17:24 +0000
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu> <B8809EB2.411E%jonathan@onegoodidea.com> <a3fisr$9j6$0@216.39.172.122> <B881642D.41AD%jonathan@onegoodidea.com> <a3ht5h$2o0$0@216.39.172.122>
Message-ID: <B882BEB4.425F%jonathan@onegoodidea.com>

On 2/2/2002 23:36, in article a3ht5h$2o0$0 at 216.39.172.122, "Bengt Richter"
<bokr at oz.net> wrote:

> IME, you can use ScriptAlias to set suexec cgi directories on a per-virtual
> host basis. I have one test VH running cgi out of '/', another with cgi
> directory named other than 'cgi-bin', and others from the usual '/cgi-bin',
> all running as account owner. See
> 
>   http://httpd.apache.org/docs/mod/mod_alias.html#scriptalias
> 
> Perhaps one path per installation applies to another version or mode of use?

The suexec binary itself would refuse to execute anything not under a
particular root directory which you supply at configure time
(--suexec-docroot=DIR). This directory is meant to be your server document
root. However if you use virtual hosting and have multiple unrelated
document roots, you're in trouble.

I've never found it to be a particularly useful security check as Apache's
own cgi restrictions are more flexible. Generally I just set it to '/' and
forget about it. Though depending on the installation I have been able to
use a more restrictive path in the past, like '/customers' or something. It
also follows symlinks so in an NFS automounted environment where you're
automounting a local directory you get stuffed again.

(This is because '/customer/bob' might be '/export/customer/bob' on the
local machine.)

Jonathan



From stephennaicken at hotmail.com  Sat Feb 16 14:25:19 2002
From: stephennaicken at hotmail.com (Stephen Naicken)
Date: 16 Feb 2002 11:25:19 -0800
Subject: Python Gnutella Library
Message-ID: <999cc599.0202161125.64272cb7@posting.google.com>

I am currently working on dissertation project that makes use of
Peer-to-Peer networks.  I have been trying to find a Gnutella Library
for Python but without success.  There are many Gnutella libraries for
Java (JTella) and other languages, but so far I have yet to find one
for Python.

I would be grateful for any help you could provide me.

Thanks

Stephen.


From sameer_ at email.com  Tue Feb  5 15:00:50 2002
From: sameer_ at email.com (sameer)
Date: 5 Feb 2002 12:00:50 -0800
Subject: distutils on win32
Message-ID: <6bd9f01b.0202051200.74b0267b@posting.google.com>

i'm using distutils to make a windows installer for a python/wxpython
app.  what's the fastest way for me to create shortcuts on the desktop
as well as the Start menu for the file I want the users to be able to
run.  I don't want them to have to go into the correct package, and
run the app from there.


From loewis at informatik.hu-berlin.de  Mon Feb 25 08:52:57 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 14:52:57 +0100
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de> <s%qe8.598$Zj1.7273540@newsserver.ip.pt>
Message-ID: <j4664lllc6.fsf@informatik.hu-berlin.de>

"Hugo Martires" <hmartires at cultalg.pt> writes:

> And how can i count in a list ?
> It is possible ?

Read the source code. Alternatively, read the newsgroup archives,
searching for "size" and "object", e.g.

http://groups.google.com/groups?hl=en&selm=j43d3m93iu.fsf%40informatik.hu-berlin.de

This article also discusses the size of lists.

Regards,
Martin


From sabren at manifestation.com  Tue Feb 19 18:36:52 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 19 Feb 2002 18:36:52 -0500 (EST)
Subject: MIDI
In-Reply-To: <03b101c1b98f$e5f7d1d0$6e42a8c0@corelabs.coresdi.com>
Message-ID: <Pine.LNX.4.44.0202191835230.31461-100000@mercury.sabren.com>

On Tue, 19 Feb 2002, Mariano Corsunsky wrote:

> Does someone knows where can I find a module to play midi
> files from a python script?

If you're on win32, I have some old and unmaintained code
(a python module written in C) at:

  http://sabren.net/code/python/midi/

The precompiled pyd file is for 1.52, I believe...

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From Scirious at Scirious.com  Fri Feb 15 14:49:32 2002
From: Scirious at Scirious.com (Scirious)
Date: Fri, 15 Feb 2002 19:49:32 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <Xns91B6B52B3DA3EScirious@140.99.99.131>

Ok! Once again thaks for your rsponse.
Scirious.


From Peter.Milliken at GTECH.COM  Mon Feb 18 20:51:33 2002
From: Peter.Milliken at GTECH.COM (Milliken, Peter)
Date: Mon, 18 Feb 2002 20:51:33 -0500
Subject: Adding static typing to Python
Message-ID: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com>


-----Original Message-----
From: brueckd at tbye.com [mailto:brueckd at tbye.com]
Sent: Tuesday, February 19, 2002 11:54 AM
To: Peter Milliken
Cc: python-list at python.org
Subject: Re: Adding static typing to Python


On Tue, 19 Feb 2002, Peter Milliken wrote:

> The loose typing of Python is perceived as one of its big features :-).
> There are regular discussions on the usefulness of Python for "larger"
> projects. Many claim that they use it without any problems

I'm one of those people who like to make those sorts of claims. :)

> - I suspect that
> it is from a background of ignorance and lack of metrics

Rather than from ignorance, maybe it's from actual experience using it.
Have you considered the possibility that the claims might be true? In this
job as well as the past few our Python code has always had less time
devoted to developing, testing, maintaining, and refactoring it. It's not
that there's hidden costs involved - we tend to write it and move on to
something else because it just works.

Depends, "ignorance" was used in the context of "have you tried an
alternative?". You don't provide any examples of where you have found Python
superior over any other language development environment. So it becomes
impossible to verify the veracity of your statement :-) I provided an
alternative language in my statements to back up my claims :-). Since we are
talking loose typing versus strong typing, I would expect you to respond
with experience in a strongly typed language. The original poster used Java
as his comparision and is wishing for the feature in Python. Necessarily he
feels it is a deficiency in the Python language that I concur with :-)

There is no dispute from me that Python is the best environment for certain
situations - the library support alone is worth a very great deal! :-) A lot
also depends on your definition of "large", my definition of this is several
million lines of code - co-ordinating enough programmers to produce this
level of effort in a reasonable time can be quite a challenge :-).
"Contractual" type interfaces between modules helps a lot - something else
that Python lacks (IMO) that is useful for "large" projects. Python also
lacks features of "data hiding", it attempts to provide some support for it
but only in a desultory "procedural" way - try this with 40 programmers and
you won't get far! :-) This level of discussion has occured in the past and
people have disputed every one of these points - again, relevant experience
is important, the argument generally tends to the dismissive without
offering any supportive experience i.e. "I used data hiding features of
language X for Y+ years on projects that were over Z programmers large and I
never missed it once I moved to Python". In which case such opinions become
almost meaningless in the argument.

The meer fact that you are referring to "this job"  and the "past few" means
that (to me :-)) you haven't worked on a "large" project (by my definition).
This many of my "large" projects cover a span of 10 or more years!

Strong typing forces design up front i.e. it forces you to think through the
problem rather than just start coding - so it would be an anathema for
Extreme Programming advocates! :-) (please, no argument here, I have
probably got the idea of Extreme Programming wrong - I really don't care,
from what I have seen described the style doesn't suite me :-)). Because you
are "forced" to design up front and the compiler won't let you do sloppy
coding then you tend to write it and get it right the first time - then you
move on :-)

> The only way in Python to find these is through (exhaustive) testing -
> such rigorous testing *requires* you to develop an exhaustive test
> suite that is capable of running periodically i.e. nightly etc to
> catch such errors.

The current state of software development is that large programs need
exhaustive and rigorous testing anyway. The reason that the loose typing
doesn't kill you is that good testing finds misnamed variable bugs like
you cited above as well as bazillions of other bugs.

Finding such bugs (misnamed variables) requires multiple runs through your
test suite i.e. find a bug, fix it and re-run the tests. You *can't* find
all such bugs in a single pass! You also have to find it through (often)
painstaking use of the debugger or print statements. With a strongly typed
language, you can (often :-) I know there are exceptions :-)) do it in a
single pass with a compiler - definitely the debugger and print statements
are *never* involved! Running the same test over and over again to get these
levels of bugs out is clearly more inefficent than using a compiler. So
where your process may take possibly hours, I'm talking about minutes to fix
and recompile :-). 

> Whether the people who support Python for larger projects have the
> discipline within their project team to maintain such an effort is
> unclear - there has been no mention of it at all.

Really? Good testing comes up every single time this topic comes up, and
invariably people talk about how they *do* write tests for their code, and
how good testing helps them find all sorts of bugs.

You also never see anyone talking about testing the test code :-) -
something that is constant with any language :-) But the same scale of
efficiencies apply to the test code as the production code.

> So I doubt that the proponents of Python in these instances have any
> idea of the true cost of this language "feature" - they appear to be
> wound up in how quickly they can produce a working prototype (or
> whatever) so I suspect their day to day work involves very small,
> manageable projects where the lost efficiency due to loose typing
> isn't really that noticable for them i.e. they "live with it" :-).

Might be true for lots of people, but nearly every single Python module in
our company's code has either built-in or accompanying tests that run
every night. The same is true for our non-Python code. The dynamic or
static typing of the language makes little real difference - if you don't
test, you overlook bugs, pure and simple.

True, you obviously have a large investment in your test suite. My point is
that up front development time is reduced through the features of strong
typing. Doesn't matter what language is used, test suites are still required
to ensure that accidental changes aren't introduced.

> When I compile a Ada program cleanly I know that 95% of the bugs are out
of
> it - before I even start testing! :-) Of course, it takes me longer to
write
> up front but overall development effort (in my subjective opinion - since
I
> have no metrics :-)) is that it is much lower.

Whatever floats your boat, I guess. Very few of my bugs (in any language)
are things caught by a compiler and if anything, the warm fuzzies because
the code compiles is often a false sense of security. The 95% seems very
high, even considering Ada's strictness - to that I'd say that with Python
95% of the bugs never make it into the code in the first place. :)

Then you haven't tried a strongly typed language :-) I have tried both,
which allows me to speak from some knowledge :-) I haven't tried any other
strongly typed language than Ada, so my statements are necessarily limited
by that background. The original post states that the typing of Java was
perceived as a distinct advantage, so presumably the level of type checking
there is enough to request the feature in Python. 

Have fun,
Dave

You too mate :-)
Peter



From magnus at thinkware.se  Thu Feb 21 09:32:00 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 15:32:00 +0100
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org>
Message-ID: <3C7504E0.2050602@thinkware.se>

William Park wrote:

> On Wed, Feb 20, 2002 at 03:51:24PM -0800, Kevin Dahlhausen wrote:
> 
>>Does anyone know of an inexpensive web host that allows compilation
>>and use of Python extensions from  your own directory?
> 
> With the money you pay to ISP, why don't you get DSL with static IP, and
> run your own server?


I don't know about your part of the world, but I have an
ISP account with my own domain, telnet etc, not too bad, that
I pay about $12 per month for.

ADSL here is about $30 in the cheapest version. That will not
include a static IP, maybe you won't even be allowed to use
a server. Don't really know.

I'm lucky enough to just pay $20 for a 10Mb connection where I
live, but I can't get any static IP address at all, and I'm
not allowed to have a "server" in my end (which is a bit silly
since they allow me to upload stuff all the time with 10Mb, which
all that the ethernet delivers. In the discussions we had, they
said they were planning to offer "server licences", and they were
to cost $100 per month.

I talked to someone a while ago who had SDSL with 10 ip adresses,
and he paid $200 per month.

My ISP, www.fsdata.se has Python 2.1.2 installed, but unfortunately
they don't have mod_python or mod_snake, and they don't allow long
running processes. I'm not sure that they are interested in customers
outside Sweden though...



From feldspar at ix.netcom.com  Mon Feb  4 14:18:48 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Mon, 04 Feb 2002 15:18:48 -0400
Subject: heres a dumb question....
References: <a3mp6v$u04$1@wiscnews.wiscnet.net> <3C5EEA7C.F6380391@noaa.gov>
Message-ID: <3C5EDE98.33C039A8@ix.netcom.com>

"jim.vickroy" wrote:
> 
> How about the raw_input() function?
> 
> > x = raw_input('Enter an integer: ') # Enter 123
> > x
> 123
> 
> The prompt string (e.g., 'Enter an integer: ') is optional.

A slight correction/clarification here.  What you'll get back is not the
integer 123, but the string '123'.  The int() function will perform the
conversion for you if it's the integer you want.

	-jc


From antskip01 at optushome.com.au  Fri Feb  8 01:02:04 2002
From: antskip01 at optushome.com.au (antskip)
Date: Fri, 08 Feb 2002 17:02:04 +1100
Subject: dependency hell
References: <xIw88.500$9%6.51706@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <pan.2002.02.08.17.01.51.564647.1752@optushome.com.au>

Python 2.2.2 doesn't appear to exist yet!
How about saying exactly which python rpm you are trying to install?
On Fri, 08 Feb 2002 02:05:01 +1100, Tom F. wrote:

> I'm trying to install python2-2.2-2. I apparently need the following
> files:
> 
> libcrypto.so.2, libreadline.so.4, libssl.so.2
> 
> As far as I can search, libopenssIO and readline are supposed to supply
> them.
>  I can even find libreadline.so.4  specificly on my system (but not the
> others).
> 
> I've done many RPM searches and more, to no avail. could someone perhaps
> point me inthe right direction?
>


From newgene at bigfoot.com  Mon Feb 25 14:12:21 2002
From: newgene at bigfoot.com (newgene)
Date: 25 Feb 2002 11:12:21 -0800
Subject: Question on Python path
References: <3qnk7ug977hk4peorufobtlo38nh8j9v87@4ax.com>
Message-ID: <a50631f6.0202251112.6f6b81a6@posting.google.com>

Put your directory into "PYTHONPATH" environment variable.
or
import sys
sys.path.append("your directory")


CL

Gon?lo Rodrigues <op73418 at mail.telepac.pt> wrote in message news:<3qnk7ug977hk4peorufobtlo38nh8j9v87 at 4ax.com>...
> I have put all my own modules in a subdirectory of site-packages. How
> can I instruct Python (I'm using Python 2.2 on win2k) to also look up
> this directory when a
> 
> import <whatever>
> 
> is found?
> 
> My Thanks,
> Gon?alo Rodrigues
> 
> P.S: I have edited the registry directly by adding a User key in the
> Pythonpath folder. Needless to say I do NOT want to mess directly with
> the registry.


From jeremy at alum.mit.edu  Wed Feb 13 16:23:40 2002
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: 13 Feb 2002 13:23:40 -0800
Subject: SSL problem with Python (NEWBIE ALERT)
References: <kbua8.895$t83.126513@paloalto-snr1.gtei.net>
Message-ID: <b0f083db.0202131323.59b7422f@posting.google.com>

"AJ" <a-no-space-here-weberr at verizon-dot.net> wrote in message news:<kbua8.895$t83.126513 at paloalto-snr1.gtei.net>...
> Can anyone tell me why this is suddenly occuring?  It seemed to work in
> previous runs.  It's the ipcheck.py script on:
> http://ipcheck.sourceforge.net/.
> 
> I'm told it's probably a corrupt Python library.  Does anyone know how to
> troubleshoot this further, and/or how/which library I should re-install (and
> where to get that lib)?  If I run the script with the flag to use non-ssl
> connections it works fine...but who wants to be sending username/passwords
> over the net in the clear???
> 
> Thanks in advance for any assistance!
> --AJ
> 
> Traceback (most recent call last):
>   File "ipcheck.py", line 4083, in ?
>     _main(sys.argv)
>   File "ipcheck.py", line 3663, in _main
>     h2.putrequest("GET", updateprefix + hostlist + updatesuffix)
>   File "/usr/local/lib/python2.1/httplib.py", line 437, in putrequest
>     self.send(str)
>   File "/usr/local/lib/python2.1/httplib.py", line 379, in send
>     self.connect()
>   File "/usr/local/lib/python2.1/httplib.py", line 639, in connect
>     ssl = socket.ssl(realsock, self.key_file, self.cert_file)
> socket.sslerror: SSL_connect error

>From the traceback, it looks like you are using Python 2.1.  The SSL
support in Python 2.2 is improved in a number of ways, including bugs
fixed and much better error messages.  If you can upgrade to Python
2.2 and try again.  At the very least, you should get the SSL error
message that explains exactly what wen t wrong.

(And if you're already using Python 2.2, I guess the improved error
messages aren't as improved as I thought :-).

Jeremy


From marklists at mceahern.com  Tue Feb  5 16:52:14 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 5 Feb 2002 13:52:14 -0800
Subject: What is the command to determine an object's type?
In-Reply-To: <a3pj3e$1900rd$1@ID-98166.news.dfncis.de>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com>

MDK wrote:
> The code that you posted made perfect sense; thanks for that.
> However, I am
> not clear about what you are talking about in regards to "...the Strategy
> design pattern."
>
> I would like to investigate it but I do not know what it is I am
> investigating.  Can you clarify a little more for me?

There's a book called Design Patterns written by four folks commonly
referred to as the Gang of Four:

	http://www.amazon.com/exec/obidos/ASIN/0201633612/

Strategy is one of the Patterns.  I believe Bruce Eckel focuses on Patterns
in his Thinking in Python (a work in progress):

	http://www.mindview.net/Books/TIPython

Here's a google search that might give you some ideas:


http://www.google.com/search?hl=en&q=strategy+%22gang+of+four%22+design+patt
ern+python

Patterns seem to be a perennial topic of interest.  I'm still learning them
myself.

Cheers,

// mark




From lac at strakt.com  Wed Feb 13 14:25:59 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 13 Feb 2002 20:25:59 +0100
Subject: redirect console output to widget? 
In-Reply-To: Message from llchen223@hotmail.com (Lei Chen) 
   of "13 Feb 2002 11:09:19 PST." <de6c123c.0202131109.2708c2f4@posting.google.com> 
References: <de6c123c.0202101612.72b27f8d@posting.google.com> <mailman.1013395757.3817.python-list@python.org>  <de6c123c.0202131109.2708c2f4@posting.google.com> 
Message-ID: <200202131925.g1DJPxGo000561@ratthing-b246.strakt.com>

> I think you are right.  Is there an alternative to Tkinter that is thread safe?
> 

I forget, if I ever knew, what your original post said you 
wanted to do, but possibly this:

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/82965

is of use. 

Laura Creighton



From dsh8290 at rit.edu  Thu Feb  7 15:50:51 2002
From: dsh8290 at rit.edu (dman)
Date: Thu, 7 Feb 2002 15:50:51 -0500
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
In-Reply-To: <Pine.LNX.4.33.0202070643020.4847-100000@ns0.tbye.com>
References: <7xr8nxy0rb.fsf@ruckus.brouhaha.com> <Pine.LNX.4.33.0202070643020.4847-100000@ns0.tbye.com>
Message-ID: <20020207205051.GA29888@dman.ddts.net>

On Thu, Feb 07, 2002 at 07:01:28AM -0800, brueckd at tbye.com wrote:
| On 7 Feb 2002, Paul Rubin wrote:
| 
| > As much as I like Python, I don't see it working so well in the kind
| > of large, real-world projects that Java works pretty well in:
| 
| Why? You listed some cases where you don't think it would work well, but
| didn't give any reasons why you think that. Give us some insight into your
| line of thought. :-)

Ditto.

[examples snipped]
|
| Agreed, but to me the above 3 are examples of when Python would shine:

Ditto.

The compactness and conciseness of python, wrt C/C++/Java, makes it
really easy to grasp a new algorithm because there are fewer
instructions to get a handle on and less boilerplate.  In addition to
conciseness, python has a clarity that some other languages lack.

-D 

-- 

The Lord detests all the proud of heart.
Be sure of this: They will not go unpunished.
        Proverbs 16:7




From maxm at mxm.dk  Tue Feb 26 12:31:17 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 26 Feb 2002 17:31:17 +0000
Subject: ANN: decompyle 2.2 beta 1
References: <mailman.1014741412.30867.clpa-moderators@python.org>
Message-ID: <3C7BC665.9070104@mxm.dk>

Hartmut Goebel wrote:

> 'decompyle' converts Python byte-code back into equivalent Python
> source. It accepts byte-code from any Python version starting with 1.5
> up to 2.2.

Hmmm ... any chance of this reading an .exe from Gordons installer ??

I didn't think so, but that would have been nice.

And would have saved me from writing the same piece of software twice :-(

regards Max M



From jeff at ccvcorp.com  Wed Feb 27 13:24:44 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 10:24:44 -0800
Subject: Parse variable into variables >Re: Apology
References: <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <2240.820T2407T12845765threeseas@earthlink.net> <r29j7uo1ckvhesgc7po0thcevmdg0jeukg@4ax.com> <3C7AF404.6BC628F0@cosc.canterbury.ac.nz> <1121.822T529T14383507threeseas@earthlink.net> <3c7c7503_2@corp.newsgroups.com>
Message-ID: <3C7D246C.77C87C64@ccvcorp.com>


Joshua Muskovitz wrote:

> > testfile = open('test','r')
> > varset = testfile.readline()
> > tple = varset.split()
> > vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = tple
> > I assume this will work???
>
> Almost.
>
> vic = {}
> testfile = open('test','r')
> varset  = testfile.readline()
> tple = varset.split()
> vic[('AI','1')], vic[('AI','2')], vic[('AI','3')] = tple
> testfile.close() # don't forget to clean up after yourself

Actually, your sematics are slightly different.  In

vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = tple

vic['AI'] is a dictionary, and we're creating keys '1', '2', and '3'.  One
could presumably also have a vic['IQ'] dictionary, etc -- in other words, vic
itself is a dictionary of dictionaries.  In your example, you're creating a
single dictionary, which is keyed on tuples of ('AI', ...), and presumably at
some point ('IQ', ...), etc.  This may not be the best way to deal with it.
Since each chunk (AI, IQ, etc..) is conceptually different, having them as
separate dictionaries seems reasonable -- using them as the first element of a
key-tuple seems to me to be a bit obscure.  And Tim's usage *will* work,
provided that the dictionaries are set up properly beforehand:

vic = { 'AI': {}, 'IQ': {}, .... }

Actually, it might (or might not) make more sense to use a dictionary of lists,
i.e.

vic = { 'AI': [], 'IQ': [], ...}
...
vic['AI'].extend( list(tple) )  #convert the tuple to list before extend can
work

... wait, actually, since split() returns a list, the list call isn't needed
after all.  (Despite being called 'tuple unpacking', it also works on lists.)
So this can be simplified to

vic['AI'].extend( tple )

Doing this additionally allows a variable number of elements in tple -- you no
longer throw an exception if it's not exactly the number you expect.  (Whether
this is good or bad depends on your intent -- if you *want* to enforce
three-element lines, you can check len(tple) and raise an exception yourself.)


> and yes, you can shorten the entire thing down to:
>
> vic[('AI','1')], vic[('AI','2')], vic[('AI','3')] =
> open('test','r').readline().split()

Of course, by doing this, you're only getting the first line of the file.  If
one presumes that the rest of the file will be processed, too, then this isn't
a very good way of doing things.  ;)  I think that chaining in the open() call
is a bad idea, therefore; but the rest of the chaining works just fine.

infile = open('test.txt', 'r')
...
vic['AI'].extend( infile.readline().split() )
...
infile.close()


Jeff Shannon
Technician/Programmer
Credit International




From usenet at thinkspot.net  Fri Feb 15 12:35:35 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 15 Feb 2002 09:35:35 -0800
Subject: Mutex a file?
References: <a4h98k$ao2f$1@ID-98166.news.dfncis.de> <a4ghg4.3vv79at.1@kserver.org> <a4j8o4$mpqo$1@ID-98166.news.dfncis.de>
Message-ID: <a4ikp8.3vshuu5.1@kserver.org>

On Fri, 15 Feb 2002 10:16:19 -0500, "MDK" <mdk at mdk.com> wrote in
comp.lang.python in article <a4j8o4$mpqo$1 at ID-98166.news.dfncis.de>:

> The code that you posted is exactly what I need: simple file locking on a
> Windows machine.

Well, good. Although, if it is WinNT, Win2K or (probably ?) XP, there is
a better way to do it that affords more control and doesn't require the
looping to keep trying to acquire a lock on the file.

Be careful that locks should only be held for a very short time.

Good Luck,

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From marklists at mceahern.com  Thu Feb  7 16:56:57 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 13:56:57 -0800
Subject: FixedPoint
In-Reply-To: <20020207211613.GA1622@lilith.hqd-internal>
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEIDKBAA.marklists@mceahern.com>

Gerhard H?ring wrote:
> There are two changes:
>
> 1) it really tests for "is None", which should be ok for all cases I can
> currently imagine.
>
> 2) It also tests for a _quote() method first, so users can more easily
> override the default quoting behaviour.

I like this!  I'm embarassed to say I didn't study the code enough to see
that I could have added a _quote() to FixedPoint rather than changing it's
__repr__.  That's neat!

> Hmm., while we're at it, we should probably change this test too: "in
> [ListType, TupleType]" Does anybody know how I can reliably test if
> something is a Sequence or is iterable?

I'm just thinking out loud here...  How about something like this:

	if hasattr(value, '__getitem__'):
		...

This may only work in 2.2:

	if hasattr(value, '__iter__'):
		...

I don't really have a good grasp of all the subtleties involved in testing
for sequence in a way that works across lots of different versions of
Python.

// mark




From philh at comuno.freeserve.co.uk  Tue Feb  5 15:26:03 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Tue, 5 Feb 2002 20:26:03 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <slrna60fur.6sf.philh@comuno.freeserve.co.uk>

On Tue, 05 Feb 2002 05:36:47 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>I've come up with this concept that I'm calling "lingos": the name
>was chosen to evoke the idea of defining one's own lingo as a subset
>of the language. I'm still very much in the concept definition stage.
>
>I'm certain of little, but I'm certain that I want lingos to have
>at least the following properties.
>
>1. Lingos must be understood natively by the parser. No preprocessor
>front-ends or any other hack.

Why do you want this? Isn't it just an implementation detail?
I can see advantages in having it implemented as a preprocessor --
you could look at the preprocessor output and see what Python is 
generated. This might help with debugging complex macros.

>2. Lingos must be "hygienic". For those unfamiliar with the term,
>this means that names and symbols in the macro definition must not
>permute or modify the symbols in the namespace of the caller except
>where such a possibility is explicitly expressed: i.e., only the
>arguments to the macro itself can be permuted.

What if a macro needs to create local variables in its expansion?
Perhpas they could be called something like _lingos_local_xxx ?

>3. Lingos should be able to modify arguments as if the code in the
>lingo were used in place of the lingo itself. 

This would probably make more sense to me if you gave an example.

>I suppose that this
>is a fancy way of saying that one additional level of referentiality
>is supported. 

What's that meant to mean?

>4. Lingos should be able do deal with _code itself_ as a parameter.

Again, please give an example.

>This affords the opportunity to have lingos masquerade as proper
>Python block forms.
>
>5. Lingos might be able to support serial lingo forms, where one
>lingo knows how to aggregate multiple lingos into one lingo
>invocation. What I mean by this last is easier to describe by
>example.

Certainly

>Discarding the issue of special symbols for the moment, here are
>some possible lingos by example:
>
>1. Here "invariant" is a lingo macro which has 0 expression arguments
>   and 1 code argument:
>
>	invariant:
>		dosomething()

What is that meant to do?

>2. Here "unless" is a lingo macro taking 1 expression argument and 1
>   code argument:
>
>	unless i < 3:
>		dosomething()
>
>3. Here "factor" is a lingo macro taking 1 expression argument and
>   no code argument:
>
>	result = factor 17

Why not factor(17) ?

>4. Here "do" is a lingo macro taking 0 expression arguments, a code
>   argument, and acting as an aggregator for exactly 1 "while" lingo
>   form:
>
>	do:
>		dosomething()
>	while i <10


IMO it is more Pythonic to say:

   do:
      dosomething()
      while i <10                                                                                        

And it is even more Pythonic to say:

   while 1:
      dosomething()
      if i >= 10: break


>So. That, at last, brings me to the point of this message. Ideas?
>Feedback? This has been an interesting experiment so far, I'm just
>wondering how many folks find this sort of thing interesting and if
>anyone has any input.

I think Python could benefit from a macro system. But the devil's
in the details.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jeff at ccvcorp.com  Mon Feb 18 20:14:25 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 18 Feb 2002 17:14:25 -0800
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
References: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com> <mailman.1014024089.4625.python-list@python.org>
Message-ID: <3C71A6F1.24107297@ccvcorp.com>

Tim Legant wrote:

> I haven't read Jabberwocky in about 20 years, I think, and I don't
> have a copy here (a sad deficiency in my library!), but don't you want
> three blanks in the last line?  From (ancient) memory, isn't it:
>
> 'Twas _brillig_, and the _slithy_ _toves_
> Did _gyre_ and _gimble_ in the _wabe_.
> All _mimsy_ were the _borogoves_ (borogroves?)
> And the _mome_ _raths_ _outgabe_.

Yes, it is.


> Isn't-it-weird-having-a-mind-like-a-landfill-ly y'rs,

Yes, and it's also weird seeing someone else use the description that I've
been using of my own mind for years....   :)

Jeff Shannon
Technician/Programmer
Credit International




From jwbaxter at spamcop.net  Thu Feb 21 03:55:42 2002
From: jwbaxter at spamcop.net (John Baxter)
Date: Thu, 21 Feb 2002 00:55:42 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3c7455a8_4@corp.newsgroups.com>
Message-ID: <jwbaxter-CE1C5D.00553921022002@corp.supernews.com>

In article <3c7455a8_4 at corp.newsgroups.com>,
 "Joshua Muskovitz" <joshm at taconic.net> wrote:

> Philosophically, I would say that () and "" are different because tuples can
> contain objects of any type, while strings can only contain characters.

I don't think strings contain characters.  I think strings contain 
one-character strings.

>>> type("abc"[0])
<type 'string'>

This particular test was on
% python
Python 2.1 (#1, 04/22/01, 11:06:25) 
[GCC Apple DevKit-based CPP 6.0alpha] on darwin1

  --John


From skip at pobox.com  Tue Feb  5 20:22:02 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 5 Feb 2002 19:22:02 -0600
Subject: How do I override distutils compile options?
Message-ID: <15456.34106.697903.397614@localhost.localdomain>

I built and installed Python with the gcc flags 

    -O3 -minline-all-stringops -fomit-frame-pointer

I'm working on an extension module built with distutils.  For the time being
I'd prefer to compile with 

    -g -Wall

I can add that using distutils.core.Extension's extra_compile_args option.
Is there a way to replace the default compile args though?

Thx,

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From kkeller at speakeasy.net  Wed Feb 13 12:15:05 2002
From: kkeller at speakeasy.net (Keith Keller)
Date: Wed, 13 Feb 2002 09:15:05 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org>
    <slrna6kkpj.is0.tim@vegeta.ath.cx>
Message-ID: <pu6e4a.5ti.ln@wombat.san-francisco.ca.us>

In article <slrna6kkpj.is0.tim at vegeta.ath.cx>,
	Tim Hammerquist <tim at vegeta.ath.cx> writes:

> IIRC, "agua" _is_ feminine, but takes the masculine article when
> singular. Mi diccionario dice que "agua" es feminina, tambi?n.
> 
> OTOH, la palabra "mar" es masculina _y_ feminina. Por ejemplo:
> 
>     "el agua en la mar"     # Correct
>     "las aguas del mar"     # Correct

Es mas malo que Perl!  ;-)

(yes, it's been a decade since I spoke Spanish, why do you ask?)

--keith

-- 
kkeller at speakeasy.net
public key:  http://wombat.san-francisco.ca.us/kkeller/public_key
alt.os.linux.slackware FAQ:  http://wombat.san-francisco.ca.us/perl/fom



From MaximilianScherr at T-Online.de  Thu Feb 28 16:07:31 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Thu, 28 Feb 2002 21:07:31 -0000
Subject: listing threads?
Message-ID: <a5m66j+lg39@eGroups.com>

in the docs theres the information, that i can list all active 
threads with threading.enumerate(), but i always just get the 
Mainthread displayed in the interpreter console.

im doing somethin like:

class test(threading.Thread):
  def t(self):
       print "self"

te = test()
te.start()

then if i do threading.enumerate() i just get the mainthread int he 
list.

so how can i now list all (active) threads?




From db3l at fitlinxx.com  Mon Feb  4 15:48:27 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 04 Feb 2002 15:48:27 -0500
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <mailman.1012787645.17462.python-list@python.org> <a3kv4l$jki$1@news.efn.org> <mailman.1012795326.24429.python-list@python.org>
Message-ID: <uvgddgehw.fsf@ctwd0143.fitlinxx.com>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> writes:

> According to the help file, you need a PyOVERLAPPEDReadBuffer object here,
> whatever that is... Mark Hammond, are you out there?

You get this by using win32file.AllocateReadBuffer().  I believe
that's just a renamed buffer object, so it's possible that a normal
buffer object would work too, but there's the nice API function to use
after all.

> I've lost track; why are you passing in a buffer?  Don't you get the bytes
> read as the second return value from ReadFile?  Have you read the help
> file yet?

The second return value is actually the resultant buffer itself.  From
what I can see the only reason to pass in a buffer to ReadFile itself
is if you're using overlapped I/O and might not keep the result in
scope before the I/O completes.  In that case, you might want to
allocate the buffer in some other scope (with AllocateReadBuffer).  If
you pass the buffer in to ReadFile, than a reference to that same
buffer is what is returned by the function.

But I'm not really sure you ever need to allocate your own buffer -
after all, even if you're going to leave the scope where the function
is called, as long as you pass back the resultant buffer pointer the
object should always be alive while the I/O is ongoing.

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


From llchen223 at hotmail.com  Wed Feb 13 14:09:19 2002
From: llchen223 at hotmail.com (Lei Chen)
Date: 13 Feb 2002 11:09:19 -0800
Subject: redirect console output to widget?
References: <de6c123c.0202101612.72b27f8d@posting.google.com> <mailman.1013395757.3817.python-list@python.org>
Message-ID: <de6c123c.0202131109.2708c2f4@posting.google.com>

I think you are right.  Is there an alternative to Tkinter that is thread safe?


"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1013395757.3817.python-list at python.org>...
> << hack hack hack >>
> 
> I didn't think you were supposed to use Tkinter in parallel threads?
> 
> Am I wrong?  (Someone will tell me if I am, I'm sure...)


From martin at v.loewis.de  Tue Feb 26 04:32:41 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 10:32:41 +0100
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu>
Message-ID: <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de>

Nick Collier <nick at src.uchicago.edu> writes:

> Anyway, I thought that the Python grammar was LL(1) compliant and
> yet when I translate the grammar into the Coco/R LL(1) parser
> generator I get lots of errors like:

[...]
> Where X and Y are productions such as arglist, etc. 

My guess is that your translation into Coco/R is incorrect. I cannot
spot any ambiguity in arglist. Considering

arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**' test)

it may be that, given "argument "," argument ","', it finds itself
incapable of determining whether this is the second argument belongs
to the first subexpression or the second one. For pgen, it does not
matter: this all belongs into the production arglist, and is shifted
on the right-hand side until the first token of the follow-set is
found. If you had to split this production into multiple rules to
implement the Kleene star or the parentheses, you are likely running
into problems.

Regards,
Martin


From syring at email.com  Fri Feb  8 07:32:59 2002
From: syring at email.com (Karl M. Syring)
Date: Fri, 8 Feb 2002 13:32:59 +0100
Subject: SourceForge Python foundry ????
References: <3c6301ac.188163695@news.laplaza.org>
Message-ID: <a40hh0$1bdtaj$1@ID-7529.news.dfncis.de>

"Mats Wichmann" <mats at laplaza.org> schrieb
> What's up with the Python foundry on SourceForge? It's not listed in
> any of their lists of foundries, although if you plug in "python" in
> the standard path they use for a foundry it comes up fine...

Perhaps, that is because SF still has a grubby v1.5.2 installed?


--
Barr's Law of Recursive Futility:
If you are smart enough to use one of these... you can probably manage
without one.





From fperez528 at yahoo.com  Tue Feb 19 05:01:58 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 19 Feb 2002 10:01:58 +0000
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org> <a4ss56$akj6@news1.gtech.com> <1ovc8.54286$X64.19378143@news1.rdc2.pa.home.com>
Message-ID: <a4u0ho$e49$1@peabody.colorado.edu>

Terry Reedy wrote:

> The same compiler will probably
> 
> wrongly reject???2?*?3Liters?(=?6?Liters),
> wrongly accept??2Liters?*?3Liters?(=?6?Liters**2,?not?6?Liters),
> wrongly reject???2LitersperHour?*?3Hours?(=?6?Liters),
> 
> and similarly for division, so such a compiler is worse than useless
> unless you restrict yourself to +- operations.??Do?you?know?of?any
> compiler for any general-purpose language that correctly handles
> measurement unit arithmetic out of the box?
> 
> One can write a class (or base class and set of classes) in Python
> that will do so correctly, and I believe there is at least one such
> module available.??Such?can?probably?be?done?more?efficiently?that
> before in 2.2 by deriving from float.

Indeed:

http://starship.python.net/crew/hinsen/scientific.html

In particular:

Scientific.Physics.PhysicalQuantities: physical quantities
        with units, including unit conversion and arithmetic operations
        with unit checks.
   
An example:

In [1]: v1 = 3 m/s
In [2]: v2 = 1000 cm/h
In [3]: v1+v2
Out[3]= 3.003 m/s
In [4]: v1+v2**2
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

[ snipped traceback]

TypeError: Incompatible units

Beautiful, isn't it?

Cheers,

f.


From slhath at charter.net.nospam  Mon Feb 25 11:59:06 2002
From: slhath at charter.net.nospam (Scott Hathaway)
Date: Mon, 25 Feb 2002 10:59:06 -0600
Subject: call me
References: <u7fcig3gep8k40@corp.supernews.com> <slrna7fh30.a72.gerhard@lilith.my-fqdn.de>
Message-ID: <u7krbge1l1481b@corp.supernews.com>

Sorry all.  At moments like this, I wish that this was the stupidest thing I
have ever done, but it doesn't even come close.

:)

Scott

"Gerhard H?ring" <gerhard at bigfoot.de> wrote in message
news:slrna7fh30.a72.gerhard at lilith.my-fqdn.de...
> Scott Hathaway wrote in comp.lang.python:
> > Mom,
> >
> > Give me a call when you have a chance.
>
> Wow, a familiy of diehard Python fans :-)
>
> Gerhard
> --
> mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
> web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
> public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
> reduce(lambda x,y:x+y,map(lambda
x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))




From tc at clark.net  Sat Feb  9 12:13:00 2002
From: tc at clark.net (Thomas R. Corbin)
Date: Sat, 09 Feb 2002 17:13:00 GMT
Subject: Has anyone got a module to do logfile rotation?
Message-ID: <wMc98.73124$A51.32215049@typhoon.southeast.rr.com>

Has anyone got a module to do logfile rotation?

Thanks
Tom


From salmeri at lucent.com  Fri Feb  1 08:04:11 2002
From: salmeri at lucent.com (Joe Salmeri)
Date: Fri, 1 Feb 2002 08:04:11 -0500
Subject: Python 2.2 + Win32All 146 = Python/ASP Failures
References: <vtk68.20338$8d1.7162721@news1.rdc1.md.home.com> <3C59DA37.9070506@skippinet.com.au>
Message-ID: <a3e3p0$ss8@nntpb.cb.lucent.com>

Thanks Mark, I was hoping you would respond to the problem.

Things were working fine with 2.1.1 and Win32all build 145.

Do I have to go all the way back to framework.py in build 142 or will the
one in build 145 work?

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C59DA37.9070506 at skippinet.com.au...
> This ia a bug I am hoping to fix very soon.  If you are really keen,
> then you can grab the framework.py from win32all-142.exe - you can open
> that .exe using WinZip.
>
> Mark.
>




From brian at sweetapp.com  Mon Feb 25 02:34:46 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sun, 24 Feb 2002 23:34:46 -0800
Subject: Remote debugging ?
In-Reply-To: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com>
Message-ID: <003c01c1bdce$e9066e70$445d4540@Dell2>

Lothar wrote:
> Is there any debugger for python that allows me to do remote debugging
> from a windows 2000 client to a python script running on a Linux
> Server.

ActiveState's Komodo IDE features remote Python debugging:
http://www.activestate.com/Products/Komodo/

> It must not be freeware.

What does that mean?

Cheers,
Brian




From seeger at sitewaerts.de  Mon Feb 25 04:32:51 2002
From: seeger at sitewaerts.de (Felix Seeger)
Date: Mon, 25 Feb 2002 10:32:51 +0100
Subject: running system command in cgi problem
References: <a55a6d$4orlk$1@ID-79019.news.dfncis.de> <3C766C7D.EE4B9A40@snakefarm.org>
Message-ID: <a5d065$69es1$2@ID-79019.news.dfncis.de>

<ver?ffentlicht & per Mail versendet>

Carsten Gaebler wrote:

> You could fork a child process which executes the command and let the
> parent process wait for it to finish and print keepalive messages in
> the meantime. This could look like the following. Note the '-u'
> argument to python; it ensures that output is not buffered but sent to
> the browser immediately.
> 
> ...
>
> Hope this helps.
> cg.

Thanks for the tip.
I will use is later, at the moment all output is going to a file. The 
programm is called with &.
So it works now.


thanks again
I've looked also for a good fork example and now I have one ;) 

have fun
Felix


From jason at jorendorff.com  Fri Feb 15 19:22:16 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 15 Feb 2002 18:22:16 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <3C6D5607.3070900@wi.mit.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLAEOLDCAA.jason@jorendorff.com>

Brian Kelley writes:
> n.b. pickle and cPickle also use eval to convert representations of 
> string values into python strings so security is not a valid reason to 
> choose one over the other, yet...

I dunno, let's compare them:

  1.  pickle/cPickle
      There is a small but nonzero possibility of a yet-unknown flaw
      in either
      (a) either pickle.Unpickler._is_string_secure()
          (which I doubt, having checked the function myself) or
      (b) the cPickle equivalent
          (which I have also checked), or
      (b) the way eval() evaluates string literals
          (which might have been noticed before now).
      The chance that this hypothetical flaw might constitute a
      security hole is even smaller but still nonzero.

  2.  eval()
      Well-known gaping security hole.

It seems to me that security is a compelling reason to choose
pickle() over eval().  Am I wrong?

## Jason Orendorff    http://www.jorendorff.com/




From buzzard at urubu.freeserve.co.uk  Wed Feb 13 21:09:42 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Thu, 14 Feb 2002 02:09:42 -0000
Subject: Opening random line from file with random.shuffle()
References: <87bses6g1p.fsf@uol.com.br>
Message-ID: <a4f6d4$ju8$1@news6.svr.pol.co.uk>

"synthespian" <synthespian at uol.com.br> wrote in message
news:87bses6g1p.fsf at uol.com.br...
>
>
> Hello-
>
> I need to get a random line from a file. I read in thins NG about a
"poorly" documented random.shuffle(), but I can't get it to work:
>
>
>
>
> #!/usr/local/bin/python2.0
>
> import random
> file = open('/xxxx/yyy/zzz.txt')
>
> allLines = file.readlines()
>
> file.close()
>
> getLine = random.shuffle(allLines)
>

Shuffles it in place and returns None.  Try,

getLine = random.choice(allLines)


> print getLine,
> string = str(getLine)
>
> print string,
>
>
> All I get as output is
> "None None"
>
> Thanks for __any__ help!
>
> Regs
> Henry
>
>
>
>
>
>
> ----------------------
> _________   1001101010
> | ^   ^ |  /1111100110
> | o   o | / 0010100111
> |   v |/  1110100111
> | ----- |
> \_______/
>
> ______________________
> Micro$oft-Free Human
> 100% Debian GNU/Linux
> ----------------------
> KMFMS
>




From jason at jorendorff.com  Mon Feb 18 23:53:00 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 22:53:00 -0600
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
In-Reply-To: <3C71A6F1.24107297@ccvcorp.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEJPDDAA.jason@jorendorff.com>

Tim Legant wrote:
> isn't it:
> [...]
> And the _mome_ _raths_ _outgabe_.

Yep, for some reason I was imagining "momeraths" as one word.

Silly me.  :)

## Jason Orendorff    http://www.jorendorff.com/



From abottoni at quadrante.com  Tue Feb 19 10:56:25 2002
From: abottoni at quadrante.com (Alessandro Bottoni)
Date: Tue, 19 Feb 2002 16:56:25 +0100
Subject: Python-based "Portal System"?
Message-ID: <5.1.0.14.0.20020219163858.00a901a8@veronika.quadrante.com>

Most likely, all of you know a number of open source, pre-built "portal 
systems", like ezPublish (http://developer.ez.no), PHPNuke 
(www.phpnuke.org), PostNuke (http://www.postnuke.com), Midgard 
(http://www.midgard-project.org/) and so on.

Does anybody know if exists a Portal System like those, written in Python?

Thanks in advance

Alessandro Bottoni

PS: I know about Zope (http://www.zope.org) and WebWare 
(http://webware.sourceforge.net), already...




From scarblac at pino.selwerd.nl  Thu Feb 21 11:12:08 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Feb 2002 16:12:08 GMT
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org> <7xit8rhebq.fsf@ruckus.brouhaha.com>
Message-ID: <slrna7a6sv.ekg.scarblac@pino.selwerd.nl>

Paul Rubin <phr-n2002a at nightsong.com> wrote in comp.lang.python:
> Gustavo Cordova <gcordova at hebmex.com> writes:
> > > Next question: why is muddying the statement/expression distinction a
> > > detriment?  As opposed to the alternate view, that the 
> > > statement/expression
> > > distinction is itself a detriment?
> > > -- 
> > 
> > Or, if I may rephrase the question:
> > 
> > Why is the statement/expression distinction a detriment,
> > as opposed to the alternate view, that the statement/expression
> > distinction is a virtue?
> 
> Because we've already seen, assignments in expressions are useful.
> And blurring the statement/expression distinction has proven quite
> successful in languages like C and Lisp.

Are you sure? writing 'if (a=3) {' must be the most common bug in C
programs. I wouldn't call it successful. In fact, Python already caught
three of those for me this week as syntax errors - that would have been bugs
in C.

Also, putting multiple conceptual things on one line isn't good for code
clarity.

> Anyway, language attempts to impose "morality" on programmers tend to
> annoy programmers more than they makes code any cleaner.

When Python flagged those typos this week, I was *happy*.

> Maybe there's a principle to be found here: "design by FAQ".  If newbies
> keep arriving and asking over and over again "how do I do X?", maybe
> there's a good reason for having a way to do X.  We've seen that already
> for augmented assignments.

Most newbies tend to come from language X, think Python is great,
except it isn't quite X - why block indentation when { } would work? Why
'self'? Why not assignments in expressions? Etc.

Hollywood wanted to make a movie out of Pratchett's "Mort" book, but only if
that morbid 'Death' character could be removed.

The Python idiom for while (m := re.match('whee',x)): blah() is

while 1:
   m = re.match('whee',x)
   if not m:
      break
   blah()
   
And to me, an experienced Python programmer who hasn't really used other
languages in quite a while, that looks *much* clearer than the one liner
assignment-in-the-while version. It's a common pattern, and the eye
recognizes it immediately.

But in the end, neither of our subjective opinions matter - Guido's does.
That's probably why this language is so great.

-- 
Remco Gerlich


From robin at jessikat.fsnet.co.uk  Sat Feb 16 05:19:20 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 16 Feb 2002 10:19:20 +0000
Subject: win32 DeleteObject ??
References: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk>
 <3C6DB4D0.1020601@skippinet.com.au>
Message-ID: <dXsWSXAoIjb8EwjI@jessikat.fsnet.co.uk>

In article <3C6DB4D0.1020601 at skippinet.com.au>, Mark Hammond
<mhammond at skippinet.com.au> writes
....
>
>These few functions would be trivial to add to win32gui, but it is not 
>clear if that is all you need.
....
I've never been sure exactly how you selected what to include in any
particular module. I see that there are large swathes of swig interface.

I've managed to get a progress bar working with anygui, but typically an
installer needs some kind of pretty pictures (extracted from the exe
usually). For that one needs something of the order

        dc = GetDC(hwnd)
        bmp=LoadImage(hinst,....)
        memdc = CreateCompatibleDC(dc)
        old = SelectObject(memdc,bmp)
        BitBlt(dc,memdc,.....)
        SelectObject(memdc,old)

and then we need the way to destroy our created things eg bmp, memdc. As
others have said it's fairly easy to get these with
calldll/npstruct/windll etc etc. Installers tend to need all sorts of
win32 internals so perhaps the calldll approach is reasonable.
-- 
Robin Becker


From matthias_j at lycos.com  Thu Feb 14 11:43:02 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 14 Feb 2002 08:43:02 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <slrna6lbrp.guj.mlh@vier.idi.ntnu.no> <d7d5ebdf.0202131429.16a21cca@posting.google.com> <915a998f.0202140105.1df2e70@posting.google.com>
Message-ID: <d7d5ebdf.0202140842.1a66ca9a@posting.google.com>

> 
> There was a typo (a missing y) in the URL:
> 
>     http://www.ferg.org/easygui/index.html
>                            ^
> 
> Hamish Lawson

THANKS to all.

Matthias Janes


From thomas.heller at ion-tof.com  Fri Feb  1 05:59:21 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Feb 2002 11:59:21 +0100
Subject: py2exe version 0.3.0 - does now windows NT services
References: <a3dp5u$179scg$1@ID-59885.news.dfncis.de>
Message-ID: <a3dsee$17f557$1@ID-59885.news.dfncis.de>

Thomas Heller wrote...
> py2exe is a distutils extension to convert python scripts
> into windows exe-files (plus maybe a few dlls), able to run
> on computers without requiring a python installation.
>
>
> New in release 0.3.0:
>
>   The first release which can create windows NT services:
>   A sample service with detailed instructions is provided.
>   Builds version resources into the executables.
>   Other small bug fixes as well.
>
>
> Download from the usual location:
> http://starship.python.net/crew/theller/py2exe/
>

Peter Goode found a bug in this release which prevents
using the -O flag. Many thanks to him!

I will hide the 0.3.0 release, and prepare a 0.3.1 release
which fixes this bug. Sorry for the inconvenience.

Thomas




From brian at sweetapp.com  Tue Feb 26 01:56:02 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 25 Feb 2002 22:56:02 -0800
Subject: Proposal: local variable declarations and type advice
In-Reply-To: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <006001c1be92$a7975520$445d4540@Dell2>

Paul Rubin wrote: 

[snipped]
> Python has grown from a tiny acorn to a healthy young tree.  I believe
> any changes and additions to its design should be made with the idea
> that it may become a mighty oak one day, i.e. that it will have
serious
> native-code optimizing compilers.   Language design changes should
work
> toward that goal, or at least not impede it.

Is optimization the only purpose of your proposals? If that is the case,
they are probably not needed. The Python compiler already identifies
variables that have local-only scope. Presumably other compilers could
do the same.

Cheers,
Brian




From ae98096 at aero.iitm.ernet.in  Sun Feb 17 06:39:53 2002
From: ae98096 at aero.iitm.ernet.in (decay)
Date: 17 Feb 2002 03:39:53 -0800
Subject: Embedding python with lcc-win32
References: <335764c1.0202161031.425dd81a@posting.google.com>
Message-ID: <335764c1.0202170339.8db63a0@posting.google.com>

I got my simple python-embedded-in-C program (read below) to compile
with lcc-win32 by doing the following:

I used buildlib with the following file as argument:
-----------------
python21.dll
__imp__PyArg_Parse PyArg_Parse
__imp__PyArg_ParseTuple PyArg_ParseTuple
...
-----------------

The program now compiles, but when I execute it, it "caused an error
in <unknown>". Can anyone tell me what's happening?

yours,
Ashok.

ae98096 at aero.iitm.ernet.in (decay) wrote in message news:<335764c1.0202161031.425dd81a at posting.google.com>...
> Hello.
> 
> I've been trying to embed python in a C program. I use the lcc-win32
> compiler system. I made an import library from python21.dll using
> implib. But when I link a simple program, I get "Undefined reference
> to __imp__Py_Initialize" in test.c. I suppose its got to do with
> making the import library. Can anyone please enlighten me?
> 
> Thanks,
> Ashok.


From bougle at gluce.com  Wed Feb  6 20:08:06 2002
From: bougle at gluce.com (bougle gluce)
Date: Wed, 6 Feb 2002 17:08:06 -0800
Subject: The book "Programming Python" by van Rossum et al
References: <3C3DA255.7E22F390@inenco.no>
Message-ID: <tpk88.111$XN3.91155@news.uswest.net>

I never had the first edition but I'd say the biggest difference would be
that
it covers python 2.0. also according to guido's foreword "the book has
changed
about as much as the language and the python community" if that tells
you anything.

"Hans Kristian Ruud" <hans at inenco.no> wrote in message
news:3C3DA255.7E22F390 at inenco.no...
> I have the 1st edition. What are the differences between that and the
> 2nd edition?
>
> Thanks in advance
>
> Hans Kristian
>
>
>




From aleax at aleax.it  Mon Feb 18 11:30:05 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 16:30:05 GMT
Subject: super(obj)?
References: <mailman.1014045511.10659.python-list@python.org>
Message-ID: <h_9c8.69760$om6.1918534@news1.tin.it>

Gustavo Niemeyer wrote:

> Hello everyone!
> 
> AFAICS, this seems to be the most useful version of super:
> 
> super(obj.__class__, obj)

I don't get it.  Why would this be "most useful"?  E.g.:

class X(object):
  def foo(self):
    print 'X.foo'
    try: supermethod = super(self.__class__, self).foo
    except: pass
    else: supermethod()

class Y(X):
  def foo(self):
    print 'Y.foo'
    try: supermethod = super(self.__class__, self).foo
    except: pass
    else: supermethod()

    
This systematically raises 'maximum recursion depth
exceeded'.  More generally, if each method restarts
the search-for-super 'from the root' (self.__class__),
how is the search EVER going to terminate...?  Just
by exceeding recursion limits, no?

I guess I must be missing something, but really can't
see what.  Explain please?  Thanks!


Alex



From brueckd at tbye.com  Mon Feb 18 19:54:11 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Mon, 18 Feb 2002 16:54:11 -0800 (PST)
Subject: Adding static typing to Python
In-Reply-To: <a4s51k$8pd1@news1.gtech.com>
Message-ID: <Pine.LNX.4.33.0202181634230.4734-100000@ns0.tbye.com>

On Tue, 19 Feb 2002, Peter Milliken wrote:

> The loose typing of Python is perceived as one of its big features :-).
> There are regular discussions on the usefulness of Python for "larger"
> projects. Many claim that they use it without any problems

I'm one of those people who like to make those sorts of claims. :)

> - I suspect that
> it is from a background of ignorance and lack of metrics

Rather than from ignorance, maybe it's from actual experience using it.
Have you considered the possibility that the claims might be true? In this
job as well as the past few our Python code has always had less time
devoted to developing, testing, maintaining, and refactoring it. It's not
that there's hidden costs involved - we tend to write it and move on to
something else because it just works.

> The only way in Python to find these is through (exhaustive) testing -
> such rigorous testing *requires* you to develop an exhaustive test
> suite that is capable of running periodically i.e. nightly etc to
> catch such errors.

The current state of software development is that large programs need
exhaustive and rigorous testing anyway. The reason that the loose typing
doesn't kill you is that good testing finds misnamed variable bugs like
you cited above as well as bazillions of other bugs.

> Whether the people who support Python for larger projects have the
> discipline within their project team to maintain such an effort is
> unclear - there has been no mention of it at all.

Really? Good testing comes up every single time this topic comes up, and
invariably people talk about how they *do* write tests for their code, and
how good testing helps them find all sorts of bugs.

> So I doubt that the proponents of Python in these instances have any
> idea of the true cost of this language "feature" - they appear to be
> wound up in how quickly they can produce a working prototype (or
> whatever) so I suspect their day to day work involves very small,
> manageable projects where the lost efficiency due to loose typing
> isn't really that noticable for them i.e. they "live with it" :-).

Might be true for lots of people, but nearly every single Python module in
our company's code has either built-in or accompanying tests that run
every night. The same is true for our non-Python code. The dynamic or
static typing of the language makes little real difference - if you don't
test, you overlook bugs, pure and simple.

> When I compile a Ada program cleanly I know that 95% of the bugs are out of
> it - before I even start testing! :-) Of course, it takes me longer to write
> up front but overall development effort (in my subjective opinion - since I
> have no metrics :-)) is that it is much lower.

Whatever floats your boat, I guess. Very few of my bugs (in any language)
are things caught by a compiler and if anything, the warm fuzzies because
the code compiles is often a false sense of security. The 95% seems very
high, even considering Ada's strictness - to that I'd say that with Python
95% of the bugs never make it into the code in the first place. :)

Have fun,
Dave




From skip at pobox.com  Tue Feb 12 17:36:06 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 16:36:06 -0600
Subject: Re-throwing an exception without loosing the stack trace
In-Reply-To: <20020212174235.A17928@doublegemini.com>
References: <20020212174235.A17928@doublegemini.com>
Message-ID: <15465.39126.250327.239174@beluga.mojam.com>

    Clark> How do I catch and then re-throw an exception without loosing the
    Clark> information about the stack trace?  I've looked around alot and
    Clark> don't see the "obvious" answer.

Assuming you mean something like

    try:
        bunch
        of
        stuff
        that
        might
        barf
    except:
        clean
        up
        database
        turds
        then
        reraise

when you are finished cleaning up you can just execute the raise statement
with no arguments.  Better might be this:

    try:
        bunch
        of
        stuff
        that
        might
        barf
    finally:
        clean
        up
        database
        turds        

That way the database stuff gets cleaned up no matter what happened.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From spamtrash at softhome.net  Wed Feb 13 08:17:49 2002
From: spamtrash at softhome.net (Johannes Gijsbers)
Date: 13 Feb 2002 05:17:49 -0800
Subject: Hidden files, selecting directories, & taxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
Message-ID: <e87ce949.0202130517.32ba5974@posting.google.com>

chris_mk at hotmail.com (Christopher) wrote in message news:<cc14131d.0202121623.47ffaa99 at posting.google.com>...
> Hello everybody,
>     I had a couple things to ask:
> 
> 1)  Is it possible to generate a hidden file (under Windows)? 

I guess you'll have to use the win32all extensions[1] by Mark Hammond.
The win32api module includes the function SetFileAttributes[2, 3]
which you could use for this.

> 2)  Is there a simple and easy way to allow the end-user to select a
> directory?

I'm not very familiar with Tkinter, but I don't think such a dialog is
the standard Tkinter distribution. But there's one in Tix,
DirSelectDialog[4, 5]

[1]http://starship.python.net/crew/mhammond/win32/
[2]ASPN: http://aspn.activestate.com//ASPN/Reference/Products/ActivePython/PythonWin32Extensions/win32api__SetFileAttributes_meth.html
[3]MSDN: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/fileio/filesio_9a2b.asp
[4]Tix File Selectors:
http://www.python.org/doc/current/lib/node520.html
[5]DirSelecDialog: http://tix.sourceforge.net/dist/current/man/html/TixCmd/tixDirSelectDialog.htm


From jason at jorendorff.com  Tue Feb 26 19:20:41 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 18:20:41 -0600
Subject: thread question
In-Reply-To: <a5gld7+3jal@eGroups.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEILDFAA.jason@jorendorff.com>

> hmm if i now just have a random threadobject, like the one returned 
> from enumerate()
> 
> how can i then access the data(attributes, functions...) of that 
> thread?

The same way you get the attributes or methods of any other object:

  threadobject.attribute
  threadobject.method()

## Jason Orendorff    http://www.jorendorff.com/




From jessefw at loop.com  Mon Feb  4 15:03:56 2002
From: jessefw at loop.com (Jesse F. W)
Date: Mon, 4 Feb 2002 12:03:56 -0800
Subject: I love obfuscated python!
In-Reply-To: <zay78.23173$l93.4207854@newsb.telia.net>
Message-ID: <3C5E78AC.3893.337485@localhost>

Fredrik Lundh wrote:
> "Jesse W" wrote:
> > Horrey!  I have finally solved the piece of obfuscated python
> > posted by Greg Stein in November of 1998.*
> note that the solution was posted in october 1998.
Yes.  As I say in my message:
> > See the above mentioned solution message 
> > if you are interested)
I was aware that it had been solved before.  I was just happy that _I_ 
had just solved it, and wanted to share my thoughts on the value of 
obfuscated code as an art form.
> guess someone was using not only the time machine, but
> also some kind of yet-to-be-seen autodecoding tool...
<grin>
	not-so-magical-l'y,
			Jesse W



From QnickQm at alum.mit.edu  Fri Feb  1 02:38:37 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Fri, 01 Feb 2002 07:38:37 GMT
Subject: Comments invited:  PEP 279 Enhanced Generators
References: <a3deaj$b54$1@bob.news.rcn.net>
Message-ID: <slrna5khlg.bi4.QnickQm@localhost.localdomain>

In article <a3deaj$b54$1 at bob.news.rcn.net>, Raymond Hettinger wrote:
> Your review and comments are invited for PEP 279:
>    http://python.sourceforge.net/peps/pep-0279.html
> 
> In summary, it proposes:
> 
> 1.  Four new built-in functions:  xmap, xfilter, xzip, and indexed

I like these a great deal; I can certainly imagine using them on a 
regular basis.

One question: why isn't "indexed" named "xindexed"?  This would 
follow the (implicit?) convention that the x-versions of functions
are lazy-evaluating, and the non-x versions evaluate greedily.

Also, one might add an "indexed" builtin along with "xindexed" for
the sake of orthogonality.  
   
BTW, I think I've noticed a bug in your code:
    In your implementation of "indexed", you probably mean to write 
   "gen.next()" rather than "collection.next()"

> 2.  Generator comprehensions:  [yield int(x) for x in floatgen()]

Strongly in favor, with the current syntax. 

> 3.  Feeding data into a generator using an optional argument to .next()

Specification notes: What happens when you try to pass an argument to
a .next() that isn't expecting one?  What happens when you call a
.next() that's expecting an argument without an argument?  This needs
to go in.

Here, I have to say I'm skeptical.  I'd like this feature, as it would
seem to enable more general-purpose control flow, but I don't have a lot
of experience with this kind of, and your example didn't convince me
it was worth the mental energy it takes me to understand this code.

Of course, this wouldn't be the first time that an ingenious construction
has gone over my head!  But to enlighten me as to why this feature is
good, it might be a good idea to cite a more fleshed-out example.  Which
other languages provide features for lazy consumers?  Who uses them?

I'd especially like to see an example of real-world code that becomes
far easier to write with the use of lazy consumers.  

> 4.  Providing a mechanism to raise exceptions in a generator

Instead of just gen.throw( ExceptionClass ), you should explicitly 
allow gen.throw( object ).  

Also, you might want to clarify whether this mechanism is only useful
in the context of lazy consumers, or whether it's handy for regular
generators as well.

Yours,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From jkraska at san.rr.com  Mon Feb 25 22:15:39 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 03:15:39 GMT
Subject: Time to rename Stackless?
References: <3C7A151F.1090709@tismer.com> <mailman.1014690284.11692.python-list@python.org>
Message-ID: <mnvl7ukagdojifcev30gibrj1dr7c7euja@4ax.com>

>> It is still stackless in the sense that order of execution is not
>> restricted by the C stack.
>
>Even so, I still think it's a poor name, because it
>focuses on an implementation detail rather than
>telling you what it can do for you.

Possibly. But coming up with a glib sounding name which
is evocative of Stackless' simple purpose might not be
so easy, and meanwhile, most of the people that care about
Stacklessness and continuations are hardcore computer
science junkies, folks who've been exposed to some fairly
obscure languages (sorry, scheme guys!), or people with
very highly specific needs.

C//



From buzzard at urubu.freeserve.co.uk  Sun Feb 24 12:16:27 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Sun, 24 Feb 2002 17:16:27 -0000
Subject: Submenu in Boa 0.05?
Message-ID: <a5b79q$s6r$1@news5.svr.pol.co.uk>

Does anyone know how I can add a submenu in Boa.  I can achieve what I want
by using,

parent.AppendMenu()

but Boa won't parse the code correctly (and I lose all the menus).  Thanks
in advance.

Duncan Smith




From lfini at arcetri.astro.it  Fri Feb 15 07:04:13 2002
From: lfini at arcetri.astro.it (Luca Fini)
Date: Fri, 15 Feb 2002 13:04:13 +0100
Subject: Bug or feature? 'abc'.split('') rejects empty separator
In-Reply-To: <3c68118e.3802800336@news>
References: <mailman.1013382318.2151.python-list@python.org> <3c68118e.3802800336@news>
Message-ID: <Pine.LNX.4.44.0202151300400.5874-100000@localhost.localdomain>

On Mon, 11 Feb 2002, David C. Ullrich wrote:

> On Sun, 10 Feb 2002 18:05:01 -0500, "Tim Peters" <tim.one at home.com>
> wrote:
> 
> >[Bengt Richter]
> >>  >>> 'abc'.split('')
> >>  Traceback (most recent call last):
> >>    File "<stdin>", line 1, in ?
> >>  ValueError: empty separator
> >>
> >> Wouldn't it make sense to return list('abc') ?
> >
> >[Neil Schemenauer]
> >> It would also make sense to return ['a', 'b', 'c'].
> >

My 2 cent worth opinion is that it should return ['a', 'b', 'c'].

The characters are separated by "nothing". It also would maintain simmetry
with join() which is not bad.

BTW: this is how the corresponding functions work in perl. Maybe this is a 
reason to do it differently ;-)

-- 
--------------------------------------------------------------------------
       -- )     Luca Fini                            Tel: +39 055 2752 307
 ___    |\      Osservatorio Astrofisico di Arcetri  Fax: +39 055 2752 292
/   |   | |-_   L.go E.Fermi, 5 +-----------------------------------------
(___|___//___)  50125 Firenze  /   WWW: http://www.arcetri.astro.it/~lfini
 (_)      (_)   Italia        / e-mail: lfini at arcetri.astro.it
-----------------------------+--------------------------------------------



From sheershion at mailexpire.com  Fri Feb 22 13:56:56 2002
From: sheershion at mailexpire.com (Robert Amesz)
Date: Fri, 22 Feb 2002 18:56:56 -0000
Subject: wxlayoutconstraints inside a wxDialog ?
References: <3c76666e.18420093@news.easynews.net>
Message-ID: <Xns91BDC97DAC9B9rcamesz@amesz.demon.nl>

Marcus Stojek wrote:

> I'm using wxPython and I'm trying to use
> layout constraints in a Dialog.
> why doesn't this work?
> 
> --------------------------------
> 
> d=CompoDlg(self)
> -----
> class CompoDlg(wxDialog):
>     def __init__(self,parent):
>         BORDERPIX=20
>         ID_E11=wxNewId()
> 
>         win=wxDialog(parent, -1, "Title", wxDefaultPosition,
>                                wxSize(350, 200))
>         win.SetAutoLayout(true)
> 
>         E11 = wxStaticText(win,-1,"E11:",wxDefaultPosition,
>                                        wxDefaultSize)
>         lc=wxLayoutConstraints()
>         lc.top.SameAs(win,wxTop,BORDERPIX)
>         lc.left.SameAs(win,wxLeft,BORDERPIX)
>         lc.width.AsIs()
>         lc.height.AsIs()
>         E11.SetConstraints(lc)
> --------------------------------

There is one basic thing wrong with this code: you're not initializing 
your class correctly. You should call the constructor of the base 
class, but you're creating a new wxDialog instead. With that in mind, 
the window which needs laying out is simply 'self', not 'win'. So, 
after correcting, rearranging, and removing some unused and redundant 
stuff, you might end up with something like this (warning, untested 
code):


class CompoDlg(wxDialog):
    def __init__(self,parent):
        wxDialog.__init__(self, parent, -1, "Title", wxDefaultPosition,
                        wxSize(350, 200))

        self.SetAutoLayout(true)


        E11 = wxStaticText(self,-1,"E11:")

        BORDERPIX=20
        lc=wxLayoutConstraints()
        lc.top.SameAs(self,wxTop,BORDERPIX)
        lc.left.SameAs(self,wxLeft,BORDERPIX)
        E11.SetConstraints(lc)



By the way, I assume you know that

    d=CompoDlg(self)

will not, by itself, actually show the dialog? Do

    result = d.ShowModal()

for that.

Having gone through all that, you might want to consider using sizers 
instead of constraints. As far as development is concerned, constraints 
are pretty much dead, or so I gather, and sizers are very much alive. 
Furthermore, it has become the preferred layout method of the wxWindows 
developers, so who am I to argue with them?


Robert Amesz
-- 
P.S. Take a look at the wxPython mailing list. It's somewhere on 
http://lists.wxwindows.org


From cfelling at iae.nl  Sat Feb 23 12:47:23 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 23 Feb 2002 18:47:23 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl> <a56o39$7dg$1@animus.fel.iae.nl> <3C7761D4.7050100@student.kun.nl> <a58k1m$8t1$1@animus.fel.iae.nl>
Message-ID: <a58kjb$927$1@animus.fel.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:

> When you write to `i' you don't add arguments to the commandline used
> to start `profit',  you're writing to the inputstream of `pdbs', quit a

oeps, `pdbs' should read `profit'

-- 
groetjes, carel


From threeseas at earthlink.net  Wed Feb  6 23:28:46 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:46 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net> <3c60c05c$0$79564$3c090ad1@news.plethora.net>
Message-ID: <2106.802T789T14243611threeseas@earthlink.net>

On 06-Feb-02 00:34:20 Peter Seebach <seebs at plethora.net> wrote:
>In article <1372.801T1070T13994028threeseas at earthlink.net>,
>Timothy Rue <threeseas at earthlink.net> wrote:
>>Why would  be interested in a programming language by a company that has
>>been found guilty of Anti-trust? Not to mention that language is not
>>available anywhere near as much as something like Python.

>Because the language fits your project a lot better.  The *community* is
>perfect for you.  These people share your basic mindset.  They're heading
>where you seem to be already.

>>It'd really be a good thing if the MS cronies state so in their sig.

>I am not a MS crony.

>>This way I don't have to try and figure out why you have a problem knowing
>>the difference between what is idea and what is natural law physics. For
>>MS seems to think it own reality. It doesn't!

>This isn't about ownership, it's about finding the community most naturally
>suited to your project.  The C# people are the community least contaminated
>by existing notions of software engineering.

>-s

No, .NET is not the correct way, but rather an attempt to assume ownership
of, via evolution or changes over a period of time, about it. But will it
really ever get there?

C# is a subtopic of .NET / Mono, as is many other programing languages,
including Python.

The community you speak of is not inclusive of the end user/consumer, but
rather a community of programmers that are likley to follow MS mindset
programming.

And MS is a company that believe that by stating gsomething, makes it so.

But that's only because of the weak minded that have been given no other
option to compare to.

If you have never done it all the way right, how will you know when you
do?


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From magnus at thinkware.se  Fri Feb 22 14:11:35 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 20:11:35 +0100
Subject: Can Python COM execute a Microsoft Office macro?
References: <mailman.1014398136.1669.python-list@python.org>
Message-ID: <3C7697E7.1030706@thinkware.se>

Mike Brenner wrote:

> Is there a way for the Python COM interface to execute a macro in a Microsoft Office product like Project, Word, or Excel?


Sure. I suggest that you record macros in Excel or whatever to see
what they would look like in VBA, then it's fairly simple to translate.

I opened an Excel session from Python like this:

 >>> from win32com.client import Dispatch
 >>> xl = Dispatch("Excel.Application")
 >>> xl.Visible=1
 >>> xl.Workbooks.Add()
<COMObject Add>

Then I recorded a macro in Excel -- the one I am to run. Since I don't
remember exactly how to run macros, I recorded that too. In other words,
I did "record macro" again, and while I was recording, I opened the
macro menu and clicked to run my first macro. Then I stopped the
recording of the seconed one. Finally I opened the macro editor for this
last one. It looked like this (Swedish Excel 2000).

Sub Makro2()
'
' Makro2 Makro
' Makrot inspelat 2002-02-22 av Magnus Lyck?
'

'
     Application.Run "Bok1!Makro1"
End Sub

"Application" is my variable "xl" above and Python likes (), so

 >>> xl.Run("Bok1!Makro1")

will run the macro. Easy, right?




From eriggs at kollmorgen.com  Sat Feb 16 10:12:24 2002
From: eriggs at kollmorgen.com (eriggs)
Date: 16 Feb 2002 07:12:24 -0800
Subject: How to disable a button control in wxPython?
References: <3bff5060.0202150900.77c5f5fa@posting.google.com> <Xns91B6DE74F368Crcamesz@amesz.demon.nl>
Message-ID: <3bff5060.0202160712.14a16f46@posting.google.com>

Robert Amesz <sheershion at mailexpire.com> wrote in message news:<Xns91B6DE74F368Crcamesz at amesz.demon.nl>...
> eriggs wrote:
> 
> > I'm somewhat new to Python, and I'm trying to figure out how to
> > disable a button control.
> 
> Use use button.Enable(true) instead of button.Enable(), and use 
> button.Enable(false) instead of button.Disable(). That should do the 
> trick. The documentation seems to be wrong concerning that, at least 
> for wxPython.
> 
> Robert Amesz


That did it... Thanks!!  BTW, I'm reposting my modified source in case
anyone else can benefit.

---  Earl

######################################################################
# [ Contents of PYWRUN.bat... ]
#@echo off
#set PYTHONPATH=c:\python21\lib;c:\python21\libcust;c:\pythprgs\lib
#c:\python21\pythonw %1
#
# Usage:  pywrun my_file.py
######################################################################
from wxPython.wx import *

class ButtonFrame(wxFrame):
    'Creates a frame with a single button in the center'
    def __init__(self):
        wxFrame.__init__(self, NULL, -1, 'wxPython',
                         wxDefaultPosition, (330, 210))

        panel = wxPanel(self, -1) 

        #wxTextCtrl(wxWindow* parent, wxWindowID id, 
        #           const wxString& value = "", 
        #           const wxPoint& pos,
        #           const wxSize& size = wxDefaultSize, 
        #           long style = 0, const wxValidator& validator, 
        #           const wxString& name = "text")
        tc = wxTextCtrl(panel, -1, "", wxPoint(10, 10), \
                    wxSize(300, 100), wxTE_MULTILINE|wxTE_READONLY)
        self.msgtext = tc

        button1 = wxButton(panel, 111, 'Start', wxPoint(10, 130), \
                           wxSize(100, 30))
        EVT_BUTTON(self, 111, self.onButtonStart)

        button1.SetDefault()
        self.StartButton = button1
 

        button2 = wxButton(panel, 112, 'Quit', wxPoint(210, 130), \
                           wxSize(100, 30))
        EVT_BUTTON(self, 112, self.onButtonQuit)
        EVT_CLOSE(self, self.OnCloseWindow)

        self.QuitButton = button2

        # Disable Quit button
        self.QuitButton.Enable(false)
        # Enable Start button
        self.StartButton.Enable(true)


    def onButtonStart(self, event):
        'Create a message dialog when the button is clicked'
        dlg = wxMessageDialog(self, 'Starting...',\
                              'Start', wxOK)
        dlg.ShowModal()
        dlg.Destroy()

        self.setText("Starting program...")
        # Enable Quit button
        self.QuitButton.Enable(true)
        # Disable Start button
        self.StartButton.Enable(false)
        # Start program loop


    def onButtonQuit(self, event):
        'Create a message dialog when the button is clicked'
        dlg = wxMessageDialog(self, 'Quitting...',\
                              'Stop', wxOK)
        dlg.ShowModal()
        dlg.Destroy()

        self.setText("Stopping program...")

        self.Close(true)


    def OnCloseWindow(self, event):
        self.Destroy()


    def setText(self, msg):
        'Create a message dialog when the button is clicked'
        self.msgtext.Clear()
        self.msgtext.AppendText(msg)


class App(wxApp):
    def OnInit(self):
        'Create the main window and insert the custom frame'
        frame = ButtonFrame()
        frame.Show(true)
        return true # Yes, continue processing

# Create the app and start processing messages
app = App(0)
app.MainLoop()


From sholden at holdenweb.com  Thu Feb  7 16:01:00 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 16:01:00 -0500
Subject: IIS ASP Python 2.2 script works first time only
References: <f36760cb.0202071137.61bc8027@posting.google.com>
Message-ID: <3WB88.96393$KM2.3588702@atlpnn01.usenetserver.com>

"Stephen Arndt" <stephen.arndt at bethsteel.com> wrote ...
> Have a simple Python script hello.asp to create a web page which
> displays current date and time.
>
> <%@LANGUAGE="Python"%>
> <% python script code here %>
>
> After upgrading from Python 2.1 to 2.2 with win32all-146.exe my
> hello.asp script runs first time OK. Then just a blank page is
> displayed in IE browser window next Refresh or activation.  I am
> running Python 2.2 on WindowsNT 4.0 server and IIS 4.0.
>
> Any ideas why the hello.asp script would generate HTML code only the
> first time a new hello.asp file is copied to the server?

You've hit a bug. Mark Hammond's recent advice to another poster with the
same problem involved overwriting one particular element with an earlier
release's version: you could probably ggogle forit if you chose. Maybe it's
fixed in CVS if you want to compile your own version, but I dunno.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From huaiyu at gauss.almadan.ibm.com  Tue Feb 26 13:50:38 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Tue, 26 Feb 2002 18:50:38 +0000 (UTC)
Subject: Precompiled regular expressions slower?
References: <3c7bb531$0$3930$b45e6eb0@senator-bedfellow.mit.edu>
Message-ID: <slrna7nm7u.11s.huaiyu@gauss.almadan.ibm.com>

On Tue, 26 Feb 2002 11:17:53 -0500, Peter Bienstman <pbienst at mit.edu> wrote:
>
>re1 = compile(r"....") 
>...
>line = inputfile.readline()
>while line:
>  s = search(re1, line)
>  ...
>  line = inputfile.readline()
>
>Strangely enough this new version seems to run about 30% *slower* than the 
>old version.
>
>Am I missing something here? (I'm using Python 2.1)

Yes.  Use re1.search(line) instead.  

The times of the following three tests are:
1.94322395325   # search(string, string)
2.62431204319   # search(re_obj, string)
0.667925000191  # re_obj.search(string)

Someone care to explain why the second ends up slower than the first?

Huaiyu

# =========================== test for re speed ===============================
from re import search, compile
from time import time

a = 'dsa'
b = 'adsfasf dsaf adsf a'

def test1(a, b):
    _time = time();     
    for i in range(100000):
        search(a, b)
    print time() - _time

def test2(a, b):
    a = compile(a)
    _time = time();     
    for i in range(100000):
        search(a, b)
    print time() - _time

def test3(a, b):
    a = compile(a)
    _time = time();     
    for i in range(100000):
        a.search(b)
    print time() - _time

test1(a, b)
test2(a, b)
test3(a, b)




From op73418 at mail.telepac.pt  Mon Feb 25 18:43:03 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 23:43:03 +0000
Subject: On inner classes
References: <mailman.1014676694.26824.python-list@python.org>
Message-ID: <ouil7u412rei1i1mearruh4vjuuf8p703k@4ax.com>

On Mon, 25 Feb 2002 17:36:02 -0500, mcherm <mcherm at destiny.com> wrote:

>bjorn writes:
>> You can't. Think about it. Your inner_class is at class scope in
>> outer_class. What you're after is a value of an *instance* of
>> outer_class, however you have no handle to such an instance from within
>> inner_class, ergo you can't get to it.
>
>Maybe I can make this a little clearer. Your use of the term "inner 
>class" suggests that you might be coming from a Java background. In 
>java, there are things called inner classes, and the way that they work 
>is that each instance of the inner class is "magically" connected to an
>instance of the outer class.
>
>That's not what happens when you declare one class inside of another in 
>Python -- in Python the "inner" class is just another class, and its 
>instances are not "magically" bound to instances of the outer class. 
>It's as if (in Python language) they were declared to be STATIC inner 
>classes.
>
>Most of the time, this is the behavior that you want -- inner classes in 
>Java were invented to solve problems that are really easy to handle in 
>Python because of the existance of bound methods. (I can explain what 
>that meant if you like.) But as you might expect, if you actually DO 
>want the java-like behavior, Python allows it, you just have to be 
>explicit and explain what you mean.
>
>Here's some code illustrating what I mean:
>
>----------------<sample>--------------------
>Python 2.2c1 (#27, Dec 14 2001, 13:15:16) [MSC 32 bit (Intel)] on win32
>Type "copyright", "credits" or "license" for more information.
>IDLE 0.8 -- press F1 for help
> >>> class Outer:
>	def makeAnInner(self):
>
>		class Inner:
>			def __init__(self, myOuter):
>				self.myOuter = myOuter
>			def innerMethod(self):
>				print "I'm an inner, my outer is %s" % self.myOuter
>				self.myOuter.outerMethod()
>		anInner = Inner(self)
>		anInner.innerMethod()
>	def outerMethod(self):
>		print "This is an outer method."
>
>		
> >>> anOuter = Outer()
> >>> anOuter.makeAnInner()
>I'm an inner, my outer is <__main__.Outer instance at 0x00954E90>
>This is an outer method.
>----------------</sample>--------------------
>
>There are only two things you have to do differently here than in 
>Java... you have to declare the __init__() method of Inner to take a 
>myOuter parameter and store it, and you have to explicitly pass the 
>outer when you create it in the line "anInner = Inner(self)". By 
>requiring you to explicitly create this outer-inner relationship when 
>you want it, Python adds the ability to NOT have the relationship when 
>you DON'T need it.
>
>-- Michael Chermside
>

Thanx, Both Bjorn and Michael, great explanation!

Gon?alo Rodrigues


From cbrown at met.co.nz  Tue Feb 12 20:15:38 2002
From: cbrown at met.co.nz (Colin Brown)
Date: Wed, 13 Feb 2002 14:15:38 +1300
Subject: UTC time conversion to internal time
Message-ID: <49ja8.17872$N31.902613@ozemail.com.au>

I wish to convert UTC times to internal time representation (the inverse of
time.gmtime). There does not seem to be any function to do this. Does anyone
have a clean solution please.

Note the time.mktime() daylight savings flag does not help.

Python 2.1 on Windows NT:

>>> import time
>>> time.gmtime()
(2002, 2, 13, 0, 51, 46, 2, 44, 0)
>>> time.gmtime(time.mktime((2002, 2, 13, 0, 51, 46, 2, 44, 0)))
(2002, 2, 12, 12, 51, 46, 1, 43, 0)
>>> time.gmtime(time.mktime((2002, 2, 13, 0, 51, 46, 2, 44, 1)))
(2002, 2, 12, 11, 51, 46, 1, 43, 0)
>>> time.gmtime(time.mktime((2002, 2, 13, 0, 51, 46, 2, 44, -1)))
(2002, 2, 11, 11, 51, 46, 1, 43, 0)
>>>

This does work for localtimes:
>>> time.localtime()
(2002, 2, 13, 0, 51, 46, 2, 44, 1)
>>> time.localtime(time.mktime((2002, 2, 13, 0, 51, 46, 2, 44, 1)))
(2002, 2, 13, 0, 51, 46, 2, 44, 1)
>>>

Thanks
Colin Brown
PyNZ





From hankc at nospam.com  Wed Feb 27 02:45:34 2002
From: hankc at nospam.com (HankC)
Date: Wed, 27 Feb 2002 07:45:34 GMT
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net> <ad052e5c.0202261950.72991328@posting.google.com>
Message-ID: <3c7c8b74.63255807@news.la.sbcglobal.net>

On 26 Feb 2002 19:50:56 -0800, danb_83 at yahoo.com (Dan Bishop) wrote:

>
>If your data file is small enough to fit in memory, you can use
>
>dict = {}
>datafile = open(..., "r")
>for line in datafile.readlines():
>   name = line.split(",")[0]
>   dict[name] = line[:-1] # strip trailing \n
>datafile.close()
>

Thanks all, for the help.  I have one more question regarding resource
management.  My altered version of the above code:

dict = {}
datafile = open(..., "r")
datalines = datafile.readlines()
for line in datalines:
   name = line.split(",")[0]
   dict[name] = line[:-1] # strip trailing \n
###
datafile.close()
del(datalines)

At the ### mark, I looks like have have two versions of the data
resource - one list and one dictionary.  My questions:

- does the del(datalines) call free the memory (not necessarily
immediately) for the list resource?

- does the original version of the function, by combining readlines()
with the for statement, negate the need for freeing the list?  In
other words, will the datafile.close() statement in itself do the
trick?

I appreciate the patience!  Things like garbage collection and for
loops iterating through lists take a little getting used to.



From jkraska at san.rr.com  Sun Feb  3 14:59:12 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 19:59:12 GMT
Subject: Java vs Python Benchmarks: Java is faster
References: <sdqp5u49m52b1u8vq1fe1gtm7gjej7t1fj@4ax.com> <mailman.1012761186.28271.python-list@python.org>
Message-ID: <pr5r5ucse0m255ha67r3cuiug2l9eobps9@4ax.com>

>On behalf of all of us that don't know what we're talking about and
>do need your help...

Ah, but you're one of those who _do_ know what you're talking about.
For example, the following comment illustrates knowledge of one of the
very first things one notices in profiling Python itself:

> 1. Ultra-late binding of, and consequent repetitive lookups of,
>    global names, attributes, methods, module members, etc.

And likewise this:

> Clearly there's something to gain by moving the huge ceval.c
> loop/switch from runtime to compile time.

Type declarations target higher hanging fruit. For example, there
is quite a lot more which can be achieved in the pursuit of #1 than
has been so far, _without_ the gains which type declarations would
offer. It would therefor follow to my sorry logical mind that
_first_ attempting a declaration system is a premature optimization.

C//



From fperez528 at yahoo.com  Sun Feb 17 15:24:48 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 17 Feb 2002 20:24:48 +0000
Subject: execute file interactively
References: <e9e6dd1f.0202171731.6cf2a5e2@posting.google.com>
Message-ID: <a4psa1$5jo$2@peabody.colorado.edu>

Jason wrote:

> When using the Python interpreter interactively, how can I execute a
> file, so I can use the function definitions and data defined in it
> interactively.

See:
http://www-hep.colorado.edu/~fperez/ipython/

The @run command does exactly what you want, including passing things that 
will look to your file like command line arguments.

In regular python, you can use execfile('filename.py'). There may be 
namespace issues there, and if you want to pass arguments you need to fix 
sys.argv yourself.

Cheers,

f.


From threeseas at earthlink.net  Fri Feb 15 20:45:37 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:37 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <uk7tkylre.fsf@synopsys.com> <a492j6$18om$1@nntp6.u.washington.edu> <slrna6gegr.vsl.philh@comuno.freeserve.co.uk> <uk7tiy2m8.fsf@synopsys.com>
Message-ID: <1417.811T120T11603258threeseas@earthlink.net>

On 12-Feb-02 13:30:23 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> phil hunt writes:

>> On 11 Feb 2002 18:29:58 GMT, Donn Cave <donn at u.washington.edu> wrote:
>>> Quoth David Masterson <dmaster at synopsys.com>:

>>> | Or you haven't given us anything we can deal with.  Note, if we wanted
>>> | to remain ignorant, we could simply just stop talking to you.  You
>>> | have an audience and your "10 minutes of fame", but, if you cannot
>>> | express your problem in understandable fashion, we will go on to other
>>> | things that we do understand.

>>> Who's "we" - you have a mouse in your pocket?

>> No, he's just pleased to see us.

>All right, so I used the king's "we" a little too freely.  I had no
>intention of going on with it and this was my way of saying "You've
>lost me as an audience, Tim".

>--
>David Masterson                dmaster AT synopsys DOT com
>Sr. R&D Engineer               Synopsys, Inc.
>Software Engineering           Sunnyvale, CA


Giving in to peer pressure huh?

Just can't get out of that "we" huh?


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From NKluge at optonline.net  Sun Feb 17 14:11:30 2002
From: NKluge at optonline.net (Nicola Kluge)
Date: Sun, 17 Feb 2002 19:11:30 GMT
Subject: nested scopes
Message-ID: <CfTb8.129031$th4.36971458@news02.optonline.net>

I installed 2.1.2 binary but I can't get this to work

>>> from __future__ import nested_scopes
>>> def B():
            spam = 'bacon'
            def C():
                  print "spam comes with %s" % (spam)
             C()


SyntaxError: local name 'spam' in 'B' shadows use of 'spam' as global in
nested scope 'C' (<pyshell#1>, line 1)

I thought that once I imported nested_scopes I would be able to define
function B.




From SBrunning at trisystems.co.uk  Thu Feb 21 04:25:28 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 09:25:28 -0000
Subject: What values are considered false?
Message-ID: <31575A892FF6D1118F5800600846864DCBCCCF@intrepid>

> From:	Roy Smith [SMTP:roy at panix.com]
> If you want a really odd concept of falsity, look at SQL.  NULL is neither
> 
> true nor false!  Or, looking at it a different way, anything non-NULL is 
> neither equal to, nor not equal to NULL.
 
NULL = NULL is false, too, but then again, so is NULL != NULL.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk






-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From fbartlet at optonline.net  Tue Feb  5 15:51:26 2002
From: fbartlet at optonline.net (Frederick H. Bartlett)
Date: Tue, 05 Feb 2002 20:51:26 GMT
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com>
	 <20020130073810.382$1b@news.newsreader.com>
	 <b34d3835.0201310005.548ee6be@posting.google.com>
	 <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com>
	 <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com>
	 <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu>
	 <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com>
	 <a3h3cn$cs6$1@newshost.accu.uu.nl>
	 <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org>
Message-ID: <3C6045CC.A0A64191@optonline.net>

Well, this is fun. Perhaps the source of Python's appeal ... snakes are
surely ambidextrous if they're any kind of dextrous at all.

While I am mostly right-handed, I often use the left-hand settings on
the mouse (so as to spread the load -- and annoy folks who try to borrow
my computer). I occasionally pick up a pen to write with my left hand
before I remember that I can't write with that hand.

When playing the piano, I can easily confuse the bass and treble between
the two hands.

And I play table tennis equally well on both wings; and have been known
to execute the odd left-handed shot in tennis.

Of course, I play neither the piano nor either type of tennis
particularly well!

Fred


From p.mayers at ic.ac.uk  Sat Feb 23 15:38:09 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Sat, 23 Feb 2002 20:38:09 +0000
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk> <mailman.1014230314.2890.python-list@python.org> <c76ff6fc.0202201519.509ec73e@posting.google.com> <20020221.144033.1012502954.7349@ic.ac.uk> <c76ff6fc.0202211211.4f33f947@posting.google.com>
Message-ID: <20020223.203809.1626276121.20624@ic.ac.uk>

In article <c76ff6fc.0202211211.4f33f947 at posting.google.com>, "John
Machin" <sjmachin at lexicon.net> wrote:

> 
> BTW, how long are your own-format long numbers, what's the format, and
> where are you getting them from in the first place?

They're things like ASN.1 BitStrings in X509v3 certificates (signatures),
which can be very long indeed, Counter64s in SNMP, which can be very
close to the 64bit limit (not 63bits, as a signed long long gets you)

etc. - Basically, it's a Python DER library with the performance-critical
bits in C. There's a Flex/Bison-based ASN.1->Python compiler
and I'm using it to implement Python-native (well, nearly native)
Kerberos 5, X509, SNMP and (if I can ever get it working)
LDAP support.

Cheers,
Phil


From nas at python.ca  Wed Feb 20 17:35:11 2002
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 20 Feb 2002 14:35:11 -0800
Subject: How do you conver output of "popen('ls')" to a list?
In-Reply-To: <20020220140806.A30148@ActiveState.com>; from trentm@ActiveState.com on Wed, Feb 20, 2002 at 02:08:06PM -0800
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com>
Message-ID: <20020220143511.B2018@glacier.arctrix.com>

I'm a bit late coming in on this thread but what's wrong with:

    os.listdir('.')

It's more portable and faster.

  Neil



From jimd at vega.starshine.org  Mon Feb 25 02:09:44 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 25 Feb 2002 07:09:44 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu> <Vg%c8.4179$XT1.116344@news1.east.cox.net>
Message-ID: <a5cnvo$2mdu$1@news.idiom.com>

In article <Vg%c8.4179$XT1.116344 at news1.east.cox.net>, Jeff Hinrichs wrote:

>see:
>http://mail.python.org/pipermail/python-list/2001-August/060976.html
>for some info.   Unless you have total control over what is being eval'd you
>are at risk.
>-Jeff

>"Rajarshi Guha" <rxg218 at psu.edu> wrote in message
>news:a50udo$1dou at r02n01.cac.psu.edu...
>> On Wednesday 20 February 2002 01:24 in comp.lang.python Jeff Hinrichs
>wrote:

>>> If you wanted to get away from the dangerous eval, you could put your
>>> functions inside of a class and then,

>> Why is eval dangerous?

 In THIS case it's not a problem.  The string being eval'd is a 
 reference to be bit of code he wrote himself.  However, if it was
 being combined with a string from any untrusted source (esp. 
 user input, or data from a network connection or subprocess, even
 filenames from an os.listdir() or some such, IF that was the case
 then eval could be quite dangerous because it could be executing 
 arbitrary Python code (which could, in turn, execute arbitrary bits
 of system code).

 Regulars on this newsgroup display a knee jerk reaction to eval().
 This could be construed as a healthy wariness in some cases.  One 
 should always ask: 
	where did this data come from? 
	how did I validate it? 
   and especially, 
	what are the risks of using/trusting this data in this way?

 ... but those questions should not be limited merely to strings
 that we might be passing to eval, they must be constantly applied
 throughout our coding if we intend to write robust code which 
 works across security contexts (handles any sort of foreign data)
 and which is capable of enforcing the most basic implicit security
 policy (don't get subverted to executing arbitrary code or tricked
 into removing, or corrupting our data).




From paul at prescod.net  Mon Feb 11 18:25:45 2002
From: paul at prescod.net (Paul Prescod)
Date: Mon, 11 Feb 2002 15:25:45 -0800
Subject: Help: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254>
Message-ID: <3C6852F9.75C39170@prescod.net>

You probably need to configure your server to understand that ".py"
files are Python programs. I don't know how to do that with Xitami.

 Paul Prescod



From claird at starbase.neosoft.com  Fri Feb 15 18:48:01 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Feb 2002 17:48:01 -0600
Subject: Can Tk be connect with Copyboard of Linux
References: <a4jlhf$6uv$1@pegasus.csx.cam.ac.uk>
Message-ID: <E7074AB8E4B21A88.EDD64E91A4F24E4F.5BF62F10327BF3C4@lp.airnews.net>

In article <a4jlhf$6uv$1 at pegasus.csx.cam.ac.uk>,
Xiao-Qin Xia  <xx758 at cam.ac.uk> wrote:
>Hi, everybody,
>
>Under windows, the copyboard is shared by all the applications. So I can 
>copy some messages from different applications, and paste them into a Text 
>or Entry widgit based on Tkinter/Tk. But I found the same operation cannot 
>work under linux. Under KDE, I can not paste messages from other 
>applications into a widget based on Tkinter/Tk. Is it possible to solve 
>this problem?
			.
			.
			.
Yes it can--'least, that's what I'm told.  I don't have
KDE handy, though I'm an old hand at X.

There are a few small hurdles along the way, though.
X distinguishes the PRIMARY selection (typically <But-
tonPress-2>) from the CLIPBOARD selection (used for
explicit "Cut", "Copy", and "Paste").  You might need
a ^v for ^y (or vice-versa?).  Does that get you closer?
-- 

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


From jjl at pobox.com  Thu Feb 14 16:27:48 2002
From: jjl at pobox.com (John J. Lee)
Date: Thu, 14 Feb 2002 21:27:48 +0000
Subject: Convert String to Dictionary question
In-Reply-To: <a4h851$a8g7$1@ID-98166.news.dfncis.de>
References: <a4h851$a8g7$1@ID-98166.news.dfncis.de>
Message-ID: <Pine.SOL.4.30.0202142127220.24038-100000@mimosa.csv.warwick.ac.uk>

On Thu, 14 Feb 2002, MDK wrote:

> I saved my dictionary {'hello1': [1, 0, 0]} to a file.
>
> When I do a readline the I get back a string.
>
> How do I convert this string to a dictionary?

You probably don't want to.  Read the documentation for the pickle module.


John



From johngrayson at home.com  Sun Feb 17 09:57:05 2002
From: johngrayson at home.com (johngrayson42)
Date: Sun, 17 Feb 2002 14:57:05 -0000
Subject: [Tkinter] Bitmap Image Class
Message-ID: <a4ogc1+ua1t@eGroups.com>

Try this...

from Tkinter import *

root = Tk()

lbl = Label(root)
bmp = BitmapImage(file="bitmap.bmp")
lbl.config(image=bmp)
lbl.pack()

root.mainloop()

  ...if you want to maintain your structure...

This:

...
bmp = BitmapImage(file="bitmap.bmp")
Label(root, image=bmp).pack()
...

  ...if you're into brevity.

===================================================

Before anyone tries it and asks the question "why don't
I see an image?" DO NOT TRY ANY VARIANTS OF THIS:

Label(root, image=BitmapImage(file="bitmap.bmp")).pack()

  ... you MUST keep a reference to the bitmap around...

===================================================

      John Grayson





From vodine at direct.ca  Thu Feb 21 13:49:52 2002
From: vodine at direct.ca (Virginia O)
Date: Thu, 21 Feb 2002 10:49:52 -0800
Subject: Newbie: Crazy, but Quick
Message-ID: <u7ag65cbc0q3e@corp.supernews.com>

Hi all.

I need to check if I'm heading in the right direction.

I'm trying to write a 'protype' of a program. I'll give a quick example of
what I'm trying to do:

Say I spend a month training for Speed Skating.
I open up my Speed Skating Training Buddy Program on my computer. I just
bought the program and it should be cross-platform compatible, to work on my
PC at home and the Mac's at work.

I enter the date and time and my race times.

Then I choose the weather conditions for each day. There should already be a
DB with all the info, so I don't have to enter each time.

I would like to hit a button that calculates the correlations between
weather and my race times.
------
So far, this is what I understand I need to do:

Build my database. (I will be using Access,unless convinced otherwise.)

Learn more Python, so I can write the functions necessary to calculate
correlations. (There would be more than what I've shown here.)

Setup the Python database module, to be able to access the DB.

Then, do I want wxPython, Tkinter or the like to create an interface? I want
graphics and all the fancy stuff.....not just a buncha form fields.

Then I see py2exe helps create the executable program.
---------
So, am I way out in left field? At this point, I don't want to hire a
developer. I'll be happy if I can get the works to function in Python,
before I make it a standalone program.

Thanks all,
Virginia





From afonso at writeme.com  Thu Feb 14 09:51:01 2002
From: afonso at writeme.com (Afonso Fernandez Nogueira)
Date: Thu, 14 Feb 2002 15:51:01 +0100
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org> <slrna6kkpj.is0.tim@vegeta.ath.cx>
Message-ID: <000401c1b57b$6b01a7c0$0de2533e@fonso>

>Gustavo Cordova <gcordova at hebmex.com> graced us by uttering:
>> And also "agua" which seems female, but it's a male subject: "el agua".

>IIRC, "agua" _is_ feminine, but takes the masculine article when
>singular. Mi diccionario dice que "agua" es feminina, tambi?n.

>OTOH, la palabra "mar" es masculina _y_ feminina. Por ejemplo:
>
>   "el agua en la mar"     # Correct
>    "las aguas del mar"     # Correct

Right. "Agua" is feminine, but takes the masculine article in singular just to avoid the cacophony of the two consecutive a's.

"La _ agua" is considered difficult to say and/or ugly in Spanish, as you can't merge the sounds together as if in Catalan "L'aigua".

But this doesn't apply to every word starting with "a", only in a few cases.

almost-zero-python-knowledge-but-eager-to-share-email-address-with-spammers'ly yours
fonso




From emile at fenx.com  Thu Feb 28 14:23:00 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 11:23:00 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl> <a5lk0a$8enfq$1@ID-11957.news.dfncis.de> <a5lltf$4ra$1@animus.fel.iae.nl> <a5lvav$8614t$1@ID-11957.news.dfncis.de>
Message-ID: <a5m0co$8klqb$1@ID-11957.news.dfncis.de>

Hmmm... Gerson may have given higher marks to my prior sample, but this
code is a bit cleaner without losing any of the evilness... ;-)

class _dexedItem:
    def __init__(self, sequence, idx, idxname, idxnamespace):
        self.sequence = sequence
        self.idx = idx
        self.idxname = idxname
        self.idxnamespace = idxnamespace
    def __del__(self):
        if self.idxnamespace.has_key(self.idxname):
            self.sequence[self.idx] = self.idxnamespace[self.idxname]
        else:
            self.sequence[self.idx] = None
    def __coerce__(self, other):
        return (self.sequence[self.idx],other)

class dexing:
    def __init__(self, sequence):
        import traceback, sys
        self.sequence = sequence
        self._idxname = traceback.extract_stack()[-2][3].split('in
dexing(')[0].split()[-1]
        frame = sys._getframe(-2)
        if frame.f_locals.has_key(self._idxname):
            self._idxnamespace = frame.f_locals
        else:
            self._idxnamespace = frame.f_globals
    def __getitem__(self, idx):
        if idx < len(self.sequence):
            return _dexedItem(self.sequence, idx, self._idxname,
self._idxnamespace)
        else:
            raise IndexError

mylist = range(5)

for ii in dexing(mylist):
    ii *= 4 #* ii

print mylist

--

Emile van Sebille
emile at fenx.com

---------



From mjackson at wrc.xerox.com  Thu Feb  7 14:21:47 2002
From: mjackson at wrc.xerox.com (Mark Jackson)
Date: 7 Feb 2002 19:21:47 GMT
Subject: encapsulation, typing & other questions
References: <3C62C5C5.7FEF0B10@daylight.com>
Message-ID: <a3uk4b$q4i$2@news.wrc.xerox.com>

Ragu Bharadwaj <ragu at daylight.com> writes:

> 3. Are there plans to include method & variable encapsulation in the
> object models in python? The answer I've heard is that since it's
> possible to overcome encapsulation by long or dirty ways, it isn't worth
> it. But my comeback is that if one sticks to the rules while coding,
> this won't happen and the resulting gains are worth it.

Python's technique for "encapsulating methods and variables" should be
good enough to eliminate accidents.  They are probably easier to
overcome intentionally, but "if one sticks to the rules while coding
this won't happen."

So unless you're planning on exclusively employing programmers smart
enough to subvert Python, but not smart enough to subvert
your-B&D-language-of-choice, I don't see an issue.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Back in the Sixties we didn't have video games and the
	Internet.  All we had was drugs and naked people.
					- Scott Bateman




From lars.is-nospam at home.se  Sat Feb  2 10:36:48 2002
From: lars.is-nospam at home.se (Lars Olsson)
Date: Sat, 2 Feb 2002 16:36:48 +0100
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <jIT68.555$VU3.5043@news1.bredband.com>

As a starter, you can try this:

for i in range(12):
    res = ""
    for j in range(12):
        res = '%s %4d' % (res, (i+1)*(j+1))
    print res

"Fred" <c at tech.usd253.org> wrote in message
news:e959e7b.0202020710.417801c8 at posting.google.com...
> Hello, I am just starting out with Python. I have a very basic book
> that is not helping me much. I have done some simple C programming in
> the past and I think if I could see the below C code done in Python it
> would get me over the beginners hump. I tried to do it in Python but
> got stuck right at the start, Python did not like it when I tried to
> say: for i = 1
>
> Can anybody help here? Thanks!





From rony.steelandt at bucodi.com  Sat Feb 16 11:57:15 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 16 Feb 2002 08:57:15 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org>
Message-ID: <b34d3835.0202160857.55bf6193@posting.google.com>

You are definitifly right, i get the same error with 2.1.2 and 1.5.2
Actually i tested it with C++ , VB, Windev and it works.
It only doesn't work with Python.
I think this could be a question for our COM expert Mark Hammond
Can you keep me posted on this,i could use the control too

Thanks

Rony


"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1013487810.369.python-list at python.org>...
> I am developing a headache messing with an OCX I picked up from the
> Internet.  It seems that my forays into ActiveX never pan out; I should
> stick with Linux.
> 
> The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> have to be such a pain to find?)  I can instantiate it:
> 
> >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> >>>
> 
> (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
> 
> >>> o.AboutBox()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py", line
> 29, in AboutBox
>     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
> com_error: (-2147418113, 'Unexpected failure', None, None)
> >>>
> 
> Any method or property, same answer.
> 
> The URL for the control is:
> 
>     http://www.qwerks.com/download/3693/report.zip
> 
> It is supposed to be a general printing control, but nothing works.
> 
> Chris Gonnerman -- chris.gonnerman at newcenturycomputers.net
> http://newcenturycomputers.net


From ngps at madcap.netmemetic.com  Wed Feb 27 11:10:49 2002
From: ngps at madcap.netmemetic.com (Ng Pheng Siong)
Date: 27 Feb 2002 16:10:49 GMT
Subject: SSL Server Sockets?
References: <a5ijos$1565$1@news.idiom.com>
Message-ID: <a5j0e9$rkg$1@dahlia.singnet.com.sg>

According to Jim Dennis <jimd at vega.starshine.org>:
>  I know everyone recommends m2crypto --- but m2crypto-0.07-snap2
>  doesn't build on my Debian box.  I get errors like:
> 
> Generating wrappers for Python
> _bio.i : Line 30. Syntax error in input.
> _bio.i : Line 31. Syntax error in input.
> _bio.i : Line 32. Variable %constant multiply defined (2nd definition
> ignored).
> _bio.i : Line 32. Syntax error in input.
> _bio.i : Line 33. Variable %constant multiply defined (2nd definition
> ignored).
> _bio.i : Line 33. Syntax error in input.

What does 'swig -version' say on your box?

Only this version is supported:

    ngps at madcap:~$ swig -version
    
    SWIG Version 1.3.6u-20010725-2356
    Copyright (c) 1995-1998
    University of Utah and the Regents of the University of California
    Copyright (c) 1998-2001
    University of Chicago
    
    Compiled with CC


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



From chrisabraham at yahoo.com  Wed Feb 13 01:44:33 2002
From: chrisabraham at yahoo.com (Chris Abraham)
Date: Tue, 12 Feb 2002 22:44:33 -0800 (PST)
Subject: Zope Python User Group News
Message-ID: <20020213064433.25556.qmail@web10006.mail.yahoo.com>

PiGGies:

ZPugDC <http://www.zpug.org> Mailing List Died and Rose Again!

Well, it used to be quite a fine Mailing List. And it shall be again. First
things first, though.

Python Motivated Me!

I was so motivated by last week's Python 10 conference that I got my hands
dirty and with Jon and Ethan's help (thanks iMeme for the free hosting  of the
ZPug) I have the list up and all the old names in there.

(Yes, I am giving up being ZPug-lazy for Lent, if you really must know . . .)

ZPug.org was never down, but the list, our lifeblood, was. To get straight to
the list, check out DC-Gen  <http://lists.zpug.org/mailman/listinfo/dc-gen/>,
realising that I am in the process of building a franchising operation  wherein
all the ZPugs, the PiGs, the ZUGs, the Pugs, and the PZigs can have  either a
for real home or a referral to the official site.

Franchising Opportunities

If you need a Mailing List for your local Zope User Group, Python User Group,
or ZPug? I have one for you. Just ask. Same thing if you need web  space. It
will be a subweb off the Zpug site, but we'll learn to share the  fridge.
<chris at zdev.com>

Next Gen Zpug is a Community Wonderland

So, the next step is the ZPug Community CMF. I will be enlisting your help
(those who offered) this upcoming week.

It was wicked cool seeing all you all who were there for the Python 10
conference. Also, thank you Miss Joya for hosting the last ZPug meeting  and
Andrew for winning the Gold and speaking at the last meeting as well.

Chris Abraham
beehive North America
<http://www.beehive.de>


=====
-- 
chris abraham <chrisabraham at yahoo.com> +1 202-452-7442 tel
po box 15163 washington, dc 20003-0163 +1 703-995-0409 fax


__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com



From brueckd at tbye.com  Thu Feb 28 10:40:58 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 28 Feb 2002 07:40:58 -0800 (PST)
Subject: HTTP chunk Transfer-Encoding
In-Reply-To: <4Wpf8.13$zd.865@nreader2.kpnqwest.net>
Message-ID: <Pine.LNX.4.33.0202280723270.16013-100000@ns0.tbye.com>

On Thu, 28 Feb 2002, Francis Meyvis wrote:

> Is there some python source that sends chunked packets (RFC 2616) in
> the SimpleHTTPServer or BaseHTTPServer?
> Or do you know where to find snippet of code that can split up
> file to send them in chunk format?

Hi Francis, For curiosity's sake, why do you want to split a file and
transfer it using the chunked encoding? It's most often used when the
server side doesn't know the length of the file (dynamically generated
content), which wouldn't be the case if you already have the file and want
to split it up. It's more work for both the client and the server...

Anyway, the end of the HTTP spec has psuedo code for reading a chunked
message, and generating one is even easier: your message body is the size
of the chunk (in hex) + CRLF + the chunk data + CRLF. Repeat until you
have no more data, and write out a "0" + CRLF + CRLF. The spec lets you
write out some entity headers after the 0-length chunk (and before the
final CRLF) but these are evil and should be avoided if you can help it.

-Dave




From eppstein at ics.uci.edu  Tue Feb 26 13:31:19 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Tue, 26 Feb 2002 10:31:19 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <3C7A66A7.8060503@attbi.com> <slrn.pl.a7lbtu.3vs69hb.imbaczek@baczek.net.invalid>
Message-ID: <eppstein-803FDF.10311926022002@news.service.uci.edu>

In article <slrn.pl.a7lbtu.3vs69hb.imbaczek at baczek.net.invalid>,
 Marek "Baczek" Baczy?ski <imbaczek at poczta.fm> wrote:

> Pete Shinners wyprodukowa?[ao]?:
> 
> > Donnal Walter wrote:
> >> But what is the functional equvalent of:
> >> 
> >> for x in items:
> >>     x.f()
> > 
> > if all "items" are the same class, you could get away with this,
> > 
> > map(class_of_x.f, items)
> 
> map(type(x).f, items) solves the problem.

x looks like an unbound variable to me.  map(type(items[0]).f, items) is 
close but doesn't work for generators or mixed lists.  If you really insist 
on map, it looks like the right thing is map(lambda x: x.f(), items).
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From phr-n2002a at nightsong.com  Tue Feb 12 04:38:39 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 01:38:39 -0800
Subject: how to compare two strings?
References: <7xofivazj1.fsf@ruckus.brouhaha.com> <m3aduf839k.fsf@mira.informatik.hu-berlin.de>
Message-ID: <7xadufavkw.fsf@ruckus.brouhaha.com>

martin at v.loewis.de (Martin v. Loewis) writes:
> Can you specify "alphabetically" more precisely? Do you mean
> "lexicographically, by character ordinal value" (which is what your
> scmp code suggests)? If so, the builtin cmp() function will do;
> it compares strings just fine.

I meant exactly the same function that the sort method uses by default.
Yes it looks like cmp was what I wanted--I just didn't find it in the
obvious places in the docs (under the sort method or under string
services or string methods).  Thanks.


From mwh at python.net  Thu Feb 21 06:14:53 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 11:14:53 GMT
Subject: How do you conver output of "popen('ls')" to a list?
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com> <mailman.1014244477.1738.python-list@python.org> <3C7391F6.CB6CAE01@macrozon.net> <mailman.1014253235.29718.python-list@python.org> <slrna78of4.1jg.grante@tuxtop.visi.com>
Message-ID: <uit8rm6cj.fsf@python.net>

grante at visi.com (Grant Edwards) writes:

> Maybe I'm a bit too paranoid, but using os.system() seems to me
> to be begging for problems down the road.  If you explicitly
> want shell-command-line semantics then it's cool, but most of
> the places I see it used (including in my code) there are more
> robust solutions.

Hmm.  I find myself doing this sort of thing "by hand" (i.e. calling
pipe/fork/dup2/exec/waitpid directly) more and more at the moment
because none of the library solutions are quite flexible enough.
Yesterday I wanted to cd into a directory (without affecting the
parent), run a subprocess, pipe its output into another subprocess and
wait for the latter process to finish.

Would it be possible to design an elegant library that allowed this
much control?  Certainly, I don't think I've seen any that allow
specifying the working directory, though this wouldn't be hard.

Cheers,
M.

-- 
  In case you're not a computer person, I should probably point out
  that "Real Soon Now" is a technical term meaning "sometime before
  the heat-death of the universe, maybe".
                                     -- Scott Fahlman <sef at cs.cmu.edu>


From tanzer at swing.co.at  Mon Feb 18 12:41:11 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 18 Feb 2002 18:41:11 +0100
Subject: Performances of Pyhton programs
In-Reply-To: Your message of "Mon, 18 Feb 2002 11:18:59 EST."
             <roy-0B7BBB.11185918022002@reader2.panix.com>
Message-ID: <E16crmp-0004pW-00@tswings.swing.cluster>

Roy Smith <roy at panix.com> wrote:

> bairef2 at cti.ecp.fr (Fran?ois Baire) wrote:
> > Are there any benchmark or does anoyone know what are the performances of a
> > program written in python (in terms of CPU and RAM) compared with other
> > languages (especially C++ and compiled languages) ?
> >    I've met many articles which explain that python programs are slowler
> > because it uses the interpreter, but none says how much slower it is ! I
> > would like to know this for a rather big OO program which uses standard
> > fonctions of the language.
>
> There is little doubt that a program written in C/C++ will be faster than
> the same algorithm implemented in Python (or any of the common "scripting"
> languages such as Perl or TCL).  How much slower depends to a very large
> degree on exactly what you're doing.

True, but misleading. A Python solution might use very different
algorithms than a C/C++ solution. As the first working version of a
Python solution is achieved much earlier than a C/C++ solution, there
is a lot of time left to optimize if and where necessary.

In my experience, a Python solution might be anywhere between 100
times slower to 1000 times faster than a C/C++ solution.

YMMV.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From gimbo at ftech.net  Tue Feb 26 08:36:36 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 13:36:36 +0000
Subject: Calling baseclass method from a derivated class
In-Reply-To: <slrna7n2u6.f53.patois@ns.sicfa.org>
References: <slrna7n2u6.f53.patois@ns.sicfa.org>
Message-ID: <20020226133636.GC18498@andy.tynant.ftech.net>

On Tue, Feb 26, 2002 at 01:21:11PM +0000, Yannick Patois wrote:

> I would like to know how I can call a method defined in a base class 
> from a derivated class that overloaded it.

http://www.python.org/doc/FAQ.html#4.23

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From martin at v.loewis.de  Tue Feb 26 04:43:59 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 10:43:59 +0100
Subject: Support for resume in HTTP-server
References: <NVGe8.99$5vh.176191488@news.telia.no>
Message-ID: <m3lmdgfuhs.fsf@mira.informatik.hu-berlin.de>

"Thomas Weholt" <thomas at gatsoft.no> writes:

> Some web-servers ( and ftp-servers too ) support resume; the ability to get
> the rest of a previous partial download. How do I support this using
> BaseHTTPServer?

Make yourself familiar with RFC2068. The client includes a Range
header (14.36) in the request; the server will then reply with a 206
response (Partial Content), and typically include a Content-Range
header in the response (see 10.2.7 for other valid responses). Note
that you may also want to react on the If-Range header.

BaseHTTPServer does not offer any provisions for that protocol; you
need to implement it in your do_GET method.

Regards,
Martin


From david.abrahams at rcn.com  Tue Feb 19 09:30:35 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Tue, 19 Feb 2002 09:30:35 -0500
Subject: urllib[2]/redirection help
Message-ID: <a4tncd$7ss$1@bob.news.rcn.net>

Hi,

I'm trying to solve what should be a simple problem with Python, but I'm
afraid my lack of knowledge of www protocols is making the solution elusive.
I'm trying to move a mailing list from yahoogroups to mailman, so I want to
collect an archive of past messages. A friend of mine wrote a simple script
a few months ago which would download the yahoogroups web pages for the
messages using urllib. Unfortunately, yahoogroups added periodic redirection
to pages containing advertisements, so the old script doesn't work. The
nature of the beast is that if you visit
http://groups.yahoo.com/group/boost/message/1000 in your web browser, you'll
often end up at
http://groups.yahoo.com/group/boost/auth?done=%2Fgroup%2Fboost%2Fmessage%2F1
000 instead, a page containing an advertisement. The latter page contains a
link to "/group/boost/message/1000" which always takes you to the right
place. It looks to me as though that link needs to be in the context of the
ad page in order to work properly, because I can't figure out how to make
urllib retrieve the right one. I'm sure I'm just missing something simple.

Any help appreciated,
Dave


--
+---------------------------------------------------------------+
                  David Abrahams
      C++ Booster (http://www.boost.org)               O__  ==
      Pythonista (http://www.python.org)              c/ /'_ ==
  resume: http://users.rcn.com/abrahams/resume.html  (*) \(*) ==
          email: david.abrahams at rcn.com
+---------------------------------------------------------------+




From tdelaney at avaya.com  Mon Feb  4 20:52:37 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 5 Feb 2002 12:52:37 +1100 
Subject: question about subclassing dict
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A165@natasha.auslabs.avaya.com>

> From: Russell E. Owen 
> 
> I'm writing a subclass of dict (OrderedDict, keeps items in the order 
> entered; you can get it from http://www.astro.washington.edu/owen/ if 
> interested.).

I would prefer to call this a "ChonologicalDict" or "ChronoDict" to better
describe the order that it stores things.

"OrderedDict" implies that the objects will come out in the order that you
would get by calling sort() on a list.

Cheers.

Tim Delaney



From syring at email.com  Sun Feb  3 06:55:18 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 12:55:18 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <e63q5ukd116tkglrbp92k160bjl2jihvu5@4ax.com> <mailman.1012733826.5309.python-list@python.org>
Message-ID: <a3j8gu$18hp5g$1@ID-7529.news.dfncis.de>

"Oleg Broytmann" <phd at phd.pp.ru> schrieb
> On Sun, Feb 03, 2002 at 11:11:44AM +0100, Lawrence Oluyede wrote:
> > You're right, i use Win2k and the binary distribution..
>
>    M$ Ophice GUI is not supposed to run text-mode progrms (though some
good
> guys don't beleive in it and produced pdcurses :)

Stop it! Next time, you will come out with headlines like:
Bill Gates thrown out of the US. Goverment accuses him of terror on computer
users. Buys Argentinia.

Karl M. Syring



From tim.one at comcast.net  Sat Feb 23 17:31:29 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sat, 23 Feb 2002 17:31:29 -0500
Subject: Puzzled: Y am I ending up with extra bytes?
In-Reply-To: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4>
Message-ID: <LNBBLJKPBEHFEDALKOLCGENDNOAA.tim.one@comcast.net>

[A.Newby]
> Why is this happening? I read large chunks of data from a text file,
> according to byte locations specified on another file, and for
> some reason, this function (below), spits out a few extra bytes.
>
> Here's the code, as entered into the Python shell......
>
> ...
> 	    	chat = open('D:\cgi-bin\log.txt', 'r')

Bingo.  From the file name, you're running on Windows.  Change 'r' to 'rb'
and see what happens.

Usenet doesn't have enough space <wink> to explain what Windows does with
seeks for files opened in text mode.  ssek/tell act very strangely with text
mode files.

Note that, after opening it in binary mode instead, you're going to have
deal with Windows '\r\n' line endings yourself (indeed, the magical
transformation of \r\n to \n when you open it in text mode is the cause of
your problems (well, it's a primary cause, anyway)).

Another caution:

>    	#this opens the index file, which has precise byte locations of each
>    	#chunk of data I want to extract from the log.txt file,

You're going to have to think carefully about what "precise" means,
precisely.  If you didn't mean "precise" in the sense of the actual number
of bytes in the file *as stored on disk* (including 2 bytes for each line
end), then opening in binary mode is going to screw up too, because the
index is in fact lying about the disk file then.  The combination of "actual
byte counts" and "binary mode" works fine.

If you try to stick to text mode, you can get *parts* of it to work, by
ensuring the index contains only numbers returned directly by tell() when
the file was open in text mode.  But doing arithmetic on such numbers won't
always act the way you hope.  This isn't Python screwing you, it's how C
defines fseek and ftell in text mode:  tell doesn't necessarily return a
physical byte count for files opened in text mode, and on Windows it
actually doesn't.  Likewise for a file opened in text mode, the *only*
arguments to seek() that are guaranteed to work are

    seek(0, SEEK_SET)  # start of file
    seek(0, SEEK_END)  # end of file
    seek(0, SEEK_CUR)  # don't move at all
    seek(i, SEEK_SET)  # restore previous position, where i is exactly
                       # the value returned by a previous tell

Note too that if i and j are two values returned by a text-mode tell,
there's no guarantee that i-j (or j-i) bears any relation to the number of
physical bytes beteen the file positions they represent (and, on Windows,
they don't).  If you want to do arithmetic on tell() results, it will screw
you over and over unless you use binary mode (both when obtaining offsets
via tell(), and when using offsets with seek()).




From rony.steelandt at bucodi.com  Fri Feb  1 05:55:05 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 1 Feb 2002 02:55:05 -0800
Subject: Tkinter Text
References: <3daa1aa8.0201311053.6dd349bf@posting.google.com>
Message-ID: <b34d3835.0202010255.269e3f91@posting.google.com>

jmsun at bioeng.ucsd.edu (Jeffrey) wrote in message news:<3daa1aa8.0201311053.6dd349bf at posting.google.com>...
> Hi Everyone,
>   I have a question about the Tkinter Text Widget.  I want to make it
> so that I disable the user's ability to write text.  I know that it is
> possible to do this by doing: text_widget.configure( state =
> 'disabled' ).  But when I do this I also lose the ability to scroll
> up/down/left/right/pageUP/pageDown.  I don't want the user to lose the
> ability to use the keyboard as a scroll method.  Does anyone know how
> to solve this off the top of their head?

Here's a quick solution

1. Record a state flag in self.state couldbe enable and disable
2. In the press key or release key event of the widget test self.state
If it is disable refuse caracters



> 			
> 						Thanks,
> 							Jeff


From david.abrahams at rcn.com  Wed Feb 20 18:48:55 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Wed, 20 Feb 2002 18:48:55 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu> <a4vaqi$j8e$1@bob.news.rcn.net> <3C73E1CD.8070100@physics.ucsb.edu>
Message-ID: <a51cj9$9m1$1@bob.news.rcn.net>

"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
news:3C73E1CD.8070100 at physics.ucsb.edu...
> David Abrahams wrote:
>
> > "Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
> > news:3C7283E4.4050604 at physics.ucsb.edu...
> >
> >
> > One of the design goals for Boost.Python was to avoid the language
support
> > issues that tend to plague systems which attempt to parse C++. Parsing
C++
> > is hard; AFAIK all of the wrapping systems which try to do it break down
in
> > some areas. Boost.Python uses the builtin type inference abilities of
your
> > C++ compiler instead of parsing the C++ code separately.
>
> As I understand it, the SILOON project uses a commercial parser known as
> PDT from uoregon.  Have you, or has anyone else had any experience with
> the SILOON/PDT project?  On the face of it, my guess would be that you
> would be somewhat skeptical of their claims?

I haven't had any experience with it. I just found that site. My skepticism
just dropped by a factor of 8. If they are really using the EDG front-end as
they claim to, then I am sure the parser is excellent. I do have to wonder
who is paying for the FE. The issues of dealing with the semantics of what
they've parsed remain, but these guys at least seem to be doing the parsing
part right.

> As for the CERN Root interpereter (CINT)
> http://root.cern.ch/root/Cint.html -- which I had already heard about
> from my friends in the know in the particle physics community -- I
> would assume that many here also harbor the same skeptecism toward the
> C++ interpereter as they do toward the C++ parsing systems?

Look at the site documentation. They only claim about 85% coverage of C++.




From ol1 at v10a.com  Wed Feb 20 09:35:25 2002
From: ol1 at v10a.com (O Polite)
Date: 20 Feb 2002 15:35:25 +0100
Subject: Recursively traverse linked list -- help!
In-Reply-To: <20020220105807.B1470@ibook.distro.conectiva>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
	<mailman.1014106650.11778.python-list@python.org>
	<7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek>
	<mailman.1014126336.32560.python-list@python.org> <7x7kp9blab.fsf@ruckus.b 
	<20020220105807.B1470@ibook.distro.conectiva>
Message-ID: <1014215725.14019.0.camel@matijek>

On Wed, 2002-02-20 at 14:58, Gustavo Niemeyer wrote:
 
> That indeed would be true if you always have a handle to the insertion
> point. But 136500% of overhead in the usual case is too much for me.
> 
yep.




From alanmk at hotmail.com  Thu Feb 28 07:41:03 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 28 Feb 2002 04:41:03 -0800
Subject: How to have multi-threaded server log to one file ?
References: <97ae44ee.0202261855.15c97708@posting.google.com> <f72dac0d.0202270140.97b925f@posting.google.com> <97ae44ee.0202272013.2094b2b@posting.google.com>
Message-ID: <f72dac0d.0202280441.1078b1e7@posting.google.com>

shriek at gmx.co.uk (Stephen) wrote in message 
> I had a look at the effbot queue-example and think I understand the
> Queue class but adapting my program to share the queue instance in the
> first place is making my head spin.  Namely, how do I get the same
> Queue object into both MyHandler and MyServer in the first place ?

Stephen,

Jason Orendorff has already outlined a very similar method to the one
we use, so I won't repeat it here.

The only thing I would add is that we have a threadManager class whose
only job is to start, monitor, recreate and stop threads. This
includes the logging thread, so if the logging thread ever stops, the
threadMonitor will notice and start a replacement.

And in anticipation of people asking the question: "How do you stop
one thread from another", we do it very simply: we just send the
thread a "kill" message on its Queue, i.e. an object which tells the
thread to exit as soon as it receives the message.

Happy threading.

Alan.


From boblancaster at zxmail.com  Mon Feb 25 17:33:15 2002
From: boblancaster at zxmail.com (Bob Lancaster)
Date: 25 Feb 2002 14:33:15 -0800
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com> <slrna7l380.16u.gerhard@lilith.my-fqdn.de> <slrna7l5jn.1l8.gerhard@lilith.my-fqdn.de>
Message-ID: <b470cb58.0202251433.1728a7e@posting.google.com>

Thanks for the suggestion.  I just downloaded mxODBC, and I will see
if it works better than DCOracle2.  DCOracle2 was recently upgraded,
and it seems to work better, but still crashes at times.

-Bob

Gerhard H?ring <gerhard at bigfoot.de> wrote in message news:<slrna7l5jn.1l8.gerhard at lilith.my-fqdn.de>...
> Gerhard H?ring wrote in comp.lang.python:
> > Bob Lancaster wrote in comp.lang.python:
> >> I am in a bit of a quandry. 
> >> I am writing a python module which converts some data from various
> >> python variables, lists, etc. into SQL, which in turn is inserted to
> >> or selected from a database.
> >> 
> >> I need to support three combinations.
> >> (1) PostgreSQL and Linux, for academic customers.  No serious problems
> >> here.  In fact, I have learned why folks love PostgreSQL so much.
> >> 
> >> (2) Oracle 9i and Linux.  I haven't tried this one yet, but it seems
> >> to be at least as easy as the third combination.
> >> 
> >> (3) Oracle 9i and Windows 2000, for industrial customers.  Trouble!
> >> I have tried two approaches.  
> >>   (a) DCOracle2:  builds, connects, but then does not work properly. 
> >> I see two possible sources of the problem.
> > 
> > Or:
> > 
> > (4) Use a commercially supported DB module like mxODBC.
> 
> Oops. I meant it could be a possible solution for your Oracle
> troubles, and it does run on Linux and Windows.
> 
> Gerhard


From sholden at holdenweb.com  Mon Feb 18 07:59:00 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 18 Feb 2002 07:59:00 -0500
Subject: Newbie: Problems with File IO
References: <mailman.1013981489.9375.python-list@python.org> <Xns91B9D83F1CFC1AusYourStandingInIt@130.133.1.4>
Message-ID: <UZrc8.7203$Zp2.693727@atlpnn01.usenetserver.com>

"Flavian Hardcastle" <deathtospam43423 at altavista.com> wrote in message
news:Xns91B9D83F1CFC1AusYourStandingInIt at 130.133.1.4...
> I cracked it!
>
Congratulations! Solving a problem like this always gives you more
confidence with the next one.

> It was a lot simpler than I thought. Immensely obvious infact!
>
> In the part of the program which processes the fieldstorage, I inserted
the
> lines...
>
>
> The_Form = cgi.FieldStorage()
>
> if (The_Form.has_key("handle")):
>     user = str(The_Form["handle"].value)
> elif not (The_Form.has_key("handle")):
>     user = " "
>
A quick note here: since the first test eliminates the case where your form
has the required key, by the law of the excluded middle (something is either
true or it is not), your second test isn't needed, and you could just write

if (The_Form.has_key("handle")):
    user = str(The_Form["handle"].value)
else:
    user = " "

Some people would shorten this still further by setting the default value,
and only changing it if the key is found:

user = " "
if (The_Form.has_key("handle")):
    user = str(The_Form["handle"].value)

That's a matter of taste. Not sure what a user name of a single space
implies in your code, of course.

> So the user gets to decide their handle, and handle will be stored in  the
> string "user".
>
> Then, in the actual html form itself, I inserted ....
>
> print '<input type=text NAME="handle" SIZE="30" MAXLENGTH="30" value =
> "'+user+'">'
>
> So whatever the user puts in the  input box "handle", will come back to
> them when the script re creates the html page.
>
> Simplicity itself heh? I can't imagine why I didn't spot it before!
Perhaps
> I did, but for some strange reason I thought the script would confuse
> everyone's handles.
>
Nope. Each interaction gets its own inputs, separate from those of any other
user.

> I've already had a chat buddy test it out with me, it works ok. The only
> problem is that the posts are stored in the file "log.txt", and the script
> reproduces the file in it's entirety every chat session. The result being,
> the longer you chat, the bigger and bigger web page gets, until it's a cow
> to download.
>
> For my next feat of mental contortion, I have to figure out a way to limit
> the number of posts in the "log.txt" file that are displayed. Ideally, I
> want the user to be able to specify that amount, so that (s)he can chose
> how many old posts (s)he wishes to view.
>
Well that's something for you to work on. You're being very adventurous
working with CGI so early in your programming career, and this will mean you
have quite a bit of learning about the nature of web systems as well as
Python. I guess we'll be seeing you again ;-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From phr-n2002a at nightsong.com  Wed Feb  6 05:38:59 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 06 Feb 2002 02:38:59 -0800
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org>
Message-ID: <7xlme67v3w.fsf@ruckus.brouhaha.com>

"Orr, Steve" <sorr at rightnow.com> writes:
> Python Newbie here.
>  
> I need to migrate a web application from PHP to Python and I have a
> requirement to save state info WITHOUT using cookies where much of
> the state data MUST not be visible to the end users. The PHP
> application allows users to logon to one of several databases using
> any of a number of Oracle user/passwords. The logon info needs to be
> saved for queries on other pages in the app. With PHP it's easy to
> avoid cookies by saving this info using its session handling
> functions. Is there an EASY way to do this in Python without having
> to adopt the Mother of all frameworks?

I don't know how PHP tracks sessions without using cookies--probably
by putting a session identifier into the URL's?  You can do the same
in Python to identify sessions.  To go on from there, I have to ask:

1) What web server are you using?

2) What Python integration (cgi's? mod_python?  etc.)

3) Is there just some fixed list of database passwords that don't
   change while the application is running?

If mod_python, there may already be some PHP-like session scheme-- I
haven't checked.  There are also some simple Python template systems
that might do that.  Finally, if you're using a single-process Python
http server, you can just make a session table in the server.

If there's just a few non-changing passwords, you could number them (1
to 5, say) and include the number as part of the session ID in the URL.
The application can then decode it statelessly and look up the right
password in a table.

Finally, if you want dynamic session info using Python cgi's, you can
figure out some simple way to store the session info on the server
side, either in the file system or maybe using the mmap module.


From duncan at NOSPAMrcp.co.uk  Fri Feb 22 03:59:30 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Fri, 22 Feb 2002 08:59:30 +0000 (UTC)
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de> <Xns91BC7AB03699duncanrcpcouk@127.0.0.1> <lI9d8.1181$N7.63227@ruti.visi.com>
Message-ID: <Xns91BD5AD60A367duncanrcpcouk@127.0.0.1>

grante at visi.com (Grant Edwards) wrote in 
news:lI9d8.1181$N7.63227 at ruti.visi.com:

> In article <Xns91BC7AB03699duncanrcpcouk at 127.0.0.1>, Duncan Booth wrote:
>> gerson.kurz at t-online.de (Gerson Kurz) wrote in 
>> news:3c74c3d6.20466328 at news.isar.de:
>> 
>>> I found an easier solution:
>>> 
>>> sort = lambda x:(x.sort(),x)[1]
>>> 
>> If you are heading that way:
>> sort = lambda x:x.sort() or x
> 
> Assuming you never run into an "x" that returns a true value
> from its sort method...
> 
I was of course assuming that any sensible implementation of sort would 
always return None. :^)


-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From dmaster at synopsys.com  Mon Feb 11 12:24:37 2002
From: dmaster at synopsys.com (David Masterson)
Date: 11 Feb 2002 09:24:37 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net>
Message-ID: <uk7tkylre.fsf@synopsys.com>

>>>>> Timothy Rue writes:

> On 07-Feb-02 11:39:14 David Masterson <dmaster at synopsys.com> wrote:

>> You might be the first to make the claim that those principles can
>> be represented in "nine commands".  However, so far, your
>> definition of the "nine commands" isn't clear enough to determine
>> whether your claim is valid.

> David, the biggest hurdle I see you dealing with is your belief that
> the VIC and it's nine commands are, or make up, a language.

I didn't say that exactly -- but you're close.

> The VIC is not a language, it is an identification of unavoidable
> physics in how we use abstraction. You can change the labels
> attached to the action set all you want, but the action set is
> physics, unchangeable.

Computers deal in "language" of one sort or another.  Computers boil
everything (that they can handle!) down to a series of bits and
various processes for manipulating those bits.  Problems are
abstractly represented in one of the languages that the computer
understands so that it may deal with the problem.

If the problem of the VIC cannot be abstractly represented in a
computer language, then perhaps the VIC is not something that can be
implemented on a computer.

> So I've asked for help in creating the validation. <shrug> And I've
> gotten alot of Nos.... suggesting there is interest in staying
> ignorant or widespread (trained?) ignorance regarding how to
> advance.

Or you haven't given us anything we can deal with.  Note, if we wanted
to remain ignorant, we could simply just stop talking to you.  You
have an audience and your "10 minutes of fame", but, if you cannot
express your problem in understandable fashion, we will go on to other
things that we do understand.

That will be your fault -- not ours.

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From duncan at NOSPAMrcp.co.uk  Wed Feb 20 04:45:13 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Wed, 20 Feb 2002 09:45:13 +0000 (UTC)
Subject: I haven't seen a recursive version, yet, so here.
References: <e9e6dd1f.0202191541.30eb9f62@posting.google.com>
Message-ID: <Xns91BB634C3D2E1duncanrcpcouk@127.0.0.1>

caljason76 at yahoo.com (Jason) wrote in 
news:e9e6dd1f.0202191541.30eb9f62 at posting.google.com:

> # I am a Python novice so I don't really understand how things
> # work, yet.  So some of this may either be wrong or not the
> # Python way.  Also I have no idea why the 'if __name__ == ...'
> # construct is needed instead of just the naked main statements.

There are two ways to run the code you have written. The way you probably 
expect is that someone just runs the script, but they might decide that 
some of your classes or functions are useful in another program. If so, 
they could import your code as a module.
If your code is imported, then the lines protected by 'if __name__==...' 
will not be executed. If you just put these lines straight into the file 
then anyone trying to import it will get the whole script running instead 
of being able to pull out a few classes or functions.

So the answer to you question is that this is purely a convention, you dont 
have to protect the main code in a script, but if you follow the convention 
you will find you dont have to keep rewriting the same functions as you can 
reuse the originals.

> # Also, can somebody help me find the documentation to the builtin
> # procedures, like print().  I checked the __builting__ module docs
> # but there is nothing there.  Thanks
> #
1. print is a statement, not a builtin procedure.
2. Run Python in interactive mode, type "help()" (without the quote marks) 
and you will get a prompt that looks like "help>" type "print" at this 
prompt and you will get the documentation for the print statement.
3. Alternatively look in the Language Reference section 6.6



-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From zombor_666 at hotmail.com  Mon Feb  4 23:51:51 2002
From: zombor_666 at hotmail.com (Jeremy Bush)
Date: Mon, 4 Feb 2002 22:51:51 -0600
Subject: heres a dumb question....
References: <mailman.1012873689.8841.python-list@python.org>
Message-ID: <a3nof2$14gm$1@wiscnews.wiscnet.net>

"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1012873689.8841.python-list at python.org...
> Jeremy Bush writes:
> > [...] cant seem to figure out how to get keyboard input...
> > ive read the whole readme that came with it, the tutorial,
> > everything.
>
> raw_input() gets a line of input.
>
> Unlike C++, you ordinarily get a line at a time, and split it
> to get the information you want.
>
> For example, if you expect the user to type in numbers
> separated by commas, you'd do:
>
> line = raw_input('> ')   # see? you can provide a prompt
> fields = line.split(',')
> for s in fields:
>     x = float(s)  # convert string to float
>     print "you entered the number:", x
>
> If you "import sys", then you have sys.stdin, which is a file object.
> File objects have various methods for input.  Viz:
>   http://www.python.org/doc/current/lib/bltin-file-objects.html
>
> Enjoy Python!
>
> ## Jason Orendorff    http://www.jorendorff.com/
>
>

Thanks, everyone




From logiplexsoftware at earthlink.net  Mon Feb 11 13:28:44 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 11 Feb 2002 10:28:44 -0800
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <B88AE186.48CD%jonathan@onegoodidea.com>
References: <mailman.1013196740.22073.python-list@python.org>
	<a41oi1$a22$1@panix2.panix.com>
	<mailman.1013216835.14351.python-list@python.org>
	<B88AE186.48CD%jonathan@onegoodidea.com>
Message-ID: <20020211102844.7edc9f05.logiplexsoftware@earthlink.net>

On Sat, 09 Feb 2002 14:24:06 +0000
Jonathan Hogg wrote:

> On 9/2/2002 0:56, in article
> mailman.1013216835.14351.python-list at python.org, "Cliff Wells"
> <logiplexsoftware at earthlink.net> wrote:
> 
> > Aahz Maruch wrote:
> > 
> >> I have to admit that I'm seriously thinking about an iMac, though.
> > 
> > Yikes.  That's an expensive Linux box ;)
> 
> Hah! all smileys aside, you'd have to be mad to run Linux on it when it
> comes with Mac OS X and all of Apple's free iApps. I get to run
commercial
> applications including the Adobe and Microsoft range, it's got full Java
> support, runs all the UNIX apps I need including Apache, Python and
> PostgreSQL, and is fun to use.

Can't argue with that.  I suppose if I were to shell out for a mac I might
use OSX.  I was only joking anyway.

> Stuff cygwin, there's no comparison when you have a full BSD under the
hood.
> And when Linux runs Photoshop, MS Office, IE, and Palm Desktop, I'll
> consider using it as my main work OS.

Yeah, I've never been too happy with cygwin.  That's what you get when you
try to get GNU under Win32.  The opposite approach is much better. Running
win2k under VMware 3.0 is flawless, so I can in fact use all those
applications (and more).  It was a little sluggish on my PIII 800 with
128MB, but on my new Athlon XP 1700+ with 256MB you can't tell it's running
in a VM.  _Very_ snappy.  Of course, the main point is that my system cost
< $1000 (including a 17" Trinitron monitor).  Do that with a Mac (all
smileys aside).

The only Apple product I've seen in many years to make me a bit jealous is
the iPod.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From brueckd at tbye.com  Tue Feb 19 17:08:05 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 19 Feb 2002 14:08:05 -0800 (PST)
Subject: Adding static typing to Python 
In-Reply-To: <20020219215146.CF6548046A@regurgitate.ugcs.caltech.edu>
Message-ID: <Pine.LNX.4.33.0202191404550.6214-100000@ns0.tbye.com>

On Tue, 19 Feb 2002, Quinn Dunkan wrote:

> > An even better approach is to ask yourself: if it's a person's ID, then
> > why isn't the variable named personID?
>
> I didn't say that because the OP could have objected that "then you're just
> encoding the type information in the variable name".

No, that would be encoding more information about what it is - a person
object vs. a person's ID. It still might be a string, an integer, etc.
Encoding the type info in the variable name would be something along the
Hungarianish lines of "iPersonID".

-Dave




From alain at onesite.org  Tue Feb 26 19:03:30 2002
From: alain at onesite.org (Alain Tesio)
Date: Wed, 27 Feb 2002 01:03:30 +0100
Subject: Problems with select
References: <pan.2002.02.26.04.17.29.538545.11917@onesite.org> <slrna7nfkh.2bd.rmunn@rmunn.dyndns.org>
Message-ID: <pan.2002.02.27.01.03.24.922015.11917@onesite.org>

On Tue, 26 Feb 2002 18:01:30 +0100, Robin Munn wrote:

> On Tue, 26 Feb 2002 04:17:30 +0100, Alain Tesio <alain at onesite.org>
> wrote:
>>Hi, I can't manage to make select work, in both 1.52 and 2.1.2
>>
>>I just want to read stderr and stdout when executing a command with
>>popen3, and I must use select, read() blocks when there is more than a
>>fixed amount of data (4 KB I think)
>>
>>The problems is that select still gives something after the end of the
>>file, or sometimes never returns anything at all.
>>
>>I create the pipes with :
>>
>>(pipe_out,pipe_in,pipe_err)=popen2.popen3(command)
>>
>>and use select like this:
>>
>>pollstatus=select.select([],[pipe_out],[pipe_err],1)
>>
>>
>>(the doc says : select(iwtd, owtd, ewtd[, timeout]) I didn't put pipe_in
>>as it's closed)
> 
> I don't think you've understood the documentation completely. The three
> lists you pass to select() are lists of objects for input, output, and
> "exceptional conditions" (which are things like out-of-band data in
> TCP/IP sockets). Here, "input" and "output" means from the point of view
> of your program: if you want to be informed when a pipe has data waiting
> for you to read, you put it in the input list.
> 
> So what I think you should be doing is:
> 
> pollstatus=select.select([pipe_out, pipe_err], [], [], 1)
> 
> Hope this helps.
> 

Hi, thanks for the explanation, but select still returns files where
there is nothing to read, after the end of the stream.

Alain


#!/usr/bin/python

import sys,select,popen2,string

def execute(command,maxiter):
	(pipe_out,pipe_in,pipe_err)=popen2.popen3(command)
	pipe_in.close()
	pipes=[pipe_out,pipe_err]
	gotsomething=1
	i=0
	while 1:
		i=i+1
		if i>maxiter:
			print "maxiter reached"
			sys.exit(0)
		pollstatus=select.select([pipe_out,pipe_err],[],[])
		print "%i select returns : " % i,pollstatus
		if not(pollstatus[0]):
			return
		for file in pollstatus[0]:
			if file.fileno()==pipe_out.fileno():
				label="out"
			elif file.fileno()==pipe_err.fileno():
				label="err"
			else:
				raise ValueError
			print "something on %s : '%s'" % (label,file.readline())

execute("cat 3lines",10)



00:59:25 alain ~/tmp $./test_select.py
1 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : 'first
'
something on err : ''
2 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : 'second
'
something on err : ''
3 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : 'third
'
something on err : ''
4 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
5 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
6 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
7 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
8 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
9 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
10 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
something on out : ''
something on err : ''
maxiter reached


From max at alcyone.com  Sun Feb  3 16:38:00 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 03 Feb 2002 13:38:00 -0800
Subject: pyicqlib/icqlib crash?
Message-ID: <3C5DADB8.86675CA4@alcyone.com>

I'm seeing the following crash with icqlib 1.0.0 and pyicqlib 0.6.0:

max at oxygen:~/incoming/pyicqlib-0.6.0% python
Python 2.1.1 (#6, Sep  2 2001, 10:26:49) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import icqlib
>>> icqlib.ICQLink()
zsh: segmentation fault (core dumped)  python

gdb shows:

	...
#0  0x403adc99 in list_free () from /usr/local/lib/libicq.so.1
(gdb) bt
#0  0x403adc99 in list_free () from /usr/local/lib/libicq.so.1
#1  0x403adc75 in list_delete () from /usr/local/lib/libicq.so.1
#2  0x403aef91 in icq_TCPDone () from /usr/local/lib/libicq.so.1
#3  0x403acc32 in icq_Done () from /usr/local/lib/libicq.so.1
#4  0x4039d5cd in icqlib_DelPyICQLink (self=0x80eca88) at
icqlibmodule.c:426
#5  0x403c3dcf in CCL_call (self=0x403a5e40, arg=0x80e6184, kw=0x0)
    at ExtensionClass.c:2081
#6  0x805c8f5 in call_object (func=0x403a5e40, arg=0x80e6184, kw=0x0)
    at Python/ceval.c:2813
#7  0x805cf5d in do_call (func=0x403a5e40, pp_stack=0xbfffeb18, na=0,
nk=0)
    at Python/ceval.c:3123
#8  0x805b556 in eval_code2 (co=0x812bd08, globals=0x80e64dc, 
    locals=0x80e64dc, args=0x0, argcount=0, kws=0x0, kwcount=0,
defs=0x0, 
    defcount=0, closure=0x0) at Python/ceval.c:1975
#9  0x805898b in PyEval_EvalCode (co=0x812bd08, globals=0x80e64dc, 
    locals=0x80e64dc) at Python/ceval.c:341
#10 0x806e2c0 in run_node (n=0x8126748, filename=0x80aeaed "<stdin>", 
    globals=0x80e64dc, locals=0x80e64dc, flags=0xbfffec44)
    at Python/pythonrun.c:1045
#11 0x806d649 in PyRun_InteractiveOneFlags (fp=0x40392080, 
    filename=0x80aeaed "<stdin>", flags=0xbfffec44) at
Python/pythonrun.c:570
#12 0x806d4df in PyRun_InteractiveLoopFlags (fp=0x40392080, 
    filename=0x80aeaed "<stdin>", flags=0xbfffec44) at
Python/pythonrun.c:510
#13 0x806d3f6 in PyRun_AnyFileExFlags (fp=0x40392080, 
---Type <return> to continue, or q <return> to quit---q

Any ideas?  I'm new to icqlib and pyicqlib.  The pyicqlib seems to
suggest that pyicqlib has been integrated into the icqlib release branch
but I couldn't find it in the files available for download.  Perhaps
there's a later version that I'm missing?

Thanks!

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From j.spies at hccnet.nl  Sun Feb 17 10:10:00 2002
From: j.spies at hccnet.nl (Jaap Spies)
Date: Sun, 17 Feb 2002 15:10:00 +0000
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net>
Message-ID: <3C6FC7C8.7060409@hccnet.nl>

I.J. wrote:

> Is there a module for python that grants ussage of decimal numbers with more
> than 14 decimal places, including all scientific functions to do with those
> numbers?
> 
> 
> 

Take a look at
	http://cens.ioc.ee/projects/pyginac/

Jaap Spies



From steve at lurking.demon.co.uk  Mon Feb  4 10:29:18 2002
From: steve at lurking.demon.co.uk (Steve Horne)
Date: Mon, 04 Feb 2002 15:29:18 +0000
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com> <mailman.1012829227.9110.python-list@python.org>
Message-ID: <g39t5us89a1um3c24ord469nq404vpmh5q@4ax.com>

On Mon, 4 Feb 2002 07:25:00 -0600, "Chris Gonnerman"
<chris.gonnerman at newcenturycomputers.net> wrote:

>----- Original Message ----- 
>From: "Steve Horne" <steve at lurking.demon.co.uk>
>
>[[ hack hack hack ]]
>
>> One major advantage is the possibility that source code could be more
>> concerned with semantics, and the presentation could be an issue for
>> the editor. No more arguments about indentation against braces - tell
>> the dissenters to simply configure their editor to work with braces
>> instead of (or as well as) indentation. With a sufficiently
>> sophisticated editor, the syntax of the language becomes 'skinnable'.
>
>Heaven help the poor fool who has to work at someone else's station or 
>on someone else's account...  would be MUCH worse than just a difference
>in coding style.

Nah - just keep everyones editor configuration files in a shared area.

The whole point is that you always see the code formatted in your
preferred style, irrespective of who wrote it - the stylistic
differences are separated from the source code, so that when you load
up someone elses source code into your editor, it is automatically
formatted the way you prefer it.

You can play any MP3 file irrespective of the skin you set up for the
player. The music is independent of the user interface. There are also
skinnable music editors. Just because the waveform display is a
different colour, or uses a spectral format, or is scaled differently
or whatever, that doesn't affect what music you can edit or what it
will sound like.

So who cares whether the author uses braces for blocks - when you load
the file, it will use indentation the same as you've always used. For
that matter, who cares whether the author wrote "a b +" or "(+ a b)"
or "(add a b)" or "please add a and b" or whatever - as long as you
see it as "a + b" that's fine.

-- 
Steve Horne
steve at lurking.demon.co.uk


From fredrik at pythonware.com  Sat Feb 16 11:49:05 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 16 Feb 2002 16:49:05 GMT
Subject: Convert String to Dictionary question
References: <mailman.1013870847.12516.python-list@python.org>
Message-ID: <54wb8.19093$n4.3345145@newsc.telia.net>

Jason Orendorff wrote:
> Whereas there are no known security holes in pickle.

careful.

it's fairly trivial to construct a pickle string that calls eval
or os.system with arbitrary arguments.

</F>




From kennedy.bill at freeuk.com  Wed Feb 27 06:07:39 2002
From: kennedy.bill at freeuk.com (William Kennedy)
Date: 27 Feb 2002 11:07:39 GMT
Subject: ANN: RNG_Lotto / winLotto
Message-ID: <Xns91C271F019E47kennedybillfreeukcom@195.92.193.157>

I have updated these programs so that they now use the random module rather 
than the whRandom version.

These programs are FREEWARE UK Lottery number generators, please feel free 
to use them. I would be grateful for any comments / bug reports.

http://www.kennedy.bill.freeuk.com/lotto.html
http://www.kennedy.bill.freeuk.com/winlotto.html


From skywalkerpackage at hotmail.com  Fri Feb  1 08:20:26 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Fri, 01 Feb 2002 13:20:26 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
Message-ID: <3c5a84eb.70228013@news.online.de>

Hello all,

In Tkinter the widgets are created in the form:

wgt = Widget(master, option_tag1 = option1,  option_tag2 =
option2,...)

I want to create widgets dynamically. Therefore I have created a
dictionary for the options (well, it is a list of lists of lists which
have this dictionary included, but I will keep it simple here). The
problem is that not only the options are unknown at compile time, but
also the *option tags*. For example in Button(root, text='Hello') also
the *text* option tag should be taken from the keys of the options
dictionary. 

This is hardcoded:

from Tkinter import *
root = Tk()
bt = Button(root, text='Hello') 
bt.pack(side=TOP)
root.mainloop()


When I change it to something like

...
dictOptions = {text: 'Hello',....}
bt.config(?)
...

,Python complains that there is no defined symbol text. Also, I
wouldn't know what to write in bt.config(....)

On the other hand, I have tried to write

dictOptions={'text':'Hello'} and compose the string "text='Hello'" on
the fly, but then I built a button with the text "text='Hello'".


So, how can I change the options and the option tags dynamically?


Any help would be appreciated.


ByeBye

Johannes


From usenet at thinkspot.net  Sat Feb  9 01:17:59 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 08 Feb 2002 22:17:59 -0800
Subject: email module Bug in 2.2
References: <a4164b.3vv7l9p.1@kserver.org> <slrna69eks.c6s.quinn@vomit.ugcs.caltech.edu>
Message-ID: <a41iqo.3vshntl.1@kserver.org>

On 9 Feb 2002 05:58:53 GMT, quinn at vomit.ugcs.caltech.edu (Quinn Dunkan)
wrote in comp.lang.python in article
<slrna69eks.c6s.quinn at vomit.ugcs.caltech.edu>:

> From the libref:
> is_multipart()
> Return 1 if the message's payload is a list of sub-Message objects, otherwise
> return 0. When is_multipart() returns 0, the payload should either be a string
> object, or a single Message instance.
> 
> So notice that is_multipart() does not check if get_main_type() is 'multipart'.
> I don't know much about the MIME hairball, but the msg appears to claim it is
> multipart, but doesn't actually have multiple parts.  Message.add_payload only
> sets the payload to a list if it's called more than once, not if the msg calls
> itself 'multipart'.  I'm not sure if this is correct or not, though I'm
> inclined to agree with you that it's not (section 12.2.2.2 implies that you
> should only get a non-string, non-list payload for some message/* types), but
> is_multipart() is behaving strictly according to documentation.

OK, thanks for pointing this out. I'm sure I read that documentation
before, but right now I'm so steeped in documentation (learning the new
email module...which in general I think is fantastic, and the re module
while working with RFCs and other people on a project...I'm a bit
overwhelmend).

...<useful code sample snipped>...

I'm going to have to restrain myself a bit more from tossing the word
"bug" around in the subject headings. I hope people don't take offense
too quickly.

Thanks,

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_





From chris.gonnerman at newcenturycomputers.net  Sat Feb  9 20:56:06 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sat, 9 Feb 2002 19:56:06 -0600
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4>
Message-ID: <00eb01c1b1d6$1c7a9d80$0101010a@local>

Let's go over this step-by-step.  First, what web server did you download?
Most docs assume Apache.

You have a form (stored in a .html file or produced by a cgi script, doesn't
matter) which contains something like this:

    <form method='POST' action='http://url.to.the/cgi-bin/script'>

"script" must be a script file in the cgi-bin directory.  You probably can
shortcut it like this:

    <form method='POST' action='/cgi-bin/script'>

and this is probably better.

The 11.2 docs then describe how you decode the data sent by the form.

If you are still having problems, please post code (and tell us what
webserver software you are using).







From kalle at gnupung.net  Mon Feb 18 20:20:34 2002
From: kalle at gnupung.net (Kalle Svensson)
Date: Tue, 19 Feb 2002 02:20:34 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: <20020218204121.GA9316@lilith.hqd-internal>
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <20020218204121.GA9316@lilith.hqd-internal>
Message-ID: <20020219012034.GD27474@sandra.lysator.liu.se>

[Gerhard H?ring]
> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> > [Aahz]
> > > I think the occasional non-English post simply reflects the
> > > international nature of Python.
> > 
> > [Fran?ois Pinard]
> > > Exactement.  Bravo!  Bien dit! :-)
> > 
> > Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> > czignat?  Classifero en bragna!
> 
> Woas da Deifi wos des hoasn mog.

Beh! Yuoo shuoold ell be-a useeng Svedeesh, zee oonly lungooege-a
recugneezed by a veede-a mejureety ooff users es "Zee Next
Gruoondbreekeeng Refulooshun in Inhuncement ooff Sherehulder
Felooe-a"! Bork Bork Bork!

Peace,
  Kalle
-- 
Kalle Svensson (kalle at gnupung.net) - Laziness, impatience, hubris: Pick two!
English: http://www.gnupung.net/  Svenska: http://www.lysator.liu.se/~kalle/
Stuff: ["http://www.%s.org/" % x for x in "gnu debian python emacs".split()]



From jason at jorendorff.com  Thu Feb 14 18:42:03 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 17:42:03 -0600
Subject: help!: optimization of search problem (example code)
In-Reply-To: <HFEKILOLEFEFMKAECNDLCEJPDCAA.jason@jorendorff.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEKNDCAA.jason@jorendorff.com>

I wrote:
> More to come.

Never mind.  SciPy kills whatever I was going to do.  :)

## Jason Orendorff    http://www.jorendorff.com/



From jkraska at san.rr.com  Tue Feb 26 02:22:54 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 07:22:54 GMT
Subject: [Python-Help] advocacy help ( Python vs VB )
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01> <mailman.1014706963.7898.python-list@python.org>
Message-ID: <9cem7uc521ni47v2mo0qiv1dfougdciqp6@4ax.com>

While this is academic, things like this are going to come up again
in the future. What would really sell them is a Python which compiled
to CLR and gave them full access to the suite of COM/.NET objects.
Many "VB shops" require this. He should take a long gander at win32all
and make sure that it fits their need.

C//



From shalehperry at attbi.com  Thu Feb 28 11:45:32 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Thu, 28 Feb 2002 08:45:32 -0800 (PST)
Subject: Can use python to play MP3 file?
In-Reply-To: <a5llkf$8a75f$1@ID-12869.news.dfncis.de>
Message-ID: <XFMail.20020228084532.shalehperry@attbi.com>

On 28-Feb-2002 Ricky wrote:
> I hear that PHP can online play MP3 file. Is there any method that can make
> me play MP3 by using Python CGI?
> 
> Thanks for any reply!
> 

It does not play them, it sends them over the network so that a program on a
user's machine may play them.  Most decent mp3 players allow you to connect to
a streaming server (as they are called) and play the songs as they are
transferred.  Gives an affect like listening to the radio.



From wesc at deirdre.org  Wed Feb 13 01:24:35 2002
From: wesc at deirdre.org (Wesley Chun)
Date: Tue, 12 Feb 2002 22:24:35 -0800 (PST)
Subject: ANN: SV-SF Bay Area Python users group (BayPIGgies) mtg 2/13 7:30pm
Message-ID: <Pine.LNX.4.31.0202122223260.29997-100000@emperor.deirdre.org>

hi folks,

a friendly reminder for those in the Silicon Valley or SF Bay Area
that our monthly Python users group meeting meets on the 2nd wednesday
of every month, and that means tonite!!  more info in the listing below.
hope to see some of you there!

-wesley


Date: February 13, 2002
Time: 7:30pm - 9pm
Location: Stanford University
Agenda: Python 10 Conference Summary
Speaker: Todd Valentic, SRI International

Many of us were not be able to make it to the Python10 Conference (Feb
4-7 2002). Todd, who presented a paper at the conference, will give us
the lowdown on what happened.

More information including directions at http://deirdre.org/baypiggies

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

"Core Python Programming", Prentice Hall PTR, ? 2001
    http://starship.python.net/crew/wesc/cpp/

    wesley.j.chun :: wesc at deirdre.org
    cyberweb.consulting : henderson, nv : cyberweb at rocketmail.com
    http://www.roadkill.com/~wesc/cyberweb/




From fredrik at pythonware.com  Mon Feb  4 10:56:15 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 04 Feb 2002 15:56:15 GMT
Subject: I love obfuscated python!
References: <mailman.1012809006.11061.python-list@python.org>
Message-ID: <zay78.23173$l93.4207854@newsb.telia.net>

"Jesse W" wrote:
> Horrey!  I have finally solved the piece of obfuscated python
> posted by Greg Stein in November of 1998.*

note that the solution was posted in october 1998.

    http://groups.google.com/groups?th=75fbb17530a0fae1

guess someone was using not only the time machine, but
also some kind of yet-to-be-seen autodecoding tool...

psuedo-programming-ly y'rs /F




From mdk at mdk.com  Tue Feb  5 11:29:14 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 11:29:14 -0500
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de>
Message-ID: <a3p190$19v1rc$1@ID-98166.news.dfncis.de>

"Emile van Sebille" <emile at fenx.com> wrote in message
news:a3p0ss$19ob7s$1 at ID-11957.news.dfncis.de...
>
> "MDK" <mdk at mdk.com> wrote in message
> news:a3p01b$186eq8$1 at ID-98166.news.dfncis.de...
> > Hello,
> >
> > I am trying to create my own special type of list object.
> >
> > However, when I do x.append('yo') it uses Python's append instead of
> the one
> > from my class.
> >
> > How can I get it to use my append method?  I have tried def __append__
> and
> > then called my own class' function but that did not work.  I've played
> with
> > __getattr__ but that did not help.
> >
>
> Call the method append and not __append__?
>
>
> --
>
> Emile van Sebille
> emile at fenx.com

Yes, I figured that to be the case because it should be transparant to the
caller.  However, that is not the solution.

Thanks though.




From syring at email.com  Sat Feb 23 10:53:16 2002
From: syring at email.com (Karl M. Syring)
Date: Sat, 23 Feb 2002 16:53:16 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net>   <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de> <7x7kp7krjl.fsf@ruckus.brouhaha.com> <a52ml7$42r9l$1@ID-7529.news.dfncis.de> <7x8z9n6nwl.fsf@ruckus.brouhaha.com> <a52os1$45k2v$2@ID-7529.news.dfncis.de> <j2935a.0b.ln@127.0.0.1>
Message-ID: <a58e3p$5ctts$1@ID-7529.news.dfncis.de>

"Carl Banks" <idot at vt.edu> schrieb im Newsbeitrag
news:j2935a.0b.ln at 127.0.0.1...
> Karl M. Syring wrote:
> > But you really should to it. I have floating point code that was running
for
> > a long time without a hitch and then the horrible happened. The ugly
thing
> > is, the code often will not crash but produce wrong results that seem to
be
> > plausible. If you don't have any results to check the output, the worst
> > nightmares might come true.
>
> I'm sorry I answered this troll.
>
> I he weren't a troll, I might ask him to try updating his rational
> number 100 times per second with a differential equation and see what
> happens to the denominator.  But he is a troll, so it wouldn't matter.

You obviously are a clueless!
This fork of the thread was not referencing to problems in computational
geometry, but to insufficient knowledge of numerical analysis in general. I
have learnt my lesson, you obviosly have not. Perhaps a visit to Kahan's
site (http://www.cs.berkeley.edu/~wkahan/Math128/) will help you. Especially
the section "Plausible but spurious solutions due to roundoff" is quite an
eye opener.

Karl M. Syring





From akuchlin at ute.mems-exchange.org  Fri Feb  8 14:29:00 2002
From: akuchlin at ute.mems-exchange.org (A.M. Kuchling)
Date: 08 Feb 2002 19:29:00 GMT
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <slrna689o0.cn1.akuchlin@ute.mems-exchange.org>

In article <mailman.1013181915.301.python-list at python.org>, 
	Skip Montanaro wrote:
> access to PPT sort of out in the cold.  What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT slides
> to other, more accessible, formats?  If we can get a reasonable (but fairly

OpenOffice 6.0beta seems to work fairly well for me; it's even managed
to read files that PowerPoint wouldn't open.  You can also unzip
OpenOffice-format files to get a collection of XML, PNG, and JPG
files, which was convenient for borrowing MEMS graphics from my boss's
presentation.

One visible trend at this year's conference was HTML + a Web browser
as a presentation tool.  Skip, Jeremy and I all used HTML, and I think
other people did too.  It's a nice trend to see, and Opera's
full-screen mode can make the resulting presentation quite attractive.

--amk                                                  (www.amk.ca)
In London there is a man who screams when the church bells ring.
    -- H.P. Lovecraft, "The Descendant"


From anthony_barker at hotmail.com  Tue Feb 12 11:18:51 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 12 Feb 2002 08:18:51 -0800
Subject: proxy authentication: how to hide password
References: <9ff6b66.0201250109.1c68a130@posting.google.com>
Message-ID: <899f842.0202120818.11412d4b@posting.google.com>

Hi:

> I'm using Python 2.2 on Win2k. We have a firewall in our organization,
> so I need to connect to the Internet via proxy from my .py scripts.
> 
> So I use the following lines:
> 
> proxy=urllib2.ProxyHandler({'http':'http://login:passwd at 10.10.10.11:80'})
> opener=urllib2.build_opener(proxy)
> urllib2.install_opener(opener)
>  
> f = urllib2.urlopen(url-to-retrieve)
> 
> where login and passwd are my Windows login and password.

I am in a similar position, but luckily my password isn't in sync with
my windows password.

I tried sockscap without much success. But you will have to add your
password to sockscap anyways.

But don't feel bad - Perl has a similar issue - but uses an
environment variable instead. I also use wget - this uses the .wgetrc
(yet another location for the password).


> Everything works OK, but obviously there are some problems:
> - my password is stored in my script as a simple text;
use a local environment variable gets around this one (like perl).
>>> import os
>>> user = os.environ['USERNAME']
>>> password = os.environ['HTTP_proxy_pass']


> - I need to change my scripts regularly to correct the password;
> - my colleagues can't use the same script - they should change the
> login and password their.

prompt for the password if need be.


From grante at visi.com  Wed Feb 20 21:56:18 2002
From: grante at visi.com (Grant Edwards)
Date: Thu, 21 Feb 2002 02:56:18 GMT
Subject: How do you conver output of "popen('ls')" to a list?
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com> <mailman.1014244477.1738.python-list@python.org> <3C7391F6.CB6CAE01@macrozon.net> <mailman.1014253235.29718.python-list@python.org>
Message-ID: <slrna78of4.1jg.grante@tuxtop.visi.com>

In article <mailman.1014253235.29718.python-list at python.org>, Neil Schemenauer wrote:
> husam wrote:

>> os.listdir('.') gets you all the files in the '.' dircetory. This want
>> work if i want to process some selected files. In contrast to this,
>> "os.popen() allows you to pass arguments like "ls *txt" to retrieve only
>> *.txt files.
> 
> Sorry, that's still not a good argument for two forks and god knows how
> any system calls.  Try this:
> 
>     import glob
>     txt_files = glob.glob("*txt")

In these days of 1.whatever GHz processors, I doubt that a
couple fork/exec combinations matters a lot in most situations.
But, os.system() calls the shell. The shell does god-knows-what
to the arguments, then looks god-knows-where for the
executable.  You can lessen these problems by specifying an
absolute path and quoting the arguments -- though quoting them
properly isn't a trivial problem.

Don't even ask what happens when the wrong shell gets chosen to
start with.  :)

Maybe I'm a bit too paranoid, but using os.system() seems to me
to be begging for problems down the road.  If you explicitly
want shell-command-line semantics then it's cool, but most of
the places I see it used (including in my code) there are more
robust solutions.

Modules to look in when you're tempted to do an os.system():

   shutil
   os.path
   glob
   os
   commands   

If all else fails, os.spawn or os.popen2 are often better
choices than os.system.

I'm done preaching now...

-- 
Grant Edwards                   grante             Yow!  Gibble, Gobble, we
                                  at               ACCEPT YOU...
                               visi.com            


From ecastro at cicei.ulpgc.es  Thu Feb  7 05:12:16 2002
From: ecastro at cicei.ulpgc.es (Enrique)
Date: 7 Feb 2002 10:12:16 GMT
Subject: Delphi vs Python
References: <mailman.1013063891.28572.python-list@python.org>
Message-ID: <3C625389.FA470395@cicei.ulpgc.es>

Sean 'Shaleh' Perry wrote:
> 
> On 07-Feb-2002 Peter Hansen wrote:
> > Bo Vandenberg wrote:
> >>
> >> I'm planning learning strategy and I thought I would ask people who use both
> >> Python and Delphi their opinion.
> >
> > I used Delphi (back in the 3 and 4 days, I think) extensively for several
> > years, and had a lot of respect for the environment and language.  Very
> > productive for me, relatively few bugs (compared to C++ and C, roughly
> > similar to my levels with Java).
> >
> 
> This is mostly a "me too" post.  With the addition of attributes in 2.2 python
> now does almost everything syntactically Delphi does.  I do on occasion miss
> the 'with' keyword, nice bit of sugar that.
> 
> For me what made Delphi nice was how much it abstracted the Window's API away
> and the ease of creating GUI apps by putting things where I wanted.  I have yet
> to find another environment that made good looking apps that fast.  If I could
> have one of those that outputs python code AND was not proprietary I would be
> quite happy.  glade is a very pale comparison.
> 

  Perhaps you should have a look at Boa Constructor. 
	(http://sourceforge.net/projects/boa-constructor/)
It is a RAD IDE and a GUI-builder based on
wxWindows/wxPython.
The declared phylosophy of the author is to get Delphi
working manners (drag-and-drop frame construction etc) into
Python applications.   


> If your goal is to write GUI apps, Delphi may be a good place.  Python is great
> as an all around language and it really can churn out code fast.  On top of
> that you have the interpreter available to you by just running python without a
> script AND it is available to you in your running code (introspection).  Being
> able to quickly test an idea by just firing up python is one of the things I
> love most.  Hell, I even use it as a desk calculator (whoever made '_' return
> the last result -- I love you).

I find the combination Python/wxWindows(wxPython) very
convenient for writing GUIs, and as powerful or more than
Delphi. A progran is not its GUI, and having a good,
powerful, logic behind GUI is essential. Think in how easy
is to manage database connections, or Numeric arrays , or
image manipulation (PIL), with Python.

Enrique


From gaul at spam.utexas.edu  Tue Feb 26 03:42:37 2002
From: gaul at spam.utexas.edu (Andrew Gaul)
Date: Tue, 26 Feb 2002 08:42:37 +0000 (UTC)
Subject: Anagram
References: <3c4e8efc.1036790726@news.mch.sni.de>
Message-ID: <slrna7mijs.lfm.gaul@bifrost.cs.utexas.edu>

In article <3c4e8efc.1036790726 at news.mch.sni.de>, Nikolai Kirsebom wrote:
> Friend of mine who used to work with lisp (and is a bit interested in
> Python) asked me how compact I could write a program to evaluate the
> number of possible combinations a set of characters (string) can be
> written in - handling two identical characters as different
> characters.
> 
> Example:
> "ab" --> yields 2 ("ab", "ba")
> "abc" --> yields 6
> "aa" --> yields 2

My apologies for the delayed reply; I only recently saw your message.
There is a much better way to compute the number of unique permutations;
I'm suprised no one else has posted this (it's standard material in a
university probability course).

def uniq_perm(l):
    ''' returns the number of unique permutations of a list.

                      n!
            ------------------------
            n_1! * n_2! * ... * n_r!

        computes the number of unique permutations of n objects, of which n_1
        are alike, n_2 are alike, ... n_r are alike.  Take from _A First Course
        in Probability_, 5th edition, Ross, pg 5.
    '''

    acc = fac(len(l))
    l.sort()
    i = 0
    while i < len(l):
        cnt = l.count(l[i])
        acc /= fac(cnt)
        i += cnt
    return acc

def fac(n):
    ''' returns n! '''
    assert n >= 0
    if n == 0:
        return 1
    acc = 1
    for i in range(1, n+1):
        acc *= i
    return acc


Python 2.2 (#5, Jan 23 2002, 16:23:03)
[GCC 3.0.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from perm import *
>>> uniq_perm(list('pepper'))
60
>>> uniq_perm(list('aa'))
1
>>> uniq_perm(list('aba'))
3

-- 
Andrew Gaul
http://gaul.org/


From marklists at mceahern.com  Thu Feb  7 17:06:08 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 14:06:08 -0800
Subject: FixedPoint
In-Reply-To: <sDC88.96798$KM2.3611943@atlpnn01.usenetserver.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEIEKBAA.marklists@mceahern.com>

Steve Holden wrote:
> Oh, no, don't go changing other modules to fix this. Hadn't read your post
> carefully enough to realize your problems were occurring with database
> access.

As it turns out, Gerhard H?ring replied to the thread and he seems to want
to fix the x == None stuff.  Also, his post made me realize that PyPgSql
will look for and call a _quote() method on params so that I don't have to
resort to my goofy fix of making __repr__ return str(self).

It's all good.

Cheers,

// mark




From vtail at yandex.ru  Wed Feb 27 05:57:52 2002
From: vtail at yandex.ru (vtail)
Date: 27 Feb 2002 02:57:52 -0800
Subject: Tkinter and grid: strange alignment with and without sticky option
References: <9ff6b66.0202120802.3f20bcaf@posting.google.com> <3C73CDC6.F110D0C@mindspring.com>
Message-ID: <9ff6b66.0202270257.6a2ec8db@posting.google.com>

"Dr. David J. Ritchie, Sr." <djrassoc01 at mindspring.com> wrote in message news:<3C73CDC6.F110D0C at mindspring.com>...

> Yes, I see the same thing.  My Python version is:
> 
> Python 2.1.1 (#97, Aug  2 2001, 21:53:31)  [CW PPC GUSI2 THREADS] on mac
 
<skipped>

> Running a slightly modified version of the original poster's example
> listed below,
> I find that the non-sticky red Listbox is horizontally
> misaligned by a small amount (probably 1 pixel)
> from the following non-sticky

My Python version is
Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
on Windows 2000.

The things are becoming more interesting: when I run Your slightly
modified version on my system, I see that it's the last Entry box who
is misalligned.

I also receive the following personal answer to my post:

>I ran your program under KDE on my debian linux system.  I could not
>reproduce your bug.

Probably we should notify some of the Tkinter developers?

Sincerely Yours,

Vtail.


From jason at jorendorff.com  Sun Feb 24 21:34:33 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 24 Feb 2002 20:34:33 -0600
Subject: Parse variable into variables >Re: Apology
In-Reply-To: <2240.820T2407T12845765threeseas@earthlink.net>
Message-ID: <HFEKILOLEFEFMKAECNDLAENFDEAA.jason@jorendorff.com>

Timothy Rue wrote:
> I can read a line from a file but then I want to parse the line into a set
> of variables. How do I do the parse?

Use regular expressions!
A nice Python introduction:
  http://py-howto.sourceforge.net/regex/regex.html

Here is some sample code that parses a line of user input.

# -------------------------------------------------
import re

print "Please type a number, then a word.  Like this:"
print "> 1884384 toad"
print

x = raw_input('> ')
match = re.match(r'^\s*(\d+)\s+(\w+)\s*$', x)
if match is None:
    print "Sorry, I couldn't parse that."
else:
    number, word = match.groups()
    print "Your number was:", number
    print "Your word was:", word


## Jason Orendorff    http://www.jorendorff.com/




From empty_one at ameritech.net  Mon Feb 18 15:03:19 2002
From: empty_one at ameritech.net (Empty_One)
Date: Mon, 18 Feb 2002 20:03:19 GMT
Subject: Installing Python
Message-ID: <b6dc8.5870$pC1.1234514@newssrv26.news.prodigy.com>

I just downloaded the python tarball, and installed it on my slack system 
using sheck install.  All works good, except I cannot start IDLE.  How can 
I configure the install script to also install IDLE, since it is included 
with the base python, and not a seperate download.

-- 
Empty_One
_________
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."--Benjamin Franklin


From gh_pythonlist at gmx.de  Tue Feb 19 04:58:33 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 10:58:33 +0100
Subject: Adding static typing to Python
In-Reply-To: <slrna73h0o.18i.grante@tuxtop.visi.com>
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com> <mailman.1014085471.28317.python-list@python.org> <slrna73h0o.18i.grante@tuxtop.visi.com>
Message-ID: <20020219095832.GE2611@lilith.hqd-internal>

Le 19/02/02 ? 03:18, Grant Edwards ?crivit:
> In article <mailman.1014085471.28317.python-list at python.org>, Chris Gonnerman wrote:
> 
> > Ultimately, adding static typing to Python results in a different
> > language; it's not Python anymore.
> > 
> > If you don't like Python because the typing isn't strong, use Ada,
> > or C++, or Pascal... 
> 
> You might also want to take a look at Modula-3.  Anybody used
> to Python will recognize many of the language features.

Modula-3 also looks great, as did Sather to mee. But both seem seem to
have no compilers that are actively maintained.

So I'm still leaning towards Ada 95 and C++ for statically typed
languages.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.2 ?C      Wind: 9.0 m/s



From lac at strakt.com  Sun Feb 17 08:50:53 2002
From: lac at strakt.com (Laura Creighton)
Date: Sun, 17 Feb 2002 14:50:53 +0100
Subject: Highlights & Cursors 
In-Reply-To: Message from joost_jacob@hotmail.com (J.Jacob) 
   of "15 Feb 2002 15:01:12 PST." <13285ea2.0202151501.47080e91@posting.google.com> 
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com> <3daa1aa8.0202141551.1dae3651@posting.google.com> <2apq6ucuh9n7s446rrcsdi2fu7eis0qaqu@4ax.com>  <13285ea2.0202151501.47080e91@posting.google.com> 
Message-ID: <200202171350.g1HDorGo008623@ratthing-b246.strakt.com>

> [Lawrence Oluyede]
> > My question is: how can i change the cursor in a frame or some others
> > widgets?
> >
> 
> You mean this ?
> 
> from Tkinter import *
> widget = Button(text='new cursor', padx=10, pady=4)
> widget.pack()
> widget.config(cursor='cross')    # others: 'gumby' 'hand2' 'pencil'
> 'watch'
> mainloop()
> 
> Don't ask me the best place to look up Tkinter things like this on the
> internet i am still looking for that myself :)
> 
 Get John Shipman's 84 page guide.
http://www.nmt.edu/tcc/help/lang/python/tkinter.html
This is on page 9

Laura Creighton



From jkraska at san.rr.com  Mon Feb 11 21:17:14 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 12 Feb 2002 02:17:14 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <mt0g6ukklms8q21qsrb2kkjoohblgq3bn8@4ax.com> <Xns91B28848C4937RASXnewsDFE1@130.133.1.4>
Message-ID: <99vg6uojivd8lt3fjdcq6qpq8eahnovsjo@4ax.com>

>> I FUCKING ROCK AND YOU ALL SUCK!

>Considering the nature and meaning of you and i to depend on which person 
>is speaking, i think we can all agree with that. :)

So what you meant to say was "I fucking rock and you all suck?
Yeah, I agree with that! *I* do rock, and *you* do suck!" :) :)

C//



From m.faassen at vet.uu.nl  Mon Feb 11 21:07:08 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Feb 2002 02:07:08 GMT
Subject: OT: Perl programmers?
References: <mailman.1013225776.2733.python-list@python.org>
Message-ID: <a49tcc$5ao$3@newshost.accu.uu.nl>

hungjunglu <hungjunglu at yahoo.com> wrote:
> --- In python-list at y..., aahz at p... (Aahz Maruch) wrote:
>> ... the way we call ourselves Pythonistas?

> Never liked that name. You may call yourself Pythonista, but I prefer 
> the term Pythoneer. 

I prefer Pythoneer too. There's more than one way to do it! :)

Regards,

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


From arturs at iidea.pl  Fri Feb 15 14:01:47 2002
From: arturs at iidea.pl (Artur Skura)
Date: Fri, 15 Feb 2002 19:01:47 +0000 (UTC)
Subject: entries between labels dynamically
Message-ID: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>

I have a specific problem: I have to create dialogs on the fly, and
in those dialog text together with text entries.
Something like:


                               ^
Lavenda's ____, dilly dilliy,  |
_____ green, dilly dilly,      |
                               v

(where _____ is a text entry)

In other words: I have to put text entries between text labels.
Now can you suggest an easy, lazy way of doing this (with any
toolkit) so things don't pile up and everything looks nice?

Regards,
Artur
-- 
http://www.wolneprogramy.org


From travis at puddy.lan.kerrgulch.net  Thu Feb 14 14:48:06 2002
From: travis at puddy.lan.kerrgulch.net (Travis Shirk)
Date: Thu, 14 Feb 2002 19:48:06 -0000
Subject: AttributeError: 'getopt' module has no attribute 'GetoptError'
References: <Xns90D3ECC92390Dgustaflalgonetse@194.213.69.152> <ku7kxp32z3.fsf@lasipalatsi.fi> <Xns90D4988CE4ADEgustaflalgonetse@194.213.69.152>
Message-ID: <u6o53mb595981a@corp.supernews.com>

Gustaf Liljegren <gustafl at algonet.se> wrote:
> Erno Kuusela <erno-news at erno.iki.fi> wrote:

>>> I use Python 2.1 on Win98.  Try to use the getopt module, but get
>>> the error in the subject line. What's wrong?

>>your test program isn't called "getopt.py" by any chance?

> Oh. Yes, it is... Thank you.

I'm having the exact same problem as Gustaf, and
my program is *not* called getopt.py.  Anyone know what the
the AttibuteError is all about?

Travis


From martin at v.loewis.de  Wed Feb 20 17:29:35 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 23:29:35 +0100
Subject: socket in python 2.2 vs 2.1
References: <mailman.1014206732.25525.python-list@python.org>
Message-ID: <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de>

Ivan Bykov <ivan at kupol.ru> writes:

> It`s bug?

I think this is a bug in your program. You don't need to connect() for
a UDP socket, but if you do, you need to pass a valid address.

Regards,
Martin


From rdsteph at earthlink.net  Sun Feb 10 17:19:27 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 10 Feb 2002 22:19:27 GMT
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org>
Message-ID: <3C66F2CA.802A00BF@earthlink.net>

Well, now i believe in magic ;-)))  I added a one liner to run my
scripts at 17:00, and it will worked. But I don't know why. I even left
out the full paths, and I added the import sys and mail to lines for
traceback messages, Chris, btu I got no mail, no errors, and yet it
worked. I just wish I knew why it worked and didn't before....

My new crontab line looks like this: (I still can't paste the whole
crontab file because I only have a two button mouse and K-Edit doesn't
copy and paste into Netscape messenger ;-)))

00 17 * * * python mygale.py get all && python urls.py && python
urldb.py && python mygale.py -n 200 display all && python
mygale-uploader.py 

Which get all new links form the web and then selects the latest 200 to
make a new output.py and then uploads the new output.py to my site at
http://www.awaretek.com/plf.html

I'm glad it worked. I 'll try it again at 17:30 and see if it still
works ;-)))

But I am sorry it seems so mysterious. 

Thanks for the help!

Since id dint get any mail despite putting your insert into the top pf
mygale.py 

import sys
    sys.stdout = open("/home/ron/.error-log","a")
    sys.stderr = sys.stdout
    print "\n(job name) Starting..."

I wonder if my mail is broken or if it is record somewhere else???


From graz at mindless.com  Tue Feb 12 03:22:16 2002
From: graz at mindless.com (Graham Ashton)
Date: Tue, 12 Feb 2002 08:22:16 +0000
Subject: XP (was Re: Defending the Python lanuage... )
References: <3C62EE4A.5090305@destiny.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com> <mailman.1013434699.26587.python-list@python.org> <h1S98.86429$A51.36785660@typhoon.southeast.rr.com>
Message-ID: <Wg4a8.5754$k91.299578@wards>

On Mon, 11 Feb 2002 16:09:49 +0000, Thomas R. Corbin wrote:

> The XP folks say that just adopting part of XP just isn't as efficacious
> as doing the whole thing because each piece supports the other pieces.
> 
>         Of course, they also say, "season to taste".

Ron Jeffries says it's an ideal, which is almost never achieved:

  http://www.xprogramming.com/xpmag/Misconceptions.htm

>From which I quote:

  XP says that "all production code is written by two people sitting
  together at one machine". That's the rule. As far as I know, no project
  has ever done that, including many that I consider to deserve the name
  XP. So what's the rule about? It's an ideal.

Ron seems to be a bit of a wheel in the XP world too.

-- 
Graham


From gmuller at worldonline.nl  Thu Feb 21 13:49:34 2002
From: gmuller at worldonline.nl (GerritM)
Date: Thu, 21 Feb 2002 19:49:34 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee> <3C74DD65.3060000@thinkware.se> <m3heoa93gs.fsf@ns.bagley.org>
Message-ID: <a53fok$cfg$1@reader1.tiscali.nl>

<usenet+no_spam (" Doug Bagley ") .2002-02-21-14d7 @ (the) bagley . (not
.com) org (spam whammy)> schreef in bericht
news:m3heoa93gs.fsf at ns.bagley.org...
> Magnus Lyck? <magnus at thinkware.se> writes:
> > There is a comparision at http://www.bagley.org/~doug/shootout/
> > but it's really just a pointless play. Don't fall for this one.
>
> Harumph! The shootout isn't asking anyone to fall for anything. I hope
> I have enumerated the many deficiencies it has on the methodology
> page. I hope it's clear I'm not trying to pull the wool over anyone's
> eyes.
> <...snip...>
> http://www.bagley.org/~doug/contact.shtml

I think the shootout site is great. Of course in _using_ the information
great care and lots of common sense is needed.

It would be interesting to have some larger benchmark to compare language
behavior when scaled up in size.

Regards Gerrit

--
www.extra.research.philips.com/natlab/sysarch





From DavidA at ActiveState.com  Tue Feb 12 20:25:58 2002
From: DavidA at ActiveState.com (David Ascher)
Date: Tue, 12 Feb 2002 17:25:58 -0800
Subject: photo id ?!?!
References: <000c01c1b416$b85c9540$445d4540@Dell2>
Message-ID: <3C69C0A6.17E30809@activestate.com>

BrianQ:

> Presumably the security requirements of the Python conference are not so
> onerous that none of these pieces of non-governmental id would have been
> acceptable.

For those of you who weren't there (balaclava or not), it's interesting
to note that this year's conference probably had the best security
ever.  There was some sort of military convention in the hotel, w/
4-star generals walking around, and lots of  people in windbreakers
(always zipped up) and funny curly wires going to their ears.

As I mentioned to Andrew in private email, we'll make sure to address
his concern at future Python conferences (I can't recall if OSCON has a
similar requirement).

--david



From clarence at netlojix.com  Thu Feb 28 20:53:11 2002
From: clarence at netlojix.com (Clarence Gardner)
Date: Fri, 01 Mar 2002 01:53:11 GMT
Subject: mime multipart messages
References: <mailman.1014923455.7488.python-list@python.org>
Message-ID: <baBf8.77$GV.1890@newsfeed.avtel.net>

In article <mailman.1014923455.7488.python-list at python.org>,
	Glenn Stauffer <python at dejazzd.com> writes:
> I'm using the MimeWriter module to attach a file, submitted through an html 
> form, to an email message.  When I do this and send mail to a unix machine 
> and read the mail with mutt or kmail, the file is attached to the message as 
> I would expect.  However, when the message is opened on a Mac or Windows 
> machine, the file is mixed into the body of the message.
> 
> This is the code I use to construct the message:
> 
> -------------------------------------------------------------------------------
> 
> message = StringIO.StringIO()
> writer = MimeWriter.MimeWriter(message)
> writer.addheader('Subject', mail_subject)
> writer.startmultipartbody('mixed')
> 
> part=writer.nextpart()
> body = part.startbody('text/plain')
> body.write(msgbody)
> 
> part=writer.nextpart()
> part.addheader('Content-Type','application/mixed')
> part.addheader('Content-Transfer-Encoding', 'base64')
> part.addheader('Content-Disposition','attachment; filename=%s' % 
> form['file'].filename)
> body = part.startbody('application/binary')
> base64.encode(open(tmpfile,'rb'), body)
> 
> writer.lastpart()
> 
> -------------------------------------------------------------------------------
> 
> This is how the file looks when someone opens it with Eudora on a Windows PC:
> 
> --130.58.64.30.99.29450.1014921136.814.11340
> Content-Type: text/plain
> 
> 
> --130.58.64.30.99.29450.1014921136.814.11340
> Content-Type: application/binary
> Content-Type: application/mixed
> Content-Transfer-Encoding: base64
> Content-Disposition: attachment; filename=README
> 
> PT09PT09PT09PT09PT09PT09PT0NClRoZSB4bWxycGNsaWIgbW9kdWxlDQo9PT09PT09PT09PT09
> PT09PT09PQ0KDQpUaGlzIGtpdCBjb250YWlucyBhIFB5dGhvbiBpbXBsZW1lbnRhdGlvbiBmb3Ig
> 
> -------------------------------------------------------------------------------
> 
> Anyone have any thoughts on what I'm doing wrong or where to look to figured 
> this problem out?
> 
> Thanks,
> 
> Glenn
> 

-- 
Clarence Gardner
Software Engineer
NetLojix Communications
clarence at netlojix.com


From pobrien at orbtech.com  Tue Feb 19 23:25:24 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Wed, 20 Feb 2002 04:25:24 GMT
Subject: Python-based "Portal System"?
References: <mailman.1014134011.2051.python-list@python.org>
Message-ID: <UyFc8.2025$ZC3.168566@newsread2.prod.itd.earthlink.net>

I've used PostNuke, myself. And while Zope and Webware are probably capable
as foundations for similar systems, I know of nothing in the Python world
that really comes close to the products you mention. If I had time I would
love to implement something like PostNuke using Webware. I think Python
would really benefit from these kinds of applications that work out of the
box.

--
Patrick K. O'Brien
Orbtech


"Alessandro Bottoni" <abottoni at quadrante.com> wrote in message
news:mailman.1014134011.2051.python-list at python.org...
> Most likely, all of you know a number of open source, pre-built "portal
> systems", like ezPublish (http://developer.ez.no), PHPNuke
> (www.phpnuke.org), PostNuke (http://www.postnuke.com), Midgard
> (http://www.midgard-project.org/) and so on.
>
> Does anybody know if exists a Portal System like those, written in Python?
>
> Thanks in advance
>
> Alessandro Bottoni
>
> PS: I know about Zope (http://www.zope.org) and WebWare
> (http://webware.sourceforge.net), already...
>
>
>




From sholden at holdenweb.com  Mon Feb 18 08:28:17 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 18 Feb 2002 08:28:17 -0500
Subject: help - inserting values into database tables
References: <mailman.1013981787.10266.python-list@python.org> <7xpu33myrc.fsf@ruckus.brouhaha.com>
Message-ID: <mpsc8.7339$Zp2.716179@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xpu33myrc.fsf at ruckus.brouhaha.com...
> chajadan <python at chajadan.net> writes:
>
> > Yes, this is altogether possible.
> >
> > There is a module include in the main distribution called odbc, and
> > one called dbi.
>
> Whaaa?  That's the first I've heard of this.  In Python 2.2 under Linux,
> "import odbc" and "import dbi" both throw ImportError.
>
> There's an ODBC module in Vaults of Parnassus that looks pretty good, but
I haven't
> tried it yet.  Is that what you meant?

The dbi and odbc modules are only available, I believe, with the win32all
extensions, so you can't expect to see them on Linux or other Unixoid
platforms.

Since Joe was asking about a Microsoft Access database, however, this may be
enough for him. On Linux/Unix you can use DB driver modules directly, or
(non-commercially, without cost) Marc-Andre Lemburg's mxODBC module, though
you'll have to install an ODBC driver manager for that to work. I seem to
remember you can use either unixODBC or iODBC.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From pedronis at bluewin.ch  Mon Feb 25 12:01:32 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Mon, 25 Feb 2002 18:01:32 +0100
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com>
Message-ID: <3c7a70d2_4@news.bluewin.ch>

Dinu Gherman <gherman at europemail.com> wrote in message
gherman-035CF3.16342425022002 at news.t-online.com...
> Hi,
>
> I'm trying to get a little more familiar with Jython, but after having
> just installed the 2.1 final release on OS X I find the following
> rather surprising differences between Jython and CPython (2.1 and 2.2
> respectively, but that doesn't matter here):
>
>   1. cStringIO.StringIO.reset() is missing in Jython
>   2. list() doesn't perform as expected in Jython

[namely it does not do a copy :( ]

>
> Are these bugs and if so: how many people are actually using Jython,
> then, especially as books start to come out for it from O'Reilly and
> New Riders...?
>

Yes they are bugs. Please report them.
Since 2.1 release (two month ago) there have been approximately  10000
download.

How are such bugs possible?

1. the typical idiom for list(l) is l[:] (mildly irrelevant I know)
2. if you check the test suite for CPython 2.1 and Jython there is no test
for list(l) behavior
3. the new reset method is also not tested by the test suite and is not
reported in the NEWS file

(These are the result of some grepping, maybe I'm wrong but given the bugs
probably
I'm right).

We try to follow the big picture and the PEPs and check the NEWS file but
for the rest the test suites
are our best hope vs. delusional friend.
If a feature is old, under-used, or through usage the bug does not show
through, and un-tested things become tricky.

It is open source: Jython is as conforming as its community and CPython
community ACTIVELY want it to be.

regards, Samuele Pedroni.




From gcordova at hebmex.com  Wed Feb 20 11:25:21 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 10:25:21 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC3@SRM201002>

>
> for line in file.xreadlines():
>     parse(line)
> 
> xreadlines() does not load the whole file into memory and is 
> generally about as
> efficient as you can get.
> 
> The second item does however make re's more annoying in 
> python than they are in
> other languages where re's are native to the language.
>

Hmmm... so lets make them more readable.



>>> class WrapRX:
... 	def __init__(self, regex, String):
... 		self.regex = regex
... 		self.String = String
... 		self.pos = 0
... 	def __getitem__(self,k):
... 		# Ignore k, go for the next match.
... 		match = self.regex.search(self.String, self.pos)
... 		if not match:
... 			raise IndexError("No more regex matches.")
... 		self.pos = match.end()
... 		return match
... 
>>> some_string = "This is a lone string, with no friends or family."
>>> rx_vowels = sre.compile(r"[aeiou]+", sre.S)
>>> wrapped_re = WrapRX(rx_vowels,some_string)
>>> for match in wrapped_re:
... 	print "Found vowel: '%s'" % match.group(0)
... 	
Found vowel: 'i'
Found vowel: 'i'
Found vowel: 'a'
Found vowel: 'o'
Found vowel: 'e'
Found vowel: 'i'
Found vowel: 'i'
Found vowel: 'o'
Found vowel: 'ie'
Found vowel: 'o'
Found vowel: 'a'
Found vowel: 'i'
>>> 



Seems useful, *if* you wanna use it :-)

-gustavo



From mhammond at skippinet.com.au  Mon Feb  4 07:35:35 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 12:35:35 GMT
Subject: Compiling from Pythonwin?
References: <a3k0cf$dko$1@news.efn.org>
Message-ID: <3C5E8068.90608@skippinet.com.au>

Spencer Doidge wrote:
> Is there a way to compile .py scripts from the Pythonwin GUI screen other
> than using
> compileFile(...)?

Select File->Import (Ctrl+I).  This will create a .pyc for the file (and 
also load it into Pythonwin's main namespace).

Mark.



From deathtospam43423 at altavista.com  Tue Feb 19 08:29:33 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 19 Feb 2002 13:29:33 GMT
Subject: newbie: Some file object tips?
References: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4> <mailman.1014120512.7783.python-list@python.org>
Message-ID: <Xns91BAF4112B7BEAusYourStandingInIt@130.133.1.4>

Gerhard =?unknown-8bit?Q?H=E4ring?= <gh_pythonlist at gmx.de> wrote in
news:mailman.1014120512.7783.python-list at python.org: 

> Le 19/02/02 ? 11:00, Flavian Hardcastle ?crivit:
>> 
>> I'm writing a crude cgi chat script. I've already got it up and
>> working with Xitami server, and have chatted with a few net buddies.
>> 
>> How it works is very rough and simple... it basically just takes data
>> from people's browsers (using the FieldStorage function) and stores it
>> in a file called log.txt, and then displays the contents of that file
>> back to them. 
>> 
>> My problem is ... 
>> 
>> The longer you chat, the bigger the log.txt file gets, and
>> consequently the web page the chatters are viewing gets bigger and
>> bigger, and becomes hard to download.
>> 
>> So I want to include a "scroll back" feature ... which will enable the
>> user to control how many previous posts (s)he views. It will probably
>> take the form a simple input box on the form ... if you want to see
>> the last 8 posts, you put 8 in the box, if you want to see the last 50
>> posts, you put 50 in the box and so on.
>> 
>> So, I need the script to be able to search the log.txt file. Any tips
>> as to how I might go about this? 
> 
> You could read the file into a list with
> 
> lines = open("log.txt").readlines()
> 
> but now, all the lines will include a newline character at their end.
> To strip these, you can do the following instead:
> 
> lines = map(string.rstrip, open("log.txt").readlines())
> 
> Now you can use slicing on the list, say to return the last 8 lines,
> you can do:
> 
> lines[-8:]
> 
> HTH,
> 
> Gerhard

Cool! Thanx very much. Never saw that in any of the tutorials!

BTW, one thing I'm concerned about ... I want to avoid the script having to 
load the entire log.txt file into RAM everytime someone makes a post. You 
see, I estimate that a log file with only 200 posts worth of text would 
weigh about 60k. 

If say, ten people, are all posting posting at once that could be as much 
as 600k in the memory. If it's 20 chatters, 1.2 meg. That could take a 
couple of seconds for the servers CPU to process, might noticeably slow 
things down ... especially on my 'puter which is only 500Mhz.

So does this command ....

> lines = open("log.txt").readlines()

... actually load the whole log into RAM.

-- 
cdewin at dingoblue.net.au


From jason at jorendorff.com  Tue Feb 26 18:38:19 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 17:38:19 -0600
Subject: Class functions
In-Reply-To: <QGUe8.213$en5.1614@typhoon.sonic.net>
Message-ID: <HFEKILOLEFEFMKAECNDLGEIIDFAA.jason@jorendorff.com>

> Yeah, my function runs a loop looking for updates and
> posting them to a StringIO file object.  I want access
> to that file object in real time in my client instead
> of waiting for the function to "return".

Socket objects have a method sok.makefile('rb') that
returns a file object.  I think that's what you want.
  http://www.python.org/doc/current/lib/socket-objects.html

Anyway, if you could provide some real code, I'm sure
someone here could make it work.  :)

Good luck!

## Jason Orendorff    http://www.jorendorff.com/



From timr at probo.com  Sun Feb 17 01:32:49 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 16 Feb 2002 22:32:49 -0800
Subject: line break confusion
References: <mailman.1013827165.15861.python-list@python.org> <wz664xsbzg.fsf@nono.cs.uu.nl>
Message-ID: <8dju6uoscj0rtd7q7bkuiq84ivkugl7uua@4ax.com>

Piet van Oostrum <piet at cs.uu.nl> wrote:

>>>>>> Michael Mell <mike at nthwave.net> (MM) writes:
>
>MM> In the string module, whitespace = ' \t\n\r\v\f' (5 characters here).
>MM> Yet, when I run this:
>>>>> import string
>>>>> for c in string.whitespace:
>>>>> print '.'+ c + '.'
>
>MM> I get 7 results.
>MM> What's going on here?
>
>And then you will get more lines than 6. The \n, \v and \f will probably
>get two dots on separate lines each. The \r will rpobably print one dot
>only because the second dot overprints the first one. So you should get 11
>dots together.

I think you've nailed his 6 vs 7 problem:

C:\Tmp>python
Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import string
>>> string.whitespace.encode('hex')
'090a0b0c0d20'
>>> for c in string.whitespace:
...  print '.' + c + '.'
...
.       .      # \i
.              # \n part 1
.              # \n part 2
.?.            # \v, which prints as a male symbol in Windows
.?.            # \f, which prints as a female symbol in Windows
.              # \r, two dots overprinted
. .            # good old space
>>>

--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From winelight at spamcop.net  Mon Feb 18 02:37:53 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Mon, 18 Feb 2002 07:37:53 -0000
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com> <Ry3a8.40197$bP3.337803@NewsReader> <8f070afd.0202120550.45d8873b@posting.google.com> <Xns91B39B73814D6mhueningzedatfuberli@130.133.1.4> <gQfa8.41893$bP3.355857@NewsReader> <hjKb8.26255$l93.4949437@newsb.telia.net>
Message-ID: <lb2c8.839$S37.718@NewsReader>

> did you get anywhere on this one?

Sort of. I replaced the urllib stuff with the individual httplib statements
and so far, I haven't seen it hang.

Too soon to be 100% sure, really, but looks promising.

Andrew Ward





From phd at phd.pp.ru  Mon Feb 18 16:49:46 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 19 Feb 2002 00:49:46 +0300
Subject: Free Software For You .
In-Reply-To: <20020218093455.41c56955.logiplexsoftware@earthlink.net>; from logiplexsoftware@earthlink.net on Mon, Feb 18, 2002 at 09:34:55AM -0800
References: <1cef01c1b6da$944ca020$28eb42d9@supervisor> <20020218093455.41c56955.logiplexsoftware@earthlink.net>
Message-ID: <20020219004946.A7437@phd.pp.ru>

On Mon, Feb 18, 2002 at 09:34:55AM -0800, Cliff Wells wrote:
> On Sat, 16 Feb 2002 13:10:45 +0200
> feedback at cd-house.net wrote:
> 
> > Do You Need Free Software in the fields of
> > PC Games , Graphics , Applications , DVD , VCD , Music Audio , MP3 ,
> Educational Software and many more ............
> 
> > JUST Visit : http://www.cd-house.net
> > and don't forget to join our membership for free to get free software
> 
> Sic 'em, Oleg!

   There are different ways to promote Free Softwre. This message requires
SpamCop assistance :)

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



From robin at jessikat.fsnet.co.uk  Mon Feb  4 06:19:10 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Feb 2002 11:19:10 +0000
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
 <iJU5VJA6wdX8Ewk4@jessikat.fsnet.co.uk>
 <slrna5rpqd.19d.philh@comuno.freeserve.co.uk>
Message-ID: <qyRLLIAu4mX8EwWS@jessikat.demon.co.uk>

In article <slrna5rpqd.19d.philh at comuno.freeserve.co.uk>, phil hunt
<philh at comuno.freeserve.co.uk> writes
>>>Anyone want to add anything to this?
>>>
>>all your ideas are belong to you
>
>No, that's:
>
>all your (tim's) idea are belong to *us*.
>
><evil chuckle> he he he
I'm not sure I want them 
-- 
Robin Becker


From winelight at spamcop.net  Tue Feb 12 02:33:05 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Tue, 12 Feb 2002 07:33:05 -0000
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com>
Message-ID: <Ry3a8.40197$bP3.337803@NewsReader>

> Andrew, what version of python are you running?

I am using 2.2 under Windows XP, but in my case, the read() is of a plain
old single page of html text, no cookies, no nothing complicated!

Yes, urlopen() is what I would expect to hang, rather than the read(), so
I'm very puzzled.

Andrew Ward





From mwh at python.net  Thu Feb  7 06:40:11 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 11:40:11 GMT
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <Xns91AE6BE28FCD8michaelrcpcouk@194.238.50.13>
Message-ID: <uk7tpa574.fsf@python.net>

Michael Abbott <michael at rcp.co.uk> writes:

> I reckon to pay a cost of perhaps a factor of 100 in raw CPU for the 
> privilege of using Python.

But not over Java, surely?  I guess a JIT or native compilation will
help, but...

Cheers,
M.

-- 
  CLiki pages can be edited by anybody at any time. Imagine the most
  fearsomely comprehensive legal disclaimer you have ever seen, and
  double it                        -- http://ww.telent.net/cliki/index


From jkraska at san.rr.com  Sat Feb 16 21:16:59 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 17 Feb 2002 02:16:59 GMT
Subject: Guido wins FSF Award
References: <mailman.1013906967.6705.python-list@python.org> <7xy9hsvr40.fsf@ruckus.brouhaha.com>
Message-ID: <3s4u6ugk038esv6drk5go253u95fa5o5fg@4ax.com>

>> Not at all.  Guido's acceptance speech was a foundational masterpiece,
>> revealing the Nine Pythonic Actions underlying all that we do.

Really, Geedo didn't deserve this award. Since plainly Python is
a system designed to obfuscate code, perpetuate complexity, and other-
wise fails to recognize the simplicity of the nine command division
of the entire physical universe, I redirect the committee to Sir
Timothy Rue. My vehement protest to the comittee will be publically
filed by 1100 GMT tomorrow, I swear. And well, if none of you see or
understand why it is I have to do this, that's understandable. You are
perhaps simply incapable of comprehending my rationale.

inserting-a-marker-for-the-satirically-impaired ly yrs -C



From jkraska at san.rr.com  Wed Feb  6 11:22:09 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 06 Feb 2002 16:22:09 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk>
Message-ID: <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com>

>>That's non hygienic behavior.

>And what if I decide that for my particular application, that's just
>the behaviour I want? Or do you think that you know better than all
>the users of your macro system, even though you've not met them and don't
>know anything about the software they are writing?

Well, since one has to go through hoops to make a macro hygienic,
it's not hard at all to make them unhygienic. The plan is for hygiene
because this is generally regarded as a best practice by the community.
Which is to say, this as a default is regarded as what is most likely
to service _everyone_ best, allowing for exceptions of course.

You say "that you think you know better" as a scathing remark when what
you should know is that the idea of hygienic macros and their regard as
a best practice don't come from me at all. If what you want is a means to
make them unhygienic, just say so, don't scathe.

>It doesn't need handling; python does it OK already.

You can do it fairly trivially in Python, but there is no such
form in Python. The example was used to demonstrate the ability
to construct such a form, not acknowledge the goodness of its use.

C//



From tim-dated-1014628835.d87d33 at catseye.net  Mon Feb 18 04:20:33 2002
From: tim-dated-1014628835.d87d33 at catseye.net (Tim Legant)
Date: 18 Feb 2002 03:20:33 -0600
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
In-Reply-To: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com>
References: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com>
Message-ID: <863czzi1su.fsf@skitty.catseye.net>

"Jason Orendorff" <jason at jorendorff.com> writes:

> text = """'Twas _____, and the _____ _____
> Did _____ and _____ in the _____.
> All _____ were the _____
> And the _____ _____.
> """

I haven't read Jabberwocky in about 20 years, I think, and I don't
have a copy here (a sad deficiency in my library!), but don't you want
three blanks in the last line?  From (ancient) memory, isn't it:

'Twas _brillig_, and the _slithy_ _toves_
Did _gyre_ and _gimble_ in the _wabe_.
All _mimsy_ were the _borogoves_ (borogroves?)
And the _mome_ _raths_ _outgabe_.


Isn't-it-weird-having-a-mind-like-a-landfill-ly y'rs,

Tim



From wwaldo4 at cox.net  Wed Feb 20 13:19:21 2002
From: wwaldo4 at cox.net (william)
Date: Wed, 20 Feb 2002 18:19:21 GMT
Subject: Linux Library install
Message-ID: <JMRc8.1761$Ep1.91991@news2.east.cox.net>

I am trying to install the imaging libraries with Python.  This is what i
have:

Redhat 7.2

Installed Python 2.2 from the rpm.  Default dir is /usr/lib/python2.2

I am trying to install the 1.1.2 imaging libraries.

1.  I downloaded the tar files. and untared them to
/usr/lib/python2.2/imaging-1.1.2
2.  I then do the ./confige and then make
3.  When I run a python program that a buddy made it states that i cannot
find ImageDraw, Image, ImageFont etc.  These were in the imaging libraries
that i just installed.

Any suggestions on what the problem might be?

Also i do belive i still have python 1.5 and 2.1 installed.  could they be
causing a problem?

Thanks





From tismer at tismer.com  Mon Feb 25 05:42:39 2002
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 25 Feb 2002 11:42:39 +0100
Subject: Time to rename Stackless? (Re: Stackless Platform Independence?)
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz>
Message-ID: <3C7A151F.1090709@tismer.com>

Greg Ewing wrote:

> Seems to me that "Stackless" is a misnomer now. If it
> works by switching between multiple stacks, then it now
> uses *more* stacks than normal Python!


:-)

Well, it doesn't switch between multiple stacks, it
repeatedly replaces parts of the same stack.
(But this would have the same effect, of course).
It is still stackless in the sense that order of
execution is not restricted by the C stack.
It is the same effect as in the old design, just
a simpler approach:

Instead of begging the interpreter not to recurse,
I now enforce it. Stackless Python always had
and still has a C stack. Today it is totally
freed form the C stack's limitations, so it is
more stackless than ever before.

> Maybe something like "Microthreaded Python" or
> "Multitasking Python" would be better?


We have now tasklets, proclets, and the stack slices
are stacklets. :)

Stacklets Python - ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From James_Althoff at i2.com  Thu Feb  7 13:36:26 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 7 Feb 2002 10:36:26 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
Message-ID: <OFA32EDAC6.862A1516-ON88256B59.006552E1@i2.com>

[Paul Rubin]
>I don't know of any Python projects on such a scale.  The biggest
>ones I know of (e.g. Zope) are smaller and have extremely good
>programmers rather than just ordinary programmers.  So they don't
>face the problems that typical Java projects do.
>
>I don't know what would happen if someone tried a big project
>in Python (Zope is medium sized).  Maybe it would work ok, maybe
>not.

We tried it -- using Jython.  I haven't counted lately but the total lines
of code is in the many-hundreds of thousands.  Development has been ongoing
during the past 2.5 years with around 40 - 50 different developers involved
at various times (in and out with usually 10 - 15 at peak times).  The
development group comprises engineers in both Silicon Valley, CA and
Bangalore, India.

Seems to work -- the software is in production at happy customer accounts
(and has generated valuable revenue <wink>).

Jim




From gh_pythonlist at gmx.de  Fri Feb 15 00:02:07 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Fri, 15 Feb 2002 06:02:07 +0100
Subject: Curses & Colors
In-Reply-To: <3C6C06FF.DAFFBB43@zumnet.com.br>
References: <3C6C06FF.DAFFBB43@zumnet.com.br>
Message-ID: <20020215050206.GB2867@lilith.hqd-internal>

Le 14/02/02 ? 16:50, Rafael Mentz Aquino ?crivit:
> Hi, there...
> 
> I'm creating some simple interfaces with curses with python under
> FreeBSD, but I'm having some problems to change colors. I tried to do
> something in /etc/termcap, but when I run curses.can_change_color(), I
> always obtain 0.  My env term is cons25, but I don't know if python
> uses the same terminal.  I'm running python 2.1.1. Can anybody help
> me...? ;-))

I recently fiddled with this, too. And it took me a while to get the
curses code right. What I *can* do is give you some source that will
work if color works at all:

import curses

def mymain(stdscr):
    curses.init_pair(1, curses.COLOR_RED, curses.COLOR_WHITE)
    stdscr.clear()

    stdscr.addstr(0, 0, "This text is red on white.", curses.color_pair(1))

    stdscr.refresh()
    stdscr.getkey()

curses.wrapper(mymain)

I've no solution for how to activate color on your FreeBSD console,
though. Maybe best ask on a FreeBSD group.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: -0.7 ?C      Wind: 4.1 m/s



From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 19:32:07 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 18:32:07 -0600
Subject: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254>
Message-ID: <006a01c1b35c$b8f4d220$0101010a@local>

----- Original Message -----
From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Monday, February 11, 2002 5:12 PM
Subject: Help: stuck on first example!


> The example at http://www.devshed.com/Server_Side/Python/CGI/page3.html is
> supposed to produce the text in my browser "Hello, Python", but it doesn't
> do a thing.
>
> Instead, using the Xitami personal server, I get the error
>
> >Internal error: your request was unsuccessful
> >Cannot create CGI process - program not found
>
> Here is the text of the example...
>
> #!/usr/bin/python

Change this line to point to your Python interpreter.  Xitami requires an
extra space; using Python 2.2, the line would be:

    #! c:\python22\python.exe

Add a -u at the end for unbuffered output (recommended):

    #! c:\python22\python.exe -u

The rest of your script looks good.





From tim.one at comcast.net  Wed Feb 27 03:28:33 2002
From: tim.one at comcast.net (Tim Peters)
Date: Wed, 27 Feb 2002 03:28:33 -0500
Subject: Time to rename Stackless?
In-Reply-To: <a5gevj$77e6d$1@ID-11957.news.dfncis.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCOENANPAA.tim.one@comcast.net>

[Christian]
> Not bad!
> Should I start a contest?

I vote for "Christian Python".  This would give Python a huge advantage in
the all-imporant American Bible Belt market <wink>.

suffer-the-tasklets-to-come-unto-me-ly y'rs  - tim




From chris.gonnerman at newcenturycomputers.net  Mon Feb 18 23:56:44 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 18 Feb 2002 22:56:44 -0600
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com> <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com> <3C715962.A42F1E48@ccvcorp.com>
Message-ID: <011e01c1b901$d55e08a0$0101010a@local>

----- Original Message -----
From: "Jeff Shannon" <jeff at ccvcorp.com>
>
> Michael Kelly wrote:
> >
> > Out of curiosity, did they give you a type library
> > with the control?
>
> This is important.  As I understand it, without a type library, pythoncom
> has no way of knowing which params are [out] params.  Since Python doesn't
> have reference semantics like those of C++ or Visual Basic, it cannot use
> [out] params in the "normal" way.  So if you're using an interface that
> makes use of [out] params, you *need* the type library in order to be able
> to use it from Python.  If you're not sure whether you've got one, try
> Makepy, as suggested...

Type libraries have a .tlb extension normally, correct?  The ZIP contains
ReportOcx.lic, .lib, and .ocx; I'm a novice at Windows and had to be banged
on the head before I put the .lic and .ocx together in the SYSTEM folder.
I did register the OCX using REGSVR32 as noted on several websites, and I
ran makepy from the Pythonwin menu as I have in several other instances.

Have I missed anything?

The source files are not available (it's a form of shareware, where you pay
for source access) so I can't build a type library even if I knew how.

I should just go back to Linux where everything makes sense...





From threeseas at earthlink.net  Tue Feb  5 23:54:28 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:28 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net><iJU5VJA6wdX8Ewk4@jessikat.fsnet.co.uk><slrna5rpqd.19d.philh@comuno.freeserve.co.uk> <qyRLLIAu4mX8EwWS@jessikat.demon.co.uk>
Message-ID: <887.801T2642T14375708threeseas@earthlink.net>

On 04-Feb-02 06:19:10 Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>In article <slrna5rpqd.19d.philh at comuno.freeserve.co.uk>, phil hunt
><philh at comuno.freeserve.co.uk> writes
>>>>Anyone want to add anything to this?
>>>>
>>>all your ideas are belong to you
>>
>>No, that's:
>>
>>all your (tim's) idea are belong to *us*.
>>
>><evil chuckle> he he he
>I'm not sure I want them
>--
>Robin Becker


Honesty! It's the first step in the twelve step program. You are showing
more hope and promise than some others.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From aahz at panix.com  Thu Feb  7 20:17:55 2002
From: aahz at panix.com (Aahz Maruch)
Date: 7 Feb 2002 17:17:55 -0800
Subject: What's the equivalent in Python for Perl's CPAN?
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com>
Message-ID: <a3v903$kgk$1@panix2.panix.com>

In article <CFz88.19365$zJ3.1503300 at typhoon.maine.rr.com>,
Mike Carifio <carifio.nospam at nospam.usys.com> wrote:
>
>

That's an impressive post.  Try a Google for Parnassus.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From jimd at vega.starshine.org  Mon Feb 18 19:53:46 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 19 Feb 2002 00:53:46 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <a4s7mq$u07$1@news.idiom.com>

In article <24c39b2c.0202181531.187fad4c at posting.google.com>, 
Alexander Jerusalem wrote:

> Has anyone heard of plans to add static type checking to Python? I
> like Python very much but I don't use it for bigger projects because I
> fear the testing effort. For example I just refactored a rather large
> Java codebase. And when I change a method name for example, all I have
> to do is run everything thru the compiler and it will show me all
> places in other classes where this method is called, so I can go there
> and chenge the method call as well.

> In Python I'd have to test the whole software suite including each and
> every branch in each and every if statement to be sure that I didn't
> forget something. To build such a test suite would take almost as long
> as it takes to build the whole software and I'd have to maintain the
> test suite as well. That's just unafordable.

> On the other hand, Python is the best language that currently exists
> and I would like to use it on bigger projects not just for prototyping
> as some suggest.

> Regards,
> Alexander Jerusalem

 Let's think about this suggestion:  static typing allows you to use
 one technique for detecting a particular programming oversight
 in one particular (family of) programming languages (by raising compile 
 time errors).  So this "feature" (or constraint or put most neutrally,
 "characteristic")  must be available in all other languages so that
 you can use the same technique.

 Perhaps you should consider using a different technique.  Is there
 another technique that might detect the sorts of errors you're 
 looking for?  Maybe you should consider adding additional tools to 
 your toolbox?

 For example, you might look at PyChecker.  It might do what you 
 want, or you might extend it to do what you want.  Python has 
 wonderful "metaprogramming" (introspection/reflection) features
 (using .__class__ and .__dict__ etc).  So you could write or 
 modify a tool to walk through your modules, class trees, and their
 members and methods and report on all sorts of errors.  

 Of course there are limitations to this technique.  For example
 I see that it's common for Pythonistas to use raise NotImplementedError
 for abstract classes.  It would be nice of PyChecker could detect
 any time a concrete subclass fails to implement any of the requisite
 abstract functions.  Of course it's difficult to see how this would
 be accomplished given current language features: how would we tell
 that a given abstract function is "abstract" (is defined to raise 
 NotImplementedError)? how could we tell that a given subclass was
 "concrete" (final)?  (Obviously we can see that any class that is
 used to instantiate an object must be concrete, though not necessarily
 "final", but that only works if we're given a whole project, otherwise
 there might be classes which are intended to be concrete but not
 used within the scope of our hypothetical scanning engine)
 
 However, this all begs the original question.  You're using the 
 compiler as a "test" tool.  Historically we see that compilers make
 very poor test suites.  There's just too many bad bits of code that
 compile.

 So perhaps it's better to go back to basics: 

	Requirement:  	You need rapid feedback from 
			programming errors in large projects
			(for refactoring and enhancement)

 ... so you want automated tests that will alert you when changes
 in your project's modules, classes, packages, and other structures
 "breaks stuff."

 So, build them in as part of the development process.  Mark Pilgrim's
 "Dive Into Python" (http://www.diveintopython.org/) has a whole
 chapter on unit testing (and extensive examples of using the unittest
 modules for Python).

 Then you'll be testing for far more then just superficial syntactical 
 errors; you should catch semantic errors more quickly.  Every time
 one "slips through" try to figure out ways to add cases to the appropriate
 test suite(s).



From karthik at james.hut.fi  Fri Feb 15 15:47:53 2002
From: karthik at james.hut.fi (Karthikesh Raju)
Date: 15 Feb 2002 22:47:53 +0200
Subject: Basic Question
Message-ID: <m3y9huzd3a.fsf@mycomp.tky.hut.fi>

Hi,

Just started, so my question might be stupid. i tried this:

import random

def randomList(n):
    s = [0] * n
    for i in range(n):
        s[i] = random.random()
    return s

in a file called randomList.py. Started python, imported randomList
(>> import randomList) and then called this function (>>
randomList(5)), got this un-understandable error:

Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: object of type 'module' is not callable

But when i when type the function at the python command prompt, it
works. Couldnt figure out why..

Thankx in advance,

Best wishes
karthik


From SBrunning at trisystems.co.uk  Tue Feb 12 06:41:08 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 12 Feb 2002 11:41:08 -0000
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <31575A892FF6D1118F5800600846864DCBCC26@intrepid>

> From:	Bernhard Herzog [SMTP:bh at intevation.de]
> 
> Gui-do: the way of the python
 
I feel a t-shirt coming on...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From pobrien at orbtech.com  Mon Feb 18 09:53:02 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Mon, 18 Feb 2002 14:53:02 GMT
Subject: OT: Mike C. Fletcher
Message-ID: <iz8c8.19222$P21.1815373@newsread1.prod.itd.earthlink.net>

[Sorry about this off-topic message.]

Mike,

I'm trying to reply to the last email you sent me, but I keep getting errors
when I try to email you at geocities.com:

Connected to 209.1.224.42 but sender was rejected.
Remote host said: 550 5.7.1 Access denied

Do you have another email address? Any idea what the problem might be?

--
Patrick K. O'Brien
Orbtech






From p.vrijlandt at aig.azn.nl  Mon Feb 18 10:09:04 2002
From: p.vrijlandt at aig.azn.nl (Patrick Vrijlandt)
Date: Mon, 18 Feb 2002 16:09:04 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no>
Message-ID: <a4r5ei$pkm$1@odysseus.uci.kun.nl>

Hi,

Previous discussions have narrowed down the line in telnetlib that is buggy.
The problem is probably that the server is sending you a special (Telnet)
command.

Unfortunately, telnetlib does not support Telnet options at all. This means,
making telnetlib adhere to the RFC is quite a lot of work, and a simple bug
report won't be enough.

--
Patrick Vrijlandt




From BPettersen at NAREX.com  Wed Feb  6 11:55:27 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Wed, 6 Feb 2002 09:55:27 -0700
Subject: Autocoding project proposal -- finally implemented!
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192215138@admin56.narex.com>

> From: Timothy Rue [mailto:threeseas at earthlink.net] 
<snip>
> 
> Good, than you won't know it hit you untill it is to late, for you.
> 

Well, since you don't want to implement it I've gone ahead and done it
for you... You can simply ask it a question, and it will give you an
answer, for example:

  >>>from VIC import Question
  >>>Question("What's the meaning of life, the Universe, and
Everything?")
  42

I plan to market and sell this myself though, so I can't show you the
code -- but I'm sure with your amazing intelligence you should be able
to reverse engineer this in no time at all...

-- bjorn



From phr-n2002a at nightsong.com  Thu Feb 21 14:59:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 11:59:04 -0800
Subject: newbie floating point question
References: <pan.2002.02.21.12.47.20.378591.4031@R191.net>
Message-ID: <7x3czutxmf.fsf@ruckus.brouhaha.com>

Alex <alexhelder at R191.net> writes:
> Why does this happen ?
> 
> >>> .2
> 0.20000000000000001
> 
> Is this the correct behavior for the Python 2.2 interpreter ? (version info
> is printed below )

It's because 0.2 (= 1/5) can't be represented as an exact binary
number.  Only numbers like 0.25 (= 1/4), 0.125 (1/8), 0.375 (1/4 +
1/8), and so forth can be represented exactly.  0.2 gets converted
into an inexact approximation that differs in the 17th or so decimal
place.  When you print it out displaying all the digits, you can see
the discrepancy.  It's normal and expected behavior.  If you're doing
serious numerical computing you have to worry about this kind of thing.
For more casual purposes you probably can ignore the tiny error.


From jeff at ccvcorp.com  Mon Feb  4 19:01:04 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 16:01:04 -0800
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <3C5F08BF.95D8ED5B@ccvcorp.com> <3c5f0c6d$1_1@news.nwlink.com>
Message-ID: <3C5F20C0.309F34EC@ccvcorp.com>


Guy Gascoigne - Piggford wrote:

> Neat.  The dictionary idea might be tidier - I'll try it out.  The only
> thing that I'm unsure about is that with a list there is a predictable order
> in which the list is iterated and as such there is a clear order in which
> the patterns are tested.  With a dictionary I assume that whilst the order
> is constant, it's not at all obvious from the declaration what it'll be.

This is true.  Dictionaries are, I believe, internally "ordered" by the hash
order of their keys--constant, but essentially random.  Depending on your usage,
though, you may be able to extract the list of keys, sort that, then iterate
over the sorted list:

keys = MyDict.keys()
keys.sort()
for key in keys:
    if received == key:
        MyDict[key]()

or something of that sort.  (Spice with an appropriate cmp() function to taste
;) )


> As for the clarity of anonymous functions, I've seen it go both ways.  Where
> they are short I think that they can be very useful, but you're right, as
> they get longer they can become very distracting.

Yes, it is a bit of a tradeoff, and the exact point at which anonymous functions
cease to be beneficial is a matter of taste.  My own taste puts that point
pretty low, but judging by the number of lambda-proponents in this group, my
opinions don't necessarily represent any sort of concensus.  ;)

Jeff Shannon
Technician/Programmer
Credit International




From sheershion at mailexpire.com  Fri Feb 15 20:24:31 2002
From: sheershion at mailexpire.com (Robert Amesz)
Date: Sat, 16 Feb 2002 01:24:31 -0000
Subject: How to disable a button control in wxPython?
References: <3bff5060.0202150900.77c5f5fa@posting.google.com>
Message-ID: <Xns91B6DE74F368Crcamesz@amesz.demon.nl>

eriggs wrote:

> I'm somewhat new to Python, and I'm trying to figure out how to
> disable a button control.

Use use button.Enable(true) instead of button.Enable(), and use 
button.Enable(false) instead of button.Disable(). That should do the 
trick. The documentation seems to be wrong concerning that, at least 
for wxPython.


Robert Amesz


From s.schwarzer at ndh.net  Sun Feb  3 08:40:51 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Sun, 03 Feb 2002 14:40:51 +0100
Subject: reading ftp log file..
References: <aO078.9013$Xk4.483475@news1.east.cox.net>
Message-ID: <3C5D3DE3.C7E17153@ndh.net>

Hello Andrew

Andrew Replogle wrote:
> I'm trying to write a python script that watches an ftp log file.

What kind of "watching" do you mean? Runs your script continously or
do you start it in certain intervals via a cronjob (or what else)?

Please tell us a bit more about your application.

> Based on someone logging in and successfully transmitting a file it will use
> the smptlib to generate somewhat of a dynamic mail as an "upload
> confirmation".  But I'm so new to python and programming in general that
> this is proving very difficult.
> 
> this is what I have so far..
> -----------------------------------
> import os
> import sys
> import smtplib
> 
> ServerPath = 'c:\\Program Files\\G6 FTP Server\\'
> logExists = os.path.isfile(ServerPath + 'FTP.log')
> line = "230"
> #print logExists
> 
> log = open(ServerPath + 'FTP.log')
> lines = log.read()
> for line in lines:
>  print "someone logged in"
>  break
> -------------------------------------
> 
> the "230" you see is the code the ftp server puts in the log if X user

You can check a line whether it starts with "230" by writing

if line.startswith("230"):
    ...

> successfully logs in.. I'm trying to search for that in the file and if it
> exists then go on with what the account is & what file they transfered.

How does a complete line in the logfile look like?

> it always prints "someone logges in" even if I change 'line'  to a bazillion
> zeros..

You use read() to access your logfile which slurps the whole file into
a single string. So what your loop above actually does (if you omit the
break statement) is to iterate over each single character:

>>> contents = 'a string'
>>> for char in contents:
...     print char
...
a

s
t
r
i
n
g
 

Stefan


From dsh8290 at rit.edu  Tue Feb  5 15:18:34 2002
From: dsh8290 at rit.edu (dman)
Date: Tue, 5 Feb 2002 15:18:34 -0500
Subject: when does python free memory?
In-Reply-To: <10b3f29a.0202050840.7641923a@posting.google.com>
References: <10b3f29a.0202050840.7641923a@posting.google.com>
Message-ID: <20020205201834.GA5828@dman.ddts.net>

On Tue, Feb 05, 2002 at 08:40:12AM -0800, Robert Eanes wrote:
| Hi-
| Reading the docs and searching the list archives, it seemed that
| python should release memory as soon as the last reference to an
| object is deleted.  However, I can't seem to prove this in practice. 

It frees the memory whenever it darn well feels like (which is only
sometime after all references are gone).  Certainly the timing
semantics are different in CPython vs. Jython.

-D

-- 

Consider what God has done:
    Who can straighten what He has made crooked?
        Ecclesiastes 7:13




From eppstein at ics.uci.edu  Tue Feb 12 16:30:08 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Tue, 12 Feb 2002 13:30:08 -0800
Subject: Why no documentation of basic data structure performance?
Message-ID: <eppstein-0310EF.13300812022002@news.service.uci.edu>

Does anyone else notice a serious lack of timing information in the Python 
documentation?

For example, the Python tutorial goes through an example of using a list as 
a stack (5.1.1), with append() and pop(), and as a queue (5.1.2) with 
append() and pop(0).  It does not mention the significant differences in 
timing between these two pieces of code -- the queue operations take time 
linear in the length of the list while the stack operations take constant 
amortized time (theory), and it is easy to come up with cases where this 
makes a significant difference in actual time (practice).

We had a discussion a couple weeks ago on this board about how to implement 
queues efficiently, but I'm more concerned here with the state of the 
documentation: why doesn't it give some indication that 5.1.1 is fast and 
that 5.1.2 is slow?  If a programmer needs to add and remove things from a 
list of items, and it doesn't matter for correctness whether he uses stack 
or queue order, how is he supposed to know that the stack is a better 
choice?

Even when I knew about this issue and wanted to know whether append() was 
really constant-amortized-time, I couldn't find any mention in any of the 
Python documention of the performance of these basic list operations, 
instead I had to grope the source, find listobject.c, and see the comment 
near the start of that file:
         * This over-allocates proportional to the list size, making room
         * for additional growth.  The over-allocation is mild, but is
         * enough to give linear-time amortized behavior over a long
         * sequence of appends()

If it was worth putting in the source, why isn't it worth putting in the 
documentation?  I have the impression that one of the biggest negative 
points for Python wrt other languages is the perception that it's slow, but 
that a lot of that perception may come from programmers not having a good 
model of how much time each Python operation takes -- if they don't know 
which operations to avoid, they can't design their code to avoid them.

(Not that Python is alone in this lack -- can anyone tell me whether the 
Java Vector class uses a constant-amortized-time resizing strategy when not 
given a capacityIncrement initialization parameter?)
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From francois.lepoutre at seriatim.com  Fri Feb  8 11:42:22 2002
From: francois.lepoutre at seriatim.com (François Lepoutrre)
Date: Fri, 8 Feb 2002 16:42:22 -0000
Subject: mxODBC, stored procedure syntax
References: <e746a5a8.0202080441.4bd2c023@posting.google.com>
Message-ID: <a40s9m$aif$1@suaar1aa.prod.compuserve.com>

Hi,

As far as i understand it there is currently
no support for returning params from
storedprocedures at the moment.

I understand it is on their "to-do" list.

The trick is to have the stored procedure
return a cursor (select statement).

Works quite well for us on sybase ASA.

I'll like the support for odbc to be extended
to quite a lot of other things...

But odbc works - simple and stable.

That makes python useful of us.
Many thanks to M.A Lemburg :)

Fran?ois

cb921 wrote in message ...
>Hi,
>
>I have been trying to connect to an MS SQL Server via ODBC using mxODBC.
>It's all working, but calling of a stored procedure is causing problems.
>Without any syntax worries, it looks like:
>
>cr.execute("{call storedprocedure('cellnr','passwd')}")
>
>and it should return me two output parameters, in the example (not Py :)
code
>I've seen we call this @DATABASENAME and @ISNOM.  How do I retrieve those
>values?  So far all I've managed is a flock of errors.
>
>Anyone who can help me, even an insignificant bit, thank you...
>
>Cheerio,




From joost_jacob at hotmail.com  Fri Feb  8 08:24:37 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 8 Feb 2002 05:24:37 -0800
Subject: a portable text user interface ?
Message-ID: <13285ea2.0202080524.e2e7948@posting.google.com>

We would like a -portable- module that provides 4 functions:


setupTextUI(width, height)  # returning a text UI object, top left is
(0,0)

writeXY(textWindow, x, y, string)   # print string on position (x,y)

readXY(textWindow, x, y)    # waiting for user input with cursor
positioned on
                            # (x,y) and returning a string from user
clearWindow(textWindow)


or better function names if you come up with it.

The textWindow parameter is here for the example and not really
necessary
because the module could handle it internally.  The text window's top
left
corner should match the console window's top left.  If the text window
is
bigger than the console window it just gets clipped.  More functions
are not
needed immediately, e.g. you need to erase part of the window just
print
spaces there.

With a module like this included in the Python distribution you could
make
simple text based user interfaces that work on Windows, linux, *nix
and Mac
that dont flicker, that is, if it is well implemented.  It looks so
simple but
it really is not.  You would need to know how to do things on all
platforms.
I tried and gave up, besides i do not have access to much OS's to test
things
with.  Doing this for linux and unix is simple with the curses module,
but
Windows? Mac?  The pdcurses and WConio modules i found for Windows
need
compilation etc., couldnt test/install them properly (no MS compiler
here).
Cygwin Python has partially implemented curses but then you need to
have Cygwin installed.
What we need is something in the Python distribution -itself- that is
portable.
Portable curses would be nice, but maybe its too much to start with,
most
people would be happy with only a few functions like those above,
routines
like the low/high-Video could come later.

We think a well implemented portable text ui module in the Python
distribution would improve Python a lot.  What do other users think? 
What do
the core Python developers think?  Should it be a PEP?


From logstx at bellatlantic.net  Tue Feb 26 02:06:41 2002
From: logstx at bellatlantic.net (logistix)
Date: Tue, 26 Feb 2002 07:06:41 GMT
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <5uGe8.4184$LN5.796@nwrddc01.gnilink.net>

> 3) (Separate proposal from above): add optional type advice:
>
>      local x(foo)     # advises Python implementation that x will
>                       # only be bound to instances of class foo
>
>    This is called "advice" rather than a "declaration" because it has
>    no defined semantics and the compiler is free to ignore it, except
>    that if the program acts as the advice says it will, it will work.
>
>    Example 1:
>      local x(int)    # advise Python that x is an int
>      x = 3 + 2       # guaranteed to set x = 5 and work as you'd expect
>
>    Example 2:
>      local x(int)
>      x = "banana"    # implementation-dependent behavior
>

The type advice is really impossible in a dynamic language and it's not
going to help out a native compiler anyways.  A better example (than Example
2) of code biting you in the ass is:

def foo(x):
    local y(int)
    y = x

This will only be able to successfully statically analyze if every call made
to foo is passing in a statically typed value.  So for any of the "advice"
to work, all vars need to be typed.  This would remove all dynamic
properties from python, which is probably undesirable.  If you do need/want
to validate the type at runtime, you might as well use the builtin type()
function.  This is exactly the way you have to validate info in "real
languages" like C++ (and I imagine java). RTTI is required if your class
heirarchies get too crazy for static checking (such as a toy lisp
interpreter I wrote, all lisp objects inherited from a base "object" class).

# inline
def foo(x):
    assert type(x) == types.IntType, "Invalid parameter passed to function
foo"
    y = x

# helper functions
def isInt(x):
    assert type(x) == types.IntType, "Invalid value, should be an int"
    return x

def foo(x):
    y = isInt(x)

# or of course a class with Operator Overloading
# won't bore you with implementation




From bhoel at web.de  Tue Feb 26 15:33:43 2002
From: bhoel at web.de (Berthold Höllmann)
Date: 26 Feb 2002 21:33:43 +0100
Subject: Simple translation from python to C++
References: <mailman.1014728445.18876.python-list@python.org>
Message-ID: <m28z9g0yqg.fsf@pchoel.privat.uni-hamburg.de>

GUYOT Herv? <herve.guyot at thales-avionics.com> writes:

> And naturally, I've been asked to "model" this poorly designed code
> into Rhapsody in order to try and have a better understanding of
> what goes on, an if possible, refactor it. My first step is to
> define the classes (with methods and attributes declarations) and
> inheritance relationships. Not difficult.
> 
> But instead of dooing it by hand, I've thought of extracting from the
> python code C++ class, methods and attributes declarations, and then,
> feed the retro-engeneering tool of Rhapsody with this "skeletton"
> C++ code.

Search for pyreverse. This program generates a XMI representation. XMI
is a XML representation of UML diagrams understood by many UML tools
(at least two I know of: PoseidonUML and Rational Rose)

Greetings

Berthold
-- 
bhoel at web.de / http://starship.python.net/crew/bhoel/
        It is unlawful to use this email address for unsolicited ads
        (USC Title 47 Sec.227). I will assess a US$500 charge for
        reviewing and deleting each unsolicited ad.


From mwh at python.net  Tue Feb  5 05:22:19 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 10:22:19 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de>
Message-ID: <u4rkwql8p.fsf@python.net>

Simon Budig <simon at ping-pong.budig> writes:

> Michael Hudson wrote:
> I just downloaded readline-2.2.tar.gz. The documention does not
> mention this variable but readline.h has:
> 
> /* Non-zero means to suppress normal filename completion after the
>    user-specified completion function has been called. */
> extern int rl_attempted_completion_over;
> 
> I guess, this counts as a "yes" - missing documentation is nothing
> unusual...   :-)

No.

> >> Unfortunately I don't have an account at sourceforge to submit this
> >> the proper way.
> > 
> > Why not?
> 
> It is way too much hazzle to maintain *another* account/password to
> be able to give three or four bugreports to one or two (admittedly
> important) projects.

Well, if you want to contribute to Python, a sf account is a must.

> Ok, you can also say that I am too lazy.   ;-)

That's what I did when I submitted it:

https://sourceforge.net/tracker/index.php?func=detail&aid=513235&group_id=5470&atid=305470

:) you might want to type your email address into the "monitor" field.

Cheers,
M.

-- 
 Have you considered downgrading your arrogance to a reasonable level?
        -- Erik Naggum, comp.lang.lisp, to yet another C++-using troll


From tdelaney at avaya.com  Mon Feb  4 17:51:04 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 5 Feb 2002 09:51:04 +1100 
Subject: Jython question concerning actionPerformed
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A164@natasha.auslabs.avaya.com>

> From: mishre at hushmail.com [mailto:mishre at hushmail.com]
> 
> I have the following code in Jython running on JDK 1.4, which looks
> like it should work, but isn't:
> 
> def actionPerformed(self,event):
>     if event.getActionCommand() == "Exit":
>         java.lang.System.exit(0)
>     elif event.getActionCommand() == "comboBoxChanged":
>         print event
> 
> Then later on I have:
> 
> self.catCombo = 
> javax.swing.JComboBox(actionPerformed=self.actionPerformed)
> 
> I'm trying to update a JList when the selected item in the JComboBox
> is changed.  If I use the code above I get the following error when I
> run the program:
> 
> "TypeError: can't assign to this attribute in java instance:
> actionPerformed"

There are particular problems with JComboBox and actionPerformed.

http://aspn.activestate.com/ASPN/Mail/Message/Jython-users/541301

Tim Delaney



From kalafutj1 at yahoo.com  Sun Feb 24 14:44:27 2002
From: kalafutj1 at yahoo.com (Jim Kalafut)
Date: Sun, 24 Feb 2002 19:44:27 GMT
Subject: .exe files
References: <jpae8.66140$X64.23604369@news1.rdc2.pa.home.com>
Message-ID: <vobe8.17166$dK1.47161@rwcrnsc54>

> related to Python installed.  Is there anyway to turn a .py file into a
.exe
> file that can be executable on Windows?  It doesn't call from any other
> modules, it's completely stand alone.  Thank you for any information!

This works well:

http://www.mcmillan-inc.com/install1.html

Regards,
Jim




From phr-n2002a at nightsong.com  Wed Feb 20 02:20:08 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 23:20:08 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com> <7xu1scew8m.fsf@ruckus.brouhaha.com> <slrna76g9l.rag.yhatakka@kenny.evolvit.fi>
Message-ID: <7xeljg7h7b.fsf@ruckus.brouhaha.com>

yhatakka at dlc.nospam.fi (Yrj? Hatakka) writes:
> >> def traverse(list):
> >>     if list:
> >>         print list[0]  # or do something here
> >>         traverse(list[1:])
> >>     else:
> >>         print "done"
> >> 
> > 
> > That's ok in Scheme but I don't think any current Python implementations
> > handle tail recursion "correctly".
> 
> What do you mean? I tried above an it seems to work. When does it stop
> working?

By "correctly" I mean the code should be executed without pushing
another recursive call onto the call stack for every list element.
The recursive call should instead be compiled as a jump back to the
top of the function.  I don't remember seeing it documented anywhere
that Python does that.  Even if the current implementation happens to
transform tail recursion like that, it's an implementation-specific
optimization rather than part of the language definition like it is in
Scheme.


From support at internetdiscovery.com  Fri Feb 15 22:58:59 2002
From: support at internetdiscovery.com (Mike Clarkson)
Date: Sat, 16 Feb 2002 03:58:59 GMT
Subject: Is making Python executables a reality?
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>
Message-ID: <3c6d6e9c.3472897@test.news.rcn.net>

On Wed, 13 Feb 2002 16:55:54 -0500, "MDK" <mdk at mdk.com> wrote:

>I have tried several tools to make a Python exe, i.e. py2exe which just
>gives me errors.  It looks as though changes were made to how it is supposed
>to work but now it does not work for me.
>
>The programs that are supposed to come bundled with Python for this task do
>not seem to exist on my PC after the Python install.
>
>What percentage of people make exe files?  Do most people have end-users
>download the Python environment?
>
>What is everybody doing in this regard?

For an example of using the standard Python freeze.py under Linux and
Windows, look at the IDE Studio distribution:

	http://starship.python.net/crew/mike/

It's tricky to set up for Tkinter, but works well when you're done.

Mike.


From runyaga at noeggsorspam.runyaga.com  Mon Feb 11 17:59:12 2002
From: runyaga at noeggsorspam.runyaga.com (Alan Runyan)
Date: Mon, 11 Feb 2002 22:59:12 GMT
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader>
Message-ID: <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com>

>     mystring=u.read()
>
> works fine, but sometimes it just hangs. Is this simply because the remote
> server is not responding? I would have thought that would cause
> urllib.urlopen() to hang, not u.read().

Andrew, what version of python are you running?  A friend of mine who I am
trying to convert to Python ran into this exact problem.
He was trying to do a HTTP POST to a web page, which was assigning him
cookies and redirecting him (the Real World ;).  urllib
doesnt handle this very well at all ;'(.  he reported to me urlopen() was
hanging so I gave it a go.  I'm using Python 2.1.2 and I could not
reproduce this.

So.. what I attempted was to re-write what he assumed urlopen() would do for
him.  and now I am stuck.  I'm not quite sure how
cookies and redirect work together.  I know urllib2 kinda gives you some
more options, but this is *very* unintuitive I believe.  we really need
examples.  here is my code if someone could take a look at it and see what I
am trying to do I would greatly appreciate it.

-- snip! --
import urllib2, urllib, urlparse
from urllib2 import Request
import httplib

DEBUG = 1

class CookieHTTPRedirectHandler(urllib2.HTTPRedirectHandler,
urllib2.HTTPHandler):
    def http_error_302(self, req, fp, code, msg, headers):
        if DEBUG:
            print 'was going to ' + req._Request__original +
str(req.headers)

        import pdb; pdb.set_trace()

        if headers.has_key('location'):
            newurl = headers['location']
        elif headers.has_key('uri'):
            newurl = headers['uri']
        else:
            print 'returning'
            return
        newurl = urlparse.urljoin(req.get_full_url(), newurl)

        # XXX Probably want to forget about the state of the current
        # request, although that might interact poorly with other
        # handlers that also use handler-specific request attributes

        response_headers={}
        for head in headers.headers:
            cookie='Set-Cookie:'
            if head[:len(cookie)]==cookie:
                response_headers[cookie]=head[len(cookie)+1:]
        print 'redirect headers ' + str(response_headers)
        new = Request(newurl, req.get_data(), response_headers)

        if DEBUG:
            print 'redirected to ' + new._Request__original

        new.error_302_dict = {}
        if hasattr(req, 'error_302_dict'):
            if len(req.error_302_dict)>10 or \
               req.error_302_dict.has_key(newurl):
                raise HTTPError(req.get_full_url(), code,
                                self.inf_msg + msg, headers, fp)
            new.error_302_dict.update(req.error_302_dict)
        new.error_302_dict[newurl] = newurl

        # Don't close the fp until we are sure that we won't use it
        # with HTTPError.
        fp.read()
        fp.close()
        print 'returning : ' + str(new.headers)
        return self.parent.open(new)

    def http_open(self, req):
        return self.do_open(httplib.HTTP, req)

class HTTPConnection:
    def __init__(self, url, request_data, headers):
        self._request=urllib2.Request(url, urllib.urlencode(request_data),
{})
        self._director=urllib2.OpenerDirector()
        self._director.add_handler(CookieHTTPRedirectHandler())
        self._conn=self._director.open(self._request)

if __name__=='__main__':
    url='http://www.winemag.com/buyingGuide/login.asp'
    req_data={'LoginID':'wine',
              'LoginPassword':'enthusiast',
              'Submit':'Login' }
    winemag=HTTPConnection(url, req_data, {})





From bldrake1 at yahoo.com  Mon Feb  4 01:35:29 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 3 Feb 2002 22:35:29 -0800
Subject: Finding all of a classes methods?
References: <roy-CA3285.20323003022002@news1.panix.com>
Message-ID: <fb072181.0202032235.716279cf@posting.google.com>

Roy Smith <roy at panix.com> wrote in message news:<roy-CA3285.20323003022002 at news1.panix.com>...
> Is there any way to find all of a class's methods?  I know I can use dir(), 
> but that gives you the names of the methods (as strings), not the methods 
> themselves (as callable MethodType objects).
> 
> I'm building a production system.  Basicly, I want to run every one of a 
> class's methods on an object.  The number of methods is large, and will 
> grow in the future.  Right now, I'm building a list of the methods as I 
> define them.  Something like:
> 
> methods = []
> 
> def foo (self, x):
>    pass
> methods.append (foo)
> 
> def bar (self, x):
>    pass
> methods.append (bar)
> 
> and so on down the list (yes, the real ones do something more useful than 
> pass).  Obviously, the class is keeping track of its methods already.  If I 
> could get at that list, I could avoid having to build my own.
> 
> Also, I'm slightly mystified about what, exactly, I'm adding to my list.  
> When I do:
> 
> for method in self.methods:
>    method (x)
> 
> I get an error saying that the method expected 2 arguments, but only got 1.  
> Obviously, it's not getting self because I'm not calling it as a method of 
> self.  If I change it to:
> 
> for method in self.methods:
>    method (self, x)
> 
> I get what I want, but I'm not 100% sure that's kosher.

Roy,
PyDoc.py located in the Lib subdirectory does just what your asking
using inspect.py as mentioned in another reply to your question.  By
using PyDoc.py as an example you should be able to extract what you
need and modify to your specs.  Giving appropriate credit, of course.

Regards,
Barry


From mattams at yahoo.co.uk  Wed Feb 27 06:27:13 2002
From: mattams at yahoo.co.uk (Matt van Amsterdam)
Date: 27 Feb 2002 03:27:13 -0800
Subject: Python and Daemons
References: <02022215115601.01886@localhost.localdomain> <mailman.1014416559.28231.python-list@python.org>
Message-ID: <930f999d.0202270327.78fe1fdd@posting.google.com>

Justin Sheehy <justin at iago.org> wrote in message news:<mailman.1014416559.28231.python-list at python.org>...
> [ ... ]
>     os.umask(0)
> [ .. ]
> -Justin

Why umask 0?


From emile at fenx.com  Thu Feb 28 10:51:35 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 07:51:35 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl>
Message-ID: <a5lk0a$8enfq$1@ID-11957.news.dfncis.de>

"Remco Gerlich"
> Compare:
>
> for i in 6: print i
> for i in 6,7: print i
> for i in 6,7,8: print i
>
> You wouldn't think the first line does something different, would you?

Is that really any different from what we have now:

for i in "Spam": print i
for i in "Spam","and": print i
for i in "Spam","and", "Eggs": print i

I find the best argument for adding an iter method to ints is that it
allows me to say:

    for ii in iter(len(mylist)):

which I find _much_ less confusing than saying:

    for ii in range(len(mylist)).

The first reads as:

    'for ii iterating over the len of mylist'

whereas the second:

    'for ii in the list of numbers starting with 0 to 1 less than the
len of mylist'


Of course,  this can be done now by simply(!!?) shadowing iter with:
iter = range.

If this PEP is added, the coincidental effect of ints invoking their
iterators in appropriate contexts is, in my mind, no different than what
we do today with strings.

However, the clarity and intent conveyed by allowing this, particularly
for newbies, more than offsets the gotcha's, which, once learned, is the
same set of gotcha's in place for strings.

Learn-each-idiom-only-once-ly y'rs

--

Emile van Sebille
emile at fenx.com

---------





From aleax at aleax.it  Mon Feb 18 05:39:37 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 10:39:37 GMT
Subject: using lists in C-extensions?
References: <3c6e711a.9871609@news.easynews.net>
Message-ID: <JR4c8.68439$om6.1888911@news1.tin.it>

Marcus Stojek wrote:
        ...
> EXAMPLE:
> 
> list=[]
> for i in range(100000):
>     list.append(i)

Note in passing that this is equivalent to
    list = range(100000)
since range does return a list.  If it didn't (if it returned
some other kind of sequence or iterator), then you could
ALMOST use:
    list = list(funkyvariantofrange(100000))
EXCEPT that by calling this local variable 'list' you've
shadowed the builtin type [function in 2.1 and earlier] of
the same name [if this code is within a function] -- so,
don't: call the variable alist or thelist or mylist instead.

Peripheral issues, but sometimes they DO give headaches.


> listnew=map((lambda x:x+1),list)  # this should be coded in C

You won't get all that much gain by recoding this in C, since
map is already C-coded and well-optimized and fetching and
setting the list elements is substantially the same amount of
work.  However, if you can start with an array.array (or, of
course, a Numeric array, as Jason suggested), then potential
for performance is indeed there.

Simplest, assuming you do want to work with integers:

import array
anarray = array.array('l', alist)

import mycode
mycode.incr(anarray)

In a directory called e.g. ~/mycode put two text files, one
called setup.py with:

from distutils.core import setup, Extension

setup (name = "mycode",
       version = "1.0",
       maintainer = "Alex Martelli",
       maintainer_email = "aleax at aleax.it",
       description = "Sample use of array",

       ext_modules = [Extension('mycode',sources=['mycode.c'])]

and one called mycode.c with:

#include "Python.h"

static PyObject*
incr(PyObject *self, PyObject *args)
{
    PyObject *the_array;
    long* pbuf;
    int blen, i;

    if(!PyArg_ParseTuple(args, "O", &the_array))
        return 0;
    if(PyObject_AsWriteBuffer(the_array, (void**)&pbuf, &blen))
        return 0;
    blen /= sizeof(long);
    for(i=0; i<blen; ++i)
        pbuf[i] += 1;

    return Py_BuildValue("");
}

static PyMethodDef mycode_methods[] = {
    {"incr",   incr,   METH_VARARGS},
    {0, 0}
};

void
initmycode(void)
{
    Py_InitModule("mycode", mycode_methods);
}

Run in this directory:
    python setup.py install

and the above sample Python snippet should work.  Of course,
you may wish to take more precautions in function incr, which,
as it stands, would produce funny results should you call
it e.g. with an array.array of typecode other than 'l' or
'L'.  But don't go too far in "precautions" (typetests) that
may in the future interfere with the ability of using this
code with objects of other types.  For example, you might
want to check PyObject_GetAttrString(the_array, "typecode"),
that it works and returns "l" or "L": this makes it easy in
the future to emulate the two important aspects of an
array.array's behavior (the typecode attribute and the
fact that it has a single contiguous writable buffer) by
another type should the need arise.  You may also want to
support multiple typecodes, such as "l" or "L" for longs
(signed/unsigned), "f" for floats, "d" for doubles, to have
incr transparently work on numeric arrays of different
types of numbers (you'll need to dispatch e.g. with a
case statement in your C code between different loops of
operations in this case, of course).


Alex



From phd at phd.pp.ru  Tue Feb 12 09:34:02 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 12 Feb 2002 17:34:02 +0300
Subject: PyFT and PIL
In-Reply-To: <oqu1sm4vqr.fsf@titan.progiciels-bpi.ca>; from pinard@iro.umontreal.ca on Tue, Feb 12, 2002 at 09:31:40AM -0500
References: <mailman.987996754.9967.python-list@python.org> <9c11bn$sph@gap.cco.caltech.edu> <oqu1sm4vqr.fsf@titan.progiciels-bpi.ca>
Message-ID: <20020212173402.G8503@phd.pp.ru>

On Tue, Feb 12, 2002 at 09:31:40AM -0500, Fran?ois Pinard wrote:
> PyFT seems to be a wonderful thing.  But the `README' file roughly says:
> 
>    `Compilation of the extension is standard.'
> 
> The only standard I know is `Distutils'.

   There was older Python standard using Setup.in and Makefile.pre.in
files:

   make -f Makefile.pre.in boot
   make

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



From bergeston at yahoo.fr  Mon Feb  4 20:04:32 2002
From: bergeston at yahoo.fr (Bertrand Geston)
Date: Tue, 5 Feb 2002 02:04:32 +0100
Subject: Drag and drop from one PC to the other
Message-ID: <a3nbg4$1jgs$1@rivage.news.be.easynet.net>

Hello,

I would like to know your opinion about this: I have 2 Windows computers one
next to the other and I would like to be able to pass object from one to the
other really easily. A kind of window open on each computer could be handy
if the user can paste (by Cut 'n Paste or by Drag 'n Drop) an object in one
window and retrieve it from the other computer window. In other words, I
would like to have a shared and distributed clipboard.

Does it exist already ? Is it feaseable and easy ? I really don't know COM
or COM+ but I guess that this could be part of the solution. Any opinion or
advice is really welcome (just to know how to start or where to look for).

Thanks in advance and have good time.

B.




From mcherm at destiny.com  Thu Feb  7 16:14:50 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Thu, 07 Feb 2002 16:14:50 -0500
Subject: Defending the Python lanuage...
Message-ID: <3C62EE4A.5090305@destiny.com>

>> Really? What an unusual idea - I didn't realise we had such a surplus of
>> programmers to advocate something like this :-)
> 
> This is a little more "extreme" than I was referring to as well.  The idea
> of two programmers sitting at the same computer doesn't really go over well
> with me... I was referring having two programmers sit side-by-side at
> separate computers, working on different parts of the same program versus
> having one programmer work in isolation.  Indeed, I don't know many
> programmers with the patience to sit and watch someone else type ;) 


Ermmm.... No.

When XP or Agile Programming advocates talk about "pair programming" 
they really DO mean one keyboard, two programmers.

Peter Milliken writes that he didn't know we had such a surplus of 
programmers -- this is a common misconception. Suppose that, working on 
your own, you get X amount of coding done in a day. If we could spare a 
second programmer to stand there and just help you out -- someone to 
bounce ideas off of, to watch for typos, to suggest things that you had 
forgotten -- then you'd get MORE done... say X * k.

If k is 1.05, then you are getting a 5% speedup by having someone around 
to help out. If k were 3, then you'd get THREE TIMES as much done. If 
THIS were the case, then it'd be worth it to your company to tell 
someone else to stop working on their own and just help you out! The 
break-even point is somewhere around k=2.

The exact value of k depends on LOTS of things, including the 
personalities of the programmers involved, the kind of work involved, 
etc, but I have heard several different annecdotal estimates that place 
k at just-less-than 2. (Maybe k=1.9 +/- 0.2 for typical situations.)

If k is close to 2, then there it might NOT be a stupid idea to put 
programmers in pairs, even without a surplus. Particularly because we 
expect some additional benefits in addition to the coding speed, like 
fewer bugs and better sharing of the knowledge among the coding team.

As for Cliff Wells' contention that he doesn't know many programmers who 
would want to do this, I have to differ. There are clearly some 
programmers who absolutely hate working in pairs, but there are also 
many others who LOVE it.

Now I just wish I could convince my company to consider doing pair 
programming. Because for *ME* personally, k is around 3 or 4.

-- Michael Chermside








From jsc-news at bluesine.com  Mon Feb 11 13:10:25 2002
From: jsc-news at bluesine.com (Jeff Clement)
Date: Mon, 11 Feb 2002 11:10:25 -0700
Subject: Global Variables in Modules
Message-ID: <20020211111025.63f4c956.jsc-news@bluesine.com>

Good Afternoon All,

I have a project where it would be extremely handy (albeit maybe not so clean codeing), to have a global variable in one module be available to another module (without a lot of work.  I know I can do module.varname = varname from the importing module but I would prefer not to).

Below is my sample program.  I would like to launch test.py and declare some global variable A, then import test_mod and hopefully get access to A from within test_mod.  

What I'm actually doing is running a code block in an exec block and declaring a bunch of globals for that exec.  The code block imports some modules and I need those to be able to see the global variables.  I can make this work by adding them to __builtins__ but that doesn't seem right somehow :)

Anyone have any ideas?

Thanks,
Jeff

---- test program -----------------------------------------

$ python test.py

test.py
-------
A=100
import test_mod

test_mod.py
-----------
print A


From joshm at taconic.net  Mon Feb 11 22:05:57 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 11 Feb 2002 22:05:57 -0500
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <a49s16$5ao$1@newshost.accu.uu.nl> <pan.2002.02.11.20.34.30.302944.2302@uchicago.edu>
Message-ID: <3c688449_1@corp.newsgroups.com>

> It's a poor conspiracy that censors the messages denying its existence.

All the more reason to doubt its existence.  Or *is* it?

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('No autocoders were harmed in the making of this .sig')




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


From jason at jorendorff.com  Tue Feb 19 00:54:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 23:54:57 -0600
Subject: Adding static typing to Python
In-Reply-To: <0DFA7478F1E5D311B90900508B1208CF565712@ausmail02.mis.gtech.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEKCDDAA.jason@jorendorff.com>

Milliken, Peter wrote:
> Strong typing requires that you design before you code (Extreme 
> Programmers beware! :-)) - therefore by the time a clean compile
> is achieved 95% of the bugs *are* out :-) That is one of the
> primary advantages to strong typing, it *forces* you to do the
> design up front otherwise you will never get a clean compile.

Hmm.  Perhaps you can explain this in more detail.

Are you arguing that compilers for strongly typed languages tend to
reject ill-designed or thrown-together code?  I hadn't noticed.

Surely you're not arguing that extra-productive programming tools
are bad because they encourage the programmer to waste his or her
own time by skipping the design step.

Maybe this is true in some sense - since I enjoy coding Python so
much, perhaps I'm overeager to get started?  But if so, then the
solution is sane development practices and a touch of
self-discipline.  Which will likely come in handy on any ten-year
project.


In any case, it's not nice to pick on XP and then say:

> [...] I have probably got the idea of Extreme
> Programming wrong - I really don't care, from what I have seen
> described the style doesn't suite me :-)

Blah.  :(

## Jason Orendorff    http://www.jorendorff.com/



From achim.domma at syynx.de  Sat Feb  2 10:07:38 2002
From: achim.domma at syynx.de (Achim Domma)
Date: Sat, 2 Feb 2002 16:07:38 +0100
Subject: win32api.RegQueryValue fails
Message-ID: <a3gv20$fv8$05$1@news.t-online.com>

Hi,

still not having solved my problem importing win32com in an embedded
interpreter , I tried to debug where the error comes from. My real problem
is that win32api.RegQueryValue fails.

The following code works without problems:

import win32api
print win32api.GetFullPathName(".")

but

import win32api
keyName = "SOFTWARE\\someKey"
key = win32api.RegOpenKey(win32con.HKEY_LOCAL_MACHINE , keyName, 0,
win32con.KEY_ALL_ACCESS)
win32api.RegQueryValue(key, "Root")

fails with a 'TypeError: this constructor takes no arguments'. If I print
the handle returned by RegOpenKey it is not Null, so it seems to be ok. A
look at the implementation of RegQueryValue (head-revision in cvs) showed
that it is implemented straight forward and I have no idea what should go
wrong.

I have to use ADO from my embedded python, which will not work without
win32com, so every little hint is welcome.

greetings
Achim





From aahz at panix.com  Fri Feb  8 10:39:16 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 07:39:16 -0800
Subject: 2001 Turing Award
References: <mailman.1013109679.5469.python-list@python.org> <2jnu3a.4sq.ln@wombat.san-francisco.ca.us> <slrna663pr.d87.philh@comuno.freeserve.co.uk>
Message-ID: <a40rf4$9im$1@panix2.panix.com>

In article <slrna663pr.d87.philh at comuno.freeserve.co.uk>,
phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Thu, 7 Feb 2002 12:20:50 -0800, Keith Keller <kkeller at speakeasy.net> wrote:
>>In article <mailman.1013109679.5469.python-list at python.org>,
>>	James_Althoff at i2.com writes:
>>> 
>>> I'm happy to pass on the news that the 2001 Turing Award, the highest
>>> award in computer science
>>
>>Damn--I thought the award was for the person who best impersonated
>>a Turing machine.  :)
>
>But only using 9 commands.

I can impersonate that Turing machine in 7 commands.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From lsmithso at NOhare.SPAM.demon.co.uk  Fri Feb  8 16:17:07 2002
From: lsmithso at NOhare.SPAM.demon.co.uk (Les Smithson)
Date: 08 Feb 2002 21:17:07 +0000
Subject: threads killing other threads
References: <1013105970.135322@cswreg.cos.agilent.com> <a40h4g$kl5$1@newsreader.mailgate.org> <a40rq0$ajl$1@panix2.panix.com>
Message-ID: <a41fsq$bvf$1@newsreader.mailgate.org>

>>>>> "Aahz" == Aahz Maruch <aahz at panix.com> writes:

    Aahz> In article <a40h4g$kl5$1 at newsreader.mailgate.org>,
    Aahz> Les Smithson  <lsmithso at NOhare.SPAM.demon.co.uk> wrote:
    >>  I'm sure there's a good reason for Python not implementing
    >> this. Anyone know what it is?

    Aahz> Yes.  It doesn't work.

    Aahz> Oh, sure, there's a long technical explanation of how
    Aahz> arbitrarily killing threads leaves processes in various
    Aahz> kinds of weird states, but it all boils down to, "It doesn't
    Aahz> work."  That goes double for any kind of cross-platform
    Aahz> facility.  Note that Java has pretty much abandoned any
    Aahz> pretence that it can provide this capability.  -- --- Aahz
    Aahz> <*> (Copyright 2002 by aahz at pobox.com)


But doesn't pthread_testcancel allow cooperative threads to be
canceled? This is the portable equivalent to the thread signaling
scheme described earlier. It also gives the canceled thread a chance
to clean itself up before exiting.

Not ideal, but better than nothing.


From aries at bitstream.net  Wed Feb 20 17:41:08 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Wed, 20 Feb 2002 16:41:08 -0600
Subject: help - inserting values into database tables
References: <mailman.1014086793.5272.python-list@python.org>
Message-ID: <3c742468$0$79564$6536d21b@news.bitstream.net>

Hey chajadan,
Well I tried clicking on the ODBC icon in  control panel, I get the error
saying
I need to reinstall the ODBC Components......what a bummer.
I don't have the install CD for windows 98.
Boy, if I ever get it going I'm going to give it full thrusters and
afterburners !!!
Cause I know now it can be done.

I did however setup the ODBC on a windows 95 system, and it appears that
I set it up correctly.  I gave it a "System DSN"  to the Manufacturers.mdb
...but
running the script I get an error something about "system 7.0....  in Login"
I set the "System DSN" up with no user ID or password so I wouldn't have
to worry about that in my script file.  But it seems it gives that Login
error.
I wonder if I need to put SHARE.EXE thing in my autoexec.bat file ???

I could send you the whole error message if you want ???  on the win95 sys.
ON both systems neither one has ms-access installed just so you know.
But I did put the small database files *.mdb in the directory that contains
my script
files.

Well I will chat with ya later.....Sega,
Joe F.

chajadan <python at chajadan.net> wrote in message
news:mailman.1014086793.5272.python-list at python.org...
> Hi Joe =)
>
> Your problem shouldn't be too hard to fix.
>
> Using odbc, you can't open the database file directly like that, since
odbc
> supports so many different database standards, it needs to know which one.
>
> This requires a DNS. In order to set up a DNS for MS Access, just go to
> your control panel, and you should have an option called "OBDC
Datasources"
> or similar. Click that, and it will show you your current DSNs. There are
> several types of DSNs, but you will want to make this file a System DSN,
so
> click that tab. Select New, and choose the correct database driver, in
this
> case MS Access. Then you have to name the DSN, anything you'd like (avoid
> spaces), chose a login/password or leave that blank, and finally show it
> the database file this whole DSN should be referring to.
>
> Then, you can odbc.odbc("myDsnName")
>
> =)
>
> Let me know if you have any other questions and I'll try to help. I work
on
> Win98, so some of what I described may look slightly different under XP,
> but it should be there.
>
> Good luck!
>
> --chajadan
> aka charlie
>
> At 06:35 PM 2/18/02 -0600, you wrote:
>
> >Hi again,
> >Just wondering if you can help?
> >
> >I installed the python ver 2.0(from CD)for windowsX on a windows98
> >system.  Then I installed the windows32all stuff.
> >So are the dbi and the odbc modules should now be installed
> >on my system now, or do I need some other package ??????
> >
> >It seems I can import the dbi then the odbc with no errors...
> >But my simple script below won't run; has an error that I
> >don't understand ???????
> >
> >I am running from the command line, heres my environ stuff
> >******* Do I need to put in another path to Lib ?????
> >
> >COMSPEC=C:\WINDOWS\COMMAND.COM
> >windir=C:\WINDOWS
> >PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\JDK11~1.8\BIN;C:\PYTHON20;
> >
> >******** Heres the error I get upon running from prompt
> >
> >C:\codepy>python conntest.py
> >
> >Traceback (most recent call last):
> >   File "conntest.py", line 8, in ?
> >     conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
> >dbi.operation-error: [Microsoft][ODBC Driver Manager] Data source name
not
> >found and no default driver specified in LOGIN
> >
> >The Manufacturers.mdb is the ms-access database of ver office XP
> >2002.  It is just a small database with a small table in it.
> >Very simple kind.  The file is in that codepy directory also.
> >So I don't understand ?????????????????
> >What is DSN, does it allow for a ms-access file??????????
> >
> >Well anyways heres my test database connection script, I am
> >just trying to connect and print some output from the database.
> >I think I am doing this correct but not sure ??????????
> >
> >The script:  conntest.py listing
> >#! /usr/bin/env python    #not needed for windowspython
> >
> >#test script to connect to a ms-access database
> >
> >import dbi
> >import odbc
> >
> >conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
> >rs = conn.cursor()
> >rs.execute("select * from ManufName")
> >print rs.description
> >
> >rs.close
> >conn.close
> >
> >Can you shine any light my way ???????????
> >
> >Thank you (working on a windows system today(UGH...)),
> >Joe F.
> >#####
> >
> >----- Original Message -----
> >From: chajadan <python at chajadan.net>
> >Newsgroups: comp.lang.python
> >Sent: Monday, February 18, 2002 2:34 PM
> >Subject: Re: help - inserting values into database tables
> >
> >
> > > I always use PythonWin, part of the win32all distribution. It's a
> > > color-coded ide, with a basic syntax check, and has one nifty
advantage
> > > over any other input editor I've tried: when you call a function, it
takes
> > > the document strings and actually gives you a little note about what
kind
> > > of input it's expecting!
> > >
> > > So all in all it should be the same thing as it ever was, but now in
the
> > > environment your particular task seems to ask for. The only "new"
things
> > > will be the other possible windows-only features you might want to
learn
> > > and use in your time on windows.
> > >
> > > Happy computing!
> > >
> > > --chajadan
> > >
> > > At 08:15 AM 2/18/02 -0600, mylinuxboxroot wrote:
> > > >Hey thanks,
> > > >This is great, and I have the project now that it will be developed
on a
> > > >windows
> > > >server so I guess I don't need to worry about going from linux to the
> > > >MS-Access
> > > >database.  I will develop it right on the server using the win32all
> > > >distribution that
> >
> >snip
>
>




From grante at visi.com  Wed Feb 20 22:05:49 2002
From: grante at visi.com (Grant Edwards)
Date: Thu, 21 Feb 2002 03:05:49 GMT
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com>
Message-ID: <slrna78p0v.1jg.grante@tuxtop.visi.com>

In article <e9e6dd1f.0202201727.9741615 at posting.google.com>, Jason wrote:

> In my experimentation 0, None, "", [], and () all seem to be
> considered false.

Yup.  It seems odd, but once you get the hang of it, it's *so*
useful.

> Just to confound me none of them are == to each other.

Likewise, all non-false things need not be == to each other.

> Coming from a Scheme world this is confusing.  Are there any
> other values that are false?  Why aren't () and "" the same thing?

Because they're not.  ;)

One is a string, the other is a tuple.  They are objects of
different types.

> Don't both describe an immutable sequence of values (basically, why
> aren't strings either lists or tuples)?

Mostly because it is both useful that way and I suspect easier
to impliment.  Two of the interesting features of strings is
that strings are always flat (they don't nest), and for
non-empty strings type(s) == type(s[0]).

-- 
Grant Edwards                   grante             Yow!  Clear the
                                  at               laundromat!! This
                               visi.com            whirl-o-matic just had a
                                                   nuclear meltdown!!


From syring at email.com  Tue Feb 12 09:29:52 2002
From: syring at email.com (Karl M. Syring)
Date: Tue, 12 Feb 2002 15:29:52 +0100
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de> <mailman.1013521581.26252.python-list@python.org>
Message-ID: <a4b8tr$1dq02n$1@ID-7529.news.dfncis.de>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> schrieb im
Newsbeitrag news:mailman.1013521581.26252.python-list at python.org...
> ----- Original Message -----
> From: "Karl M. Syring" <syring at email.com>
>
>
> > "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> schrieb
> > >
> > > The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> > > have to be such a pain to find?)  I can instantiate it:
> > >
> > > >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> > > >>>
> > >
> > > (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
> > >
> > > >>> o.AboutBox()
> > > Traceback (most recent call last):
> > >   File "<interactive input>", line 1, in ?
> > >   File
"win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py",
> > line
> > > 29, in AboutBox
> > >     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0),
(),)
> > > com_error: (-2147418113, 'Unexpected failure', None, None)
> > > >>>
> >
> > Looks like it is missing a DLL, probably one of MFC's (If it is on a
bare
> > bones Win95, there are more interesting modes of failure). Can't you
> > inspect, what your OCX is trying to import?
>
> Ur.  What's the Win32 equivalent of ldd (dumps linked dll's)?  I ran
strings

The canonical command is
Dumpbin /imports

> on it and filtered out the .dll files (case-insensitive match) and came up
> with this:
>
> MFC42.DLL    <- got that
> MSVCRT.dll   <- got that
> KERNEL32.dll <- got that (system wouldn't run without it, right?)
> USER32.dll   <- ditto
> GDI32.dll    <- ditto
> OLEAUT32.dll <- got that
> What might I be missing?
>
> I tried some other ActiveX controls with reasonable success, but a few
> showed
> the same error message.

Copying dlls is not recommended, because of possible additional
dependencies. There is a VC runtime installer called vcredist.exe that you
can scrap from msdn.microsoft.com.
If the going really gets tough, especially with old Win9x systems, I always
recommend to install IE 5.5, even if you never use it.

Karl M. Syring
Karl M. Syring



From cliechti at gmx.net  Wed Feb 20 17:27:41 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 20 Feb 2002 23:27:41 +0100
Subject: HTMLifyer for Python code?
References: <mailman.1014238715.7462.python-list@python.org>
Message-ID: <Xns91BBEF384B311cliechtigmxnet@62.2.16.82>

"Bjorn Pettersen" <BPettersen at NAREX.com> wrote in 
news:mailman.1014238715.7462.python-list at python.org:
> tool for converting Python source code to syntax-highlighted HTML. 

the SciTE editor which you can finde here:
http://www.scintilla.org/

has a save as HTML option in the filemenu. (that editor component is 
used in PythonWin and wxWindows)

chris

-- 
Chris <cliechti at gmx.net>



From asim_beg at hotmail.com  Tue Feb 19 21:33:46 2002
From: asim_beg at hotmail.com (Asim Beg)
Date: 19 Feb 2002 18:33:46 -0800
Subject: MySQLdb build on Solaris 7
Message-ID: <d1dba0f2.0202191833.757c52c6@posting.google.com>

My ISP has Python 2.1 installed and I can run Python CGI programs
fine. I now need to connect to MySQL. I need to install MySQLdb and am
having difficulty getting the MySQLdb source to build. I have tried
both MySQL-python-0.9.0 and MySQL-python-0.9.1.

The machine is a solaris 7 with Python 2.1 on it. When I try to build
it starts building and then goes into an infinite loop with strange
libmysqlclient.a error messages..

If you are going to help, please be specfic and don't rant about the
do's and don't of why I am doing this. I am new to Python and would
really like to use CGI+Python+MySQL on an ISP hosted account.

Also since I cannot install MySQLdb in the standard places on Solaris,
how should I go about after the build step. Where and which files
should I copy to my home dir toget Python to work with MySQLdb.

Thanks

Asim


bash-2.03$ uname -a
SunOS europa 5.7 Generic_106541-10 sun4u sparc SUNW,Ultra-80

bash-2.03$ python
Python 2.1 (#1, May 25 2001, 16:58:42) 
[GCC 2.95.2 19991024 (release)] on sunos5
Type "copyright", "credits" or "license" for more information.

bash-2.03$ cd MySQL-python-0.9.0
bash-2.03$ python setup.py build
running build
running build_py
not copying CompatMysqldb.py (output up-to-date)
not copying _mysql_exceptions.py (output up-to-date)
not copying MySQLdb/__init__.py (output up-to-date)
not copying MySQLdb/converters.py (output up-to-date)
not copying MySQLdb/connections.py (output up-to-date)
not copying MySQLdb/cursors.py (output up-to-date)
not copying MySQLdb/sets.py (output up-to-date)
not copying MySQLdb/times.py (output up-to-date)
not copying MySQLdb/constants/__init__.py (output up-to-date)
not copying MySQLdb/constants/CR.py (output up-to-date)
not copying MySQLdb/constants/FIELD_TYPE.py (output up-to-date)
not copying MySQLdb/constants/ER.py (output up-to-date)
not copying MySQLdb/constants/FLAG.py (output up-to-date)
not copying MySQLdb/constants/REFRESH.py (output up-to-date)
not copying MySQLdb/constants/CLIENT.py (output up-to-date)
running build_ext
building '_mysql' extension
skipping _mysql.c (build/temp.solaris-2.7-sun4u-2.1/_mysql.o
up-to-date)
gcc -shared build/temp.solaris-2.7-sun4u-2.1/_mysql.o
-L/usr/local/mysql/lib -lmysqlclient -lz -o
build/lib.solaris-2.7-sun4u-2.1/_mysql.so
Text relocation remains                         referenced
    against symbol                  offset      in file
mysql_read_query_result             0x2da4     
/usr/local/mysql/lib/libmysqlclient.a(libmysql.o)
shutdown                            0x20c      
/usr/local/mysql/lib/libmysqlclient.a(violite.o)
printf                              0x1244     
/usr/local/mysql/lib/libmysqlclient.a(charset.o)
printf                              0x428      
/usr/local/mysql/lib/libmysqlclient.a(default.o)
printf                              0x3fc      
/usr/local/mysql/lib/libmysqlclient.a(default.o)
printf                              0x1238     
/usr/local/mysql/lib/libmysqlclient.a(charset.o)
printf                              0x13c8     
/usr/local/mysql/lib/libmysqlclient.a(charset.o)
printf                              0x125c     
/usr/local/mysql/lib/libmysqlclient.a(charset.o)
printf             
...................


From jgardn at alumni.washington.edu  Thu Feb 28 03:11:32 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Thu, 28 Feb 2002 17:11:32 +0900
Subject: regular expression help
References: <mailman.1014853248.711.python-list@python.org>
Message-ID: <bslf8.397$Az6.1785@news.hananet.net>

Drew Fisher wrote:

> Hello,
> 
> I'm having some issues doing a small but irritating regular expression
> match.
> 
> Here's an example of text that I want to search through:
> 
> str = """
> Location:Home
> 555-1212
> 555-3434
> 555-5656
> 
> Location:Work
> 555-9999
> """
> 
> Just for reference, there is newline characters at the end of each line
> and the phone numbers have a single tab character (\t) in front of them.
> 
> I'm trying to get the location of the record and the associated phone
> number(s) that go along with it.
> 
> I've tried things like:
> 
> import re
> test1 = re.compile ('Location:(\w+)\s+(\d+\-\d+)', re.DOTALL)
> print test1.findall (str)
> [('Home', '555-1212'), ('Work', '555-9999')]
> 
> But this only gets the location and the first phone number in the
> record.  How do I get all of the phone numbers for each location without
> writing a second regular expression?
> 

APPROACH 1:
You'll need to do something fun like this:
re.compile('Location:(\w+)$\s((?:\s\d+\-\d+$\s)*)', re.MULTILINE)

Of course, this is a bit strict...

You'll get (with findall):
[('Home', '\t555-1212\n\t555-3434\n\t555-5656\n'),...]

Take the second element in the pair and use split('\n'). Then use strip(). 

APPROACH 2:
Write a basic parser.
loc_re = re.compile('Location:(\w+)')
pnum_re = re.compile('(\d+-\d+)')

locs = {}
last_loc = None
for i in str.split():
    found = loc_re.findall(i)
    if found:
        locs[found[0]] = []
        last_loc = locs[found[0]]
        continue
    found = pnum_re.findall(i)
    if found and last_loc:
        last_loc.append(found[0])

I think the parser approach is a bit easier to maintain, extend, and 
customize.

Jonathan


From jkraska at san.rr.com  Wed Feb  6 23:52:38 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 04:52:38 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net>
Message-ID: <jb246uci04pj6q2l0aqu18ch17kmv8ujof@4ax.com>

> With that in mind, it's not possible for me to convince
> you of anything beyond your own limited two deminsional
> perceptions.

If you ever became delusional, how could you be sure that
it was or wasn't so?

C//



From phr-n2002a at nightsong.com  Tue Feb 12 03:13:22 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 00:13:22 -0800
Subject: how to compare two strings?
Message-ID: <7xofivazj1.fsf@ruckus.brouhaha.com>

Is there any built-in function in Python for comparing two strings
alphabetically?  Something suitable for passing as the optional
comparison function for the list "sort" method.  I could code up
something like

  def scmp(a,b):
    if a<b: return -1
    if b<a: return 1
    return 0

but it's pretty ugly to need to do that.  I've looked in the docs but
don't see any obvious counterpart to perl's lozenge (<=>) operator.
Is there one?

Thanks.


From kp87 at lycos.com  Sun Feb 17 07:38:37 2002
From: kp87 at lycos.com (kevin parks)
Date: 17 Feb 2002 04:38:37 -0800
Subject: help with silly algorhythm
References: <mailman.1013794830.12783.python-list@python.org>
Message-ID: <5e8bd451.0202170438.1867b1b3@posting.google.com>

Thanks to all for your help. I am learning much by looking at this
code. I am always amazed at how many things are built in to Python.
bisect is one of those things that you always forget is there... I am
going to look at what folks posted a play around some. Though I am
really happy too see those lambdas get explained I was about to pop a
vein trying to get my head 'round those things.

I sort of guess you were goofing on me, but the funny thing is that
those functions did mostly work....

Thanks for your help everyone I am amazed that you could all figure
out what i was getting at. I think that I made a mistake or two in my
examples (i did on the fly) but i will play with the code now to see
that it give the results that I am after.

cheers,

-kevin


From usenet at thinkspot.net  Mon Feb 25 19:00:53 2002
From: usenet at thinkspot.net (Sheila King)
Date: Mon, 25 Feb 2002 16:00:53 -0800
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5dia2.3vs4847.1@kserver.org> <3c7abd92_1@corp.newsgroups.com>
Message-ID: <a5dn3m.3vs4847.1@kserver.org>

On Mon, 25 Feb 2002 17:51:36 -0500, "Joshua Muskovitz" <joshm at taconic.net>
wrote in comp.lang.python in article <3c7abd92_1 at corp.newsgroups.com>:

> 
> They are the tech faculty for the local tech school.  It is meant to
> demonstrate my ability to teach (something). 

If this is the case, they are more interested in HOW you present, than in
WHAT you present. Still, I can appreciate you wanting to show off the fun
of Python.

Don't try too much to "wow" them with the topic. Pick something that is
fairly clear to explain. Keep in mind, that you will have to explain the
data structures, etc... before you can even begin to talk about the actual
application you want to show. (At least, that is how I would start. First
with an overview of some Python data types/structures that will occur in
the example you are about to present, and then show the actual example.)

15 minutes is not very much time, either. Keep to something fairly simple.

> Maybe writing something like wc, then perhaps an indexer (using a dict to
> construct a list of references, then producing pretty output).  Hmm...  I
> think I answered my own question.

I'm not catching this reference here. But even something as simple as
grabbing the text from an HTML document, getting only the links, and
listing them, would be fine.

Or how about, testing whether the links are dead or resolve? Generating a
list of dead links in  the web page? This would be fairly simple. Start by
just showing how you would grab all the links and put them into some Python
data structure. Maybe a dictionary, where the URL is the key and it could
have value 'untested', 'dead', and 'valid', or something like that.

Then, after you've constructed the list, go through and test each one to
see whether you get an error response or a good response, updating your
dictionary as you go. Then output the results.

Well, not a "wow", but anyhow, it's an idea. I'm sure we could all come up
with quite a few examples.

Personally, I'm into email examples. Generate and send a live email. (Will
you have a live connection?)

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From lac at strakt.com  Mon Feb 11 10:38:27 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 11 Feb 2002 16:38:27 +0100
Subject: A newbie question about class 
In-Reply-To: Message from "Mark McEahern" <marklists@mceahern.com> 
   of "Sun, 10 Feb 2002 14:27:17 PST." <NCBBLFCOHHDIKCAFGCFBKEADKCAA.marklists@mceahern.com> 
References: <NCBBLFCOHHDIKCAFGCFBKEADKCAA.marklists@mceahern.com> 
Message-ID: <200202111538.g1BFcRGo026664@ratthing-b246.strakt.com>

You can't write 

class doIt(value=something-mutable):

and get the behaviour you want.

Instead, the idiom is

class doIt(value=None):

and then test for None.

If you use == instead of is in your test then it will fail when it is
absolutely fine for what you get passed to redefine __cmp__ or __eq__.
And won't you look silly.

Laura Creighton



From jason at jorendorff.com  Fri Feb  1 00:47:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 31 Jan 2002 23:47:02 -0600
Subject: Conventions for iterators
In-Reply-To: <a3cu29$grp$1@newsreaderm1.core.theplanet.net>
Message-ID: <HFEKILOLEFEFMKAECNDLKEJLCPAA.jason@jorendorff.com>

> So, obviously, behavior 2 is "best."  But I was wondering what the
> python community expects/requires.  Is the idiom such that solution 2 is
> "required," and everyone will expect an object's iterator to work that
> way, or is it more of a caveat emptor situation, and you had better red
> the module's documentation to see how the module handles this situation?

Caveat emptor.  Look at what dictionaries do.

x = {'a': 3, 'b': 9, 'c': 12, 'd': 18}
for k in x:
    if x[k] > 9:
        del x[k]

RuntimeError: dictionary changed size during iteration

## Jason Orendorff    http://www.jorendorff.com/



From jimmy at retzlaff.com  Sat Feb  2 11:43:23 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Sat, 2 Feb 2002 08:43:23 -0800
Subject: How to distribute code with required modules?
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F556D@MAYOR.retzlaff.com>

>> I created a subdirectory of tools called pylib, and
>> check all my class modules into there.  Then I set by
>> PYTHONLIB environment variable to
>> "$HOME/sandbox/src/tools/pylib".

You could turn pylib into a package by placing an empty file inside of
it called '__init__.py'. Then, supposing RoysLib is a module, you could
simply change any occurrence of

import RoysLib

to

import pylib.RoysLib as RoysLib

and everything should work as before. You only need to make this change
in code appearing in "tools" and not to code appearing in "pylib".

Jimmy



From spamtrash at softhome.net  Thu Feb 14 03:51:59 2002
From: spamtrash at softhome.net (Johannes Gijsbers)
Date: 14 Feb 2002 00:51:59 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <slrna6lbrp.guj.mlh@vier.idi.ntnu.no> <d7d5ebdf.0202131429.16a21cca@posting.google.com>
Message-ID: <e87ce949.0202140051.59601ce5@posting.google.com>

matthias_j at lycos.com (Matthias Janes) wrote in message news:<d7d5ebdf.0202131429.16a21cca at posting.google.com>...
> Hi Steve Ferg ;
> 
> I tried the link to get more info but it just displayed page not found.
> 
> Please could you let me know where to get more info for EASYGUI
> 
> Thanks

He forgot a 'y' in the URL. This is the right one:
http://www.ferg.org/easygui/index.html


From jcromwell at ciena.com  Fri Feb  1 13:45:45 2002
From: jcromwell at ciena.com (Cromwell, Jeremy)
Date: Fri, 1 Feb 2002 10:45:45 -0800 
Subject: Comments invited:  PEP 279 Enhanced Generators
Message-ID: <FB77A8A317F5D31199F1009027DE4A8701B001DF@wntcsdexg01.csd.ciena.com>

>-----Original Message-----
>From: Raymond Hettinger [mailto:othello at javanet.com]
>Subject: Comments invited: PEP 279 Enhanced Generators
>
>
>Your review and comments are invited for PEP 279:
>   http://python.sourceforge.net/peps/pep-0279.html
>
[snip good suggestions and change history]
>
>Raymond Hettinger

I like all of these.  It might be worth noting that these functions would
come in at 2.3, the same time that generators are final (no longer imported
from the __future__). 

I'm a little confused by your example in Specification for two-way Generator
Parameter Passing 

>From the PEP:

    Example of a Consumer Stream

        def filelike(packagename, appendOrOverwrite):
            cum = []
            if appendOrOverwrite == 'w+':
                cum.extend( packages[packagename] )
            try:
                while 1:
                    dat = yield None
                    cum.append(dat)
            except FlushStream:
                packages[packagename] = cum
        ostream = filelike('mydest','w')   # Analogous to
file.open(name,flag)
        ostream.next()                     # Advance to the first yield
        ostream.next(firstdat)             # Analogous to file.write(dat)
        ostream.next(seconddat)
        ostream.throw( FlushStream )       # This feature proposed below

I don't see why the no-argument first call to .next() is necessary.  I know
you previously proposed control flow so that the code up to the first yield
was executed at iterator construction, rather than the lazy way they are
now.  The current flow is this:
	process, yield output
Unless I'm wrong, you are proposing this: 
	accept input (.next arguments), process, yield output
It seems more natural to do this:
	process, yield output, accept input
even though from the caller's perspective, you see this:
	resultFromLastInput = myGen.next(inputToBeProcessed)
since internally, the iterator is doing this:
	inputToBeProcessed = yield resultFromLastInput
Assuming that throwing the exception causes the processing to proceed to the
yield before it is raised, this would allow the above example to be
unchanged, except for the removal of the now-unnecessary pump-priming call
to ostream.next()
(If you didn't understand my explanation, you obviously can't understand the
nine commands that constitute all are actions. ;)

Also, this example appears before the section that describes the .throw()
addition.



I also have minor quibble with Specification for Generator Comprehensions

>From the PEP:
    If a list comprehension starts with a 'yield' keyword, then
    express the remainder of the statement as generator.  For example:

        g = [yield (len(line),line) for line in file.readline() if
len(line)>5]
        print g.next()
        print g.next()

    This would be implemented as if it had been written:

        def __temp_gen():
            for line in file.readline():
                if len(line) > 5:
                    yield (len(line), line)
        g = __temp_gen()
        print g.next()
        print g.next()

Currently, this code:
	g = [(len(line),line) for line in file.readline() if len(line)>5]
would leave the variable "line" (with its last value) left in the locals
(and clobber an existing "line"), since it implemented as if it had been
written:
	g = []
       for line in file.readline():
           if len(line) > 5:
               g.append(line)

When many read this, they assumed that it wouldn't _actually_ create a local
called "line".  They were wrong.  Will your code: 
a) actually create a local called "__temp_gen"
b) not create a local called "line"


Otherwise, it's very impressive and I hope they're all implemented.

-Jeremy Cromwell



From dmaster at synopsys.com  Wed Feb  6 12:12:24 2002
From: dmaster at synopsys.com (David Masterson)
Date: 06 Feb 2002 09:12:24 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net>
Message-ID: <ug04ey1on.fsf@synopsys.com>

>>>>> Timothy Rue writes:

> On 04-Feb-02 12:59:12 David Masterson <dmaster at synopsys.com> wrote:

>> Your (extended) program appears to be no better than the previous
>> one I looked at in that it doesn't appear to do anything useful.
>> My memory of AREXX is ~10 years old, so some things in your program
>> are not obvious (like the PARSE command).  Overall, though, it
>> appears to merely do some convoluted printouts that I can't make
>> heads or tails of.

> Programming one step at a time. This step is just the main parser.

> Certainly a Software engineer such as yourself knows this to such a
> degree that it's obvious to you. And as far as arexx goes....well
> Like I said, it doesn't go thru the complete framework, But it
> doesn't mean there is not a framework spelled out in english.

Of course I can recognize this as a (simple) parser.  That much was
obvious sometime back.  A simple Lex/YACC program what I saw there.

What is not obvious is what all the variations of the "expected" input
for the parser will be, what the program is supposed to do with the
parsed input, and what the "expected" output of the program
(ie. V.I.C.) is supposed to be for a given input.

> We both know it can be done, and in more than one way. At this stage of
> development, that is all that is needed. Once the tool is built then we
> can focus on getting the tool to output what we want from the input we
> provide it with.

Given a clear definition of the three key things required in
programming (input, program, output), we can give you lots of
different ideas in lots of different computer languages on how to
create the program.

> In either case, more information is needed and that's what we work on
> after the tool is created. 

That is putting the cart in front of the horse.  We cannot create a
tool to solve a problem until we know what the problem is.

> There is only the creation of a configuration of very common code
> functionality, carried out to it's natural conclusion in support of
> versatility.

That has already been done.  It's been done over and over for the last
50 years of computer programming.  Without a clear definition of what
you mean by "common code functionality" to differentiate your approach
versus what's already out there, I see nothing worthwhile in your
project.

> The parse framework is the place to start in creating the whole
> VIC. And at some point we will be able to integrate What was done in
> the standalone IQ program. The PK multi-deminsional array is the
> main data collection that gets manipulated.

> Handle the parse routine and then fill in one by one that functions the
> parse routine points to given selected input.

Again, this is putting the cart before the horse.

> A great deal of thought has gone into this already, alot more has been
> thoughtout and defined than this parse routine framework. Enough to know
> doing the parse routine is no waste of effort, but something that needs to
> be done.

Unproven statement...

> <shrug>

<quizzocal look>

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From sabren at manifestation.com  Tue Feb 12 16:58:47 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 12 Feb 2002 16:58:47 -0500 (EST)
Subject: leo
Message-ID: <Pine.LNX.4.44.0202121653310.14393-100000@mercury.sabren.com>


I haven't seen this mentioned here in a while, so I thought
I'd point out the neat stuff Edward Ream is doing over on
the leo project....

Leo is a Literate Editor with Outlines - a tool for literate
programming. It has been scriptable in python for quite a
while, and over the past few weeks, Edward has moved the
whole thing to python. (He even wrote a tool to translate C
code to python)

Anyway, just thought some of you might be interested. I'm
using Leo to organize my pyunit test cases into a sort of
executable spec, and I've found it to be a really nice tool.

     http://personalpages.tds.net/~edream/front.html

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From stevena at neosynapse.net  Tue Feb 19 16:53:50 2002
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Tue, 19 Feb 2002 16:53:50 -0500
Subject: command-line history problems
Message-ID: <71319333921.20020219165350@neosynapse.net>

I'm using Python 2.2.  On a particular machine, command history isn't
working.  When I hit the up-arrow, I get:

>>> ^[[A

The same version of Python works correctly on a different machine.  On
the problem machine, in the bash shell, command history works fine.
Anyone know what I can do to correct this?

Thanks in advance,

steve





From vvainio at karhu.tp.spt.fi  Mon Feb 18 09:15:36 2002
From: vvainio at karhu.tp.spt.fi (Ville Vainio)
Date: 18 Feb 2002 16:15:36 +0200
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net>
Message-ID: <yoxr8niyiyf.fsf@karhu.tp.spt.fi>

"Gillou" <nospam at bigfoot.com> writes:

> Alternate solution, use a direct connection to your DBMS client (no ODBC)
> Search from http://www.python.org/sigs/db-sig/

Direct connection doesn't work in my case (my DBMS only offers ODBC
and JDBC). However, I overreacted a little bit w/ mxODBC - I thought
it was $500 (as it used to be), but I was emailed that it is in fact
$75.

-- 
Ville Vainio - http://www.tp.spt.fi/~vvainio - ICQ #115524762
 Wild geese have no intention to cast a reflection
 Water has no mind to assume their form


From dyoo at hkn.eecs.berkeley.edu  Tue Feb 19 23:11:10 2002
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Wed, 20 Feb 2002 04:11:10 +0000 (UTC)
Subject: This isn't recursive
References: <e9e6dd1f.0202191424.51a23ba8@posting.google.com>
Message-ID: <a4v7ku$2kdf$1@agate.berkeley.edu>

Jason <caljason76 at yahoo.com> wrote:
: # Ummm... I think the posted wanted a recursive solution.
: #
: # I am a Python novice so I don't really understand how things
: # work, yet.  So some of this may either be wrong or not the
: # Python way.  Also I have no idea why the 'if __name__ == ...'
: # construct is needed instead of just the naked main statements.

Python files don't necessarily have main() functions; when Python
processes a file, it actually executes the code as-is.  That is,
things like 'def' are actual statements in the language.  This makes
certain dynamic things possible, like this:

###
choice = raw_input("numeric or verbal?")
if choice == 'numeric':
    def yes(): return 1
else:
    def yes(): return 'Yes!'
print yes(), "master"
###

If you're familiar with Scheme, you will feel quite at home.  *grin*


The "__name__ == '__main__'" expression is true only when we're
running a file as a standalone Python program.  If we 'import' a file
as a module, however, things are quite different.  Here's an example:

###
[dyoo at hkn dyoo]$ cat hello.py
print "Hello, my name is", __name__
[dyoo at hkn dyoo]$ python hello.py
Hello, my name is __main__
[dyoo at hkn dyoo]$ python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import hello
Hello, my name is hello
>>> 
###

Variables with double underscores around them should be considered
special varaibles that are set up by the Python runtime system.


: # Also, can somebody help me find the documentation to the builtin
: # procedures, like print().  I checked the __builtin__ module docs
: # but there is nothing there.  Thanks

You can find documentation on the builtin functions here:

    http://python.org/doc/lib/built-in-funcs.html

However, you won't find 'print' in there because Python's print is a
statement; it's not a function but a "special form".  There are
reasons for this: it allows for a simplified syntax for printing
things.


: # -j (nordwick at xcf.berkeley.edu)

Nice to see another person from Berkeley!  *grin*

Good luck to you!


From chris_mk at hotmail.com  Thu Feb  7 15:30:32 2002
From: chris_mk at hotmail.com (Christopher)
Date: 7 Feb 2002 12:30:32 -0800
Subject: cgi POST question
Message-ID: <cc14131d.0202071230.415f2447@posting.google.com>

Hi,
     I am relatively new to programming in general and Python in
particular (Two years programming basic stuff and the last six months
with Python).  For the most part, I use Python purely to process and
extrapolate data.  I have absolutely no Network Programming
experience.  Lately, I have been using the urllib module to construct
cgi queries on the fly (get a list of keywords, etc. and plug them
into a generic cgi query).  As an example (not real code):

ListOfQueries = []
cgiQuery = "http://www.site.com/query.cgi?keyword=MYWORD&defaults=blahblah"
for x in [MyListOfStuff]:
        newQuery = cgiQuery.replace('MYWORD', x)
        ListOfQueries.append(newQuery)


I would then use the formatted queries for whatever.  My boss asked me
to add  the ability to do the same on our intranet server.  I thought
it would be easy.  I was wrong.  Here is what I think is going on
(though I am sure I have some details wrong).
     A query like the one above is what is passed when a method of
'GET' is specified (the intranet search specifies a method of 'POST'
instead, I realized much later).  I carefully constructed the cgi
query, but it wasn't returning the correct results.  I then took the
source itselt, modified it to submit using a method of 'GET' and
tried.  The cgi query that was produced matched what I had constructed
but still no results.  I spoke with the guy in charge of the intranet
search engine and he said that the cgi script he wrote is incompatible
with a 'GET' method (my ignorance shows itself, I thought it didn't
matter which method was used) and he doesn't have the time or the
inclination to change it (he was also a little bitter that I would
even be 'messing' with his 'stuff' even though it was his boss' boss
that told me to do it).  I looked at the cgi module documentation, but
it doesn't seem to be able to do what I want.
     What I want, by the way, is to know if there is a python module
that submits information in the 'POST' method.  With urllib, I can
submit information in the 'GET' method (all it is is a website, after
all), but I can't figure out how to do the same with the POST method
(I was thinking it would be an object, with attributes in a dictionary
and a method to post it to the cgi script in the correct format,
soemthing like:

myTest = cgimodule.cgiPostObject
myTest['keyword'] = "MyWord"
myTest['keyword2'] = "MyOtherWord"
myTest.cgiSubmit("http://www.site.com/query.cgi", method = 'POST')


or something to that effect).  I would write it myself, but I wouldn't
know how.
     I did have my boss approve the purchases of a couple networking
books and "Python Web Programming" by Steve Holden (<= very good so
far, btw) but I am still not up to speed yet (basically, I am learning
Networking stuff, but reading at home, since I am so busy at work).  I
really need to figure this out because patience is relatively low
right now.  Any help would be appreciated.  Thank you very much.  I
would be happy to answer any additional questions. Talk to you all
later.

Chris


From sholden at holdenweb.com  Sun Feb 10 18:57:57 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 10 Feb 2002 18:57:57 -0500
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org> <afe3b146.0202100036.f221c9a@posting.google.com>
Message-ID: <ZND98.140690$KM2.6229113@atlpnn01.usenetserver.com>

"Brian Lenihan" <brian_l at yahoo.com> wrote in message
news:afe3b146.0202100036.f221c9a at posting.google.com...
[ ... ]
> I eat, write, mouse, and play golf right handed.  I kick equally well
> with both feet, at least while playing soccer.  My surfing, skate
> boarding, throwing, tennis, darts, archery and shooting are done
> left-handed. My left eye is dominant. I agree with the cultural bias
> against left-handers. However, nuns with rulers were probably
> responsible for me writing with my right hand.  My mother is probably
> responsible for me eating with my right hand. I play golf right-handed
> because there were no left-handed clubs available to me when a learned
> as a child.
>
> I don't throw like a girl right-handed, so maybe I am actually
> ambidextrous.
>
Not to mention gender-biased? :-)

> I have a brother who does everything right-handed except eating. He's
> always been the stubborn one.
>
>
There-are-other-things-that-one-does-with-one-hand-or-the-other-but-we-don't
-mention-ly-yours
>
> Oh, and if throwing was really so "coarse-grained" I'd have been a
> Major League pitcher.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jgardn at alumni.washington.edu  Tue Feb 12 04:59:00 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Tue, 12 Feb 2002 18:59:00 +0900
Subject: Networking Problems
References: <3C68E17C.3030009@xtra.co.nz>
Message-ID: <zy5a8.306$ab7.1316@news.hananet.net>

Simon Crosby wrote:

> Hi
> 
> I have recently tried programming a text network game using sockets in
> python as detailed in "Programming Python" 2nd ed. I developed this on
> linux and used telnet as a client. Then I tried connecting to the game
> server with windows telnet. This showed the text very differently,
> ignoring newlines and when prompting for input takes any character as a
> newline. It does not matter if the server is run on linux or windows.
> Attached is my networking code.
> 

You'll want to use "\012\015" explicitly as a newline. This is both for 
input and output.

Windows telnet sends characters as they are typed. Linux telnet defaults to 
sending whole lines rather than individual characters. The best way is to 
collect input from the client until the client sends "\012\015".

Also, your whole design is lacking. You'll need a server that can 
handle multiple connections. I suggest using non-blocking sockets and 
select. It's better to build the server like this from the beginning 
because it affects how you do everything.

Jonathan


From mwh at python.net  Mon Feb 18 06:54:30 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 18 Feb 2002 11:54:30 GMT
Subject: nested scopes
References: <CfTb8.129031$th4.36971458@news02.optonline.net> <zsTb8.129039$th4.37009591@news02.optonline.net> <a4p13e$1r4ju$1@ID-11957.news.dfncis.de>
Message-ID: <ur8njc89g.fsf@python.net>

"Emile van Sebille" <emile at fenx.com> writes:

> "Nicola Kluge" <NKluge at optonline.net> wrote in message
> news:zsTb8.129039$th4.37009591 at news02.optonline.net...
> > Nevermind, I found that it doesn't work in IDLE, but it works in
> cmdline
> > Python.
> >
> 
> from __future__ import nested_scopes  needs to be the first thing done,
> and in idle you never get to be first...
> 
> It does work, of course, in idle under 2.2.

__future__ statements work in idle under 2.2, too (i.e. generators,
future division).

Cheers,
M.

-- 
  ... when all the programmes on all the channels actually were made
  by actors with cleft pallettes speaking lines by dyslexic writers
  filmed by blind cameramen instead of merely seeming like that, it
  somehow made the whole thing more worthwhile.   -- HHGTG, Episode 11


From warlock at eskimo.com  Fri Feb 22 03:12:33 2002
From: warlock at eskimo.com (Jim Richardson)
Date: Fri, 22 Feb 2002 00:12:33 -0800
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1> <a54qfl$vl4$0@216.39.172.122>
Message-ID: <hhu45a.4d1.ln@127.0.0.1>

On 22 Feb 2002 07:03:17 GMT,
 Bengt Richter <bokr at oz.net> wrote:
> On Thu, 21 Feb 2002 17:59:14 -0800, Jim Richardson <warlock at eskimo.com> wrote:
> 
>>
>>Pythonic bit shuffling question
>>
>>I have an application that does a calculation on a single byte of data,
>>after breaking it into an upper and lower nibble of four bits. How can I
>>do this in python? To clarify.
>>Take a single byte, break off the upper 4 bits, and do some stuff, then
>>do similar stuff with the lower nibble, and then munge them back
>>together to create a byte again. This is for a checksum to inteface with
>>a bit of embedded stuff. I know how I can do this in C, but as that
>>would be the sole bit of copiled code in the app, I'd prefer to avoid
>>it if possible. 
>>
> Unless you are mixing in other data somehow into the byte, there is only
> 256 possible states of your input and output, so you should able to
> pre-construct a translation table and say
>     outByte = transTable[inByte]
> 
> and be done. This may not be smaller than defining a function
> to do it, but if you need speed, it ought to run faster.
> If your in and out bytes are characters, the above transTable
> becomes a string and you have
>     outByte = transTable[ord(inByte)]
> 
> It sounds like you should be able to describe your function in detail
> pretty easily. Why not post a cut at a function definition?
> 
> Then you can decide to use it directly or use it to compute a
> translation string, per your priorities. If you are doing exactly
> the same thing to both nybbles, you can do a 16-item lookup for
> that part. It might be easier to enumerate than to program
> the logic, depending.
> 
> Regards,
> Bengt Richter
> 


Thanks, this isn't a bad idea (I am not really python guru, so am easily
overwhelmed sometimes :) I will have to think on it. Speed isn't much of
an issue, as the data is only 9600 and each packet needs a checksum, and
usually, only 8-10 packets a second. 
 Another reply mentioned a "byte" operator, but I can't find that
anywhere in my python docs. Is it something not in the default setup? 

Thanks

-- 
Jim Richardson
	Anarchist, pagan and proud of it
www.eskimo.com/~warlock
    Linux, for when you need to get work done, Or you could just play UT...


From huaiyu at gauss.almadan.ibm.com  Fri Feb 15 14:24:16 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Fri, 15 Feb 2002 19:24:16 +0000 (UTC)
Subject: Newbie: Object assignment return value
References: <e81be8b2.0202141051.157160d6@posting.google.com>
Message-ID: <slrna6qo2v.15d.huaiyu@gauss.almadan.ibm.com>

On 14 Feb 2002 10:51:40 -0800, Dean Goodmanson <ponderor at lycos.com> wrote:
>(I think) I'm trying to understand if object assignment has a return value.
>
>Scenario A:
>>>> str( spam= 5 )
>''
>>>> str((spam=5))
>  File "<stdin>", line 1
>    str((spam=5))
>            ^
>SyntaxError: invalid syntax
>>>> #  continued : Scenario B:
>>>> print spam
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>NameError: name 'spam' is not defined
>
>A: Why does including an extra set of paranthesis throw when the original doesn't?
>
>B: What happend to my spam?
>Espcially when this works:
>>>> str(int())
>'0'
>>>> str((int()))
>'0'
>>>>
>
>Wishin'-I-had-something-practical-for-an-example,
>
>Dean

As others have pointed out, assignments are statements.  Expressions can
appear in statements but not vice versa.  Assignments do return values in
the sense that it can be used in chained assignments, but nothing else.

But you examples look strange anyway.  Did you copy and paste verbosely?
Which version of Python are you using?  I get

>>> str( spam= 5 )
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: this function takes no keyword arguments
>>> str(int())
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: int requires at least 1 argument; 0 given
>>> str((int()))
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: int requires at least 1 argument; 0 given

Huaiyu


From gerson.kurz at t-online.de  Mon Feb  4 12:27:07 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 17:27:07 GMT
Subject: how can i get the screen width and height in pixels?
References: <a3mclk$pjh$1@news5.svr.pol.co.uk>
Message-ID: <3c5ec34b.27704203@news.t-online.de>

On Mon, 4 Feb 2002 16:26:42 -0000, "G. Willoughby"
<sab at NOSPAM.freeuk.com> wrote:

>how can i get the screen width and height in pixels?

On Windows:
----------------
from win32api import *
from win32con import *

print "Screen width:", GetSystemMetrics(SM_CXSCREEN)
print "Screen height:", GetSystemMetrics(SM_CYSCREEN)
----------------
On Linux:

Buy a metric ruler, then construct a lego mindstorm robot to measure
the screen with that ruler. 

(Sorry folks, couldn't resist, I have nothing against linux, in fact I
like it :).



From jason at tishler.net  Mon Feb 11 09:16:51 2002
From: jason at tishler.net (Jason Tishler)
Date: Mon, 11 Feb 2002 09:16:51 -0500
Subject: cx_Oracle on cygwin
In-Reply-To: <Xns91AD8D95B2B69andy47halfcookedcom@203.109.250.24>
References: <Xns91AD8D95B2B69andy47halfcookedcom@203.109.250.24>
Message-ID: <20020211141651.GE2256@dothill.com>

Andy,

> Now everything is fine until we reach this stage;
> 
> gcc -shared -Wl,--enable-auto-image-base build/temp.cygwin-1.3.9-i686-
> 2.2/cx_Oracle.o -L/Tools/Lib -L/usr/lib/python2.2/config -loci -
> lpython2.2 -o build/lib.cygwin-1.3.9-i686-2.2/cx_Oracle.dll -s
> /usr/lib/gcc-lib/i686-pc-cygwin/2.95.3-5/../../../../i686-pc-
> cygwin/bin/ld: cannot find -loci
> collect2: ld returned 1 exit status
> error: command 'gcc' failed with exit status 1
> 
> Reading 'man' for gcc this *seems* to be a problem because the linker is 
> looking for a file called liboci.a which, naturally, doesn't exist under 
> windows. I think I should be linking to oci.dll, but I don't know enough 
> to be sure.
> 
> Any suggestions, pointers or re-education would be most welcome.

You *should* be able to link directly with the DLL -- although, it is
recommended to link against the corresponding import library instead.

Create an import library, if necessary.  I recommend pexports:

    http://www.cygwin.com/ml/cygwin/2000-03/msg00694.html

Here are some (possibly useless) suggestions:

    o Try using the "-v" option to see what is really going on.
    o Did you supply the correct -L option to find -loci?
    o Try replacing -loci with oci.dll.

HTH,
Jason



From mojo at geofinity.com  Sun Feb 10 13:03:31 2002
From: mojo at geofinity.com (Mojo)
Date: Sun, 10 Feb 2002 13:03:31 -0500
Subject: OOP Newbie mixing COM, wxPython, BOA
Message-ID: <u6ddfmsu6ug32@corp.supernews.com>

The subject line could easily have been "fool in way over his head..."

That said, I got farther than I thought possible only to be stymied by what
I'd guess is either an idiom I don't understand (being an OOP newbie) or a
question of variable scope/namespace...

My problem is that my app has two classes (a class encapsulating the GUI,
and a class that implements an event handler for a COM object.)  I don't
know how to access GUI objects in the former, from code in the latter.

I've built a GUI using the latest BOA Constructor (very nice), wxPython
(2.3.2.1) on Python 2.1.1 (ActiveState).  The GUI runs nicely, and is
encapsulated in a nifty class I named "VpFrame".  VpFrame includes a menu
with event bindings to internal (VpFrame) methods, and a tree widget.  The
tree widget is created inside a VpFrame method named "_init_ctrls__" using
the command:

self.buddyTree = wxTreeCtrl(id = wxID_VPFRAMEBUDDYTREE, name = 'buddyTree',
parent = self.mainSplit, pos = wxPoint(354, 2), size = wxSize(157, 407),
style = wxTR_HAS_BUTTONS, validator = wxDefaultValidator)

Next, I used win32com to add access to the JabberCOM object and created a
class to process events coming from the COM object.  This event processing
class is called and "JabberEvents" and works also.  My menu from VpFrame
successfully executes a "Connect" method via the JabberCOM object.  This
object, in turn, sends a message to the remote server (verified) and
receives an "OnConnect" event (verified) from the server.  I know that
"JabberEvents" is working because it's "OnConnect" event handler contains
the code that makes my "presence" visible to other users of the server.
Only when this event is processed do I become "Online" to other users.  I've
verified that all of this works simply by logging in (as a different user,
on a different machine) and watching for my presence to appear.  It does.

My next step is to add code to the "OnConnect" event handler (in the
"JabberEvents" class) that adds some items (e.g. my list of buddies) to the
tree object in VpFrame.  This is where I'm stumped.  For code inside the
VpFrame class, I can simply call "self.buddyTree.AddRoot("Root Item On
Tree") and it works.  Unfortunately none of my (scattershot) attempts to do
this from the JabberEvents class work.  I've tried "buddyTree.AddRoot" (no
global attribute named "buddyTree"), "VpFrame.buddyTree.AddRoot" (class
VpFrame has no attribute 'buddyTree'), etc.  I also tried creating a method
inside the VpFrame class (FillTree()) to do the tree management.  I can call
this method from the menu (self.FillTree() inside the VpFrame class) but not
from code in the JabberEvents class.

How do I call the "AddRoot" method of the "buddyTree" object in the VpFrame
class, from the separate class "JabberEvents"?  Any help, pointers,
explanations greatly appreciated.  I've include the code (trimmed as much as
possible for brevity) below.

Thanks,

David
---
--
-


#########  Code For "vp.py" #########
from wxPython.wx import *
import VpFrame
modules ={'VpFrame': [1, 'Main frame of Application', 'VpFrame.py']}

class BoaApp(wxApp):
    def OnInit(self):
        self.main = VpFrame.create(None)
        self.main.Show(true)
        self.SetTopWindow(self.main)
        return true

def main():
    application = BoaApp(0)
    application.MainLoop()

if __name__ == '__main__':
    main()

#########  Code For "VpFrame.py" #########
def create(parent):
    return VpFrame(parent)

[wxID_VPFRAME, wxID_VPFRAMEBUDDYTREE, wxID_VPFRAMEMAINSPLIT] = map(lambda
_init_ctrls: wxNewId(), range(10))

class JabberEvents:
    """Main handler for Jabber events"""

    def OnConnect(self):
        """Event occurs when a sucessfull connection to the Jabber server is
established"""
        jab.Available = 1
        jabRoster.Fetch()
        try:
            VpFrame.buddyTree.AddRoot('This is some text')
        except:
            TBStrings = traceback.format_exception(*sys.exc_info())
            for Line in TBStrings:
                print Line

jab = win32com.client.DispatchWithEvents("JabberCOM.JabberSession",
JabberEvents)
jabRoster = jab.Roster

[wxID_WXFRAME1FILEMENUITEMS0, wxID_WXFRAME1FILEMENUITEMS1,
wxID_WXFRAME1FILEMENUITEMS2, wxID_WXFRAME1FILEMENUITEMS3] = map(lambda
_init_coll_FileMenu_Items: wxNewId(), range(4))

class VpFrame(wxFrame):
    """ Main Application "Frame", provides the (GUI) user interface"""

    def _init_coll_menuBar1_Menus(self, parent):
        parent.Append(menu = self.FileMenu, title = 'File')

    def _init_coll_FileMenu_Items(self, parent):
        parent.Append(checkable = false, helpString = 'Connect to the VP
System', id = wxID_WXFRAME1FILEMENUITEMS0, item = 'Connect')
        parent.Append(checkable = false, helpString = 'Disconnect from the
VP system', id = wxID_WXFRAME1FILEMENUITEMS1, item = 'Disconnect')
        parent.Append(checkable = false, helpString = 'Exit VP', id =
wxID_WXFRAME1FILEMENUITEMS3, item = 'Exit')
        EVT_MENU(self, wxID_WXFRAME1FILEMENUITEMS0, self.OnFileMenuConnect)
        EVT_MENU(self, wxID_WXFRAME1FILEMENUITEMS1,
self.OnFileMenuDisconnect)
        EVT_MENU(self, wxID_WXFRAME1FILEMENUITEMS3, self.OnFileMenuExit)

    def _init_utils(self):
        self.FileMenu = wxMenu(title = '')
        self._init_coll_FileMenu_Items(self.FileMenu)
        self.menuBar1 = wxMenuBar()
        self._init_coll_menuBar1_Menus(self.menuBar1)

    def _init_ctrls(self, prnt):
        wxFrame.__init__(self, id = wxID_VPFRAME, name = '', parent = prnt,
pos = wxPoint(367, 442), size = wxSize(521, 485), style =
wxDEFAULT_FRAME_STYLE, title = 'VP Version 3.0')
        self._init_utils()
        self.SetClientSize(wxSize(513, 451))
        self.SetMenuBar(self.menuBar1)

        self.mainSplit = wxSplitterWindow(id = wxID_VPFRAMEMAINSPLIT, name =
'mainSplit', parent = self, point = wxPoint(0, 0), size = wxSize(513, 412),
style = wxSP_3D)
        self.mainSplit.SetMinimumPaneSize(100)
        self.mainSplit.SetSashSize(4)

        self.buddyTree = wxTreeCtrl(id = wxID_VPFRAMEBUDDYTREE, name =
'buddyTree', parent = self.mainSplit, pos = wxPoint(354, 2), size =
wxSize(157, 407), style = wxTR_HAS_BUTTONS, validator = wxDefaultValidator)
        self.buddyTree.SetToolTipString('Buddy Tree')
        self.mainSplit.SplitVertically(self.mainNote, self.buddyTree, 350)

        self._init_coll_mainNote_Pages(self.mainNote)

    def __init__(self, parent):
        self._init_ctrls(parent)

    def FillTree(self):
        myStat = "My Roster (" + `jabRoster.Count` + ")"
        btRoot = self.buddyTree.AddRoot(myStat)
        btOnline = self.buddyTree.AppendItem(btRoot, 'Online Buddies')
        btOffline = self.buddyTree.AppendItem(btRoot, 'Offline Buddies')
        self.buddyTree.Expand(btRoot)
        self.buddyTree.Expand(btOnline)
        self.buddyTree.Expand(btOffline)

    def OnFileMenuConnect(self, event):
        jab.Username = 'userid'
        jab.Password = 'password'
        jab.Resource = 'resname'
        jab.Server = 'servername'
        jab.Port = 'portnum'
        jab.DoConnect(0,5)

    def OnFileMenuExit(self, event):
        self.Close()

    def OnFileMenuDisconnect(self, event):
        jab.DoDisconnect(1)





From dyoo at hkn.eecs.berkeley.edu  Mon Feb 25 23:25:46 2002
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Tue, 26 Feb 2002 04:25:46 +0000 (UTC)
Subject: What's wrong with this subroutine?
References: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>
Message-ID: <a5f2oa$e7m$1@agate.berkeley.edu>

A.Newby <deathtospam43423 at altavista.com> wrote:
: First, here's what I'm trying to do ......

: I'm working on a html, cgi based chat script. It's working pretty well so 
: far. 

: I want to get this subroutine so that the user just has to post a url into 
: the chat box to get a post a picture or a link in chat. For e.g, to post a 
: pic of a walrus, you would simply type ....

: http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg

: ... and the script would take that, and before writing it to the chat log 
: it would convert it to ...

: <img src = http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg>

: ... or to post a link, the user would simply type ... 

: http://www.goodvibes.com/

: ... and the script would convert it to ....

: <a href = "http://www.goodvibes.com/>Link!</a>


Sounds good!  You might want to use a regular expression to detect
this "url" pattern.  Here's a translation of Tom Christiansen's (of
Perl fame) HTTP url regular expression:


###
## This is a regular expression that detects HTTP urls.
##
## This is only a small sample of tchrist's very nice tutorial on
## regular expressions.  See:
##
##     http://www.perl.com/doc/FMTEYEWTK/regexps.html
##
## for more details.

urls = '(%s)' % '|'.join("""http telnet gopher file wais ftp""".split())
ltrs = r'\w'
gunk = '/#~:.?+=&%@!\-'
punc = '.:?\-'
any = "%(ltrs)s%(gunk)s%(punc)s" % { 'ltrs' : ltrs,
                                     'gunk' : gunk,
                                     'punc' : punc }

url = r"""
    \b                            # start at word boundary
    (                             # begin \1 {
        %(urls)s    :             # need resource and a colon
        [%(any)s] +?              # followed by one or more
                                  #  of any valid character, but
                                  #  be conservative and take only
                                  #  what you need to....
    )                             # end   \1 }
    (?=                           # look-ahead non-consumptive assertion
            [%(punc)s]*           # either 0 or more punctuation
            [^%(any)s]            #  followed by a non-url char
        |                         # or else
            $                     #  then end of the string
    )
    """ % {'urls' : urls,
           'any' : any,
           'punc' : punc }

url_re = re.compile(url, re.VERBOSE)


def _test():
    sample = """hello world, this is an url:
                http://python.org.  Can you find it?"""
    match = url_re.search(sample)
    print "Here's what we found: '%s'" % match.group(0)

if __name__ == '__main__':
    _test()
###



Good luck to you!


From dysfate at gmx.de  Fri Feb  1 07:34:10 2002
From: dysfate at gmx.de (Johannes Stiehler)
Date: 01 Feb 2002 13:34:10 +0100
Subject: automaton package for python
Message-ID: <a3e201$dtc$1@news.informatik.uni-muenchen.de>

Does anybody know about an elaborate package for automaton
construction? It should support DFA/NFA - Conversion,
automaton-reduction and converting to regular expressions.

So far I've only seen this kind of thing for Java
(http://www.brics.dk/~amoeller/automaton/) which is very impressive.

If at the moment there is no such library, would there be any interest
to develop one?

Features (besides the above mentioned) I long for are:

- reading an automaton-specification from an XML-File (with tags for
  the components: start-state, accept-states, transition table, set of
  states)

- outputting the automaton to graphviz dot

- generating a re-compliant regular expression from the automaton
  specification as a raw string

Regards,
Johannes Stiehler

-- 
Johannes Stiehler				 stiehler at cis.uni-muenchen.de
Centrum f?r Informations- und Sprachverarbeitung 089/2178-2706
Ludwig-Maximilians-Universit?t 			 Oettingenstr. 67
Systemverwaltung & Anwendungsbetreuung		 80538 M?nchen


From mhammond at skippinet.com.au  Thu Feb 14 18:44:13 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 14 Feb 2002 23:44:13 GMT
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk>
Message-ID: <3C6C4C37.6020209@skippinet.com.au>

Robin Becker wrote:
> I am having bad problems with zope not starting up properly. I think
> this is caused by having Python-2.1.2 + win32all installed. It seems
> that paths set up in the registry by win32all are getting into sys.path
> when zope runs. This is causing massive confusion between modules
> installed in the base python and those I wish to test with zope.

IMO, the zope install should be fixed.  The Zope installer should decide 
if it wants to work with an installed Python OR it wants its own, 
discrete Python.

If the former, it should just make a regular standard Python install, 
then install Zope as an extension.  Then, even if "\Python21" was not 
the base directory, it is still a regular Python install - ie, just as 
if the user had manually specified the installation directory.  win32all 
etc would then happily install into this directory.

If the latter (which is more professional) it should just take steps to 
ensure that it is partitioned correctly.  Generally, this should only 
mean changing *one* string in the compiled Python DLL, which will 
completely change the registry key used by Python to load sys.path. 
Obviously the installer itself then changes to reflect this new key, but 
that should be all that is necessary.

Mark.



From fperez528 at yahoo.com  Mon Feb 18 19:01:31 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 19 Feb 2002 00:01:31 +0000
Subject: Adding static typing to Python
References: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com> <mailman.1014092911.31096.python-list@python.org> <gft37u8csm1ukd923nr933iq7meimqu5i7@4ax.com>
Message-ID: <a4stbu$ptq$1@peabody.colorado.edu>

Courageous wrote:

> 
> One of the things I like about C sharp is their use of attributes,
> where you can define getters and setters on an attribute, but
> otherwise the attribute is simply used as if it were a public one.
> An objects attributes can thereby be used as a published aspect of
> its interface safely, but no one will be the wiser if you manipulate
> the getter and setter code.

Just curious, isn't there something quite similar to what you describe in 
py2.2? I've only glanced at it so I could be wrong...

cheers,

f


From fredrik at pythonware.com  Wed Feb 13 17:20:50 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 22:20:50 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <c76ff6fc.0202131330.66bd9d93@posting.google.com>
Message-ID: <6FBa8.25612$l93.4773309@newsb.telia.net>

John Machin wrote:
> Yes, unfortunately there is a lot of octal in the library. Some of it
> has the reasonable excuse of copying existing Unix practice or
> mechanical translation of existing C code:
>
> stat.py:    return mode & 0170000
> stat.py:S_IFDIR  = 0040000

fixing the library would be trivial.

encouraging people to fix their code would also be trivial
(add both spellings, use the warning system to phase out
the old spelling).

code that expects eval(str) not to blow up on things read
from configuration files, things in data files, things received
over the network (yeah, I know about security), etc, is a
bit more difficult to deal with.

Traceback (most recent call last):
  File "effbot.py", line 105, in ?
  File "SimpleXMLWriter.py", line 103, in end
    assert self.__tags, "unbalanced end(%s)" % tag
AssertionError: unbalanced end(F)




From mats at laplaza.org  Thu Feb 14 12:14:40 2002
From: mats at laplaza.org (Mats Wichmann)
Date: Thu, 14 Feb 2002 17:14:40 GMT
Subject: Re-throwing an exception without loosing the stack trace
References: <20020212174235.A17928@doublegemini.com> <15465.39126.250327.239174@beluga.mojam.com> <mailman.1013554227.2313.python-list@python.org>
Message-ID: <3c6beb68.82233805@news.laplaza.org>

On Tue, 12 Feb 2002 18:08:29 -0500, "Clark C . Evans"
<cce at clarkevans.com> wrote:

:Yes.  I didn't try using raise without any arguments.  Cool.
:Funny, this isn't in Learning Python... and raise all by
:itself isn't listed in the syntax in the book.  Hmm.


A teaching text is not going to serve as a complete language reference
- those are separate topics.  I glanced at LP and notice it did
mention re-raising exceptions, but didn't actually give the details
(p. 197). I'm not going to throw stones; as the author of a Python
course, every time I look at it I see something I wish I had said
differently, or included, or ...  

You might consider keeping a Python Essential Reference on hand as a
supplement.

Mats Wichmann



From threeseas at earthlink.net  Tue Feb  5 23:54:44 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:44 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com>
Message-ID: <4898.802T2975T395701threeseas@earthlink.net>

On 04-Feb-02 12:59:12 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> NEED TO PUT MORE THOUGHT INTO THIS!!!
>> *************************************

>I couldn't have said it better myself!

>Your (extended) program appears to be no better than the previous one
>I looked at in that it doesn't appear to do anything useful.  My
>memory of AREXX is ~10 years old, so some things in your program are
>not obvious (like the PARSE command).  Overall, though, it appears to
>merely do some convoluted printouts that I can't make heads or tails
>of.

Certainly you can recognize the creation of an multi-deminsional array
of data, writing and reading it to and from a file as well as test
prints. And the manipulation of the data can't possibly be out of your
realm of understanding. Of course you do know what you are looking at is
the framework of taking command line arguements, parsing them and then
applying the manipulations to the data array. One big select statement to
handle the many options/commands, where the functionality that will be
tied to the selected option can be filled in later.

Programming one step at a time. This step is just the main parser.

Certainly a Software engineer such as yourself knows this to such a degree
that it's obvious to you. And as far as arexx goes....well Like I said, it
doesn't go thru the complete framework, But it doesn't mean there is not a
framework spelled out in english.

>Since you can't seem to make us understand the value of V.I.C. from a
>"whitebox" look at it, let's try a "blackbox" approach.  In a
>"blackbox" approach, you give some examples of (expected) input and
>(expected) output from V.I.C. and leave it to the reader to figure out
>how V.I.C. will (or should) work.  Obviously, for this to work, your
>examples must be well thought out and very concrete.  They must show
>the usefulness of what V.I.C. does and they shouldn't need much in
>explanation of how the input became the output.  And the more examples
>you can provide, the better are the chances that someone will
>understand what you're trying to do and develop the program to do it.


take the math symbols 1 2 3 4 5 6 7 8 9 0 - + = x /, etc..

input is 156, 963, 87753.98635 and the output is 3

Show me how these symbols, these tools can do this.

We both know it can be done, and in more than one way. At this stage of
development, that is all that is needed. Once the tool is built then we
can focus on getting the tool to output what we want from the input we
provide it with.

In either case, more information is needed and that's what we work on
after the tool is created. If we find something the tool cannot handle
within the scope of what it is to handle, then just like with the science
of physics, we determine what needs to be adjusted/corrected in order to
handle all exceptions.

To ask a question like how do you get the VIC to create a Hello World
program is foolish. There is not enough information in such a question!
Just like the above math question.


>My recommendation, though, is that you put these examples on your
>web-page and let us go look at them.  Posting them here would be bad
>because they would probably get lost.

Nothing gets lost in usenet. Or at least there is alot less chance of it.

But on the other hand, I do still need to and want to take all that I've
posted here and extract and compile a web page. But right now I don't have
the time. Maybe Next week If the company I'm working for doesn't have
enough work to keep me on. (show biz)


There is no white or black box to reverse engineer or clean room create.
There is only the creation of a configuration of very common code
functionality, carried out to it's natural conclusion in support of
versatility.

The parse framework is the place to start in creating the whole VIC. And
at some point we will be able to integrate What was done in the standalone
IQ program. The PK multi-deminsional array is the main data collection
that gets manipulated.

Handle the parse routine and then fill in one by one that functions the
parse routine points to given selected input.

A great deal of thought has gone into this already, alot more has been
thoughtout and defined than this parse routine framework. Enough to know
doing the parse routine is no waste of effort, but something that needs to
be done.

<shrug>



>--
>David Masterson                dmaster AT synopsys DOT com
>Sr. R&D Engineer               Synopsys, Inc.
>Software Engineering           Sunnyvale, CA


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From wolfson at uchicago.edu  Tue Feb 26 16:05:04 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Tue, 26 Feb 2002 15:05:04 -0600
Subject: Time to rename Stackless?
References: <m3pu2sancq.fsf@chvatal.cbbrowne.com> <a5fqbi$p5q$1@thorium.cix.co.uk>
Message-ID: <pan.2002.02.26.15.05.03.465651.1294@uchicago.edu>

On Tue, 26 Feb 2002 05:08:34 -0600, gbreed wrote:

>> There have got to be some puns heading in the direction of "seriously
>> stacked."  Hmm...
> 
> Wonder Python!

Builds strong threads and coroutines twelve ways?

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From sholden at holdenweb.com  Mon Feb  4 10:04:11 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 10:04:11 -0500
Subject: Making a list-inhereted object return standard repr?
References: <mailman.1012777326.7083.python-list@python.org>
Message-ID: <fsx78.39685$KM2.1225436@atlpnn01.usenetserver.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote ...
> This brings up the interesting question of whether there should be a an
> object method which returns the "canonical repr" of an object i.e. the
> <module.Class object at address> or possibly even a builtin method (so it
> would work with classic classes too).
>
> I know I've wanted to use this in classic classes many times for debugging
> purposes - I want to be easily able to distinguish instances from each
> other. Each time I end up having to code it myself.
>
Although, of course, even this would not necessarily be unambigous if you
had code that produced classes from factory functions, as you would then
have to identify the class by address rather than by name (each class
returned by the factory having either the same name or no name at all,
depending on your point of view).

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From tc at clark.net  Mon Feb 11 11:09:49 2002
From: tc at clark.net (Thomas R. Corbin)
Date: Mon, 11 Feb 2002 16:09:49 GMT
Subject: XP (was Re: Defending the Python lanuage... )
References: <3C62EE4A.5090305@destiny.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com> <mailman.1013434699.26587.python-list@python.org>
Message-ID: <h1S98.86429$A51.36785660@typhoon.southeast.rr.com>

Chris Gonnerman wrote:

> ----- Original Message -----
> From: "Laura Creighton" <lac at strakt.com>
>> 
>> Don't formalise the process too much.  You are not trying to bring
>> order to your establishment by building rules that work top-down.
>> You are trying to foster order that works from the bottom up.
> 
> Here is the ideal of XP which I think is being lost on managers
> trying to implement it for buzzword compliance.  They don't really
> understand what they are doing, so they will *mandate* all parts
> of XP even if they are incompatible with the people doing the work.
> 
> XP is only going to work if you adapt it together with your people.
> Each side may have to give; if the programmers you are working with
> are hardheaded (and which of us aren't), forcing them to pair-program
> or do other XP things is *not* going to improve anything.

        No - but if you are going to adopt XP, you might as well go all the 
way.  The XP folks say that just adopting part of XP just isn't as 
efficacious as doing the whole thing because each piece supports the other 
pieces.   Leaving out one pieces is leaving out one wall of a house, it's
sure to fall down.

        Of course, they also say, "season to taste".


> 
> I've lived through too many "paradigm shifts" already.  I remember
> the fiasco of TQM being applied to a scientific laboratory, where
> commitment to quality was already so high that the technicians were
> insulted.  Managers blindly applying buzzwords like Band-Aids makes
> me curse... whether it's TQM, or Java, or XP, *no* "new way" will
> work for everyone.
> 
> It's just not a one-size-fits-all world.



From mhammond at skippinet.com.au  Sat Feb 16 00:36:45 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 16 Feb 2002 05:36:45 GMT
Subject: impossible to reboot OS in python?
References: <a4k9as$ak8$1@nntp.itservices.ubc.ca> <3C6DB92F.50107@skippinet.com.au>
Message-ID: <3C6DF059.5050805@skippinet.com.au>

Mark Hammond wrote:
> For Windows NT/2000, do a google search - I have posted code that does 
> this a number of times.
> 
> For Win9x, MSDN documents that you can execute:

Oops - and as Chris pointed out, first check if win32api.ExitWindows(Ex) 
will do what you need :)

Mark.



From eric.brunel at pragmadev.com  Fri Feb  1 08:54:24 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Fri, 1 Feb 2002 14:54:24 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de>
Message-ID: <a3e6g8$18ip$1@norfair.nerim.net>

Hi Johannes,

> I want to create widgets dynamically. Therefore I have created a
> dictionary for the options (well, it is a list of lists of lists which
> have this dictionary included, but I will keep it simple here). The
> problem is that not only the options are unknown at compile time, but
> also the *option tags*. For example in Button(root, text='Hello') also
> the *text* option tag should be taken from the keys of the options
> dictionary.

Here are two simple solutions:

- If I remember well, all Tkinter widgets accept configuration at creation
time using either named parameters, or a dictionary. So doing:
b = Button(root, text='Hello')
or:
b = Button(root, {'text': 'Hello'})
should have the same result (notice the quotes around 'text'...)

- If it doesn't succeed, here's a solution that will work: just call the
"config" method not directly, but using "apply":
b = Button(root)
apply(b.config, (), {'text' : 'Hello'})
The empty tuple () in the second argument just means "no positional
parameters". The last argument gives the named parameters as a dictionary
(again, notice the quotes around 'text'). So the last line above is the
exact equivalent of:
b.config(text='Hello')

HTH
 - eric -




From aahz at panix.com  Fri Feb  8 10:47:10 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 07:47:10 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013158814.22949.python-list@python.org>
Message-ID: <a40rtu$b0r$1@panix2.panix.com>

In article <mailman.1013158814.22949.python-list at python.org>,
Simon Brunning  <SBrunning at trisystems.co.uk> wrote:
> From:	aahz at panix.com [SMTP:aahz at panix.com]
>>
>> Heh.  My preferred working style is to use two computers, one with each
>> hand when I'm not doing anything that requires actual typing.  And, of
>> course, I use one mouse in each hand.
> 
>Only two? Haven't you got feet?

Haven't figured out a good/cheap way to put more than two monitors on a
desk, particularly when both are 17" or larger.  It's hard enough
stuffing *one* monitor on most crapola office desks.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From andy47 at halfcooked.com  Mon Feb 25 18:27:13 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Mon, 25 Feb 2002 23:27:13 +0000 (UTC)
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de>
Message-ID: <Xns91C16A332ADFCandy47halfcookedcom@203.109.252.31>

"Emile van Sebille" <emile at fenx.com> wrote in news:a5ef3h$6p6g1$1 at ID-
11957.news.dfncis.de:

> "Joshua Muskovitz" <joshm at taconic.net> wrote in message
> news:3c7ab974_4 at corp.newsgroups.com...
>> Hey all,
>>
>> If you had to give a 15 minute introduction to Python, what would you
>> cover? What is a good example of a simple, but useful, text-based app
>> for a Wintel environment, that has good "wow" quality?
>>
> 
> Hmm... text-based... "wow" appeal... oxymoron?  ;-)
> 
> Otherwise, what Sheila said.
> 
> --
> 
> Emile van Sebille
> emile at fenx.com
> 
> 
> 

With a nod to the great man, how about borrowing Mark Pilgrim's roman 
numeral converter from Dive into Python  (http://diveintopython.org/). If 
you want to get really ambitious you could add an XML-RPC or SOAP call to a 
remote server and convert the stock price of a local corporation or the 
some local weather information (temperature usually) to roman numerals on 
the fly.

Regards,
Andy
-- 
Contents free posts a speciality


From ron.l.johnson at cox.net  Wed Feb 20 23:32:00 2002
From: ron.l.johnson at cox.net (Ron Johnson)
Date: Thu, 21 Feb 2002 04:32:00 GMT
Subject: No space left on device
References: <mailman.1014247475.13555.python-list@python.org>
Message-ID: <20020220223156.2ea3a905.ron.l.johnson@cox.net>

Creating 100,000 files like a pretty sub-optimal solution to almost
any problem.  Have you considered using a database, or an indexed file?
Maybe 1 file where each chunk of data is appended to the end, and a 
2nd file containing offset pointers into The Really Big File.

If, on the other hand, creating 100,000 small files is the best, or
only, way to do it, then think about moving to linux and using the
Reiser FileSystem.  It does _extremely_ well with small files.  And,
as a side-benefit, it is journaled, like NTFS.

On Wed, 20 Feb 2002 16:23:13 -0700 sag at hydrosphere.com (Sue Giller) wrote:

> Curiouser and curiouser
> 
> I don't think the problem is in the space on the drive.
> 
> I can get the program to run completely with no slowdown if I 
> generate filenames without using the mktemp function
> 	fileName = base + number + extension
> 
> I can also get the program to run completely, with no slowdown, if I 
> generate filenames, BUT don't make any files.
> 
> There seems to be some issue with the combination of using 
> mktemp to get a unique filename (the slowdown is probably due to 
> some sort of checking mktemp does for each name generation) and 
> also creating a file.
> 
> My original code generated a series of temp files using one of three 
> different seeds, but it failed also at the 21800 level.  So changing the 
> seeds doesn't cure this.
> 
> I will try a test where I build a directory tree and see if that fixes 
> things.  Currently, I am inclined to think there is something in 
> mktemp and my computer (ram size or some such) that might be the 
> problem.  I need to try this on a drive not on my main machine  as 
> well.
> 
> Thanks for replies.
> 


-- 
+------------------------------------------------------------+
| Ron Johnson, Jr.        Home: ron.l.johnson at cox.net        |
| Jefferson, LA  USA      http://ronandheather.dhs.org:81    |
|                                                            |
| 484,246 sq mi are needed for 6 billion people to live, 4   !
! persons per lot, in lots that are 60'x150'.                |
! That is ~ California, Texas and Missouri.                  !
! Alternatively, France, Spain and The United Kingdom.       |
+------------------------------------------------------------+


From mrchameleon at hotmail.com  Sat Feb 23 00:05:35 2002
From: mrchameleon at hotmail.com (Chris Reay)
Date: 22 Feb 2002 21:05:35 -0800
Subject: Python and Daemons
References: <02022215115601.01886@localhost.localdomain> <mailman.1014416559.28231.python-list@python.org>
Message-ID: <7652139e.0202222105.1220442c@posting.google.com>

Justin Sheehy <justin at iago.org> wrote in message news:<mailman.1014416559.28231.python-list at python.org>...
> Dieter Lunn <theguru at mb.sympatico.ca> writes:
> 
> > I was wondering how to write a linux daemon using Python.
> 
> This is what I use:
> 
... and an attribution to W. Richard Stevens is also considered good mojo.

Yours

Chris


From db3l at fitlinxx.com  Fri Feb  1 19:24:37 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Feb 2002 19:24:37 -0500
Subject: Windows, RS232, and Python
References: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org>
Message-ID: <ur8o4spbu.fsf@ctwd0143.fitlinxx.com>

Spencer Ernest Doidge <spencer at efn.org> writes:

> I want to have control of every aspect of RS232 transmitting and
> receiving in Win9x/NT with Python commands.

Well, the latest win32all packages pretty much wraps all the Win32 API
functions provided for managing the serial port, so you should be able
to do it the same way you would with Win32 API calls, just using that
wrapper.  Documentation for win32all contains any Python-specifics in
the wrappers, but you would use the standard Win32 API documentation
for the calls themselves.

> A. Extend Python as needed, then embed this extended Python into my GUI
> app. This would necessitate maintaining two applications: the extended
> command-line Python for the behind-the-scenes users, and the GUI app for
> the customers.

I'm not quite sure what you mean by "this extended Python".
Typically, when one talks about "extending" Python they are referring
to writing an extension module.  Such a module acts just like a
standard Python module and can be imported into standard production
interpreters and code.

So you could just write an extension module that covers any APIs that
aren't already wrapped.  Then just import that module into your
overall script and you're set.  This is the approach we took before
the serial API calls were fully wrapped in win32all, and is similar to
some of the other serial access packages mentioned here at times.

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


From emile at fenx.com  Fri Feb  1 14:38:12 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 11:38:12 -0800
Subject: telnetlib read_until()( question
References: <u5lqt76m0ogfd8@corp.supernews.com>
Message-ID: <a3er56$17h5lp$1@ID-11957.news.dfncis.de>

"DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
news:u5lqt76m0ogfd8 at corp.supernews.com...
> For telnetlib's read_until(expected [,timeout])
>
> a.  Does anyone know the default value of the timeout when no timeout
is
> given, and
> b.  When a timeout is given:  Is the numeric value in seconds or
mintues?

telnetlib uses select.select, passing the timeout value.  From the
select docs:

FUNCTIONS
    select(...)
        select(rlist, wlist, xlist[, timeout]) -> (rlist, wlist, xlist)
[snip]
        The optional 4th argument specifies a timeout in seconds; it may
be
        a floating point number to specify fractions of seconds.  If it
is absent
        or None, the call will never time out.


HTH,

--

Emile van Sebille
emile at fenx.com

---------



From daniels at DevelopNET.com  Tue Feb 12 15:21:52 2002
From: daniels at DevelopNET.com (Scott David Daniels)
Date: 12 Feb 2002 12:21:52 -0800
Subject: Drawing in "Slow motion".
References: <3C692698.C9DAD820@inenco.no> <a4ba6i$14bg$1@norfair.nerim.net>
Message-ID: <5bb527bf.0202121221.1b20e868@posting.google.com>

Eric Brunel <eric.brunel at pragmadev.com> wrote in message news:<a4ba6i$14bg$1 at norfair.nerim.net>...
> Hans Kristian Ruud wrote:
> > ...(The actual drawing procedure is triggered by an event bound to a
> > Button).
> > My problem is that I would like the program to draw one line at a time;
> > then pause, then draw the next line... and so on....
> Two solutions come to my mind:
> - keep your time.sleep calls, but call the update_idletasks method on the 
> canvas after each sleep. The call should redraw the screen.
> - a bit trickier: use the delayed callback facility in Tk via the "after" 
> method to draw while your application is running...

Here is a solution that requires less "program inversion" on your 
actual drawing.  The "yield" of the drawing generator is used to do
the inversion.  Obviously, 2.2 is required.  Code shamelessly stolen 
from Eric Brunel's fine example.

from __future__ import generators
import random
from Tkinter import *

w, h = 500, 500
root = Tk()
cnv = Canvas(root, width=w, height=h)
cnv.pack(side=TOP)

def draw(count):
    """A generator that draws with a 'yield' at each time point."""
    x, y = random.randrange(w), random.randrange(h)
    for i in range(count):
        oldx, oldy = x, y
        x, y = random.randrange(w), random.randrange(h)
        cnv.create_line(oldx, oldy, x, y)
        yield None

# Put in a button to walk through the drawing
Button(root, text='Click', command=draws(10).next).pack(side=TOP)

# Put in a button to do a "timed" walk through the drawing

def per_tick():
    """This gets executed once per 'tick' to do the drawing"""
    try:
        draw_step()
        root.after(500, per_tick)
    except StopIteration:
        pass

def timed():
    "Once per button push: set up the generator for per_tick, and start it."""
    global draw_step
    draw_step = draw(17).next
    root.after(500, per_tick)

Button(root, text='Timed', command=timed).pack(side=TOP)


daniels at DevelopNET.com


From starx at pacbell.net  Tue Feb  5 02:57:15 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 5 Feb 2002 07:57:15 GMT
Subject: tt = a, b = t
References: <5bl34uc7rumgpnbloe4o5cttotuiaj99g0@4ax.com> <mmm34u4vd1no3r56lofsmj83v51dbad6i3@4ax.com> <qfm34uch7hrkc46jqqndpqc9nb657pv49j@4ax.com> <FAA08.11363$k87.1258193@atlpnn01.usenetserver.com>
Message-ID: <Xns91ABF3ABE9041RASXnewsDFE1@130.133.1.4>

Steve Holden || Mon 14 Jan 2002 04:54:38a:

> "rihad" <rihad at mail.ru> wrote in message
> news:qfm34uch7hrkc46jqqndpqc9nb657pv49j at 4ax.com...
>> On Sun, 13 Jan 2002 18:58:25 GMT, Courageous <jkraska at san.rr.com>
>> wrote: 
>>
>> >>>> None
>> >2
>> >>>>
>> Huh? This works! What are the reasons behind allowing to assign to a
>> language construct???
>>
> This is Pythonicity at its most Pythonic. In actual fact the singleton
> value None is by default associated with the name __builtins__.None --
> which is to say, any program component will be able to access this
> value, assuming that the name None does not exist in any namespace
> taking prioirity over __builtins__, which is the namespace of last
> resort. 
> 
> So normally, when you execute a statement equivalent to
> 
>     None = "Hi, Mom!"
> 
> you're creating a *local* binding to the name None, which your other
> code will access in perference to the built-in as a result of the
> namespace searching rules.
> 
> [ Other Bad Things... ]

I got bit by this quite accidentally as well. I wrote 'locals = globals()' 
(or some such, preparing to spawn a debug shell), and it spits back 'type 
module not callable' or something to that effect. About twenty minutes 
later i realized it was because i have a project module named 'globals.py' 
that's imported here, and that's superceeding it. At least you can always 
do __builtins__.x() :).

(FWIW, i -was- careful when i named this, i went over the reserved words 
and language reference and all that, but i didn't think to check builtins 
and i'd never heard of this function. I should probably change the name but 
I can't really think of anything else that is short enough (maybe just glo 
:) )

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From gh_pythonlist at gmx.de  Tue Feb 12 00:36:19 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 12 Feb 2002 06:36:19 +0100
Subject: Python MySQL and MAC OS X
In-Reply-To: <20020211192310.GC1686@lilith.hqd-internal>
References: <3C656ACD.1020907@inter.nl.net> <mailman.1013294776.15973.python-list@python.org> <3C663807.AA608724@xs4all.nl> <a487t7$qhb$1@highway.leidenuniv.nl> <20020211192310.GC1686@lilith.hqd-internal>
Message-ID: <20020212053618.GA1648@lilith.hqd-internal>

Le 11/02/02 ? 20:23, Gerhard H?ring ?crivit:
> Le 11/02/02 ? 10:54, Peter H.J. v.d. Kamp ?crivit:
> > [building MySQLdb on MacOS X]
> I have almost zero knowledge of Macs, but AFAIK MacPython is built with
> CodeWarrior, so you'll probably need this to compile MySQLdb.

> I strongly suggest you use the unixoid Python for MacOS X, as it will
> allow you to use GNU autoconf et al. for building the MySQL client
> libraries and gcc for building the C part of the MySQLdb Python
> extension.

For the fun of it, I built Python 2.2 for MacOS X on Sourceforge's
compilefarm, it compiled out of the box, but from what I've heard, it
still has some issues.

> [...] 
> Just recognized that I have access to MacOS X on Sourceforge's compile
> farm. I'll try to build a MySQLdb.

There's also a prebuilt unixoid Python for MacOS X available here:

    http://tony.lownds.com/macosx/

Personally, I'd just install the development stuff from the MacOS X CD
and recompile Python 2.2 myself. Then install the MacOS X MySQL
binaries, then compile and install MySQLdb.

You could also give my compilation a try, I'put it on my homepage:
http://www.cs.fhm.edu/~ifw00065/

But I've no idea if it will work on your machine. Of course, I also
cannot support this in any way, as I don't even have a Mac.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.9 ?C      Wind: 8.6 m/s



From Arnaud.De_La_Fortelle at inria.fr  Wed Feb 27 13:08:44 2002
From: Arnaud.De_La_Fortelle at inria.fr (Arnaud de La Fortelle)
Date: 27 Feb 2002 19:08:44 +0100
Subject: wxPython blocks in thread?
Message-ID: <rk63czmu79v.fsf@pegase.inria.fr>

I am a beginner with wxPython and tried to embed a GUI within a
thread.  So I came with a simple implementation, but it does not work.
I do not understand why the following application blocks. What the
matter?

Arnaud

from wxPython.wx import *
class MyApp(wxApp):
  def OnInit(self):
    frame = wxFrame(NULL, -1, "It blocks!")
    frame.Show(true)
    self.SetTopWindow(frame)
    return true

import threading
class GUIThread(threading.Thread,object):
  def __init__(self):
    threading.Thread.__init__(self,target=self.loop)
    self.start()
  def loop(self):
    print gui_thr
    self.app = MyApp(0)
    self.app.MainLoop() # if I comment out this line, it works
    print "Ending wxLoop", self.isAlive()


gui_thr=GUIThread()
print "Ending Main thread"


From redskin at tut.by  Sun Feb 24 16:57:03 2002
From: redskin at tut.by (Andrey Sidorenko)
Date: 24 Feb 2002 13:57:03 -0800
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com> <mailman.1014533499.3218.python-list@python.org>
Message-ID: <26428104.0202241357.61c3035c@posting.google.com>

Kirill Simonov <kirill_simonov at mail.ru> wrote in message news:<mailman.1014533499.3218.python-list at python.org>...
> On Sat, Feb 23, 2002 at 04:57:19PM -0800, Andrey Sidorenko wrote:
> > I use Active Python 2.2 on Windows 2000 and have trouble - "Entry"
> > class can't display cyrillic letters (cp1251 encoding) correctly in
> > entry field. Can anybody help me with this weird problem?
> 
> Try to replace "Python22\Lib\lib-tk\FixTk.py" with the following:
> 
> import sys, os
> 
> if not os.environ.has_key('TCL_LIBRARY'):
>     tcl_library = os.path.join(sys.prefix, "tcl", "tclX.Y")
>     os.environ['TCL_LIBRARY'] = tcl_library
> 
> 
> Kirill.

"It's working, it's working!" (c) Anakin Skywalker ;-)

Thank you very much.


From roy at panix.com  Sat Feb 16 18:08:55 2002
From: roy at panix.com (Roy Smith)
Date: Sat, 16 Feb 2002 18:08:55 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <ad052e5c.0202161240.73499031@posting.google.com>
Message-ID: <roy-2F6171.18085516022002@reader2.panix.com>

danb_83 at yahoo.com (Dan Bishop) wrote:
> Everyone should learn BASIC.  It's the perfect example of how *not* to
> design a programming language :-)

I learned basic in high school, in 1975 or so.  We had an ASR-33 and a 110 
baud modem connecting us to a time-share system a few towns away.  Compared 
to what else was available at the time (fortran, PL/1, cobol, or assembler, 
most likely on punch cards), it was an excellent system for introducing 
somebody to programming.

Granted, the syntax and data structures aren't up to what we now consider 
modern, 25 years later.  But, again, compare it to its competitors at the 
time, and it looks pretty good.  It certainly got you used to the 
fundamental concepts of programming which are still valid today -- 
variables as containers, looping and conditional flow of control, input and 
output, subroutines and function calls.  And, of course, the instant 
feedback of an interpreted language.

I can remember teaching a 10-week class in basic to a bunch of scientists 
and graduate students in the mid 80's who had no other computer training.  
I used it as a springboard to talk about algorithms.  I had my students 
program a couple of different kinds of simple sorting algorithms and 
compare how long they took to run.  Surely, if the language is simple 
enough to learn that you can go from zero to comparitive experiments in 
algorithmic complexity in just a few weeks, there's something to be said 
for it?

Why people are still teaching or using basic today is beyond me, since 
there are so many better alternatives.  Of course, some of the basic 
variations around today are so different from what I taught myself 25 years 
ago, they might as well be different languages.


From fperez528 at yahoo.com  Sun Feb  3 08:41:52 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 03 Feb 2002 13:41:52 +0000
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org> <3c5d4c21.38617375@news.t-online.de> <mailman.1012752906.26372.python-list@python.org> <3c5d624d.44293625@news.t-online.de>
Message-ID: <a3k7cn$ogl$1@peabody.colorado.edu>

Gerson Kurz wrote:

> And, of course, my examples were not taken "from real life". There is
> however one nice usage of lambda: passing context.

[snip]

Quick comment. I actually need this frequently precisely for numerical 
methods, but prefer to use the alternate approach of writing the context-part 
as extra arguments. The reason is that the lambda approach generates an extra 
nested function call for each of your original function calls, and in a 
numerical algorithm this can be very expensive (particularly considering how 
expensive function calls are in Python).

A quick example:

#-----------------------------------------------------------------------------
def constrained_svdfit(x,y,sig,fix_i,fix_v,funcs,*args,**kw):
    
    # this will be a (num_params X num_datapts) matrix
    phi_base=funcs(x,*args,**kw)

    [ snipped rest of code...]


Here, funcs is a function that returns a set of fit functions evaluated at 
the array of input data x. All extra context is passed to funcs via *args and 
**kw. If on top of that I also want constrained_svdfit() to take keyword 
args, I just set up a naming convention so that they all begin with _svd_ or 
something similar, and strip them from the **kw dict before passing it to 
funcs().


Cheers,

f


PS. If by any chance the actual code is of interest to you, just email me. I 
removed the numerical details from this post but the above is taken from 
actual working code.


From jjl at pobox.com  Sat Feb  2 17:22:05 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 22:22:05 +0000
Subject: Python System Administration
In-Reply-To: <3c5c241a$0$2203$7586b60c@news.frii.com>
References: <LwK68.7172$Xk4.381742@news1.east.cox.net> <3c5c241a$0$2203$7586b60c@news.frii.com>
Message-ID: <Pine.SOL.4.30.0202022215560.597-100000@mimosa.csv.warwick.ac.uk>

On Sat, 2 Feb 2002, Marc Poulin wrote:

> In article <LwK68.7172$Xk4.381742 at news1.east.cox.net>, "Andrew Replogle"
> <replogle2 at cox.net> wrote:
>
> > Anyone know of any books out there for python system administration or
> > should I just stick to perl for these tasks?
>
> I've been wondering the same thing.I searched Amazon and found
>
> 	Perl for System Administration by David N. Blank-Edelman
> 	Automating Windows With Perl by Scott McMahan
>
> but nothing for Python.
[...]

Which system?  Windows, unix...?

Of course, the standard library itself contains reasonable documentation
for everthing there, which covers a lot of unix stuff, but if that isn't
enough, Beazley's book is often spoken well of, for example.  The O'Reilly
Windows Python book (forget the name) by Hammond and Robinson is essential
for doing anything Windows-specific with Python, and has a chapter or two
on system administration.  Non python-specific books will certainly be
applicable too, of course.  Don't know about the Mac.


John



From tdelaney at avaya.com  Mon Feb 25 21:28:25 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 26 Feb 2002 13:28:25 +1100
Subject: Time to rename Stackless?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1AE@natasha.auslabs.avaya.com>

> From: Greg Ewing [mailto:greg at cosc.canterbury.ac.nz]
> Christian Tismer <tismer at tismer.com>:
> 
> > It is still stackless in the sense that order of execution is not
> > restricted by the C stack.
> 
> Even so, I still think it's a poor name, because it
> focuses on an implementation detail rather than
> telling you what it can do for you.
> 
> > We have now tasklets, proclets, and the stack slices are
> > stacklets. :)
> 
> Hmmm... maybe it should be called Pythonlet...

Pylet (Pilot).

Tim Delaney



From shalehperry at attbi.com  Sun Feb 10 19:14:58 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Sun, 10 Feb 2002 16:14:58 -0800 (PST)
Subject: No explanation for weird behavior in re module!
In-Reply-To: <a470re$1cl30l$1@ID-78052.news.dfncis.de>
Message-ID: <XFMail.20020210161458.shalehperry@attbi.com>

On 11-Feb-2002 synthespian wrote:
> Hi-
> 
>       I'm really intrigued by this behavior:
> 
>>>> import re
>>>> p = re.compile('^(der|die|das(\s\w+))')

The \w symbol makes decisions based on your locale.  So, set LC_ALL=de_DE and
umlauts should work.  This assumes you have the german locale installed (under
Debian it is /etc/locale.gen which decides this).  More info on \w can be found
in the re module docs.

sre is simply the faster implementation of re, you are probably using it unless
you force python not to do so.



From jeff at ccvcorp.com  Fri Feb  8 15:30:24 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 12:30:24 -0800
Subject: range(seq)?
References: <slrna67foh.dl7.mlh@vier.idi.ntnu.no>
Message-ID: <3C643560.B6FF51A0@ccvcorp.com>


Magnus Lie Hetland wrote:

>   for x in range(seq): ...
>
> I.e. let range() take a sequence (or iterable) as an argument, and
> generate range(len(seq)).

This seems more logical and intuitive to me, than the previous
suggestions.

Whether that's enough to justify its being added to Python is another
question entirely, of course.  ;)

Jeff Shannon
Technician/Programmer
Credit International




From rxg218 at psu.edu  Wed Feb 20 14:46:55 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Wed, 20 Feb 2002 14:46:55 -0500
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net>
Message-ID: <a50udo$1dou@r02n01.cac.psu.edu>

On Wednesday 20 February 2002 01:24 in comp.lang.python Jeff Hinrichs wrote:

> If you wanted to get away from the dangerous eval, you could put your
> functions inside of a class and then,

Why is eval dangerous?

-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
Dept. Of Chemistry             | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
There is no truth to the allegation that statisticians are mean.
They are just your standard normal deviates.



From tripps81 at yahoo.com  Sat Feb 23 03:34:01 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Sat, 23 Feb 2002 15:34:01 +0700
Subject: "pass by reference?"
Message-ID: <5.1.0.14.2.20020223153217.04178e80@202.95.131.154>

Suppose I want to make a function that modifies its argument:

  a = -1
  make_abs(a)
  print a # = 1

What is the Pythonic way of doing this? Wrap it in a list?

  make_abs([a])

t




From jkraska at san.rr.com  Tue Feb 19 01:45:10 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 19 Feb 2002 06:45:10 GMT
Subject: Adding static typing to Python
References: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com> <mailman.1014092911.31096.python-list@python.org>
Message-ID: <gft37u8csm1ukd923nr933iq7meimqu5i7@4ax.com>

>> Python also lacks features of "data hiding", it attempts to provide
>> some support for it but only in a desultory "procedural" way - try
>> this with 40 programmers and you won't get far! :-)
>
>This is silly - let me guess, ...

One of the things I like about C sharp is their use of attributes,
where you can define getters and setters on an attribute, but
otherwise the attribute is simply used as if it were a public one.
An objects attributes can thereby be used as a published aspect of
its interface safely, but no one will be the wiser if you manipulate
the getter and setter code.

This has to be one of the best ideas I've seen in encapsulation in
quite some time. It's encapsulated, but it isn't. It's syntactically
compact, but can be used to do a significant job.



C//



From ullrich at math.okstate.edu  Fri Feb  8 13:54:57 2002
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Fri, 08 Feb 2002 18:54:57 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <3c641e04.3543807919@news>

On 7 Feb 2002 17:18:12 GMT, "Donn Cave" <donn at u.washington.edu> wrote:

>Quoth ullrich at math.okstate.edu (David C. Ullrich):
>
>| An amusing aspect of all this: I just browse clp occaisionally,
>| since I'm an abstraction manipulator, not a programmer. So I
>| overheard comments about this guy before seeing any examples
>| of his output. Before seeing the things he wrote I was gonna
>| say "you guys got it easy - you should see sci.math". But
>| the funny thing is the guy reminds me _exactly_ of the
>| sci.math guy I had in mind: If you can't see what a genius
>| he is you're incompetent, you're all gonna be fired when
>| the Truth comes out, etc...
>
>So is sci.math choked with posts from people who normally seem
>to have their heads screwed on, endlessly replying this guy as
>if they sincerely think there's some purpose in it? 

Yup.

> People who
>take time several times every day to post about how useless his
>ideas are and how they have no interest in helping?  Or are
>they saner there than we are?

People take turns. Everyone agrees there's no point to replying,
but at any given time there are plenty of people with nothing
better to do (or too pissed off at the guy to be rational about
it, or both.)

>The existence of people with unusual ideas isn't surprising.

So far you haven't been invaded by anyone insisting that
there's information about computer programming in the
Vedas that far surpasses modern methods in power. I haven't
seen anyone say that computer programmers are more likely
than most people to be mass murderers. Could be a lot worse.

>I'm kind of taken aback by the response here, though.
>
>	Donn Cave, donn at u.washington.edu


David C. Ullrich


From bcboy at thecraftstudio.com  Fri Feb  8 16:11:57 2002
From: bcboy at thecraftstudio.com (Brian Craft)
Date: Fri, 8 Feb 2002 13:11:57 -0800
Subject: threading problem
Message-ID: <20020208131157.A20149@bcboy-linux.cisco.com>

Hi all --

I'm trying to use pygtk and orbit-python. They won't work together for reasons
I don't fully understand, but seem to have something to do with the python
threading model.

Both of these libraries work by providing a wrapper to invoke a glib main loop.
Since both orbit and gtk use the glib main loop, they can work together. The
main loop gets events (corba or gtk) and calls into the python interpreter to
handle them.

Both of them attempt to handle threading by using PyEval_SaveThread(), etc.,
like so:

  o The script invokes orb.run(), or gtk.mainloop()
  o   PyEval_SaveThread() is called.
  o   g_main_run() is invoked via the corresponding library.
  o     An event occurs, and a callback is made into pygtk or orbit-python.
  o       PyEval_RestoreThread() is called.
  o       A python callback for the event is invoked.
  o       PyEval_SaveThread() is called.
  o     The handler returns control to glib main.
  o   An event may exit the glib main loop (if user invokes File->quit)
  o   PyEval_RestoreThread is called.
  o Control returns to the python script.

It appears to me that what's happening is PyEval_RestoreThread() is being
invoked with a bogus value. If you initialize both libraries and call
gtk.mainloop(), then pygtk stores the state info. Then when a corba event
occurs, orbit-python attempts to restore the state info (which is uninitialized
in that library) and barfs with a thread error on this code:

        dict = PyObject_GetAttrString(servant->impl,"__dict__");

I don't understand these calls at all, so I'm not sure how this should be
fixed. In addition, pygtk uses these sick and wrong constructs, like:

         if (counter == -INITIAL_LOCK_COUNT) {                               \
           PyThreadState *_save;                                             \
           Py_UNBLOCK_THREADS;                                               \
           g_static_private_set(&pythreadstate_key, _save, NULL);            \
         }                                                                   \
 
where a variable defined in a macro is being referenced, for extra clarity.

Anyone with a clue?

Btw, putting the main loop in python isn't an option, because orbit doesn't
support a polling mode.

b.c.



From gerson.kurz at t-online.de  Thu Feb 21 04:55:01 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 09:55:01 GMT
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr>
Message-ID: <3c74c3d6.20466328@news.isar.de>

I found an easier solution:

sort = lambda x:(x.sort(),x)[1]



From gustav at morpheus.demon.co.uk  Thu Feb 14 16:34:22 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Thu, 14 Feb 2002 21:34:22 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk>
Message-ID: <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com>

On Thu, 14 Feb 2002 17:56:11 +0000, Robin Becker
<robin at jessikat.fsnet.co.uk> wrote:

>well his .pyz is a compressed archive form and I have been using it to compress the entire
>python library. All that's required is a special importer to allow imports from such an archive
>to work. I believe that the archive is allowed to contain data files as well so provided you
>alter the open/read you can at least consider doing that.

Hmm. Looks more messy than I'd hoped. Needs a bootstrap module, and then
extra hacking to handle open/read. And I guess the fact that I'm doing
this manually means that there's not much chance of something like a
wxPython toolbar "just working" with icons stored in the archive.

I guess it's not worth the hassle.

The reason I liked the Java approach, BTW, is that a complete
application comes in a single file, which you can just double click on
and it works. Nothing to unpack, put it anywhere you like, run off a
floppy/CD/zip disk, no dependencies (other than needing Java installed).
But it's just a convenience thing - not enough to endure writing all my
code in Java :-)

Nice to keep an eye on what the competition can do, though...

Thanks for the explanation,
Paul.


From gcordova at hebmex.com  Fri Feb 15 16:36:18 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 15:36:18 -0600
Subject: MemoryError using NumPy
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A91@SRM201002>

> 
> OK, so the Python process keeps the memory. Does the memory that the
> Python process holds get fragmented? If so, same problem. If not, then
> what is going on?
> 

Whenever there isn't enough memory (reserved by the Python
runtime) it simply asks for another block, at least as
large as is needed.

Linux's VM system is responsible for that request. If there's
enough virtual memory, then it answers positivly. As the memory
is being physically used, it's being swapped in and out of ram.

So. If you need 1Gb of ram for an allocation, you could make
a 2Gb swap partition. Most probably, the memory request would
succeed, if there's enough VM free.

> Also, a few years back there was a discussion on the MATLAB newsgroup
> about a similar problem, and The Mathworks claimed that it was not
> possible to return memory to the os once a process has used 
> it. This was
> the case on Linux and I'm pretty sure, all the other *nixs that they
> supported. Perhaps it's a library issue, but is there a 
> different one(s)
> that we all should be using to compile Python?
> 
> -Chris
> 

I wouldn't really know. Allocated memory becomes fragmented,
so maybe it isn't possible to return it to the OS, since not
the whole page/block/whatever is unused.

On the other hand, maybe the Python runtime can be smart
enough to periodically shrink it's allocated memory size,
by freeing the top area of the heap whenever possible. But,
since it isn't possible to relocate an object once it's
been instantiated, it'd be up to luck to wait for a moment
when the heap has enough free memory on the top to free.

Is it possible to relocate an active object?

If not, what would be necessary to make that happen?

If it can be done, then a garbage-collector can just
pack everything at the bottom of the heap and free
all allocated memory above the new heap-top.

Anyhow.

That's just my 2 ct.

-gustavo



From greg at cosc.canterbury.ac.nz  Thu Feb 14 20:44:45 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 15 Feb 2002 14:44:45 +1300
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com>
Message-ID: <3C6C680D.B35E7719@cosc.canterbury.ac.nz>

Courageous wrote:
> 
> And while vim indeed does just what I said, when you backspace over
> these runs of spaces while in insert mode, it knows how those spaces
> where generated and so:
> 
> "deletes any such spaces inserted through a tab, just as if
> it were deleting a single tab."

This sounds ingenious, but what happens when you save the
file and load it back? Does it remember this information
somehow, or does it get lost?

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


From nospam at bigfoot.com  Wed Feb 27 09:33:38 2002
From: nospam at bigfoot.com (Gillou)
Date: Wed, 27 Feb 2002 15:33:38 +0100
Subject: SSL Server Sockets?
References: <a5ijos$1565$1@news.idiom.com> <jw5f8.129$5vh.170404352@news.telia.no>
Message-ID: <a5iqab$sji$1@norfair.nerim.net>

distutils is great but there are some missing (future ?) stuffs...

I'm dreaming about a pythonic,  platform independant, Web enabled  RPM like
utility for that purpose.
That checks dependancies, platform issues (...) and downloads/install the
proper version of a package (if available) best suited to the user's
installation.
But I'm disgressing from SSL Server Sockets...:)

--Gilles

"Thomas Weholt" <thomas at gatsoft.no> a ?crit dans le message news:
jw5f8.129$5vh.170404352 at news.telia.no...
> I got the same problem. Wish there were more precompiled binary packages
of
> M2Crypto. Why can't Python be shipped with SSL-support compiled into it?
> Could M2Crypto be included in the distro? If it's dependant on OpenSSL
> that's open source too, so why not join forces and create a real
> security-focused programming-package?
>
> The ability to compile from source is great, but when it has a lot of
> dependencies it would be easier to include the package in the standard
> distro so people don't have to hassle with it.
>
> To make life alot easier for many of us, either :
> a) get more packages into the standard distro
> or
> b) offer some precompiled and/or easy to install add-on package containing
> the most widely used modules ( PIL, M2Crypto, BSDDB 3.x-wrappers,
> StandaloneZODB, PyXML ).
>
> Just my two cents.
>
> Thomas Weholt
>
> "Jim Dennis" <jimd at vega.starshine.org> wrote in message
> news:a5ijos$1565$1 at news.idiom.com...
> > Is the native Python2.2 socket.ssl support still client only,
> >  or is it possible to use it for simple ssl servers, yet?
> >
> >  I know everyone recommends m2crypto --- but m2crypto-0.07-snap2
> >  doesn't build on my Debian box.  I get errors like:
> >
> > Generating wrappers for Python
> > _bio.i : Line 30. Syntax error in input.
> > _bio.i : Line 31. Syntax error in input.
> > _bio.i : Line 32. Variable %constant multiply defined (2nd definition
> > ignored).
> > _bio.i : Line 32. Syntax error in input.
> > _bio.i : Line 33. Variable %constant multiply defined (2nd definition
> > ignored).
> > _bio.i : Line 33. Syntax error in input.
> >
> >  ... meanwhile I can simply apt-get install python2.2-ssl
> >
> >
>
>




From mapler at erinet.com  Thu Feb  7 21:41:10 2002
From: mapler at erinet.com (RMaple)
Date: Thu, 07 Feb 2002 21:41:10 -0500
Subject: tkinter checkbutton rendering problem
Message-ID: <3c633acd$0$93061$4c5ecdc7@news.erinet.com>

Hello,
    I have something strange going on with a tkinter checkbutton. It is 
created using

Checkbutton(self, bitmap="@small_light.xbm",indicatoron=0)

In my .Xdefaults file, I have the following two items.

tk*background:                  gray80
tk*activebackground:            gray90

When the button is selected but inactive, the shadow borders of the the 
sunken representation are drawn in shades of pink!  When active (cursor 
over the button) the button is drawn correctly.  The unselected (raised) 
rendering is also drawn correctly.  Finally, if I don't set the 
activebackground color, the button is always drawn correctly. 
(Unfortunately, the default is black, which I just don't like.  Is this 
some strange colormap problem or a problem with tkinter?

I am running under Mandrake linux (8.1), python 2.1.1, truecolor X 
display.

Thanks,
Ray Maple





From sjmachin at lexicon.net  Wed Feb 27 19:11:10 2002
From: sjmachin at lexicon.net (John Machin)
Date: 27 Feb 2002 16:11:10 -0800
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz>
Message-ID: <c76ff6fc.0202271611.5366f835@posting.google.com>

Beda Kosata <kosatab at vscht.cz> wrote in message news:<3C7CF0EE.3010009 at vscht.cz>...
> Hi,
> for a piece of code I need to store relatively large amount of records 
> (100-1000).

Sorry, a 1000-record amount is relatively tiny.

> Each record will contain few (10-20) pieces of information 
> (mostly strings or integers).
> For convenience it would be better for me to make this records as dicts, 
> so I don't have to remember what data has what position in the list. 

There are *two* levels to think about: how you store the fields in
each record, and how you store the collection(s) of records

Fields in each record: for *convenience*, contemplate using a class,
not a dict.

>>> class Person(object):
...    __slots__ = ("name", "salary", "title", "empno")
...
>>> p1 = Person() # new record
>>> p1.name = "Attila the Hun"
>>> p1.salary = 1000000
>>> p1.title = "CEO"
>>> p1.empno = 666
>>> p2 = Person()
>>> p2.name = "Marmaduke Murgatroyd"
>>> p2.salary = 20000
>>> p2.tittle = "Clerk" # __slots__ gives you typo-checking on
assignments
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
AttributeError: 'Person' object has no attribute 'tittle'
>>> p2.title = "Clerk"
>>> p2.empno = 1
>>> for x in (p1, p2):
...   print x.name, x.salary, x.title
...
Attila the Hun 1000000 CEO
Marmaduke Murgatroyd 20000 Clerk
>>> print ["false", "true"][p1.salary < p2.salary]
false

Now, how you store your collection(s) of records depends on many
things, but speed is *not* one of those things when you have only 1000
records.

If the record has a unique key, such as an employee number, and you
need to access records by employee number, then you can set up a dict
with employee number as the key and the class instance as the value --
employee_dict[p1.empno] = p1 --

If you need merely to access the records sequentially, you can use a
list.

employee_list = []
employee_dict = {}
for buffer in file("employees.data"):
   fld = buffer.rstrip().split("~") # assuming data separated by "~"
   p = Person()
   p.name = fld[0]
   p.salary = int(fld[1])
   p.title = fld[2]
   p.empno = int(fld[3])
   # The above is where you do have to "remember" the
   # correspondence between positions and names
   # but this could be automated using the __slots__ and
   # a parallel sequence of functions to apply:
   # conv_funcs = (None, int, None, int)
   employee_list.append(p)
   employee_dict[p.empno] = p
# note -- above needs much error checking and exception handling
# to make it robust -- e.g. salary or empno not an int, 
# empno not unique, too few/many fields in input file, ...

> However the most important for me is speed

Speed of what? Speed of running? Speed of implementing a robust
functional application? Speed of maintenance when bugs surface or
requirements change?

Hope this helps,
John


From theivo at abv.bg  Fri Feb 15 16:38:22 2002
From: theivo at abv.bg (I.C.)
Date: 15 Feb 2002 13:38:22 -0800
Subject: How can I make a screenshot using BitBlt or something else???
Message-ID: <6ac1ecad.0202151338.15d3c4b@posting.google.com>

I am new in PythonWin programming so I have a simple question -
How can I make a screenshot of the whole screen and then get the colors
of some pixels?

Thank you very much!!!


From dsavitsk at e-coli.net  Mon Feb 11 00:04:59 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 05:04:59 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com>
Message-ID: <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com>

the truncate part is the problem.  I am building VB Hex strings to send to a
VB server.  so

>>> red = hex(56)[2:4]
>>> green = hex(156)[2:4]
>>> blue = hex(12)[2:4]
>>> thestring = '&H' + blue + green + red

but here, since the blue string is only 1 character it throws the rest off.

-d

"Joshua Muskovitz" <joshm at taconic.net> wrote in message
news:3c674cbd_2 at corp.newsgroups.com...
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> >>> ''.join(map(max,zip('00','AA')))
> 'AA'
> >>> ''.join(map(max,zip('00','A')))
> 'A'
> >>> ''.join(map(max,zip('00','')))
> ''
>
> Damn, I thought I had it.  :-)  Is there an variant of zip which doesn't
> truncate?  BTW, this does work, provided that the input strings are the
same
> length, and even allows for more than two strings!
>
> >>> s = ['abte','scrt','sdfw','cxvs']
> >>> ''.join(map(max,zip(*s)))
> 'sxvw'
>
> --
> # Joshua Muskovitz
> # joshm at taconic.net
> def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
> lyyrs('Hire me!  I need the work!')
>
>
>
>
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----




From sholden at holdenweb.com  Fri Feb  1 18:21:18 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 18:21:18 -0500
Subject: Equivalent of Perl chomp?
References: <mailman.1012403361.2370.python-list@python.org> <u5i8b8jtee2r13@corp.supernews.com> <7xbsfag023.fsf@ruckus.brouhaha.com> <3C59D2EE.3937ABBB@ndh.net> <a3f7dh$lq9$1@news.island.liu.se>
Message-ID: <isF68.111010$8b.4218974@atlpnn01.usenetserver.com>

"Thomas Bellman" <bellman at lysator.liu.se> wrote in message
news:a3f7dh$lq9$1 at news.island.liu.se...
> Stefan Schwarzer <s.schwarzer at ndh.net> writes:
>
> > It might be interesting to give some examples of code that "wants"
chomp.
> > We then could see if there is a pythonic way to make a chomp-like
function
> > unnecessary. Maybe there are better ways.
>
> I have typically needed to strip newlines when I'm processing a
> text file where each line is a record with fields separated by
> some character.  If you just .split() on the separator, you end
> up with an unwanted newline in the last field.
>
> /etc/passwd is a good example of such a file under Unix.  Sure,
> there's the pwd module, but 1) it only handles *that* format, not
> other, formats, and 2) it really is an interface to the current
> machines passwd database; if you want to process some random file
> in /etc/passwd format, you can't use it.
>
>
> Another occasion I have wanted a chomp(), is when I want to check
> if the line ends in some special character (like \).  I can't
> then just do
>
>     for line in fp.xreadlines():
> if line[-1:] == "\\":
>     blahonga(line)
>
> but instead have to do
>
>     for line in fp.xreadlines():
> if line[-2:] == "\\\n":
>     blahonga(line)
>
> and I find that less readable.
>
Perhaps I'm just lazy, but why not process

    for line in file(filename).read().split("\n"):

under such circumstances? True, this can be a pain if you need to read
Windows files under Unix or if you are using binary mode under Windows.
Otherwise, as long as the files aren't uncomfortably large, it avoids the
need to chomp.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From ykingma at accessforall.nl  Mon Feb  4 14:34:54 2002
From: ykingma at accessforall.nl (Ype Kingma)
Date: Mon, 04 Feb 2002 20:34:54 +0100
Subject: Jython question concerning actionPerformed
References: <f6b73ec8.0202040425.4c3e0bb4@posting.google.com>
Message-ID: <3C5EE256.1AE52D70@accessforall.nl>

Mishre,

you wrote:
> 
> I have the following code in Jython running on JDK 1.4, which looks
> like it should work, but isn't:
> 
> def actionPerformed(self,event):
>     if event.getActionCommand() == "Exit":
>         java.lang.System.exit(0)
>     elif event.getActionCommand() == "comboBoxChanged":
>         print event
> 
> Then later on I have:
> 
> self.catCombo = javax.swing.JComboBox(actionPerformed=self.actionPerformed)
>

Have a look at http://www.jython.org/docs/properties.html:
you can pass an ActionListener object, or a function with one argument.

You might try one of these: (untested)

self.catCombo = javax.swing.JComboBox(self)
# This assumes that actionPerformed method is available
# and that the current class implements ActionListener.

or

self.catCombo = javax.swing.JComboBox(actionPerformed = (lamda event, self=self:
                                                          self.actionPerformed(event))
# No need to inherit from ActionListener here.

I prefer to use the last form since it allows to choose a method from the current
class to be used as action listener.
You might call eg. self.exit(event) from the lambda.
 
> 
> Does anyone have any ideas on what to try next?
> 
> Thanks,
> 

My pleasure,
Ype


From wiedeman at gmx.net  Wed Feb 20 07:41:03 2002
From: wiedeman at gmx.net (Christoph Wiedemann)
Date: 20 Feb 2002 04:41:03 -0800
Subject: compiling c-extensions with distutils
Message-ID: <b2cc9907.0202200441.7056c743@posting.google.com>

Hello,

i actually have a problem building non-debug extensions (using linux).
I use 'python setup.py build_ext' to build the extensions, but the
debug flag -g is sent to the compiler. What i expected was that _only_
'python setup.py build_ext -debug' would make a debug version.
Any suggestions ?

Regards,
Christoph


From news at fairford.co.uk  Mon Feb  4 03:45:46 2002
From: news at fairford.co.uk (JH)
Date: Mon, 4 Feb 2002 08:45:46 -0000
Subject: How Can I add serial comms events to Tkinter mainloop?
Message-ID: <ORr78.26230$oK1.225674@NewsReader>

Hi,
 I am trying to write a simple serial console. I was hoping to use Tkinter
for the GUI
but I cannot see how to add RX chars from my serial port to the console
while Tk.mainloop
is running is it me ?
 thanks for any help.





From phr-n2002a at nightsong.com  Wed Feb 27 10:14:19 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 07:14:19 -0800
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com> <a5hur4$3qg$1@bob.news.rcn.net>
Message-ID: <7xk7szq7n8.fsf@ruckus.brouhaha.com>

"Raymond Hettinger" <othello,javanet at rcn.com> writes:
> 4) Code is generated along the lines of:
> if type('banana')==int:
>    <run optimized code that assumes 'banana' to be an int>
> elseif USESTRICT:
>    raise TypeError
> else:
>    <run normal, slow code which binds 'banana' to int regardless of type>
> 
> In other words, generate optimized code in case the type advice is correct;
> otherwise, run normally unless strict type checking is in effect.

But part of the optimization is to skip the type checking.  For example,
something declared int can be stored in a machine register with no type
bits, rather than as a tagged word or as a pointer to something in the
heap.

> This allows optimization based on the advice and error checking only when
> wanted.

Yes, the proposal is the programmer decides when to request optimization
and when to request error checking.  Normally you'd leave error checking on
during development.

> In general, I like your proposals (after Greg Ewing's correction).
> It reminds of importing packages in that it is a simple tool for
> building large, fast, reliable programs while staying out of the way
> (and out of mind) for more casual programming.

Thanks, that summarizes the intention pretty well.


From tim.one at home.com  Sat Feb  2 14:35:29 2002
From: tim.one at home.com (Tim Peters)
Date: Sat, 2 Feb 2002 14:35:29 -0500
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
In-Reply-To: <yu99wuxv7n7r.fsf@europa.research.att.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEHENJAA.tim.one@home.com>

[Andrew Koenig]
> It seems to be universal among programmers that they get hung up about
> tiny syntax details until they get used to those details.  For example,
> C and C++ programmers still argue about where to put the curly braces
> in their programs.

That's because they don't have Guido to force them to do it the right way:

	if (guido) {
		tab(indent);
	}

If Guido had written the first C compiler, that's the only style it would
have accepted <0.5 wink>.

> At least Python doesn't have that particular problem...

Oddly enough, none of the Python developers argue about brace placement,
except for the occasional radical who sneaks in

	} else {
      }

instead of

	}
	else {
	}

The indentation of "case" wrt an enclosing "switch" is so hopeless nobody
even bothers to argue.

What the Python developers do get testy about is living with Guido's
seemingly inconsistent demand that Python code use 4-space indents with no
hard tabs, C code use 8-space hard-tab indents.  Oddly enough, the more
intelligent of us agree with him, but nobody can explain why <wink>.

to-call-it-a-matter-of-taste-would-be-too-forgiving-ly y'rs  - tim




From dale at riverhall.NOTHANKS.co.uk  Fri Feb 15 14:07:35 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Fri, 15 Feb 2002 19:07:35 +0000
Subject: Object copying itself
References: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com>
Message-ID: <qumq6ustholkguucuv3k95mautf32cvcn2@4ax.com>

It occurs to me that I could use copy.copy to do most of the work and
patch up the bits it missed afterwards

What's the best approach?

class data:
    def __init__(self, defaults=None):
        if defaults:
            self.__dict__ = copy.copy(defaults.__dict__)

Is that safe?

or what about:
            self = copy.copy(defaults)

Any other ideas?

Cheers.
--
Dale Strickland-Clark
Riverhall Systems Ltd


From fperez528 at yahoo.com  Tue Feb 26 10:54:21 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 26 Feb 2002 15:54:21 +0000
Subject: what I would like to see in python to make a better "glue" language (Newbie)
References: <899f842.0202261325.6ff3dd54@posting.google.com>
Message-ID: <a5h3qd$2df$2@peabody.colorado.edu>

Anthony_Barker wrote:

> One more idea for Administrators/Glue Implementers; lots of code
> examples is key.

100% agreed. The python docs are very nice, but they could definitely use 
more example code. Often a single well written example is worth 10 pages of 
ultra-verbose text. 

The doc maintainers do a great job already, this is just bugging them a bit :)

Cheers,
f


From flognat at flognat.myip.org  Sun Feb  3 14:59:35 2002
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Sun, 03 Feb 2002 19:59:35 GMT
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>
Message-ID: <m37kpuz68o.fsf@localhost.localdomain>

/ Lawrence Oluyede <rhymes at myself.com> wrote:
| Why in the std distribution of Python 2.2 there isn't the curses
| module?

New news.. Throw an eye at the cygwin compiled python, they probably
have squeezed in ncurses support.

        /Andy
-- 
 The eye of the beholder rests on the beauty!


From sholden at holdenweb.com  Thu Feb  7 08:40:17 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 08:40:17 -0500
Subject: Why it does NOT work on Linux ?
References: <mailman.1012871110.28776.python-list@python.org>
Message-ID: <Vsv88.94274$KM2.3404324@atlpnn01.usenetserver.com>

"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1012871110.28776.python-list at python.org...
> Steve Holden wrote:
> > Markus Sch?nhaber wrote:
> > > 1.) Here
> > > > IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> > > > Results=re.findall(IDs,Text)
> > >
> > > you call re.findall with a regular expression object as a first
> > > parameter which should be a string. What you want to do is
> > >
> > > Results = IDs.findall(Text)
> > >
> > > i. e. call the appropriate method on the re object you created.
> > >
> > While your suggestion is correct, it is incorrect to say that Ladislav's
> > method is wrong.
>
> Ladislav's use of findall works, but this feature of re.findall()
> is not documented and should be avoided, as undocumented features
> have been known to disappear.
>
Nonsense. The function interface continues to be a fully-documented part of
the re module. You are apparently confusing

    http://www.python.org/doc/current/lib/node99.html

(the re module contents description, which describes the functional
interface to regular expressions) with

    http://www.python.org/doc/current/lib/re-objects.html

(the method descriptions for re objects). It may not be the way *you* do it,
but that doesn't mean it's not documented <wink>. For now, either approach
is both documented and valid, as it is for strings (string.replace(s, p, r)
versus s.replace(p, r)).

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From joshm at taconic.net  Mon Feb 25 17:34:02 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 25 Feb 2002 17:34:02 -0500
Subject: 15 minute presentation on python?
Message-ID: <3c7ab974_4@corp.newsgroups.com>

Hey all,

If you had to give a 15 minute introduction to Python, what would you cover?
What is a good example of a simple, but useful, text-based app for a Wintel
environment, that has good "wow" quality?

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From jason at jorendorff.com  Fri Feb 22 16:31:20 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 22 Feb 2002 15:31:20 -0600
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <u1yfdlr4i.fsf@python.net>
Message-ID: <HFEKILOLEFEFMKAECNDLIEJFDEAA.jason@jorendorff.com>

Michael Hudson wrote:
> Parsing C++ is not much easier than compiling it, I believe.
> Templates!

Okay, but this is the kind of thing I don't understand:

  int* p1;   // OK
  int *p2;   // OK
  int * p3;  // OK
  int*p3;    // Cint parsing limitation, space character needed

## Jason Orendorff    http://www.jorendorff.com/



From andda715 at student.liu.se  Mon Feb 18 06:32:43 2002
From: andda715 at student.liu.se (Anders Dahlberg)
Date: Mon, 18 Feb 2002 12:32:43 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c6fc507$0$159$bb624dac@diablo.uninet.ee>
Message-ID: <a4qotb$iu5$1@news.island.liu.se>

"Eduard Smirnov" <Ed.Smirnov at mail.ee> skrev i meddelandet
news:3c6fc507$0$159$bb624dac at diablo.uninet.ee...
> I'm writing a diploma paper on the difference between scripting (e.g. in
> Python or Tcl) and system programming (e.g. in C, C++ or Java), and one of
> the claims I want to make is that you can often develop applications 10x
or
> more faster in a scripting language.

Yep, notice that this time is *only* development - if you also count
debug/maintain time (that is, time spent on code after it is "done") you
will get somewhat more realistic figures.

> I apologue that more applications can be written or re-written in
scripting
> language faster and with more complex functionality then in system
> programming language.
> If you have any experience in writing applications in system programming
> language (e.g. in C, C++) and re-writing it in scripting language (e.g.
> Python) or vice versa, please compare your work in both implementations,
> describe your work/application's functionality, approximate number of
lines
> and time of development of each implementation, skill's level needed.
> Any more comments about are welcome.
>
> More thanks for copy your answer to my e-mail.
>
> Eduard.Smirnov at mail.ee
> Tallinn Technical Univercity
> Estonia
>
>
>




From jmfbahciv at aol.com  Fri Feb 15 06:09:20 2002
From: jmfbahciv at aol.com (jmfbahciv at aol.com)
Date: Fri, 15 Feb 02 11:09:20 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net> <a4geh2$hir$12@bob.news.rcn.net> <140220020703593280%jwbaxter@spamcop.net>
Message-ID: <a4j24g$jao$9@bob.news.rcn.net>

In article <140220020703593280%jwbaxter at spamcop.net>,
   "John W. Baxter" <jwbaxter at spamcop.net> wrote:
>In article <a4geh2$hir$12 at bob.news.rcn.net>, <jmfbahciv at aol.com> wrote:
>
>> In article 
>> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345 at lp.airnews.net>,
>>    claird at starbase.neosoft.com (Cameron Laird) wrote:
>> >In article <130220020751360934%jwbaxter at spamcop.net>,
>> >John W. Baxter <jwbaxter at spamcop.net> wrote:
>> >                       .
>> >                       .
>> >                       .
>> >>Did the elves at AT&T pick up the "leading 0 means octal" from
>> >>something earlier, or did they invent this stupidity themselves?
>> >                       .
>> >                       .
>> >                       .
>> >OK, folks, how far back *can* we trace this?  My
>> >mind associates it with DEC systems, going back
>> >to the '60s, but I couldn't find any confirmation
>> >of that in a quick search.
>> 
>> You will never find a DEC standard that ever defined
>> a number with a leading zero to be octal.  I never
>> typed up anything that ever had that kind of standard. 
>
>Well, Digital's hardware was what the AT&T elves worked with, so the
>connection is certainly there, but not necessarily coming in from DEC.

Ah, but they had to define their own standards ;-).


>
>I can't remember from my not-extensive messing about with PDP-1 (serial
>1, provided at no charge and with support by DEC) at MIT whether that
>assembler used a leading 0 to signal octal, or used something else. 

That is before my time.  By the time I started doing stuff, there
were pseudo-ops in the assembler.

I do have two memoes that I found in my office boxes that were
addressed to PDP-1 users.  I don't know the people who wrote
them (Roland Silver, Edward Fredkin) nor where they were. 

The first defines the terms we still use today and the second
specifies all of the info that has to go along with a program.
The example given is a writeup of dvd and i-dvd by Bill 
Fletcher.

There doesn't seem to be any indication about how numbers
are to be viewed in these memoes.




>[Octal was natural on that machine...hex wasn't an issue.]  That
>assembler pretty much moved to the PDP-1 from the older TX-0
>("Transistorized eXperimental" and the round thing is zero) next door.
>
>And then moved on to DEC along with a bunch of people, who were key
>parts of the technical staff for a long time.  That was a very fertile
>"seeding" by the General.  And produced "SpaceWar."

Well, if you can give me a clue about these memoes that I have, I'd
appreciate it.

/BAH

Subtract a hundred and four for e-mail.


From robert_kuzelj at yahoo.com  Mon Feb 25 04:55:00 2002
From: robert_kuzelj at yahoo.com (Robert Kuzelj)
Date: 25 Feb 2002 01:55:00 -0800
Subject: pythonisten stammtisch in muenchen - step 2
References: <eaef2e43.0202220517.7cc555ed@posting.google.com>
Message-ID: <eaef2e43.0202250154.639c80ca@posting.google.com>

hi,

also nachdem die zahl auf 5 1/2 angestiegen ist,
der zeitpunkt und ort den den meisten egal ist,
sollten wir konkreter werden.

deshalb verfuege ich kraft der mir von gott gegebenen 
gewalt den 18.3.2002 19:00 uhr zum zeitpunkt der
konstituirenden sitzung des python stammtisches zu
muenchen. ;-)

was die wahl der geweihten hallen angeht, so muss
ich nochmals nachfragen - bisher sind haidhausen
und schwabing als grobe treffpunkte vorgeschlagen 
worden. ich wuerde das ganze gerne weiter 
einschraenken. es sollte eine eher ruhige kneipe
sein, in der man sich in normaler lautstaerke
unterhalten kann (damit fallen wohl die meisten 
in-places aus).

vorschlaege?

ciao robertj


From tim.one at home.com  Fri Feb  1 04:17:37 2002
From: tim.one at home.com (Tim Peters)
Date: Fri, 1 Feb 2002 04:17:37 -0500
Subject: Conventions for iterators
In-Reply-To: <a3cu29$grp$1@newsreaderm1.core.theplanet.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEBLNJAA.tim.one@home.com>

[jlbec at evilplan.org]
> 	I was wondering what (if any) were the conventions used when
> generating iterators.  I know the next()/StopIteration protocol.  I'm
> speaking more of consistency conventions.
> 	Say I have a list-like object containing the objects (obj1,
> obj2, obj3, obj4, obj5).  Each successive iteration returns the next
> object.  Pretty simple.  However, what if I call
> "listthing.delete(cur_obj)", where cur_obj is the current object
> returned by the iteration?  There are many things that *could* happen
> (let's say that cur_obj is obj3).
>
> 1. Stop iterating, because the change renders the iterator invalid.
> 2. The next element (obj4) is correctly iterated.
> 3. The internal structure isn't exactly a list, so the next element
>    (when queried) now may be obj2 again, or might skip obj4 to obj 5)

People who mutate a container while iterating over it are living
dangerously, and Python doesn't try to stop them from blowing off their
feet, except to guarantee that this doesn't happen:

> 4. The list-like object is broken due to a state inconsistency created
>    here.

Short of that, anything goes:  the behavior is undefined.  It may be any of
your 1-3, or even a mix.  2.2 introduced dict iterators (dict.iteritems()
and its friends), and they currently make a good-faith attempt at your #1,
but only so far as they can while remaining cheap.  IOW, you *may* get your
#2 and #3 behaviors at times when mutating a dict while iterating over it,
but will usually get #1.  Here's a contrived example; nothing about its
behavior is guaranteed across releases:

>>> d = {1:1, 2:2}
>>> for k in d:
...     print k
...     del d[k]
...     d[10*k] = 10*k
...
1
2
20
10
>>> d
{200: 200, 100: 100}
>>>

If you can't live with that, good!  Do the sensible thing and iterate over a
*copy* (for k in d.keys(); for k, v in d.copy(); etc).

Mutating sequences while iterating over them is well-defined, but not via
any of your 1-3.  You can read "the rules" in section 7.3 ("The 'for'
statement") of the Language Reference manual.  Relying on those rules is
generally stupid, but I confess I've often used

all = [root]
d = {root: 1}
for node in all:
    for child in node.children():
        if child not in d:
            d[child] = 1
            all.append(child)

to get a breadth-first list of the nodes in a rooted, directed graph.  This
relies on that growing a list while iterating over it works exactly the way
it's defined to work.

Most often, though, it's again best to iterate over list[:] (i.e., a copy),
if you do size-changing mutations during iteration.

> 	Now, obviously, 3 and 4 are bad.

#4 is right out:  no way.  #3 doesn't offend Pythonic sensibilities, though.

> ...
> With "one modification invalidates the iterator," you cannot have
> code like:
>
> for item in list:
>     if matches(item):
>         list.remove(item)

Code like that won't complain today, but won't do what you probably hope for
either.  Change it to "for item in list[:]" and there's not even a
possibility for confusion.

> So, obviously, behavior 2 is "best."

It would be in Scheme, but not in Python.  Python enjoys making tradeoffs
that drive *someone* crazy <wink>.  Since the semantics of mutating while
iterating are subtle even in an "anally correct" implementation, Python
punts and gives you a different choice:  accept whatever you get, or make
the issues utterly clear by iterating over a copy.  In practice I've never
found this to be a real limitation (and have found many clever ways to avoid
mutating while iterating <wink>), but it is true that *everyone* gets burned
by your list.remove() example once.

> But I was wondering what the python community expects/requires.

Most of all, they want clarity.  If you can't easily explain what mutating
during iteration will do, exactly, leave it undefined beyond guaranteeing
the object remains sane "no matter what".  Then if you can gripe if they try
to do something undefined, so much the better, but that's not worth gross
complications or slowdowns in your implementation.  Python doesn't stick its
head up its ass here trying to swallow mutatarrhea, so why should you
<wink>.

caveat-mutator-ly y'rs  - tim




From gh_pythonlist at gmx.de  Mon Feb 18 15:41:22 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 21:41:22 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com>
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com>
Message-ID: <20020218204121.GA9316@lilith.hqd-internal>

Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> [Aahz]
> > I think the occasional non-English post simply reflects the
> > international nature of Python.
> 
> [Fran?ois Pinard]
> > Exactement.  Bravo!  Bien dit! :-)
> 
> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> czignat?  Classifero en bragna!

Woas da Deifi wos des hoasn mog.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.4 ?C      Wind: 3.1 m/s



From emile at fenx.com  Wed Feb 13 21:26:48 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Feb 2002 18:26:48 -0800
Subject: Python License Issues
References: <20020213191640.13956.qmail@mail.yifansoft.com> <mailman.1013651063.14021.python-list@python.org>
Message-ID: <a4f7j8$1ek0dm$1@ID-11957.news.dfncis.de>

"Cliff Wells"
> Besides, if there were any future legal issues regarding Python, I
imagine
> Guido would already know about it ;)
>

I always thought of Guido's time machine more like Rocky and
Bullwinkle's (or was it Mr Sherman's) way-back machine... ;-)

remembering the way it hasn't happened yet,

--

Emile van Sebille
emile at fenx.com

---------



From sdm7g at virginia.edu  Thu Feb 21 13:09:22 2002
From: sdm7g at virginia.edu (Steven Majewski)
Date: Thu, 21 Feb 2002 13:09:22 -0500 (EST)
Subject: Hosting
In-Reply-To: <283adf56.0202210631.f083922@posting.google.com>
Message-ID: <Pine.OSX.4.40.0202211245070.423-100000@d-128-54-224.bootp.virginia.edu>


On 21 Feb 2002, Kevin Dahlhausen wrote:

> That's actually not such a bad idea.  I'm only paying $10/month for
> dialup, and looking to pay about the same for hosting.   Does DSL
> require you to continue phone service once it is installed?  That
> would make up the difference and it would be affordable then.
>
> > With the money you pay to ISP, why don't you get DSL with static IP, and
> > run your own server?

Prices will vary, but where I am you'll pay a LOT more for DSL
service with a static IP than $10/month. However, sometimes you
can get a package plan ( w/ phone service and a bunch of extras:
caller ID, etc. ) so it might make sense -- you'll probably still
pay more but the connection is worth it.

I don't have DSL right now -- I'm too far out of town for it to
reach, so I have ISDN instead. I'll probably get DSL when I move
back into town next year.

I don't know if it's a commentary on the reliability of DSL | ISDN,
but they all though it was strange that I wanted digital service WITHOUT
also an analog phone line. You will experience less uptime for
your phone if you go all digital, but we use cell phones for backup.
( It does feel a bit strange when I have to go home and "reboot"
 my phone system when there's a problem because my wife isn't
 that 'technical' ! )


The other thing is that you still might end up paying an ISP :
 For DSL service, it was a package provided by Sprint (my local
phone Co.) and Earthlink, so you HAD to get an account with
Earthlink. ( Somebody told me that since Sprint doesn't know
what they're doing with DSL, that it's really a marketing arrangement
with Earthlink, who really provide most of the "service" )
With my ISDN, I'm paying Sprint for the ISDN lines and I'm paying
our local ISP (cstone.net) for an ISDN to internet connection on the
other end.

 For small to moderate usage, I think an ISP is always going to
be cheaper than paying for a thick pipe and doing it yourself --
but if you want the thick pipe anyway, then it might be worth
setting up your own server. ( But if you do anything "commercial"
with the server, they won't want to let you use any of the
end-user oriented packages and pricing! )

-- Steve Majewski






From andy47 at halfcooked.com  Mon Feb 25 18:09:17 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Mon, 25 Feb 2002 23:09:17 +0000 (UTC)
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com>
Message-ID: <Xns91C167290C907andy47halfcookedcom@203.109.252.31>

boblancaster at zxmail.com (Bob Lancaster) wrote in 
news:b470cb58.0202250816.7839c1c4 at posting.google.com:

> I am in a bit of a quandry. 
> I am writing a python module which converts some data from various
> python variables, lists, etc. into SQL, which in turn is inserted to
> or selected from a database.
> 
> I need to support three combinations.
> (1) PostgreSQL and Linux, for academic customers.  No serious problems
> here.  In fact, I have learned why folks love PostgreSQL so much.
> 
> (2) Oracle 9i and Linux.  I haven't tried this one yet, but it seems
> to be at least as easy as the third combination.
> 
> (3) Oracle 9i and Windows 2000, for industrial customers.  Trouble!
> I have tried two approaches.  
>   (a) DCOracle2:  builds, connects, but then does not work properly. 
> I see two possible sources of the problem.
> 
> The documentation for DCOracle2 says:
> 
> """Oracle installs the C headers to build OCI programs in
> $ORACLE_HOME/rdbms/demo, $ORACLE_HOME/network/public, and
> $ORACLE_HOME/plsql/public. If these directories do not have the proper
> include files, the build will not succeed. Oracle typically only
> installs these files when a complete server installation is
> performed."""
> 
>  
> Okay, under my ORACLE_HOME directory, I have rdbms/demo, which has
> lots of header files, I have plsql/public, which has a header file,
> but my ORACLE_HOME/network dircetory does not have a /public
> subdirectory.  I not sure why DCOracle2 would even build without this
> subdirectory.  Have the folks at Oracle, in their infinite wisdom,
> chosen to place the files elswhere so that I could copy into a
> /network/public directory? Or am I screwed?
> 
> The other possible problem is that I am using Python 2.1.1.  I rewrote
> the install.py to accept 2.1.1 in addition to 2.1, since the Python
> home page no longer lists 2.1 as a possible download.  (In fact, they
> only list 2.1.2 these days.) Would using 2.1.1 instead of 2.1 make
> that big a difference?
> 
> (b) I tried using cx_Oracle.  This seemed like the best solution,
> until I started doing more complicated stuff.  First, it sometimes
> crashes when I do a fetchall().  (Error message 50). I have looked at
> newsgroups, etc. and I saw that I am not the only one to ever have
> this problem.  Second, although I could insert and update
> TIMESTAMP_WITH_TIMEZONE and user-defined types, I could not retrieve
> them.  I checked the source code, and cx_Oracle is VERY limited in the
> types it handles.  I would hate to have to rewrite that module and
> recompile the whole thing, especially since I can't solve the
> fetchall() problem.
> 
> 
> It appears that I have three options at this point, none of which are
> very good.  All of them may require some substantial reworking of my
> database, alas.
> 
> I.  Somehow get DCOracle2 to work with Oracle 9i.  This seems to be my
> best bet, if I can figure out how.
> 
> II.  Figure out how to get around the fetchall() problem with
> cx_Oracle, and put all my user-defined types into VARCHARs, or
> seperate tables, or whatever.
> 
> III.  Use Jython and connect with the JDBC and a JDBC:ODBC bridge. 
> This seems to be the least satisfying, since I want to be able to put
> everything into a single package.
> 
> 
> Any advice would be most welcome.
> 
> Thanks,
> Bob
> 
> BTW, I have an exinct email listed, to avoid spam.  If you really need
> to contact me, try rlancasteratbruker-axsdotcom, converting the at and
> the dot to symbols.

Bob,

I think the DCOracle2 documentation is incorrect. All of the header and 
library files needed to compile OCI clients for Oracle should now (Oracle8i 
release 2 and above) be included in the %ORACLE_HOME%\oci directory on 
Windows. For information on which components are necessary to build OCI 
client modules have a look at the "Oracle Call Interface Getting Started 
Release x.x.x for Windows" manual provided with your database installation. 
Mine is for 8.1.6, ymmv. 

Something that has caught me out in the past is that I have performed a 
server install but forgot to include the Oracle client software. This then 
doesn't include the appropriate dlls which will result in run time errors 
as soon as you try to call them. If you do a 'full' install of Oracle on 
Windows this shouldn't be a problem, but if you have performed a 'custom' 
install you may have missed out the appropriate headers and dll files and 
that would definitely cause issues.

As far as I know you should have no problems using DCOracle2 with Python 
2.1.1 or 2.1.2. I'm running Python 2.1.2 as part of Zope on my Windows2000 
machine and I have successfully managed to install DCOracle2 into that 
environment. All I did was change line 14 of install.py from;

version = string.split(sys.version)[0]

to

version = string.split(sys.version)[0][0:3]

and everything was hunky dory. Hope this is of some help.

Regards,
Andy
-- 
Contents free posts a speciality


From mhammond at skippinet.com.au  Mon Feb  4 02:12:41 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 07:12:41 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com> <3C5B90F3.3020503@skippinet.com.au> <CkN68.6145$F6.3643599033@newssvr14.news.prodigy.com>
Message-ID: <3C5E34BA.6030702@skippinet.com.au>

Keith Farmer wrote:
> My mistake:
> 
> lr = win32gui.DefWindowProc(self._GetAppBarWnd().GetSafeHwnd(),
>     WM_NCHITTEST,
>     0,
>     self._LowHighWordsToLparam(win32api.GetCursorPos())
>    )
> 
> I'm catching the message with:
>     self._appbarwnd.HookMessage(self._OnWMNCHitTest, WM_NCHITTEST)
> 
> And yet though I return HTBORDER, the edges are still resizeable.  This is
> the C-code I'm trying to replicate:
> 
>  case WM_NCHITTEST:
>     {
>         LRESULT lr = DefWindowProc (hwnd, message, wParam, lParam) ;
> 
>         if (lr == HTBOTTOM || lr == HTBOTTOMLEFT || lr == HTBOTTOMRIGHT)
>             return HTBORDER;
>         else
>             return lr;
>     }

The major difference I see if the call to DefWindowProc.  You seem to be 
passing 0 for wParam, and GetCursorPos() for lParam.

Try passing wParam and lParam exactly as passed to you, as per the C++ code.

Mark.



From fredrik at pythonware.com  Mon Feb  4 11:03:10 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 04 Feb 2002 16:03:10 GMT
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com>
Message-ID: <2hy78.23175$l93.4207585@newsb.telia.net>

Alan Kennedy wrote:

> I don't understand what's wrong with
>
> <script language="python" xml:space="preserve">
>     def messiah():
>         print "He's NOT the Messiah."
>         print "He's a very naughty boy."
> </script>

something tells me that you're a pragmatic programmer
working in real-life projects, right? ;-)

</F>





From SteveC at innocent.com  Thu Feb 14 05:47:40 2002
From: SteveC at innocent.com (Steven M. Castellotti)
Date: Thu, 14 Feb 2002 23:47:40 +1300
Subject: Recommendations for cross-platform GUI Bindings ?
Message-ID: <xBMa8.217$B92.52511@news.xtra.co.nz>

Greeting all--

	I'm writing an Open Source software application that allows children to
create their own video games, and would like to hear anyone's
recommendations for GUI bindings. The key things I'm looking for are:

 - Cross Platfrom (I need to run on at least Linux and Windows, though
Mac would be great if possible.)


 - Compiles to something easily executable (Ideally can be run from the
web, or can at least be wrapped by an installer)


 - Has good Image/Graphics routines


 - Is well supported (Online documentation/examples are available, or at
least an active mailing list. I don't want to get stuck coding for a
target the dissapears after the developer gets bored)

 - Uses Python or python-esqe syntax.



	So far I'm looking at:

Java 1.2 (Using Jython)
Java 1.4 (Using Jython)
pyGTK


	Java 1.2 is nice because most current web browsers support it, and Swing
has some nice features, but saving and loading games to the hard drive is
tough without a paid-for signed applet.

	Java 1.4 is the most current version of Java, and
I'm told includes some nice ImageIO routines, but isn't in web browsers
yet, and I don't want to have to ask kids to upgrade their software.
There's still the problem of saving and loading to the hard drive.

	pyGTK is what I'm currently using, but GnomeCanvas, the most usful way
to manipulate on-screen images is not available under win32. In theory
a pyGTK program can be compiled into a .exe for win32 distribution, but I
haven't worked out the details yet.



	If anyone has any recommendations as to technologies I should look at, I
would be very interested to hear your thoughts. If Mono was further along
on the Linux side, I might consider targetting .Net for win32 development
ride now, but I like to code under linux and am willing to wait to move
onto Mono when its ready.


Cheers,

-Steve Castellotti
 SteveC at Innocent.com
 http://cogengine.sourceforge.net/


From tim-dated-1015298553.dcf11c at catseye.net  Mon Feb 25 22:22:32 2002
From: tim-dated-1015298553.dcf11c at catseye.net (Tim Legant)
Date: 25 Feb 2002 21:22:32 -0600
Subject: What's wrong with this subroutine?
In-Reply-To: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>
References: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>
Message-ID: <864rk5x6yv.fsf@skitty.catseye.net>

"A.Newby" <deathtospam43423 at altavista.com> writes:

> First, here's what I'm trying to do ......
> 
> I want to get this subroutine so that the user just has to post a url into 
> the chat box to get a post a picture or a link in chat.

Not sure why you're getting a ValueError, but how about this, instead.
The code in the "__name__ == '__main__'" section executes if you run
this as a script from the command line:

$ python URLReplace.py


--- URLReplace.py ---

import os
import re

RE_HTTP = re.compile(r'http://\S+')

def replaceURL(user_text, startidx=0):
    imgtag = '<img src="%s">'
    linktag = '<a href="%s" target="blank">Link</a>'

    matchobj = RE_HTTP.search(user_text, startidx)
    if matchobj:
        url = matchobj.group()
        urltype = os.path.splitext(url)[1]
        if urltype in ('.jpg', '.gif', '.png'):
            tag = imgtag % url
        else:
            tag = linktag % url
        user_text = "%s%s%s" % (user_text[0:matchobj.start()],
                                tag, user_text[matchobj.end():])
        return user_text, matchobj.start() + len(tag)
    return user_text, 0


if __name__ == '__main__':
    text = "http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg\n" \
                  "is a really cool picture.  In fact, the whole site,\n" \
                  "http://www.pbs.org/kratts/world/oceans/ , is pretty awesome.\n"
    print text

    nextidx = 0
    while 1:
        (text, nextidx) = replaceURL(text, nextidx)
        if not nextidx:
            break

    print text

--- end URLReplace.py ---


Tim



From joshm at taconic.net  Mon Feb  4 22:24:38 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 4 Feb 2002 22:24:38 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb  4)
References: <C587C06E0D2D15BE.08C92F8B9CB3521B.3849AAF8C8E92DF5@lp.airnews.net>
Message-ID: <3c5f4e13_4@corp.newsgroups.com>

> "One less file is one less thing to get screwed up in the distribution,
> IMHO."  Joshua Muskovitz

Yay!  I've finally arrived!  Now where's that damn job offer?  :-)

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From alanmk at hotmail.com  Thu Feb 28 10:47:49 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 28 Feb 2002 07:47:49 -0800
Subject: Pointers for solving windows GUI problem.
Message-ID: <f72dac0d.0202280747.2360c608@posting.google.com>

Greetings all,

I'm posting this question in the hope that someone has already done
this, or something similar, and can perhaps direct me some code or a
tutorial that will shorten the path to a solution.

I have a Windows COM object which renders a series of images into a
window. I can control the COM object using provided API methods. I
don't have the source for the COM object.

I want to do screen grabs of the display of this window at various
stages, so that I can dump them to files, in GIF, JPEG or PNG format.

What I'm looking for is perhaps some sample code that does this or
something similar. Failing that, a pointer to a good MFC tutorial
would be most helpful.

I am totally unfamiliar with MFC and Win32 GUI programming, although
I've developed some X-Windows stuff in the distant past.

However, given the excellent tools available, namely Python and Mark
Hammond's excellent Windows extensions, I have every faith that the
job can be done without having to go near VC++ (which I can do, but
really don't want to :-)

TIA for any pointers.

Cheers,

Alan.


From pawelczyk at lucent.com  Tue Feb  5 04:46:09 2002
From: pawelczyk at lucent.com (Lukasz Pawelczyk)
Date: Tue, 05 Feb 2002 10:46:09 +0100
Subject: Binary Python file.
Message-ID: <3C5FA9E1.DF4AC37F@lucent.com>

Hello,

Is there a possibility to make a binary python file?
I would like to hide the code in unix environment.

Regards,
Lukasz


From gerson.kurz at t-online.de  Wed Feb  6 12:02:33 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Wed, 06 Feb 2002 17:02:33 GMT
Subject: for item in (pydoc, docs, lib): question_regarding(item)
Message-ID: <3c615b18.47723171@news.t-online.de>

1) Shouldn't

c:\python22\lib\pydoc.py -w c:\python22\lib

work (provided you have python installed in c:\python22)? Well, it
starts working just fine, but then all those ugly "test" macros hit
in, and finally the processing is aborted with

Traceback (most recent call last):
  File "C:\python22\lib\pydoc.py", line 2101, in ?
    if __name__ == '__main__': cli()
  File "C:\python22\lib\pydoc.py", line 2068, in cli
    writedocs(arg)
  File "C:\python22\lib\pydoc.py", line 1361, in writedocs
    writedocs(path, pkgpath + file + '.', done)
  File "C:\python22\lib\pydoc.py", line 1368, in writedocs
    writedoc(modname)
  File "C:\python22\lib\pydoc.py", line 1347, in writedoc
    html.document(object, object.__name__))
  File "C:\python22\lib\pydoc.py", line 267, in document
    if inspect.ismodule(object): return apply(self.docmodule, args)
  File "C:\python22\lib\pydoc.py", line 528, in docmodule
    cdict[key] = cdict[value] = '#' + key
TypeError: 'str' object is not callable

and the rest of the files is not processed. 

2) There are some minor things I would like to see changed in the
official python documentation. e.g., the example in section "6.4 stat
-- Interpreting stat() results" uses the line

pathname = '%s/%s' % (dir, f)

which IMHO should read

pathname = os.path.join(dir, f)

I think that IF you at last give in and RTFM, it should at least
silently guide you to the python-way of doing things. Now, where to
submit things like these?

And saving the best for last:

3) Is there something like a TODO-list for "The Python Standard
Library"? And how to contribute? 

I'm thinking of tiny things like a hexdump function or administrative
functions (like "granting/denying a user access to a file on NT") and
things like that. [If you speak out regarding "OS dependancy",
remember "sunau" and those IRIX modules]. I'd also like to add bigger
things like a stripped down generic SMTP/POP3 server. 

I know we had threads on this before and they never got anywhere, but
hey this time it could all be totally different :)




From mats at laplaza.org  Thu Feb 14 12:14:31 2002
From: mats at laplaza.org (Mats Wichmann)
Date: Thu, 14 Feb 2002 17:14:31 GMT
Subject: photo id ?!?!
References: <mailman.1013585541.16266.python-list@python.org>
Message-ID: <3c6bef71.83266190@news.laplaza.org>

On Wed, 13 Feb 2002 02:31:45 -0500, "Tim Peters" <tim.one at home.com>
wrote:

:It's curious that the FAA doesn't require photo ids (see
:http://www/faa.gov/).  I've read that the airlines made up that rule, not
:for security, but to make it harder for person A to use a cheap ticket
:obtained by person B.  Is that true?  Beats me, but my cynical half can't
:not believe it <wink>.


There are more than a few FAA regulations/guidelines that the airlines
choose to enforce a particular way because it benefits their
operations.  Prior to Sept 11, your take was essentially true.


And as to the whole ID-at-conference question, I believe it's mostly
CYA on the part of organizers.  It's not "payment verification" alone,
I was in New York for the Linux World Expo the week before and while
security was much less tight than we had been warned, even
exhibitor/speaker passes needed an ID to obtain (I had someone
standing there who knew me, didn't help).  On the other hand, security
all over New York has gone a bit haywire.  Some of you know that I do
some teaching for Learning Tree for time to time, I went to drop by
their NY center and the combination of photo ID plus company badge
(non-photo variety) was not sufficient, someone had to come down to
escort me since I wasn't on the list of instructors for that week.  Of
course, that's not Learning Tree's own building.  Goes by the name of
the "CBS Building" (although there are many other tenants) so the
anthrax scare probably has something to do with it.


I suspect we're not going to have much luck with this for a while, at
least not for conferences held in places like NY and DC.  You're
welcome to head out my way (Taos, NM) for future conferences; I might
even be able to attend!


Mats

Mats Wichmann



From vlindberg at verio.net  Fri Feb 22 15:50:10 2002
From: vlindberg at verio.net (VanL)
Date: Fri, 22 Feb 2002 13:50:10 -0700
Subject: Request for comments: Search and replace script
Message-ID: <3C76AF02.2070101@verio.net>

Hello,

I recently finished writing a search and replace script to automate some 
of the things I do at work. It does the following:

Will search and replace in a single file, all files in a directory, or 
recursively
Will search and replace from a regular expression or a string
Excludes all binary files unless --binary argument is given
Has include and exclude lists (search and replace in all *.html, but
exclude donttouch.html)

and a bunch of other stuff too. Most of the work is encapsulated in
some fairly reusable classes.

You can find it at http://lindbergs.org/vsar.py

I would be very appreciative of any comments or patches to it.
Particularly, I am interested in:

- Things that could make it faster
- Things to make it more 'pythonic'
- Bugs or logic errors

After any suggested improvements have been made, I will be submitting it
to useless python, in the hope that it may be more generally useful.

Tear into it,

Van









From claird at starbase.neosoft.com  Thu Feb 14 09:08:18 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Feb 2002 08:08:18 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <c76ff6fc.0202131330.66bd9d93@posting.google.com>
Message-ID: <92C4C5D30DAFCA11.C9B327E56C7CCF63.8F1104D8FC7D157C@lp.airnews.net>

In article <c76ff6fc.0202131330.66bd9d93 at posting.google.com>,
John Machin <sjmachin at lexicon.net> wrote:
			.
			.
			.
>I've been semi-bitten just in the last week; grabbed a table of data
>that included month numbers with leading zeroes and pasted it into a
>script -- the 08 and 09 caused a syntax error which puzzled me for a
>minute or two (I haven't needed to think about octal much since I
>worked on an ICL 1905 (24-bit words)). Fortunately no harm done.
			.
			.
			.
This is a classic, by the way:  build an application,
test it, put it into production, it makes people happy
for months, go on vacation, and it mysteriously halts
the system on the first Monday you're gone--during the
first week of August.

August is both a traditional vacation time, and the
eighth month.
-- 

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


From akuchlin at ute.mems-exchange.org  Fri Feb  8 14:23:59 2002
From: akuchlin at ute.mems-exchange.org (A.M. Kuchling)
Date: 08 Feb 2002 19:23:59 GMT
Subject: What's the equivalent in Python for Perl's CPAN?
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com> <a3v903$kgk$1@panix2.panix.com> <msU88.19568$zJ3.1529583@typhoon.maine.rr.com> <mailman.1013193915.2932.python-list@python.org>
Message-ID: <slrna689ej.cn1.akuchlin@ute.mems-exchange.org>

In article <mailman.1013193915.2932.python-list at python.org>, 
	Skip Montanaro wrote:
> People are working on stuff.  Search the list archives for Ciphon.  There
> was a lightning talk presented yesterday about a server called Gideon that's
> supposed to serve up package metadata to tools like Ciphon I think.

Gideon's Web page is at
http://www.zope.org/Members/k_vertigo/Stories/Gideon/FrontPage ; I
don't think there's a running installation on-line.  Kapil demoed it
to David Ascher while I eavesdropped, and it looked quite nice.

--amk                                                  (www.amk.ca)
I suppose you could say "the yoke's on him", if you were the sort of
person who said that sort of thing, which fortunately I'm not.
    -- The Doctor, in "The Creature From the Pit"



From gerson.kurz at t-online.de  Thu Feb 21 13:07:03 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 18:07:03 GMT
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <3c753661.18311796@news.t-online.de>

On Wed, 20 Feb 2002 21:58:50 +0000, Gon?alo Rodrigues
<op73418 at mail.telepac.pt> wrote:

> a nice set of questions others answered already. Still,

Here are two good links:

"Continuations Made Simple and Illustrated" (and its even for Python!)
http://www.ps.uni-sb.de/~duchier/python/continuations.html

"Coroutines in C"
http://www.chiark.greenend.org.uk/~sgtatham/coroutines.html



From hmartires at cultalg.pt  Thu Feb 21 11:42:26 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Thu, 21 Feb 2002 16:42:26 -0000
Subject: MySQLdb
References: <XD7d8.203$zZ4.2649220@newsserver.ip.pt> <mailman.1014303155.8362.python-list@python.org>
Message-ID: <6q9d8.209$V05.2692159@newsserver.ip.pt>

That ones i have already read, but they don't explain a lot.
Don't you know more documentation ?

"Andy Gimblett" <gimbo at ftech.net> wrote in message
news:mailman.1014303155.8362.python-list at python.org...
> > Where i can find an "how to" to learn working with MySQLdb module.
> > I have downloaded from sourceforge and install it, but there is no
> > much information on how to use it.  Any one knows where i can get
> > it?
>
> (Think I just sent this to the wrong address by mistake, but apologies
> if it comes through twice...)
>
> http://python.org/topics/database/DatabaseAPI-2.0.html
>
> http://sourceforge.net/docman/?group_id=22307
>
> --
> Andy Gimblett - Programmer - Frontier Internet Services Limited
> Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
> Statements made are at all times subject to Frontier's Terms and
> Conditions of Business, which are available upon request.
>
>




From ykingma at accessforall.nl  Thu Feb 28 15:49:19 2002
From: ykingma at accessforall.nl (Ype Kingma)
Date: Thu, 28 Feb 2002 21:49:19 +0100
Subject: Simple threading thing
References: <Rlsf8.136$5vh.170729984@news.telia.no>
Message-ID: <3C7E97BC.8C9E3CD@accessforall.nl>


Thomas Weholt wrote:
> 
> I got a webserver based on BaseHTTPServer using threads. When it starts I
> want to create a seperate process, a thread, that goes into the background
> and checks the server on given intervals, clearing expired sessions etc. I
> cannot get the created thread to pass control back to the server and let it
> continue. It just sleeps the specified interval and then does it thing then
> sleeps again. The server never gets into action at all.
> 
> Any clues?

You are probably creating a process to work for you on given intervals,
and your server seems to be waiting for that process to finish.

Have a look at the thread and threading modules.
Make a function for your background worker and start this function on 
a separate thread.
The moment this thread sleeps (or earlier) the thread that created
it will continue.

Regards,
Ype


From jeff at ccvcorp.com  Tue Feb 26 19:56:04 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 26 Feb 2002 16:56:04 -0800
Subject: dlg.Destroy() behaviour
References: <a5e561$q8v$1@news5.svr.pol.co.uk> <mailman.1014680321.7184.python-list@python.org> <a5eqpf$8fn$1@news5.svr.pol.co.uk>
Message-ID: <3C7C2EA3.9B8EA9F9@ccvcorp.com>


Duncan Smith wrote:

> "Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
> >
> > I don't know if this works (never tried it on a dialog), but
> dlg.SetFocus()
> > ?
>
> Thanks, but no.  A computer scientist mate of mine initially suggested that.
> It didn't work.  He then suggested I just got rid of dlg.Destroy(), which
> works, but he couldn't tell me whether it was a safe thing to do or not.  He
> suggested that as the 'deleted' dialogue wouldn't be destroyed immediately
> with dlg.Destroy() it might retain the focus.  I don't know.

As far as I understand it, you should be safe omitting the dlg.Destroy().  The
wxWindows dialog object *should* be cleaned up when the Python reference to it
goes out of scope.  If you're particularly worried about reclaiming memory as
soon as possible, then once the dialog is hidden (i.e., your dlg.ShowModal()
call returns), you should be able to 'del dlg', but I don't think that's
particularly necessary (especially if you re-use dlg for the next dialog).

If you are worried about memory leaks, use the hybrid build of wxPython.  At
program  end, it will dump a list of any leaked wxWindows objects.

Jeff Shannon
Technician/Programmer
Credit International




From sjoshi at ingr.com  Wed Feb 20 16:45:54 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 20 Feb 2002 13:45:54 -0800
Subject: Is there a bug in CGI-POST
Message-ID: <8f8ffe67.0202201345.66fa13ed@posting.google.com>

Hello All
I'm trying use the cgi module and I have seen that if I use the POST
method in the form then I donot get any form fields. This however
works with the GET method. Is this known, or am'i doing something
wrong here. I'm using Xitami Server on Win2000

thanks
Sunit


From neal at metaslash.com  Mon Feb  4 18:30:44 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Mon, 04 Feb 2002 18:30:44 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org> <7xu1sx41cl.fsf@ruckus.brouhaha.com> <3C5F04BF.3090806@yumpee.org> <7x4rkwdfx2.fsf@ruckus.brouhaha.com>
Message-ID: <3C5F19A4.4767F524@metaslash.com>

Paul Rubin wrote:
> The issue is there's no way to
> use integer division that works in both 2.1 and 3.0, without using
> something like divmod that would shock the unwary.

#from __future__ import division # uncomment this line if in 2.2+

def divide(n, d): 
    if type(n) == type(d) == type(0):
        return int(n / d)
    return n / d 

print divide(5.3, 2.3)
print divide(5, 2)
print divide(4, 2)

[neal at epoch neal]$ python1.5 /tmp/div.py 
2.30434782609
2
2
[neal at epoch neal]$ python2.1 /tmp/div.py 
2.30434782609
2
2
[neal at epoch neal]$ python2.2 /tmp/div.py 
2.30434782609
2
2
[neal at epoch neal]$ python2.2 /tmp/div.py # with from __future__ line uncommented
2.30434782609
2
2

So unless I'm missing something, this is what you want. 
And the .pyc is only 406 bytes. :-)

Neal


From jeff at ccvcorp.com  Wed Feb 27 18:49:44 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 15:49:44 -0800
Subject: PythonWin repeatedly crashing on trying to create COM object
References: <e8f1f186.0202261024.6370a155@posting.google.com>
Message-ID: <3C7D7098.BD827BD2@ccvcorp.com>


pj wrote:

> Greetings
>
> I use python to do some COM scripting on my windows 2000 (SR2) box. It
> was working well.. but it once crashed once.
>
> Then it just keeps crashing on
>
>     session = win32com.client.dynamic.Dispatch("Lotus.NotesSession.1")
>
> I just cant seem to get past this point.

What exactly does "crashing" mean?  Are you getting any error messages?
Does Lotus work properly when *not* accessed through COM?  Can you
instantiate any *other* COM objects?

Btw, you can probably eliminate the 'dynamic' portion, and simply use
win32com.client.Dispatch().  That will attempt to use a type library to
speed up access, if one is available, and will default to the purely
dynamic dispatch if one isn't available.  The only time you need to
specify 'dynamic', is if you really must *force* purely dynamic dispatch
even in the presence of a type library.  (And if you don't have a clue
what the difference is between purely dynamic dispatch and dispatch
through a type library is, then odds are that you're better off allowing
the type library to be used. :) )

Jeff Shannon
Technician/Programmer
Credit International




From nw_moriarty at yahoo.com  Wed Feb 13 18:43:58 2002
From: nw_moriarty at yahoo.com (Nigel Moriarty)
Date: 13 Feb 2002 15:43:58 -0800
Subject: spawn and standard output
Message-ID: <ce0b6758.0202131543.1707e80c@posting.google.com>

Hi all,

When I spawn a process from a python script on MSW, the standard
output doesn't go to the terminal where I started the script.  I would
like to view this output for debugging.  Where does the output go and
how can I view it?  Do I need to use popen or can I redirect to a
file?  Any other suggestions?

Nigel


From christophertavares at earthlink.net  Sat Feb 16 14:27:37 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Sat, 16 Feb 2002 19:27:37 GMT
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com> <b16e4ef7.0202130950.67f4004d@posting.google.com> <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4> <52e5ab5f.0202141045.5867bfbb@posting.google.com> <Xns91B71A6CA36C3RASXnewsDFE1@130.133.1.4>
Message-ID: <Joyb8.3109$tu6.336822@newsread2.prod.itd.earthlink.net>

"Philip Swartzleonard" <starx at pacbell.net> wrote in message
news:Xns91B71A6CA36C3RASXnewsDFE1 at 130.133.1.4...
> Michael Robin || Thu 14 Feb 2002 10:45:39a:
>
[... snip ...]
>
> Also... what's this 'filename completion' you speak of? Are you talking
> about windows? (Though i s'pouse it may have been added past 98, i've
> never known nor can find a key that does this...)
>

Windows finally got real, working filename completion in Win 2000. Works
great in every file selection dialog. And there *is* no completion char -
you just type and it gives you a dropdown with the names.

-Chris





From robin at jessikat.fsnet.co.uk  Wed Feb 13 12:06:01 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 13 Feb 2002 17:06:01 +0000
Subject: cgi binary stdout
Message-ID: <xNp9EKB5zpa8EwvQ@jessikat.demon.co.uk>

Does anyone know how stdout should be configured in cgi scripts. It
suddenly occurred to me that IIS/Apache running on win32 might not pass
stdout configured as binary. Does the CGI script have to ensure that it
is writing to a binary stream or can it assume the server does that?
-- 
Robin Becker


From alife at narod.ru  Wed Feb 13 06:40:30 2002
From: alife at narod.ru (Serge Terekhoff)
Date: 13 Feb 2002 03:40:30 -0800
Subject: PocketPC
Message-ID: <98322540.0202130340.616e2be7@posting.google.com>

Dear all!

Are there any Pythons or plans for PocketPC?

Sincerely, Serge A.Terekhoff
http://alife.narod.ru/eng/


From bkelley at wi.mit.edu  Fri Feb 15 15:47:26 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 15 Feb 2002 15:47:26 -0500
Subject: Warning about "for line in file:"
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org>
Message-ID: <3C6D73DE.3090201@wi.mit.edu>

Neil Schemenauer wrote:

> Russell E. Owen wrote:
> 
>>The readline or xreadline file methods work fine, of course.
>>
> 
> Why "of course"?  iter(file) does the same thing as file.xreadlines().
> Have you tested xreadlines?
> 
>   Neil
> 
> 


I had the same problem with xreadlines but "for line in file" is MUCH 
less explicit and leads to erros like this.

file = open(...)

count = 0
for line in file:
     if count > 10: break
     print line
     count = count + 1

for line in file:
     print line

Doesn't work like I would expect.  This is essentially doing the following:

file = open(...)

count = 0
for line in file.xreadlines():
     if count > 10: break
     print line
     count = count + 1

for line in file.xreadlines():
     print line

So what is REALLY happening is that you are creating two seperate 
iterators in the above examples.  Writing "for line in file" instead of 
"for line in file.xreadlines()" simply hides and confuses this.

The problem with spawning multiple iterators is that their is a read 
cache going on behind the scenes and file.xreadlines() doesn't rewind 
the file to the starting point.

The behavior that I would expect is exemplified in the following correct 
example that actually prints out all lines of a file.

file = open(...)

count = 0
iterator = iter(file)
for line in iterator:
     if count > 10: break
     print line
     count = count + 1

for line in iterator:
     print line

So from my point of view, unless you are reading the whole file, don't 
use "for line in file"

Brian Kelley
Whitehead Institute for Biomedical Research




From jeff at ccvcorp.com  Wed Feb 13 14:58:36 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 13 Feb 2002 11:58:36 -0800
Subject: Random Module help
References: <20020210215559.13516.00000455@mb-cq.aol.com> <whK98.289$ab7.1102@news.hananet.net> <QCqa8.373000$pa1.87470284@news3.rdc1.on.home.com>
Message-ID: <3C6AC56C.E6D92EE3@ccvcorp.com>


Bo Vandenberg wrote:

> Wouldnt you have to dump your initial deck into a string or tuple and then
> remove cards as they were played?
>
> That would set you up to deal elegantly with multiple hands from the same
> deck.

Those wouldn't work, because strings and tuples are immutable--you can't
remove anything from them.  You could, however, put them in a list.

deck = random.shuffle(newdeck())
for card in range(len(hand)):
    hand[card] = deck.pop()

Or some such.

Jeff Shannon
Technician/Programmer
Credit International




From cliechti at gmx.net  Tue Feb  5 15:32:49 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 5 Feb 2002 21:32:49 +0100
Subject: Jython Embedding - how do I compile and exec with params
References: <cefe12a8.0202050531.33412d38@posting.google.com>
Message-ID: <Xns91ACDBB33C1E8cliechtigmxnet@62.2.16.82>

tonygreen at techie.com (Tony Green) wrote in 
news:cefe12a8.0202050531.33412d38 at posting.google.com:
> I am writing a scripting extension mechanism to a big server project
> we've built over the last year and a half.
> 
> What I want to acheive is
> - system authors upload scripts (via a web front end)
> - the system checks and compiles the script
> - stashes the compiled script away against a named event.
> 
> When the even occurs the compiled script is called by the system 
> with a list of parameters (current user id, various numbers 
> describing the system state).
[...]
> (3) put the variables into the local namespace and then execute the
> script (without a named method or class) - but this means that I 
> would have to clean down the local namespace each time (- and anyway 
> what is it local to?).

you can pass a dictionary to exec: "exec code in dict" so you don't 
need to clean up. just build one root dict, clone it and update it 
with additional params. no clreanup needed.
 root = {...}
 copy = {}
 copy.update(root)
 copy[additional] = param
 exec code in copy

you can also make the user code import some module whitch provides the 
data throug variables or function calls.

chris

-- 
Chris <cliechti at gmx.net>



From max at alcyone.com  Sun Feb 24 22:49:15 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Feb 2002 19:49:15 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com> <3C751FEA.24CEFBBC@ford.com> <slrna7i5v6.7kb.scarblac@pino.selwerd.nl> <3C7953CE.8D430F71@alcyone.com> <bpie8.42812$TG1.1731146956@newssvr14.news.prodigy.com>
Message-ID: <3C79B43B.97152A75@alcyone.com>

Zero wrote:

> In the strictest of definitions, a "function" that does not provide a
> return value is a "method"

No, a method is a function that's a part of a class (and has the
instance of the class passed in as the first argument).  Perhaps you
meant "procedure," which is a common name for a function which has no
return value (i.e., a function which is called only for its side
effects), but still this distinction doesn't apply too well to Python
since _every_ function, whether it explicitly returns or not, returns a
value -- if it implicitly returns then the value returned is None.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From whoRunBartertown at masterBlaster.net  Sun Feb  3 07:15:02 2002
From: whoRunBartertown at masterBlaster.net (wrbt)
Date: Sun, 03 Feb 2002 12:15:02 GMT
Subject: nice to meet you
Message-ID: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net>

Hi and hello all, I'm the latest Python convert... 2 months and counting.

I LOVE IT.

You'll see me around.





From jwbaxter at spamcop.net  Tue Feb 19 23:06:36 2002
From: jwbaxter at spamcop.net (John Baxter)
Date: Tue, 19 Feb 2002 20:06:36 -0800
Subject: Finding nonprintable characters?
References: <mailman.1014145891.20575.python-list@python.org>
Message-ID: <jwbaxter-E19035.20063319022002@corp.supernews.com>

In article <mailman.1014145891.20575.python-list at python.org>,
 VanL <vlindberg at verio.net> wrote:

> isBinary(filehandle)

On Unix, you execute
   file the.unknown.file
and parse the result.

But that's not what you meant.

% file /etc/passwd
/etc/passwd: English text

% file /etc/ssh_host_rsa_key.pub
/etc/ssh_host_rsa_key.pub: ASCII text

% file /bin/sh
/bin/sh: Mach-O executable ppc

And so on, and on.

  --John


From gerson.kurz at t-online.de  Mon Feb  4 09:40:00 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 14:40:00 GMT
Subject: Autocoding project proposal.
References: <83bsfkk31e.fsf@panacea.canonical.org> <743.789T1807T13135321threeseas@earthlink.net> <slrna53q9j.d8v.huaiyu@gauss.almadan.ibm.com> <5372.791T2458T1516056threeseas@earthlink.net> <B8785A8F.39B2%jonathan@onegoodidea.com> <5203.791T524T9503446threeseas@earthlink.net> <B878EBE4.3A2B%jonathan@onegoodidea.com> <slrna58en4.4cd.philh@comuno.freeserve.co.uk> <B87A406F.3B07%jonathan@onegoodidea.com> <83sn8rgmc7.fsf@panacea.canonical.org>
Message-ID: <3c5e9cb9.17829734@news.t-online.de>

On 27 Jan 2002 22:56:24 -0500, Kragen Sitaker <kragen at pobox.com>
wrote:

>COBOL has the same advantage, really.  Not that I'd want to use or
>learn COBOL, but if I had to write a program to do some kind of simple
>calculation and it needed to be readable by nontechnical people, COBOL
>would be the tool for the job!

Proof that you can write unintelligible code in a language that is as
COBOLesque as can get:

http://p-nand-q.com/sorted.htm

Here is a fibonacci program in that language (note that everything is
more or less english):

This code uses the numbers twenty, seventyone, three, two, and one.
This code sometimes goes to the second label if the first condition is
true, sometimes goes to the fourth label if the second condition is
true, always goes to the third label, always goes to the fifth label,
and always goes to the first label.
This code writes the eleventh number as a english cardinal.
This code cannot read.
This code uses the sums of the tenth number and the eleventh number, 
and of the second number and the fifth number.
This code uses the condition 
that the second number is equal to the first number, 
and the condition that the second number is less than the third
number.
This code uses five labels.
This code does not use any ordered differences.
This code assigns the fourth number to the tenth number,
the fifth number to the eleventh number,
the fifth number to the second number,
the second sum to the second number,
the eleventh number to the fourth number,
the eleventh number to the tenth number,
and the first sum to the eleventh number.
This code does not use any products.
This code implements
the third assignment,
the first label,
the first jump,
the second jump,
the third jump,
the fourth label,
the second assignment,
the sixth assignment,
the fourth jump,
the third label,
the fifth assignment,
the seventh assignment,
the first assignment,
the fifth label,
the first output,
the fourth assignment,
the fifth jump, and
the second label.
This code does not use any ratios.
This code does not use any logical operations.
Cool.




From jobs at seaburyic.com  Tue Feb 19 12:24:35 2002
From: jobs at seaburyic.com (Seabury Insurance Capital)
Date: 19 Feb 2002 09:24:35 -0800
Subject: Contractor job, Westchester country, NY: local candidates only
Message-ID: <68b53c2c.0202190924.2c6395a6@posting.google.com>

Hello,

We are looking for a temporary contractor to help with a project here.
Must be local to Westchester County, NY, USA. No telecommuting. No
recruiters, please.

Job type: contractor
Length: 3 months

Required: Experience in Visual C++ and Python, object-oriented
programming on Windows platform. Strong analytical skills. SQL Server
or other database server. Excel VBA programming. Experience in
webserver or application server.

Pluses: Experience in statistics, or Monte Carlo simulation. Internet
programming experience (TCP/IP, HTTP protocol, RPC, socket
programming.) Networking. Linux box setup. Linux programming
experience. COM interfacing with Access/Excel. Source code revision
control systems. Bug report database. InstallShield or InnoSetup.
Technical writing.

Please send resume, hourly rate, and inquiries to jobs at seaburyic.com.

Thanks.


From travis at puddy.lan.kerrgulch.net  Thu Feb 14 18:48:12 2002
From: travis at puddy.lan.kerrgulch.net (Travis Shirk)
Date: Thu, 14 Feb 2002 23:48:12 -0000
Subject: AttributeError: 'getopt' module has no attribute 'GetoptError'
References: <Xns90D3ECC92390Dgustaflalgonetse@194.213.69.152> <ku7kxp32z3.fsf@lasipalatsi.fi> <Xns90D4988CE4ADEgustaflalgonetse@194.213.69.152> <u6o53mb595981a@corp.supernews.com> <a4h64c$99rm$1@ID-7529.news.dfncis.de>
Message-ID: <u6oj5s5uq3osb7@corp.supernews.com>

Karl M. Syring <syring at email.com> wrote:
> "Travis Shirk" <travis at puddy.lan.kerrgulch.net> schrieb
>> Gustaf Liljegren <gustafl at algonet.se> wrote:
>> > Erno Kuusela <erno-news at erno.iki.fi> wrote:
>> 
>> >>> I use Python 2.1 on Win98.  Try to use the getopt module, but get
>> >>> the error in the subject line. What's wrong?
>> 
>> >>your test program isn't called "getopt.py" by any chance?
>> 
>> > Oh. Yes, it is... Thank you.
>> 
>> I'm having the exact same problem as Gustaf, and
>> my program is *not* called getopt.py.  Anyone know what the
>> the AttibuteError is all about?

> There is not accidentally a file getopt.pyc laying around?

Well actually, there is.  I'm running Red Hat 7.2 and I've
got python and python2 rpms installed.  My script starts
with '#!/usr/bin/python', but when I change it to python2,
I don't get the AttributeError when the exception is caught.

<sigh>

I thought that python and python2 knew not to step on each
others toes, evidently I was wrong.  I'd rather not hardcode
'python2' into my script, but I guess I have no choice.

I've not followed the python 2 development closely, but is
there any reason to keep 1.5.x installed?

Thanks for your help.

> Once bitten, twice shy
> Karl M. Syring



From sholden at holdenweb.com  Thu Feb  7 16:31:36 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 16:31:36 -0500
Subject: Amount of files on a drive?
References: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk> <a3ov9n$dn9$1@newsg3.svr.pol.co.uk> <Xns91ACD5444241CRASXnewsDFE1@130.133.1.4> <slrna65a5i.1m1.rmunn@rmunn.dyndns.org>
Message-ID: <LmC88.96656$KM2.3601767@atlpnn01.usenetserver.com>

"Robin Munn" <rmunn at pobox.com> wrote ...
> On 6 Feb 2002 04:57:54 GMT, Philip Swartzleonard <starx at pacbell.net>
wrote:
> >G. Willoughby || Tue 05 Feb 2002 07:56:57a:
> >
[file counting code]
> >
> >Well, first determine how long it actually takes (time it), and see how
it
> >compares to say, selecting everything on the drive and hitting properties
> >in the context menu to get a count (i.e. see how good windows itself is
at
> >it). It took my 800mhz athlon about 35 seconds to tell me that i have
6.15
> >gigs in 101,064 files in 3,930 folders on my C drive. Counting on that
> >scale can get slow.
>
> Also, the bottleneck here is going to be the drive's access time, not
> the processor speed.

True to a degree, although multi-level caching will likely mean that it's
difficult to determine the effects directly.

>     The access time is usually measured in milliseconds
> and is the average time for the drive's read/write head to move from one
> location to another on the disk surface. This operation is sometimes
> called a "seek". Since the drive has to perform at least one seek for
> every file, and usually more, the total time spent waiting for the hard
> drive far outweighs the time spent by the processor.
>
This is a completely incorrect assumption. Even the original code is only
visiting the directories, not the actual files themselves. A directory is
just a special file containing data about the files and directories it
contains.

If the whole directory is on a single cylinder (likely for most directories,
but false if a directory is extended after the cylinder it was created on is
full), at most one seek is required to get the heads into position to read
it. Since the content of the files is not being read, no further seeks are
required per file.

> On the other hand, I do see one thing you could do to improve your code.
> Chage the countFiles() function thus:
>
> ------------------------------------------------------------------
> def countFiles(arg, dir, files):
>     count = count + len(files)
> ------------------------------------------------------------------
>
That's certainly a good suggestion. Although disk accesses are almost
certain to dominate timings to the extent that it will make hardly any
difference <0.8 wink>. Also, as you point out below, the original code
couldn't possibly work as shown anyhow, so we're reading only the flavor of
the real program.

> Also, I presume you're using a global variable to hold your count,
> unlike the trimmed-down example code you showed us. In the code you
> showed us above, the count variable would be a variable local to
> countFiles() and would not hold its value across separate invocations of
> countFiles(). Instead, you would do:
>
> ------------------------------------------------------------------
> count = 0
>
> def countFiles(arg, dir, files):
>     global count
>     count = count + len(files)
> ------------------------------------------------------------------
>
This is good advice. Of course, the only reason the original was cast the
way it was is so that the OP could use the walk() function - countFiles() is
the callback for walk().

> The "global count" statement inside countFiles() specifies that from
> here on, the name count will refer not to a variable local to
> countFiles() but instead to the global variable of that name.
>
> Gee, this is reminding me of that student blooper (probably an urban
> legend) that went, "The Iliad was not written by Homer, but by another
> man of that name." :)

someone-also-called-steve-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From core.lists.python at core-sdi.com  Tue Feb 12 10:45:32 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Tue, 12 Feb 2002 12:45:32 -0300
Subject: 19 y/o blonde want to know something about Python working with IPX protocol
Message-ID: <001b01c1b3dc$4e243fa0$6e42a8c0@corelabs.coresdi.com>

DISCLAIMER:

Hi!, I tried to get this same question answered before, but nobody did. So
I'm using the "19 y/o blonde strategy"


----

Hi boys!

I want some examples of Python dealing with the IPX protocol.
I do only need a simple, so so simple example which sends and receive strings between 2 computers using IPX.

Could you help me?

I need someting like ........

(PC number 1)
SEND "I'm gonna aply some sun-tan oil over my californian body"

(PC number 2)
[Received-->] "I'm gonna aply some sun-tan oil over my californian body"

SEND "...but you are only pretending to be a 10aged girl just to get this question answered!, STOP that, or everybody will start thinking you're gay!"

(PC number 1)
[Received-->]
"...but you are only pretending to be a 10aged girl just to get this question answered!, STOP that, or everybody will start thinking you're gay!"



-----

That's it.
Thank you all! 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020212/7f1ef5a7/attachment.html>

From aahz at panix.com  Wed Feb 20 15:52:41 2002
From: aahz at panix.com (Aahz Maruch)
Date: 20 Feb 2002 12:52:41 -0800
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com>
Message-ID: <a512ap$dep$1@panix2.panix.com>

In article <b03e80d.0202201228.76ac8525 at posting.google.com>,
Sandy Norton <sandskyfly at hotmail.com> wrote:
>
>This simple little program retrieves web pages using threads and
>urllib. Now it seems to work fine with many websites, however, more
>often than not it just hangs there and the thread (or is it the
>socket?) does not time out and die.
>
>Is there anyway to specify a time limit in which the operation can
>happen and enforce it in the code?

http://www.timo-tasi.org/python/timeoutsocket.py
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"Not everything in life has a clue in front of it...."  --JMS


From phr-n2002a at nightsong.com  Sun Feb  3 01:32:10 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 02 Feb 2002 22:32:10 -0800
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com>
Message-ID: <7xbsf7dqj9.fsf@ruckus.brouhaha.com>

Joshua Macy <l0819m0v0smfm001 at sneakemail.com> writes:
> try:
>      from __future__ import division
> except ImportError:
>      pass
> 
> Of course, the semantics of division will then vary depending upon
> whether the type is integer...but that's the problem that future
> division is trying to solve, so I don't see any way around that.

So much for compatibility then.


From philh at comuno.freeserve.co.uk  Wed Feb  6 08:44:27 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 6 Feb 2002 13:44:27 +0000
Subject: Autocoding project proposal.
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net> <mailman.1012972032.30205.python-list@python.org> <gbl16u4ug6pau2l393eh50t85o031csmtm@4ax.com>
Message-ID: <slrna62cpr.afn.philh@comuno.freeserve.co.uk>

On Wed, 06 Feb 2002 06:58:46 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>>Figure it out already!!!!
>
>Has anyone happened to have seen the BCG research? One of their
>data ranked the what traits opens source developers value most
>in the project leader. The top trait was, at 48%, "makes significant
>early contribution to the code."
>
>Basically, you're not much of an open source software developer
>unless you walk into the announcement with a prototype that people
>can sink their teeth into. Hordes of people with ideas and little
>more than 1,000 LOC attention spans soon follow. As always, these
>efforts are brought to fruition off of a small group of dynamos
>who really know how to make their keyboards ring.

You can sometimes get a lot into 1000 LOC. Herbrip is currently
about 4000 LOC, but 1000 of that is the test suite. Of the rest,
about half is comments and blank lines. So Herbrip is about
1500 LOC of actual code. But some of the code in the main Herbrip
source is not executed (it's a library module) or is test code for
the module, or implements logging (which isn't necessary apart from
testing and debugging).

I bet I could write a working herbivore system in less than 1000 LOC,
if I wanted to. It wouldn't be pretty, mind. But it would testify
to Python being expressive: 1 Python LOC is about 5 C++ LOC or
3 Java LOC or 8 C LOC.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From boris at cantal.lip6.fr  Thu Feb 14 11:25:12 2002
From: boris at cantal.lip6.fr (Boris Boutillier)
Date: Thu, 14 Feb 2002 17:25:12 +0100
Subject: Is there a reverse operator to the '%' operator on string ?
References: <20020214.121107.1335354340.24152@cantal.lip6.fr> <a4glqq$2oj$1@animus.fel.iae.nl>
Message-ID: <20020214.172512.7684930.24152@cantal.lip6.fr>

Yes, this is something like that that I'll have to do...
Thanks for Help

Boris Boutilier


In article <a4glqq$2oj$1 at animus.fel.iae.nl>, "Carel Fellinger"
<cfelling at iae.nl> wrote:

> Boris Boutillier <boris at cantal.lip6.fr> wrote: ...
>> Now  what I'd like to do is revert the '%' operator, I mean I want to
>> get the dictionary from the name and the Format something like :
> 
>>>NameDict = Format  op  Name .. = 'w%(width)dn%(nWord)d_%(fileName)s' op
>> "w16n12_Toto" which will give {'width':16,'nWord':12,'fileName':
>> 'Toto'}
> 
>> Is there such an operator, function or simple way to do this ?
>> Something like the 'scanf' C function ?
> 
> What about regex's, like:
> 
>>>> import re
>>>> matcher = re.compile(r"^w(\d+)n(\d+)_(.*)$").match
>>>> matcher("w16n12_Toto").groups()
> ('16', '12', 'Toto')
>


From fperez528 at yahoo.com  Sun Feb 17 09:28:45 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 17 Feb 2002 14:28:45 +0000
Subject: Q. on make install.
References: <mailman.1013978007.26076.python-list@python.org>
Message-ID: <a4p7ec$nbc$1@peabody.colorado.edu>

infotechsys wrote:

> Hi,
> I'm in the process of installing Pythonw 2.1.2 on my
> Redhat release 7.2. I don't what to remove 1.5.2 that
> came with the Redhat release for fear of messing thier
> code up. So, do I do a make altinstall and should I
> set any env. variables? The reason for the upgrade
> is to use GNUe software.
> Thanks.
> Wayne

Why don't you just install the redhat 7.2 python2 rpms? That will give you 
python 2.1.1 (I think) with no directory conflicts. Look in your cds for rpms 
that begin python2...rpm

hth,

f


From max at alcyone.com  Thu Feb 21 04:25:28 2002
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 21 Feb 2002 01:25:28 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3c7455a8_4@corp.newsgroups.com> <jwbaxter-CE1C5D.00553921022002@corp.supernews.com>
Message-ID: <3C74BD08.FD8CAACE@alcyone.com>

John Baxter wrote:

> I don't think strings contain characters.  I think strings contain
> one-character strings.

Naturally, because there is no distinct concept of a character in Python
as opposed to a one-character string.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From mdk at mdk.com  Tue Feb 12 16:46:52 2002
From: mdk at mdk.com (MDK)
Date: Tue, 12 Feb 2002 16:46:52 -0500
Subject: tcl question
Message-ID: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de>

In PythonWin I entered this script from a book:

from Tkinter import Label
widget = Label(None, text='Hello GUI world!')
widget.pack()
widget.mainloop()

When I run it in PythonWin it runs fine.

When I double click on it in Windows explorer I get a message that tcl80.dll
can not be found.  That file does not exist on my PC.  How then is it
running under PythonWin if that file does not exists and how can I make it
run when double-clicked?

My path includes:

PATH=C:\Python22\;c:\python22\dlls\;C:\Python22\Lib;C:\Python22\Lib\lib-tk

Thanks.




From janfbr at sex.ifi.uio.no  Mon Feb 18 18:00:50 2002
From: janfbr at sex.ifi.uio.no (Jan-Fredrik Braseth)
Date: 19 Feb 2002 00:00:50 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl>
Message-ID: <xyuvgcujsyl.fsf@sex.ifi.uio.no>

"Patrick Vrijlandt" <p.vrijlandt at aig.azn.nl> writes:

> Hi,
> 
> Previous discussions have narrowed down the line in telnetlib that is buggy.
> The problem is probably that the server is sending you a special (Telnet)
> command.

Even though I haven't looked very closely it looks like they just forgot
to initiate the 'opt' variable in one part of the if-else branch

-- 
Jan-Fredrik Braseth		www.ifi.uio.no/~janfbr/
2B|!2B


From logiplexsoftware at earthlink.net  Mon Feb  4 12:21:53 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 4 Feb 2002 09:21:53 -0800
Subject: Autocoding finally explained (was Re: READ THE WEB-SITE THAT THE KIKES TRY TO CENSOR)
In-Reply-To: <A042AF87D5C59948FB6368C76F958@news.wanadoo.fr>
References: <A042AF87D5C59948FB6368C76F958@news.wanadoo.fr>
Message-ID: <20020204092153.77a1ce35.logiplexsoftware@earthlink.net>

On Mon, 4 Feb 2002 14:35:23 GMT
charlie at news.wanadoo.fr wrote:

> Other sticky bizarre ointments will order annually before shirts.  
> They explain once, sow partly, then fear in back of the tyrant 
> at the monolith.  She'd rather scold lovingly than answer with 
> Yolanda's younger hat.  She wants to call elder units against 
> Neil's window.  
> 
> It can recollect wickedly if Norbert's pear isn't lost.  Try not to 
> tease the cans halfheartedly, clean them usably.  Lots of fresh 
> diets through the clean stadium were arriving towards the durable 
> cave.  Both playing now, Corinne and Byron dyed the polite barns 
> towards cold spoon.  
> 
> She might sneakily irritate empty and attacks our heavy, hot 
> raindrops in back of a ceiling.  For Edna the tree's closed, 
> in me it's stale, whereas beneath you it's moving lower.  Hardly any 
> abysmal painter or sunshine, and she'll wastefully cook everybody.  
> 
> Charlene attempts, then Sherry virtually fills a weak jug in 
> Ronnie's hallway.  Ben, beside yogis humble and proud, laughs 
> alongside it, tasting easily.  
> 
> If you will behave Norris's road beneath cases, it will daily 
> learn the plate.  What will we excuse after Lionel converses the 
> ugly castle's ache?  If the filthy buckets can measure eventually, the 
> pathetic bandage may lift more camps.  It's very sad today, I'll 
> creep firmly or Nydia will change the wrinkles.  How doesn't 
> Peter kill fully?  Tomorrow, it covers a cup too dry alongside her 
> long lake.  

Oh! So _that's_ what Rue meant by "autocoding"....

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From ajeru at gmx.net  Wed Feb 20 21:05:40 2002
From: ajeru at gmx.net (Alexander Jerusalem)
Date: Thu, 21 Feb 2002 03:05:40 +0100
Subject: Adding static typing to Python 
In-Reply-To: <200202210750.g1L7o3Go032360@ratthing-b246.strakt.com>
References: <Message from ajeru@vknn.org (Alexander Jerusalem)  of "20 Feb 2002 13:00:29 PST." <24c39b2c.0202201300.3dfe7dbc@posting.google.com>
 <24c39b2c.0202181531.187fad4c@posting.google.com>
 <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com>
 <24c39b2c.0202191646.776c1993@posting.google.com>
 <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com>
 <24c39b2c.0202192354.765da777@posting.google.com>
 <mailman.1014210154.9297.python-list@python.org>
 <24c39b2c.0202201300.3dfe7dbc@posting.google.com>
Message-ID: <5.1.0.14.0.20020221025222.02aad830@pop.gmx.net>

In a static type system I would use an interface to tell everyone who wants 
to use my function: If you want to use this function, give your object this 
interface and everything will be fine. No one has to look at my code to 
find out what they have to provide for to use my function. But I agree with 
you that it'd be even better if it wasn't necessary to provide a fixed 
interface, that forces you to implement all the methods even though the 
function you want to use only makes use of one of them.

In the paper you cite, the author speaks of the detection of protocols and 
wrappers. I'm completely sure if he intends this to be a dynamic process at 
runtime or a kind of "compile-time" tool. I can imagine using it both at 
runtime and to check a code base.

But there's another feature of static typing that I find extremely useful 
and that's type based dispatching, ideally multi methods. And that's not 
covered by dynamic protocoll detections as suggested in the paper as far as 
I understand it.

Regards,
Alexander

At 08:50 21.02.2002 +0100, Laura Creighton wrote:
> > Alexander Jerusalem:
> > The proposals of the types-SIG say that static typing is used on a per
> > module basis. That means that weather there is or is not a performance
> > improvement it's only going to affect you if you choose to use the
> > static types.
> >
> > Alexander
>
>The problem is that static typing effects everybody in terms of code
>reusability.  Let us say you make a terrific function which works on
>a string.  And in some way you specify that, yes, the first and only
>argument to this function is a string.
>
>Now, let us assume I love your function and want to use it a lot.
>But I want to use it on my own objects which are in some way 'string
>like'.  Right now I have to read your code, and determine what
>were the properties of a string you used, to determine whether my
>stringlike object can already be used with your function.  I would
>like to automate this, and be able to interrogate both your function
>and my object and get a result  'will work' or 'won't work: reason
>object does not have an islower method which function requires' and
>the like.
>
>What I don't want to have to do is rewrite the code, because
>Alexander Jerusalem was only thinking about strings when he wrote it.
>Static type checking makes you think that 'strings' and 'ints' and the
>like a real meaningful things rather than a convenient shorthand for
>'is a sequence and has the following methods':
>
>['capitalize', 'center', 'count', 'encode', 'endswith', 'expandtabs', 'find',
>  'index', 'isalnum', 'isalpha', 'isdigit', 'islower', 'isspace', 'istitle',
>  'isupper', 'join', 'ljust', 'lower', 'lstrip', 'replace', 'rfind', 'rindex',
>  'rjust', 'rstrip', 'split', 'splitlines', 'startswith', 'strip', 'swapcase',
>  'title', 'translate', 'upper']
>
>See 'PEP 246: Object Adaptation - (Clark C. Evans)'
>http://python.sourceforge.net/peps/pep-0246.html  for one attempt to
>get the safety you want without the restriction I don't.
>
>Laura Creighton




From google at cb921.org  Fri Feb  8 07:41:13 2002
From: google at cb921.org (cb921)
Date: 8 Feb 2002 04:41:13 -0800
Subject: mxODBC, stored procedure syntax
Message-ID: <e746a5a8.0202080441.4bd2c023@posting.google.com>

Hi,

I have been trying to connect to an MS SQL Server via ODBC using mxODBC. 
It's all working, but calling of a stored procedure is causing problems.
Without any syntax worries, it looks like:

cr.execute("{call storedprocedure('cellnr','passwd')}")

and it should return me two output parameters, in the example (not Py :) code
I've seen we call this @DATABASENAME and @ISNOM.  How do I retrieve those
values?  So far all I've managed is a flock of errors.

Anyone who can help me, even an insignificant bit, thank you...

Cheerio,


From tjreedy at home.com  Mon Feb  4 13:24:00 2002
From: tjreedy at home.com (Terry Reedy)
Date: Mon, 04 Feb 2002 18:24:00 GMT
Subject: working with files with the same extenstion
References: <3C5EAF50.80004@rmb.co.za>
Message-ID: <4lA78.25495$X64.8227413@news1.rdc2.pa.home.com>

"Ulrich Pfisterer" <ulrich.pfisterer at rmb.co.za> wrote in message
news:3C5EAF50.80004 at rmb.co.za...
> Can anyone please tell me how I would go about doing the following.
> I would like to write a script that would take in a command line
> parameter "*.txt". The script would then loop through every file
with a
> "txt" extension and something with it.

Try the glob module




From peter at engcorp.com  Mon Feb 25 20:00:02 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 25 Feb 2002 20:00:02 -0500
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de>
Message-ID: <3C7ADE12.C6D895A0@engcorp.com>

Emile van Sebille wrote:
> 
> "Joshua Muskovitz" <joshm at taconic.net> wrote in message
> news:3c7ab974_4 at corp.newsgroups.com...
> > Hey all,
> >
> > If you had to give a 15 minute introduction to Python, what would you
> cover?
> > What is a good example of a simple, but useful, text-based app for a
> Wintel
> > environment, that has good "wow" quality?
> >
> 
> Hmm... text-based... "wow" appeal... oxymoron?  ;-)
> 
> Otherwise, what Sheila said.

Something text-based that retrieves some useful information
from the web (a weather site?) and parses the result and
displays something useful with a few lines of Python code
using the standard modules would have enough "wow" appeal
for anyone used to other languages...

At least, it "wow"ed me when I did this sort of thing
from the interactive prompt for the first time... :)

-Peter


From caljason76 at yahoo.com  Wed Feb 20 20:27:05 2002
From: caljason76 at yahoo.com (Jason)
Date: 20 Feb 2002 17:27:05 -0800
Subject: What values are considered false?
Message-ID: <e9e6dd1f.0202201727.9741615@posting.google.com>

In my experimentation 0, None, "", [], and () all seem to be
considered false.  Just to confound me none of them are == to each
other.  Coming from a Scheme world this is confusing.  Are there any
other values that are false?  Why aren't () and "" the same thing? 
Don't both describe an immutable sequence of values (basically, why
aren't strings either lists or tuples)?


From emile at fenx.com  Wed Feb  6 11:53:54 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 08:53:54 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de>
Message-ID: <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de>

"MDK" <mdk at mdk.com> wrote in message
news:a3rbad$19u08k$1 at ID-98166.news.dfncis.de...
> I need to convert a list into a list of characters.
>
> For example:
>
> ("cat",5,['dog',[3,3,1]],"zoo")
>
> Would become:
>
> ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
>

def flatten(ary, rslt):
    try:
        flatten(ary[0], rslt)
        flatten(ary[1:], rslt)
    except:
        if ary: rslt.append(ary)

rslt = []
flatten(("cat",5,['dog',[3,3,1]],"zoo"), rslt)
print 'rslt = ',tuple(rslt)

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From grante at visi.com  Mon Feb 25 12:13:00 2002
From: grante at visi.com (Grant Edwards)
Date: Mon, 25 Feb 2002 17:13:00 GMT
Subject: call me
References: <u7fcig3gep8k40@corp.supernews.com> <slrna7fh30.a72.gerhard@lilith.my-fqdn.de>
Message-ID: <wgue8.3490$N7.975881@ruti.visi.com>

In article <slrna7fh30.a72.gerhard at lilith.my-fqdn.de>, Gerhard H?ring wrote:
> Scott Hathaway wrote in comp.lang.python:
>> Mom,
>> 
>> Give me a call when you have a chance.
> 
> Wow, a familiy of diehard Python fans :-)

I was proud that my mom cut an article about Linus Torvalds out
of the paper and sent it to me, but I don't ever expect to run
into her on Usenet.

-- 
Grant Edwards                   grante             Yow!  My pants just went to
                                  at               high school in the Carlsbad
                               visi.com            Caverns!!!


From mdk at mdk.com  Fri Feb  8 15:13:07 2002
From: mdk at mdk.com (MDK)
Date: Fri, 8 Feb 2002 15:13:07 -0500
Subject: What is the most efficient way to do this?
References: <mailman.1013197035.25987.python-list@python.org>
Message-ID: <a41bgk$1bi72p$1@ID-98166.news.dfncis.de>

"Mark McEahern" <marklists at mceahern.com> wrote in message
news:mailman.1013197035.25987.python-list at python.org...
> [MDK]
> > What is the most efficient way to remove empty items from this list?
> >
> > ['', '1', '2', '3', '', '456', '789', 'wow']
>
> Try filter:
>
> >>> l = ['', '1', '2', '3', '', '456', '789', 'wow']
> >>> filter(None, l)
> ['1', '2', '3', '456', '789', 'wow']
>

Very nice.  Thanks!




From jason at jorendorff.com  Thu Feb 21 13:56:03 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 21 Feb 2002 12:56:03 -0600
Subject: Recursively traverse linked list -- help!
In-Reply-To: <mR%c8.57438$X64.20569489@news1.rdc2.pa.home.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOEGEDEAA.jason@jorendorff.com>

Terry Reedy wrote (to Paul Rubin):
> So, are you suggesting
> 
> A. general tail call optimization?
> 
>   If so, 'jump to' is problematic, since there is no frame object to
> jump to, the avoidance of creating such being part of the purpose of
> the optimization.  If the call is not recursive, I could imagine
> replacing the contents of the current frame.

That is not feasible, generally.

Here is how it might work.  Python could create a new frame.
The optimization is that the frame performing the tail call can be
culled right away, saving some memory and shortening the stack.
The culled frame would disappear from traceback information too,
unfortunately.

I'm sure this has been discussed endlessly before.

## Jason Orendorff    http://www.jorendorff.com/



From bokr at oz.net  Sat Feb 16 15:07:28 2002
From: bokr at oz.net (Bengt Richter)
Date: 16 Feb 2002 20:07:28 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no>
Message-ID: <a4me60$brk$0@216.39.172.122>

On 16 Feb 2002 18:08:10 +0100, Jan-Fredrik Braseth <janfbr at sex.ifi.uio.no> wrote:

>Sheila King <usenet at thinkspot.net> writes:
>
>> On 16 Feb 2002 13:56:35 +0100, Jan-Fredrik Braseth
>> <janfbr at sex.ifi.uio.no> wrote in comp.lang.python in article
>> <xyuvgcxlh4s.fsf at sex.ifi.uio.no>:
>> 
>> >  but after some time I get this error message:
>> > 
>> > Traceback (most recent call last):
>> >   File "./quizbot.py", line 366, in ?
>> >     foo = QuizBot(name,password,channel)
>> >   File "./quizbot.py", line 28, in __init__
>> >     line = self.tn.read_until("\n").split()
>> >   File "/usr/lib/python2.2/telnetlib.py", line 297, in read_until
>> >     self.process_rawq()
>> >   File "/usr/lib/python2.2/telnetlib.py", line 424, in process_rawq
>> >     self.msg('IAC %d not recognized' % ord(opt))
>> > UnboundLocalError: local variable 'opt' referenced before assignment
>> > > 
>> You have an error in your code. Show us the code. I'm guessing, that you
>> have a variable 'opt' that under some circumstances is not defined
>> before entering a particular loop or something.
>
>If you look closely at the traceback you see that the error with opt is in the
>telnetlib.py, that's a module for python. But it's unlikely that there is a bug
>in python itself, isn't it?
>
>BTW: it's python 2.2
>
I am not familiar with the code, but from a cursory look ISTM
there is a bug, and UnBoundLocalError nails it well.
I suspect line 424
    self.msg('IAC %d not recognized' % ord(opt))
should be
    self.msg('IAC %d not recognized' % ord(c))
since lacking a recognized value for c for the character gotten
after IAC, opt does not seem to be be set:
(snip from Python22\lib\telnetlib.py)
---------
                if c != IAC:
                    buf = buf + c
                    continue
                c = self.rawq_getchar()
                if c == IAC:
                    buf = buf + c
                elif c in (DO, DONT):
                    opt = self.rawq_getchar()
# ok here
                    self.msg('IAC %s %d', c == DO and 'DO' or 'DONT', ord(opt))
                    if self.option_callback:
                        self.option_callback(self.sock, c, opt)
                    else:
                        self.sock.send(IAC + WONT + opt)
                elif c in (WILL, WONT):
                    opt = self.rawq_getchar()
# ok here
                    self.msg('IAC %s %d',
                             c == WILL and 'WILL' or 'WONT', ord(opt))
                    if self.option_callback:
                        self.option_callback(self.sock, c, opt)
                    else:
                        self.sock.send(IAC + DONT + opt)
                else:
# not ok here: c seems to be the problem. opt would come from the *next* char,
# stored by  opt = self.rawq_getchar(), but that isn't happening here,
# and printing opt is not appropriate.
# I would change opt to c in the following line
                    self.msg('IAC %d not recognized' % ord(opt))
---------

HTH get to the real problem ;-)

Regards,
Bengt Richter



From frr at ThouShallNotSpam.EasyJob.NET  Wed Feb 27 10:31:12 2002
From: frr at ThouShallNotSpam.EasyJob.NET (Fernando Rodríguez)
Date: Wed, 27 Feb 2002 16:31:12 +0100
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
Message-ID: <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>

On Wed, 27 Feb 2002 15:05:01 +0000, Chris <nospam@[127.0.0.1]> wrote:

>In the olden days I played around with BASIC
>just for fun, writing trivial games and making patterns move on the 
>screen.
>
>Can I do that sort of thing with Python?

Yes, and Python a better and easier language than Basic.

>I've at last got a new machine, running XP Home.
>Will this be OK?

Yes, that's more than enough.


>Before I start wrestling with Python - and with thinking - and with 
>making my brain hurt - is Python the best language for this sort of 
>play?

If you want to learn how to program, then I wouldn't recommend using Python.
Python _is_ a great language for beginners, but there many _great_ (and I mean
great) books for beginners that use it.

I recommend you buy a copy of this book: "The Schemer's Guide" at
www.schemers.com. It's an excellent, short and gentle (let me stress the word
_gentle_) intro to programming and computer science. It uses Scheme, a
language designed for teaching, that is quite similar to Python. IMHO, this is
the most painless way to quickly become a viable programmer.

Search for "DrScheme" at google to get yourself a free Scheme environment.

>Would Java be better?

Nope.  Java is a good second language if you want to get serious about
programming. Personally, I would start with Scheme and read that book.

>Will I be able to make a ball bounce around inside a box etc.?

Once you finish with the book, you will be able to write a game that learns
from its own mistakes (that's the final chapter).  :-)

PS I'm not the author and I'm not related in any way to the editors.




| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Fernando Rodr?guez   frr at EasyJob.NET 
| http://www.EasyJob.NET/
| Expert resume and cover letter creation system. 
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


From mhuening at zedat.fu-berlin.de  Mon Feb 18 02:32:28 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 18 Feb 2002 07:32:28 GMT
Subject: win32com and MS Excel
References: <mailman.1013967115.15656.python-list@python.org>
Message-ID: <Xns91B956DBE9835mhueningzedatfuberli@130.133.1.4>

"Jimmy Retzlaff" <jimmy at retzlaff.com> wrote in 
news:mailman.1013967115.15656.python-list at python.org:

> 
> The approach I've used is as follows:
> 

Thank you very much! I'll give that a try.

Matthias


From fperez528 at yahoo.com  Fri Feb 15 23:01:00 2002
From: fperez528 at yahoo.com (Fernando Perez)
Date: Fri, 15 Feb 2002 21:01:00 -0700
Subject: line break confusion
References: <mailman.1013827165.15861.python-list@python.org>
Message-ID: <a4klhs$63u$1@peabody.colorado.edu>

Michael Mell wrote:

> Using Python 2.2 running on any arbitrary OS (Linux, Win, Mac), I want
> to
>     1) read several files having an arbitrary mix of line break types
> (Linux, Win, Mac)
>     2) then standardize the breaks of each of the files to a single type
> and
>     3) then concatenate them all into a string for writing.

This is probably a useful start:

#-----------------------------------------------------------------------------
def native_line_ends(file,backup=1):
    """Convert (in-place) a file to line-ends native to the current OS.

    If the optional backup argument is given as false, no backup of the
    original file is left.  """

    backup_suffixes = {'posix':'~','dos':'.bak','nt':'.bak','mac':'.bak'}

    bak_file = file + backup_suffixes[os.name]
    
    original = open(file).read()
    shutil.copy2(file,bak_file)
    try:
        new = open(file,'w')
        new.write(os.linesep.join(original.splitlines()))
        new.write(os.linesep) # ALWAYS put an eol at the end of the file
        new.close()
    except:
        os.rename(bak_file,file)
    if not backup:
        try:
            os.remove(bak_file)
        except:
            pass

Modify it to suit your needs.

Cheers,

f.


From loewis at informatik.hu-berlin.de  Thu Feb 21 10:40:15 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 21 Feb 2002 16:40:15 +0100
Subject: No space left on device
References: <mailman.1014249279.18923.python-list@python.org>
Message-ID: <j4d6yyq1wg.fsf@informatik.hu-berlin.de>

sag at hydrosphere.com (Sue Giller) writes:

> So, I am assuming there is some problem with the FAT32 system 
> when a directory gets lots of files in it.  Since I can make the 100K 
> files if I don't use mktemp, I am also assuming there is some 
> problem with mktemp in this scenario.

Unlikely. More likely, the length of each individual filename matters.
Also, as you've reported, it is the open() that is failing, not the
mktemp. Since, in your test, mktemp returns the file names in a
predictable way, just try removing that call and create the files
explicitly in the order in which they get created.

> If anyone figures this out, it would be interesting to know what is 
> going on.  Meanwhile, I guess I'll need to redesign my file 
> generation algorithm to split things up.

It is very easy. You are exceeding a limit of the operating
system. What the precise limit in terms of number of files is is
difficult to say, as it depends on the specific file names, in
particular, on whether some of them are long filenames.

Regards,
Martin



From just at xs4all.nl  Sat Feb  9 17:40:58 2002
From: just at xs4all.nl (Just van Rossum)
Date: Sat, 09 Feb 2002 23:40:58 +0100
Subject: Python MySQL and MAC OS X
References: <3C656ACD.1020907@inter.nl.net> <mailman.1013294776.15973.python-list@python.org>
Message-ID: <3C663807.AA608724@xs4all.nl>

Gerhard H?ring wrote:
> 
> Le 09/02/02 ? 19:30, Peter HJ van der Kamp ?crivit:
> > Hello,
> >
> > I'm trying to install the Python MySQL module, without success.
> > According to the MAC OS X documentation of Python 2.0 I have
> > to drop the setup.py file onto the interpreter and pass the
> > build argument by setting a Unix style command line (which can
> > be set by holding the option key during interpreter startup).
> 
> Sounds frightening ;-) Don't you have a shell or terminal window on
> MacOS X?

Sure there is. I think maybe he's not yet aware that there are still two
very different version of Python available for OSX: MacPython, which
is still backwards compatible with MacOS 9 and earlier, and then there's
plain vanilla "unix Python". It sounds like he's trying with MacPython
where he really needs unix Python.

Just


From jeff at ccvcorp.com  Wed Feb 20 13:10:00 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 20 Feb 2002 10:10:00 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <3C72AB39.28618CC3@ccvcorp.com> <24c39b2c.0202191642.1bfbfaad@posting.google.com>
Message-ID: <3C73E678.84B9551F@ccvcorp.com>


Alexander Jerusalem wrote:

> Jeff Shannon <jeff at ccvcorp.com> wrote in message news:<3C72AB39.28618CC3 at ccvcorp.com>...
>
> > This is exactly what Python does, except that it does it at runtime
> > instead of compile time; and really, it is *not* any slower to test this
> > than it is to try a couple of compile cycles.
>
> From my experience I can say it is much slower but maybe one has to
> get used to this style of programming.

Just yesterday, I was refactoring a Python COM object I'm working on.  Within the space of
five minutes, I found an error I'd made, found that my correction was incomplete, and found
that *that* correction was incomplete too.  This is including a space of about a minute each
time to restart the COM client that was using my object (It has to be restarted in order to
have it release the reference, or I don't see my changes).  Had I been using a compiled
language, then each of these errors would've require a separate full compile, and in at
least one case would've still required testing -- the compiler wouldn't have found the
error.  I doubt that I could've done that within five minutes.  And the exception tracebacks
I got pointed me to exactly where the problem was, each time.  Quick, easy, painless.
>shrug<  YMMV, of course, but I'm happy with the Python way.

Jeff Shannon
Technician/Programmer
Credit International




From andreas at andreas-jung.com  Wed Feb 20 12:24:07 2002
From: andreas at andreas-jung.com (Andreas Jung)
Date: Wed, 20 Feb 2002 12:24:07 -0500
Subject: Python 2.1.1 on Solaris 7 broken
References: <26C20D46-2625-11D6-B442-0030656BCC12@laplaza.org>
Message-ID: <098501c1ba33$70554760$3e17a8c0@suxlap>

try to set LD_LIBRARY_PATH to the path where gcc has its libs
installed.

Andreas
----- Original Message ----- 
From: "Alex Chavez" <alex at laplaza.org>
To: <python-list at python.org>
Sent: Wednesday, February 20, 2002 12:13
Subject: Python 2.1.1 on Solaris 7 broken


> I just upgraded Python from 1.5.1 to 2.1.1 on a Solaris 7 Sparc machine 
> and I get the following error when I try running a scrip.
> 
> root at elvis cron $ python -i run_queue
> Traceback (most recent call last):
>    File "run_queue", line 26, in ?
>      from Mailman import OutgoingQueue
>    File "/usr/local/mailman/Mailman/OutgoingQueue.py", line 41, in ?
>      import time
> ImportError: ld.so.1: python: fatal: libgcc_s.so.1: open failed: No such 
> file or directory
>  >>> ZZ^Z
> [2]+  Stopped                 python -i run_queue
> 
> 
> I'm not sure if the gcc version had anything to do with it but here is 
> that info.
> 
> root at elvis cron $ gcc -v
> Reading specs from /usr/local/lib/gcc-lib/sparc-sun-
> solaris2.7/2.8.1/specs
> gcc version 2.8.1
> root at elvis cron $
> 
> 
> 
> Help would be greatly appreciated.
> 
> alex
> 
> ......:::  Alex Chavez  :  System Admin :::......
> La Plaza Telecommunity : www.laplaza.org
> :   Taos.  New Mexico   :   505.758.1836    :
> ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 




From greg at cosc.canterbury.ac.nz  Thu Feb 14 21:05:00 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 15 Feb 2002 15:05:00 +1300
Subject: Proposal for quoted octal
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <slrna6o1i2.u4q.huaiyu@gauss.almadan.ibm.com>
Message-ID: <3C6C6CCC.C5049297@cosc.canterbury.ac.nz>

Huaiyu Zhu wrote:
> 
>    from __future__ import zerolead_octal so that 041 will mean 33.

Shouldn't that be from __past__ import zerolead_octal? :-)

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


From weeks at vitus.scs.agilent.com  Tue Feb 12 11:48:33 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Tue, 12 Feb 2002 16:48:33 +0000 (UTC)
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4>
Message-ID: <1013532513.171199@cswreg.cos.agilent.com>

: Aahz Maruch || Fri 08 Feb 2002 11:10:47a:
: I know some of y'all still keep roots to the Perl community, so I was
: wondering if there's any kind of definitive nickname for Perl
: programmers, the way we call ourselves Pythonistas?

Where I come from they're called "hardware engineers".  (Honest.)


Greg


From warlock at eskimo.com  Fri Feb 22 18:48:41 2002
From: warlock at eskimo.com (Jim Richardson)
Date: Fri, 22 Feb 2002 15:48:41 -0800
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1> <a54qfl$vl4$0@216.39.172.122> <hhu45a.4d1.ln@127.0.0.1> <7xy9hlubo8.fsf@ruckus.brouhaha.com>
Message-ID: <pcl65a.1s5.ln@127.0.0.1>

On 22 Feb 2002 01:07:51 -0800,
 Paul Rubin <phr-n2002a at nightsong.com> wrote:
> Jim Richardson <warlock at eskimo.com> writes:
>> Thanks, this isn't a bad idea (I am not really python guru, so am easily
>> overwhelmed sometimes :) I will have to think on it. Speed isn't much of
>> an issue, as the data is only 9600 and each packet needs a checksum, and
>> usually, only 8-10 packets a second. 
>>  Another reply mentioned a "byte" operator, but I can't find that
>> anywhere in my python docs. Is it something not in the default setup? 
> 
> That was me.  There's no byte operator.  I was using 'byte' as a
> variable name to denote an integer variable that held an 8-bit number.
> Sorry for the confusion.


Ah, that explained it, you assumed I was competent python programmer,
and failed to read my mind. all your fault :)

-- 
Jim Richardson
	Anarchist, pagan and proud of it
www.eskimo.com/~warlock
    Linux, for when you need to get work done, Or you could just play UT...


From gh_pythonlist at gmx.de  Sun Feb  3 23:17:15 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 4 Feb 2002 05:17:15 +0100
Subject: Forwarding messages with the email package
Message-ID: <20020204041714.GA2141@lilith.hqd-internal>

I'm having some problems with the new email package in Python 2.2. I
guess the reason is that I don't really understand MIME and I fear my
head will explode when I try to make sense of the source code of the
email package.

What I'm trying to do is forward an email message with MIME. I. e. not
inline. I'm trying to duplicate what my MUA does. It creates a
multipart/mixed MIME doc with a text/plain part and a message/rfc822
part, that contains the email to forward. So, my current code is:


#!/usr/bin/env python2.2
import sys, email
from email.Message import Message
from email.MIMEBase import MIMEBase
from email.MIMEText import MIMEText
from email.MIMEMessage import MIMEMessage

forwarded_mail = sys.stdin.read()

me = "gerhard at bigfoot.de"
to = ["ghmail at gargamel.hqd-internal"]
msg = MIMEBase('multipart', 'mixed')
msg['Subject'] = 'Re: (no subject)'
msg['From'] = me
msg['To'] = ", ".join(to)

msg.attach(MIMEText("I send you the following in order to have your advice."))

# Hmm. Do I really need to do this in this two steps?
raw = Message()
raw.set_payload(forwarded_mail)
mail = MIMEMessage(raw)

msg.attach(mail)
print msg


This doesn't seem to work quite right when the message I'm trying to
forward is itself mime/multipart. Does I *really* need to parse the
forwarded message into some MIME object and then attach it to the
"container" message?

Has anybody code to share that does this already (forward an arbitrary
message with the email module)?

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.0 ?C      Wind: 3.3 m/s



From owen at astroNOJNK.washington.edu.invalid  Wed Feb 20 12:39:29 2002
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Wed, 20 Feb 2002 10:39:29 -0700
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <a50n0m$24ge$1@nntp6.u.washington.edu>

In article <m2vgcs5hgf.fsf at desk.crynwr.com>, Russell Nelson 
<nelson at crynwr.com> wrote:

>There are two Python ideoms that bother me:
>
>    while 1:
>        line = file.readline()
>        if not line: break

I agree this is ugly. Fortunately, the language designers did too and 
offer a better way in Python 2.2. If you are willing to only read the 
file from a single for loop, you can write:

for line in file:
   ...process the line

However, if you want to read part of the file now and the rest later in 
your program, do not use this syntax. The first for loop usually reads 
more of the file than it gives you back (for the sake of efficiency), so 
you'll miss lines if you then read more later. In that case, use 
iterators...

There was a recent discussion of this, but I was never clear if there 
was a best way to get an efficient file iterator. There is no 
file.iter() call, which is a great pity (since it would be unambiguous). 
In lieu of that, the following is probably reasonable, but use at your 
own risk:

fileiter = file.xreadlines()
for line in fileiter:
  # process some of the file

# then later...
for line in fileiter:


Regarding assignments returning values...it is discussed every once in 
awhile and I don't think it'll change. (For the record, I find 
assignments returning values somewhat confusing and error-prone. But I 
realize others do like them and admit that they can shorten code.)

-- Russell


From othello at javanet.com  Fri Feb  1 01:56:17 2002
From: othello at javanet.com (Raymond Hettinger)
Date: Fri, 1 Feb 2002 01:56:17 -0500
Subject: Comments invited:  PEP 279 Enhanced Generators
Message-ID: <a3deaj$b54$1@bob.news.rcn.net>

Your review and comments are invited for PEP 279:
   http://python.sourceforge.net/peps/pep-0279.html

In summary, it proposes:

1.  Four new built-in functions:  xmap, xfilter, xzip, and indexed
2.  Generator comprehensions:  [yield int(x) for x in floatgen()]
3.  Feeding data into a generator using an optional argument to .next()
4.  Providing a mechanism to raise exceptions in a generator

This version incorporates all of the critiques I received on earlier
discussion drafts.  The key changes from the earliest draft are:

- xmap, xfilter, and xzip now exactly mimic map, filter, and zip
- index() was renamed to indexed()
- ints() is no longer being proposed
- generator comprehensions now exactly mimic list comprehensions
- .next(arg) is used instead of a separate .submit() method
- control flow is now the same whether or not 'x=yield None' is used
- .throw(exception) is now a separate part of the proposal


Raymond Hettinger








From phr-n2002a at nightsong.com  Thu Feb 21 03:56:25 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 00:56:25 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de>
Message-ID: <7x8z9njjra.fsf@ruckus.brouhaha.com>

"Karl M. Syring" <syring at email.com> writes:
> > Physical simulations and computational geometry don't have much in common.
> 
> Think of simulating a robot arm that does a brain operation on you. I think
> you would immediately turn to proper rational number coordinates.

Not likely.  


From dan at eevolved.com  Wed Feb 13 22:32:20 2002
From: dan at eevolved.com (Dan P)
Date: Wed, 13 Feb 2002 22:32:20 -0500
Subject: RELEASED [OOPS - The Python Datastore v0.9] 
Message-ID: <JmGa8.5516$MR3.311561@wagner.videotron.net>

OOPS - The Python Datastore.

OOPS stands for object-oriented persistent storage which should give you an 
idea of its purpose. It was designed to be as easy to learn as python, to 
have an intuitive interface, to be conservative in memory and processor 
usage; basically to be as simple as possible without sacrificing 
scalability. It runs everywhere Python 2.2 runs (as long as bsddb is 
installed).

OOPS makes it possible to merge database design and application design in 
Python. This results in less code which in turn is more manageable and more 
readable. All OOPS classes are inheritable. In fact you'll see that it has 
been designed with the expressed purpose of leveraging python's 
object-oriented programming concepts to their fullest.

OOPS achieves transparent persistence in the Python environment by 
mirroring the python namespace over top of a shelve-like database. 
Behavioral methods were added to classes to make the bind to each other and 
receive extra identification information so each object could build a 
unique identifying database key in which it stores persistent data.

As this is my first public release, the documentation is not as full as I 
want it to be and I'm certain it isn't as stable either. I still recommend 
trying it out to get a feel for this new dimension in Python programming.

You can download it at sourceforge.net here:

http://sourceforge.net/projects/pyobjfs/

This module is going to be the basis for many projects in the future so try 
it out, recommend it to your friends and happy hacking!

Dan Parisien <dan at eevolved.com>
Software Systems Consultant
eEvolved Real-Time Technologies



From phr-n2002a at nightsong.com  Mon Feb 18 19:13:57 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 18 Feb 2002 16:13:57 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <mailman.1013816158.26552.python-list@python.org> <7934d084.0202172009.75324c2a@posting.google.com> <mailman.1014058862.32020.python-list@python.org>
Message-ID: <7xy9hqib0a.fsf@ruckus.brouhaha.com>

Cliff Wells <logiplexsoftware at earthlink.net> writes:
> > Everyone: Does anyone know of an accessable concurrency text?  One
> > that moves from discussing the basics (data-races, critical sections,
> > mutural exclusion, lock-order based deadlock, resource based deadlock,
> > etc) to the various solutions (mutex/cond-var, semaphore, monitor,
> > rendezvous) to the various design approaches (master-slave, pipeline,
> > work-pile, etc).  Possibly even a discussion of various performance
> > considerations (convoy's, wait-free synchronisation, cache-aliasing,
> > spin vs. blocking locks, etc).
> 
> I would certainly like to see a good discussion of these.  Do you have a
> URL for the tutorial you wrote?

A lot of this stuff isn't much of an issue with Python threads, which
always run in a single process.


From alanmk at hotmail.com  Mon Feb  4 04:42:33 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 4 Feb 2002 01:42:33 -0800
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org>
Message-ID: <f72dac0d.0202040142.43e3a90e@posting.google.com>

Gerard Braad <g_braad at survion.com> wrote:
> Any help would be appreciated. We hope this will become a standard for
> inclusion of Python in XML documents.

Gerard,

I don't understand what's wrong with

<script language="python" xml:space="preserve">
    def messiah():
        print "He's NOT the Messiah."
        print "He's a very naughty boy."
</script>

or something similar.

Alan.

P.S. It's from "The Life of Brian", in case you're wondering.


From paul at nlhas.no  Thu Feb 21 07:51:32 2002
From: paul at nlhas.no (paul at nlhas.no)
Date: Thu, 21 Feb 2002 12:51:32 GMT
Subject: Bug in e-mail package?
Message-ID: <3fq97uoqvtbq4mkf9p6hjl5vereg30n93u@4ax.com>

I have a simple e-mail forwarding/filtering module which essentially
does this:

msg = email.message_from_string('\n'.join(pop3connection.retr(xx)[1]))
.....stuff to determine send-to address (without modifying msg)
smtpconnection.sendmail(email.Utils.parseaddr(msg.get('from'))[1], \
			'send-to address' , \
                   	msg.as_string(unixfrom=0))

The thing died when passed a piece of e-mail spam, producing the
following trailback:

===================================================================
  File "/home/public/nlh/nlhapp/utils/message/postman.py", line 49, in
processmail
    msg.as_string(unixfrom=0))
  File "/usr/local/lib/python2.2/email/Message.py", line 59, in
as_string
    g(self, unixfrom=unixfrom)
  File "/usr/local/lib/python2.2/email/Generator.py", line 83, in
__call__
    self._write(msg)
  File "/usr/local/lib/python2.2/email/Generator.py", line 104, in
_write
    self._dispatch(msg)
  File "/usr/local/lib/python2.2/email/Generator.py", line 134, in
_dispatch
    meth(msg)
  File "/usr/local/lib/python2.2/email/Generator.py", line 240, in
_handle_multipart
    for part in msg.get_payload():
  File "/usr/local/lib/python2.2/email/Message.py", line 151, in
__getitem__
    return self.get(name)
  File "/usr/local/lib/python2.2/email/Message.py", line 214, in get
    name = name.lower()
AttributeError: 'int' object has no attribute 'lower'
===================================================================

I see nothing in the spam that could have caused this (integer header
identifier presumably), but have no doubt that there is something
nasty in there. On the other hand it's a bit worrying that the e-mail
module lets itself be knocked out by this. Or is it? Comments?

pm


From paul at boddie.net  Mon Feb  4 07:26:36 2002
From: paul at boddie.net (Paul Boddie)
Date: 4 Feb 2002 04:26:36 -0800
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org>
Message-ID: <23891c90.0202040426.7b13e9b9@posting.google.com>

Gerard Braad <g_braad at survion.com> wrote in message news:<mailman.1012751045.16555.python-list at python.org>...
> 
> For a long time now, I had the idea of XML-Formed Python. This way it
> would be possible to send Python code across a network connection. The
> only problem was, I didn't had much time to realise this. Way back in
> August I saw someone (mlh) had the same idea and in December the project
> page was included on the DDJ Python-URL. JEEJ!

Yes, I like to look out for forgotten projects and include them in the
list from time to time. An exercise for the reader is to discover
which other Python-URL!-mentioned package is related so significantly
to Xython that you might want to take a closer look.

Paul


From idbaxter at semdesigns.com  Tue Feb 26 15:36:42 2002
From: idbaxter at semdesigns.com (Ira D. Baxter)
Date: Tue, 26 Feb 2002 14:36:42 -0600
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com> <a5gk1u$hlu$1@bob.news.rcn.net>
Message-ID: <3c7befb8$1@giga.realtime.net>

"David Abrahams" <david.abrahams at rcn.com> wrote in message
news:a5gk1u$hlu$1 at bob.news.rcn.net...
>
> "Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
> news:6bf54244.0202260920.5920efc6 at posting.google.com...
> > Hi David.
> >
> > ... I would presume that any compiler (eg g++) would
> > create some internal representation, ...

> > Presumably the EDG front end is geared toward producing this information
> > in a format more suited to being used by the "wrapper generators" in
> > contrast with a compiler which produces this information only for itself
and with
> > the sole goal of compiling an object file?
>
> Not exactly. EDG makes a front-ends which they sell to vendors of full
> compiler solutions (Intel, KAI, Comeau, Compaq, ...). As such, their
> requirements for a well-defined and accessible intermediate representation
> (IR) are higher than other front-end writers.
>
> > What I'm getting at is that I'm confused as to why people wouldn't have
> > used information from the compiler *itself* to automate the process of
> > wrapper generation.
>
> They did. EDG's front end is part of many of the best C++ compilers.
>
> > Obviously the compiler writer people would have to
> > collaborate with the wrapper generator people...
> > but there must be  technical obstacles?
>
> What's in it for the compiler writer people?

Full employment? More importantly for EDG anyway, a broader market base.

There are other ways to tackle this problem.
One could build generic compiler front ends capable
of collecting all this data in an "accessible representation"
that isn't biased towards just a particular compiler,
and then populate a space of front-ends using
the generic infrastructure.  The value in
such a scheme is that you can use it for all
kinds of analysis and/or massive automated
change activitiies. of which wrappers is just
a particular application.

This is exactly what we've done with our DMS Software Reengineering Toolkit.
See http://www.semdesigns.com/Products/DMS/DMSToolkit.html.
That link also points to a wide variety of possible applications.

One of the issues to face is the problem of parsing nasty langauges
like C++, and deciding just exactly what you want from the "parse".
The EDG front end (I don't know a lot about it) is a "traditional"
compiler front end.   My unconfirmed guess is that for C++
it does preprocessing, and then builds intermediate representations
of the preprocessed text.

Now, if what you want to do is compiler-like, then throwing
away the comments, line numbering, number formats,
and preprocessor instructions is just fine.  And you
generally only have to focus on one "file" at time.

If, however, you want to modify the sources of the program,
*retaining* all that information, then you can't use
a conventional compiler technology front end.
You need to have something that can parse the sources,
*including* the preprocessor directives.
One of DMS's unique properties is its ability
to do that (with some constraints) for C and C++
code.

A second unique property is the ability to parse
entire suites of files in the same session, so
that you can build analysis tools that operate
across file boundaries (i.e., multiple compilation units)
and make modifications that move code/data
from one source file to another to accomplish
refactoring activities.

A third unique property of such generalized front ends
is the ability to *mix* notations from multiple programming
languages in patterns to enable translation from one langague/dialect
to another.

Of course, our hope is to be fully employed building/selling
reengineering tools rather than compilers.

--
Ira D. Baxter, Ph.D. CTO Semantic Designs, Inc.
http://www.semdesigns.com




From mwh at python.net  Thu Feb  7 08:25:54 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 13:25:54 GMT
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <Xns91AE6BE28FCD8michaelrcpcouk@194.238.50.13> <uk7tpa574.fsf@python.net> <Xns91AE86CF4427Cmichaelrcpcouk@194.238.50.13>
Message-ID: <uy9i58lqi.fsf@python.net>

Michael Abbott <michael at rcp.co.uk> writes:

> Michael Hudson <mwh at python.net> wrote in news:uk7tpa574.fsf at python.net:
> 
> > Michael Abbott <michael at rcp.co.uk> writes:
> > 
> >> I reckon to pay a cost of perhaps a factor of 100 in raw CPU for the 
> >> privilege of using Python. 
> > 
> > But not over Java, surely?  I guess a JIT or native compilation will
> > help, but...
> 
> I think so.  Java byte codes are pretty low level, and don't have to do 
> very much, and of course JIT or native compilation can be *very* effective.  
> Python byte codes are, on the other hand, much more powerful, and have to 
> do quite a lot of run-time work.

Fair enough.  Don't know much about the Java world.

> If you're doubtful, take a quick peek at what the Python byte code 
> interpreter does to implement a method call (I'm thinking of call_object in 
> ceval.c).  There's nothing wrong with this (that's not my point), but 
> there's quite a lot of work to be done (including memory allocation, I'm 
> afraid).

I do know quite a lot about the Python world though :)

> This is one of the most widely used operations in the language (excepting, 
> perhaps, method lookup), and can be reduced in a natively compiled language 
> to a small handful of instructions or even eliminated (when we're lucky).
> 
> My factor of a 100 is a guess; am I wrong?  I believe it's fair.

In my experience, it's a bit of an over-estimate.  Depends what you're
doing, of course.  I find I tend to optimize the straight line bits of
code, which probably doesn't yield the greatest improvement, in some
sense, but not writing the not-straight-line bits of code in C is the
reason I use Python in the first place...

Cheers,
M.

-- 
  Every day I send overnight packages filled with rabid weasels to
  people who use frames for no good reason.
                             -- The Usenet Oracle, Oracularity #1017-1


From wurmy at earthlink.net  Mon Feb 25 12:37:34 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Mon, 25 Feb 2002 17:37:34 GMT
Subject: Iterators and generators for the masses
References: <5e8bd451.0202250924.70f47df1@posting.google.com>
Message-ID: <3C7A7699.EC239F65@earthlink.net>

kevin parks wrote:

> A note in my Python 2.2 folder says that:
> 
> Iterators (PEP 234) and generators (PEP 255) were added.

[snip]

> what do they do? what can they do, what tasks do they make easier,
> what is the simplest possible example, what is a more complex
> example...etc. & co.

"What's new in Python 2.2" describes them, maybe this will help
you:

  http://www.amk.ca/python/2.2/index.html#SECTION000400000000000000000

HTH,

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From roy at panix.com  Wed Feb 20 23:11:08 2002
From: roy at panix.com (Roy Smith)
Date: Wed, 20 Feb 2002 23:11:08 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com>
Message-ID: <roy-630F36.23110820022002@reader1.panix.com>

Erik Max Francis <max at alcyone.com> wrote:
> Many languages (other than Scheme) have a loose definition of truth
> and falsity, simply because at times it's useful to be able to test
> any object for truth.

If you want a really odd concept of falsity, look at SQL.  NULL is neither 
true nor false!  Or, looking at it a different way, anything non-NULL is 
neither equal to, nor not equal to NULL.

This actually causes some strange interactions in the python SQL modules.  
You put '' into a database, and get None back.  They are both false, but 
they're not the same thing.

Anybody remember .TRUE. and .FALSE. :-)

PS -- In the lists of false things in python that have been mentioned on 
this thread, I don't think I've seen anybody mention {}, the empty 
dictionary.


From syring at email.com  Sun Feb  3 08:41:39 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 14:41:39 +0100
Subject: HELP?
References: <a3jebb$cg8$1@knossos.btinternet.com>
Message-ID: <a3jf2o$18l85a$1@ID-7529.news.dfncis.de>

"Callum Golding" <Rave_Dj at btopenworld.com> schrieb
> I'm sorry to bother you, but i was wondering if you could give me a little
> advice. I've recently downloaded Python 2.2. Ive read a begginers guide to
> programming and tried creating a small program, I used the input function.
> Now as soon as I open the program it goes into dos and asks me the
question,
> and when i type in an anwser, dos shuts down. Now its really starting to
bug
> me. If you could give me some advice I would be real grateful. The program
I
> wrote is below; -
>
>
> tempetature = input("What is the temperature of the spam?")
>
> if temperature > 50:
>     print "The salad is properly cooked"
> else:
>     print "Cook the salad some more"

Spelling error: "tempetature" is distinct from "temperature". And you should
not cook the salad.

Karl M. Syring




From syring at email.com  Thu Feb 21 04:52:19 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 10:52:19 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com>
Message-ID: <a52g60$4630o$2@ID-7529.news.dfncis.de>

"Paul Rubin" <phr-n2002a at nightsong.com> schrieb im Newsbeitrag
news:7x8z9njjra.fsf at ruckus.brouhaha.com...
> "Karl M. Syring" <syring at email.com> writes:
> > > Physical simulations and computational geometry don't have much in
common.
> >
> > Think of simulating a robot arm that does a brain operation on you. I
think
> > you would immediately turn to proper rational number coordinates.
>
> Not likely.

You really should have a look at high precison libraries (e.g.
http://www.hipilib.de). They will use rational numbers internally.

Karl M. Syring



From dinu at reportlab.com  Wed Feb 13 11:01:51 2002
From: dinu at reportlab.com (Dinu C. Gherman)
Date: Wed, 13 Feb 2002 17:01:51 +0100
Subject: Proceedings CD-ROM of IPC10?
Message-ID: <dinu-781536.17015113022002@news.t-online.com>

Hi,

I'm not sure if Foretec plans to do this or not, but
is there already a CD-ROM with Proceedings for IPC10?
I vagualy remember having seen something similar for 
IPC9, or so...

Thanks,

Dinu


From jmsun at bioeng.ucsd.edu  Thu Feb 14 20:18:16 2002
From: jmsun at bioeng.ucsd.edu (Jeffrey)
Date: 14 Feb 2002 17:18:16 -0800
Subject: Tkinter and Greek Symbols
Message-ID: <3daa1aa8.0202141718.3500c462@posting.google.com>

Hi Everyone,
  You can probably guess what I want to know from the title.  But hey
I'll explain anyways :)  I want to be able to display Greek Symbols
like alpha and theta.

     Thanks in Advance!!!
                      Jeff


From michael at rcp.co.uk  Fri Feb  1 04:11:36 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Fri, 1 Feb 2002 09:11:36 +0000 (UTC)
Subject: An improvement to Queue
References: <Xns91A793581DD79michaelrcpcouk@194.238.50.13> <a3blop$a29$1@serv1.iunet.it> <Xns91A7A14C232EBmichaelrcpcouk@194.238.50.13> <a3bqsu$bab$1@serv1.iunet.it> <Xns91A7AEE05E65Fmichaelrcpcouk@194.238.50.13> <3C59C1AA.58A02B4E@ccvcorp.com>
Message-ID: <Xns91A85D8586450michaelrcpcouk@194.238.50.13>

Jeff Shannon <jeff at ccvcorp.com> wrote in
news:3C59C1AA.58A02B4E at ccvcorp.com: 

> Michael Abbott wrote:
> 
>> It's the transition from empty to non-empty that's important: ....
> 
> What's the point of putting an incomplete command event into the queue
> at all? 

I'm not putting incomplete commands in the queue.  However, the consumer is 
in a position to merge multiple commands together, and I'd like it to have 
the option.

Also, it is in general more efficient to perform jobs in batches, which is 
what I'm trying to do here.


From timr at probo.com  Mon Feb 25 02:15:48 2002
From: timr at probo.com (Tim Roberts)
Date: Sun, 24 Feb 2002 23:15:48 -0800
Subject: Parsing date strings into tuples
References: <3C74EC73.6030204@olen.to>
Message-ID: <gooj7usb6ses69rbgm8i5uq6uie5e8tl28@4ax.com>

Joonas Paalasmaa <joonas at olen.to> wrote:
>
>Question 2:
>Why doesn't this work?
>
> >>> time.strptime('Wed, 20 Feb 2002 09:25:29 EET',
>     "%a, %d %b %Y %H:%M:%S %Z")
>Traceback (most recent call last):
>   File "<pyshell#5>", line 1, in ?
>     time.strptime('Wed, 20 Feb 2002 09:25:29 EET', "%a, %d %b %Y 
>%H:%M:%S %Z")
>ValueError: unconverted data remains: 'EET'

Because your C library strptime call does not recognize EET as a time zone.
EET is also not acknowledged in RFC [2]822.  You get the American zones,
/[PMCE][SD]T/, and the recommended numeric zones, like -0700 and +0530, but
not any arbitrary zone.

Also note that time.strptime is not available on Windows systems at all.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From aahz at panix.com  Thu Feb 28 22:57:53 2002
From: aahz at panix.com (Aahz Maruch)
Date: 28 Feb 2002 19:57:53 -0800
Subject: Simple threads (was Re: listing threads?)
References: <mailman.1014930527.20065.python-list@python.org> <a5mmoj$2ji$1@panix2.panix.com> <uadttot9v.fsf@ctwd0143.fitlinxx.com>
Message-ID: <a5mu81$qom$1@panix2.panix.com>

In article <uadttot9v.fsf at ctwd0143.fitlinxx.com>,
David Bolen  <db3l at fitlinxx.com> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>>
>> DON'T DO THAT.  I'm serious.  You really do NOT want to use
>> threading.enumerate(); you'll only confuse yourself.  I've pointed you
>> at my web page on threading; is there some reason you haven't read it,
>> or is there something you don't understand?
>
>I wouldn't go all that far.  Clearly you have to understand the nature
>of the snapshot that threading.enumerate() returns with respect to the
>asychronous nature of threads (which of course is what was confusing
>the OP, since the example had a thread that barely got off the ground
>before exiting).
>
>But I don't see that enumerate() is in and of itself a bad method,
>when used properly - certainly not something to warrant such a
>sweeping statement.  We certainly use it in some of our applications
>as a passive way to get a snapshot of thread activity and it works
>fine.

I am generally more interested in "practicality over purity".  And the
fact remains that 99% of the time that someone wants enumerate(), they
should have been storing references to thread objects in a list or dict;
that way, they're guaranteed to not make a mistake in trying to perform
some operation on the main thread.

I advocate against enumerate() for the same reasons I advocate against
Lock(), Condition(), Event(), and Semaphore().  (I strongly recommend
sticking with RLock() and Queue().)  Writing correct threaded
applications is difficult enough -- KISS, KISS, KISS.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"How come we choose from just two people for President but more than 50
for Miss America?"  --AEN


From donn at u.washington.edu  Tue Feb 12 15:45:25 2002
From: donn at u.washington.edu (Donn Cave)
Date: 12 Feb 2002 20:45:25 GMT
Subject: catching error output
References: <a4br6c$lt1$1@news-int.gatech.edu>
Message-ID: <a4but5$2a8m$1@nntp6.u.washington.edu>

Quoth gte743n at cad.gatech.edu (Joseph Holland King):
| in a small script i am using the popen command to run a command 
| and then parse the output:
| fs = os.popen('foo some_data')
| fs.readline()
| ...
| however when foo fails i need to catch that error and be able to parse
| it. is there anyway to do this? fs = .. does not catch it just the 
| normal output. thanks.

That sounds about like popen2.popen3().  You can close the input pipe
it also returns.

Might be safer to use the Python class that popen3() creates, though.

>>> import popen2
>>> c = popen2.Popen3('echo hi; echo ouch >&2; exit 1', 1)
>>> c.poll()
256                            # this is exit status 1 * 256
>>> c.childerr.readline()      # OK because process has exited
'ouch\012'

	Donn Cave, donn at u.washington.edu


From frederic.giacometti at arakne.com  Sat Feb 23 12:16:24 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Sat, 23 Feb 2002 17:16:24 GMT
Subject: Stackless Platform Independence?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com>
Message-ID: <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com>

"Courageous" <jkraska at san.rr.com> wrote in message
news:987e7uoo5v87fsjruq0ug6bff0vha5ear0 at 4ax.com...
>
> >I am personally still surprised that stackless has not be considered more
> >seriously in the python-dev group, especially in regard to the threading
> >problems and limitations of the current python versions.
>
> Well, keep in mind that stackless threads aren't OS threads at all.

Yes, this is very much the point, and Python os thread furthermore require
global interpreter lock operations. There's no free lunch there; and some
significant limitations too.
In addition to all the benefits that Christian has described so far,
stackless python would make it possible to have a different OS
multi-threading implementation for calling external C/C++ libraries; one
that does not require interpreter lock operations.
However, stackless is a pre-requisite for such things.


> The way I have it, Stackless _was_ considered by Guido and team,
> it was just regarded as both too dirivative and too complicated to
> make part of core Python. Look out in the real world and try to
> find out how many languages have continuation support in them.
> Not very many.

>From what I have perceived, there has been confusion in the debate, between
stackless and continuations.
Most of the remarks I've been aware of are against pure continuations
(basically, arbitrary use of continuations, like C goto, would just break
program flow structure, with all the consequences one can imagine).
Complexity problems are also regarding continuations.

I'm not too familiar with other scripting language implementations, but Java
and Perl, at least, I think, are stackless!!

The concept of stackless is a lot simpler, and its implementation has no
particular technical obstacles; especially since Christian made the proof of
concept.

Stackless just means that function calls executed by the bytecode are
executed by a C function that returns immediately; thus disconnecting the C
call stack from the virtual machine call stack.
In current python, they are interleaved one-on-one; making it impossible to
switch context within the VM unnless you do Christian's last 'non-portable'
assembly language manipulation.

Christian's first implementation was at the C source level, and wasn't that
complicated to integrate; its implementation had to be reengineered, but
there was no out-of-this-world wizardry there.

Provided a proper source organization, stackless implementation would not
add much complexity to the python VM (one additional data structure,
basically). However, it's true that, because of source management issues
(Christian did not create additional files), Christian implementation could
not be integrated as such.

> I personally make great use of continuations in a simulator core
> I wrote. While I grok them, I have the feeling that an average
> programmer following in my footsteps would have to read my
> comments very very carefully before he could come to understand
> how a certain central function call in my simulator actually
> returns millions of times.

Once a stackless implementation of the Python VM is provided, continuations
can always be proposed as optional module, external to the python standard
distribution.
Continuations is one item of what stackless would enable.
Python microthreads would certainly be the most significant outcome, but
there will be others too.
The debate on continuations should not be mixed with stackless; and very
much valid arguments against continuations should not be used to argue
against a stackless Python VM implementation.

Frederic Giacometti







From sameer_ at email.com  Mon Feb 18 10:03:59 2002
From: sameer_ at email.com (sameer)
Date: 18 Feb 2002 07:03:59 -0800
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <a4ps2l$5jo$1@peabody.colorado.edu>
Message-ID: <6bd9f01b.0202180703.19f2eee@posting.google.com>

this doesn't look like an installer.

Fernando P?rez <fperez528 at yahoo.com> wrote in message news:<a4ps2l$5jo$1 at peabody.colorado.edu>...
> sameer wrote:
> 
> > sameer_ at email.com (sameer) wrote in message
> > news:<6bd9f01b.0202160255.77244c9e at posting.google.com>...
> >> Here's the situation:
> >> I have a python package (or two) to be installed on a windows machine
> >> (for now) which has python installed.  I am using distutils to create
> >> a windows installer which installs my python packages just fine.  Now,
> >> I want to also create some shortcuts on the Desktop as well as the
> >> Start Menu.  As of now, distutils doesn't allow me to do either of
> >> those things.  What are my options here to do either or both of these
> >> things.  Can the distutils package be modified easily to provide this
> >> kind of support on windows?  Or is there another python package
> >> available that would allow me to create the desktop and Start Menu
> >> shortcuts?
> > 
> > it's a pity no one has replied.  I suppose there are no sane options for
> > this.
> 
> Sorry, I hadn't seen this.
> 
> Go to http://www-hep.colorado.edu/~fperez/ipython and download the program. 
> 
> You'll find besides the standard setup.py a setup_Windows.py which does 
> exactly what you're looking for. Feel free to use it/modify it. 
> Changes/fixes/suggestions welcome.
> 
> Cheers,
> 
> f.


From jason at jorendorff.com  Fri Feb  1 00:39:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 31 Jan 2002 23:39:28 -0600
Subject: Fun with lambda and map
In-Reply-To: <a3b6e1$5m8$1@news1.xs4all.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLIEJKCPAA.jason@jorendorff.com>

Ivo wrote:
> I implemented some of the pure (?) lambda calculus in python:
> http://vanderwijk.info/lambda
> Who needs '+' , 'if' and numbers? :)

Wow.  It's extra cool that this should come up, because I just
recently tore through the first chapter of SICP, and wrote this:

  http://www.jorendorff.com/toys/truth.html

## Jason "mul = lambda m,n: lambda f: m(n(f))" Orendorff



From nordwick at scam.xcf.berkeley.edu  Thu Feb 21 15:10:06 2002
From: nordwick at scam.xcf.berkeley.edu (Jason Nordwick)
Date: 21 Feb 2002 12:10:06 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <cb4ba455.0202210525.6687556f@posting.google.com>
Message-ID: <kpd6yybnq9.fsf@scam.xcf.berkeley.edu>


sorry, but I had to test posting a follow-up article, please ignore.


From justin at iago.org  Fri Feb 22 17:20:54 2002
From: justin at iago.org (Justin Sheehy)
Date: Fri, 22 Feb 2002 17:20:54 -0500
Subject: Python and Daemons
In-Reply-To: <02022215115601.01886@localhost.localdomain> (Dieter Lunn's
 message of "Fri, 22 Feb 2002 15:11:56 -0600")
References: <02022215115601.01886@localhost.localdomain>
Message-ID: <x74rk917ll.fsf@ra.iago.org>

Dieter Lunn <theguru at mb.sympatico.ca> writes:

> I was wondering how to write a linux daemon using Python.

This is what I use:

class Log_out:
    '''file-like dummy device, used for daemonization
    writes are redirected into our logging object'''
    def write(self, s):
        debugmsg(s)

def daemonize():
    "cause this process to becaome a daemon"
    if os.fork() != 0:
        os._exit(0)
    os.setsid()
    if os.fork() != 0:      # 2nd fork is widely recommended
        os._exit(0)         # though not always needed
    debugmsg("daemonized")
    os.chdir("/")   # not strictly necessary, but a good idea
    os.umask(0)
    sys.stdin.close()
    sys.stdout = Log_out()
    sys.stderr = Log_out()

debugmsg is a local function which calls a function in my logging library.

-Justin

 




From magnus at thinkware.se  Thu Feb 21 06:43:33 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 12:43:33 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>
Message-ID: <3C74DD65.3060000@thinkware.se>

Eduard Smirnov wrote:

 > I'm writing a diploma paper on the difference between scripting
 > (e.g. in Python or Tcl) and system programming (e.g. in C, C++ or
 > Java), and one of the claims I want to make is that you can often
 > develop applications 10x or more faster in a scripting language.

I wouldn't really say that Python is a scripting language though. The
name scripting language suggests a language suited for small snippets.
Python is as much a general language for large scale software
development as C++ is. But Python is surely a rapid development
language.

Bash, awk, Tcl and Perl are in my opinion scripting languages, but
not Python. (Well, Perl probably has the theoretical requirements for
developing large systems, but it's fairly obscure and bolted on as an
afterthought.)

 > I apologue that more applications can be written or re-written in
 > scripting language faster and with more complex functionality then
 > in system programming language. If you have any experience in
 > writing applications in system programming language (e.g. in C,
 > C++) and re-writing it in scripting language (e.g. Python) or vice
 > versa, please compare your work in both implementations,

Actually, I think many comparisions are made like this, and a problem
is that if you first write in C, and then translate, you are likely to
use your C design for Python, and end up with a program that is almost
as big, but much slower. If you go the other way, there is no way you
can use your Python solution directly, since C can't handle the level
of abstraction, and you will probably get a fairer comparision.

 > describe your work/application's functionality, approximate number
 > of lines and time of development of each implementation, skill's
 > level needed. Any more comments about are welcome.

Personally, my Python applications worked well, thank you, so I
never rewrote them in any other language...

There are some links at
http://www.thinkware.se/cgi-bin/thinki.cgi/PythonQuotes
which might guide you do relevant sources. (Glyph Lefkowitz,
Greg Stein and Eric S. Raymond for instance.)

Also, have a look at
http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprt_computer2000.pdf
For this experiment, the productivity gain with Python is roughly
factor three, but this is also

There is a comparision at http://www.bagley.org/~doug/shootout/
but it's really just a pointless play. Don't fall for this one.
It has two fundamental flaws: It only uses small programs, so
it doesn't show you anything about how the language scales. If you
believed in that, do all your coding in C and bash. It also ignores
one of greatest assets of Python, its rich standard library.

So instead of writing the "random generator" as

import random
print random.random()

They reinvent the wheel like this:

import sys

IM = 139968
IA = 3877
IC = 29573

LAST = 42
def gen_random(max):
     global LAST
     LAST = (LAST * IA + IC) % IM
     return( (max * LAST) / IM )

def main():
     N = int(sys.argv[1])
     if N < 1:
         N = 1
     gr = gen_random
     for i in xrange(1,N):
         gr(100.0)
     print "%.9f" % gr(100.0)

main()

In other words, many of the programs are about as long
as C programs since they "level the playfield" by not
using the "included batteries". On the other hand Python
always show more CPU and memory usage since it loads the
interpreter and a lot of modules before it runs the those
twenty lines of code. No wonder that C seems slim and fast
there. If we talked about a REAL application, and compared
50 000 lines of C++ with 10 000 lines of Python code, things
would look very different.







From eppstein at ics.uci.edu  Sun Feb 10 17:43:07 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Sun, 10 Feb 2002 14:43:07 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net>
Message-ID: <eppstein-A87A6C.14430710022002@news.service.uci.edu>

In article <eSz98.4759$zb.260473 at news1.east.cox.net>,
 "Jeff Hinrichs" <jlh at cox.net> wrote:

> Pythonista - Pie thun ee sta
> I took it as a unintended play on Sandanista, which is a group of  "freedom
> fighting" rebels in Nicaragua.

I think it's a normal Spanish suffix, but it seems to be moving into 
English -- I've also seen "Gapatista" used by some Calif. activists
<http://www.ics.uci.edu/~eppstein/pix/sv/j4p/Gapatista.html>
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From amuys at shortech.com.au  Wed Feb 20 20:00:40 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 20 Feb 2002 17:00:40 -0800
Subject: Adding GUI to application
References: <9120a667.0202200814.1bd679cb@posting.google.com>
Message-ID: <7934d084.0202201700.3eae4d5f@posting.google.com>

mgash at trisystems.co.uk (Mark Gash) wrote in message news:<9120a667.0202200814.1bd679cb at posting.google.com>...
> I wish to amend the application so that (i) it runs as before, (ii)
> there is an icon in the taskbar, via which the application can be
> stopped.
> 
> As a newbie to programming, I am finding this task difficult, to say
> the least. The problem that I am having, is that by using the
> win32gui_taksbar.py from the win32/demos module, I can get the icon
> down in the taskbar - but as the diskchecking application 'goes to
> sleep' rather than completes/exits, the control is never returned to
> the win32gui_taskbar to await the next interupt.
> 
Mark,
   While I've never written anything with Win32 before (regardless of
language), I've written applications like this numerous times under
unix.  The standard solution I tend to use (which I in turn borrowed
off numerous other applications ;) is to store the process id in a
file/known location (maybe the registry? or is this a bad idea under
windows?) and to write a second program that reads the pid, and sends
the appropriate signal to the background process to terminate it
(under unix generally a SIGHUP [hangup signal]).  I believe under NT
daemons are called "services", there should be a programatic way of
terminating them from another process.  I don't know about windowsXXX.

Andrae Muys


From printers at sendme.cz  Mon Feb 25 10:51:45 2002
From: printers at sendme.cz (A)
Date: Mon, 25 Feb 2002 16:51:45 +0100
Subject: Is it possible in Python too?
Message-ID: <3C7A6BA1.26999.19EEC27@localhost>

Hi,
Is possible to set up a proxy name (address) in MS Explorer 
browser from a running  Python program?
Thank you for your help.
Ladislav






From fredrik at pythonware.com  Mon Feb 11 05:09:16 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 10:09:16 GMT
Subject: No explanation for weird behavior in re module!
References: <a470re$1cl30l$1@ID-78052.news.dfncis.de> <c76ff6fc.0202101814.5d18209d@posting.google.com> <87pu3c7osj.fsf@uol.com.br>
Message-ID: <gLM98.25033$l93.4622976@newsb.telia.net>

"synthespian" wrote:

> As I understood it from the other posts, the "\w+" on
> the regex will depend on my locales.

not necessarily.  I suggest reading the RE documentation
again; look for the ?L and ?u flags.  if neither flag is present,
the engine assumes plain ASCII.

> Seems like non-ASCII is a real bother in Python...

it isn't.  you just have to know how things work.

1) text files contain *encoded* data.  each character in the
text is encoded as one or more bytes.

2) if you read a line of text from a file, you get an encoded
string.

3) to decode an encoded string into a string of well-defined
characters, you have to know what encoding it uses.

4) to decode a string, use the decode method on the input
string, and pass it the name of the encoding:

    fileencoding = "iso-8859-1"

    raw = file.readline()
    str = raw.decode(fileencoding)

the result is a unicode string.

5) to create a regular expression pattern that uses Unicode
character classes for \w, use the "(?u)" prefix, or the re.UNICODE
flag:

    pattern = re.compile("(?u)pattern")
    pattern = re.compile("pattern", re.UNICODE)

that's it.

6) to print a unicode string to your output device, you have to
convert it to the encoding used by your terminal.

    import locale
    language, output_encoding = locale.getdefaultlocale()

    print str.encode(output_encoding)

(there are lots of shortcuts, including coded streams, using default
locales for pattern matching, iso-8859-1 as a subset of unicode, etc,
but that's outside the scope of this post).

</F>




From daniel.dittmar at sap.com  Fri Feb  8 10:07:53 2002
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Fri, 8 Feb 2002 16:07:53 +0100
Subject: readlines() a file in 2 parts.
References: <87vgd85a7h.fsf@cosette.nampaz.com>
Message-ID: <a40pka$nol$1@news1.wdf.sap-ag.de>

> I have several files to read that logically break into 2 parts.  It would
> be much cleaner to write 2 different 'for line in file.readline() :'
loops,
> but the break point is different in each file, so I don't know the
starting

import xreadlines

class TwoStepReadline:
    def __init__ (self, stream):
        self.iterator = xreadlines.xreadlines (stream)

    def __iter__ (self):
        return self

    def next (self):
        line = self.iterator.next ()
        if self.condition (line):
            self.next = self.iterator.next   # just a speed hack to remove
the test
            raise StopIteration
        else:
            return line

    def condition (self, line):
        must be implemented


iterator = TwoStepReadline (sys.stdin)
for line in iterator:
    process part 1

for line in iterator:
    process part2

You'll have to change this a bit depending on whether the triggering line
should be part of the first iteration, the second iteration or ignored
entirely (as in the implementation above).

Of course there are variations which might be more flexible:

iterator = xreadlines (sys.stdin)
for line in ReadUntil (iterator, re.compile ('some pattern')):
    process part 1

for line in iterator:
    process rermainder

Daniel





From fperez528 at yahoo.com  Mon Feb 11 00:35:39 2002
From: fperez528 at yahoo.com (Fernando Perez)
Date: Sun, 10 Feb 2002 22:35:39 -0700
Subject: Using pydoc
References: <a46upn$u92$2@gonzo.circuit.com>
Message-ID: <a47lat$rfj$1@peabody.colorado.edu>

> Calls seem to be made like:
>
>         drvobj = partition.Partition(drv)
>         drvobj.create_partition_table()
>         drvobj.commit_changes()
>
> My question is:  How do I find documentation for those things?
>
>     ~$ pydoc partition.Partition
>         no Python documentation found for 'partition.Partition'

<blatant plug for my pet project follows>

In [1]: import os

In [2]: os.path?
Type:           module
Base Class:     <type 'module'>
String Form:    <module 'ntpath' from 'C:\Program
Files\Python22\lib\ntpath.pyc'
>
Namespace:      Currently not defined in user session.
File:           c:\program files\python22\lib\ntpath.py
Docstring:
    Common pathname manipulations, WindowsNT/95 version.

    Instead of importing this module directly, import os and refer to this
    module as os.path.

The above is a copy/paste of an IPython session. It's actually a Linux
project originally, but right now I happen to be cleaning up a Windows port.
With object? IPython will give you all sorts of details about any object it
can access (uses pydoc internally).

You can get it at

http://www-hep.colorado.edu/~fperez/ipython/

</blatant plug>

Cheers,

f.





From pedronis at bluewin.ch  Wed Feb 27 20:47:35 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Thu, 28 Feb 2002 02:47:35 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org>
Message-ID: <3c7d8f26_3@news.bluewin.ch>

Maybe this is a useful insight:

in mathematetics it is not uncommon to
identify the natural numbers with the set
of their predecessors

0 = {}
1 = {0} = {{}}
2 = {0, 1}  {{},{{}}}
etc

e.g. in the notation for power-sets

powerset(X) = 2^X  where 2 stands for
the set {0,1}

so having
  for i in 5
iterating over 0,1,2,3,4
is natural in mathematical sense
(which does not mean automatically that it is newbie friendly <wink>)

because 5 as an ordered set is the sequence (0,1,2,3,4)

so also the notation   1 in 5  is natural.

regards, Samuele Pedroni.




From jason at jorendorff.com  Tue Feb 12 12:09:05 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 12 Feb 2002 11:09:05 -0600
Subject: searching backwards in a string
In-Reply-To: <Jlaa8.174222$KM2.7680225@atlpnn01.usenetserver.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAEBGDCAA.jason@jorendorff.com>

Steve Holden wrote:
> If you'd like to find the "last match before a given position" then,
> supposing string p to represent the regex matching the pattern you are
> looking for, you can try match()'ing
> 
>     ".*(%s)" % p
> 
> Given the greedy nature of matching, this should give you the rightmost
> match in groups(1). Maybe. Anyway, an idea to play with.

Hey, that's pretty clever!  I wonder about the performance, though...
Anyway, remember that by default .* won't cross newlines.
So: r"(.|\n)*(%s)".  Or you could re.DOTALL, but that may change
the meaning of the other regular expression you're plugging in.

## Jason Orendorff    http://www.jorendorff.com/



From robin at jessikat.fsnet.co.uk  Wed Feb 27 18:58:04 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 27 Feb 2002 23:58:04 +0000
Subject: PIL vs eps
References: <HRCwzIADVVf8EwB0@jessikat.fsnet.co.uk>
 <mailman.1014849261.21729.python-list@python.org>
Message-ID: <YjxXPIAMKXf8EwAE@jessikat.fsnet.co.uk>

.
.....
>I don't know what kind of constraints you're dealing with, but the correct
>way to handle an eps file (i.e. make it display) would be to actually
>embed it in what is essentially a dummy .ps file, e.g. take the eps data
>and surround it with the minimal ps header and footer necessary and then
>process that instead of the original. (The problem you're having is rooted
>in the fact that you're trying to treat an eps file like a ps file, so the
>solution is to use the eps to make a real ps file).

well I didn't actually expect anything, someone else was using PIL to
import images into ReportLab. On the other hand given that the PIL
module concerned is called EPSImagePlugin.py I would somehow have
expected it to cope with eps files. Also inside there is an explicit
class called EpsImageFile which I reckon is supposed to handle them.

Unfortunately I can't seem to get a nox11 ghostscript to build on
freeBSD 4.5 so I am guessing here. Certainly the error is reported to
occur exactly where the .ppm output would be read back.
-- 
Robin Becker


From roy at panix.com  Sun Feb 10 17:44:12 2002
From: roy at panix.com (Roy Smith)
Date: Sun, 10 Feb 2002 17:44:12 -0500
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com> <7xlme1hv0y.fsf@ruckus.brouhaha.com> <3C66EA99.B22478E3@alcyone.com>
Message-ID: <roy-F1119D.17441210022002@news1.panix.com>

Erik Max Francis <max at alcyone.com> wrote:
> That is simply not comparable.  The purpose of split is to divide up a
> string according to a delimiter.  Asking for it to split a string given
> a null delimiter doesn't make much sense at all;

I think it makes plenty of sense.

> if you're calling
> string.split with the intent of dividing it up and you're passing in ''
> as a delimiter, that's almost certainly a mistake.

If the delimter was written as a constant '', I might be inclined to agree, 
but it could be computed on the fly, and having a delimter of '' mean 
"split the thing up with nothing between the parts" is a very logical 
extension of the general idea.

Should we outlaw raising things to the zero power because somebody thinks 
it "doesn't make much sense at all"?  I'll be a lot of people think it 
doesn't make much sense, but the mathematicians would howl if you tried to 
take it away.

In everyday usage, raising something to the zero power, I'll agree, doesn't 
make a lot of sense.  But, the math folks figured out a reasonable, 
consistant, and logical meaning for it, by extrapolating from other 
exponentiation and seeing what would fit into a logical whole.  The same 
thing should be done with split().  Having split('') mean "split every 
character" is the logical extension of its other uses.

None of the following are illegal..

for i in []:
   print "can't get here"

if ('a' in ''):
   print "can't get here either"

while (0):
   print "what do you think?"

class theNullClass:
   pass

Not to mention a zillion variations on identity operations like adding zero 
or multiplying by one.

You could make rational arguments that none of them "make much sense at 
all", and that they are all likely to be errors, but you'd get arguments 
that to outlaw them would destroy the orthogonality of the language.

But I think the strongest argument in favor of allowing '' as an argument 
to split() is the join() symmetry argument.


From paul1brian at yahoo.com  Tue Feb 26 10:09:07 2002
From: paul1brian at yahoo.com (Paul Brian)
Date: Tue, 26 Feb 2002 15:09:07 -0000
Subject: clearing down all instances in python
Message-ID: <1014736114.26600.0.nnrp-12.c1c3e191@news.demon.co.uk>

Dear all,

forgive me for being a bit vague:

A while ago I am sure I read that someone (steve purcell?) had problems with
unittesting and worked out that python was keeping versions of modules in
memory that they had changed in testing, thus running a version that they
thought no longer existed.

They wrote something that cleared out all modules and classes in python
memory each time they ran their unittests. Apparently it was a long chore.
It seemed to be more convoluted than "reload"ing modules for some reason.

I am having similar problems, and wondered

1) Can anyone remember reading similar? Deja / google no help.
2) py_finalize from C API seems to do a similar job.  How feasible is
calling a C API function from within the python interpeter (I think I am
getting confused)

Basically I am interested in clearing down the python interpreter as if from
new, but without going through the actual from new.

Any ideas?

----------------------------------
Paul Brian
"No sig and lovin' it!"





From gh_pythonlist at gmx.de  Thu Feb  7 17:51:09 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Thu, 7 Feb 2002 23:51:09 +0100
Subject: FixedPoint
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBIEIEKBAA.marklists@mceahern.com>
References: <sDC88.96798$KM2.3611943@atlpnn01.usenetserver.com> <NCBBLFCOHHDIKCAFGCFBIEIEKBAA.marklists@mceahern.com>
Message-ID: <20020207225108.GA2034@lilith.hqd-internal>

Le 07/02/02 ? 14:06, Mark McEahern ?crivit:
> Steve Holden wrote:
> > Oh, no, don't go changing other modules to fix this. Hadn't read your post
> > carefully enough to realize your problems were occurring with database
> > access.
> 
> As it turns out, Gerhard H?ring replied to the thread and he seems to want
> to fix the x == None stuff.  Also, his post made me realize that PyPgSql
> will look for and call a _quote() method on params so that I don't have to
> resort to my goofy fix of making __repr__ return str(self).
> 
> It's all good.

Well, not all. Somebody needs to fix that fixed point class. Failing to
compare to None is clearly a bug.

I don't think pyPgSQL really needs fixing for SQL-ing None, but I'm
willing to do it, if only because "is None" is faster <0.7 wink>.

I think I also found a way to test for iterability:
operator.isSequenceType() seems to do it, but I risk catching too much
(Strings are also sequences, for example). So I could try testing for
strings first.

I fear this might introduce more bugs. Sigh.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 2.1 ?C      Wind: 2.7 m/s



From marknenadov at winningteam.com  Mon Feb 25 15:49:21 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Mon, 25 Feb 2002 15:49:21 -0500
Subject: mcrypt core dumping
References: <20020225.150537.1605908235.23755@d57-33-82.home.cgocable.net>
Message-ID: <20020225.154920.1476153275.23916@d57-33-82.home.cgocable.net>

In case anyone is wondering, I am using:

Python 2.2
Python "mcrypt" 1.0
Libmcrypt 2.4.17


From phr-n2002a at nightsong.com  Tue Feb 12 03:31:31 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 00:31:31 -0800
Subject: searching backwards in a string
Message-ID: <7xk7tjayos.fsf@ruckus.brouhaha.com>

I have a string and want to search backwards for a regexp, starting at
a given index, like if you search backwards in a text editor.  Is
there a way to do that with re?  The docs aren't clear about what's
supposed to happen if the ending position to the pat.search method is
less than the starting position, but I tried it and it doesn't seem
to do what I want.  Any advice appreciated--thanks.


From phd at phd.pp.ru  Thu Feb  7 04:07:20 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 7 Feb 2002 12:07:20 +0300
Subject: Autocoding evolves from........
In-Reply-To: <00b901c1af9e$9de1d080$0101010a@local>; from chris.gonnerman@newcenturycomputers.net on Thu, Feb 07, 2002 at 12:13:46AM -0600
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <00b901c1af9e$9de1d080$0101010a@local>
Message-ID: <20020207120720.I497@phd.pp.ru>

NAY

On Thu, Feb 07, 2002 at 12:13:46AM -0600, Chris Gonnerman wrote:
> The time has come, my friends and colleagues, to bring the issue at hand
> to a vote:
> 
> Should Timothy Rue go away?
> 
> All in favor, post AYE.  All opposed, post NAY.
> 
> (Yes, this is a waste of bandwidth, but in my opinion not as bad as Rue.)
> 
> I recommend you top-post your answers, or better yet, remove all this text
> from your posting and just post the single word.

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



From phr-n2002a at nightsong.com  Mon Feb  4 18:01:18 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 15:01:18 -0800
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu>
Message-ID: <7xg04gsvgh.fsf@ruckus.brouhaha.com>

"John" <nobody at nobody.com> writes:
> Is there any random number generator in Python, that allows users to
> specify "skewness"? Thanks!

I'm not too up on that statistics stuff, but I'd think you'd care
about the shape of the whole distribution and not just the skewness.
Is there a particular distribution you're interested in?


From afp_randjohnson at yahoo.com  Sun Feb 24 05:08:49 2002
From: afp_randjohnson at yahoo.com (Mike Johnson)
Date: 24 Feb 2002 02:08:49 -0800
Subject: Retrieve a variable from an embedded interpreter
Message-ID: <zo1zjioe.fsf@fido.trinitycapital.com>

Hi all!

I have been searching through the documentation, and it doesn't
mention how to retrieve variables from the python interpreter. :-(

I'm trying to keep my embedding work as simple as possible. I'd like
to set a bunch of variables, run a script, and collect some values
after it has finished.

The following is sorta what I had in mind. This will work for my basic
needs, the only thing missing is to somehow retrieve the python
variable "a" and store it in my application's "sa".

------
void cb_script_execute ( void ) {
        char *sa;

        Py_Initialize ();

        PyRun_SimpleString ( "a = \"hi\"" );
        PyRun_SimpleString ( "print \"a=\" + a" );
     /* PyArg_ParseTuple? */
     /* printf ( "sa = %s\n", sa ); */

        Py_Finalize ();

}
-----

Any help would be very appreciated! I've been searching through the
header files for the last couple of days, but I haven't found a
function dealing with this yet.

Thank you,

- Mike Johnson



From nospam at bigfoot.com  Thu Feb  7 09:31:44 2002
From: nospam at bigfoot.com (Gillou)
Date: Thu, 7 Feb 2002 15:31:44 +0100
Subject: re difference between python 1.5 and python 2.1.2
References: <mailman.1013016508.6475.python-list@python.org>
Message-ID: <a3u2n2$mpn$1@norfair.nerim.net>

Many thanks Jason !

"Jason Orendorff" <jason at jorendorff.com> a ?crit dans le message news:
mailman.1013016508.6475.python-list at python.org...
> Gillou wrote:
> > re.compile(r'((ftp|http)://[\w-]+(?:\.[\w-]+)*(?:/[\w-\.?=]*)*)')
>                                                        ^
>
> Put a backslash before the last - in the regular expression.
> 1.5.2 should not have accepted this.  It's a bug in 1.5.2.
>
> ## Jason Orendorff    http://www.jorendorff.com/
>




From nospam at [127.0.0.1]  Wed Feb 27 15:31:29 2002
From: nospam at [127.0.0.1] (Chris)
Date: Wed, 27 Feb 2002 20:31:29 +0000
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
 <3c7d199c.17109656@news.t-online.de>
Message-ID: <8wHsiXAhIUf8Ew0M@[127.0.0.1]>

In article <3c7d199c.17109656 at news.t-online.de>, Gerson Kurz 
<gerson.kurz at t-online.de> writes
>On Wed, 27 Feb 2002 15:05:01 +0000, Chris <nospam@[127.0.0.1]> wrote:
>>In the olden days I played around with BASIC
>>just for fun, writing trivial games and making patterns move on the
>>screen.
>>Can I do that sort of thing with Python?
>>Will I be able to make a ball bounce around inside a box etc.?

>Yup! I recommend you take a look at http://www.vpython.org/, get
>amazed, and motivated to learn python really quickly.

Thanks.  I've looked at that - and been impressed.
-- 
Chris


From xscottgjunk at yahoo.com  Thu Feb 21 20:26:51 2002
From: xscottgjunk at yahoo.com (Scott Gilbert)
Date: 21 Feb 2002 17:26:51 -0800
Subject: Embedding: How to set globals for PyObject_Call
References: <mailman.1014234635.18641.python-list@python.org> <79b2b0bb.0202210841.457c0b81@posting.google.com> <uit8qlqdm.fsf@python.net>
Message-ID: <79b2b0bb.0202211726.7e8ab7f1@posting.google.com>

Michael Hudson <mwh at python.net> wrote:
> xscottgjunk at yahoo.com (Scott Gilbert) writes:
> 
> >   // This one breaks
> >   Var g = VarEval("globals");
> >   g(); // This causes a segfault
> 
> Do we get to see the backtrace?
> 

- For "globals", it segfaults in builtin_globals(...) where it tries
to Py_INCREF the null result from PyEval_GetGlobals().  The dbx stack
trace is below if you really want it.  :-)

- For "locals", it segfaults when trying to Py_INCREF the null return
result from PyEval_GetLocals().  Similar stack trace as below, but
this is on line 1104 of bltinmodule.c

- For "vars", it sets up a Python error of "no locals!?" on line 1992
of bltinmodule.c

- Finally for "dir", PyObject_Dir [object.c, 2085] is called from
builtin_dir [bltinmodule.c, 1992] and returns a null, but doesn't
setup the Python error, so up the call stack PyObject_Call sets up a
system error of "NULL result without error in PyObject_Call"
[abstract.c, 2034].


Many other functions work without a hitch.  (For instance, "list" is a
builtin that can take no args, and I correctly get a new list object
when I do my equivalent of apply() on that.)

Well, I hope this is sufficient info if you view this as enough of a
problem to fix it.  Thanks for your interest.




  0 builtin_globals(self = (nil)) ["Python/bltinmodule.c":665,
0x3ffbffa6814]

  1 PyCFunction_Call(func = 0x140006d70, arg = 0x140003398, kw =
0x1400fb2d8) ["Objects/methodobject.c":83, 0x3ffbff61320]

  2 PyObject_Call(func = 0x140006d70, arg = 0x140003398, kw =
0x1400fb2d8) ["Objects/abstract.c":1665, 0x3ffbff5bb6c]

// Below here are my wrappers around the standard Python stuff.

  3 pv_call(x = 0x140006d70, args = 0x1400e7d98, kwds = 0x1400fb2d8,
err = 0x11fffec20) ["xpyvar.c":572, 0x3ffbff13a5c]

  4 VarApply__X3Var(__result_pointer = 0x11fffeca0, x = @140006d70)
["./xpyvar.h~alt~deccxx_945E6458":1433, 0x120001fd8]

  5 __cl__3VarXv(__result_pointer = 0x11fffeca0, this = 0x11fffec98)
["./xpyvar.h~alt~deccxx_945E6458":868, 0x120001e20]

  6 main() ["main.cc":5, 0x120002204]


From maxm at mxm.dk  Mon Feb 25 08:40:46 2002
From: maxm at mxm.dk (Max M)
Date: Mon, 25 Feb 2002 13:40:46 +0000
Subject: strings and sort()
References: <mailman.1014641682.25067.python-list@python.org> <7xadtxln11.fsf@ruckus.brouhaha.com>
Message-ID: <3C7A3EDE.4010606@mxm.dk>

Paul Rubin wrote:
> "Jason Orendorff" <jason at jorendorff.com> writes:


>>But:
>>  Flat is better than nested.
>>  Sparse is better than dense.
>>  Readability counts.
>>
>>That's why Python is the way it is.
>>
> 
> But: dense
>              can
>                    be
>                            more
>           readable
>   than
>                        sparse.
> 
> See?

1=absolutely
2=be
3=to

if:
     You 1 WANT 3 2 redundancy free, 2cause you 1 MUST adhere 3 the DRY 
principle.
then:
     Things can 2 harder 3 read

regadrs Max M



From philh at comuno.freeserve.co.uk  Tue Feb  5 20:53:08 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 6 Feb 2002 01:53:08 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com>
Message-ID: <slrna61344.8s1.philh@comuno.freeserve.co.uk>

On Wed, 06 Feb 2002 00:04:24 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>>What if a macro needs to create local variables in its expansion?
>
>This is perfectly legitimate. What's not legitimate is...
>
>	deflingo myterm (i, codeblock):
>		j=3
>
>...where j=3 modifies the variable "j" in the context of the caller
>if this variable is preexisting in the caller. That's non hygienic
>behavior. Past experience with macros has shown this to be bad
>behavior; hence there is a design in the community for good "hygiene,"
>as it were.

And what if I decide that for my particular application, that's just
the behaviour I want? Or do you think that you know better than all
the users of your macro system, even though you've not met them and don't
know anything about the software they are writing? And if you take that
attitude, why not just use Pascal or Java?

>>IMO it is more Pythonic to say:
>>
>>   do:
>>      dosomething()
>>      while i <10                                                                                        
>
>Correct. I'm not quite sure how to handle that one.

It doesn't need handling; python does it OK already.

>But of course macros themselves significantly erode simplicity, making
>them _per se_ unpythonic, wouldn't you say? :)

If you are writing code for a complex system, your code needs to have a minimum
level of complexity. You can't get away from that, whether you use python or
any other language. The question is: will using a macro here introduce more
complexity than doing it another way. This is a question that can only be 
answered by having experience and, for want of a better word, taste.

Any dogmatic pronouncement that a particular programming technique (such as
macros) is inherently good or bad is just plain brain dead.



-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jgardn at alumni.washington.edu  Mon Feb 11 02:42:50 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Mon, 11 Feb 2002 16:42:50 +0900
Subject: defining functions
References: <20020210214159.13516.00000450@mb-cq.aol.com>
Message-ID: <_sK98.290$ab7.1150@news.hananet.net>

Patio87 wrote:
> I am pretty new to python, and I dont understand the parameters when
> defining functions. Whenever I see a function definition with a argument
> It seems like the argument already is a variable, but it goes threw the
> function and it makes no sense. I dont know if what I just said made any
> sence but if you can help me please reply
> 
> def factorial(n):
>     if n == 1:
>         return 1
>     else:
>         return n * factorial(n-1)
> What the hell does 'n' have assingned to it?

Remember algebra, or crack open an algebra book if you don't remember.
 
What does "f(x) = 4x" mean?

Asking what "n" represents is like asking what "x" represents in my 
example. You don't know until you actually *do* something like f(4).

By the same token, factorial(4) means "evaluate the factorial code, with 
n=4". When you end up calling factorial(n-1) at the bottom, you will 
evaluate the same function again with n=3, because 4-1 is 3.

Jonathan




From chad.maine at broadlink.com  Tue Feb 26 17:06:00 2002
From: chad.maine at broadlink.com (Chad Maine)
Date: Tue, 26 Feb 2002 22:06:00 GMT
Subject: Class functions
References: <NVSe8.151$en5.1429@typhoon.sonic.net> <3C7C0250.7361AABD@alcyone.com>
Message-ID: <cFTe8.170$en5.1490@typhoon.sonic.net>

I still have the same problem.  I need the function to return control to the
program so I can access the updates to the file object.

"Erik Max Francis" <max at alcyone.com> wrote in message
news:3C7C0250.7361AABD at alcyone.com...
> Chad Maine wrote:
>
> > import theclass
> > X = theclass.CLASS
>
> You want
>
> X = theclass.CLASS(...)
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ Laws are silent in time of war.
> \__/ Cicero
>     Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
>  An Esperanto reference for English speakers.




From phr-n2002a at nightsong.com  Sun Feb 17 11:33:27 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 08:33:27 -0800
Subject: high precision mathematics
References: <mailman.1013961207.23598.python-list@python.org>
Message-ID: <7xbseo5aqw.fsf@ruckus.brouhaha.com>

"Stephen L Arnold" <sarnold at arnolds.dhs.org> writes:
> Having just browsed the pyginac page for the first time, I'm 
> probably just flapping my gums, but why would you want a symbolic 
> math package for high-precision math?  I would think a python 
> wrapper for GMP would do it.  If such a wrapper already exists then 
> you're good to go.  Otherwise, go get swig and generate one.  The 
> Pyginac thing looks cool (if unfinished) but why use it if you 
> don't have to?

There's already a python wrapper for GMP (gmpy.sourceforge.net) but
the OP wanted scientific functions, and gmpy doesn't provide those but
ginac does.  If you just want to call the functions, using pyginac is
probably the easiest thing to do, rather than trying to implement
your own wrapper or function library.


From emile at fenx.com  Fri Feb  1 19:14:04 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 16:14:04 -0800
Subject: Error opening file ! Help, please
References: <Xns91A9CA356801ilen712l@193.70.192.103>
Message-ID: <a3fba3$17nc83$1@ID-11957.news.dfncis.de>

"gimar" <ilen712 at libero.it> wrote in message
news:Xns91A9CA356801ilen712l at 193.70.192.103...
> Sometimes when I try to create an output file executing:
>
> f=open('\paul','w')
>
> I receive the following error message:
>
> "TypeError: object of type 'list' is not callable".
>
>
> This has no sense for me. Can you explain me what's happening ?
>
>

Any chance you're assigning to open?

>>> open = [1,2,3]
>>> f=open('\paul','w')
Traceback (most recent call last):
  File "<pyshell#16>", line 1, in ?
    f=open('\paul','w')
TypeError: 'list' object is not callable
>>>



--

Emile van Sebille
emile at fenx.com

---------



From phr-n2002a at nightsong.com  Tue Feb 19 21:13:29 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 18:13:29 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com>
Message-ID: <7xu1scew8m.fsf@ruckus.brouhaha.com>

"Mike Carifio" <carifio.nospam at nospam.usys.com> writes:
> # simple is good too
> 
> def traverse(list):
>     if list:
>         print list[0]  # or do something here
>         traverse(list[1:])
>     else:
>         print "done"
> 
> # even simplier
> 
> def traverse1(list):
>     if list:
>         # do something
>         traverse(list[1:])

That's ok in Scheme but I don't think any current Python implementations
handle tail recursion "correctly".


From claird at starbase.neosoft.com  Tue Feb 26 10:30:45 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Feb 2002 09:30:45 -0600
Subject: Python and Daemons
References: <02022215115601.01886@localhost.localdomain> <mailman.1014416559.28231.python-list@python.org> <7652139e.0202222105.1220442c@posting.google.com> <mailman.1014484778.28709.python-list@python.org>
Message-ID: <418DCD0D9DCBF13F.B177565408B75219.16813DEFD6A72B5E@lp.airnews.net>

In article <mailman.1014484778.28709.python-list at python.org>,
Justin Sheehy  <justin at iago.org> wrote:
>mrchameleon at hotmail.com (Chris Reay) writes:
>
>>> This is what I use:
>>> 
>> ... and an attribution to W. Richard Stevens is also considered good mojo.
>
>True enough.  Everyone who is going to do any serious network
>programming or much of anything on UNIX should read his books.
>
>A few of his idioms have become so standard that I sometimes
>forget to make the attribution explicit.
>
>Thank you, Richard.  You are missed.
			.
			.
			.
'Worth repeating.  See <URL: http://www.kohala.com/start/ >
for details.

I'll add a few words on the subject of daemons.  Daemons
roughly correspond to what Win* people call "services".
Some Unixois use the word loosely to mean, "long-running
process" or even, in an extreme case, "anything in the
background".  Stevens' heritage generally intends "daemon"
only in a more restricted sense, involving these notions,
but also including specific signal- and I/O-handling qual-
ifications, and umask and process group leader assignment.

The first surprise Pythoneers typically have in "daemonizing",
though, is to discover an unsuspected dependence on environ-
ment variables.  Daemons typically run disconnected from any
TTY, and they also have minimal environment dictionaries (to
mix metaphors slightly).  The latter often flushes out
forgotten assumptions about PATH, *LIB*PATH, and so on.
-- 

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


From neal at metaslash.com  Thu Feb  7 10:40:41 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Thu, 07 Feb 2002 10:40:41 -0500
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de>
Message-ID: <3C629FF9.CF4670FF@metaslash.com>

MDK wrote:
> 
> I have this:
> 
> # myprog.py
> 
> from mymod import *
> 
> def dosomething(x):
>     dosomething(1000)
> 
> # End code
> 
> I want the dosomething(1000) run from the function in mymod.  However,
> instead it is running dosomething(x) in myprog.py.
> 
> Note: Changing the names of dosomething() in either module is not an option.
> (The names depicted here are not the real names in the program but have been
> changed for clarity.)
> 
> I know that one way to fix this is to change
> 
> from mymod import *
> 
> to
> 
> import mymod
> 
> But I would like to know if, without changing the import line, there is a
> way to tell Python to use the dosomething() from mymod.

from  mymod import *
mymod_dosomething = dosomething

def dosomething(x):
    mymod_dosomething(1000)

Neal


From peter at engcorp.com  Sun Feb 24 23:46:47 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 23:46:47 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014418936.10783.python-list@python.org> <3C799833.AC08A401@cosc.canterbury.ac.nz>
Message-ID: <3C79C1B7.AB9287D0@engcorp.com>

Greg Ewing wrote:
> 
> Gustavo Cordova wrote:
> >
> > Das Computermaschine is not for your fingerpoken?
> > .... blinkenlighten?
> >
> > (can't remember nor find the whole thing)
> 
> From memory, something like this:
> 
>      ACHTUNG!
> 
> Dies Komputermaschine ist nicht fur gerfingerpoken
> und mittengrabben. Ist easie schnappen der Springenwerken,
> mit Poppenkorken und Spitzensparken. Die rubbernecken
> Seitseer ist keepen der Hans in der Pocketsrelaxen,
> watchen das Blinkenleits.

And from a google search, several alternatives:

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

ACHTUNG !!

DAS MACHINE IST NICHT F?R DER 
FINGERPOKEN UND MITTENGRAPEN.
IST EASY SCHARPPEN DER SPRINGENWERK, 
BLOWENFUSEN UND POPPENCORKEN
MIT SPIZTENSPARKEN. IST NICHT F?R
GEWERKEN BY DAS DUMM KOPPEN. DAS
RUBBERNECKEN SIGHSEERE KEEPEN
HANDS IN DAS POKETS. RELAXEN UND
WATCH DAS BLIKENLIGHTS.

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

Achtung!!

Das machiene is nicht fur der fingerpoken und mittengraben. Is easy
schnappen der springenwerk, blowenfusen und poppencorken mit spitzensparken.
Ist nicht fur gewerken by das dumm kopfen. Das rubbernecken sightseere
keepen hands in das pockets. Relaxen und watch das blinkenlights.

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

ACHTUNG!

Der computen ist nicht furger fingerpoken und mittengrabben.
ist easy shcnappen der springenwerk, poppencorken, und
blowenfusen mit spitzensparken und smoken.

Das rubbernecken sightseeun keepen das hands in das
pockets, relaxen, und watchen das blinkenlights.

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

ACHTUNG! ACHTUNG!

Alles touristen und non-technishen looken peepers! Das machine is nicht 
fur der fingerpoken und mittengrabben. Is easy schnappen der springenwerk, 
blowenfusen und popencorken mit spitzen sparken. Das machine is diggen by 
experten only. Is nicht fur gerwerken by das dummkopfen. Das rubbernecken 
sightseeren keepen das cottenpiken hands in das pockets.  Relaxen und 
watchen das blinkenlights.

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

Apparently German, as Perl, has many ways of saying the same thing.
And it's equally readable.  <grin>

-Peter


From piet at cs.uu.nl  Sat Feb 23 10:10:34 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 23 Feb 2002 16:10:34 +0100
Subject: basic books/guides on multithreading programming?
References: <mailman.1013813246.12449.python-list@python.org> <7xlmdrmymb.fsf@ruckus.brouhaha.com>
Message-ID: <wzvgcodyj9.fsf@ucu-106-58.ucu.uu.nl>

>>>>> Paul Rubin <phr-n2002a at nightsong.com> (PR) writes:

PR> Tripp Scott <tripps81 at yahoo.com> writes:
>> which ones do you recommend? examples in any language is fine, though
>> i prefer python, perl, or ruby. what i would like is guidance on what
>> is thread safety, how to achieve it, what are the issues in
>> multithreading programming, the global interpreter lock, and other
>> basics.

PR> There's a pretty good book called (I think) Java Threads.  I looked at it
PR> ages ago so I don't remember too much specific from it.  Python's threading
PR> module is sort of a subset of Java threads.

Maybe this one: http://www.mcs.drexel.edu/~shartley/ConcProgJava/
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From -$P-W$- at verence.demon.co.uk  Mon Feb  4 07:03:57 2002
From: -$P-W$- at verence.demon.co.uk (Paul Wright)
Date: 4 Feb 2002 12:03:57 -0000
Subject: Attention Phil Hunt
References: <mailman.1012327189.31750.python-list@python.org> <585.799T42T11732924threeseas@earthlink.net> <slrna5rqpa.19d.philh@comuno.freeserve.co.uk>
Message-ID: <a3ltbd$3oi$1@verence.demon.co.uk>

In article <slrna5rqpa.19d.philh at comuno.freeserve.co.uk>,
phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Mon, 04 Feb 2002 00:00:59 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>>On 29-Jan-02 12:58:27 Laura Creighton <lac at strakt.com> wrote:
>>>Every minute you spend arguing here is a minute that the release of
>>>Herbivore is delayed.
>>
>>protecting once psuedo position can be very important.
>
>ITYM "protecting one's psuedo position can be very important"
>
>In what way is my psotion "pseudo"? Are you suggesting I didn't
>originate the herbivore projevct or write Herbrip? If so, who did?
>
>Or did you mean something else entirely?


                      ______
                     /      \
                   .' PLEASE `.
                   |  DO NOT  |      _____
                   | FEED THE |    ,'.....`.
                   `. LOONY  ,'  ,'........ )
                     \_    _/   |........ ,'
                       |  |     `. .... _/
                       |  |      ,'.,'-'
                       |  |     /../
                       |  |   ,'.,'
                       |  |  /../
                     . |  | /..'
                   .\_\|  |/_/,
                   ___ |  | ___
                     . `--' .
                      .    .

It was fun at first but now it's tiresome. 

-- 
Paul Wright | http://pobox.com/~pw201 |


From joost_jacob at hotmail.com  Sun Feb  3 07:04:09 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 3 Feb 2002 04:04:09 -0800
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org>
Message-ID: <13285ea2.0202030404.5dfe9b66@posting.google.com>

> > Why in the std distribution of Python 2.2 there isn't the curses
> > module?
[Gerhard H?ring]
> If you're using Windows, that explains it ;-) There is indeed no curses
> module included in the Python binary distribution for Windows. IIRC the
> reason is that there is no ncurses library readily available on win32 to
> build the curses support against.

There have been many more questions about curses for Windows.  There
is even no alternative in the Python distribution, you can download
3rd party packages like WConio or console but you have to compile them
yourself and you are out of luck if you do not have that MSVC++
compiler the Python team choose if you want to communicate with their
libraries.  This is a bad situation and it comes up again and again
when you [have to] use Windows.  For a programming class in school
[never did see a school using linux except university] it would be
good if Python came with something like curses for Windows, the
flickering menus look bad and give Python a bad reputation.  And no,
you do not want to start a programming class with explaing TKinter and
callbacks hehe :-)


From phr-n2002a at nightsong.com  Thu Feb 21 09:19:24 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 06:19:24 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014299857.23656.python-list@python.org>
Message-ID: <7xy9hmj4sz.fsf@ruckus.brouhaha.com>

Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> I didn't express myself properly.  What I meant is, in a CGI application if
> you're including the session ID in the query string, then you have to make
> sure all of your relative links within the site are constructed to include
> that session ID.  In other words, your links will have to be written like:
> 
> '<a href="SomeOtherScript.cgi?_SID_=%s">Link to Some Other Script</a>' %
> sessionID

No no, that's what the BASE tag does for you.  It maps

  <a href=SomeOtherScript.cgi>

into a request to http://yoursite.com/_SID_=12345/SomeOtherScript.cgi.

You then use a server side rewrite rule to map the SID path component
into a query parameter.


From BPettersen at NAREX.com  Mon Feb 25 17:00:46 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Mon, 25 Feb 2002 15:00:46 -0700
Subject: On inner classes
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158D37@admin56.narex.com>

> From: Gon?alo Rodrigues [mailto:op73418 at mail.telepac.pt] 
> 
> On Mon, 25 Feb 2002 12:31:38 -0700, "Bjorn Pettersen" 
> <BPettersen at NAREX.com> wrote:
> 
> >> From: Gon?alo Rodrigues [mailto:op73418 at mail.telepac.pt]
> >> 
> >> Suppose we have something like
> >> 
> >> class <name>:
> >>     class <name>:
> >>         <whatever>
> >>     <whatever>
> >> 
> >> that is, a class defined inside another. How can the inner
> >> class acess the methods/attributes of the outer one?
> >
> >I'm assuming you're meaning something like:
> >
> >  class Outer:
> >     def outerMethod(self):
> >        class Inner:
> >           def innerMethod(this):
> >              self.outerMethod() # call class Outer's outerMethod
> >
> >-- bjorn
> 
> Actually no. Let me be more specific
> 
> class outer_class:
>     def __init__(self):
>         self.example = "just a test"
> 
>     class inner_class:
>         def __init__(self):
>             <whatever>
> 
> and now, specifically:
> 
> How in the __init__ of the inner class do i acess the 
> instance attributes declared in the __init__ of the outer 
> class? In this case the instance attribute example?

You can't. Think about it. Your inner_class is at class scope in
outer_class. What you're after is a value of an *instance* of
outer_class, however you have no handle to such an instance from within
inner_class, ergo you can't get to it.

-- bjorn



From martin at v.loewis.de  Wed Feb 27 14:54:34 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 20:54:34 +0100
Subject: Thread support in NetBSD
References: <a5i40k$1k59$1@news.uit.no> <m3lmdf1dv8.fsf@mira.informatik.hu-berlin.de> <a5iarm$1nvp$1@news.uit.no>
Message-ID: <m3zo1uittx.fsf@mira.informatik.hu-berlin.de>

"Anders ?berg" <andersaa at stud.cs.uit.no> writes:

> So i should use
> ./configure --with-threads=/usr/pkg/pthread/lib
> instead of
> ./configure --with-threads=/usr/pkg/pthread ?

You can try, but I doubt it will help.

> > If the pthreads module is suited for your thread library, you can try
> > setting CPPFLAGS and LDFLAGS before invoking configure. If that fails,
> > you need to adjust configure.in.
> 
> What should i set these flags to? Sorry but i don't know much about these
> things..

Then I guess you are on your own. Hire somebody to do this for
you. Or, on a Unix newsgroup, ask what environment variables are.

Regards,
Martin


From wagner.nils at vdi.de  Fri Feb 22 09:30:57 2002
From: wagner.nils at vdi.de (My VDI Freemail)
Date: Fri, 22 Feb 2002 15:30:57 +0100
Subject: Problems with xreadlines and SuSE 7.3
In-Reply-To: <1014386773.20881.ezmlm@suse.com>
Message-ID: <E16eGix-0002s5-00@mail.python.org>

Hi,

I can't import xreadlines

Python 2.1.1 (#1, Sep 24 2001, 05:28:47)
[GCC 2.95.3 20010315 (SuSE)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import xreadlines
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: No module named xreadlines
>>>

Any idea ?

Nils





From timo77 at gmx.net  Mon Feb 25 06:46:15 2002
From: timo77 at gmx.net (Timo Rainio)
Date: Mon, 25 Feb 2002 11:46:15 GMT
Subject: Remote debugging ?
References: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com> <7xadtxajuo.fsf@ruckus.brouhaha.com>
Message-ID: <1014637646.677472@news.kkf.net>

"Paul Rubin" <phr-n2002a at nightsong.com> schrieb im Newsbeitrag
news:7xadtxajuo.fsf at ruckus.brouhaha.com...
> Lothar Scholz <llothar at mailandnews.de> writes:
> > Is there any debugger for python that allows me to do remote debugging
> > from a windows 2000 client to a python script running on a Linux
> > Server.
> >
> > It must not be freeware.
>
> If freeware is unacceptable, you better stop using Python as well.

That's a common error of native german speakers.

Lothar meant to say: "It does not have to be freeware".
that's what saying "Es muss nicht freeware sein" in german means (it's not a
must, but it can)

a slight difference in logic, than can cause big missunderstandings as we
see here ... ;-)




From James_Althoff at i2.com  Tue Feb  5 14:29:17 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Tue, 5 Feb 2002 11:29:17 -0800
Subject: lambda expressions
Message-ID: <OF349452E2.282CE2D7-ON88256B57.0069DC79@i2.com>

[Jason Orendorff]
>I should have said "Lack of tail call optimization could
>be a problem."
>
>For spectators:  the reason tail recursion is important
>at all is, any tail call can be compiled into (essentially)
>a goto.  This allows functions like WHILE() to loop indefinitely
>without eating up memory.
>
>In Python, tail calls are not optimized.  So in Python this
>uses O(N) memory, where N = the number of times the loop
>executes. Eventually you get "RuntimeError: maximum recursion
>depth exceeded".

As Jason points out, any tail call -- not just recursive calls --
potentially can be optimized.  A downside (the upside is mentioned above)
-- in addition to some added complexity of implementation -- is that you
pay the piper sometimes when debugging since you lose all intermediate
context (by overwriting the stack frame early -- hence the savings in stack
memory).

Jim




From cqxiao at charter.net  Mon Feb 25 20:09:21 2002
From: cqxiao at charter.net (chongqing xiao)
Date: Mon, 25 Feb 2002 19:09:21 -0600
Subject: call vb dll (will pop up vb form) form python
Message-ID: <u7loa4o4kt5t48@corp.supernews.com>

Hi,

I have a visual basic component (dll) which will pop up a form in run time).
When I try to call this dll in python (using activie python on win2000), it
give the
following error message like this:
Unable to show modal form in this enviroment

Any ideas or suggestions are welcome.

thanks
chong




From max at alcyone.com  Wed Feb 20 22:09:56 2002
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 20 Feb 2002 19:09:56 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com>
Message-ID: <3C746504.C2C70DF0@alcyone.com>

Jason wrote:

> In my experimentation 0, None, "", [], and () all seem to be
> considered false.  Just to confound me none of them are == to each
> other.  Coming from a Scheme world this is confusing.  Are there any
> other values that are false?

Sure.  There's also 0l, 0.0, and any class which has had its __nonzero__
method overridden to return false at times.  Many languages (other than
Scheme) have a loose definition of truth and falsity, simply because at
times it's useful to be able to test any object for truth.  That's true
in Scheme too, except that in Scheme the only false value is #f*. 
That's the way Scheme works, but it's not universal.

> Why aren't () and "" the same thing?
> Don't both describe an immutable sequence of values (basically, why
> aren't strings either lists or tuples)?

Strings, lists, and tuples are sequences, but they're not the same
thing.  'abc' is distinct from ['a', 'b', 'c'], which is distinct from
('a', 'b', 'c').

. 

* And, of course, the empty list (), should you be running on a system
where #f and () are identical.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From gaul at spam.utexas.edu  Tue Feb 12 16:08:26 2002
From: gaul at spam.utexas.edu (Andrew Gaul)
Date: Tue, 12 Feb 2002 21:08:26 +0000 (UTC)
Subject: Fun with lambda and map
References: <3c566a21.22129828@news.t-online.de> <slrna6i14c.sp.gaul@bifrost.cs.utexas.edu> <slrna6i45q.sp.gaul@bifrost.cs.utexas.edu>
Message-ID: <slrna6j12a.sp.gaul@bifrost.cs.utexas.edu>

In article <slrna6i45q.sp.gaul at bifrost.cs.utexas.edu>, Andrew Gaul wrote:
> In article <slrna6i14c.sp.gaul at bifrost.cs.utexas.edu>, Andrew Gaul wrote:
>> In article <3c566a21.22129828 at news.t-online.de>, Gerson Kurz wrote:
>>> fibonacci = lambda x:map(lambda o:(map(lambda c:map(lambda l:o.__setslice__(l[0],l[1],l[2]),([o[2]+3,o[2]+4,[o[0]]],[0,3,[o[1],reduce(lambda x,o:x+o,o[:2]),o[2]+1]])),range(x)),o)[1],[[1,1,0]+range(x)])[0][3:]
>> 
>>     fibonacci = lambda z: map(lambda n, f=(lambda f,x,a,b: \
>>     [x<=0 and b, x>0 and f(f,x-1,b,a+b)][x>0]): f(f,n,0,1), range(z))
> 
>     fibonacci = lambda z, f=(lambda f,x,l: \
>     [x<=0 and l, x>0 and f(f,x-1,l+[l[-1]+l[-2]])][x>0]): f(f,z-2,[1,1])

In my zealousness to cut characters, I also sacrificied correctness for
the base case.  This should be proof enough that code like this should
never be written.  Here's a final, working version:

    Python 2.2 (#5, Jan 23 2002, 16:23:03)
    [GCC 3.0.3] on linux2
    Type "help", "copyright", "credits" or "license" for more information.
    >>> fibonacci = lambda z, f=(lambda f,x,l=[0,1]: \
    ... [x<=0 and l, x>0 and f(f,x-1,l+[l[-1]+l[-2]])][x>0]): f(f,z-1)[1:]
    >>> fibonacci(1)
    [1]
    >>> fibonacci(10)
    [1, 1, 2, 3, 5, 8, 13, 21, 34, 55]

-- 
Andrew Gaul
http://gaul.org/


From andersjm at dancontrol.dk  Tue Feb  5 12:23:54 2002
From: andersjm at dancontrol.dk (Anders J. Munch)
Date: Tue, 5 Feb 2002 18:23:54 +0100
Subject: distutils, C extensions, Borland compiler, wxPython
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk> <ulme8ovdf.fsf@python.net>
Message-ID: <3c601545$0$219$edfadb0f@dspool01.news.tele.dk>

"Michael Hudson" <mwh at python.net> wrote in message
news:ulme8ovdf.fsf at python.net...
> "Anders J. Munch" <andersjm at dancontrol.dk> writes:
>
> > I have built Python from source on Win2K using cygwin except with a
> > Borland compiler (BCB4) instead of gcc.  It took a fair amount of
> > hacks to get the configure script and makefiles to work (s/.o/.obj/ is
> > just the beginning) but it was worth it.
>
> Which version of Python?  I thought there was at least *some* support
> for this in recent versions of python.

I began with a not-so-recent version. When I had the choice I
preferred wrestling with the vaxocentrisms of the main build mechanism
over wrestling with the MSVC++'isms in the PC subdirectory. Seemed
like the right thing to do.

Currently using 2.2.

>
> Have you looked at google to see if people have tried this before?

Yes.  I don't like to reinvent the wheel, but in this case all the
wheels I could find were square.

> Try using
>
> $ python setup.py build --compiler=bcpp install
>
[...]
>
> Hopefully the author of ../distutils/bccpcompiler.py has done the
> messing already...

It's worth looking into.  I presume the bcpp code is written with
interfacing an MSVC++ Python build in mind.  Perhaps I can patch it.

>
> > and also my python.exe is not a cygwin app. and hence doesn't
> > understand the cygwin paths from the make process.
>
> Don't really follow you here, but as I've never used the borland
> compiler and try to avoid understanding windows issues...

Let me illustrate by example.

bash is cygwin-enabled:
  $ ls /usr/local/lib
  python2.2

In native cmd.exe the same directory is called something else:
  C:\>dir /b c:\cygwin\usr\local\lib
  python2.2

But cmd.exe, like my python.exe, isn't cygwin-enabled:
  C:\>dir /usr/local/lib
  Invalid switch - "usr".

  C:\>dir \usr\local\lib
  The system cannot find the path specified.

>
> > Now what I'd like to have is for distutils to just tell me which
> > .c/.c++ files need to be compiled into python and with which options,
> > so that I could add these files to Setup.local manually and rebuild
> > python.exe.
>
> This *shouldn't* be that hard; the setup.py file should contain a list
> of source files that need compiling, like
>
> setup(...
>       extensions = [Extension("wx", ["wxfile1.c", "wxfile2.c"]),
>         ...]
>       ...)
>
> this translates to lines in Setup.local along the lines of
>
> wx wxfile1.c wxfile2.c

I've taken this path before with simpler stuff, but the problem is
that I'll miss out on whatever clever stuff the setup script does in
addition to running distutils.core.setup.  E.g. the wxPython setup
will run SWIG to generate .c files on the fly if you ask it to.

Maybe that's what I'll end up doing -- I probably will be able to
decipher wxPythons setup script and emulate it -- but I was hoping to
come up with a procedure that didn't require me to reverse-engineer
the setup script for every package with C extensions that I install.

> > Are there any docs on how to tell distutils which compiler to use
> > and how?
>
> There's
>
> http://python.sourceforge.net/devel-docs/inst/inst.html

The "Borland C++" section begins by saying "Python is built with
Microsoft Visual C++".  Non sequitur.

>
> but I don't know how thorough this is -- I generally read the source
> when I need to understand what distutils is up to...

320079 bytes of source code and no overview document.  Even Python can
be a write-only language if the code is large enough and there's no
way of getting the big picture.

- Anders




From philh at comuno.freeserve.co.uk  Sat Feb 16 13:13:54 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sat, 16 Feb 2002 18:13:54 +0000
Subject: Apology
References: <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net>
Message-ID: <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk>

On Sat, 16 Feb 2002 01:45:42 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 12-Feb-02 10:46:49 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>On Mon, 11 Feb 2002 18:37:51 -0600, Chris Gonnerman
>><chris.gonnerman at newcenturycomputers.net> wrote:
>>>
>>>Indeed.  He uses a nonstandard definition of "language" and applies
>>>"physics"
>>>to programming.  He can't program, you know; ten years on this project and
>>>nothing to show for it.  If I had tried to develop something like this for
>>>ten years and had nothing to show for it but a website, I'd give up.
>
>>At least his code is all neatly lined up!
>
>Chris's comprehension level is greatly lacking and you've shown your
>superiority attitude.
>
>Ten years???? it was Feburary 1988 when I identified the nine commands.

14 years then. You do realise, that makes you look even worse.

>I never claimed to be a programmer 

14 years is plenty of time to lesarn, if you want to.

>and never wanted to make the VIC but
>rather someone to use it. 

This implies you'd rather someone else coded it...

>But I also don't, and never will want to be
>abused by any programmer wanting to take credit and property rights to the
>VIC.

...but if they do, you want to take the credit and copyright for their
code. 

>The computer industry has alot of ethical problems in this area of taking
>everything the can from the users, from their ideas, to their debugging
>efforts (feedback), to their money in payment for what came from them, and
>now rental....and always it is never quite right but lacking always... for
>upgrading recycling of consumer entrapment abuse.

If you refer to Microsoft, I agree. But free software / open source *isn't*
like that, in fact it is deliberately in opposition to that sort of thing.

>It's good to know that such practice is going to soon come to an end.

Because of VIC? Or because of something else?


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From chrishbarker at attbi.com  Fri Feb 15 15:17:28 2002
From: chrishbarker at attbi.com (Chris Barker)
Date: Fri, 15 Feb 2002 12:17:28 -0800
Subject: MemoryError using NumPy
References: <a4gqd9$68e5$1@ID-11957.news.dfncis.de>
Message-ID: <3C6D6CD8.FA03AA2B@attbi.com>

Emile van Sebille wrote:

> Numeric.sum(Numeric.take(myDataArray, indices)).astype(Numeric.Float64)

> This causes a temporary jump in memory usage (from ~800Mb to ~1.2Gb)
> before settling down again, but it delivers the speed I'm looking for
> (~7 secs vs ~30 summing in loop).  All starts well, but soon I get
> MemoryError "can't allocate memory for array" only when indices is
> large.  I suspect that memory has fragmented such that a sufficiently
> large contiguous block is not available to set up the array.  Is this
> the likely cause, and if so is there a way to un-fragment memory?  Or
> might there be an alternative to Numeric.take that gets to the same
> place?

I think you have two functions in one line that create a copy (take()
and astype()). I have never gotten an answer as to when CPython frees
the memory for temporary variables in one statement, but if you can
delete myDataArray before calling astype(), you might save something
there.

If not, I'm interested in finding oput what the options are for memory
defragmenting...

I'd also be interested in what happens under Linux, I still have a much
better feeling about the robustness of Linux over Win2k, I'd like to see
if they are different in this respect.

-Chris






-- 
Christopher Barker,
Ph.D.                                                           
ChrisHBarker at attbi.net                ---           ---           ---
                                     ---@@       -----@@       -----@@
                                   ------@@@     ------@@@     ------@@@
Oil Spill Modeling                ------   @    ------   @   ------   @
Water Resources Engineering       -------      ---------     --------    
Coastal and Fluvial Hydrodynamics --------------------------------------
------------------------------------------------------------------------


From logiplexsoftware at earthlink.net  Wed Feb  6 02:38:29 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 5 Feb 2002 23:38:29 -0800
Subject: wxPython book
In-Reply-To: <88bc63c6.0202040505.2a9c3016@posting.google.com>
References: <88bc63c6.0202040505.2a9c3016@posting.google.com>
Message-ID: <20020205233829.0d6f866a.logiplexsoftware@earthlink.net>

On 4 Feb 2002 05:05:53 -0800
Doug Farrell wrote:

> Hi all,
> 
> I'm busy evangilizing my office co-workers on the wonders of Python
> and wxPython, to which they are responding pretty well. We are
> currently working on a test project that uses both, and the most
> frequent complaint I get is "there's no book on wxPython..... wah wah
> wah". I have seen mention of a book being written, presumably by Robin
> Dunn. Is this true and if so, any idea when it might be coming out?

No idea, but do they all really say "wah wah wah"?  

Weird.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From marklists at mceahern.com  Sat Feb  9 01:03:52 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Fri, 8 Feb 2002 22:03:52 -0800
Subject: FixedPoint
In-Reply-To: <HFEKILOLEFEFMKAECNDLKEFHDBAA.jason@jorendorff.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEMJKBAA.marklists@mceahern.com>

[Jason Orendorff]
> How about:
>
>     def __cmp__(self, other):
>         try:
>             xn, yn, p = _norm(self, other)
>             return cmp(xn, yn)
>         except:
>             return cmp(float(self), other)

I like that!  Would it be better style to trap only the specific error I
know about?

Thanks,

// mark




From ole at cuttlefish.anu.edu.au  Thu Feb  7 00:20:54 2002
From: ole at cuttlefish.anu.edu.au (Ole Moller Nielsen)
Date: Thu, 7 Feb 2002 16:20:54 +1100
Subject: Pypar 1.0 - Simple and efficient MPI binding for Python
Message-ID: <20020207162054.A18830@anu.edu.au>

PyPAR 1.0 - Simple and efficient MPI binding for Python
Copyright (C) 2001, 2002 Ole M. Nielsen
-------------------------------------------------------

This is to announce GPL publication of pypar, ver 1.0 - 
a simple and efficient Python binding to MPI for parallel 
programming using Python.
	
Background:
-----------
The use of multi processor computers is becoming increasingly 
common and they appear in many forms: Desktop computers with more 
than one processor sharing memory, clusters of PC's connected with 
fast networks known as Beowulff clusters, and high-end super computers
all make use of parallelism.
To efficiently use these machines in a portable way one must be able to
control communication among programs running in parallel. 
One such standard is the Message Passing Interface (MPI) for 
inter-processor communication. 

Python and MPI:
---------------
There are a number of other Python bindings to MPI that are 
more comprehensive than pypar (PyMPI, Scientific Python).
However, pypar stands out by not requiring the Python interpreter to be 
modified, by being very easy to install and by
by shielding the user from many details involving data types and 
MPI parameters without sacrificing the full bandwidth provided by the 
underlying MPI implementation.

Download:
---------
Pypar can be downloaded from 

  http://datamining.anu.edu.au/pypar
    
where there is also more information available.

Credentials:
------------
Pypar was developed by Ole Nielsen at the Australian National 
University in 2001 for use in the APAC Data Mining Expertise Program
and has been published under the GNU Public License
(http://www.gnu.org/licenses/gpl.txt) as a small contribution

Contact:
  Ole.Nielsen at anu.edu.au
  
<P><A HREF="http://datamining.anu.edu.au/pypar">Pypar 1.0</A> - 
A simple and efficient MPI binding for Python.  (07-Feb-02)
	      



From phr-n2002a at nightsong.com  Sun Feb 17 23:35:24 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 20:35:24 -0800
Subject: Editable strings anyone?
References: <mailman.1013991447.20337.python-list@python.org>
Message-ID: <7xr8njv243.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> > Do a half million characters get copied around for each of the 
> > delete operations?
> 
> Yes.  There's no gap in arrays.

OK.  I guess the simplest approach to editable strings is to have a new type of array
that has multiple segments and implements the buffer interface.  Then the re module
can be extended to handle the buffer interface, if it doesn't handle it already.
Doesn't sound too bad--if I get industrious I might hack this up.


From jason at jorendorff.com  Tue Feb 26 19:38:29 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 18:38:29 -0600
Subject: what I would like to see in python to make a better "glue" language (Newbie)
In-Reply-To: <a5h3qd$2df$2@peabody.colorado.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLOEINDFAA.jason@jorendorff.com>

Fernando Perez wrote:
> Anthony_Barker wrote:
> > One more idea for Administrators/Glue Implementers; lots of code
> > examples is key.
>
> 100% agreed. The python docs are very nice, but they could definitely use
> more example code. Often a single well written example is worth
> 10 pages of ultra-verbose text.
>
> The doc maintainers do a great job already, [...]

Indeed they do.

I would love to see a "wiki" on python.org, initially populated
with the latest Python documentation (in some easy-to-edit form),
just as an experiment.  I think some examples might magically
appear in it.

The cookbook is a bit of a pain.

## Jason Orendorff    http://www.jorendorff.com/




From m.hadfield at niwa.co.nz  Thu Feb 28 00:12:49 2002
From: m.hadfield at niwa.co.nz (Mark Hadfield)
Date: Thu, 28 Feb 2002 18:12:49 +1300
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu>
Message-ID: <a5m3e7$f6f$1@newsreader.mailgate.org>

"Fernando P?rez" <fperez528 at yahoo.com> wrote in message
news:a5kcf4$qei$1 at peabody.colorado.edu...
> Python is a general purpose language, not one to be used only by
> number theorists. Putting at a level as basic as the integers a
> convention which allows things like 'for i in 3' to be valid
> iteration code really sounds bizarre and counter intuitive to me.
>
> But maybe it's just me.

No, it's not just you.

---
Mark Hadfield
m.hadfield at niwa.co.nz  http://katipo.niwa.co.nz/~hadfield
National Institute for Water and Atmospheric Research





From bmitch at copper.net  Fri Feb  1 18:53:55 2002
From: bmitch at copper.net (Brian Mitchell)
Date: Fri, 1 Feb 2002 18:53:55 -0500
Subject: Windows 2000
Message-ID: <01C1AB52.89FFDF80@BRIANMIT>

I tried an install of Python 2.2 on a Windows2000 box today. IDLE
seemed to function okay but I couldn't close the command mode
with control-Z.  Is there anything special about using Python under
Windows2000 as compared to Windows98?  TIA.



From webmaster at prankcallsunlimited.com  Mon Feb 18 00:34:50 2002
From: webmaster at prankcallsunlimited.com (Bob Pranky)
Date: 17 Feb 2002 21:34:50 -0800
Subject: FREE SOUND EFFECTS
References: <638cbd51.0202171525.6f5345ee@posting.google.com>
Message-ID: <638cbd51.0202172134.10342754@posting.google.com>

webmaster at prankcallsunlimited.com (Bob Pranky) wrote in message news:<638cbd51.0202171525.6f5345ee at posting.google.com>...
> Free downloads of sound effects in lots of different categories at
> http://www.a1freesoundeffects.com
> Professional sounds also are created for all needs
> http://www.a1freesoundeffects.com/service.htm

Forgot to mention the CD's:
http://www.a1freesoundeffects.com/donation.htm


From justin at iago.org  Mon Feb  4 10:21:30 2002
From: justin at iago.org (Justin Sheehy)
Date: Mon, 04 Feb 2002 10:21:30 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org>
	<7xsn8hjxzw.fsf@ruckus.brouhaha.com>
Message-ID: <x7g04h1ddx.fsf@ra.iago.org>

Paul Rubin <phr-n2002a at nightsong.com> writes:

>> Not so:  introducing "//" was 100% backward-compatible, so isn't controlled
>> by __future__.  "//" works out of the box in 2.2, no __future__ required.
>
> If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> it isn't backwards compatible.

You don't seem to understand what "backward-compatible" means.  What
you are talking about is more related to forward compatibility, and is
generally far less important than backward compatibility.

Backward compatibility is broken when code that does work in an older
version does not work in a newer version.

What concrete example do you have of your own working code that was
correct and working, but broke in 2.2?

(Note that the "correct" bit there rules out socket-argument problems,
since the code that broke in those cases was not really conforming to
the admittedly misleading docs.)

I believe that there are some minor and unfortunate cases, but I'd
like to know what you were actually bitten by.  If you don't have an
answer, you might be better off programming than ranting.

> If the official Python maintainers think they might release a 1.5.3
> someday, then it's fair to say that the 1.5 series is still current.

No, it isn't.

2.2 is the only current series, exactly because the Python dev team
says so.  Older versions may receive bugfix releases, but that does
not make them current.

> More importantly, in the real world, a lot of people are still using
> 2.1 and 1.5, and few are using 2.2.  So coding anything in a way that
> requires 2.2 inconveniences a lot of users. 

So, if you don't want to inconvenience those users, don't make your
code require 2.2.  This seems fairly obvious and easy.

> That might be worth it for serious features like metaclasses or
> generators, but is silly to do over this division nonsense

What division nonsense?

Show me the division-using code that works pre-2.2 but breaks in 2.2.

I don't believe that there is any.

-Justin

 




From mlh at vier.idi.ntnu.no  Mon Feb 11 13:24:09 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 11 Feb 2002 18:24:09 +0000 (UTC)
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <mailman.1013444540.5738.python-list@python.org>
Message-ID: <slrna6g329.3ul.mlh@vier.idi.ntnu.no>

In article <mailman.1013444540.5738.python-list at python.org>, Michael
Chermside wrote:
>Mike Fletcher writes:
> > [complicated workaround to annoying problem with "obvious and
> > pythonic" implementation of common design pattern]
>
>Ouch.
>
>I'm no expert, but I see your point, and can't beat your workaround. 
>Thanks for sharing.

I missed the original post, unfortunately... I'd just like to mention
that we have some workarounds for instance methods etc. and weakrefs
in the References module of Anygui (anygui.org).

>-- Michael Chermside

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From joshm at taconic.net  Tue Feb 26 18:41:46 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 26 Feb 2002 18:41:46 -0500
Subject: .NET only solves problems I don't have! (Was: advocacy help ( Python vs VB ))
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01><mailman.1014706963.7898.python-list@python.org><3C7BB093.3020600@mxm.dk> <mailman.1014745305.15528.python-list@python.org>
Message-ID: <3c7c1ad8_3@corp.newsgroups.com>

> > What I find most amusing about com and .NET is that they are trying to
> > solve a problem I only had when programming using MS tools.

This gets my vote for quote of the week.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From starx at pacbell.net  Thu Feb 14 05:14:41 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 14 Feb 2002 10:14:41 GMT
Subject: IPython 0.2.5
References: <a4ainp$gts$1@peabody.colorado.edu> <fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com>
Message-ID: <Xns91B516D3A1E11RASXnewsDFE1@130.133.1.4>

Terry Reedy || Tue 12 Feb 2002 07:55:55a:

> 
> "Fernando P?rez" <fperez528 at yahoo.com> wrote
>> Portability: Linux (and other unices, including Mac OSX), Windows XP.
>> Should run fine on all WinNT, and probably also on Win9x (I can't test
>> that). 
> 
> You wrote further on your web page: "Hopefully Windows users will
> quickly migrate to XP (which is actually quite good in many respects)
> and we'll bury Win95 and children were they should have been buried
> years ago."
> 
> Don't hold your breath.  XP is a greater resource hog than ME is a
> greather resource hog than W98 is a greater resource hog than W95, so
> many (most?) users can't upgrade, even if they would want to pony up
> the upgrade fee and run the risk (in some cases, certainty) of
> disabling a hardware system and set of software than currently work.
> (XP does NOT run everything that runs under Wxx.)
> 

Then there are those of us who refuse to upgrade to XP because of MS's 
recent we-own-your-computer you're-a-pirate-until-proven-otherwise paying-
customers-come-dead-last registration BS, among other things. (And 
generally think 98 is the best thing to come out of them in general :).

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From djc at object-craft.com.au  Fri Feb 15 05:25:39 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 15 Feb 2002 21:25:39 +1100
Subject: Sybase module 0.33pre4 released
Message-ID: <m3r8nnvy70.fsf@ferret.object-craft.com.au>

What is it:
The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

Note:

Since the 0.32 release I have come to the realisation that the Sybase
dynamic SQL functionality is not nearly general enough to use for
correct DB-API functionality.  I have thrown up my hands and gone back
to the drawing board with the high level Sybase.py module.

Status of 0.33pre2, 0.33pre3:

The 0.33pre2 release seems to be quite solid.  I have not received any
complaints from any of the 10 (count them) people who downloaded the
module.  Ben Gertzfield managed to implement a multi-threaded whois
server on top of the 0.32pre2 release.

The only problems reported for 0.33pre3 were the Solaris build needed
tweaking and the tar file unpacks to an unversioned directory.

Changes for 0.33pre4:

1) In implementing some sample code to assist a user I discovered a
   threading locking problem with the Cursor.__del__() method.
   Management of the Cursor related Connection locks has been
   reimplemented.

2) Solaris build has been fixed again - thanks to Kent Polk.

3) The tar file now unpacks to a directory called sybase-<release>.

The changes for this release are so slight that I am inclined to mark
this as the next stable release in a week or so.

Extra Extra Pleading:

Please download the pre-release version, try it out, and then send me
some mail telling me what you found.  You would be surprised at how
little mail I get regarding the Sybase module, I really do need your
help.  If more than 7 to 10 people tried the module we would all
benefit...

The module is available here:

    http://www.object-craft.com.au/projects/sybase/sybase-0.33pre4.tar.gz

The module home page is here:

    http://www.object-craft.com.au/projects/sybase/

- Dave

-- 
http://www.object-craft.com.au


From tatebll at aol.com  Fri Feb  1 08:23:28 2002
From: tatebll at aol.com (Bill Tate)
Date: 1 Feb 2002 05:23:28 -0800
Subject: Java vs Python Benchmarks: Java is faster
References: <Ldc68.67$8u4.28600@wagner.videotron.net> <cb4ba455.0201311347.11a60647@posting.google.com> <crl68.1346$dz5.287185@weber.videotron.net>
Message-ID: <cb4ba455.0202010523.11b52c22@posting.google.com>

Bryan <bryan at eevolved.com> wrote in message news:<crl68.1346$dz5.287185 at weber.videotron.net>...
> Bill Tate wrote:
> 
> > Bryan,
> > Very interesting stuff.  Do you have numbers for Java 1.3.x - just
> > curious to see if there were major gains between 1.3 and 1.4?
> 
> Unfortunately I don't have the time to do so, but I encourage you to. I 
> hope Glyph will post my results on his site, his seem very skewed in light 
> of new python and java developments.

I'll give it shot - I'll post when I get it done.  Some of the Jython
comments are starting to peek my interest as well.
Thanks again for the info.


From jmfbahciv at aol.com  Sat Feb 16 05:47:43 2002
From: jmfbahciv at aol.com (jmfbahciv at aol.com)
Date: Sat, 16 Feb 02 10:47:43 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <140220020703593280%jwbaxter@spamcop.net> <a4j24g$jao$9@bob.news.rcn.net> <3C6DA497.9AB8D624@ev1.net> <D8AB2F5FFDFFBD4F.A40B57999CA5FB13.BD8D4DC731F86177@lp.airnews.net>
Message-ID: <a4ll87$6bt$11@bob.news.rcn.net>

In article 
<D8AB2F5FFDFFBD4F.A40B57999CA5FB13.BD8D4DC731F86177 at lp.airnews.net>,
   claird at starbase.neosoft.com (Cameron Laird) wrote:
>In article <3C6DA497.9AB8D624 at ev1.net>,
>Charles Richmond  <richmond at ev1.net> wrote:
>>jmfbahciv at aol.com wrote:
>>> 
>>>      [snip...]     [snip...]     [snip...]
>>> 
>>> I do have two memoes that I found in my office boxes that were
>>> addressed to PDP-1 users.  I don't know the people who wrote
>>> them (Roland Silver, Edward Fredkin) nor where they were.
>>> 
>>>      [snip...]     [snip...]     [snip...]
>>> 
>>> Well, if you can give me a clue about these memoes that I have, I'd
>>> appreciate it.
>>> 
>>Ed Fredkin is semi-famous. He was mentioned prominantly in David
>>Levy's book _Hackers_. He also invented some data structures that
>>are used today, notably the "trie". 
>			.
>			.
>			.
>Oh, yes.  There's a whole cult around Fredkin
><URL: http://www.digitalphilosophy.org/ >.
>
>It's one to which I'm sympathetic, as it happens.


Thanks.  Now I'll have to arrange a trip to the library to
see if this site has these memoes on-line.  

/BAH

Subtract a hundred and four for e-mail.


From cs1spw at bath.ac.uk  Wed Feb 20 04:17:43 2002
From: cs1spw at bath.ac.uk (Simon Willison)
Date: Wed, 20 Feb 2002 09:17:43 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <G5Bc8.1$Az6.364@news.hananet.net>
Message-ID: <3C7368A0.5050702@bath.ac.uk>

Jonathan Gardner wrote:

> There are security problems. I tried to force the guy to be coming from one 
> unique IP address, but I found that too many places have firewalls or 
> something that change the IP address of the user each time they access the 
> page, so I had to abandon that.

I hadn't realised some places had dynamically changing IP addresses. My 
security method was to save the User Agent, IP address and the time the 
session was created (so I can expire it after X minutes). I suppose you 
could deal with IP addresses by ensuring that the first part of the IP 
address remains the same i.e

User logs on initially as 172.21.16.189
Session then allows access from 172.21.16.* or 172.21.*.*

It's pretty important to provide some kind of checking mechanism like 
that to avoid people copying/pasting the URL of the page they are on 
into an e-mail / instant message and inadvertantly giving their session 
to someone else.

Cheers,

Simon



From mkelly2002NOSPAM at earthlink.net  Sun Feb 17 13:00:03 2002
From: mkelly2002NOSPAM at earthlink.net (Michael Kelly)
Date: Sun, 17 Feb 2002 18:00:03 GMT
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com>
Message-ID: <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com>

On 16 Feb 2002 08:57:15 -0800, rony.steelandt at bucodi.com (Rony) wrote:

>You are definitifly right, i get the same error with 2.1.2 and 1.5.2
>Actually i tested it with C++ , VB, Windev and it works.
>It only doesn't work with Python.


Something I found in Python using COM,
(after Mark pointed it out to me that is)
is that if you have "out" parameters other than
the retval you should use them in multiple assignment.
I wrote a wrapper for Delphi TIniFile that got a similar
error only in Python.  It turned out I was passing a
param to procedures with "out" parameters like:

myIniFile.ReadSection('section', sectionlist)

when I should have done:

sectionlist = myIniFile.ReadSection('section')

In VB or VC++ I would have to pass the 2 params
in the parens as expected.

IOW, since Python has multiple assignment and
VB and VC++ do not, any out param should be
assigned to rather than passed as a param.

Anyway the error was something to do with
a type error so I'm pretty sure you're getting
burnt by the same thing.



--

"I don't want to belong to any club that would have me as a member."
    -- Groucho Marx


From syring at email.com  Thu Feb 21 06:42:23 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 12:42:23 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de> <7x7kp7krjl.fsf@ruckus.brouhaha.com>
Message-ID: <a52ml7$42r9l$1@ID-7529.news.dfncis.de>

"Paul Rubin" <phr-n2002a at nightsong.com> schrieb
> "Karl M. Syring" <syring at email.com> writes:
> > > Not likely.
> >
> > You really should have a look at high precison libraries (e.g.
> > http://www.hipilib.de). They will use rational numbers internally.
>
> Yes, but who actually uses those libraries, and for what?

The author of the Piologie library said somewhere that extended precision
floating point arithmetic is completely useless, but it is fun.  Then there
is the secret message coded into pi ... locate your name in pi
(http://pi.nersc.gov)

Karl M. Syring




From killspam at darwinwars.com  Wed Feb 20 06:41:33 2002
From: killspam at darwinwars.com (Andrew Brown)
Date: Wed, 20 Feb 2002 11:41:33 GMT
Subject: confusion regarding os.path.walk()
References: <mailman.1014141812.1766.python-list@python.org> <Xns91BABAD1C9BE8andrewdarwinwarscom@194.168.222.9> <slrna75bg5.hg.rmunn@rmunn.dyndns.org>
Message-ID: <Xns91BB770F47AD4andrewdarwinwarscom@194.168.222.19>

rmunn at pobox.com (Robin Munn) wrote in
news:slrna75bg5.hg.rmunn at rmunn.dyndns.org: 

> assuming
> burrows.html is the "real" file and Burrows.htm is the symlink you
> want to create, then the command you really want to be running is:
> 
> ln -s burrows.html ./wombats/habitat/Burrows.htm
> 
> This will create the symlink Burrows.htm -> burrows.html in the
> directory wombats/habitat. If you did:
> 
> ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm
> 
> then what you would get would be: in the directory wombats/habitat,
> there would be a symlink named Burrows.htm that tried to reference
> wombats/habitat/wombats/habitat/burrows.html -- this is almost
> certainly not what you wanted.
> 

Yes. This is exactly the gotcha that got me. It's not at all obvious, and 
it's not really a python problem either. It seems to arise from the way 
that Unix makes symlinks: I had assumed that it first looks for the file 
to link to, and then makes the link file. But it turns out to be the 
other way around. So, though I had tried all sorts of path manipulations 
to ensure that the link was found relative to the file linked to, I never 
tried chopping the path of the target file, which would have solved my 
problem.

Thanks for your help, and I hope this sometime helps a passing googler. 
Other people will surely fall into the same pit. 


From eppstein at ics.uci.edu  Wed Feb 27 19:23:01 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 27 Feb 2002 16:23:01 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org>
Message-ID: <eppstein-A2DA01.16230127022002@news.service.uci.edu>

In article <mailman.1014854027.3045.python-list at python.org>,
 James_Althoff at i2.com wrote:

> Below is a revision (not yet submitted) of PEP 276.

What was changed from the previous version?

> Rather than restart the entire debate on PEP 276, please send me
> suggestions for specific changes to the PEP.

Since this seems to be a matter of some confusion, I would like to see it 
much more clearly pointed out that this affects all contexts in which 
iterators are allowed (e.g. "in" operator) rather than just for-loops.

Although I fall into the camp you describe in the PEP of people who want a 
more substantive change to integer iteration, my main objection to this PEP 
does not seem to be addressed: that by blurring the boundary between 
integer and iterator types you add a nonobvious and confusing meaning to 
the expression "x in y" when y is an integer.

In the earlier discussion of this PEP, I made a different argument, that 
this PEP breaks the current property that "for something" always loops over 
exactly the values that make "something" true, but I now see that I was 
incorrect: it is still the case under the PEP that, e.g. "for x in 5" loops 
over exactly the values that make "x in 5" true.  My dislike now stems from 
the ugliness of the expression "x in 5".
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From syring at email.com  Thu Feb 21 03:21:08 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 09:21:08 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com>
Message-ID: <a52arn$4340r$3@ID-7529.news.dfncis.de>

"Paul Rubin" <phr-n2002a at nightsong.com> schrieb im Newsbeitrag
news:7xpu2zdzhh.fsf at ruckus.brouhaha.com...
> "Karl M. Syring" <syring at email.com> writes:
> > But this is really dilettantic work and NASA today is an organization
run by
> > amateurs. If you look at real computational geometry programs like CGAL,
you
> > can see, how it is done.
> > Anyway, it is not necessary to appeal to authorities to get a grasp of
the
> > problem.: Imagine, you have a polygon in the plane and you do a series
of
> > rotations and translations on it. Now, you want to know whether the
position
> > is exactly identical to another polygon. If you do this with floating
point,
> > you will never get an exact result, while in rational coordinates you
always
> > will.
>
> Physical simulations and computational geometry don't have much in common.

Think of simulating a robot arm that does a brain operation on you. I think
you would immediately turn to proper rational number coordinates.

Karl M. Syring




From simon at ping-pong.budig  Tue Feb  5 10:14:48 2002
From: simon at ping-pong.budig (Simon Budig)
Date: 5 Feb 2002 16:14:48 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net>
Message-ID: <3c5ff6e8$1@si-nic.hrz.uni-siegen.de>

Michael Hudson wrote:
> Simon Budig <simon at ping-pong.budig> writes:
[...]
>> BTW: The stupid fixed width limit of SF messed the patch up.
>> Maybe creating an attachment is more useful...
> 
> Well, anyone can see roughly what it does.  If it gets accepted (or
> more likely, noone complains for a couple of weeks), I'll check it in
> from your patch.

Great! If you do this, please make sure, that a similiar patch goes
into python 2.2.

Again: Thanks for your help.

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From phd at phd.pp.ru  Sun Feb 10 05:26:56 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sun, 10 Feb 2002 13:26:56 +0300
Subject: More email module troubles...<sigh>
In-Reply-To: <a448mo.3vsj90f.1@kserver.org>; from usenet@thinkspot.net on Sat, Feb 09, 2002 at 10:43:35PM -0800
References: <a448mo.3vsj90f.1@kserver.org>
Message-ID: <20020210132656.A15596@phd.pp.ru>

Hi!

On Sat, Feb 09, 2002 at 10:43:35PM -0800, Sheila King wrote:
> AttributeError: 'int' object has no attribute 'lower'

   There is a bug in the email package. Barry said it was fixed in the CVS
(I think it is mimelib CVS, not Python CVS).

   BTW, I am going to provide some other patches to the mimelib CVS, so
stay tuned.

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



From martin at v.loewis.de  Tue Feb 26 16:24:01 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 22:24:01 +0100
Subject: [Q] building a fortran extension module
References: <d737306c.0202261146.5dcf6e02@posting.google.com>
Message-ID: <m3n0xwueby.fsf@mira.informatik.hu-berlin.de>

yoonseo at altavista.net (SungPil Yoon) writes:

> Any help will be appreciated.

I suggest that you report the compiler line that was invoked to link
the executable. Most likely, the compiler options are incorrect. Study
cc(1) for further information.

Regards,
Martin



From dickey at saltmine.radix.net  Fri Feb 15 08:07:15 2002
From: dickey at saltmine.radix.net (Thomas Dickey)
Date: 15 Feb 2002 13:07:15 GMT
Subject: Curses & Colors
References: <mailman.1013726151.3247.python-list@python.org>
Message-ID: <a4j163$45q$1@news1.Radix.Net>

Rafael Mentz Aquino <pythonlist at zumnet.com.br> wrote:

> I'm creating some simple interfaces with curses with python under FreeBSD,
> but I'm having some problems to change colors.  I tried to do something in
> /etc/termcap, but when I run curses.can_change_color(), I always obtain 0. 
> My env term is cons25, but I don't know if python uses the same terminal. 
> I'm running python 2.1.1.  Can anybody help me...?  ;-))

can_change_color() applies to terminals that have the ability to change
the content of a particular color, e.g., making blue slightly brighter.
FreeBSD console doesn't do this.

-- 
Thomas E. Dickey <dickey at radix.net> <dickey at herndon4.his.com>
http://dickey.his.com
ftp://dickey.his.com


From tbryan at python.net  Wed Feb 13 22:46:25 2002
From: tbryan at python.net (Tom Bryan)
Date: Thu, 14 Feb 2002 03:46:25 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org>
Message-ID: <lqGa8.138848$a07.38963095@typhoon.southeast.rr.com>

Cliff Wells wrote:

> On Thu, 14 Feb 2002 02:21:03 GMT
> Tom Bryan wrote:

> I don't use IDLE, so I can't answer your question (I use XEmacs). 

Same here. :-)

Our QA department is likely to use IDLE, I would guess.  The development 
team will probably use a mixture of editors (CodeWright, vi, XEmacs, etc.). 
 The developers should have already set their tools to always use tabs 
because of our existing coding style guidelines.  If QA uses IDLE, then 
it'll cause a mess.  If a developer has to use tabs for Java and spaces for 
Python, things could get messy.  (Not in XEmacs, but I'm not sure how good 
some of the other editors are at per language configuration.)

perhaps-we-should-all-just-use-ed-ly y'rs
---Tom



From imbaczek at poczta.fm  Tue Feb 26 13:07:40 2002
From: imbaczek at poczta.fm (Marek Baczek =?iso-8859-2?Q?Baczy=F1ski?=)
Date: Tue, 26 Feb 2002 18:07:40 +0000 (UTC)
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <3C7A66A7.8060503@attbi.com>
Message-ID: <slrn.pl.a7lbtu.3vs69hb.imbaczek@baczek.net.invalid>

Pete Shinners wyprodukowa?[ao]?:

> Donnal Walter wrote:
>> But what is the functional equvalent of:
>> 
>> for x in items:
>>     x.f()
> 
> if all "items" are the same class, you could get away with this,
> 
> map(class_of_x.f, items)

map(type(x).f, items) solves the problem.

Pozdrawiam,
Baczek
-- 
Marek Baczy?ski aka Baczek UIN:57114871 RLU:201791 http://republika.pl/imbaczek
It is better to be a man without a home page than a man with a homepage created
using Microsoft Frontpage.


From QnickQm at alum.mit.edu  Wed Feb 27 12:07:39 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Wed, 27 Feb 2002 17:07:39 GMT
Subject: Equations with Regular Expressions
References: <mailman.1014805905.18241.python-list@python.org>
Message-ID: <slrna7q4ov.1pr.QnickQm@localhost.localdomain>

In article <mailman.1014805905.18241.python-list at python.org>, 
         Alves, Carlos Alberto - Coelce wrote:
> Hi everybody,

Hello there!

> Could someone of you give me and example of how can I translate
> '2x^3-2x^2+15' into '2*(x**3)-2*(x**2)+15' using regular expressions.

I can't tell (from the form of your problem) what kind of inputs you're
expecting.  Do you have to accept:
   1) Polynomials in a single variable?
   2) Arbitrary expressions?
   3) Something else entirely?

Here's the beginnings of a general solution you can use.  It accepts
a parenthesized algebraic expression of single-letter variables, and
turns in into a Python expression.  

It doesn't do _exactly_ what you want: it doesn't put parenthesis
around the ** operator.  (e.g., you asked for "2x^2"--> "2*(x**2)",
but this code gives "2*x**2.") I'm leaving this out because ** has
precedence over * in Python, and so the two expressions above are
equivalent anyway.

This code requires Python 2.0 or higher.

(I wouldn't do it this way in real life; I'd probably use more code to
write a tokenizer instead and try to do real re manipulation. Your
solution asked for a regular expression, though, so that's what you'll
get. :) )

=======================================================================
import re


# Translation is for the most part easy: Just turn all '^' into 
# "**", and insert a "*" between:
#            A number, variable, or close paren followed by an
#             open paren, number, or variable.

#
# This re matches 
#      1) a number when it's followed by a ( or a variable,
#   or 2) A ) or a variable when it's followed by a (, a number, or a variable.
#
boundary_re = re.compile(r'''(
                               (?:\d+(?:\.\d*)?|\.\d+)\s*
                               (?=\(|[A-Za-z])
                            |
                               (?:\)|[A-Za-z])\s*
                               (?=\(|\d+(?:\.\d*)?|\.\d+|[A-Za-z])
                             )''', re.X)

def translate(string):
    s2 = boundary_re.sub(r'\1*', string)
    return s2.replace("^", "**")

if __name__ == '__main__':
    print translate("1")
    print translate("")
    print translate("x")
    print translate("x+y")
    print translate("xy")
    print translate("5x^(3xy)+9x/4")
    print translate('2x^3-2x^2+15')

=======================================================================
Pretty-sure-that-putting-explicit-parenthesis-around-the-exponentiations-
  would-require-a-real-tokenizer-ly yrs,
 						Nick


From steve at ferg.org  Thu Feb 14 15:11:33 2002
From: steve at ferg.org (Stephen Ferg)
Date: 14 Feb 2002 12:11:33 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com>
Message-ID: <b16e4ef7.0202141211.3fb6b1f@posting.google.com>

Here's a bit more about what I was trying to do in EasyGui.  It took a
message from Kevin Altis really to make it clear.  Kevin wrote:

> I went ahead and looked at what you have 
> done so far and it looks like all
> you're doing is throwing up different 
> sorts of dialogs. Are you just wanting
> dialogs for input and output in a linear 
> program? GUI programs are generally
> event driven.

Kevin pinpointed exactly what is different about EasyGui ... it is
just putting up dialogs for simple input and output in a linear
program.

What makes EasyGui different from other GUIs is that it is NOT
event-driven.

The reason is this:  For some (a few? many?) novice Pythonistas (or
even novice programmers?), the event-driven paradigm is new -- it will
take some learning and some mental re-adjusting.  EasyGui is designed
to meet their immediate needs until they can make that transition. 
Until he/she is ready to move on to an event-driven paradigm, EasyGui
will allow a programmer accustomed to a linear programming style to do
simple GUI functions so he/she can be productive (with very basic
tasks) immediately.  When he/she is ready, he/she can move on to an
event-driven style with a more powerful GUI package such as anygui,
PythonCard, Tkinter, wxPython, etc.

Kevin mentions what's been done in EasyGui "so far".   At least in my
vision of it, EasyGui shouldn't go much farther.  EasyGui is there
just to do very simple, very basic stuff.   More elaborate stuff
should be done with more powerful tools.

I honestly don't know if there is a need for such an animal as
EasyGui.  But I think there might be...

-- Steve  (steve at ferg.org)


From vlindberg at verio.net  Tue Feb 19 14:14:39 2002
From: vlindberg at verio.net (VanL)
Date: Tue, 19 Feb 2002 12:14:39 -0700
Subject: Finding nonprintable characters?
Message-ID: <3C72A41F.20908@verio.net>

Hello,

I have a function

isBinary(filehandle)

that I'm not sure how to implement.  I've decided to define binary as 
containing characters above \x80.  But  what is the best way to do this?

1. iterate through xreadline, so the whole thing doesn't get loaded into 
memory?

2. String searching?  If so, for what string?  Searching for anything 
greater  than \x7f?

3. Re searching?  for what class?


Thanks in advance,

Van




From netzapper at magicstar.net  Thu Feb 28 01:30:15 2002
From: netzapper at magicstar.net (A. Jones)
Date: Thu, 28 Feb 2002 06:30:15 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <slrna7rdna.5m1.QnickQm@localhost.localdomain>
Message-ID: <3c7dcc76.261969897@mammoth.usenet-access.com>

On Thu, 28 Feb 2002 04:46:31 GMT, QnickQm at alum.mit.edu (Nick
Mathewson) wrote:

>First of all, this cipher is a bit suspicious.  It has the undesirable
>property that the frequency counts of the characters in the ciphertext
>are the same as in the plaintext.  That is to say, if I know you're
>sending either "attack at dawn" or "run for your lives", it's easy for
>me to tell which one you sent.

Yeah.  I've designed dozens of ciphers before, all of which fail
horribly the second I think about frequency analysis.  On the other
hand, I've never tried actual encryption... I don't quite grok modular
mathematics yet enough to do anything but monkey-see-monkey-do copies
of other opensource ciphers.

>Secondly, this cipher doesn't buy you a lot: The key is as long as the
>plaintext.  If you have a secure place to store the key, you would
>presumably use that place to store the plaintext as well.

This is one of the things I noticed as I played with it (not that I
didn't know it before I started, but it seemed to matter less at the
time).

>Third, the scrambling step is more crucial than you think.  Without
>scrambling, if I see the key-portion of a message, I'll have a pretty
>good idea which parts of the message go where: indices for low letters
>earlier than indices for high ones.  But with scrambling, the quality
>of your cipher is limited by the power of your random number
>generator. 

I figured as much.  But, I don't know enough Python yet to want to
play with the offsets randomly, ya' know?

>For a deeper introduction to cryptography, you may want to get a copy
>of Schneier's _Applied Cryptography_, or read the FAQs from sci.crypt.

Thanks.

>============================================================
>
>Well, I hope this that didn't dampen your entusiasm for Python!
>You're off to a good start (esp. for two days!), and I hope we'll hear
>more of you in the future!

Good Lord, no, it didn't dampen my enthusiasm.  Hell, it only
heightens it knowing that there're people in this community who're
willing to take a look at a total newbie's code and then spend the
time not only commenting but actually *re-writing* it.  Thank you.

Aubrey Jones
______________________________________________________________________________
Posted Via Binaries.net = SPEED+RETENTION+COMPLETION = http://www.binaries.net


From db3l at fitlinxx.com  Fri Feb 22 22:23:34 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 22 Feb 2002 22:23:34 -0500
Subject: Stackless Platform Independence?
References: <mailman.1014420937.15379.python-list@python.org>
Message-ID: <uvgcoq3t5.fsf@ctwd0143.fitlinxx.com>

Tim Peters <tim.one at comcast.net> writes:

> [Christian Tismer]
> > As you might have noticed, Stackless has changed quite
> > much:
> >  - Platform independence is gone.
> 
> [Christopher Browne]
> > I hope not...  I would hope to hear that "platform indepedence has
> > _ARRIVED_"
> 
> No, he meant what he said.  All the good changes are bought at the cost of
> needing some platform-specific assembler, to transform the platform C stack
> from an Insurmountable Obstacle into just another piece of manipulable
> state.

And given that it's a relatively small piece of the overall puzzle, it
need not be much more of a headache to maintain than the various
platform specific pieces inside the Python core are (e.g., the
threading portions).  Probably less, since I expect CPUs and/or things
with impact on the stack switching may change less frequently than
OS/platforms.  In any case, it seems like it's worth the tradeoff,
especially if it makes it Python version independent (or easily
trackable).

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


From s.schwarzer at ndh.net  Wed Feb  6 17:54:27 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Wed, 06 Feb 2002 23:54:27 +0100
Subject: get all paths
References: <d5534716.0202061219.60f2af7e@posting.google.com>
Message-ID: <3C61B423.8F614038@ndh.net>

Hello Ed

Ed wrote:
> Hi
> I'm trying to create a treeview widget, but i am having problems
> getting alll directories and files from a specified directory

I recommend to look at os.path.walk :-)

Stefan


From clarence at netlojix.com  Sun Feb 10 05:12:33 2002
From: clarence at netlojix.com (Clarence Gardner)
Date: Sun, 10 Feb 2002 10:12:33 GMT
Subject: Confusion with __del__
References: <2jq98.54$Mk5.2196@newsfeed.avtel.net>
Message-ID: <lIr98.55$Mk5.2188@newsfeed.avtel.net>

In article <2jq98.54$Mk5.2196 at newsfeed.avtel.net>,
	clarence at netlojix.com (Clarence Gardner) writes:
> It seems to me that what should happen is:
>     __init__ raises an exception
>     __del__ gets called; it checks whether the "db" attribute exists;
>          it doesn't; object is now deleted
> Why does it call __del__ (or close) a second (or millionth) time? 

Mea culpa. The class has __getattr__ where I meant to put __getitem__.

The separate thread about pair programming, arguing about what the
factor 'k' (denoting productivity improvement of P1 given P2 helping)
doesn't take into account the savings on broken furniture, etc., caused
by the solitary P1 :(

-- 
Clarence Gardner
Software Engineer
NetLojix Communications
clarence at netlojix.com


From moro at dotcom.ee  Fri Feb 15 01:51:36 2002
From: moro at dotcom.ee (Martin Roos)
Date: Fri, 15 Feb 2002 08:51:36 +0200
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <3C6CAFF8.4D3B5092@dotcom.ee>

Scirious wrote:

> Dear friends;
>
> I've been learning JAVA for a while, but now I've heard avout Python.
> Any coments concerning these two languages?
>
> TIA;
> Scirious.

java is powerful and quite fast with jit,
python is more fun and slower ....

 That's it ...

 ps. java has more api's (till now)

i like both of them ...
 java is for serious kinda stuff , python is for testing etc.





From mhammond at skippinet.com.au  Fri Feb 15 20:33:49 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 16 Feb 2002 01:33:49 GMT
Subject: win32com and Passing SafeArrays of Doubles
References: <mailman.1013814685.21004.python-list@python.org>
Message-ID: <3C6DB769.9060906@skippinet.com.au>

I believe it will simply be a matter of getting the correct "type tuple" 
in the generated code.  I am afraid I don't know VB well enough to know 
what the extra set of parens used by VB is.  It could be some special 
"byref" annotation, or maybe even specifying another level of 
indirection (ie, an array of arrays of doubles).

I suggest you try and determine exactly what that VB syntax does - you 
may then need to craft a type typle that indicates a 2 dimensional 
array, and actually pass ((1,2,3),) in your Python code.

Good luck, and feel free to mail if you need more help - getting this 
sorted would be good - we may even identify that pythoncom is being less 
then optimal <wink> in these edge cases.

Mark.

Jeffrey N. Shelton wrote:
> In the process of using win32com to hook Python into SolidWorks (a 3D
> modeler), I've run into the same problem as others before me -- namely, some
> strange things happen when passing SafeArrays of doubles into SolidWorks.
> Having created the appropriate gen_py file, I'm usually successful when I
> pass in a tuple with the appropriate values. But when passing SafeArrays of
> doubles into SolidWorks, it thinks I've inserted an extra value of 0.0
> between each parameter I've passed. So it thinks that an array (1, 1, 1) is
> instead (0, 1, 0, 1, 0, 1).
> 
> Past discussion at:
> http://groups.google.com/groups?hl=en&selm=39943819.65368645%40news.onlineis
> p.com
> 
> Mark Hammond's reply (back in August of 2000) indicated that he thought the
> problem was with the SolidWorks code, and I'm quite willing to accept that.
> However, I've noted from the SolidWorks API documentation (quoted below)
> that, even when using VB, some special massaging is required.
> 
> So, do I have to do any special 'wrapping' to get SolidWorks to accept my
> arrays, or does this still look like its just a shortcoming in the
> SolidWorks implementation?
> 
> Any suggestions would be appreciated!
> 
> Jeff
> 
>>From SolidWorks API Documentation:
> 
>>Passing array data to SolidWorks in Visual Basic requires the use of a
>>Variant variable to hold the data. The procedure is to put the data
>>into an array then to assign a Variant to the array. The variant is then
>>passed to an API function with parentheses round it:
>>
>>     Dim varArray As Variant
>>     Dim DataArray(9) As double
>>
>>     ' Assign values
>>     DataArray(0) = 0.1
>>     ...
>>
>>     ' Pass the data to SolidWorks
>>     varArray = DataArray
>>     Result = Object.Method( (varArray) )  ' Note the extra parentheses
>>
> are required
> 
> 
> 




From jason at jorendorff.com  Thu Feb 28 00:07:58 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 27 Feb 2002 23:07:58 -0600
Subject: How to have multi-threaded server log to one file ?
In-Reply-To: <97ae44ee.0202272013.2094b2b@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEOGDFAA.jason@jorendorff.com>

Stephen wrote:
> [...] how do I get the same Queue object into both MyHandler and
> MyServer in the first place ? [...] Or is this somewhere that a
> global variable Queue object should be used ?

Try making a separate module that says:

---------------------------------------------------------------
# logging.py - used by everyone else that needs to log messages

import Queue

# This is the queue object that's used to hold log events.
# LoggingThread knows about it, and the log() function below
# knows about it.  No one else is allowed to see it.
_log_queue = Queue.Queue()

class LoggingThread:
    ...
    def run(self):
        while 1:
            message = _log_queue.get()
            self.logfile.write(message + '\n')  # (or whatever)

# Kick off the logging thread.
LoggingThread().start()

def log(msg):
    _log_queue.put(msg)

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

Then your MyHandler and MyServer code can just use the log()
function, like this:

  from logging import log
  ...
  ...
  log("Something happened")
  ...

## Jason Orendorff    http://www.jorendorff.com/



From stephan.schumacher at andra.fr  Wed Feb 27 14:45:02 2002
From: stephan.schumacher at andra.fr (Stephan)
Date: Wed, 27 Feb 2002 20:45:02 +0100
Subject: writing efficiently a numpy array
Message-ID: <MPG.16e755abd178b85e989682@news.wanadoo.fr>

Hi,

I want to write a large numpy array on an ascii file. There must be N 
numbers per line (except for the last line...). 

Does anyone knows something more efficient than the function below :

from Numeric import *
from string import *

def writeArray(file, my_array, N) :
    my_string_array=map(repr,my_array.tolist())
    tmp = []
    ind = 0
    for i in my_string_array :
    	tmp.append(i)
	if ind == N-1 :
       	tmp.append('\n')
	ind = (ind + 1)%N
    file.write(join(tmp," "))


# exemple : 

a = arange(300000)
file = open("exemple","w")
writeArray(file, a, 8)
file.close()






From gh_pythonlist at gmx.de  Mon Feb 18 10:34:52 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 16:34:52 +0100
Subject: Subclassing a read-only class
Message-ID: <20020218153451.GA7162@lilith.hqd-internal>

Hello,

For compatibility testing, I'd like to put an additional method into
window object I get from the curses module. Unfortunately, if I try to
do so, I get:

    TypeError: object has read-only attributes

I'm pretty sure there is a solution, probably with a proxy object. But
currently I just don't get it.

Any ideas?

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 8.1 ?C      Wind: 3.7 m/s



From db3l at fitlinxx.com  Wed Feb 13 19:25:34 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Feb 2002 19:25:34 -0500
Subject: Is making Python executables a reality?
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>
Message-ID: <ubses9af5.fsf@ctwd0143.fitlinxx.com>

"MDK" <mdk at mdk.com> writes:

> What percentage of people make exe files?  Do most people have end-users
> download the Python environment?
> 
> What is everybody doing in this regard?

I definitely depend on being able to distribute some of our
applications as standard Windows installation programs without any
requirement for Python on the target system.

We use Gordan McMillan's installer (but others use py2exe) to package
the application and required files, and then use InnoSetup to build
the Windows installer.  Works great.

Perhaps if you posted some more specifics about what didn't work for
you with py2exe someone might be able to suggest a solution.

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


From gmcm at hypernet.com  Thu Feb 14 08:54:26 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Feb 2002 13:54:26 GMT
Subject: Is making Python executables a reality?
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>
Message-ID: <Xns91B55AA48E7DCgmcmhypernetcom@199.171.54.215>

MDK wrote:

> What percentage of people make exe files?  Do most people have
> end-users download the Python environment?

I can't even guess about percentages. Lately, Installer has
been downloading at a rate of around 200 per week (about
50 or those for Linux / Unix). The mailling list has a few
dozen subscribers. I have over 1,000 private emails on
Installer[1]. Posts on c.l.py for py2exe seem about on par
with Installer posts. So, as a WAG, I'd say 10K people
making exe's isn't unreasonable (it's certainly thousands).

-- Gordon
http://www.mcmillan-inc.com/

[1] Another project of mine downloads at about 15 / week,
and I have never received an email from a user.


From afonso at writeme.com  Tue Feb 12 11:38:54 2002
From: afonso at writeme.com (Afonso Fernandez Nogueira)
Date: Tue, 12 Feb 2002 17:38:54 +0100
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <Pine.LNX.4.33.0202101938190.17089-100000@ns0.tbye.com>
Message-ID: <00a301c1b3e8$2225ed20$e6e9533e@fonso>

Dave wrote:
>Could be from a number of languages; I always assumed quasi-Esperanto:

>"-ist" - an individual professionally or avocationally occupied with the
>idea or activity defined by the root [1].

>Of course, in either Spanish or Esperanto it would more commonly be
>"Pythonisto" anyway, so who knows.

Esperanto tries to be phonetically written, and usually adapts nouns to its phonetics / own vocabulary. So, if Python is called in Esperanto "(la programlingvo) Pitono", I would expect "Pitonisto" ("Pitonistino" if female, "gepitonistoj" to refer to both genders).

Anyway, if you take "Pythonista" as "member of the Python community", then "Pitonano" would perhaps be more accurate. Esperantists use to start their post with "Koraj gesamideanoj", ie "Dear members (of both genders) of the same idea" or whatever it translates to ;). 

Esperantists always use the -ist- suffix ("Esperantistoj") to refer to themselves though. I've seen several webpages about Python written in Esperanto, so I'm sure there has to be an established way to call Pythonistas in this beautiful language.

Cheers,
Fonso




From martin at v.loewis.de  Wed Feb 27 04:19:07 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 10:19:07 +0100
Subject: Thread support in NetBSD
References: <a5i40k$1k59$1@news.uit.no>
Message-ID: <m3lmdf1dv8.fsf@mira.informatik.hu-berlin.de>

"Anders ?berg" <andersaa at stud.cs.uit.no> writes:

> In the pthread directory, there are three subdirectories: bin, include and
> lib. Should I have passed one of these as parameter to configure, instead
> of the parent directory?

I think this kind of setup is not supported. The directory it asks for
is the library directory (it goes into the compiler line as a -L
option), but there is no provision for locating header files in a
different place.

> Anyway, it doesn't work and I am stuck now. Does anybody know how
> to fix this?

I'm not sure what MIT threads are, but unless they closely resemble
Posix threads, you are out of luck. In that case, you'd first need to
port Python to MIT threads.

If the pthreads module is suited for your thread library, you can try
setting CPPFLAGS and LDFLAGS before invoking configure. If that fails,
you need to adjust configure.in.

Regards,
Martin


From pinard at iro.umontreal.ca  Wed Feb 20 14:52:52 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 20 Feb 2002 14:52:52 -0500
Subject: Python praise [was Re: two ideoms at one blow ...]
In-Reply-To: <15475.50237.602572.554832@desk.crynwr.com>
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
	<oq8z9o2mx6.fsf@carouge.sram.qc.ca>
	<15475.50237.602572.554832@desk.crynwr.com>
Message-ID: <oq1yfg2anf.fsf_-_@carouge.sram.qc.ca>

[Russell Nelson]

> Hello, Fran?ois.  Yes, I discovered Python about a year ago [...] I thought
> that was a great idea ... and then much later I discovered Python using
> those rules.

I wish you will be happy with Python.  For one, I'm very, very satisfied,
to the point I lost much of my interest for C programming (and away from
all those heavy portability efforts that plague most of my previous works).

Python is excellent for quickly writing prototypes, and moreover, for reading
them back while revisiting them later.  This is especially important for
code that you (or I) write quickly, because you (I) have less time than
usual for learning what we do while doing it, per our previous habits.
When I was writing a lot of assembler, I was meditating for weeks on the
code and the project.  When one writes equivalent code in one hour or so,
one surely looses something.  To counter-balance, Python could be written
very legibly all along.  Of course, a minimum of comments always help. :-)

And you know, prototypes that work well enough often turn into production
programs, that stay un-rewritten forever, and with everybody staying happy.
If arises a special need for extra speed, a few profiling runs, followed
by the writing of a specialised C extension, is often the good way to go.

There are many Python libraries (or modules, or packages) already, they
may not be as numerous as in some other languages, but on the other hand,
many of them are extremely well designed, and are pure pleasure to use.  In a
way, my feeling is that in Python, there is less garbage to chose from :-).

Yesterday, I was playing with PIL (Python Imaging Library) and PyFT (Python
Freetype interface) in various ways, and found astonishing how easily one can
get a lot of quality out of relatively short Python programs (also provided,
in this case, that both packages allow easy handling of alpha channels).
This has been my experience all along with Python, and in various fields.
Lot of results for modest investments in code which always stay legible.

I somewhat regret having discovered something like Python so late in my
programmer life.  On the other hand, I guess that younger people, lacking
experience, may not fully appreciate the worth of the tool given to them...

In a word, Nelson, welcome in the gang! :-)

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




From suca at manning.com  Wed Feb 27 15:58:12 2002
From: suca at manning.com (Susan Capparelle)
Date: Wed, 27 Feb 2002 15:58:12 -0500
Subject: Seeking Python book authors
Message-ID: <v04220821b8a2f88c019f@[192.168.1.101]>

Hi,

We're a publisher of high quality computer books currently seeking 
authors for a number of Python related books.  A couple of titles or 
topics would be; 'Enterprise system development with Python,' 
'Practical Python' and 'Effective Python.'  If you can recommend 
anyone with the necessary experience and skills to undertake any of 
these books, please feel free to contact us.

Sincerely,

=======================================
Susan W. Capparelle
Assistant Publisher
Manning Publications Co.
209 Bruce Park Avenue, Greenwich, CT 06830
suca at manning.com
tel. 203.629.2211   www.manning.com
fax. 203.629.2084
=======================================



From schneida at seiu.org  Wed Feb 27 16:15:55 2002
From: schneida at seiu.org (Anders Schneiderman)
Date: 27 Feb 2002 13:15:55 -0800
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]> <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>
Message-ID: <54abe19d.0202271315.4e699ed0@posting.google.com>

Fernando Rodr?uez <frr at ThouShallNotSpam.EasyJob.NET> wrote in message news:<a0up7usn1v48q14i1eopb5n8mfdthfpqkm at 4ax.com>...

> If you want to learn how to program, then I wouldn't recommend using Python.
> Python _is_ a great language for beginners, but there many _great_ (and I mean
> great) books for beginners that use it.

I'd agree that there aren't any fabulous books for learning Python as
a beginner--nothing as fun as, say, "Oh Pascal!"--but there is plenty
of intro material, some of it written for and used by high school
kids, so I don't think you'd have a hard time picking it up.  Python
is very easy to learn, and you can do a lot of neat stuff w/o knowing
the language inside-out.

If in addition to getting books/materials on Python you also want some
general books on becoming a better programmer, you should definitely
ask folks on the list.  My vote would be for _Code Complete_ right now
and _The Pragmatic Programmer_ in six months.  If anyone has a
recommendation for a good book on object-oriented programming, that'd
be good to read too; I've never found one that was deep and was aimed
at beginners.

> I recommend you buy a copy of this book: "The Schemer's Guide" at
> www.schemers.com. It's an excellent, short and gentle (let me stress the word
> _gentle_) intro to programming and computer science. It uses Scheme, a
> language designed for teaching, that is quite similar to Python. IMHO, this is
> the most painless way to quickly become a viable programmer.

Scheme is "quite similar to Python"???  Scheme is a dialect of Lisp,
no?  Perhaps at a deeper level, but for a beginner, they sure look
different to me.  For example, Lisp uses parantheses hell vs. Python's
elegant approach of using white space/indentation.

Also, I've heard Lisp described many ways--powerful, etc.--but
painless ain't one of them.  I went from Basic & Pascal to Lisp, and
it was _not_ a pleasant experience.

I'd choose Python over Java; it's easier to learn, and it's a much
less wordy language.  Every once in a while I'll see something new in
the Java world I want to check out, and every time I do, I'm reminded
of how much I like Python.  Java might be a good second language.

One more thing.  One of the reason it's easy to learn Python is that
when you run into a really tricky problem you can't solve, if you post
it to the newsgroup, someone will give you an answer.  I just posted
something an hour ago, and already a really nice guy sent me the
answer!

Just my two cents.

Thanks,
Anders Schneiderman


From sjmachin at lexicon.net  Wed Feb 13 21:57:55 2002
From: sjmachin at lexicon.net (John Machin)
Date: 13 Feb 2002 18:57:55 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com> <7x1yfq2gdf.fsf@ruckus.brouhaha.com> <c76ff6fc.0202130114.6c1ab5ae@posting.google.com> <3C6AA0C0.CA9FFAA5@ford.com>
Message-ID: <c76ff6fc.0202131857.4a42aeda@posting.google.com>

Jeff Henry <jhenry19 at ford.com> wrote in message news:<3C6AA0C0.CA9FFAA5 at ford.com>...
> John Machin wrote:
> > 
>  [snip]
> > pending bag ... using a regex backwards to find the opening tag is so
> > innovative that I'm totally gobsmacked.
> > 
> I love this group! Yesterday I learned "wackyparse", and today
> "gobsmack"!
> 
> [snip]
> > 
> > Looks like the effbot gets to be gobsmacked too.
> > 
>  Ah, but will he be wackyparsed? Inquiring minds want to know! :)

I would have said, no, not much chance, but look at this, found at the
bottom of a recent effbot post in the place of his normal </F>
signature:

Traceback (most recent call last):
  File "effbot.py", line 105, in ?
  File "SimpleXMLWriter.py", line 103, in end
    assert self.__tags, "unbalanced end(%s)" % tag
AssertionError: unbalanced end(F)

Looks like effbot implemented back-ass-ward regexing for opening HTML
tags already and has been hoist by his (Paul's?) own petard. Can our
hero recover? Will he take the principled approach and insert <F> at
the start of his posts or will he remove the assertion? Tune in
tomorrow for the next exciting episode!


From mudrunner5 at hotmail.com  Sun Feb  3 00:36:38 2002
From: mudrunner5 at hotmail.com (slipknot)
Date: Sun, 03 Feb 2002 05:36:38 GMT
Subject: newbie.py
Message-ID: <G%378.488$1e.299052@news2.news.adelphia.net>

I heard of the computer programming language called Python on Jan26th for
the very first time.. it was mentioned just in passing in one of my
undergrad classes at UMaine.. so i looked it up on the web, found a few
tutorials, and homebase (python.org)...

I'm excited about coding once again.. i've been getting rather bored lately
with programming in general, but python renewed my interests.  I can't wait
to read more about it, i went out to Borders bookstore friday and bought a
"Complete Reference" book..

I've also decided to choose it as my "study language" for another class i'm
taking.. (ProgrammingLanguages) which focuses on design and features.

Long story short, I'll have plenty of questions for you guys over the next
few months..  :-)

~DM




From martin at v.loewis.de  Thu Feb 28 17:43:41 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 28 Feb 2002 23:43:41 +0100
Subject: Simple threading thing
References: <Rlsf8.136$5vh.170729984@news.telia.no> <j4vgchebcy.fsf@informatik.hu-berlin.de> <Wcwf8.1866$eJ6.34464@news2.ulv.nextra.no>
Message-ID: <m3u1s19qhu.fsf@mira.informatik.hu-berlin.de>

"Thomas Weholt" <thomas at weholt.org> writes:

> I need a thread, no forking. Should run in Linux and Win32. I need a thread
> that functions in the background, running beside the main server, with a
> reference to the main server so it can maintain states etc. in it. The
> purpose of the process is to mainly remove expired sessions ( remember this
> is a webserver ), log certain values and monitor the server, and work as
> sort of a janitor.
> 
> Does that make it any clearer?

Not really. The main process starts. What happens next? I interpret
your posting as

- creates a new thread, to observe the main thread
- the main thread should continue to run, but doesn't

Is that correct? How do you know it does not run?

If this are all the fact, it is really hard to believe. There must be
details that you've omitted.

Regards,
Martin


From mdk at mdk.com  Wed Feb  6 09:21:41 2002
From: mdk at mdk.com (MDK)
Date: Wed, 6 Feb 2002 09:21:41 -0500
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <mailman.1013003651.8981.python-list@python.org>
Message-ID: <a3re5o$1ad8nu$1@ID-98166.news.dfncis.de>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message
news:mailman.1013003651.8981.python-list at python.org...
> ----- Original Message -----
> From: "MDK" <mdk at mdk.com>
>
>
> > I need to convert a list into a list of characters.
> >
> > For example:
> >
> > ("cat",5,['dog',[3,3,1]],"zoo")
>
> That's not a "list", it's a tuple containing strings, ints, and lists;
> further, you have nested lists.  Ouch.
>
> > Would become:
> >
> > ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
> >
> > Any information would be appreciated.
>
> This is doable, but there is no Python builtin I am aware of that can
> do it.  Why do you need to do this?  If we understand what you are trying
> to do, we may be able to steer you away from this mess.
>
>

This is for a project where Python is embeded in a piece of hardware and in
order to move the data around it must be "flattened".  After I get the list
of characters I am going to convert it to ASCII.




From mojo at geofinity.com  Tue Feb 12 22:10:48 2002
From: mojo at geofinity.com (Mojo)
Date: Tue, 12 Feb 2002 22:10:48 -0500
Subject: OOP Newbie mixing COM, wxPython, BOA
References: <u6ddfmsu6ug32@corp.supernews.com> <3C687007.4ACF94EE@ccvcorp.com>
Message-ID: <u6jm9nfgjhr905@corp.supernews.com>

"Jeff Shannon" <jeff at ccvcorp.com> wrote in message
news:3C687007.4ACF94EE at ccvcorp.com...
<snip>
> What you should do, is instead of making the Jabber object be global to
your
> VpFrame module, you should encapsulate its creation into a function.  Have
your
> frame call this function in its __init__(), and then save the reference on
the
> frame.  At this point, you will also want to stick a reference to the
frame onto
> the Jabber object.

Yep, that did it.   I kept thinking of the Jabber Object as a literal COM
object and immutable (interfaces as API).  It might never have dawned on me
that it was actually a python object, and that I could add an attribute at
runtime.  Once my head stopped spinning, it made perfect sense.  Thanks a
bunch!

<snip>
>... In fact, you might
> look at defining custom wxPython command events, and having the Jabber
object
> notify the GUI by posting those events. (This is a somewhat advanced
subject,

I've got some basic functionality to test first, but that is exactly what I
want to try next.

>
> Also, by having both objects holding references to each other, we've
created a
> reference cycle.  Python's garbage collector should be able to handle this
okay,
> but just in case, it's probably a good idea to explicitly destroy the
Jabber
> object in your frame's OnCloseWindow() handler (or whatever name is given
to the
> method that shuts down your frame).  But it's a common idiom in wxPython
for
> parent and child windows to each have references to the other.
(Typically, the
> wxWindows framework will clean up child windows for you, but obviously
wxWindows
> knows nothing of your Jabber object...)
>

Will do.  I appreciate the advice and explanations.  Now, I'm off to make a
bigger mess...

Thanks again,

David
---
--
-





From cfelling at iae.nl  Wed Feb 27 22:28:37 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 04:28:37 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch>
Message-ID: <a5k855$d97$1@animus.fel.iae.nl>

Samuele Pedroni <pedronis at bluewin.ch> wrote:
> Maybe this is a useful insight:

> in mathematetics it is not uncommon to
> identify the natural numbers with the set
> of their predecessors

> 0 = {}
> 1 = {0} = {{}}
> 2 = {0, 1}  {{},{{}}}
> etc

I can immagine a mapping from the natural numbers to such sets, but I
guess you really mean they are thought of as being identical. Right?
And I guess those sets are ordered too, or is that just an extra
you added to make it work?

> so having
>   for i in 5
> iterating over 0,1,2,3,4

> is natural in mathematical sense

> (which does not mean automatically that it is newbie friendly <wink>)

Not only newbie unfriedly, you need a firm grasp of modern set based
number theory to appreciate it, it seems.

> because 5 as an ordered set is the sequence (0,1,2,3,4)

> so also the notation   1 in 5  is natural.

> regards, Samuele Pedroni.


-- 
groetjes, carel


From sjmachin at lexicon.net  Thu Feb 21 15:11:30 2002
From: sjmachin at lexicon.net (John Machin)
Date: 21 Feb 2002 12:11:30 -0800
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk> <mailman.1014230314.2890.python-list@python.org> <c76ff6fc.0202201519.509ec73e@posting.google.com> <20020221.144033.1012502954.7349@ic.ac.uk>
Message-ID: <c76ff6fc.0202211211.4f33f947@posting.google.com>

"Phil Mayers" <p.mayers at ic.ac.uk> wrote in message news:<20020221.144033.1012502954.7349 at ic.ac.uk>...
> In article <c76ff6fc.0202201519.509ec73e at posting.google.com>, "John
> Machin" <sjmachin at lexicon.net> wrote:
> 
> > Gustavo Niemeyer <niemeyer at conectiva.com> wrote in message
> > 
> > It is possible that the OP wants, but is not looking very hard for,
> > this:
> > 
> > [from Python 2.2 source, in include/longobject.h] """
> > /* _PyLong_FromByteArray:  View the n unsigned bytes as a binary integer
> > in
> >    base 256, and return a Python long with the same numeric value. If n
> >    is 0, the integer is 0.  Else:
> >    If little_endian is 1/true, bytes[n-1] is the MSB and bytes[0] the
> >    LSB; else (little_endian is 0/false) bytes[0] is the MSB and
> >    bytes[n-1] the LSB.
> >    If is_signed is 0/false, view the bytes as a non-negative integer. If
> >    is_signed is 1/true, view the bytes as a 2's-complement integer,
> >    non-negative if bit 0x80 of the MSB is clear, negative if set. Error
> >    returns:
> >    + Return NULL with the appropriate exception set if there's not
> >      enough memory to create the Python long.
> > */
> > extern DL_IMPORT(PyObject *) _PyLong_FromByteArray(
> >         const unsigned char* bytes, size_t n, int little_endian, int
> >         is_signed);
> > """
> 
> Ah, I should probably point out I am restricted to using Python 1.5.2 for
> various dull reasons, but in theory that is what I'm looking for.

So why not grab the source of _PyLong_FromByteArray() from the 2.2
distribution and put in your own code?

> However, I note that API isn't listed in the C API documentation.

So raise a doco enhancement request. Adequate comments in the .h file
of a module would be good enough for me.

> 
> I'll probably try the PyLong_FromString, and see if that helps.

Not sure how it could help, given your "in theory that is what I'm
looking for" above.
char string_513[] = "513";
char byte_array_513_little_endian[] = {1, 2};
"String" means external character representation. You would have to
convert your data to external format yourself.

BTW, how long are your own-format long numbers, what's the format, and
where are you getting them from in the first place?


From g_braad at survion.com  Sun Feb  3 10:49:54 2002
From: g_braad at survion.com (Gerard Braad)
Date: 03 Feb 2002 16:49:54 +0100
Subject: Xython - XML-Formed Python
Message-ID: <1012751401.4431.13.camel@acropolis>

Hello,


For a long time now, I had the idea of XML-Formed Python. This way it
would be possible to send Python code across a network connection. The
only problem was, I didn't had much time to realise this. Way back in
August I saw someone (mlh) had the same idea and in December the project
page was included on the DDJ Python-URL. JEEJ!

I wasn't aware of this and hadn't worked on it, until now... Work on the
initial draft has started. If you are willing to help, don't hesitate
and contact us...

  //sourceforge.net/projects/xython/
  //hetland.org/python/xython.php
  //proj.survion.com/?page=xython

Any help would be appreciated. We hope this will become a standard for
inclusion of Python in XML documents.


Gerard Braad          &   Magnus Lie Hetland
g_braad at survion.com       magnus at hetland.org






From tim.one at home.com  Thu Feb  7 23:06:08 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 7 Feb 2002 23:06:08 -0500
Subject: Please stop responding to Rue
In-Reply-To: <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGKNKAA.tim.one@home.com>

[Donn Cave]
> So is sci.math choked with posts from people who normally seem
> to have their heads screwed on, endlessly replying this guy as
> if they sincerely think there's some purpose in it?  People who
> take time several times every day to post about how useless his
> ideas are and how they have no interest in helping?  Or are
> they saner there than we are?

On whole, no; neither in sci.physics, or even sci.logic (in fact, if you
think about it, sci.logic attracts more than its share for a reason, and on
both sides <wink>).

indulging-an-urge-to-cruelty-is-anti-pythonic-ly y'rs  - tim




From SBrunning at trisystems.co.uk  Thu Feb 21 04:17:23 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 09:17:23 -0000
Subject: Any Interest in EDIFACT Parser/Generator?
Message-ID: <31575A892FF6D1118F5800600846864DCBCCCD@intrepid>

> From:	Jim Kraai [SMTP:jkraai at paypal.com]
> Just curious.
 
Not just at the moment, but six months ago, this would have been *dead*
handy.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From paul at spondo.com  Tue Feb 19 04:43:16 2002
From: paul at spondo.com (Paul Berg)
Date: Tue, 19 Feb 2002 19:43:16 +1000
Subject: newbie here...
Message-ID: <001c01c1b929$ddb4dce0$4d00000a@david>

Hello all,

Im just new to Python and don't know anything about it. I've been programming in Miva Script if anyone has heard of it and also a little Php.

Could someone send me some script so i can so  i can relate to it. And also things like databases types it can use(if so), Small things like that. Im known to be a quick learner so hopefully will be programming in days.

Paul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020219/aad510b5/attachment.html>

From bosahv at netscapenospam.net  Fri Feb  1 07:38:03 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Fri, 01 Feb 2002 12:38:03 GMT
Subject: Is thera a Metakit for Python 2.2  release ??
References: <11e94203.0201311433.519b556@posting.google.com>
Message-ID: <L_v68.295688$pa1.74188709@news3.rdc1.on.home.com>

This is really off topic but I'm very curious about 2nd source opinion on
Metakit as it looks like a great package.

Would you mind saying what sort of thing you do with it? Any details would
be most helpful?

Are you running it on Windows?




From mickey at tm.informatik.uni-frankfurt.de  Sun Feb 10 04:32:12 2002
From: mickey at tm.informatik.uni-frankfurt.de (Michael 'Mickey' Lauer)
Date: 10 Feb 2002 11:32:12 +0200
Subject: Python IDEs
References: <zfC08.15504$Hg7.1623775@news11-gui.server.ntli.net> <a1uv39$o79$1@rex.ip-plus.net>
Message-ID: <3c664c2b@nntp.server.uni-frankfurt.de>

F. GEIGER <fgeiger at datec.at> wrote:
> 
> "Stewart hector" <stewart.hector at excite.co.uk> schrieb im Newsbeitrag
> news:zfC08.15504$Hg7.1623775 at news11-gui.server.ntli.net...
>> Hi
>>
>> Apart from IDLE, what other good opensource Python IDEs are there that run
>> under Linux?

IMHO there is none. Each of the available IDEs incorporate a few good ideas,
but there is none with all or at least a majority of good features. What has
to go in an IDE and what not is unfortunately a controversial subject.

>From my point of view, I'd like to have the time to add Python capabilities to
kdevelop, which seems the most promising open source IDE around.

Yours,

:M:



From quinn at hurl.ugcs.caltech.edu  Sun Feb 10 16:45:02 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 21:45:02 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com>
Message-ID: <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu>

On 10 Feb 2002 00:22:57 -0800, wooks <wookiz at hotmail.com> wrote:
>So they are just syntatical sugar. That description would immediately
>explain their raison d'etre. Shame it wasn't used. Anyway now you are
>saying that my understanding of polymorphism isn't what it was so it
>looks like I need to go and reread. I think I ought to understand it
>as it is a fundamental OO concept which I would like to be able to
>recognise when I come across it.

Hmm, what I was trying to say is that you don't *need* an understanding of
polymorphism.  Your understanding may be perfectly good, but if you spend a lot
of time thinking about it, you may be thinking too hard.

Polymorphism has traditionally been a big deal for static languages, because
their compile time typechecking machinery needs to be powerful enough to
generate polymorphic signatures.  Dynamic languages don't have compile time
typechecking machinery, so it's not an issue.

My understanding of polymorphism is (and I've never taken a CS class so who
knows) that it's when a function can operate on more than one type.  Very
simple, and orthogonal to OO.  For example, python's '+' works on numbers
and strings, so it's polymorphic.  If you have 'sum(a)' which uses '+' to add
up the elements of a list, it will also work to concatenate a list of strings.

For that matter, '+' works on both floats and ints in C, so it's polymorphic
there too.  But if you wanted to *write* a function like '+' in C, it would
want you to declare the types of the arguments, at which point you'd be stuck
between saying "add() takes two ints" and "add() takes two floats" (in the C
case, ints have built in magic that let them automatically turn into floats,
but that's just for numbers (and I assume to make C's lack of polymorphism more
bearable) and not all static languages do automatic conversion like that (e.g.
in haskell "let {x=1::Int; y=2.0} in x+y" is an error)) so you can see why
static languages make a ruckus about supporting (or not) polymorphism, while
dynamic programmers might not even know what it is, but use it all the time
anyway.


From sassman at nucentrix.net  Sat Feb  9 22:49:59 2002
From: sassman at nucentrix.net (Mark Sass)
Date: Sat, 9 Feb 2002 21:49:59 -0600
Subject: HTML Generators
Message-ID: <I5m98.1105$dy4.39461@eagle.america.net>

Are there any HTML generator modules for Python?  I just want to create
static html and not use scripts i.e. CGI to do it.  Any available?

Thanks,
    Mark




From alien275 at libertysurf.fr  Wed Feb 27 12:34:02 2002
From: alien275 at libertysurf.fr (Riss Nicolas)
Date: Wed, 27 Feb 2002 18:34:02 +0100
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de>
Message-ID: <a5j53i$oms$1@wanadoo.fr>

I jave used it but when y tried to connect it makes an error:
   File "c:\test.py", line 9, in ?
    tn = telnetlib.Telnet("192.168.0.9",32)
  File "C:\PROGRA~1\python22\lib\telnetlib.py", line 190, in __init__
    self.open(host, port)
  File "C:\PROGRA~1\python22\lib\telnetlib.py", line 219, in open
    raise socket.error, msg
socket.error: (10061, 'Connection refused')

Programm:

import getpass
import sys
import telnetlib
password = getpass.getpass()
tn = telnetlib.Telnet("192.168.0.9",32)
tn.read_until("Administrator Password : ")
tn.write(password + "\n")
tn.write("session\n")


Where is the mistake ?

Thanks
"Emile van Sebille" <emile at fenx.com> a ?crit dans le message de news:
a5iuq9$7701i$1 at ID-11957.news.dfncis.de...
>
> "Riss Nicolas"
> > Hello,
> > I must do a program who can connect with telnet. But I don't know how
> to
> > make a such connection with python, can somebody help me please.
> >
>
> Start with telnetlib
>
> --
>
> Emile van Sebille
> emile at fenx.com
>
> ---------
>




From db3l at fitlinxx.com  Tue Feb 26 19:15:07 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 26 Feb 2002 19:15:07 -0500
Subject: Socket stream
References: <a5g5a7$1mt$1@bambi.zdv.Uni-Mainz.DE>
Message-ID: <uofibaigk.fsf@ctwd0143.fitlinxx.com>

Christian Maus <maus at netz.klinik.uni-mainz.de> writes:

> Hello,
> 
> I try to control a device via a com server. Setting a command to the device 
> works fine, but when I try to read the answer, my loop for reading from the 
> socket does not terminate. My loop looks like this:
> 
> while 1:
>        data = sockobj.recv(1024)
>        if not data: break
>        print data
> sockobj.close()
> 
> I get the output:
(...)
> and then I'm stuck in the loop. Any idea what the problem is?

If there is no data on the socket and you issue a recv() it will block
until data arrives.  I don't know your device protocol but you don't
seem to have a termination condition in your loop.  You probably want
to detect when the device has completed sending you information and
exit the loop on that condition, rather than looking for an empty
string (which should only happen when the other end closes down the
socket).

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


From tdickenson at devmail.geminidataloggers.co.uk  Mon Feb 18 12:34:15 2002
From: tdickenson at devmail.geminidataloggers.co.uk (Toby Dickenson)
Date: Mon, 18 Feb 2002 17:34:15 +0000
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org>
Message-ID: <3cd27uo3p0suf0ds56gmmr9ra9heasfaan@4ax.com>

(reply posted, and cc to Mark)

Mark Mitchell <mark at codesourcery.com> wrote:

>Python provides synchronous signals, even though signals are
>asynchronous events at the operating system level.  In many ways,
>this is a Good Thing.  For example, in C you can't do much of anything
>from within a signal handler since you don't know much about the
>current state of the application,

>but in Python you can do just
>about anything.

The rest of your post proves that you know this isnt quite true.

>This code, however, is not signal-safe.  For example, if a signal
>occurs between the two calls to os.close, we will end up not closing
>the second descriptor.

No major damage is done if the exception ends up terminating this
python process, or if the exception is only caught after sufficient
objects have been destroyed that it has lost all reference counts on
the crucial file objects. 

>One solution, of course, is to make sure that the signal-handlers do
>not throw exceptions.  But that's not a reasonable solution in library
>code; the library can't go around deciding what signal handlers should
>be doing.

A library working to those standards is already in trouble without
introducing anything so tricky as a signal. In python almost any
operation can potentially raise an exception: function calls, slices,
and even assignments to globals or class attributes may involve memory
allocation which may fail.

Im guessing you have a C++ background, right? For the benefit of the
luck ones who dont, in C++ it is considered good style to use classes
with destructors to tidy up *all* resources used by a program, often
one class instance per resource, to ensure that all resources are
freed as an exception propagates up to its handler. Any half-complete
changes to an objects state are rolled back. These destructors can
only be written using operations that are guaranteed to *never* throw.

Python does not have this type of guaranteed-never-throws operation.
Therefore it is not possible to code in the C++-style.

For almost any non-trivial object, an exception can leave it in an
inconsistent state. The easiest option is for whoever catches the
exception to ensure it gets destroyed.


Toby Dickenson
tdickenson at geminidataloggers.com


From lac at strakt.com  Mon Feb 18 10:28:52 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 18 Feb 2002 16:28:52 +0100
Subject: tab order for Tkinter widgets 
In-Reply-To: Message from Laura Creighton <lac@strakt.com> 
   of "Mon, 18 Feb 2002 14:44:57 +0100." <200202181344.g1IDivma012130@ratthing-b246.strakt.com> 
References: <200202181344.g1IDivma012130@ratthing-b246.strakt.com> 
Message-ID: <200202181528.g1IFSqGo012695@ratthing-b246.strakt.com>

Replying to my own message because I figured it out.

> I thought that the order in which you tabbed through widgets was the
> order in which you gridded them (assuming you are using the grid manager).

That was the part that was wrong.  it was the order that you create them.
So I moved my sort to a different part of the funciton and all is well.

Laura



From m.faassen at vet.uu.nl  Tue Feb 12 15:31:55 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Feb 2002 20:31:55 GMT
Subject: Coming of age...
References: <mailman.1013538140.7691.python-list@python.org>
Message-ID: <a4bu3r$ic8$1@newshost.accu.uu.nl>

Andrew MacIntyre <andymac at bullseye.apana.org.au> wrote:
> On 12 Feb 2002, Martijn Faassen wrote:

>> Another interesting thing the other day is that on comp.object someone
>> was asking which language they should try to experiment with OO, and
>> there came lots of suggestions to use Ruby (Python was in there, but Ruby was
>> mentioned a *lot*). Seems that language has been crossing a barrier
>> as well recently.

> My take: Ruby turns out to be a language of refuge for escapees from Perl
> who can't cope with white-space significance (more of them than not, I
> expect).

That doesn't explain the recommendations in comp.object, though. 
They use smalltalk, Java, C++ yes, but I don't expect most of the people
there are Perl users at all. I'm not entirely sure any more if the question
about a OO language per se or if it was about experimenting with dynamic
languages, by the way.

Regards,

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


From phr-n2002a at nightsong.com  Fri Feb  8 20:06:20 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 08 Feb 2002 17:06:20 -0800
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com> <slrna6649e.d87.philh@comuno.freeserve.co.uk>
Message-ID: <7xaduj79bn.fsf@ruckus.brouhaha.com>

philh at comuno.freeserve.co.uk (phil hunt) writes:
> You could go by the client's IP address. Though IMO this is a bit iffy.

It's not iffy, it's completely unreliable.  For example, all requests
from AOL users arrive through a small number of IP addresses of AOL
proxy servers.  Millions of AOL users share a few hundred IP addresses
and there's no way to tell them apart by address.  Also, there's no
consistent relation between an AOL client session and which AOL proxy
server it uses.  If an AOL user visits your page and hits "reload"
five times, the resulting six requests can come from six different IP
addresses.


From logiplexsoftware at earthlink.net  Fri Feb  1 14:14:29 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 11:14:29 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <CVy68.104504$8b.3960936@atlpnn01.usenetserver.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<b34d3835.0202010038.6a907bff@posting.google.com>
	<CVy68.104504$8b.3960936@atlpnn01.usenetserver.com>
Message-ID: <20020201111429.1ffdbe44.logiplexsoftware@earthlink.net>

On Fri, 1 Feb 2002 10:54:44 -0500
Steve Holden wrote:

> "Rony" <rony.steelandt at bucodi.com> wrote ...
> 
> > I'm sorry but i definitifly don't agree with that, and here i'm
> > speaking trough personnal experience !
> > Somewhere beginning 90'ths my development team came to me saying it is
> > time to abandon language X, that we were using, because now you have
> > languge Y who has all the possible benefits; better librarys, faster
> > development, blablabla...
> > And i trusted my chief development engineer and said yes.
> > To make a long story very short, the end of it was that the new
> > language din't have all the necessary function we needed, they never
> > found a solution (even not hired outside experts on that language) and
> > we trowed away one man year of development. And that is reality...
> > I know all the arguments, like your chief engineer didn't do his job
> > and how comes that it took a year and .... but since i'm the owner of
> > the company the problem  was mine at the end. So i'm convinced that
> > managment has to be involved in such decissions.
> >
> Rony, I feel you are making the mistake of generalizing from your own
> experience. This is perfectly understandable, and something we all do.
> However, my own feeling is you should solve this problem not by becoming
> better technically, but by becoming a better recruiter.

Clearly recruiting is a difficult chore.  I myself picked two candidates to
be junior developers at a job.  One of them had me convinced that he had a
lot of aptitude (and his prior experience seemed to support his claims)
while the other was far more restrained in his assessment of his skills. 
We hired both of them and guess which one turned out to be the better hire?
 The guy with the experience was so sure he was always right that he
couldn't hear anyone's critisism and turned out to be a complete loss.  The
other guy was looking for input and advanced much faster.  He appreciated
any critisism as long as it helped him improve.  But the situation could
just have easily turned out just the opposite.

> Ultimately your mistake ws not in trusting your chief engineer, but in
> hiring an untrustworthy chief engineer.

Well, we all make mistakes.  You can't be sure that the chief engineer was
incompetent or untrustworthy based upon one incident, regardless of the
outcome.  If he consistently made bad choices then that's another matter
altogether.
 
> Fortunately it seems that you do have a natural aptitude for technical
> tasks, and so you are avoiding the problem in future by developing your
> technical knowledge and, effectively, acting as chief technical officer
as
> well as manager.

Absolutely agree.

-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From mcherm at destiny.com  Wed Feb 13 15:00:25 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Wed, 13 Feb 2002 15:00:25 -0500
Subject: Python License Issues
Message-ID: <3C6AC5D9.7030602@destiny.com>

> Hello from rainy London
> 
> We have an issue with the python license

            [...]

> We can consider accepting liability 
> on all third party code we use, for such gross misconduct situations but is 
> it technically possible to do so given the PSF license? Ideally we would 
> like to negotiate a single license agreement with these clients (our license 
> agreement) which provides additional liability over and above to that which 
> PSF provides. However given clause 4 it seems that the client *must* 
> directly be bound by the PSF license. 
> 
> Any comments? 
> 
> Pavlos


First of all -- THANKS for your vote of confidence to be willing to 
accept the liability: No, we (the Python community) try hard not to 
engage in gross misconduct, but we appreciate your vote of confidence to 
back that up with real money (well... liability is LIKE money sometimes).

Secondly: this is just not a situation which the Python license is 
really intended to solve -- clause 4 is there to protect Guido, the PSF, 
and such folks, not to restrict the use of Python. Other folks with more 
legal background than I may come up with a clever legal workaround.

However, if they don't, then that's EXACTLY why the PSF holds the 
copyright on the code. Python is licensed under the PSF license to just 
about anyone who is willing to abide by a few rules (like clause 4: 
can't sue us 'cause you don't like how it works), but that doesn't mean 
that this is the ONLY license. If you can't work out something better, 
why not approach the PSF and ask them to license Python to YOU under a 
different license? A non-transferable license specifically allowing you 
to redistribute w/o restrictions if you accept legal liability might 
just be workable. Of course, IANAL and all that.

-- Michael Chermside







From fgeiger at datec.at  Wed Feb  6 02:24:32 2002
From: fgeiger at datec.at (F. GEIGER)
Date: Wed, 6 Feb 2002 08:24:32 +0100
Subject: wxPython book
References: <88bc63c6.0202040505.2a9c3016@posting.google.com>
Message-ID: <a3qlpq$ibi$1@rex.ip-plus.net>

You can find many ideas and samples here: http://wiki.wxpython.org/

Cheers
Franz

"Doug Farrell" <writeson at earthlink.net> schrieb im Newsbeitrag
news:88bc63c6.0202040505.2a9c3016 at posting.google.com...
> Hi all,
>
> I'm busy evangilizing my office co-workers on the wonders of Python
> and wxPython, to which they are responding pretty well. We are
> currently working on a test project that uses both, and the most
> frequent complaint I get is "there's no book on wxPython..... wah wah
> wah". I have seen mention of a book being written, presumably by Robin
> Dunn. Is this true and if so, any idea when it might be coming out?
>
> Thanks,
> Doug




From laz at strakt.com  Tue Feb  5 03:46:21 2002
From: laz at strakt.com (Fredrik Juhlin)
Date: Tue, 5 Feb 2002 09:46:21 +0100
Subject: Get current cursor position in ScrolledText
In-Reply-To: <4feefc55.0202042304.62cc5cc0@posting.google.com>
References: <4feefc55.0202042304.62cc5cc0@posting.google.com>
Message-ID: <20020205084621.GC3733@strakt.com>

On Mon, Feb 04, 2002 at 11:04:46PM -0800, Andrea wrote:
> On every seperate insert i need the cursor-position and the current
> length of all characters together. But if i bind an event to every
> possible key (e.g. <Key-A>) and i make a <ScrolledTextName>.get() i
> get the content of the ScrolledText before the 'A'-insertion happens.
> So the problem is if i paste a longer text i've no possibility to get
> the real length after insertion.
> Do anybody know how i can evade this problem?
You can call your event handler with after_idle(). That way all other
event handlers finish before your is executed.

//FJ



From skip at pobox.com  Thu Feb 28 16:00:05 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 28 Feb 2002 15:00:05 -0600
Subject: map-like function on dict values?
In-Reply-To: <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com>
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
        <mailman.1014861466.23611.python-list@python.org>
        <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com>
Message-ID: <15486.39510.22262.185622@beluga.mojam.com>

    >> Check out the operator module.  In particular, operator.setitem will
    >> probably do what you want:

    Huaiyu> This works.  Thank you.  It speeds up by more than a factor of
    Huaiyu> four againt for loop iteration.

Don't assume for loop iteration is your only culprit... ;-) List creation
overhead can be as big a problem.  I added another test function called iter
which relies on 2.2's new "for k in dict" capability to completely avoid
list creation.  (I also changed your range() to an xrange() to avoid the
extra list allocation, and called clock() instead of time() to measure CPU
time instead of elapsed time.  This is on a Linux system) Here's what I get
on my laptop:

    iter 6.69
    iteritems 28.36
    mapvalue 8.12

Modified version appended...

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)

import operator
from time import clock

def iter(f, d):
        for k in d:
                d[k] = f(d[k])

def iteritems(f, d):
        for k, v in d.items():
                d[k] = f(v)

def mapvalue(f, d):
        keys = d.keys()
        map(operator.setitem, [d]*len(keys), keys, map(f, d.values()))
                
def timeit(test, f, d):
        _time = clock()
        test(f, d)
        print test.__name__, clock() - _time

if __name__ == '__main__':
        d = {}
        for i in xrange(1000000):
                d[i] = None
        def f(x): pass

        timeit(iter, f, d)
        timeit(iteritems, f, d)
        timeit(mapvalue, f, d)



From tom at peresys.co.za  Sat Feb 23 08:33:31 2002
From: tom at peresys.co.za (Tomasz Stochmal)
Date: 23 Feb 2002 05:33:31 -0800
Subject: fun with ADO
References: <2ec1bc1c.0202221202.1e0ea6aa@posting.google.com>
Message-ID: <f20ea932.0202230533.6a01ac4b@posting.google.com>

> 
>     rs=win32com.client.Dispatch(r'ADODB.Recordset')
>     rs.Open(sql,connectString)
>     while not rs.EOF:
>         name=str(rs.Fields("clientname").Value)  #oink!
>         gid=str(rs.Fields("gid").Value)
>         (- do something fun with fields -)
>         rs.MoveNext()
>     rs.Close()
> 

Use rs.GetRows() to retrieve all rows & fields with one call only
on python side you can extract the field values that you need


From martin at v.loewis.de  Tue Feb 12 19:44:00 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 01:44:00 +0100
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de>
Message-ID: <m3lmdyw6r3.fsf@mira.informatik.hu-berlin.de>

"MDK" <mdk at mdk.com> writes:

> When I double click on it in Windows explorer I get a message that tcl80.dll
> can not be found.  That file does not exist on my PC.  How then is it
> running under PythonWin if that file does not exists and how can I make it
> run when double-clicked?

One possible explanation is that you have an additional Python
installation whose _tkinter.pyd is linked with an older Tcl version,
or that you merely have an older _tkinter.pyd on your system.

Regards,
Martin



From xander at yifan.net  Tue Feb  5 00:43:29 2002
From: xander at yifan.net (Alex Holkner)
Date: Tue, 5 Feb 2002 16:43:29 +1100
Subject: Floating-point multiplication
In-Reply-To: <LNBBLJKPBEHFEDALKOLCGEMPNJAA.tim.one@comcast.net>
Message-ID: <OJEJKEOFJEGJNCEHNGHKEEGCCBAA.xander@yifan.net>

> I don't know what "embedded python" means to you, and the answer
> is probably
> crucial.

Nothing special, it's a win32 app which runs Python scripts with
Py_Initialize / PyRun_SimpleString / Py_Finalize.

> "precision control" mode.  Core Python doesn't muck with the FPU control
> registers, so something outside of core Python you're doing, or
> calling, is
> doing the damage.  Scour the sources for calls to control87, _control87,
> controlfp or _controlfp.  If you find exactly one such call, it's wrong
> <wink>.

No luck, unfortunately.  My application makes use of DirectX (which I know
uses single-precision, but obviously I can't check its source, and I doubt
there's a bug there), and ODE (a dynamics library.. but no FPU control
functions used that I can find).

I tried putting _control87(_PC_64, _MCW_PC) before Py_Initialize and
PyRun_SimpleString (and restoring it afterwards); which makes everything
work as expected.  This is obviously not a fantastic solution, but it does
show that you're right about the cause of the problem.

I might play around with moving my _control87 around the code a bit to try
and find the cause, unless someone already knows of a DirectX issue (using
Direct3D7 interfaces)?

Thanks heaps for your help,
Alex.




From stojek at part-gmbh.de  Sat Feb 16 09:48:10 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Sat, 16 Feb 2002 14:48:10 GMT
Subject: using lists in C-extensions?
Message-ID: <3c6e711a.9871609@news.easynews.net>

Hi,

Hi,

I'm still fighting with my optimization problem.
After I got a lot of very useful hints from this
group I have accelerated my search problem by a
factor of app. 50. As always this was achieved by
structuring the data efficiently, avoiding expensive
operations in loops and so on. THANKS.

But I am still far too slow. So I have to go for C.
My problem is, that I have to pass a list (or actually
a couple of lists) to the C routine and after performing
the calculations, returning them to Python. As far as
I understand things, I have to use PyArg_Parse() and
Py_BuildValue() for this, right?

Both expect a format string like "s", "i" or something.
I didn't find a format string for a list ;-)
[items] could work, but my lists are huge and I can't
list the members one by one.

There is this strange "O" respectively "O&" which might
be suitable, but I don't know how.

So could anyone give an example or tell me where to look?

------------
EXAMPLE:

list=[]
for i in range(100000):
    list.append(i)

listnew=map((lambda x:x+1),list)  # this should be coded in C

------------
Maybe things are easier with scipy.weave, but one important
point is that I can use py2exe or installer. Is that possible
with weave? (Windows NT)

Does SWIG handle this problem?

Thanks again,
Marcus




From logiplexsoftware at earthlink.net  Fri Feb  8 14:47:49 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 11:47:49 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <0BV88.108473$KM2.4226261@atlpnn01.usenetserver.com>
References: <3C62EE4A.5090305@destiny.com>
	<20020207140755.364d5d77.logiplexsoftware@earthlink.net>
	<200202081311.g18DBQGo020260@ratthing-b246.strakt.com>
	<mailman.1013194035.3597.python-list@python.org>
	<0BV88.108473$KM2.4226261@atlpnn01.usenetserver.com>
Message-ID: <20020208114749.621672d8.logiplexsoftware@earthlink.net>

On Fri, 8 Feb 2002 14:23:50 -0500
Steve Holden wrote:

> "Cliff Wells" <logiplexsoftware at earthlink.net> wrote ...
> > However, this discussion does give me the idea that pair-programming
could
> > be very beneficial if programmers were rotated on some schedule, so
that
> > one day they watched someone else and on another someone else watched
> them,
> > especially given a group of several programmers that could be rotated
> > through.  This way, programmers would get enough time doing development
on
> > their own but still have a chance to share ideas with other programmers
in
> > their group.  Too bad I'm not in a position to give it a try :P
> >
> The idea behind pair programming is exactly that such rotations should
take
> place. You seem to think that two prograqmmers are joined at the hip when
> teams are formed, but in fact the pairings arise fairly spontaneously in
> most cases.

Well, I obviously didn't know much about pair-programming at the start of
this thread (and I have learned quite a bit so far), so I was under a
mistaken impression regarding rotation.  Seen in this new light I would
have to reconsider my earlier position (although I still retain the right
to be skeptical ;).

> You should try it. I spent six months in Stockholm thirty years ago, and
> would still love to go back to Sweden. Also AB Strakt are serious about
> Python, so you wouldn't be forced to usae something ghastly like C++.
> 

Hrm.  You aren't just trying to get rid of me, are you?

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From bokr at oz.net  Fri Feb  1 15:28:15 2002
From: bokr at oz.net (Bengt Richter)
Date: 1 Feb 2002 20:28:15 GMT
Subject: Equivalent of Perl chomp?
References: <mailman.1012513005.17840.python-list@python.org> <%ok68.88316$8b.3281206@atlpnn01.usenetserver.com>
Message-ID: <a3etov$hgu$0@216.39.172.122>

On Thu, 31 Jan 2002 18:24:16 -0500, "Steve Holden" <sholden at holdenweb.com> wrote:
[... big chomp test ...]
>
>My results follow. You may find you need to add a couple more decimal places
>to get sensible results, my Thinkpad is a bit of a cronker ...
>
>D:\Steve\Projects\Python>python time1.py
>Endswith 0.00000511
>sliceIt  0.00000434
>
>D:\Steve\Projects\Python>python time1.py
>Endswith 0.00000511
>sliceIt  0.00000186
>
>D:\Steve\Projects\Python>python time1.py
>Endswith 0.00000505
>sliceIt  0.00000209
>
Using CPU ticks (actual RDTSC) and finding the _minimum_ 
(best of 100 in this case - usually 10 is enough)
times on a 300mhz P2:

[12:30] C:\pywk\chompt>python chompt.py
min:   1595 nullTime
min:   3498 endsWithTime, ex nulltime: 1903
min:   2702 sliceItTime, ex nulltime: 1107

[12:30] C:\pywk\chompt>python chompt.py
min:   1598 nullTime
min:   3537 endsWithTime, ex nulltime: 1939
min:   2681 sliceItTime, ex nulltime: 1083

[12:30] C:\pywk\chompt>python chompt.py
min:   1594 nullTime
min:   3541 endsWithTime, ex nulltime: 1947
min:   2679 sliceItTime, ex nulltime: 1085

[12:30] C:\pywk\chompt>python chompt.py
min:   1598 nullTime
min:   3528 endsWithTime, ex nulltime: 1930
min:   2675 sliceItTime, ex nulltime: 1077

The variability in minimums from trials of 100 is not so much.
Four ticks (in the nullTimes) is 4/300 microseconds here ;-)

>As you can see there is still some variability but slicing looks like a
>clear winner. Increasing the iteration count might produce convergence on
I think slicing is probably a winner because endswith subsumes more
complex slicing options that have to be tested for, and the argument can be
multicharacter. I would be surprised if single-character string handling
wasn't special-cased a lot for speed, benefiting s[-1:] == '\n'.

>the timings. You might also ponder why sliceIt appear more variable than
>Endswith (tell me if you think of a good reason, I have no clue - this isn't
>a trick question).
I suspect the variability is lumpiness in memory resource access timings, not in the
routines per se. Changing the order may make the other appear more variable. The state
of your OS's memory resources will not be identical each time you start the program.

Also, you can make a shocking difference in results by inserting e.g. an apparently
innocuous print or info-storing code inside a timing loop (but outside of the timed
interval). I strongly suspect this is due to the extra code's messing up the CPU cache
for the next iteration.

I like minimum single shot times to get an idea of what an algorithm
is capable of, which you can't get as well or as quickly with averages.
Then the trick is to make it do it most of the time ;-)

Regards,
Bengt Richter



From jkraska at san.rr.com  Tue Feb 19 21:08:05 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 20 Feb 2002 02:08:05 GMT
Subject: Adding static typing to Python
References: <3C726575.5090400@destiny.com> <mailman.1014133892.1456.python-list@python.org> <q3057us1qab29uvm2c6f551p4vs6j7pjvr@4ax.com> <a4ut9p$ngd$0@216.39.172.122>
Message-ID: <7e167u8k6em49blfe16o7cag6kh9on4fp0@4ax.com>

>It would be an interesting measure of how dynamic things really are
>in practice.

Generally not very. The vast majority of all Python dictionaries
have <= 4 entries in them, for example. I verified this by mucking
with the code and adding a counter to the set method of the native
Python dictionary and then running a wide suite of code through
Python and dumping the statistics for all dictionaries that had
ever existed over the life of the interpreter.

Once a certain attribute exists in a dictionary, it has a tendency
to be set in stone over the life of the dictionary. This alone
suggests a large optimization path, but looking at the Python
implementation (been a while, 1.5.2 when I looked last, IIRC),
it would be difficult to implement without a lot of reengineering.

As far as rebinding labels to new types of objects, I think that
actually might be difficult to calculate. IIRC, labels aren't
objects -- they're just labels.

I'm working on something like this over in my experimental
environment, but this is a clean reimplementation of Python in
C++. It's unlikely to ever to come to any real fruition.

C//



From starx at pacbell.net  Wed Feb  6 04:21:41 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 09:21:41 GMT
Subject: Binary Python file.
References: <3C5FA9E1.DF4AC37F@lucent.com>
Message-ID: <Xns91ADDDAD65A6RASXnewsDFE1@130.133.1.4>

Lukasz Pawelczyk || Tue 05 Feb 2002 01:46:09a:

> Hello,
> 
> Is there a possibility to make a binary python file?
> I would like to hide the code in unix environment.

Well, the compile function listed here will give you a pyc file...
http://www.python.org/doc/current/lib/module-pycompile.html#l2h-3816

...but from what i've heard it's only a little bit better then say, rot-13, 
for keeping people out of your source. It will probably keep out the 
slightly curious, the incapable, and those of us who just don't care what's 
in there :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From syring at email.com  Sun Feb  3 04:52:54 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 10:52:54 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org>
Message-ID: <a3j1a4$18e7ov$1@ID-7529.news.dfncis.de>

"Gerhard H?ring" <gh_pythonlist at gmx.de> schrieb
> Le 02/02/02 ? 16:59, Lawrence Oluyede ?crivit:
> >
> > Why in the std distribution of Python 2.2 there isn't the curses
> > module?
>
> If you're using Windows, that explains it ;-) There is indeed no curses
> module included in the Python binary distribution for Windows. IIRC the
> reason is that there is no ncurses library readily available on win32 to
> build the curses support against.

But it shouldn't it build with pdcurses?

Karl M. Syring



From mlh at vier.idi.ntnu.no  Tue Feb 19 04:39:10 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Tue, 19 Feb 2002 09:39:10 +0000 (UTC)
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <slrna7479u.2ad.mlh@vier.idi.ntnu.no>

In article <24c39b2c.0202181531.187fad4c at posting.google.com>,
Alexander Jerusalem wrote:
>Has anyone heard of plans to add static type checking to Python? I
>like Python very much but I don't use it for bigger projects because I
>fear the testing effort. For example I just refactored a rather large
>Java codebase. And when I change a method name for example, all I have
>to do is run everything thru the compiler and it will show me all
>places in other classes where this method is called, so I can go there
>and chenge the method call as well.
>
>In Python I'd have to test the whole software suite including each and
>every branch in each and every if statement to be sure that I didn't
>forget something. To build such a test suite would take almost as long
>as it takes to build the whole software and I'd have to maintain the
>test suite as well. That's just unafordable.
[snip]

Have you heard of PyChecker?

>
>Regards,
>Alexander Jerusalem

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From max at alcyone.com  Tue Feb 26 16:46:56 2002
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Feb 2002 13:46:56 -0800
Subject: Class functions
References: <NVSe8.151$en5.1429@typhoon.sonic.net>
Message-ID: <3C7C0250.7361AABD@alcyone.com>

Chad Maine wrote:

> import theclass
> X = theclass.CLASS

You want

	X = theclass.CLASS(...)

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From weeks at vitus.scs.agilent.com  Sat Feb 23 19:07:17 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Sun, 24 Feb 2002 00:07:17 +0000 (UTC)
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <1014509237.42236@cswreg.cos.agilent.com>

Tripp Scott (tripps81 at yahoo.com) wrote:
:   a = -1
:   make_abs(a)
:   print a # = 1

It is good to see that I'm not the only person who still uses "pass by
reference" to mean that a function can rebind its argument variable.


Greg


From peter at engcorp.com  Wed Feb 27 01:30:36 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Feb 2002 01:30:36 -0500
Subject: advocacy help ( Python vs VB )
References: <mailman.1014748548.28891.python-list@python.org> <a5gutk$h1e4@news1.gtech.com>
Message-ID: <3C7C7D0C.63B1E195@engcorp.com>

Peter Milliken wrote:
> 
> I just checked the "Programming Language Levels and Ranges of Source Code
> Statements per Function Point" Table in Capers Jones book "Applied Software
> Measurement" (2nd edition) and got the following data:
> 
> Visual Basic 2   35loc/FP  Level 9.0
> Visual Basic 3   32loc/FP  Level 10.0
> Perl  27loc/FP  Level 12.0
> 
> Unfortunately I couldn't find Python in the table but the figures for Perl
> is probably close enough. So assuming that these average figures are correct
> and that there is a 1:1 relationship between Perl and Python (I assume
> Python would be higher!) then using Python should show approximately a (32 -
> 27) / 32 * 100 = 16% productivity improvement - hardly earth shattering! :-)
> But since Python is better than Perl perhaps we could expect about 25%
> ???????

loc/FP may be a good indication of relative program length, but it
should not be used blindly for comparisons of productivity.  Not that
I have any data to back this up, but anecdotally speaking I find Python 
development goes faster more as a result of its very high readability
than as a result of low loc/FP.  

-Peter


From cliechti at gmx.net  Mon Feb 25 11:57:44 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 25 Feb 2002 17:57:44 +0100
Subject: Question on Python path
References: <3qnk7ug977hk4peorufobtlo38nh8j9v87@4ax.com>
Message-ID: <Xns91C0B74CA73EEcliechtigmxnet@62.2.16.82>

Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote in 
news:3qnk7ug977hk4peorufobtlo38nh8j9v87 at 4ax.com:

> I have put all my own modules in a subdirectory of site-packages. 
> How can I instruct Python (I'm using Python 2.2 on win2k) to also 
> look up this directory when a
> 
> import <whatever>
> 
> is found?

if you put it in site packeages it should be a package... this means 
the directory should have an __init__.py in it. then you can use it 
with
>>> import yourdir.yourmodule

or you can put your modules anywhere and create a ".pth" file with 
the path in it. this is appended on sys.path on startup.

and then there is the PYTHONPATH environment variable.

or you can add you path to sys.path in sitecustomize.py
-- 
Chris <cliechti at gmx.net>



From jason at jorendorff.com  Wed Feb  6 12:56:51 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 11:56:51 -0600
Subject: Remaining nimble in the face of evolution
In-Reply-To: <3C6158A2.9060906@shinners.org>
Message-ID: <HFEKILOLEFEFMKAECNDLIEMGDAAA.jason@jorendorff.com>

jim.vickroy wrote:
> Now a reality check occurs and the requirement evolves such that the
> archival behavior is to be optional.
> 
> What is the "best" strategy to handle these types of evolution.

One excellent strategy you can employ *now* to ensure that unforeseen
future changes go smoothly is to write a complete test suite,
use it, and keep it up-to-date.

When it comes time to make an incompatible change, you then know
*exactly* what existing behavior you're breaking.  So you can try out
different strategies when you get there, rather than trying to guess
now what problems you'll face in the future.

  The seminal article "Test Infected" explains why.
  http://members.pingnet.ch/gamma/junit.htm

  The documentation of the standard unittest module tells how.
  http://www.python.org/doc/current/lib/module-unittest.html

## Jason Orendorff    http://www.jorendorff.com/



From chappa at math.washington.edu  Mon Feb  4 20:47:14 2002
From: chappa at math.washington.edu (Eduardo Chappa)
Date: Mon, 4 Feb 2002 17:47:14 -0800
Subject: OT again... is there an IMAP expert in the house?
In-Reply-To: <mailman.1012224450.11377.python-list@python.org>
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org>
 <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org>
 <mailman.1012224450.11377.python-list@python.org>
Message-ID: <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu>

*** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:

:) Urrr... it doesn't solve my problem. Maybe IMAP isn't the solution
:) either.
:)
:) The customer(s) I am trying valiantly to support want to be able to
:) share a mailbox *concurrently*. The Mercury and UW servers don't seem
:) to work that way. I haven't studied the protocol so I don't know if it
:) is even possible.

This is not a problem about the protocol you are using to read your
folder. IMAP is the way that you communicate to the server. The way that
the mail is stored in the server is the relevant point here. For example,
the default for the UW server is unix, and unix (format) does not allow
concurrent access to a mailbox. If you use the UW server, and you use mbx
format, then you can have concurrent access to your folders. Convert your
folders to mbx format and you'll be done.

-- 
Eduardo
http://www.math.washington.edu/~chappa/pine/



From python at chajadan.net  Thu Feb 21 22:44:34 2002
From: python at chajadan.net (chajadan)
Date: Thu, 21 Feb 2002 19:44:34 -0800
Subject: Unintelligible Lambda Tricks
In-Reply-To: <3c74c211.20012718@news.isar.de>
Message-ID: <5.1.0.14.0.20020221193817.00a6c6a0@63.118.78.184>

data.__setitem__(4,42)

def swap(a,x,y):
a[x] = (a[x], a[y])
a[y] = a[x][0]
a[x] = a[x][1]
You can write this as a lambda expression like this:

>swap = lambda a,x,y:(lambda
>f=a.__setitem__:(f(x,(a[x],a[y])),f(y,a[x][0]),f(x,a[x][1])))()


I'm going to indulge in a question that is preceeding research...if I'm 
spanked and told to rtfm then I will do so, but I figured with the zealous 
attraction some have to lambdas, someone would be more than happy to give 
me a quick run through.

I don't understand why __setitem__ takes 2 arguments, but in the lambda () 
there are 3...

I'm also not sure how it gets its values for a, x, or y---are these pure 
abstract variable names, or do the refer to previously defined values for 
variables with the same name?

Basically, I have no freakin' clue how python (or any other language) runs 
through that lambda. I don't understand the sequence of events. What does 
it evaluate, where does it pull the values from, if f is defining a 
function, why is f used ~in~ the function... AAarrgh!

I guess I better go read...

--chajadan




From hartmut at goebel.noris.de  Sun Feb 24 09:00:23 2002
From: hartmut at goebel.noris.de (Hartmut Goebel)
Date: Sun, 24 Feb 2002 15:00:23 +0100
Subject: ANN: decompyle 2.2 beta 1
Message-ID: <mailman.1014741412.30867.clpa-moderators@python.org>

Announcing:

             decompyle -- A Python byte-code decompiler

                      version 2.2 beta 1


'decompyle' converts Python byte-code back into equivalent Python
source. It accepts byte-code from any Python version starting with 1.5
up to 2.2.

The generated source is very readable: docstrings, lists, tuples and
hashes get pretty-printed.

'decompyle' may also verify the equivalence of the generated source by
by compiling it and comparing both byte-codes.


New Features 
------------

Since Release 0.6.0:

  * Now decompyles byte-code from Python versions 1.5 up to 2.2

  * Now requires Python 2.2 but is able to decompyle older byte-coe,
    too. Prior version of decompyle had to be run with the Python
    version which generated the byte-code.

  * pretty-prints docstrings, hashes, lists and tuples

  * decompyle is now a script and a package

  * Now adds an emacs mode-hint and tab-width to the output files

  * enhanced test suite: more test patterns, .pyc/.pyo included

  * avoids unnecessary 'global' statements

  * many internal changes and code overhouls

  Please refere to the file 'CHANGES' for a list of changes in this
  release.


Where to get it?
----------------

The source archives and instructions for isntalling the package can be
found at:

             http://www.goebel-consult.de/decompyle/


Simple Help Needed!
-------------------

Please help testing 'decompyle'! 

The EXTENDED_ARG token is untested (this is a new token for Python 2.0
which is used only if many items exist within a code object). If you
have byte-code which does or may include this token, please try to
decompyle your application.

It's easy: the script 'test_pythonlib' which is included in the source
           distribution does the job for you. Just read the simple
           instruction there.


Requirements
------------

'decompyle' requires Python 2.2 or later.


Known Bugs/Restrictions
-----------------------

* The EXTENDED_ARG token is untested (this is a new token for Python
  2.0 which is used only if many items exist within a code object).

* Byte-code generated by Python 2.2 differs byte-code generated by a
  prior version even for the same source. This is due the intruduction
  of iterators. Currently 'Decompyle' fails verifying the source if
  the byte-code was generated by an older version of Python.

* Verifying decompyled source with optizimzed byte code (.pyo) when
  running without optimizations (option '-O' not given) fails in most
  cases. Same is true for vis-a-versa. This is due to the fact that
  Python generated different bytecode depending on option '-O'.

-- 
| hartmut Goebel       | hartmut at goebel.noris.de  // |
| Software Manufactur  |                        \X/  |



From gh_pythonlist at gmx.de  Tue Feb 19 07:07:46 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 13:07:46 +0100
Subject: newbie: Some file object tips?
In-Reply-To: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4>
References: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4>
Message-ID: <20020219120745.GA3538@lilith.hqd-internal>

Le 19/02/02 ? 11:00, Flavian Hardcastle ?crivit:
> 
> I'm writing a crude cgi chat script. I've already got it up and working 
> with Xitami server, and have chatted with a few net buddies.
> 
> How it works is very rough and simple... it basically just takes data from 
> people's browsers (using the FieldStorage function) and stores it in a file 
> called log.txt, and then displays the contents of that file back to them.
> 
> My problem is ... 
> 
> The longer you chat, the bigger the log.txt file gets, and consequently the 
> web page the chatters are viewing gets bigger and bigger, and becomes hard 
> to download.
> 
> So I want to include a "scroll back" feature ... which will enable the user 
> to control how many previous posts (s)he views. It will probably take the 
> form a simple input box on the form ... if you want to see the last 8 
> posts, you put 8 in the box, if you want to see the last 50 posts, you put 
> 50 in the box and so on.
> 
> So, I need the script to be able to search the log.txt file. Any tips as to 
> how I might go about this? 

You could read the file into a list with

lines = open("log.txt").readlines()

but now, all the lines will include a newline character at their end. To
strip these, you can do the following instead:

lines = map(string.rstrip, open("log.txt").readlines())

Now you can use slicing on the list, say to return the last 8 lines, you
can do:

lines[-8:]

HTH,

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.8 ?C      Wind: 6.1 m/s



From DavidA at ActiveState.com  Thu Feb 14 20:06:52 2002
From: DavidA at ActiveState.com (David Ascher)
Date: Thu, 14 Feb 2002 17:06:52 -0800
Subject: Reminder: Python track at OSCON -- Deadline March 1!
Message-ID: <3C6C5F2C.9E9B1B79@activestate.com>

Reminder:

The O'Reilly Open Source Convention (July 22-26, 2002 -- San Diego,
CA) is accepting proposals for tutorials, talks, panels, and lightning
talks.  See the Call for Participation in the Python and Zope track on
python.org.  Proposals are due by March 1, so don't wait a moment
longer!

Details available at:

CFP URL: http://www.python.org/workshops/oscon2002/cfp.html
form:    http://conferences.oreillynet.com/cs/os2002/create/e_sess

Cheers,

-- David Ascher
   [Guido's the program chair, but he's on the road, so I'm filling in]

PS: Feel free to resend this to whatever mailing lists may be
interested.



From sholden at holdenweb.com  Fri Feb 22 07:55:32 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 07:55:32 -0500
Subject: running a program via exec*
References: <a54294$l0e@r02n01.cac.psu.edu>
Message-ID: <Dcrd8.78243$n73.1509781@atlpnn01.usenetserver.com>

"Rajarshi Guha" <rxg218 at psu.edu> wrote in message
news:a54294$l0e at r02n01.cac.psu.edu...
> Hi,
>   this is somewaht related to my earlier post. When I use execl(progname)
> it seems to work. However it always looks for the progrma in the current
> directory. Is there any way to specify the path from the environment
> without having to type out the path to the program itself?
>
You forgot about execlp() ... the exec*p() variants search the usual
suspects for the executable.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From mwh at python.net  Wed Feb  6 09:54:01 2002
From: mwh at python.net (Michael Hudson)
Date: Wed, 6 Feb 2002 14:54:01 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de>
Message-ID: <u8za6pskh.fsf@python.net>

Simon Budig <simon at ping-pong.budig> writes:

> In article <uheowosxs.fsf at python.net>, Michael Hudson wrote:
> > Well, it can't go into 2.2.  If you mean 2.2.1, that's a little
> > greyer; there might be people out there who rely on the current
> > behaviour.
> 
> Hmm. Maybe it is possible to export this variable as an optional
> parameter to readline.set_completer with the default on the
> current behaviour?

But then that's a new feature...

> I think it is really annoying, that readline in the current
> python-module implementation insists on completing filenames when
> the completer returns no matches.

Sorry.  But I don't think you can claim it's a pure bug, and so it's
not 2.2.1 material.

Cheers,
M.

-- 
  Ya, ya, ya, except ... if I were built out of KSR chips, I'd
  be running at 25 or 50 MHz, and would be wrong about ALMOST
  EVERYTHING almost ALL THE TIME just due to being a computer!
                                              -- Tim Peters, 30 Apr 97


From magnus at thinkware.se  Thu Feb 21 09:00:38 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 15:00:38 +0100
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <258fd9b8.0202191512.4f4693ff@posting.google.com> <4iNc8.17011$Zp2.1752237@atlpnn01.usenetserver.com>
Message-ID: <3C74FD86.4060306@thinkware.se>

Hm... I thought I replied to this, but it's nowhere to be seen...

 > "Magnus Lyck?" <magnus at thinkware.se> wrote ...
 >> I'm trying to adopt the XP habit of writing tests first, and then only
 >> code enough to make the test pass. It's sometimes difficult. It
 >> requires some diciplin not to get carried away while writing the code,
 >> and do more than I've written tests for. When I do get carried away
 >> I end up having to write a whole bunch of boring tests
 >> afterwards, and that feels like a drag, because then the writing
 >> of tests is not part of the creative process. And it's difficult
 >> to be sure that I actually have all the tests I need then.

Steve Holden wrote:
 > As far as this goes, EITHER you aren't being sufficiently rigorous in
 > defining your tests, OR you are over-specifying unnecessarily during

Sorry for my bluntness, but are you referring to some theory you read
about, or is this what you personally practice? ;-) I didn't see any
complete unit test suites included in the zip supplementing your book. ;-)

 > implementation. This latter is a particularly troublesome waste of
 > programmer time, which XP attempts to eliminate with its well-known
 > (?) "You aren't gonna need it" credo:

No, that's not it. Why wouldn't I write a test for it if I felt that
it was a good function. Whether I need it now or not! If I think it's
a good thing, I would normally start by writing a test case. That's how
I define its interface! Functions you don't need now needs test cases
even more than other code or they will rot, but if you don't test them
and don't use them, they won't normally cause any other trouble than
being dead weight that you can do without.

I'm normally thorough in my unit tests. For me, writing these tests
have become a way to design interfaces, predict problems and create
examples for proper use of a class. Before, I usually wrote comments
before I wrote the actual code. Now I mainly write the tests instead,
and my customer complains about the lack of comments... :-(

No, the problem lies elsewhere.

As you know, it's possible to write code without writing tests first,
and when you are pressured to deliver, there is really a short term gain
in most cases to skip the tests. The problem is that it's both more
difficult and boring to write test afterwards, so in the long run you
will spend more time, and have less reliable tests. This is probably a
good enough reason to always ignore the short term gain, but sumetimes
you get pressured to deliver now! Maybe because you've been posting at
comp.lang.python when you ought to have been programming....

A particular problem I see is that I haven't found any good way to
unit test my GUI code (wxPython) automatically. I'm not going to try
tools like WinRunner or Rational Robot. They would cost too much in
money and time for my needs, and also be to slow to run as a part of
a normal unit test sequence. And there is no way you can write the
tests in advance of coding with points and click tools like that.

[Successful GUI unittesters, please stand up, and explain how you
do it.]

I am trying to keep the GUI layer as thin as possible, but I realize
that my opinion of what is thin enough changes as the project evolves.
As the code grows, I tend to refactor code from the GUI modules to
pure logic modules that are independent of the GUI. It might be such a
thing as how objects are sorted in a GUI listbox. Initially that seems
to be a pure presentation issue, but then new requirements arise and
there are logic functions that relate to the order. Then the ordering
has to take place in the logic layer. If the required features creep
on you slowly, you might have implemented a bit more in the GUI than
you ought to have... There is often a gray zone between the GUI layer
and the logic layer. Accepting such a thing and being able to refactor
it is one of the points with XP, but in reality it doesn't work so well
when you move funtions between systems/layers such as between GUI a
logic layer or between logic layer and database. These refactorings
can't really utilize the unit tests well.

All the logic modules use unit tests, but the GUI modules don't. So here
I basically get chunks of code, or at least well defined behaviour, that
I move from GUI to logic, and thus into the realm of unit tests. When
all is well, I solve this situation like this:

I see the function in the GUI code, and I think. "Hm, this logic should
be moved to that module, and I should access it via such and such
interface." I then code that interface in the unit tests, implement the
function in the logic module, and when all tests pass, I change my GUI
to use that interface and remove the logic from there. But there is a
problems.

Since the code is really already written, it's difficult (and wasteful)
to forget about the written code, and start all over from scratch. And I
don't have a selective clear memory switch in my brain, so I really can't.
This means, that however you turn it around, the unittests are a bit of
an afterthought, not the integral part of the creative process they should
be.

Even if I did have GUI unittests, I would probably have to rewrite them
a lot to make them call the logic layer directly. Sure I can test the
logic layer indirectly with the GUI tests, but a) It might be better to
test GUI against logic stubs most of the time so that the tests and
development of the GUI layer is less dependent of the current state of the
logic layer. b) Once the functions are in the logic layer I might want
to call them outside a GUI context. Maybe from a web framework. Then
I need to be able to test it without the GUI. The web interface might
use features the GUI doesn't use or vice versa.

To be honest, I feel that the whole XP unit test philosophy works well
when you can use it all the time, but as soon as you have to deal with
including already written code, without unittests, into your project,
you are in trouble. You might have to pay the price by throwing away
perfectly functional code due to lack of unit tests, or suffer the
consequences of living with poor unit tests.

I know, many people write test code after they have coded, and these
tests can be very useful, but they are not unit tests in the XP sense.

/Magnus



From mac at ceti.pl  Tue Feb 12 07:15:59 2002
From: mac at ceti.pl (Maciej Starzyk)
Date: Tue, 12 Feb 2002 12:15:59 +0000 (UTC)
Subject: Help with Installer.
Message-ID: <Xns91B3864EE6013maccetipl@149.156.4.11>

Here is the problem:

I generated a single-directory program with the
McMillan Installer.
When I launch the executable from bash, it works OK, 
however, when run as a shebang line in a shell script it 
fails to start. It says it cannot open the archive. 

It does not seem to be fault of the Installer, though.

It looks like bash cuts off the full path to the 
executable, when run as a shebang. The main() function
receives as parameters something like:

argv[0] = the_exe
argv[1] = /path/to/script/my_script

And the shebang line in my_script is:
#! /path/to/exe/my_exe

Does anyone know how it can be fixed ?
Thanks in advance, 
Maciek



 


From p.vrijlandt at aig.azn.nl  Tue Feb 19 05:30:16 2002
From: p.vrijlandt at aig.azn.nl (Patrick Vrijlandt)
Date: Tue, 19 Feb 2002 11:30:16 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl> <xyuvgcujsyl.fsf@sex.ifi.uio.no> <a4shr2$49s$0@216.39.172.122>
Message-ID: <a4t9fr$500$1@odysseus.uci.kun.nl>

"Bengt Richter" <bokr at oz.net> schreef in bericht
news:a4shr2$49s$0 at 216.39.172.122...
> It's not clear to me that it's generally advisable to ignore
> all the command codes telnetlib.py ignores (which is the apparent
> effect of the self.msg() call), but that will depend on what
> you are communicating with, and what you're doing with it, I suppose.
>

You're quite right.

In fact, telnetlib should keep a database of all these options, and possibly
the extended options, on the client side, on the server side, and wether
these have been negotiated or not. It should be able to handle all these
requests by providing an adequate answer.
In the past I worked with an implementation that denied everything the
server requested and it served my purpose.
I didn't go deep enough to make a new version of telnetlib.

Although telnetlib seems rather buggy, there is not much request for
improvement on c.l.py, so I think it is barely used, or
it works fine with most servers.

> Where is the mystery telnetlib.py author? ;-)
At least, he does not seem interested now...

BTW: New syntactic sugar (generators) might be nice for telnetlib too! This
would mean the necessary improvements (patches) wouldn't make it back to
1.5.2. though, which might or might not be an argument (Are we still taking
bugs out of 1.5.2?)

--
Patrick Vrijlandt




From piet at cs.uu.nl  Sat Feb 16 10:05:55 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 16 Feb 2002 16:05:55 +0100
Subject: line break confusion
References: <mailman.1013827165.15861.python-list@python.org>
Message-ID: <wz664xsbzg.fsf@nono.cs.uu.nl>

>>>>> Michael Mell <mike at nthwave.net> (MM) writes:

MM> In the string module, whitespace = ' \t\n\r\v\f' (5 characters here).
MM> Yet, when I run this:
>>>> import string
>>>> for c in string.whitespace:
>>>> print '.'+ c + '.'

MM> I get 7 results.
MM> What's going on here?

You are counting wrong.
Fist, as aomeone else already noticed, the string contains 6 characters. On
my system they are even in a different order. Try 
print string.whitespace.encode('hex') if your python isn't too old.

And then you will get more lines than 6. The \n, \v and \f will probably
get two dots on separate lines each. The \r will rpobably print one dot
only because the second dot overprints the first one. So you should get 11
dots together.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From buzzard at urubu.freeserve.co.uk  Fri Feb 15 12:39:40 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Fri, 15 Feb 2002 17:39:40 -0000
Subject: A few wxPython questions
Message-ID: <a4jh93$1f7$1@news8.svr.pol.co.uk>

Hopefully these are relatively straightforward and don't require me to post
too much code (Win 2000, Python 2.0, wxPython 2.2.5):

How can I create a wxBitmap which picks up the background colour of my
panel?  Currently I can (apparently) set it to white by using dc.Clear() as
below, but that's not really what I need.

Before I tried double buffering my drawing would resize on a panel resize
without a problem.  But this no longer happens.  On minimising the panel the
drawing remains the same size, but the lines are much heavier.  (I'm sure
the answer to this is obvious, but I just can't see it.  Should I be
deleting something somewhere?)  Ropy code below.  Thanks in advance.

Duncan Smith


    def OnToolbar1tools2Tool(self, event):
        (x, y) = self.GetClientSizeTuple()
        temp_dc = self.GetDC()
        temp_dc.SetUserScale(x/self.width, y/self.height)
                                                    #original panel height
and width
        *Draws something to temp_dc*
        self.OnDraw(temp_dc)

    def GetDC(self):
        (x, y) = self.GetClientSizeTuple()
        bmp = wxEmptyBitmap(x, y)
        temp_dc = wxMemoryDC()
        temp_dc.SelectObject(bmp)
        temp_dc.Clear()

        return temp_dc

    def OnDraw(self, temp_dc, dc=0):
        if not dc:
            dc = wxClientDC(self.panel1)
        (x, y) = self.GetClientSizeTuple()
        dc.Blit(0,0, x, y, temp_dc, 0,0)





From starx at pacbell.net  Sat Feb 16 06:21:36 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 11:21:36 GMT
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com> <b16e4ef7.0202130950.67f4004d@posting.google.com> <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4> <52e5ab5f.0202141045.5867bfbb@posting.google.com> <Xns91B71A6CA36C3RASXnewsDFE1@130.133.1.4> <tnkaeJAL$jb8EwGo@jessikat.fsnet.co.uk>
Message-ID: <Xns91B7222A074A4RASXnewsDFE1@130.133.1.4>

Robin Becker || Sat 16 Feb 2002 03:17:31a:

> .
> .....
>>
>>Also... what's this 'filename completion' you speak of? Are you 
talking 
>>about windows? (Though i s'pouse it may have been added past 98, i've 
>>never known nor can find a key that does this...)
>>
> well in cmd.exe (WinNT Win2k etc) you can configure a key to auto-
> complete filenames just like bash (I use the tab key). In the standard
> file dialogs entering a prefix usually results in a dropdown list of
> files/directories beginning with that prefix; you can then use the 
down
> arrow key to select and carriage return to finish.

Ah, 98 does that in the 'location' box in explorer windows, i guess 
they've expanded that funcion to other areas. The autocomplete is nice, 
but i use cygwin for a shell anyway, so... =).

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From mal at lemburg.com  Tue Feb 19 09:38:35 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 19 Feb 2002 15:38:35 +0100
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi> <23891c90.0202190434.147ee6d1@posting.google.com>
Message-ID: <3C72636B.C131CE2A@lemburg.com>

Paul Boddie wrote:
> 
> Ville Vainio <vvainio at karhu.tp.spt.fi> wrote in message news:<yoxr8niyiyf.fsf at karhu.tp.spt.fi>...
> >
> > Direct connection doesn't work in my case (my DBMS only offers ODBC
> > and JDBC). However, I overreacted a little bit w/ mxODBC - I thought
> > it was $500 (as it used to be), but I was emailed that it is in fact
> > $75.
> 
> You could always make a SWIG wrapper for iODBC - I started one to see
> how hard it might be - but in the end it's probably cheaper (time is
> money, so they say) to license mxODBC.

Just to correct the figures:

	mxODBC 1 CPU license: USD 75,00

The developer seat license is (no royalties for redistribution):

	mxODBC 1 DevCPU license: USD 1.250,00

mxODBC supports iODBC and unixODBC on Unix (as well as a whole
bunch of ODBC drivers directly).

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From cimarron+google at taylors.org  Wed Feb 27 16:15:19 2002
From: cimarron+google at taylors.org (Cimarron Taylor)
Date: 27 Feb 2002 13:15:19 -0800
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com>
Message-ID: <29e28c51.0202271315.59481070@posting.google.com>

Here are some online materials you might find useful to include/reference:

http://www.ifi.uio.no/in228/lecsplit/index.html
http://www.norvig.com/python-iaq.html
http://www.awaretek.com/plf.html
http://gnosis.cx/publish/tech_index_cp.html
http://manatee.mojam.com/~skip/python/fastpython.html
http://www.brunningonline.net/simon/python/quick-ref2_0.html

Cim


From magnus at thinkware.se  Thu Feb 21 18:57:50 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 00:57:50 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee> <3C74DD65.3060000@thinkware.se> <m3heoa93gs.fsf@ns.bagley.org>
Message-ID: <3C75897E.9080009@thinkware.se>

Doug Bagley the not .com spam whammy wrote:

> Magnus Lyck? <magnus at thinkware.se> writes:
> 
>>There is a comparision at http://www.bagley.org/~doug/shootout/
>>but it's really just a pointless play. Don't fall for this one.
>>
> 
> Harumph! The shootout isn't asking anyone to fall for anything. I hope
> I have enumerated the many deficiencies it has on the methodology
> page. I hope it's clear I'm not trying to pull the wool over anyone's
> eyes.


I agree with everything you write, and I'm sorry for my
careless formulation. And of course I'm aware of the
problem. A "real" application might take man-years to
develop in ONE language. Who on earth would have the
resources to do this in 30 languages. IBM maybe?

I should have said "Don't draw any conclusions regarding
real world application performance from these benchmarks."

I do think that your benchmark can be very misleading if
people think the numbers are relevant for real world
applications. Obviously any system that loads a significant
runtime system has a huge disadvantage when the test programs
are much smaller than typical, real programs. Also, making
only CPU bound applications higly exagerates the gains with
using languages like C. I rarely write programs that never
wait for file i/o, network or user interaction.

I think the main value of your site is to give a flavour
of all these languages, but frankly, I don't think the
benchmarking values are meaningful. And people will get
the wrong adea about Python. They will see some syntax,
but they won't realize how the language is really used
by python programmers.

I think it would be more valuable to try to show a few
applications where you really try to do something meaningful,
and not just implement basic algorithms. Instead of molding
all programs in the same form, try to make them utilize the
strengths (and expose the weaknesses) of each language.
But that is another benchmark than yours...

E.g. the pseudo random generator will not teach us anything about
how to use random numbers in Python. It's just a misleading example
of correct syntax. Why do you want to show something that the normal
Python programmer would hardly ever code? Using the rich standard
libraries is one of the essential ingredients in Python programming!

Actually, since your random generator is simpler than the
standard, and since the standard one is also implemented in
Python, your one is faster! But your heap sort takes 6 times
longer than Python's standard sort-method. (And if speed was
an important requirement, the Python programmer would actually
code that little part in C. But maybe throwing in mixed language
solutions in the benchmark as well would be too much... At least
if you want to try a decent amount of permutations for thirty
languages. :-) ) Anyway, running your code on my computer gives
the "impression" that C is 26 times faster than Python in sorting
doubles. But if I'm lazy and use .sort() instead of a python
implementation of heapsort, the difference is only factor 4.

I guess quicksort would be faster in C as well, but the point
is to show what each language has to offer. Surely in performance,
but first of all in how much effort it requires of the programmer!
Using the standard library random and builtin sort surely makes
that Python program nice and short.

Perhaps a few of those "give what you've got" tests could be
woven in with your tests? It seems very arbitrary to suddenly
require a certain implementation which is very far from ideal
for some languages. The least one could hope for is to have two
versions of Python programs for for instance random numbers and
sorting. At least it will give those who get their first exposure
to Python a chance to see what the language really has to offer.
I'm sure it's not only Python which is "mistreated" in this way.

It would be great with more studies like
http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprt_computer2000.pdf

 From my perspective, Python compared to the "big ones", such as
Java, Perl, C, C++, ksh and maybe VB would suffice. (I'm not joking,
VB is used to write a lot of code!) I'm curious about other languages
like icon, dylan, ocaml and haskell as well, but I could sacrifice
them.

I do think that small specific benchmarks similar to yours can have
a value. For instance, they can show us certain weaknesses, which
might give us clues about what to improve in a language. But they
don't really play a role in deciding what language to use in a
real world application. Not unless we only plan to build a system
with only tiny, short running and entirely CPU bound programs.



From philh at comuno.freeserve.co.uk  Mon Feb 11 16:39:39 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 11 Feb 2002 21:39:39 +0000
Subject: (Long) Re: Autocoding project proposal.
References: <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <uk7tkylre.fsf@synopsys.com> <a492j6$18om$1@nntp6.u.washington.edu>
Message-ID: <slrna6gegr.vsl.philh@comuno.freeserve.co.uk>

On 11 Feb 2002 18:29:58 GMT, Donn Cave <donn at u.washington.edu> wrote:
>Quoth David Masterson <dmaster at synopsys.com>:
>...
>| Or you haven't given us anything we can deal with.  Note, if we wanted
>| to remain ignorant, we could simply just stop talking to you.  You
>| have an audience and your "10 minutes of fame", but, if you cannot
>| express your problem in understandable fashion, we will go on to other
>| things that we do understand.
>
>Who's "we" - you have a mouse in your pocket? 

No, he's just pleased to see us.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From mcherm at destiny.com  Tue Feb 12 14:03:49 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 12 Feb 2002 14:03:49 -0500
Subject: I wish.............
Message-ID: <3C696715.40000@destiny.com>

> Hm, then... is Stackless a reimplementation or a port? (Or something else?)
> I don't know, myself, how extensively it is rewritten....


Stackless is more like a "branch". It's based on CPython, but with some 
changes.

-- Michael Chermside




From phr-n2002a at nightsong.com  Wed Feb 20 21:01:29 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 18:01:29 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com>
Message-ID: <7xg03vsidi.fsf@ruckus.brouhaha.com>

caljason76 at yahoo.com (Jason) writes:
> How do I sort the characters in a string, so far I use this:
>         a="qwerasdfzxcv"
>         b=[x for x in a]
>         b.sort()

You could also say 

  a = "qwerasdfzxcv"
  b = list(a)
  b.sort()

That might run a little faster than the list comprehension version,
plus it's more portable to older Python versions.

> Why doesn't sort() return the sorted list.  I would like to chain it
> to other operations:
>         b=[x for x in a].sort()

In Python, that's considered immoral--it puts you in danger of
forgetting that the sort operation modifies the list that you use it
on.  For example, if you say

   a = b.sort()

expecting to get a sorted copy of b without clobbering b, you'd get
a rude surprise.


From bokr at oz.net  Wed Feb  6 18:37:19 2002
From: bokr at oz.net (Bengt Richter)
Date: 6 Feb 2002 23:37:19 GMT
Subject: What is the command to determine an object's type?
References: <mailman.1012941850.17781.python-list@python.org> <Z9Z78.23594$l93.4286654@newsb.telia.net> <bm036ukfdun2it79e91sllcjf2n909242d@4ax.com>
Message-ID: <a3senf$3fg$0@216.39.172.122>

On Wed, 06 Feb 2002 20:31:44 +0100, Tino Lange <tl_news at nexgo.de> wrote:

>On Tue, 05 Feb 2002 22:38:49 GMT, "Fredrik Lundh"
><fredrik at pythonware.com> wrote:
>
>Hi Fredrik!
>
>(1) if type(l) == types.ListType:
>>or better:
>(2) if type(l) is types.ListType: 
>>or even better:
>(3) if isinstance(l, types.ListType):
>
>Why is "if isinstance(l, types.ListType):" even better than the rest?
>What do you mean by "better"?
>
>I just compared the speed of execution (using Python 2.1 on Linux) and
>found consistently:
>(2) is the fastest, the comes (3), (1) is much slower.
>
I suspect you might not have been timing what you thought you
were timing, but I could be wrong ;-)

>So I would guess (2) is the most effective implementation?
>
Per quick check, apparently not so on Windows Python 2.1. 
(3) is fastest, (2) is next, and (1) slowest.

[15:30] C:\pywk\crdtsc>python endsvslice9.py 10 10 850 1000 5
10 loops, 1000 avgloops, 850 tadj 5 ntrials
  0: eqeq_types:  4429, avg  4673
  1: eqeq_types:  4457, avg  4679
  2: eqeq_types:  4487, avg  4680
  3: eqeq_types:  4527, avg  4685
  4: eqeq_types:  4501, avg  4680

[15:33] C:\pywk\crdtsc>python endsvslice9.py 11 10 850 1000 5
10 loops, 1000 avgloops, 850 tadj 5 ntrials
  0:   is_types:  4105, avg  4333
  1:   is_types:  4146, avg  4322
  2:   is_types:  4204, avg  4319
  3:   is_types:  4152, avg  4328
  4:   is_types:  4096, avg  4318

[15:34] C:\pywk\crdtsc>python endsvslice9.py 12 10 850 1000 5
10 loops, 1000 avgloops, 850 tadj 5 ntrials
  0: is_instance:  4168, avg  4191
  1: is_instance:  4037, avg  4147
  2: is_instance:  4032, avg  4154
  3: is_instance:  4068, avg  4139
  4: is_instance:  4003, avg  4140



From printers at sendme.cz  Mon Feb 25 03:34:22 2002
From: printers at sendme.cz (A)
Date: Mon, 25 Feb 2002 09:34:22 +0100
Subject: Is it possible in Python too?
Message-ID: <3C7A051E.2606.E6EC3@localhost>

Hi,
Is possible to set up a proxy name (address) in MS Explorer 
browser from a running  Python program?
Thank you for your help.
Ladislav






From brueckd at tbye.com  Fri Feb 15 10:17:08 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Fri, 15 Feb 2002 07:17:08 -0800 (PST)
Subject: Java vs Python ... ???
In-Reply-To: <df3b00ed.0202150453.49e54903@posting.google.com>
Message-ID: <Pine.LNX.4.33.0202150711090.25420-100000@ns0.tbye.com>

On 15 Feb 2002, Bertrand Geston wrote:

> Martin Roos <moro at dotcom.ee> wrote in message news:<3C6CAFF8.4D3B5092 at dotcom.ee>...
> > >
> > > I've been learning JAVA for a while, but now I've heard avout Python.
> > > Any coments concerning these two languages?
> >
> > java is powerful and quite fast with jit,
> > python is more fun and slower ....
> >
[snip]
> >  java is for serious kinda stuff , python is for testing etc.
>
> I fully agree with that (but never had speed issues with Python), and
> I would like to add:
> - Java is for larges applications developed by teams, Python is for
> your own needs, for prototyping (even new functionnalities in a Java
> app with Jython and that's really cool)
> - Java is probably easier to maintain, Java works on a kind of
> platform which is robust, reliable, provide security facilities, ...
> and Python is easy to extend, gives powerfull access to the system
> (but, for obvious reasons, sometimes in a platform dependant way).

Well, since we're tossing out opinions without any explanation of them,
I'd like to add that:

- Python works extremely well for personal use, prototyping, and
professional, team development
- Python programs are just as easy, if not easier, to maintain
- Python programs are more robust and reliable
- Java gives powefull access to the system (but, for obvious reasons,
sometimes in a platform dependent way).

The OP probably needs to try both to get a clear feeling for what will
suit him best, but he/she shouldn't base the decision on the ridiculous
misinformation that has come out in this thread so far.

-Dave




From mlh at vier.idi.ntnu.no  Thu Feb 14 13:58:26 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Thu, 14 Feb 2002 18:58:26 +0000 (UTC)
Subject: PullDOM docs?
References: <slrna6l7i2.8hs.mlh@vier.idi.ntnu.no>  <m33d05jhzx.fsf@mira.informatik.hu-berlin.de>
Message-ID: <slrna6o26h.fbh.mlh@vier.idi.ntnu.no>

In article <m33d05jhzx.fsf at mira.informatik.hu-berlin.de>, Martin v.
Loewis wrote:
>mlh at vier.idi.ntnu.no (Magnus Lie Hetland) writes:
>
>> I just wondered if perhaps some more documentation could be added to
>> the standard docs about PullDOM? 
>
>Contributions of more documentation are always welcome; please submit
>them to the SF patch tracker.

I'll see if I get the time ;)

(I thought perhaps the existing documentation (i.e. Prescod's) could
just be dropped into the standard distribution...)

>> It seems to me that for people who want simple XML processing (like
>> me), but who might at times find minidom a bit inefficient
>
>I think the solution to that problem would be to speed up the minidom
>implementation. If they find it significantly too slow, pulldom won't
>help, either, and they may need to use 4Suite's cDomlette, or perhaps
>PIRXX (dunno what performance characteristics that has).

Yes, the cDomlette is extremely fast...

But is pulldom really no more efficient thatn minidom? E.g. if I'm
looking for a record in a Really Large XML file? I would have thought
that not having to build a Really Large DOM tree would help
significantly... Oh, well.

>Regards,
>Martin

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From alanmk at hotmail.com  Thu Feb  7 08:47:25 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 7 Feb 2002 05:47:25 -0800
Subject: How did TimBL's speech at Python 10 go yesterday?
Message-ID: <f72dac0d.0202070547.8cb548a@posting.google.com>

Greetings,

Unfortunately, I wasn't able to attend Python 10, even though I'd love
to have heard what Tim Berners Lee had to say about "Webizing Python".

Does anyone have any reports that they could share with us, or better
still, a transcript of his speech?

Regards,

Alan.


From shippy at nmt.edu  Wed Feb 27 10:44:20 2002
From: shippy at nmt.edu (Jeff Shipman)
Date: Wed, 27 Feb 2002 08:44:20 -0700
Subject: python stripping LD_LIBRARY_PATH?
References: <3C7BDD3A.9050905@nmt.edu> <3C7BFA13.EDEFC99F@bioreason.com>
Message-ID: <3C7CFED4.9000507@nmt.edu>

Mitch Chapman wrote:
> 
> 
> Hi Jeff,
> 
> Python is not removing LD_LIBRARY_PATH.  The runtime loader 
> ld.so is doing so.  I think this is for security reasons, 
> i.e. to prevent someone from slipping a bogus shared library
> implementation into the runtime environment of a privileged Unix
> executable.  Or something like that :)
> 
> See the ld.so(8) man page.  Here's an online copy turned up 
> by Google:
> 	http://www.cwi.nl/~aeb/linux/man2html/man8/ld.so.8.html
> 
> The relevant excerpts:
> 	The shared libraries needed by the program are searched 
> 	for in various places:
> 	...
> 	o Using the environment variable LD_LIBRARY_PATH. Except 
> 	  if the executable is a setuid/setgid binary, in which 
> 	  case it is ignored. 
> 
> 

What can be done about this? Is there a place in
the python internals I can add the necessary information
to allow me to start up? I really must run this wrapper
suid to another user and it really must connect
to Oracle.

-- 
Jeff "Shippy" Shipman     E-Mail: shippy at nmt.edu
Computer Science Major    ICQ: 1786493
New Mexico Institute of Mining and Technology
Homepage: http://www.nmt.edu/~shippy



From warkid at storm.ru  Wed Feb 20 05:42:35 2002
From: warkid at storm.ru (Kerim Borchaev)
Date: Wed, 20 Feb 2002 13:42:35 +0300
Subject: execfile question
In-Reply-To: <a4v5ug$i68$0@216.39.172.122>
References: <a4v5ug$i68$0@216.39.172.122>
Message-ID: <0571.020220@storm.ru>

Thanks to all who helped me to resolve this issue.
As I understand now dict passed as globals participates in a reference
circle which clear() nicely breaks.

Bengt Richter wrote:

> On Tue, 19 Feb 2002 16:23:19 +0300, Kerim Borchaev <warkid at storm.ru> wrote:
> 
>>Mark Barclay answered:
>>
>>> "del c" will cause your __del__() to be called.
>>yes
>>
>>> So will any other action that causes your instance to be garbage-collected.
>>I don't get it.
>>And can you explain who holds a reference on C instance in my example
>>when 'execfile' completes?
>>
> I suspect the anonymous directory you passed to execfile. Apparently
> dictionaries don't always get cleared automatically. I don't know
> if this is a bug. But if you do:
> 
> xdict = {}
> execfile('executable.py', xdict)
> xdict.clear()    # just del xdict doesn't do it
> 
> I think your example will give you what you expected.
> 
> Using default dicts like this:
> 
> execfile('executable.py')
> 
> works too.
> 
> Regards,
> Bengt Richter
> 
> 
>>Kerim.
>>
>>----------------------------------
>>My original question:
>>
>>> Can someone explain why while running this script I don't see C.__del__
>>> eventually called?
>>>
>>> For me the output of it's execution is:
>>>
>>> >main.py
>>> it's me
>>> init
>>> executed
>>> >
>>>
>>> ##main.py######################
>>> execfile('executable.py', {})
>>> print 'executed'
>>> ########################
>>>
>>> ##executable.py######################
>>> print "it's me"
>>> class C:
>>>     def __init__(self):
>>>         print 'init'
>>>     def __del__(self):
>>>         print 'deleted'
>>>
>>> c = C()
>>> ########################
>>>
>>> Thanks in advance.
>>
>>
>>
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list





From quinn at dinar.ugcs.caltech.edu  Mon Feb  4 04:01:40 2002
From: quinn at dinar.ugcs.caltech.edu (Quinn Dunkan)
Date: 4 Feb 2002 09:01:40 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrna5sjfj.fng.quinn@dinar.ugcs.caltech.edu>

On 01 Feb 2002 18:54:09 -0500, David Bolen <db3l at fitlinxx.com> wrote:
>Alex Martelli <aleax at aleax.it> writes:
>
>> If you have to *think* about it, as "memorization" would suggest, then
>> you'll probably be happier going with the rodent.  The power of vi, for
>> me (and most other vi-enthusiasts I've known) is that the *fingers*
>> learn the patterns -- sort of like in needlepoint; your conscious brain
>> can focus on higher-abstraction stuff.  (...)

Replying to this because I never saw Alex's original post...

Regarding vi-isms, I don't really have to think about them.  The vi sequence I
gave an example of earlier was touched typed as soon as I thought it.

What I'm not sure of is, even though my fingers know the patterns, is it
faster or slower than the rodent?  I have no trouble imagining my fingers
learning a slow pattern and me thinking that it's fast because I'm used to
it and because there's sure a lot of key-tapping going on.  I sometimes
catch myself doing the "arrow key thing", which is holding down 'w' or 'l'
to get to where I want to go, which is a slow and error-prone way of doing
things (but it usually doesn't *feel* like it!).

For example, in vi, the 'w' command gets stuck on the various special
characters that are so common in code, so I'm pretty trained to always use 'W'
instead.  If I want to go to a certain spot, I can count n spaces and then type
'nW'.  It's beyond my ability to count n words keeping in mind vi's set of
special characters that mark a new work in a short enough time to make it
worthwhile.

Then usually 'nW' doesn't take me exactly where I want to go, maybe I want to
be inside a set of parens that 'W' has skipped over.  So sometimes I do
'(n-1)WEhhh...' to get there.  It takes a long time to do the word-counting,
subtract one, then "arrow" over with 'h', but it doesn't *feel* like a long
time because my brain has plenty of busywork the whole time.

As a result of this clumsiness, sometimes I opt to type '/blahblah' to jump to
where I want to go instead of the 'W' stuff.  Now I need to engage the regex
brain cells, and especially vi's special character quoting rules (different
from the perl/python re rules), since I usually want to jump in the vicinity of
some )s or *s or }s.  Wait... does ')' require a backslash?  Is ':set nomagic'
on?  Once again, my brain is busy, so it feels like I'm blazing along... but
how long does it take to make all those extra decisions?

For example, I originally wrote 'saw' above as 'say'.  So I typed
'?say A<cr>erw' (incsearch is nice for that) and then '}}}}}' until I got back
down here.  Would it have been faster to click?

>As an Emacs-die-hard, I'll also agree strongly with this.  The way I
>like to put it is that I touch type Emacs editing commands nearly as
>quickly as I touch type normal text.  Most of the time, I'm rarely
>even conscious of the mechanics of achieving an edit - rather I know I
>want to "move that line over there" or something, and my fingers just
>perform the right operations.  And whether it's vi or Emacs, I find
>not having to leave the home keys and reach for a mouse (with all the
>associated eye hand coordination involved) is an enormous time saver
>for most editing tasks.

But how long does it take your fingers to perform the right operations?  In my
case, with vi, the given operations rarely get me exactly where I want to go,
and I wind up with a combination of paragraph move, word move, and letter move.
I used to not be concious of it, but I am more and more nowadays.  It feels
sloppy and clumsy and awkward.

I'm the same way for X---my setup has no window borders or titles and all
cursor movement and window focussing is done with the keyboard.  This works ok
when I have a simple tiled window layout, but when I get lots of overlapping
small windows I'd probably be better off with the mouse.  But I've developed
such an aversion to reaching for the mouse...

Perhaps it's time to try out that one-handed kbd layout + mouse on top of kbd
setup :)


From fperez528 at yahoo.com  Tue Feb 19 10:22:37 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 19 Feb 2002 15:22:37 +0000
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <a4ujb1$oau$1@peabody.colorado.edu>

Craig Maloney wrote:

> 
> I have no experience with Boost.Python, but from what I
> understand, it would not involve writing wrapper definitions
> (ala SWIG or IDL), right???Are?there?other?options?

You should take a look at weave:

http://www.scipy.org/site_content/weave

Right now it's under heavy development, and not exactly what you want (it's 
not a wrapper for already-existing complete libraries). But it's definitely 
one tool you should consider as part of the arsenal for python/C/C++ 
integration.

Cheers,

f.


From jeff at ccvcorp.com  Thu Feb 14 19:20:33 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 16:20:33 -0800
Subject: Convert String to Dictionary question
References: <a4h851$a8g7$1@ID-98166.news.dfncis.de> <Pine.SOL.4.30.0202142127220.24038-100000@mimosa.csv.warwick.ac.uk> <a4hacl$argt$1@ID-98166.news.dfncis.de>
Message-ID: <3C6C5451.6FB81F07@ccvcorp.com>


MDK wrote:

> "John J. Lee" <jjl at pobox.com> wrote in message
> news:Pine.SOL.4.30.0202142127220.24038-100000 at mimosa.csv.warwick.ac.uk...
> > On Thu, 14 Feb 2002, MDK wrote:
> >
> > > I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> > >
> > > When I do a readline the I get back a string.
> > >
> > > How do I convert this string to a dictionary?
> >
> > You probably don't want to.  Read the documentation for the pickle module.
> >
>
> Pickling seems too slow.

Are you using Pickle or cPickle?  They provide the same functionality, but
cPickle is implemented in C and is thus notably faster.  Pickle is kept around
because it's possible to subclass from it, whereas it's not possible to subclass
C code (though for most purposes, subclassing a pickler is unnecessary).

I suppose this may change soon--as of 2.2, types should be subclassable, so
there shouldn't be any further need to maintain the slower all-Python version.

Jeff Shannon
Technician/Programmer
Credit International




From phr-n2002a at nightsong.com  Mon Feb  4 18:12:32 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 15:12:32 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <a3n3q9$n1i$1@panix2.panix.com>
Message-ID: <7xg04gon8f.fsf@ruckus.brouhaha.com>

aahz at panix.com (Aahz Maruch) writes:
> >  a) If I put together a 1.5.2 bugfix release and submit it, will it go
> >     up on Python.org with a link on the front page like the current
> >     1.5.2 link?
> 
> In theory, yes.  In practice, you'll have document your work so
> carefully that it probably won't be worth it.  I don't think anyone will
> commit to making such a release available in the absence of existing
> work to be submitted.  Have you read PEP 6 yet?

That's ok, it's just a theoretical question.  The other question was
whether a 1.2.1 bugfix release would also go up, since 1.2 is not
linked from the front page now.  Versions from 1.5.2 to 2.2 are
currently linked from the front page.  Versions older than 1.5.2 are
not.  That reinforces the impression that 1.5.2 is still current
while versions like 1.2 are ancient history.


From jeff at ccvcorp.com  Fri Feb  8 16:51:01 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 13:51:01 -0800
Subject: Usenet fascism
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <kk186ugk4hpo7m0eca81bv2lm2d0smnjiv@4ax.com> <1HV88.108504$KM2.4230280@atlpnn01.usenetserver.com>
Message-ID: <3C644845.6C9805E1@ccvcorp.com>


Steve Holden wrote:

> "Courageous" <jkraska at san.rr.com> wrote in message
> news:kk186ugk4hpo7m0eca81bv2lm2d0smnjiv at 4ax.com...
> >
> > >1. Nobody or nothing gives you that right.
> >
> > That's the _one_ thing you're mistaken about. I can
> > entertain the idea of everything else you said, but
> > the notion that people don't have the right to do this
> > is just plain wrong. Plainly is the right not only held,
> > but it can be exercised.
> >
> I suspect you are confusing rights and abilities.
>
> If we are talking, and I am carrying a gun when something you say gets up my
> nose, I have the ability to shoot you in the head just to make you feel
> sorry you uttered inflammatory words.
>
> I don't have the *right* to do that, however. Most civilized countries would
> have some legal sanction if I exercised that ability.

This is true, but the express nature of Usenet (at least in unmoderated groups)
is that it is an open forum in which subjects can be freely discussed.  Part of
"free discussion" is the ability to ask people to *not* discuss certain things
(or with certain individuals).  It is, of course, the right of everyone else to
ignore or ridicule such requests, as well....  ;)

In America, we (at least some of us) are constantly reminding ourselves (and I
*hope* that it never gets forgotten!) that the right to free speech includes the
right to say potentially offensive things.  Indeed, that's the *only* sort of
speech that's worth having legal protection for.  (Where's the freedom in not
being able to complain, because complaints are "offensive"?)

Of course, just because one has a right, doesn't necessarily mean that it's
advisable to exercise it.  :)

So, to my mind, Tim Rue has a right to post whatever sort of drivel he wants.
Everyone else has the right to ignore him, and to ask others to ignore him.
Those asked to ignore him have the right to ignore those requests.  And anyone
who's really bothered by the whole mess, can just start killfiling and the
problem will rapidly disappear.  :)

Jeff Shannon
Technician/Programmer
Credit International




From gerson.kurz at t-online.de  Thu Feb 21 13:25:04 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 18:25:04 GMT
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de> <3C753E6A.9050201@free.fr>
Message-ID: <3c753acf.19446437@news.t-online.de>

On Thu, 21 Feb 2002 19:37:30 +0100, Christophe Delord
<christophe.delord at free.fr> wrote:

>The problem with sort is that it sorts the list as a side effect.

Well, if that is the whole problem:

sort = lambda x:(lambda x=x[:]:(x.sort(),x)[1])()

But your lambdas are very nice, too...




From aldo at nullcube.com  Wed Feb 20 20:51:52 2002
From: aldo at nullcube.com (Aldo Cortesi)
Date: Thu, 21 Feb 2002 12:51:52 +1100
Subject: Warning about "for line in file:"
In-Reply-To: <20020220143307.A2018@glacier.arctrix.com>
References: <just-8C3043.11330718022002@news1.xs4all.nl> <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com> <20020220143307.A2018@glacier.arctrix.com>
Message-ID: <20020221015152.GA31637@nullcube.com>

Thus spake Neil Schemenauer (nas at python.ca):

> Jason Orendorff wrote:
> > I agree!  This could be improved by having
> > file.__iter__() always return the same object.  But that
> > doesn't solve the problem in full.
> 
> Perhaps the buffer should be attached to the file object
> itself instead of to the iterator.  read() and readline()
> could then be changed to use the buffer if it exists.

... but this would leave a discrepancy between the actual
file position and the notional "read" position. Among other
things, file.tell() would then give the wrong information,
and file.write() would write to an unexpected location. 

It looks like the only complete solution (short of changing
the sematics by fiat) would be to keep the file descriptor
offset correct by performing a seek after reading every
line. Then again, this would probably kill any efficiencies
gained by xreadlines...



Cheers,


Aldo



-- 
Aldo Cortesi
aldo at nullcube.com
www.nullcube.com


-- 
Aldo Cortesi
aldo at nullcube.com
www.nullcube.com



From tim at eltig.freeserve.co.uk  Tue Feb 19 17:28:50 2002
From: tim at eltig.freeserve.co.uk (Tim Rowe)
Date: Tue, 19 Feb 2002 22:28:50 -0000
Subject: Idiots guide to code objects, frame objects and settrace?
Message-ID: <VoAc8.7320$k91.367319@wards>

Is there an idiots guide to code objects, frame objects and settrace
anywhere? I want one program to monitor the execution of another. I've
looked at coverage.py from the Vaults of Parnassus, but can't make enough
sense of it to adapt it to do what I want.




From robin at jessikat.fsnet.co.uk  Thu Feb 14 08:43:12 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 13:43:12 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk> <3C6BB8B5.8020903@mxm.dk>
Message-ID: <jPLZkiAw77a8EwM7@jessikat.demon.co.uk>

In article <3C6BB8B5.8020903 at mxm.dk>, Max M <maxm at mxm.dk> writes
>Robin Becker wrote:
>
>> I am having bad problems with zope not starting up properly. I think
>> this is caused by having Python-2.1.2 + win32all installed. It seems
>> that paths set up in the registry by win32all are getting into sys.path
>> when zope runs. This is causing massive confusion between modules
>> installed in the base python and those I wish to test with zope.
>
>I have several versions of Zope and Python + win32all installed and has 
>never had any problems.
>

well I think you just have to look at the path printed out to see that something is wrong. There should
logically be no path elements starting without the prefix c:\Python\ZOPE_2_5_0 as that is my Zope home. The
other path elements creep in because of the win32all registry paths.

I'm not doing anything here except running the python.exe that came with zope. If the path is wrong in
zopexxx\bin\python.exe there's not much I can do about it except to try and eliminate path elements in z2.py
which I'm not yet doing.

When I run zope it's via the standard service mechanism. Using procexp I can see mixed pyds if I don't
eliminate the registry path.

Certainly in the past I had different versions of python from zope and installed and they coexisted.

It seems to me that python should allow a registry free startup somehow and then the pythonservice could
pass that flag to the service.  


My feeling is that in z2.py there ought to be something like

sys.path = filter(lambda p,swhome=swhome: not p or swhome.lower()==p[0:len(swhome)].lower(),sys.path)
but it seems hard to guarantee that we get the right sys unless we mess with imp et al.

>Are you starting zope via your own script or via z2.py ??
>
....
>So that version runs on another port.
>
>All pretty painless.
>
>regards Max M
>

-- 
Robin Becker


From richie at entrian.com  Fri Feb 22 10:16:12 2002
From: richie at entrian.com (Richie)
Date: Fri, 22 Feb 2002 15:16:12 +0000
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <sknc7usegdm7b5lvu37ocaehfebv089n0i@4ax.com>

Hi Kevin,

> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

I use Westhost (www.westhost.com) - $9/month for 50MB of space with full
shell access, your own cgi-bin, Python (or plenty of space to compile your
own version if you need to), Perl, MySQL, cron, SSL and a few more
trimmings.  I've been very happy with them - I haven't done a huge amount
with my site, but everything I've done has been trouble-free.

(Plug/disclaimer: I get a month's free hosting if you mention entrian.com
 when you sign up.  8-)

-- 
Richie Hindle, rjh at entrian.com



From jason at jorendorff.com  Sat Feb 16 21:30:42 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 20:30:42 -0600
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <roy-2F6171.18085516022002@reader2.panix.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEBMDDAA.jason@jorendorff.com>

Roy Smith writes:
> Why people are still teaching or using basic today is beyond me, since 
> there are so many better alternatives.

Well, there must be reasons.  Here are my guesses.

1.  BASIC is the most popular programming language of all time.
    A lot of potential teachers don't know any easy languages
    *except* BASIC.

2.  Most Windows boxes have a BASIC implementation lying around.

3.  BASIC is very easy to learn.  In my experience with
    non-programmers, BASIC is easier than Python.  There's less
    to learn.

## Jason Orendorff    http://www.jorendorff.com/



From sjoshi at ingr.com  Sat Feb 23 09:14:14 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 23 Feb 2002 06:14:14 -0800
Subject: Problems with POST
Message-ID: <8f8ffe67.0202230614.7ea56c9c@posting.google.com>

Hello
Somehow I can't seem to get any data with POST although the same
script (with modifications in the first one) works for GET . Here's
the scripts: First one is to POST and next one to recieve the POSTED
data.

#!C:\Python21\Python.exe -u 
#To POST and output the data - postTest.py

from httplib import HTTP
from urllib import quote

quoteData = "Sunit Joshi"
postData = 'txtName=' + quote(quoteData)

print 'Will POST ', len(postData), 'bytes:', postData

# begin HTTP request
req = HTTP('sunitw2k:6080')
req.putrequest("POST" ,'/PyCgi/formTest.py')
req.putheader("Accept", "text/html")
req.putheader("User-Agent", "postTest.py")

# Set Content-length to length of PostData
req.putheader("Content-length", str(len(postData)))
req.endheaders()

# Send POST data after ending headers
req.send(postData)

replycode, message, headers = req.getreply()
print "HTTP RESPONSE: ", replycode, message
print 'HEADERS: ', str(headers)

# Get file-like object from HTTP response
# and print received HTML to screen.
fd = req.getfile()
txtlines = fd.read()
fd.close()
print '\nReceived following HTML:\n'
print txtlines

#************************
#!c:\Python21\Python.exe -u
# To send the data posted - formTest.py
import sys
import cgi

sys.stderr = sys.stdout

form = cgi.FieldStorage()
print "Content-type: text/html"
print
if not form:
	print "<h2>Field 'txtName' not filled in.</h2>"
	print str(form)
else:
	print '<h3>You entered <font color="#FF6600">%s</font></h3>' %
form.getvalue('txtName','none')


From jcollins at boulder.net  Mon Feb 11 00:03:01 2002
From: jcollins at boulder.net (Jeffery D. Collins)
Date: Sun, 10 Feb 2002 22:03:01 -0700
Subject: Is pippy alive?
References: <slrna68a44.qph.ressler@cheetah.jpl.nasa.gov>
Message-ID: <01cf01c1b2b9$623a5b10$c57ba8c0@VAIO>

Pippy is still under development, though at a much slower pace these days.
I am
considering making some changes to improve development speed and will make
an announcement, hopefully in a few days.

With regard to floating point objects, others have also asked for this
feature to
be reenabled.  I'm now looking into the code to determine the extent of the
necessary
modifications.


--
Jeffery Collins (http://www.boulder.net/~jcollins)

----- Original Message -----
From: "Michael Ressler" <ressler at cheetah.jpl.nasa.gov>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, February 08, 2002 12:35 PM
Subject: Is pippy alive?


> Is any active development still going on with pippy? I've found it to be
> quite valuable as a learning tool for python, but now that I'm actually
> ready to "do something" with it, the lack of a finished palm interface
(and
> floating point! though I can work around that) really hurts. I've checked
> the CVS and various python archives, but little has been said about it
> recently ...
>
> Mike
>
> --
> Mike Ressler
> mike.ressler at alum.mit.edu
>
>
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From gh_pythonlist at gmx.de  Tue Feb  5 15:20:03 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 5 Feb 2002 21:20:03 +0100
Subject: distutils on win32
In-Reply-To: <6bd9f01b.0202051200.74b0267b@posting.google.com>
References: <6bd9f01b.0202051200.74b0267b@posting.google.com>
Message-ID: <20020205202002.GC6560@lilith.hqd-internal>

Le 05/02/02 ? 12:00, sameer ?crivit:
> i'm using distutils to make a windows installer for a python/wxpython
> app.  what's the fastest way for me to create shortcuts on the desktop
> as well as the Start menu for the file I want the users to be able to
> run.  I don't want them to have to go into the correct package, and
> run the app from there.

If you want to make an installer for an application, distutils alone
won't help you much. You could either use py2exe or Gordon McMillan's
installer to create a standalone executable, then use a "normal" free
installer maker to create the installer (for shortcuts and stuff).

I've done that once with the McMillan installer and NSIS.

http://www.nullsoft.com/free/nsis/

It can create shortcuts, uninstall entries, ...

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.6 ?C      Wind: 2.1 m/s



From gh_pythonlist at gmx.de  Fri Feb  1 00:26:36 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Fri, 1 Feb 2002 06:26:36 +0100
Subject: VIM - python errors after upgrading to Python 2.2 (Windows OLE)
In-Reply-To: <3281a460.0201300608.2f40b08d@posting.google.com>
References: <3281a460.0201290759.78cfcd0@posting.google.com> <mailman.1012324429.18834.python-list@python.org> <3281a460.0201300608.2f40b08d@posting.google.com>
Message-ID: <20020201052635.GA1871@lilith.hqd-internal>

Le 30/01/02 ? 06:08, P Adhia ?crivit:
> Gerhard ?iso-8859-15?Q?H?ing? <gh pythonlist at gmx.de> wrote in message
> news:
> [...] 
> > Sorry, you need to recompile VIM against Python 2.2. If you can't do
> > that or find a precompiled one on the web, just mail me directly and
> > I'll build o ne this evening. If you do find a precompiled one,
> > however, would you please meail me the link?
> Unfortunately I don't have MSVC compiler (nor the expertise). But if
> you do that, please share it if you can. [...]

I've now built an installer: http://www.cs.fhm.edu/~ifw00065/vim/

What it is:

Like the Vim 6.0 installer from the Vim folks (VisVim, OLE, ...), but:

- Vim from current CVS (6.0.158+)
- Python scripting enabled and linked against Python 2.2
- Installer built with a newer version of NSIS; uses bzip2 compression,
  so the full installer is only 2.3 MB :-)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 6.8 ?C      Wind: 2.4 m/s



From christian.schmidt at REMOVEaifb.uni-karlsruhe.de  Wed Feb 27 11:33:33 2002
From: christian.schmidt at REMOVEaifb.uni-karlsruhe.de (Christian Schmidt)
Date: Wed, 27 Feb 2002 17:33:33 +0100
Subject: kill childs
Message-ID: <a5j1p9$ftb$1@news.rz.uni-karlsruhe.de>

I use Python to start other commands via os.system("cmdline")
When I kill the script with 'kill <scriptpid>' the command is not killed,
only the script.
How can I kill the child, too?
Can I use something like os.kill(-1) in an except-statement?

Thanks for any help,
Christian





From jaco at puxedo.org  Tue Feb 12 12:26:18 2002
From: jaco at puxedo.org (Jaco Greeff)
Date: 12 Feb 2002 17:26:18 -0000
Subject: Extracting bzip2 & gzip files
Message-ID: <20020212172618.10479.qmail@www2a.your-server.co.za>

Hi,

I have a function to read compressed and uncompressed files into a string.
Unfortionately, I need some help in getting the thing to actually work. As
you can see from the function below, I'm using system calls to read the file
into a string and then return it, but am receivcing errors from the actual
bzip2 execution:

"It is possible that the compressed file(s) have become corrupted.
You can use the -tvv option to test integrity of such files.

You can use the `bzip2recover' program to *attempt* to recover
data from undamaged sections of corrupted files."

I'm stumped. If I could get it to work for bzip2 only, I'll
be happy - I have looked at the pybz2 0.02 module, but it seems to not want
to compile on my system, hence my "workaround":

def readFile(file, comp='None'):
	if   comp == 'bz2': 
		r, s = commands.getstatusoutput(string.join(['bzcat', file], ' '))
	elif comp == 'gz': 
		r, s = commands.getstatusoutput(string.join(['gunzip -c', file], ' '))
	else: 
		r, s = commands.getstatusoutput(string.join(['cat', file], ' '))
	r == 1:
		print 'Error:', s
		s = ''
	return s

Any ideas? Help on this?

Thanks and greetings,
Jaco




From jason at jorendorff.com  Sun Feb  3 12:03:38 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 11:03:38 -0600
Subject: lambda expressions
In-Reply-To: <20020203161455.GA16171@kub.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLIEBDDAAA.jason@jorendorff.com>

Bas van Gils writes:
> I value readability a lot. Guess I won't be using lamba's a
> lot.

Same here.

> are lambda-statments _faster_ ?

No.

## Jason Orendorff    http://www.jorendorff.com/



From sholden at holdenweb.com  Mon Feb  4 09:48:36 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 09:48:36 -0500
Subject: Why it does NOT work on Linux ?
References: <3C5AAA9E.18034.11AB3F0@localhost> <mailman.1012677967.8260.python-list@python.org>
Message-ID: <Fdx78.39388$KM2.1215789@atlpnn01.usenetserver.com>

"Markus Sch?nhaber" <mks99 at t-online.de> wrote in message
news:mailman.1012677967.8260.python-list at python.org...
> > I have the following part of program that finds ItemID numbers.
> > Here, for example, are two
> > 146759 and 146700 .
> > This program works well under windows but on Linux it does not
> > find any number. Can you please help?
> > Thanks.
> > Ladislav
> >
> > ####################
> > import re
> > Text="""<tr BGCOLOR="#FFFFFF">
> >                       <td valign="top" align="left"><a
> > href="lead.asp?ItemID=146759">[CN] Oak, Foiled &amp; Antique
> > Furniture</a></td>
> >                       <td valign="top" align="center">18/12/2001</td>
> >                     </tr><tr BGCOLOR="#FFFFFF">
> >                     <td valign="top" align="left"><a
> > href="lead.asp?ItemID=146700">[CN] Oak, Foiled &amp; Antique
> > Furniture</a></td>
> >                       <td valign="top" align="center">18/12/2001</td>
> >                     </tr>"""
> >
> > IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> > Results=re.findall(IDs,Text)
> > print Results
>
> The interesting thing is, that it works at all for you. It definitely
> doesn't on my WinXP machine.
>
> 1.) Here
> > IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> > Results=re.findall(IDs,Text)
>
> you call re.findall with a regular expression object as a first parameter
> which should be a string. What you want to do is
>
> Results = IDs.findall(Text)
>
> i. e. call the appropriate method on the re object you created.
>
While your suggestion is correct, it is incorrect to say that Ladislav's
method is wrong. There's the same method/function pairing as there is on
strings with the string module, so it is just as possible to write

    Results=re.findall(IDs,Text)

as

    Results = IDs.findall(Text)

>
> 2.) There are two whitespaces (a space and a newline - the latter may be
> inserted by your or my mail agent) between "<a" and "href...". So you
should
> replace your re with something like this:
>
> IDs = re.compile('<a\s*href="lead.asp\?ItemID=(\d{5,10})', re.MULTILINE)
>
> Since you are using findall, the enclosing ".*" expressions are
superfluous.
>
>
> BTW: Be careful reagarding backslashes in REs since the string gets
> interpreted two times.
>
This is why most recommend using raw strings (the r"..." form) - this
removes most of the backslash's escaping behavior, making the expressions
easier to read.

Ultimately, unless you *know* how the HTML is generated and can guarantee
its format, processing HTML (or XML, or *ML ;-) with re's is likely to get
very complex very quickly, and some sort of parsing solution (such as
htmllib/sgmllib) will be more fruitful in the long-term.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From idot at vt.edu  Fri Feb 15 18:36:41 2002
From: idot at vt.edu (Carl Banks)
Date: Fri, 15 Feb 2002 18:36:41 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <a4k2si$1skr$1@norfair.nerim.net>
Message-ID: <926k4a.3a.ln@127.0.0.1>

Gillou wrote:
>>
>> * NO LINE NUMBERS! Enough said.
> ...
>>   IF condition GOSUB line-number (I never did understand the necessity
>>   for that kind of statement until well after I'd stopped using BASIC
>>   entirely and moved to more "modern" languages like... Pascal).
> 
> Only old men like you and me remember this

Hi.

I remember using calculated GOSUB, something like this, in GW-BASIC:
220 ON LEVEL GOSUB 1000,2000,3000,4000,5000,...
It was for a crude level-based video game.  I'm 25 now.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From inigoserna at terra.es  Sun Feb  3 16:51:37 2002
From: inigoserna at terra.es (=?ISO-8859-1?Q?I=F1igo?= Serna)
Date: 03 Feb 2002 22:51:37 +0100
Subject: Text to Speech
In-Reply-To: <1012342535.12379.27.camel@lethe.infernus.org>
References: <1103_1012320649@news.tin.it> 
	<1012342535.12379.27.camel@lethe.infernus.org>
Message-ID: <1012773097.1138.1.camel@lethe.infernus.org>

El mar, 29-01-2002 a las 23:15, I?igo Serna escribi?:
> El mar, 29-01-2002 a las 17:03, puCK at supereva.it_ escribi?:
> Festival (http://www.cstr.ed.ac.uk/projects/festival.html or
> http://www.festival.org) 

obviously, the second one is wrong, the correct is
http://www.festvox/festival

I?igo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 240 bytes
Desc: This is a digitally signed message part
URL: <http://mail.python.org/pipermail/python-list/attachments/20020203/27c770d4/attachment.sig>

From aahz at panix.com  Sat Feb  9 00:42:40 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 21:42:40 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org>
Message-ID: <a42csg$o9l$1@panix2.panix.com>

In article <mailman.1013216835.14351.python-list at python.org>,
Cliff Wells  <logiplexsoftware at earthlink.net> wrote:
>On 8 Feb 2002 15:55:45 -0800
>Aahz Maruch wrote:
>>
>> * Yes, I'm aware that prices have come down a lot, but the equivalent
>> flat-panel display is still twice as expensive.  And I *won't* buy
>> anything at Fry's.
>
>What's wrong with Fry's?  As long as you don't listen to the salespeople,
>that is.

Returning anything that's broken, for starters.  There have been
innumerable Fry's threads on ba.* if you want more details.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From max at malva.com.ua  Mon Feb  4 09:39:55 2002
From: max at malva.com.ua (Max Ischenko)
Date: Mon, 4 Feb 2002 16:39:55 +0200
Subject: VIM - python errors after upgrading to Python 2.2 (Windows OLE)
References: <3281a460.0201290759.78cfcd0@posting.google.com> <mailman.1012324429.18834.python-list@python.org> <3281a460.0201300608.2f40b08d@posting.google.com> <mailman.1012537622.4893.python-list@python.org>
Message-ID: <rf6m3a.189.ln@barloo.malva.com.ua>

 Gerhard H?ring wrote:


> I've now built an installer: http://www.cs.fhm.edu/~ifw00065/vim/

May be you should announce it at vim.sf.net?

> What it is:

> Like the Vim 6.0 installer from the Vim folks (VisVim, OLE, ...), but:

> - Vim from current CVS (6.0.158+)
> - Python scripting enabled and linked against Python 2.2
> - Installer built with a newer version of NSIS; uses bzip2 compression,
>  so the full installer is only 2.3 MB :-)


-- 
	"During the race
	 We may eat your dust,
	 But when you graduate,
	 You'll work for us."
	-- Reed College cheer


From steve at lurking.demon.co.uk  Tue Feb  5 12:39:44 2002
From: steve at lurking.demon.co.uk (Steve Horne)
Date: Tue, 05 Feb 2002 17:39:44 +0000
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com> <f72dac0d.0202050751.68bc0a7@posting.google.com>
Message-ID: <3j406uo6g577f6djre8jeunin6q11o8p6n@4ax.com>

On 5 Feb 2002 07:51:10 -0800, alanmk at hotmail.com (Alan Kennedy) wrote:

>Steve Horne <steve at lurking.demon.co.uk> wrote in message 
>> I haven't looked at their site, but XML (using tags for structures,
>> statements etc) may have considerable benefits over plain text as a
>> source code representation - the only problem being the need for good
>> XML editors...
>
>This has been raked over quite a few times in the past on xml-dev.
>Although it seems like a good idea at first, it rapidly degrades when
>you start looking at details.
>
>For example, how do you represent the following simple python function
>in XML, in a way that is either
>
>A: simpler to read (or write) than it already is?
>B: simpler to process than it already is, using the python compiler,
>or a parsing toolkit like SPARK?
>
>#------------
>import math
>
>def sphereVol(r):
>    return 4/3*math.pi*r*r*r
>#------------
>
>one-abstract-syntax-tree-to-rule-them-all-ly-yrs
>
>Alan.

XML is bound to be easier to parse for the compiler than the existing
Python language, because it avoids the following issues...

1.  Resolution of ambiguities using associativity, precedence and
    preference for shift over reduce.
2.  Classifying words as identifiers or keywords, particularly where
    this depends on 'from __future__'.
3.  Awkward lexical handling such as recognition of indentation.

Error reporting should also be simpler. In fact, syntax errors should
be impossible (assuming the editor is configured properly) - only
semantic errors should get as far as the compiler.

The XML would certainly not be easy for people to edit, but it's not
meant to be - the presentation is an issue for the editor.

But yes, I can see problems. For instance, the error reports would
relate to the XML AST - not to the visible representation of the
source code.

-- 
Steve Horne
steve at lurking.demon.co.uk


From brian at sweetapp.com  Tue Feb 12 17:43:41 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 12 Feb 2002 14:43:41 -0800
Subject: photo id ?!?!
In-Reply-To: <04a501c1b40c$7d17bf00$0201a8c0@josiah.dalkescientific.com>
Message-ID: <000c01c1b416$b85c9540$445d4540@Dell2>

Sorry about the first joke reply. If I understand correctly, your
objection is that you are dependant on the government for your photo id
and would prefer not to be?

Presumably you could get photo id from some non-governmental source? I
took a look at my various pieces of photo id and found:
1. An (old) student card
2. A PADI diving certification card
3. A video rental card
4. An (old) bank card with my picture on it
5. A season's (once a week) skiing pass

Plus all the government stuff.

Presumably the security requirements of the Python conference are not so
onerous that none of these pieces of non-governmental id would have been
acceptable. 

Cheers,
Brian




From jeff at ccvcorp.com  Fri Feb 22 13:09:08 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 22 Feb 2002 10:09:08 -0800
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com> <3C74AAA0.D3A050EE@macrozon.net> <3C7571A2.41D4E9E2@ccvcorp.com> <3C76551A.7060107@student.kun.nl>
Message-ID: <3C768944.59BDD67E@ccvcorp.com>

husam wrote:

>  >>> passout,passin=os.popen2("/home/husam/profit_1/profit")
>   # Also tried  calling this with "popen2.popen2()" no go.
> ....
>  >>> passin.read()
> # Here it hangs, the only way to stop it is by pressing: ctrl+c

Yes, just to make it explicit, this will block until the pipe closes, i.e.,
the external program terminates.  This can be a long time, depending on the
external program.  :)

This can be avoided by reading single lines (which blocks until a full line
is available), or by specifying a number of bytes to read (which blocks
until that many bytes are available).  I think there is also an optional
timeout, but it's been a while since I poked around with popen*() ...

(I see that you already found a solution, but I wanted to post this for the
benefit of anyone else who might be following along...)

Jeff Shannon
Technician/Programmer
Credit International




From m.faassen at vet.uu.nl  Mon Feb 11 20:44:06 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Feb 2002 01:44:06 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk>
Message-ID: <a49s16$5ao$1@newshost.accu.uu.nl>

phil hunt <philh at comuno.freeserve.co.uk> wrote:
> Either that or we're part of the Evil Programmers Conspiracy to make
> programming hard.

I'm *not* a member of the PSU. The Python Secret Underground does not
exist. I have no idea why people would think it does. Probably because
they're all incompetent. Yeah..

There is no conspiracy. Even if such a conspiracy existed, I could tell you
that there is no reason to


From grayshade at libero.it  Sun Feb 24 07:47:58 2002
From: grayshade at libero.it (Domenico)
Date: 24 Feb 2002 04:47:58 -0800
Subject: problems with pythoncom and VB-generated OLE/ActiveX DLL
References: <e34e8ce6.0202231659.609787bd@posting.google.com> <3C7841F6.8040206@skippinet.com.au>
Message-ID: <e34e8ce6.0202240447.59f5cc2e@posting.google.com>

> Assuming self.CLSID is the CLSID of the *object*, not the *DLL*, this 
> should work.

well it doesn't work the same :o\

> Have you run regsvr32.exe on the DLL?  Can you create the object from 
> another VB program, or using WSH?

I can instantiate a Class1 object from another VB program, and it
works just like it should. (well it's the last win32all version on
python 2.2b1 on a winXP pro machine)

> >>>>import pythoncom
> >>>>d = pythoncom.new("{***CLSID of the DLL***}")
> >>>>
> 
> The CLSID of the DLL will never work - it must be the object's CLSID.

Sure it is -- sorry I guess I should have checked this message once
more.

> I assume you have a good reason for using pythoncom.new() over 
> win32com.client.Dispatch()?

no I haven't -- actually win32com.client.Dispatch() fails with the
same old error:

com_error: (-2147221164, 'Unregistered interface.', None, None)

Domenico


From alanmk at hotmail.com  Tue Feb  5 10:51:10 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 5 Feb 2002 07:51:10 -0800
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com>
Message-ID: <f72dac0d.0202050751.68bc0a7@posting.google.com>

Steve Horne <steve at lurking.demon.co.uk> wrote in message 
> I haven't looked at their site, but XML (using tags for structures,
> statements etc) may have considerable benefits over plain text as a
> source code representation - the only problem being the need for good
> XML editors...

This has been raked over quite a few times in the past on xml-dev.
Although it seems like a good idea at first, it rapidly degrades when
you start looking at details.

For example, how do you represent the following simple python function
in XML, in a way that is either

A: simpler to read (or write) than it already is?
B: simpler to process than it already is, using the python compiler,
or a parsing toolkit like SPARK?

#------------
import math

def sphereVol(r):
    return 4/3*math.pi*r*r*r
#------------

one-abstract-syntax-tree-to-rule-them-all-ly-yrs

Alan.


From tim-dated-1015305638.c02155 at catseye.net  Tue Feb 26 00:20:37 2002
From: tim-dated-1015305638.c02155 at catseye.net (Tim Legant)
Date: 25 Feb 2002 23:20:37 -0600
Subject: What's wrong with this subroutine?
In-Reply-To: <a5f2oa$e7m$1@agate.berkeley.edu>
References: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>
	<a5f2oa$e7m$1@agate.berkeley.edu>
Message-ID: <86lmdgx1i2.fsf@skitty.catseye.net>

Daniel Yoo <dyoo at hkn.eecs.berkeley.edu> writes:

> Sounds good!  You might want to use a regular expression to detect
> this "url" pattern.  Here's a translation of Tom Christiansen's (of
> Perl fame) HTTP url regular expression:
> 
> 
> ###
> ## This is a regular expression that detects HTTP urls.
> ##
> ## This is only a small sample of tchrist's very nice tutorial on
> ## regular expressions.  See:
> ##
> ##     http://www.perl.com/doc/FMTEYEWTK/regexps.html
> ##
> ## for more details.
> 
> urls = '(%s)' % '|'.join("""http telnet gopher file wais ftp""".split())
> ltrs = r'\w'
> gunk = '/#~:.?+=&%@!\-'
> punc = '.:?\-'
> any = "%(ltrs)s%(gunk)s%(punc)s" % { 'ltrs' : ltrs,
>                                      'gunk' : gunk,
>                                      'punc' : punc }
> 
> url = r"""
>     \b                            # start at word boundary
>     (                             # begin \1 {
>         %(urls)s    :             # need resource and a colon
>         [%(any)s] +?              # followed by one or more
>                                   #  of any valid character, but
>                                   #  be conservative and take only
>                                   #  what you need to....
>     )                             # end   \1 }
>     (?=                           # look-ahead non-consumptive assertion
>             [%(punc)s]*           # either 0 or more punctuation
>             [^%(any)s]            #  followed by a non-url char
>         |                         # or else
>             $                     #  then end of the string
>     )
>     """ % {'urls' : urls,
>            'any' : any,
>            'punc' : punc }
> 
> url_re = re.compile(url, re.VERBOSE)

Hmmm.  Here's an URL found in "the wild" that doesn't work with this
regular expression:

http://www.wired.com/news/medtech/0,1286,50394,00.html

I don't know if commas are supposed to be encoded or not, but they're
fairly common on news sites and Christiansen's RE breaks on them.

Perhaps the OP doesn't really need to validate the URLs.  He can just
spit them back out in the chat script and let the chatters complain to
one another if someone posts a bogus URL. :)


Tim



From tom.wilkason at cox.net  Thu Feb 21 16:08:34 2002
From: tom.wilkason at cox.net (Tom Wilkason)
Date: Thu, 21 Feb 2002 16:08:34 -0500
Subject: Record Audio (Automation)
References: <140415ad.0202211224.40d8b34b@posting.google.com>
Message-ID: <oldd8.323$Bd4.2942@dfw-service2.ext.raytheon.com>

"T H" <tth_1 at yahoo.com> wrote in message
news:140415ad.0202211224.40d8b34b at posting.google.com...
| I need a simple way to record audio with a scripting language in
| Windows.  I
| think this can be done by executing sndrec.exe or sndrec32.exe (the
| sound recorder that comes with windows that can be executed by going
| to Programs->Accessories->Entertainment->Sound Recorder), however I
| can't seem to be able to make it record.  I know you can use this
| program to PLAY a WAV file by executing
| sndrec32.exe /play /close filename.wav
|
| However I was not able to make it record by executing
| sndrec32.exe /record /close filename.wav
|
| The command above I tried is just a guess, i have no idea if there is
| a /record parameter.  Any idea where I can find a list of command line
| parameters for sndrec32.exe?
|
| If anyone knows any other way of automating the task of recording
| audio using a scripting language (which executes a free audio
| recording program), please let me know.  Any help is much appreciated.

You need snack. A sound extension for Tcl/Python.
http://www.speech.kth.se/snack/

There are a number of demos to do exactly want you want.

Tom Wilkason



From mwh at python.net  Thu Feb 21 06:35:37 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 11:35:37 GMT
Subject: giant stdlib tuples (was: Re: Help with code)
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com> <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu>
Message-ID: <uadu3m5dz.fsf@python.net>

quinn at schilling.ugcs.caltech.edu (Quinn Dunkan) writes:

> On Wed, 20 Feb 2002 08:41:31 -0500, Steve Holden <sholden at holdenweb.com> wrote:
> >Besides the underscores another poster mentioned, note also that you have
> >replaced a period with a comma. Your line which reads
> >
> >            self,dst=time.localtime(self.t)
> >
> >should read
> >
> >            self.dst=time.localtime(self.t)
> 
> Speaking of time.localtime() and giant tuple matches, I always wind
> up writing my own wrappers for times and pwents so access them as
> proper objects.
> 
> It's always seemed to me that things like time, pwd, grp, and stat
> are pretty non-pythonic.  You can't even really say it's because of
> the thin wrapper thing because writing and using a StructType or
> something wouldn't be much harder than using tuples.
> 
> Of course, I've learned to carry around my own little wrappers, but
> I've always wondered why they were that way in the first place.

Used Python 2.2 yet?

Cheers,
M.

-- 
  Every now and then, Google doesn't throw up what I need so I start
  checking Altavista, Yahoo, etc.  In almost every single case, I am
  brutally reminded why I use Google in the first place.
                                                  -- John Riddoch, asr


From carlca at dircon.co.uk  Fri Feb 15 22:05:21 2002
From: carlca at dircon.co.uk (Carl Caulkett)
Date: Sat, 16 Feb 2002 03:05:21 -0000
Subject: COFF2OMF C++ Builder 4
References: <2830c89c.0202150629.5a6b711f@posting.google.com>
Message-ID: <MPG.16d7dccfbfb730039896d6@newnews.dircon.co.uk>

In article <2830c89c.0202150629.5a6b711f at posting.google.com>, 
andrew.gregory at npl.co.uk says...
> Trying to create BCB4 DLLs which can be imported into Python 2.2
> COFF2OMF python22.lib new.lib gave me error message "Invalid COFF file".
> What is the easiest way to fix this?

Hi Andrew,

This page may help. 

http://www.cyberus.ca/~g_will/pyExtenDL.shtml

It's oriented towards the Borland C++ v5.5 command line tools which is 
based on the BCB5 compiler back-end. Hopefully it will be relevant to 
BCB4 as well. I can confirm that I have been able to compile PYDs using 
the info on this page.

-- 
Cheers,
Carl


From jcollins at boulder.net  Tue Feb 19 22:36:38 2002
From: jcollins at boulder.net (Jeffery D. Collins)
Date: Tue, 19 Feb 2002 20:36:38 -0700
Subject: Python and GIS
References: <3C72F52F.5050901@sympatico.ca> <20020219173633.6da14f22.logiplexsoftware@earthlink.net> <006801c1b9be$4f2564a0$8a7ba8c0@Ron>
Message-ID: <00b201c1b9bf$d15e12e0$8a7ba8c0@Ron>

(Garbage removed)

You might also check http://openev.sourceforge.net.

--
Jeffery Collins (http://www.boulder.net/~jcollins)


>
> ----- Original Message -----
> From: "Cliff Wells" <logiplexsoftware at earthlink.net>
> To: "Ramkumar Kashyap" <rkashyap at sympatico.ca>
> Cc: <python-list at python.org>
> Sent: Tuesday, February 19, 2002 6:36 PM
> Subject: Re: Python and GIS
>
>
> > On Tue, 19 Feb 2002 20:00:31 -0500
> > Ramkumar Kashyap wrote:
> >
> > > Hi,
> > >
> > > I was wondering if there are any modules for GIS, Mapping, Layering,
or
> > > hooks to Arcinfo, Mapinfo, etc.
> >
> > A quick search on Google turned up:
> >
> > http://www.geocities.com/brucedodson.rm/avpython.htm
> >
> > --
> > Cliff Wells, Software Engineer
> > Logiplex Corporation (www.logiplex.net)
> > (503) 978-6726 x308  (800) 735-0555 x308
> >
> > --
> > http://mail.python.org/mailman/listinfo/python-list
> >
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From starx at pacbell.net  Tue Feb  5 23:57:54 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 04:57:54 GMT
Subject: Amount of files on a drive?
References: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk> <a3ov9n$dn9$1@newsg3.svr.pol.co.uk>
Message-ID: <Xns91ACD5444241CRASXnewsDFE1@130.133.1.4>

G. Willoughby || Tue 05 Feb 2002 07:56:57a:

> i am using this at the minute:
> ------------------------------------------------------------------
> import win32api
> import string
> import os
> 
> def countFiles(arg, dir, files):
>     for file in files:
>         count=count+1
> 
> fileListings=[]
> driveList=string.split(win32api.GetLogicalDriveStrings(),'\0')[:-1]
> 
> for drive in driveList:
>     os.path.walk(drive, countFiles, 0)
> ------------------------------------------------------------------
> but its pretty slow any other suggestion how i could speed it up??

Well, first determine how long it actually takes (time it), and see how it 
compares to say, selecting everything on the drive and hitting properties 
in the context menu to get a count (i.e. see how good windows itself is at 
it). It took my 800mhz athlon about 35 seconds to tell me that i have 6.15 
gigs in 101,064 files in 3,930 folders on my C drive. Counting on that 
scale can get slow.

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From aahz at panix.com  Fri Feb  1 12:46:07 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 09:46:07 -0800
Subject: isinstance() necessary and helpful, sometimes
References: <83n0z44t11.fsf@panacea.canonical.org> <a2qqd1$7q6$1@bob.news.rcn.net> <a2r97q$dpi$1@serv1.iunet.it>
Message-ID: <a3ek8v$qmj$1@panix2.panix.com>

In article <a2r97q$dpi$1 at serv1.iunet.it>, Alex Martelli <aleax at aleax.it> wrote:
>
>Wouldn't we want in this case to treat X**1.0 the same as X**1 the
>same as X**1L...?  A test such as exponent==int(exponent) would
>then seem to be more general and useful than isinstance(exponent,int).

I'd tend to code exponent==long(exponent) if I were targetting versions
of Python < 2.2.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From hungjunglu at yahoo.com  Fri Feb  8 22:32:07 2002
From: hungjunglu at yahoo.com (hungjunglu)
Date: Sat, 09 Feb 2002 03:32:07 -0000
Subject: OT: Perl programmers?
In-Reply-To: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <a4257n+mgfm@eGroups.com>

--- In python-list at y..., aahz at p... (Aahz Maruch) wrote:
> ... the way we call ourselves Pythonistas?

Never liked that name. You may call yourself Pythonista, but I prefer 
the term Pythoneer. 

regards,

Hung Jung





From bokr at oz.net  Wed Feb 27 21:03:14 2002
From: bokr at oz.net (Bengt Richter)
Date: 28 Feb 2002 02:03:14 GMT
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de> <a5j53i$oms$1@wanadoo.fr> <8z9f2rks.fsf@fido.trinitycapital.com> <a5j7fb$fb$1@wanadoo.fr>
Message-ID: <a5k352$etg$0@216.39.172.122>

On Wed, 27 Feb 2002 19:14:22 +0100, "Riss Nicolas" <alien275 at libertysurf.fr> wrote:

>No, the port is correct. The telnet server is cingured to listen of the port
>32.
>Another suggestions ?
>
>Thanks
>"Mike Johnson" <afp_randjohnson at yahoo.com> a ?crit dans le message de news:
>8z9f2rks.fsf at fido.trinitycapital.com...
>> "Riss Nicolas" <alien275 at libertysurf.fr> writes:
>>
>> > I jave used it but when y tried to connect it makes an error:
>> >    File "c:\test.py", line 9, in ?
>> >     tn = telnetlib.Telnet("192.168.0.9",32)
>> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 190, in __init__
>> >     self.open(host, port)
>> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 219, in open
>> >     raise socket.error, msg
>> > socket.error: (10061, 'Connection refused')
>> >
>> > Programm:
>> >
>> > import getpass
>> > import sys
>> > import telnetlib
>> > password = getpass.getpass()
>> > tn = telnetlib.Telnet("192.168.0.9",32)
>> > tn.read_until("Administrator Password : ")
                    ^^^^^^^^^^^^^^-?      ^-^-??
                    \_ worth double checking _/
>> > tn.write(password + "\n")
                          ^^--need \r\n for default telnet protocol, IIRC
>> > tn.write("session\n")
                      ^^--need \r\n ?
>> >
>> >
>> > Where is the mistake ?
>>
>> Wrong port. Try:
>>
>> tn = telnetlib.Telnet("192.168.0.9", 23)
>>

I'd log in manually with any normal telnet client
to double check the prompts.

HTH,
Regards,
Bengt Richter




From spahievi at vega.bg  Wed Feb  6 15:07:27 2002
From: spahievi at vega.bg (Niki Spahiev)
Date: Wed, 6 Feb 2002 22:07:27 +0200
Subject: Which GUI toolkit to use?
In-Reply-To: <3c607ffd_3@corp.newsgroups.com>
References: <3c607ffd_3@corp.newsgroups.com>
Message-ID: <2311889773.20020206220727@vega.bg>

06.2.2002, 03:07:41, Joshua Muskovitz wrote:

JM> Specifics:

JM> 1.  Needs to run on Windows platform.
JM> 2.  Needs to easily support completely custom paint (buttons, scrollers,
JM> listboxes, etc).
JM> 3.  Needs to handle simple input (editboxes, onLeftClick events).

JM> The goal is to produce an app that runs full screen on a Win platform, and
JM> is interacted with via a touchscreen.  I would like to just paint the whole
JM> damn thing myself, but a widget manager will make input handling easier.
JM> There will be a lot of bitmaps making the whole thing pretty.

I have made such app using only pygame. For touchscreen you will need
virtual keyboard which is easy with it. Nowadays you can look at PyUI and
ParaGUI python bindings.

-- 
Best regards,
 Niki Spahiev


______________________________________
VEGA(TM) Internet Service Provider
      <http://www.vega.bg>
Scanned and protected by Inflex
Tova pismo e provereno i zashtiteno
ot programite "Inflex/Antivir".





From sholden at holdenweb.com  Thu Feb  7 16:49:24 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 16:49:24 -0500
Subject: FixedPoint
References: <mailman.1013099052.17373.python-list@python.org>
Message-ID: <sDC88.96798$KM2.3611943@atlpnn01.usenetserver.com>

"Mark McEahern" <marklists at mceahern.com> wrote in message
news:mailman.1013099052.17373.python-list at python.org...
> Steve Holden wrote:
> > Since the *intent* for None is that there should only ever be one
> > object of type None, the better test would be
> >
> >     if something is None
>
> Thanks, that's how I generally test for None--I didn't know the bit about
> "there should only ever be one."
>
Well, all I really meant was that binding another name or reference to None
doesn't create a new one. The ref man says in section 3.2

"""None
This type has a single value. There is a single object with this value. This
object is accessed through the built-in name None. It is used to signify the
absence of a value in many situations, e.g., it is returned from functions
that don't explicitly return anything. Its truth value is false. """

> However, you wrote:
>
> > which can easily be used without requiring any changes to FixedPoint.
>
> Well, sure, I could easily change my local copy of PyPgSql's PgSQL.py.
But
> it seems like in this case it makes more sense for me to change my local
> copy of FixedPoint.py.  Hmm, I'm having a hard time articulating *why*
> though.  I guess because we use PyPgSql more, it's actively maintained,
> etc.--I'm better off not maintaining our own branching of that code.
>
Oh, no, don't go changing other modules to fix this. Hadn't read your post
carefully enough to realize your problems were occurring with database
access.

> Of course, I could ask the PyPgSql folks not to test for None like that.
> But then they're going to say, "how's the bug actually happening?"  And
I'm
> going to have to tell them I'm sending a FixedPoint instance as a param to
> cursor.execute().  And *that* only works because I modified
FixedPoint.py's
> __repr__ to return str(self).  As you can tell, at the moment, I'm just
> flying by the seat of my pants trying to make something work.
>
Hmm. Really what's needed is an extension to PyPGSql's NUMBER type object,
but I don't think the maintainers will want to start getting involved in
special cases.

> > Good general rule: if Tim made it that way, it's supposed to be
> > that way :-)
>
> I tend to agree.  However, do you think he purposefully avoided making the
> code resilient to frippery like this?
>
> if FixedPoint(1.0) == None:
> ...
>
> Should it be resilient to code like that?  In the case of PyPgSql, one
could
> argue I'm using it incorrectly by passing it a FixedPoint to squirt into
the
> db.  But then that merely means I haven't framed my question correctly.
;-)
>
Equality to None is something pretty basic, and should be easy to test for.
You have my permission (like you need it) to fix FixedPoint :-0

I think we might admit that Tim's code was insufficiently polymorphic with
the built-in numeric types here.

but-he-knows-where-i-live-so-i'm-saying-it-quiet-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From sjmachin at lexicon.net  Wed Feb 13 04:14:06 2002
From: sjmachin at lexicon.net (John Machin)
Date: 13 Feb 2002 01:14:06 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com> <7x1yfq2gdf.fsf@ruckus.brouhaha.com>
Message-ID: <c76ff6fc.0202130114.6c1ab5ae@posting.google.com>

Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7x1yfq2gdf.fsf at ruckus.brouhaha.com>...
> "Steve Holden" <sholden at holdenweb.com> writes:
> > Paul, this thread's probably now old enough for you to tell us what the real
> > problem is! Why exactly do you need to search backwards from the 50,000th
> > character to find the beginning of an HTML tag?
> 
> Suppose I'm parsing the file and I see a </table> tag and I want to
> find the matching <table> tag.  It could be pretty far back in the file.
> That's what I was doing when I encountered this question.

Paul, I really admire your energy, writing your own HTML parser. I'm a
lazy old so-and-so, if I had the slightest interest in HTML I'd be
looking at the HTMLParser and htmllib modules. What was there about
them that didn't suit your purpose?

... and if I had the energy to write an HTML parser I probably would
have used some dumb old technique like reading the file forwards and
if I met <foo> I'd put it in a stack (or some other data structure) of
"unclosed" tags (together with the position in the file where I found
it) and when I met the corresponding </foo> I'd take the appropriate
"foo" action (which would probably involve the use of the data that
I'd found between <foo> and </foo>)and then rip "foo" out of the
pending bag ... using a regex backwards to find the opening tag is so
innovative that I'm totally gobsmacked.

> But searching
> backwards is a normal thing to want to do in general--for example it's
> a standard command in any decent text editor.

... and in many indecent text editors. However AFAIK the
implementation is to go back a line at a time and do a forward regex
search in each line.

> 
> Anyway, I just entered a sourceforge bug about it being missing from
> Python's re module.

Looks like the effbot gets to be gobsmacked too.

> Thanks

No, Paul, thank *you* -- you've made my day.


From m.faassen at vet.uu.nl  Sun Feb  3 07:08:28 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Feb 2002 12:08:28 GMT
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com> <a3hnft$cgu$1@newshost.accu.uu.nl> <3C5CC325.7B3FC988@engcorp.com> <flhp5u4jalahtcm1att17i5vuj2evqoott@4ax.com>
Message-ID: <a3j97s$d98$1@newshost.accu.uu.nl>

Courageous <jkraska at san.rr.com> wrote:

[upon my question why they put the capslock key there]
>>Come on, that's *obvious*.  Look at your keyboard: if you took
>>away the Caps Lock key, there'd be a big empty space right there!
>>Where else would they have put it?  Sheesh...

> :)

> Some place where I wouldn't accidentally hit it and then spend
> the next few seconds wonder why vim is collapsing all my lines and
> otherwise having one giant freak fest?

> :)

By the way, to my great and enduring happiness I managed to turn it off
under XFree86. Even the light doesn't go on when I press it! I've actually 
remapped it to be an extra control key, but perhaps escape would be better
for vi use. But I don't really dare getting used to it as I'd screw up
as soon as I'd switch to anyone else's keyboard..

Regards,

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


From tanzer at swing.co.at  Sat Feb 16 02:35:33 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sat, 16 Feb 2002 08:35:33 +0100
Subject: Convert String to Dictionary question
In-Reply-To: Your message of "Fri, 15 Feb 2002 18:22:16 CST."
             <HFEKILOLEFEFMKAECNDLAEOLDCAA.jason@jorendorff.com>
Message-ID: <E16bzNd-0006sf-00@tswings.swing.cluster>

"Jason Orendorff" <jason at jorendorff.com> wrote:

>   2.  eval()
>       Well-known gaping security hole.
>
> It seems to me that security is a compelling reason to choose
> pickle() over eval().  Am I wrong?

With regard to eval, yes. Check out the two optional arguments to
eval, before spreading FUD.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From sjmachin at lexicon.net  Tue Feb  5 08:05:26 2002
From: sjmachin at lexicon.net (John Machin)
Date: Tue, 05 Feb 2002 13:05:26 GMT
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de>
Message-ID: <3c5fccf3.9456177@news.lexicon.net>

On Tue, 05 Feb 2002 10:16:36 +0100, Thomas Guettler
<st-newsgroups at thomas-guettler.de> wrote:

>I have two python version installed:
>with the cygwin version "print 2**32-1" works
>but with the zope version I get OverFlowError.
>
>Background:
>I want to get a random 32-Bit integer like this randrange(0, 2*32-1)
>
>thomas
>

Do you mean 2**31-1? Putting it another way, do you want a 32-bit
signed integer or an unsigned one?

It would help if you looked at the version *numbers* of the two
Pythons that you have; you will probably find that the "cygwin
version" is 2.2 and the "zope version" is 2.1

The problem is that 2**32 won't fit in 32 bits; Pythons before 2.2
cracked an overflow exception, whereas 2.1 promotes the result to a
long integer. Further, even 2**32-1 won't fit in a Python plain
integer which is a *signed* 32-bit integer on most/many platforms,
including yours.

Here is a quote from the Language Reference manual, section 3.2
"""
Plain integers 
These represent numbers in the range -2147483648 through 2147483647.
(The range may be larger on machines with a larger natural word size,
but not smaller.) When the result of an operation would fall outside
this range, the exception OverflowError is raised.
"""
Yes, that's from the 2.2 version, and yes, it needs updating :-)

So ... if you really want the value 2**32-1 on older Pythons, you need
to say 2L**32-1

AND in any case it looks like you're plumb out of luck with
random.randrange() .... it seems to be expecting a plain integer ...

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>>> import random
>>> random.randrange(2L**32-1)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "c:\python22\lib\random.py", line 287, in randrange
    istart = int(start)
OverflowError: long int too large to convert to int

Do you really need a 32-bit integer? Would you like to be (brace
yourself!) portable across Python versions and platforms? Try this:
>>> import sys
>>> print random.randrange(sys.maxint)
1632975901
>>> print random.randrange(sys.maxint)
156138328
>>> print random.randrange(sys.maxint)
67281065

Oh and we'd better read the doco carefully: randrange(x) is designed
to act like Python's built-in range function --- 0 <= randrange(x) <
x.

To obtain a random 32-bit unsigned integer, you will need
long(random.random() * 2L ** 32)






From carifio.nospam at nospam.usys.com  Mon Feb  4 17:45:38 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Mon, 04 Feb 2002 22:45:38 GMT
Subject: using sys.stdin.raw_input and sys.stdin.readline together in a platform independent way
Message-ID: <maE78.18262$zJ3.1413933@typhoon.maine.rr.com>

In "Programming Python" by Lutz, 2nd Ed, pp. 50-51, he indicates that if
you're
reading input from sys.stdin which has been redirected from a file, then
sys.stdin.raw_input
will be confused and will also read from that file. He then goes on to
implement a
function getreply() which will figure out how to get a character from the
console.
Is there now a platform independent way to do this in python 2.2? Pl. advise
at
your convenience. Thanks.





From loewis at informatik.hu-berlin.de  Mon Feb 25 05:18:43 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 11:18:43 +0100
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com> <m3eljb9tcv.fsf@mira.informatik.hu-berlin.de> <mailman.1014602200.11854.python-list@python.org>
Message-ID: <j4lmdhn9to.fsf@informatik.hu-berlin.de>

Kirill Simonov <kirill_simonov at mail.ru> writes:

> Tk doesn't allow to enter Cyrillic characters in the Entry and Text
> widgets if it doesn't load its encodings at startup. But Tk displays
> Cyrillic letters correctly if I set the contents of the entry from
> my program.

Do you know whether I can reproduce this on a non-Russian Windows
installation, eg. on a German XP installation?

Regards,
Martin



From shriek at gmx.co.uk  Thu Feb 28 12:01:45 2002
From: shriek at gmx.co.uk (Stephen)
Date: 28 Feb 2002 09:01:45 -0800
Subject: Want to monitor process..
References: <87550ef1.0202272315.15779bbd@posting.google.com> <f2lf8.385$Az6.1854@news.hananet.net>
Message-ID: <97ae44ee.0202280901.3153bdfc@posting.google.com>

> > At present I use a naive way to write a pid file when initiate the
> > process, but in that case I can't restart it when the process has been
> > down. (i.e., if that when the process has gone, I should kill the
> > processes and remove pid file manually.)

I'm chuckling to myself. This is exactly what I'm doing at the moment
too and it sucks. I'm so glad you asked this question, Haeyoung, and
described it so well :)


> Check out the module "atexit". You can set it up so that when the process 
> is dying, it will call the functions. One of the functions should remove 
> the pid lock file.

Unfortunately, my own applications seem to crash as opposed to
shutdown normally, so it's probably not going to help me. From the
library reference ~
"Note: the functions registered via this module are not called when
the program is killed by a signal, when a Python fatal internal error
is detected, or when os._exit() is called. "


Thanks for the great examples and explanations, Jonathan. 


> > I want to run or kill a process(program?) using python.
> 
> from os import *
> # Start a process
> ...
> pid = fork()
> if not pid:
>         # I'm a child!
>         exec*(what you want to do)
> # I'm a parent
> ...


In this case, though, if the parent process were to die, then we lose
all control.  In my own scenario I have ~

1. A SocketServer application written in Python.  This server tends to
crash quite often for reasons I'm still trying to debug.

However, even once I have debugged most the problems, I'm quite
paranoid and would like to have a totally second program ~

2. A Python program that wakes up (eg. on cronjob) every minute and
checks if the SocketServer application is running.

My current method of doing this is to write the pid to a file in (1.)
and then the monitoring program (2.) does a "ps aux" and checks if the
process in the file is running. But I really don't like this approach.

Is there a lightweight IPC mechanism that my monitoring program could
talk to the the socketserver. Yes, I could connect to the socket like
a client, but that wouldn't actually tell me if the process is
running, but rather if the server is accepting connections (and my app
server is so flaky right now that it has been known to not accept
connections but not die yet).

Thank you, 

Stephen.


From skip at pobox.com  Sat Feb  9 09:30:11 2002
From: skip at pobox.com (Skip Montanaro)
Date: Sat, 9 Feb 2002 08:30:11 -0600
Subject: Perhaps I am just dumb
In-Reply-To: <8852d33c.0202090222.34be7da2@posting.google.com>
References: <8852d33c.0202090222.34be7da2@posting.google.com>
Message-ID: <15461.12915.309189.279878@12-248-41-177.client.attbi.com>

Sheila already provided pointers to some good introductory material.  I will
comment on just one thing you said:

    wookiz> .... Yet I feel I need to be doing a university degree course in
    wookiz> philosophy or mathematics to follow almost any discussion....

Most of the discussion tends to be about more controversial aspects (or
proposed aspects) of the language.  This generally means people are
discussing stuff that is around the margins.  It sometimes tends to get a
bit esoteric, but once you have a firm grasp of the core language - the 95%
everybody tends to use for most of their work - you'll begin to understand
some of the marginal issues more.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From jeff at ccvcorp.com  Wed Feb 27 18:26:48 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 15:26:48 -0800
Subject: Python COM -- What is the name of the file?
References: <mailman.1014737443.16139.python-list@python.org> <a5gdfs$6b8$1@thorium.cix.co.uk>
Message-ID: <3C7D6B38.91A3DF8C@ccvcorp.com>


gbreed at cix.compulink.co.uk wrote:

> Mike Brenner wrote:
>
> > (And, of course, hopefully the same technique will work in Microsoft
> > Project, etc.)
>
> I think "Workbooks" is Excel specific.

But there's likely a similar set of properties for whatever Project uses to
represent a document, possibly with the same name.

Once again, you need to read the Visual Basic reference for the application
in question.  It should document the object model exposed to any scripting
language.  You just need to translate VB calls to Python -- usually a fairly
simple procedure.

Jeff Shannon
Technician/Programmer
Credit International




From jim at jim.jim  Mon Feb 18 16:00:55 2002
From: jim at jim.jim (jimh)
Date: Mon, 18 Feb 2002 13:00:55 -0800
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com> <Pine.SOL.4.30.0202182029410.13673-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <a4rq2g$70o$1@web1.cup.hp.com>

> Fix your Jython installation?!  Seriously, what is the bottleneck here?
> Clearly not the execution of the code 'print "hi"'.  I presume you are
> talking about entirely server-side execution -- or not?
>
> Google points to JPE, might be worth a look.
>

Not sure how to "fix" my Jython installation - I just installed it per the
instructions, created the following file:

   #!/usr/bin/env jython
   print "hi"

and ran it.  Took 5-6 seconds on my 400mhz Linux box.

Yes, it would be entirely server-side.

But, thanks for the JPE pointer!   I did see that when I was searching for a
solution, but I only saw the part about it gives access to Python from Java
(which I didn't need).  I didn't realize it also provides access to Java
from Python.  I will definitetly check it out!

Jim




From pobrien at orbtech.com  Mon Feb 11 14:59:17 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Mon, 11 Feb 2002 19:59:17 GMT
Subject: Best python way to do 'object messaging'
References: <Xns91AFB91DF83A2RASXnewsDFE1@130.133.1.4>
Message-ID: <poV98.2470$yV3.183554@newsread2.prod.itd.earthlink.net>

Philip,

You might find this code helpful:

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/87056

--
Patrick K. O'Brien
Orbtech


"Philip Swartzleonard" <starx at pacbell.net> wrote in message
news:Xns91AFB91DF83A2RASXnewsDFE1 at 130.133.1.4...
>
> Ok, so i have this pre-prototype (embrionic? :) game system, rogueish,
> that  basically involves a bunch of different objects running around a
> map, containing each other, beating each other up, that kind of thing
> :). What i'm trying to do is set up a general purpouse system for
> passing messages between the interface, the objects, the AI's, the game
> world and timing system, and whatever else comes up. Here are my current
> ideas/contarints:
>
[snip]
>
> This is -probably- what i'm going to go with, but i'd like to see if
> anyone has any ideas/comments/insperations/pointers-to-relevant-
> information, because i've never really made a system like this in any
> language, and i'd like to get it right before i build most of the rest
> of the system over it (i alredy have to refactor(i think that's how you
> use it) all of my display stuff out of the game data stuff, and remake
> the resource system -again- :)
>
> Thanks for anything :)
>
> --
> Philip Sw "Starweaver" [rasx] :: www.rubydragon.com
>




From feldspar at ix.netcom.com  Thu Feb 14 15:30:46 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Thu, 14 Feb 2002 16:30:46 -0400
Subject: hex to int
References: <mailman.1013552904.30812.python-list@python.org>
Message-ID: <3C6C1E76.8AE06332@ix.netcom.com>

Bjorn Pettersen wrote:
> 
> > From: maximilianscherr [mailto:MaximilianScherr at T-Online.de]
> >
> > how can i convert hex to int?
> >
> 
> I'm assuming you're asking how to convert a string representation of a
> hexadecimal to an integer...
> 
> >>> int('0xff', 16)
> 255
> 
> -- bjorn

In what version of Python do they introduce that?  In the documentation
I have int() is listed as taking one and exactly one argument.  The code
you list doesn't work for me.

	-jc


From philh at comuno.freeserve.co.uk  Fri Feb 15 10:37:37 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 15 Feb 2002 15:37:37 +0000
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131> <3C6CAFF8.4D3B5092@dotcom.ee> <df3b00ed.0202150453.49e54903@posting.google.com>
Message-ID: <slrna6qaq1.f1h.philh@comuno.freeserve.co.uk>

On 15 Feb 2002 04:53:02 -0800, Bertrand Geston <bergeston at yahoo.fr> wrote:
>
>You know, Java is my wife and Python is my mistress. 

UIs that bevcause they both demand lots of resources, and often
won't run, complaining of "insufficient cache"?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From cbarber at curl.com  Wed Feb  6 10:37:31 2002
From: cbarber at curl.com (Christopher Barber)
Date: 06 Feb 2002 10:37:31 -0500
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk>
Message-ID: <psor8nywric.fsf@jekyll.curl.com>

philh at comuno.freeserve.co.uk (phil hunt) writes:

> On Wed, 06 Feb 2002 00:04:24 GMT, Courageous <jkraska at san.rr.com> wrote:
> >
> >>What if a macro needs to create local variables in its expansion?
> >
> >This is perfectly legitimate. What's not legitimate is...
> >
> >	deflingo myterm (i, codeblock):
> >		j=3
> >
> >...where j=3 modifies the variable "j" in the context of the caller
> >if this variable is preexisting in the caller. That's non hygienic
> >behavior. Past experience with macros has shown this to be bad
> >behavior; hence there is a design in the community for good "hygiene,"
> >as it were.
> 
> And what if I decide that for my particular application, that's just
> the behaviour I want? 

There is no reason that the macro system cannot provide a means to create
"unhygienic" references, but you would not want that to be the default.

- Christopher


From rony.steelandt at bucodi.com  Fri Feb  1 03:38:10 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 1 Feb 2002 00:38:10 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com>	<20020130073810.382$1b@news.newsreader.com>	<b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org>
Message-ID: <b34d3835.0202010038.6a907bff@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1012498757.26825.python-list at python.org>...
...
> Seriously though, I do think that management-related questions are
> appropriate (as they relate to Python) and if you think about it, those
> questions aren't just management questions.  Any developer should ask
> himself these questions when considering new tools.  Actually, I have seen
> these same questions asked before albeit in a somewhat different way.
> 
> I will say that it is refreshing to see a manager active on a technical
> forum. 
 
Thank you, i was allready thinking i was intruding some kind of secret
grail only accessible for developers here <wink>

As you are probably well aware (and the last couple of posts on
> this thread should make it obvious) there is a rift between management and
> developers - developers tend to think that management is clueless and
> management tends to think that developers just want to play with new toys. 
> While there is a grain of truth there (managers tend to be less aware of
> the technical details and developers do like to try new stuff), neither
> view is true enough to ever take seriously (although when you find yourself
> in the situations where it is true it can jade you for life :P )  
>
This is exactly true, at least in my opnion, sometime i have
developers that wants to use some tools <just for the fun of it> or,
like in other messages stated here, while they think it will look good
on their CV...
On the other hand i have a naturly ***bad*** reaction on 'changing'
proposals :)
 
> However, your last statement that managers make _all_ the decisions was an
> unfortunate one.  It's better if the technicians make the decisions and
> then managers _approve_ them.  This may be splitting hairs, but when it
> comes to deciding on things like what language to develop in, who is going
> to be better informed: the people /using/ that language or someone who is
> more focused on the overall picture?  Tool choices are a detail and best
> left to those using the tools.  Managers should focus on personnel
> coordination, feature specifications, overall design, etc - the big
> picture.  It's when they start trying to specify which
> editor/debugger/compiler will be used that friction occurs.  

I'm sorry but i definitifly don't agree with that, and here i'm
speaking trough personnal experience !
Somewhere beginning 90'ths my development team came to me saying it is
time to abandon language X, that we were using, because now you have
languge Y who has all the possible benefits; better librarys, faster
development, blablabla...
And i trusted my chief development engineer and said yes.
To make a long story very short, the end of it was that the new
language din't have all the necessary function we needed, they never
found a solution (even not hired outside experts on that language) and
we trowed away one man year of development. And that is reality...
I know all the arguments, like your chief engineer didn't do his job
and how comes that it took a year and .... but since i'm the owner of
the company the problem  was mine at the end. So i'm convinced that
managment has to be involved in such decissions.


Granted, if
> you have a team and one person wants to do things differently than everyone
> else, management needs to step in, but if there is a general consensus
> amongst your developers, then you need to listen, or be prepared to look
> for new developers when your existing ones start looking elsewhere for
> employment.  If you won't use the technology they prefer, they'll find
> someone who will.
> 
> Speaking-from-personal-experience-ly yours,
> 
> -- 
> Cliff Wells
> Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308
> (800) 735-0555 x308
> 
> "Then with your new power you'll accomplish all sorts of cool stuff 
>  in no time, and We'll All Be Sorry.  At that point you can either 
>  gloat a bit, and then relent, or go ahead and send the robot army 
>  after us." - Quinn Dunkan


From philh at comuno.freeserve.co.uk  Mon Feb  4 15:45:37 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 20:45:37 +0000
Subject: Autocoding finally explained (was Re: READ THE WEB-SITE THAT THE KIKES TRY TO CENSOR)
References: <A042AF87D5C59948FB6368C76F958@news.wanadoo.fr> <mailman.1012843927.21386.python-list@python.org>
Message-ID: <slrna5tsnh.3r0.philh@comuno.freeserve.co.uk>

On Mon, 4 Feb 2002 09:21:53 -0800, Cliff Wells <logiplexsoftware at earthlink.net> wrote:
>On Mon, 4 Feb 2002 14:35:23 GMT
>charlie at news.wanadoo.fr wrote:
>> [...]
>> If you will behave Norris's road beneath cases, it will daily 
>> learn the plate.  What will we excuse after Lionel converses the 
>> ugly castle's ache?  If the filthy buckets can measure eventually, the 
>> pathetic bandage may lift more camps.  It's very sad today, I'll 
>> creep firmly or Nydia will change the wrinkles.  How doesn't 
>> Peter kill fully?  Tomorrow, it covers a cup too dry alongside her 
>> long lake.  
>
>Oh! So _that's_ what Rue meant by "autocoding"....

It all makes perfect sense now!

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From savithari at mindspring.com  Mon Feb  4 22:59:42 2002
From: savithari at mindspring.com (Tarle)
Date: Mon, 04 Feb 2002 22:59:42 -0500
Subject: python and windows short file names
References: <3C5F5711.990519D@mindspring.com>
Message-ID: <3C5F58AE.F04C3014@mindspring.com>

Oops I forgot to mention, It is for the WINDOZE platform.(xp, me, 2k, 98se
nt)


Tarle wrote:

> Hello All,
>                 I think this question must have been asked a lot of
> times before.  I could not find in the searches I did.
>
> I want to be able to use java to python ( thru jython is fine) to be
> able to manipulate the full file name and the long dir structures etc.,
>
> I would like to be able short file names and other useful file i/o's
> including ability to delete directory, create directory, rename, move
> etc., etc.,
>
> How can I accomplish this without COMMERCIAL SOFTWARE ?
>
> Thanks for your help.
>
> -Narahari.



From rupert.kolb at med.uni-tuebingen.de  Wed Feb 13 07:09:16 2002
From: rupert.kolb at med.uni-tuebingen.de (Rupert Kolb)
Date: Wed, 13 Feb 2002 13:09:16 +0100
Subject: problems with a proxy & authentication
References: <899f842.0202121213.603e6617@posting.google.com>
Message-ID: <3C6A576C.2050205@med.uni-tuebingen.de>

Anthony_Barker wrote:
> I am having problems with python and our company's proxy server, which
> requires  authentication.
 >
[...]
 > import urllib2
[...]
> 
> Both perl and wget work. wget only supports Basic Authentication, so I
> assume it is the same. Thinking it might be a network type issue I
> changed the domain names to IP addresses. (results follow)
> 
> Any help would be appreciated................

My comment does not solve the problem above, but:

In urllib2 there is proxy support.

What about ftplib?

Is there anybody who would like to implement proxy support for the
ftplib, similar to perl (or at least wget)?
Maybe there should be code for different proxy types.

--> needs some hours, if he/she is familiar with it ;-)
     I would need some weeks :-(

-- 
Rupert Kolb




From dsavitsk at e-coli.net  Sun Feb 10 22:50:08 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 03:50:08 GMT
Subject: xor on strings
Message-ID: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com>

maybe this isn't xor, but i would like something that will smush strings ...
that is, something that would compare two strings position by position and
return the higher (or lower i suppose) value piece in each position, i.e.

>>> fun('00', 'AA')
'AA'

>>> fun('00', 'A')
'A0'

>>> fun('00', '')
'00'

i understand that this is easy to write, but i was looking for a builtin.
does it exist?

-d





From vdbergh at luc.ac.be  Fri Feb  1 07:02:30 2002
From: vdbergh at luc.ac.be (Michel Van den Bergh)
Date: Fri, 01 Feb 2002 12:02:30 GMT
Subject: Authentification
Message-ID: <3C5A840D.85C673A0@luc.ac.be>

I need to add username/password authentification to my application
(a la apache or samba). So I need an encrypted password file to
authentificate against, together with some utilities to add/delete users
etc..

Is there a standard (proven) python framework I can use?

Regards,
Michel



From ewestra at wave.co.nz  Mon Feb 25 19:50:29 2002
From: ewestra at wave.co.nz (Erik Westra)
Date: 25 Feb 2002 16:50:29 -0800
Subject: py2exe for Python 2.2 not finding modules? [long]
Message-ID: <ebcd2ef6.0202251650.5207cb28@posting.google.com>

Hi,

Has anyone had any joy getting the latest version of py2exe (0.3.1)
working with Python 2.2?  I've been using py2exe successfully for
months with Python 1.5.2, but after upgrading my dev environment to
Python 2.2 I suddenly find py2exe failing to find some of the modules
in my rather large wxPython app.

My app has the following directory structure:

    bin
    build
    doc
    files
    src
    temp

The 'bin' directory contains my distutils setup script, as well as a
batch file to run py2exe.  The resulting built app gets put into the
'build' directory.  All the source files (including the modules which
can't be found by py2exe) are in the 'src' directory.

I'm running under win2k, and uninstalled Python 1.5.2 before
installing Python 2.2.  Everything else seems to work fine -- it's
just py2exe which is behaving strangely.

It's really strange that py2exe is only failing to find *some* of the
modules listed in the distutils setup script, while other files listed
in exactly the same way work fine...any ideas?

Thanks in advance,

 - Erik.

PS: Here's a copy of my distutils setup script:

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

""" 
Distutils setup script for the Rosetta application.

This is used to build a standalone runnable version of Rosetta from
the
Python sources.
"""
from distutils.core import setup
from glob import glob
import py2exe

setup(name="Rosetta",
      scripts=["../src/rosetta.py"],
      package_dir = {'': "../src"},
      py_modules = ["chatbox",
		    "constants",
		    "dialogs",
		    "inbox",
		    "jabber",
		    "mainwindow",
		    "roleinterface",
		    "rosetta",
		    "roster",
		    "server",
		    "storage",
		    "utils",
		    "wizard"],
      data_files=[("..\\temp", []),
		  ("..\\files", glob("..\\files\\*.*"))],
     )

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

And here's a log of the results of running py2exe over this setup
script:

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

"Building executable version of Rosetta..."
running py2exe
running build
running build_py
copying ../src\chatbox.py -> build\lib
copying ../src\constants.py -> build\lib
copying ../src\dialogs.py -> build\lib
copying ../src\inbox.py -> build\lib
copying ../src\jabber.py -> build\lib
copying ../src\mainwindow.py -> build\lib
copying ../src\roleinterface.py -> build\lib
copying ../src\rosetta.py -> build\lib
copying ../src\roster.py -> build\lib
copying ../src\server.py -> build\lib
copying ../src\storage.py -> build\lib
copying ../src\utils.py -> build\lib
copying ../src\wizard.py -> build\lib
running build_scripts
copying ..\src\rosetta.py -> build\scripts-2.2
running install_lib
copying build\lib\chatbox.py -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\Lib\site-packages
copying build\lib\constants.py -> build\bdist.win32\winexe\lib\Program
Files\Pyt
hon\Lib\site-packages
copying build\lib\dialogs.py -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\Lib\site-packages
copying build\lib\inbox.py -> build\bdist.win32\winexe\lib\Program
Files\Python\
Lib\site-packages
copying build\lib\jabber.py -> build\bdist.win32\winexe\lib\Program
Files\Python
\Lib\site-packages
copying build\lib\mainwindow.py ->
build\bdist.win32\winexe\lib\Program Files\Py
thon\Lib\site-packages
copying build\lib\roleinterface.py ->
build\bdist.win32\winexe\lib\Program Files
\Python\Lib\site-packages
copying build\lib\rosetta.py -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\Lib\site-packages
copying build\lib\roster.py -> build\bdist.win32\winexe\lib\Program
Files\Python
\Lib\site-packages
copying build\lib\server.py -> build\bdist.win32\winexe\lib\Program
Files\Python
\Lib\site-packages
copying build\lib\storage.py -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\Lib\site-packages
copying build\lib\utils.py -> build\bdist.win32\winexe\lib\Program
Files\Python\
Lib\site-packages
copying build\lib\wizard.py -> build\bdist.win32\winexe\lib\Program
Files\Python
\Lib\site-packages
running install_scripts
copying build\scripts-2.2\rosetta.py ->
build\bdist.win32\winexe\lib\Program Fil
es\Python\Scripts
running install_data
copying ..\files\Active.bmp -> build\bdist.win32\winexe\lib\Program
Files\Python
\..\files
copying ..\files\Blocked.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
copying ..\files\dNetLogo.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pyth
on\..\files
copying ..\files\Folder.bmp -> build\bdist.win32\winexe\lib\Program
Files\Python
\..\files
copying ..\files\Freeform.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pyth
on\..\files
copying ..\files\HideRoster.bmp ->
build\bdist.win32\winexe\lib\Program Files\Py
thon\..\files
copying ..\files\Idle.bmp -> build\bdist.win32\winexe\lib\Program
Files\Python\.
.\files
copying ..\files\MultiChat.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pyt
hon\..\files
copying ..\files\Preferences.bmp ->
build\bdist.win32\winexe\lib\Program Files\P
ython\..\files
copying ..\files\Print.bmp -> build\bdist.win32\winexe\lib\Program
Files\Python\
..\files
copying ..\files\Received.wav -> build\bdist.win32\winexe\lib\Program
Files\Pyth
on\..\files
copying ..\files\Rosetta.ico -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
copying ..\files\Sent.wav -> build\bdist.win32\winexe\lib\Program
Files\Python\.
.\files
copying ..\files\ShowRoster.bmp ->
build\bdist.win32\winexe\lib\Program Files\Py
thon\..\files
copying ..\files\SingleChat.bmp ->
build\bdist.win32\winexe\lib\Program Files\Py
thon\..\files
copying ..\files\SnapToGrid.bmp ->
build\bdist.win32\winexe\lib\Program Files\Py
thon\..\files
copying ..\files\SubBoth.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
copying ..\files\SubFrom.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
copying ..\files\SubNone.bmp -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
copying ..\files\SubTo.bmp -> build\bdist.win32\winexe\lib\Program
Files\Python\
..\files
copying ..\files\userDetailsFields.txt ->
build\bdist.win32\winexe\lib\Program F
iles\Python\..\files
copying ..\files\version.txt -> build\bdist.win32\winexe\lib\Program
Files\Pytho
n\..\files
+----------------------------------------------------
| Processing script ../src/rosetta.py with py2exe-0.3.1
+----------------------------------------------------
Copying data files
not copying ..\files\Active.bmp (output up-to-date)
not copying ..\files\Blocked.bmp (output up-to-date)
not copying ..\files\dNetLogo.bmp (output up-to-date)
not copying ..\files\Folder.bmp (output up-to-date)
not copying ..\files\Freeform.bmp (output up-to-date)
not copying ..\files\HideRoster.bmp (output up-to-date)
not copying ..\files\Idle.bmp (output up-to-date)
not copying ..\files\MultiChat.bmp (output up-to-date)
not copying ..\files\Preferences.bmp (output up-to-date)
not copying ..\files\Print.bmp (output up-to-date)
not copying ..\files\Received.wav (output up-to-date)
not copying ..\files\Rosetta.ico (output up-to-date)
not copying ..\files\Sent.wav (output up-to-date)
not copying ..\files\ShowRoster.bmp (output up-to-date)
not copying ..\files\SingleChat.bmp (output up-to-date)
not copying ..\files\SnapToGrid.bmp (output up-to-date)
not copying ..\files\SubBoth.bmp (output up-to-date)
not copying ..\files\SubFrom.bmp (output up-to-date)
not copying ..\files\SubNone.bmp (output up-to-date)
not copying ..\files\SubTo.bmp (output up-to-date)
not copying ..\files\userDetailsFields.txt (output up-to-date)
not copying ..\files\version.txt (output up-to-date)
Searching modules needed to run '../src/rosetta.py' on path:
['C:\\Documents and Settings\\Erik Westra\\My
Documents\\Rosetta-Client\\bin\\bu
ild\\bdist.win32\\winexe\\lib\\Program
Files\\Python\\Lib\\site-packages', '', '
C:\\Documents and Settings\\Erik Westra\\My
Documents\\Rosetta-Client\\bin', 'C:
\\Program Files\\Python\\DLLs', 'C:\\Program Files\\Python\\lib',
'C:\\Program F
iles\\Python\\lib\\lib-tk', 'C:\\Program Files\\Python', 'C:\\Program
Files\\Pyt
hon\\lib\\site-packages']
force_imports =
Resolving binary dependencies:
  C:\WINNT\System32\python22.dll
  C:\Program Files\Python\DLLs\_sre.pyd
  C:\Program Files\Python\lib\site-packages\wxPython\wxmsw232h.dll
  C:\Program Files\Python\lib\site-packages\py2exe\run_w.exe
  C:\Program Files\Python\lib\site-packages\wxPython\wxc.pyd
  C:\Program Files\Python\lib\site-packages\wxPython\utilsc.pyd
ext_mapping = {
 '_sre': ('_sre.pyd', ('.pyd', 'rb', 3))
 'wxPython.utilsc': ('utilsc.pyd', ('.pyd', 'rb', 3))
 'wxPython.wxc': ('wxc.pyd', ('.pyd', 'rb', 3))
}
copying C:\Program Files\Python\lib\site-packages\py2exe\support.py ->
build\bdi
st.win32\winexe\collect\rosetta\Scripts.py2exe
byte-compiling C:\Program Files\Python\lib\pre.py to pre.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\lib\__init__.p
y to wxPython\lib\__init__.pyc
byte-compiling C:\Program Files\Python\lib\__future__.py to
__future__.pyc
byte-compiling C:\Program Files\Python\lib\copy_reg.py to copy_reg.pyc
byte-compiling C:\Program Files\Python\lib\sre_compile.py to
sre_compile.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\windows.py to
wxPython\windows.pyc
byte-compiling C:\Program Files\Python\lib\locale.py to locale.pyc
byte-compiling C:\Program Files\Python\lib\macpath.py to macpath.pyc
byte-compiling C:\Program Files\Python\lib\popen2.py to popen2.pyc
byte-compiling C:\Program Files\Python\lib\stat.py to stat.pyc
byte-compiling C:\Program Files\Python\lib\imputil.py to imputil.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\lib\colourdb.p
y to wxPython\lib\colourdb.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\events.py to w
xPython\events.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\utils.py to wx
Python\utils.pyc
byte-compiling C:\Program Files\Python\lib\string.py to string.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\printfw.py to
wxPython\printfw.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\misc.py to wxP
ython\misc.pyc
byte-compiling C:\Program Files\Python\lib\tempfile.py to tempfile.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\streams.py to
wxPython\streams.pyc
byte-compiling C:\Program Files\Python\lib\sre_constants.py to
sre_constants.pyc

byte-compiling C:\Program Files\Python\lib\re.py to re.pyc
byte-compiling C:\Program Files\Python\lib\ntpath.py to ntpath.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\controls2.py t
o wxPython\controls2.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\windows3.py to
 wxPython\windows3.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\stattool.py to
 wxPython\stattool.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\controls.py to
 wxPython\controls.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\__version__.py
 to wxPython\__version__.pyc
byte-compiling C:\Program Files\Python\lib\UserDict.py to UserDict.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\wx.py to wxPyt
hon\wx.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\sizers.py to w
xPython\sizers.pyc
byte-compiling C:\Program Files\Python\lib\repr.py to repr.pyc
byte-compiling C:\Program Files\Python\lib\sre_parse.py to
sre_parse.pyc
byte-compiling C:\Program Files\Python\lib\copy.py to copy.pyc
byte-compiling C:\Program Files\Python\lib\types.py to types.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\filesys.py to
wxPython\filesys.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\windows2.py to
 wxPython\windows2.pyc
byte-compiling C:\Program Files\Python\lib\dospath.py to dospath.pyc
byte-compiling C:\Program Files\Python\lib\sre.py to sre.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\misc2.py to wx
Python\misc2.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\frames.py to w
xPython\frames.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\__init__.py to
 wxPython\__init__.pyc
byte-compiling C:\Program Files\Python\lib\posixpath.py to
posixpath.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\cmndlgs.py to
wxPython\cmndlgs.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\image.py to wx
Python\image.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\mdi.py to wxPy
thon\mdi.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\gdi.py to wxPy
thon\gdi.pyc
byte-compiling C:\Program Files\Python\lib\os.py to os.pyc
byte-compiling C:\Program
Files\Python\lib\site-packages\wxPython\clip_dnd.py to
 wxPython\clip_dnd.pyc
copying ../src/rosetta.py ->
build\bdist.win32\winexe\collect\rosetta\Scripts.py
2exe\__main__.py
changing into 'build\bdist.win32\winexe\collect\rosetta'
C:\cygwin\bin\zip.exe -rq "C:\Documents and Settings\Erik Westra\My
Documents\Ro
setta-Client\bin\build\bdist.win32\winexe\rosetta.zip" .
changing back to 'C:\Documents and Settings\Erik Westra\My
Documents\Rosetta-Cli
ent\bin'
creating ..\build\rosetta\rosetta.exe
warning: py2exe: could not parse version number ''
  No VersionInfo will be created
copying C:\WINNT\System32\python22.dll -> ..\build\rosetta
copying C:\Program Files\Python\DLLs\_sre.pyd -> ..\build\rosetta
copying C:\Program Files\Python\lib\site-packages\wxPython\wxmsw232h.dll
-> ..\b
uild\rosetta
copying C:\Program Files\Python\lib\site-packages\wxPython\wxc.pyd ->
..\build\r
osetta
copying C:\Program Files\Python\lib\site-packages\wxPython\utilsc.pyd
-> ..\buil
d\rosetta
warning: py2exe: ***************************************************************
**********
warning: py2exe: * The following modules were not found:
warning: py2exe: *   utils
warning: py2exe: *   jabber
warning: py2exe: *   roleinterface
warning: py2exe: *   os.path
warning: py2exe: *   storage
warning: py2exe: *   server
warning: py2exe: *   mainwindow
warning: py2exe: ***************************************************************
**********
Built File ..\build\rosetta\rosetta.exe
Press any key to continue . . .

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

Thanks!


From phr-n2002a at nightsong.com  Mon Feb 25 06:27:29 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 03:27:29 -0800
Subject: Time to rename Stackless? (Re: Stackless Platform Independence?)
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org>
Message-ID: <7xd6yt6btq.fsf@ruckus.brouhaha.com>

Christian Tismer <tismer at tismer.com> writes:
> > Maybe something like "Microthreaded Python" or
> > "Multitasking Python" would be better?
> 
> We have now tasklets, proclets, and the stack slices
> are stacklets. :)

What about generators (yield statement)?  Since you have microthreads,
is it now ok to yield through multiple levels of function calls?


From eric.brunel at pragmadev.com  Tue Feb  5 04:40:18 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 5 Feb 2002 09:40:18 +0000
Subject: Get current cursor position in ScrolledText
References: <4feefc55.0202042304.62cc5cc0@posting.google.com>
Message-ID: <a3o5iv$1o3s$1@norfair.nerim.net>

Hi Andrea,

> But if i bind an event to every
> possible key (e.g. <Key-A>) and i make a <ScrolledTextName>.get() i
> get the content of the ScrolledText before the 'A'-insertion happens.

I think all you've got to do is to bind on the events <KeyRelease-a>, 
<KeyRelease-b>, and so on. Your binding should then be called after the 
character(s) insertion.

HTH
 - eric -


From mail at snowfactory.com  Wed Feb 27 10:44:02 2002
From: mail at snowfactory.com (The SNOWFactory)
Date: Wed, 27 Feb 2002 07:44:02 -0800 (PST)
Subject: The SNOWFactory.com - SnowSEARCH
Message-ID: <200202271544.g1RFi2w67026@pluto.powweb.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020227/29950d42/attachment.html>

From db3l at fitlinxx.com  Mon Feb 25 20:12:29 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 20:12:29 -0500
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <ud6ytvyf6.fsf@ctwd0143.fitlinxx.com>

mtaylor at lowebroadway.com (Moray Taylor) writes:

> Before all you Emacs/vi fanatics start ranting at me, I have tried,
> and hate, them both.
> 
> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.

Given that Emacs (maybe vi too, not sure) satisfies both of your only
two requirements, what was it that led you to hate them both?  Knowing
that might help when pointing out other options to try.

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


From sjmachin at lexicon.net  Thu Feb 14 16:58:50 2002
From: sjmachin at lexicon.net (John Machin)
Date: 14 Feb 2002 13:58:50 -0800
Subject: help!: optimization of search problem (example code)
References: <3c6bcf70.16509609@news.easynews.net>
Message-ID: <c76ff6fc.0202141358.4335c894@posting.google.com>

stojek at part-gmbh.de (Marcus Stojek) wrote in message news:<3c6bcf70.16509609 at news.easynews.net>...
> Hi,
> 
> after some days of playing around I am not getting
> any further and have to ask for some more help.
> 
> Here is my problem. I have two large lists ("la" and "lb") of tuples.
> Each tuple represents a point in space and has four entries:
> x, y, z, pointID
> 
> Now I have to find out for each point of list "la" which point of
> "lb" is closest.
[snip]
> for i in range(20000):
>     a[i]=(100*random(),100*random(),100*random())
>     b[i]=(100*random(),100*random(),100*random())
[snip]
> maxdist=5.0 # I know that there are no two points with a distance > 5

The source of your knowledge is not obvious. Your random test data
(100 * random() for each coordinate) violates this assertion. You may
like to contemplate how you could write your code with no prior
knowledge of a presumed maximum distance, and no crutches like
sys.maxfloat.

> 
> for i in range(500): #should be range(len(la))
> #slicing a part out of lb that is closer than maxdist for the first
> #coordinate
>     lp=(la[i][0]-maxdist,la[i][1],la[i][2],la[i][3])
>     rp=(la[i][0]+maxdist,la[i][1],la[i][2],la[i][3])
>     left=bisect.bisect_right(lb,lp)
>     right=bisect.bisect_left(lb,rp)
>     if right==len(lb): right-=1

The above line strikes me as being very suspicious. It would seem to
be valid only as compensation for irregular behaviour by
bisect.bisect_left, which I find hard to believe. I suggest that you
check your boundary conditions very carefully. If the above line is
necessary, add an explanatory comment for the benefit of future
maintenance programmers :-)

>     lsearch=lb[left:right] # only compare la[i] to this list

Firstly, this excludes the lb[right] element; are you sure that's what
you want?

Secondly, this copies a slice of the lb list. As you are not mutating
the slice, there is no need to copy it. Try changing the inner for
loop to

for lb_index in range(left, right):
    j = lb[lb_index]

This may be a win or a loss, depending on the size of the slice.

Note the above is suggested as a formal equivalent of your code. You
really should not be using the name "j" like this ... i is an index of
la, but j is an element of lb! Take pity on those future maintenance
programmers and be consistent.
      
> 
>     mindist=10000

What happened to 5.0 ?????

[snip] 
> This takes 5.3 sec on my machine for 500 points. That is much too
> long. Any idea how I could significanty improve this code?

1. Avoid the sqrt(). If x and y are floats and x >= 0 and y >= 0 and x
< y, then you know that sqrt(x) < sqrt(y) without having to calculate
the sqrts.

2. Evaluate (for yourself, and please publish the results) which of
these is fastest in Python for squaring a float: (a) pow(x, 2) (b) x
** 2 (c) x * x

3. Hoist some of your loop-invariant expressions out of your innermost
loop:

lai0 = la[i][0]
lai1 = la[i][1]
lai2 = la[i][2]
for j in lsearch:
    if abs(j[1]-lai1) < maxdist \
    and abs(j[2]-lai2) < maxdist:
        d=sqrt(pow((j[0]-lai0),2)+pow((j[1]-lai1),2)+pow((j[2]-lai2),2))

4. Look for a better algorithm. You wouldn't be the first to have to
solve this problem.

5. Recode your best Python algorithm in C.


From rxg218 at psu.edu  Fri Feb 22 12:50:56 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Fri, 22 Feb 2002 12:50:56 -0500
Subject: PyGTK problem
Message-ID: <a560ch$i8q@r02n01.cac.psu.edu>

Hi,
  I'm not sure whether this is the right place to ask this....
I have a PyGTK app built using libglade which basically consists of a 
window containiong a text entry box.

I would like to take an action when the Return key is pressed. I made a 
signal handler for the 'changed' even in GtKEditable. However the callback 
is triggered whenever any text is entered into the box. Which is OK - but 
the callback does'nt get called when a Return is entered.

I have got around it by adding a button - but it seems kludgy compared to 
just pressing the Return key.

Does anybody have any suggestions?

-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
Dept. Of Chemistry             | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
A debugged program is one for which you have not yet found the conditions
that make it fail.
                -- Jerry Ogdin



From achim.domma at syynx.de  Sun Feb  3 13:23:34 2002
From: achim.domma at syynx.de (Achim Domma)
Date: Sun, 3 Feb 2002 19:23:34 +0100
Subject: mysterious Py_Finalize
Message-ID: <a3juta$cam$02$1@news.t-online.com>

Hi,

I'm still writing a ISAPI dll for IIS which embeds Python. After hours
searching for a reason why I could not import win32com I found out that all
my problems are gone if I NOT call Py_Finalize at the end of my code !?
If I use threads Py_Finalize crashes my IIS, if I don't use them Python
seems to remain in an undefined state after Py_Finalize.

Can somebody explain to me for what Py_Finalize is good for if it's best to
simply ignore it ?

greetings
Achim




From Michael.E.Ressler at jpl.nasa.gov  Mon Feb 11 13:54:24 2002
From: Michael.E.Ressler at jpl.nasa.gov (Michael.E.Ressler at jpl.nasa.gov)
Date: Mon, 11 Feb 2002 10:54:24 -0800 (PST)
Subject: Is pippy alive?
In-Reply-To: <01cf01c1b2b9$623a5b10$c57ba8c0@VAIO>
Message-ID: <Pine.LNX.4.33L2.0202111041410.8657-100000@cheetah.jpl.nasa.gov>

On Sun, 10 Feb 2002, Jeffery D. Collins wrote:
> With regard to floating point objects, others have also asked for this
> feature to be reenabled.  I'm now looking into the code to determine
> the extent of the necessary modifications.

I'm no programmer, but if you need an enthusiastic beta tester, I hereby
volunteer. I am also on the documentation teams of a couple of open-source
projects. If you should need assistance with the docs, perhaps I can lend
a hand there.

Concerning floating point, I assume you are aware of the "mathlib" package
for the Palm? It comes with many other opensource Palm apps, and is built
into the Visors if I recall correctly. Having Pippy use mathlib for most
of the higher order math functions (sin, cos, log, etc.) should save quite
a bit on space; of course, you will still have to provide the core
floating point add, mult, etc. since the Palm doesn't have an FPU. I
played around with Quartus Forth for a while, so I know the trouble
floating point can cause on a Palm. It's also because of (the drudgery of)
Quartus Forth that I got interested in Pippy :-)

Anyway, let me know if I can contribute in some non-programming capacity.
I'll gladly repay my "whine" with a little bit of work. I actually became
aware of Python through Pippy, and its a neat language that I've already
begun to use in a number of work projects. Being able to do the same
things (at least the simpler ones) on my Palm would be way too cool :-)

Mike

-- 
Dr. Michael Ressler
Research Scientist
Astrophysics Research Element
Jet Propulsion Laboratory
ressler at cheetah.jpl.nasa.gov





From just at xs4all.nl  Mon Feb 18 05:33:07 2002
From: just at xs4all.nl (Just van Rossum)
Date: Mon, 18 Feb 2002 11:33:07 +0100
Subject: Warning about "for line in file:"
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu>
Message-ID: <just-8C3043.11330718022002@news1.xs4all.nl>

Russell E. Owen wrote:

> I tried taking advantage of
>   for line in file:
> (a recent addition to the language) and ran into some odd behavior.
> 
> I read some lines using one for loop, to get to the start of the data of 
> interest. E.g.:
> for line in file:
>   if startingcondition(line):
>     break
> 
> I then some more lines from the same file using a similar for loop, but 
> found that the lines were not contiguous! Instead there was a huge gap 
> between the two sets of reads, as if the first for loop read in a lot of 
> extra data that it never spit out.
> 
> [ ... ]

It took me a while to understand what you meant, and the followups 
didn't make it much clearer either... But now that I do, I'm just as 
baffled as you are...

Here's what happens on my box:

  >>> f = open("tmp.txt", "w")
  >>> for i in range(10000):
  ...   f.write("%s\n" % i)
  ... 
  >>> f.close()
  >>>
  >>> f = open("tmp.txt")
  >>> for line in f:
  ...   print line.strip()
  ...   break
  ... 
  0
  >>> for line in f:
  ...   print line.strip()
  ...   break
  ... 
  1861
  >>> 

This is highly unintuitive to me, indeed so much so that I would like to 
call it a bug!

Just


From trentm at ActiveState.com  Mon Feb 11 23:12:11 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Mon, 11 Feb 2002 20:12:11 -0800
Subject: problem compiling PyWin32: no 'activdbg.h' for "Active Debugging.dsp"
Message-ID: <20020211201211.B25782@ActiveState.com>

Howdy,

I am trying to build the PyWin32 extensions from source but am unable to
build the Active Debugging project (com/Active Debugging.dsp) because its
'stdafx.h' (com/win32comext/axdebug/src/stdafx.h) includes 'activdbg.h' which
is nowhere to be found.

My internet searches indicate that 'activdbg.h' is (was) part of MS's
"Internet Client SDK". This is still mentioned in some old articles on
microsoft.com relating to IE *4*, but the actual download is gone.

I have all pieces of the latest MS Platform SDK but this does not include
'activdbg.h' (as I would have expected).

Does anybody have any clue where I might find 'activdbg.h'? Or, perhaps,
'activdbg.h' has been rolled into some other header? Stdafx.h seems to be
expecting to get the 'SOURCE_TEXT_ATTR' symbol. No current header in any of
the PyWin32, Python core, MS Visual C++, or MS Platform SDK headers seem to
define this symbol (unless, of course, some pre-processor magic is going on).

Any thoughts would be appreciated. Thanks.

Trent

-- 
Trent Mick
TrentM at ActiveState.com



From magnus at thinkware.se  Fri Feb 22 10:25:24 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 16:25:24 +0100
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014161659.19869.python-list@python.org>
Message-ID: <3C7662E4.8060402@thinkware.se>

Maybe you should have a look at weave?
http://www.scipy.org/site_content/weave

rasmussn at lanl.gov wrote:

> 
> On Tuesday, February 19, 2002, at 09:57 AM, Craig Maloney wrote:
> 
>>
>> I'm doing some scientific computing and have
>> amassed a moderate set of classes.  Due to the
>> heavy computational demands, I don't think I
>> can justify rewriting everything in an interperated
>> language; nice as that would be now that I'm
>> playing around with what I've created.
> 
> 
>> I would, of course, like to have some sort of scripted
>> environment (e.g. Python) in which to "play"
>> with my objects.
> 
> 
>> I'm making essential use of templates and polymorphism.
>>
>> I have no experience with Boost.Python, but from what I
>> understand, it would not involve writing wrapper definitions
>> (ala SWIG or IDL), right?  Are there other options?
> 
> 
> You should consider SILOON which was created for scripting
> scientific code.  It works particularly well with C++
> and works by parsing user source directly.  No wrapper definitions
> or IDL is required.  See http://www.acl.lanl.gov/siloon.
> 
> A new version is coming out in a couple of months which generates
> Python extensions for Fortran 95 (again by parsing user source
> code to discover Fortran procedures to export to Python).
> 
>> Or should I just wait for .NET/mono to come out? ;)
> 
> 
> Of course.
> 
> Craig
> 
> 



From roymath at yahoo.com  Fri Feb  8 14:40:10 2002
From: roymath at yahoo.com (Roy Mathew)
Date: 8 Feb 2002 11:40:10 -0800
Subject: py code to dial a modem
Message-ID: <7b408077.0202081140.aa2d05f@posting.google.com>

#!/usr/bin/python

"""
------------------------------------------------------------------------
Here is a simple python program to dial a modem. The tty flag setting
is heavily borrowed from a perl/C program called dialstring somewhere
on the web. I don't have the original source anymore, so this meager
acknowledgement is all I can provide. Use freely, and if you would
like to let me know that you found this code useful/buggy, I would
like to hear from you; drop me a line at "roymath @ yahoo.com". (note
spam spoiled address). So far, only tested on linux (debian).

Roy Mathew (Feb/8/2002)

USAGE COMMENTS:
Invoke the program with any number of args; the args are 
concatenated, and only the digits in the args-strings are used.

I usually grep for a a string from my contact info file, and feed it to
the program. For example:
echo "bob smith: 305-555-2222" | xargs dial.py
will cause the sequence 305-555-2222 to be dialed.
------------------------------------------------------------------------
"""

import termios, sys, os, time, string, re

# a simple function to set the modem flags for dialout.
# I found that setting the termios flags was the only slightly tricky bit,
# so this is the most useful code here.
def setFlags(fd):

  [iflag, oflag, cflag, lflag, ispeed, ospeed, cc] = termios.tcgetattr(fd)

  # input flags
  iflag = iflag & ~(termios.ISTRIP|termios.ICRNL)
  
  # output flags
  oflag = oflag & ~(termios.OPOST)
  
  # control flags
  cflag = cflag & ~(termios.CSIZE|termios.PARENB)
  cflag = cflag | termios.CS8 | termios.CLOCAL
  
  # local flags
  lflag = lflag & ~(termios.ICANON|termios.ISIG|termios.IEXTEN|termios.ECHO)
  
  cc[termios.VMIN]  = 1;
  cc[termios.VTIME] = 0;

  termios.tcsetattr(fd, termios.TCSADRAIN,
    [iflag, oflag, cflag, lflag, ispeed, ospeed, cc])

# invoke the dialing sequence; trap exceptions/interrupts such as Ctrl-C,
# and reset the modem.
def dial(device, dialStr):
  fd = os.open(device, os.O_RDWR)
  old = termios.tcgetattr(fd)
  
  try:
      setFlags(fd)  
      os.write(fd, dialStr)
  
      # this could be a loop; could log to a file.
      print os.read(fd, 1024)
      sys.stdout.flush()
          
      time.sleep(10)
  finally:
      termios.tcsetattr(fd, termios.TCSADRAIN, old)

def constructDialSequence(args):
  # remove all non-numeric digits
  num = re.sub("[^\d\*,]", "", args)
  
  # add a 1 to all non local #s within the US.
  # Don't further modify intl. numbers either.
  # (remember match only looks
  # at the beginning of the string. I live in Florida, hence 305 is local)
  if not re.match("1", num) \
     and not re.match("305", num) \
     and not re.match("786", num) \
     and not re.match("011", num):
    num = "1%s" % num
  
  # if doing international, prepend a *70
  if re.match("011", num):
    num = "*70,%s" % num
  
  #initString = "ATZ\r\d ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0\r\d"
  initString = ""
  
  dialString = "%sATS11=55DT%s\r" % (initString,num)
  return dialString


if __name__=='__main__':

  # concatenate all command line args to form a number.
  args = string.join(sys.argv[1:], "")
  dialString = constructDialSequence(args)
  
  print "dialing %s" % dialString
  dial("/dev/modem", dialString)


From graz at mindless.com  Tue Feb 12 14:10:14 2002
From: graz at mindless.com (Graham Ashton)
Date: Tue, 12 Feb 2002 19:10:14 +0000
Subject: popen3 Object
References: <3c69138a$0$12226$4d4efb8e@news.be.uu.net>
Message-ID: <WIda8.5918$ZO2.295633@stones>

On Tue, 12 Feb 2002 13:07:30 +0000, DelPiccolo Ivano wrote:

> How can I print the stdout of the popen3
> 
> exemple :
> 
> fileobjet = os.popen3("ls -la", "true" ) print ??????????

Using Python 2.1 on Linux:

ratchet% python
Python 2.1 (#8, Aug 22 2001, 16:46:14) 
[GCC 2.95.2 20000220 (Debian GNU/Linux)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import os
>>> os.popen3("echo foo", "t")
(<open file '(fdopen)', mode 'w' at 0x8134478>, <open file '(fdopen)',
mode 'r' at 0x8134168>, <open file '(fdopen)', mode 'r' at 0x8132e00>)

So we know that popen3 returns three open file objects, the first one
open for writing and the second two for reading. If you check the docs 
you'll see that they are the child's stdin, stdout and stderr. So,
returning to the interactive python session:

>>> stdIn, stdOut, stdErr = os.popen3("echo foo")
>>> while 1:
...     line = stdOut.readline()
...     if line == "": 
...         break
...     else:
...         print line,
... 
foo
>>>

Does that help any?

-- 
Graham


From logiplexsoftware at earthlink.net  Fri Feb  8 17:37:27 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 14:37:27 -0800
Subject: Please stop responding to Rue
In-Reply-To: <a41jmf$cd2$1@newshost.accu.uu.nl>
References: <b03e80d.0202060227.368f6ea7@posting.google.com>
	<3c62a989.3448437947@news>
	<a3ucsk$12h8$1@nntp6.u.washington.edu>
	<a41c4m$bf4$1@newshost.accu.uu.nl>
	<mailman.1013203955.3594.python-list@python.org>
	<a41jmf$cd2$1@newshost.accu.uu.nl>
Message-ID: <20020208143727.3d33e0c4.logiplexsoftware@earthlink.net>

On 8 Feb 2002 22:32:47 GMT
Martijn Faassen wrote:

> Cliff Wells <logiplexsoftware at earthlink.net> wrote:
> > Arguing with irrational oddballs on usenet is today's equivalent of the
> > Rubik's Cube.
> 
> I never managed to solve those either. Is it still equivalent for those
who
> do mananage to solve the cube?

I'm guessing Rue could solve it, if that answers your question ;)

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From max at alcyone.com  Thu Feb 21 13:52:56 2002
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 21 Feb 2002 10:52:56 -0800
Subject: Adding static typing to Python
References: <3C73E7EE.D0F6DE21@alcyone.com> <a52rbc$p1j$1@thorium.cix.co.uk>
Message-ID: <3C754208.A62E3E4A@alcyone.com>

gbreed at cix.compulink.co.uk wrote:

> Okay, so we have a continuum and Perl isn't quite purely weak typing. 
> But
> I still don't see a trace of strong typing, by this definition, in C. 
> I
> can also find plenty of contradictory definitions ...

One can argue where to put the fenceposts.  Certainly you can circumvent
the type system in C more easily than you can in, say, Java.  But
certainly C is more strongly typed than Perl, in which objects don't
even have any intrinsic concept of type whatsoever.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From s.schwarzer at ndh.net  Fri Feb 15 16:09:04 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Fri, 15 Feb 2002 22:09:04 +0100
Subject: Searching program for sorting cvs log output of several files
Message-ID: <3C6D78F0.F5A9A4E1@ndh.net>

Hello Pythonistas :)

I'm looking for a program who will generate the output of "cvs log"
for several files but make a single merged log of it with _all_
accumulated entries sorted by date and time.

Does anyone know such a program/module, preferably in Python? Thanks
in advance for all help.

Stefan


To clarify a bit more what I mean imagine (please excuse the long
text) the following command line session:

$ cvs log logfile.py  # ordinary cvs log
RCS file: /srv/idw3/data/cvsroot/idw3/bin/tools/logfile.py,v
Working file: logfile.py
head: 1.2
branch:
locks: strict
access list:
symbolic names:
keyword substitution: kv
total revisions: 2;     selected revisions: 2
description:
----------------------------
revision 1.2
date: 2001/11/17 16:40:31;  author: svss;  state: Exp;  lines: +13 -10
Removed method name().
Added methods error() and warning().
----------------------------
revision 1.1
date: 2001/11/16 22:40:13;  author: svss;  state: Exp;
Module to write customized logfiles.
=============================================================================

$ cvs log query_result.py  # ordinary cvs log
RCS file: /srv/idw3/data/cvsroot/idw3/bin/tools/query_result.py,v
Working file: query_result.py
head: 1.2
branch:
locks: strict
access list:
symbolic names:
keyword substitution: kv
total revisions: 2;     selected revisions: 2
description:
----------------------------
revision 1.2
date: 2002/02/09 20:12:08;  author: ifdp;  state: Exp;  lines: +5 -1
Changed QueryResult to iteration object.
----------------------------
revision 1.1
date: 2001/09/25 15:16:04;  author: ifdp;  state: Exp;
Now dbinterface is split into two modules: dbrow and query_result.
=============================================================================

$ cvslogmerge logfile.py query_result.py  # merged log, also sorted
revision 1.2 of query_result.py
date: 2002/02/09 20:12:08;  author: ifdp;  state: Exp;  lines: +5 -1
Changed QueryResult to iteration object.
----------------------------
revision 1.2 of logfile.py
date: 2001/11/17 16:40:31;  author: svss;  state: Exp;  lines: +13 -10
Removed method name().
Added methods error() and warning().
----------------------------
revision 1.1 of logfile.py
date: 2001/11/16 22:40:13;  author: svss;  state: Exp;
Module to write customized logfiles.
----------------------------
revision 1.1 of query_result.py
date: 2001/09/25 15:16:04;  author: ifdp;  state: Exp;
Now dbinterface is split into two modules: dbrow and query_result.
=============================================================================


From sdm7g at Virginia.EDU  Tue Feb 19 13:02:08 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 13:02:08 -0500 (EST)
Subject: confusion regarding os.path.walk()
In-Reply-To: <Xns91BA65E93C0ECandrewdarwinwarscom@194.168.222.19>
Message-ID: <Pine.OSX.4.43.0202191256530.1674-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, Andrew Brown wrote:

> Part 1 went fine, and left me with a text file recording the changes,
> which could then be reconstructed as a dictionary of file==>symlink
> needed pairs. But when I came to do the second part, with os.path.walk, I
> found that all the calls to make symlinks produced broken links with or
> without os.path.join(). I had to put into the visitor function a call to
> os.chdir(os.curdir) and then make the symlink with os.system
> (linkcommand). ("linkcommand" was a previously constructed string like
> 'ln-s newname oldname')
>
> I found later that I could have used os.symlink() instead of the call to
> os.system() but I'm still puzzled that you can't apparently create
> symlinks in a directory that's a subdirectory relative to where you are.
> Is this a unix/linux feature or a python one?

You *can* create symlinks in a directory that's a relative subdirectory.

You and Ron seem to be having the same sort of confusion.

I think perhaps you both are somehow thinking that os.path.walk
changes it's current directory as it 'walks' the tree: it doesn't.

The current working directory stays the same and all relative
paths are relative to it. When os.path.walk gives you a directory
and a list of files in that directory, unless that directory is
your current working directory, you need to either join the paths
or change your working directory to access the files.


-- Steve Majewski






From deathtospam43423 at altavista.com  Mon Feb 25 20:45:58 2002
From: deathtospam43423 at altavista.com (A.Newby)
Date: 26 Feb 2002 01:45:58 GMT
Subject: What's wrong with this subroutine?
Message-ID: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>

First, here's what I'm trying to do ......

I'm working on a html, cgi based chat script. It's working pretty well so 
far. 

I want to get this subroutine so that the user just has to post a url into 
the chat box to get a post a picture or a link in chat. For e.g, to post a 
pic of a walrus, you would simply type ....

http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg

... and the script would take that, and before writing it to the chat log 
it would convert it to ...

<img src = http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg>

... or to post a link, the user would simply type ... 

http://www.goodvibes.com/

... and the script would convert it to ....

<a href = "http://www.goodvibes.com/>Link!</a>


Here's the subroutine I'm using to try and do this.... it is called from a 
function named "infilter" which filters the posts before writing them to 
the chat log ...

def search4URL(string):
    #the following bits are used to alter the post when it finds the URL
    beginpic = " <img src =" 
    endpic = ">"
    beginlink = " <a href ="
    endlink = " target = blank>Link</a>"
    
    URLStarts = string.find(' http:') #space is so when the function     
	    	    	    	    	    	  #repeats, it won't convert the same 
    	    	    	    	    	    	  #block of text again

    URLEnds = string.find(' ', URLStarts+1, len(string))
    
    plaything = string[URLStarts: URLEnds]
    if plaything[-4:] == ".jpg" or plaything[-4:] == ".gif": #if the URL is     
	    	    	    	    	    	    	    	    	    	 #a pic
        nstring = string.replace(plaything,beginpic+plaything[1:]+ endpic)

    elif not plaything[-4:] == ".jpg" or plaything[-4:] == ".gif": 
    	#for some reason, "else:" return a "ValueError: empty pattern string"
        nstring=string.replace(plaything,beginlink+plaything[1:]+ endlink)
    return nstring

...... The problem (aside from the fact that someone clever could probably 
neaten it up), is that it only works ONCE per post. That means, that if a 
chatter tries to post two pics, or a pic and a link, the URLs just come out 
as plain text. Any more than once and get a "ValueError: empty pattern 
string".

I have tried fixing the routine in "infilter" so it repeats the function as 
many times as it finds the substring " http:". I have tried doing this ...

x = searchstring.count(' http:')
for c in range(0, x):
    searchstring = search4URL(searchstring)

...  And I've also tried using a while loop instead. But I keep getting the 
dread ValueError. What am I missing here? 


-- 
cdewin at dingoblue.net.au


From jeff at ccvcorp.com  Thu Feb 14 15:00:20 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 12:00:20 -0800
Subject: preference tree like in jdk 1.4
References: <mailman.1013704956.3992.python-list@python.org>
Message-ID: <3C6C1754.CA559345@ccvcorp.com>


Tripp Scott wrote:

> i was looking into the preferences api in the jdk 1.4
> documentation [1] and wonder how this could be done as simple as
> possible in python. in short, preferences is a tree branch of
> options you can store and load, just like ini files or registry settings.

Well, there's the ConfigParser standard module; I'm not sure how this compares to
what you have in mind though.  (Haven't used ConfigParser myself; the apps I've
made that needed such things were wxPython apps, so I used the similar wxConfig
instead.)

Jeff Shannon
Technician/Programmer
Credit International




From othello  Wed Feb 27 16:57:55 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 16:57:55 -0500
Subject: Status of PEP's?
References: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com> <m3k7syiqye.fsf@mira.informatik.hu-berlin.de>
Message-ID: <a5jkpa$g48$1@bob.news.rcn.net>

"Martin v. Loewis" <martin at v.loewis.de> wrote in message
news:m3k7syiqye.fsf at mira.informatik.hu-berlin.de...
> Gon?alo Rodrigues <op73418 at mail.telepac.pt> writes:
>
> > Anyone knows how is the "acceptance status" of the following PEP's, that
> > is, is there any idea if they will be accepted for the next Python
> > realease:
> >
> > PEP 276: Iterator for ints.
> > (I really love this one - as simple and "obvious" as it gets)
> >
> > PEP 279: Enhanced generators
>
> You'll need to ask the authors. If they feel it is ready, they need to
> ask for BDFL pronouncement.
>
> Regards,
> Martin
>

PEP 279 is still being tweaked. I expect to put in a version 1.5
update this weekend and go through another round of comments.
Hopefully, it will trigger some discussion on the iterators maillist.

It's pre-mature to ask for pronouncement because only one
of the top Python people has given any commentary.  I've gotten
many comments from other folks and have remedied any
defects they've pointed out.  At some point, however, I need
some input from 'bots, dictators, and gurus.

I have not gotten feedback from anyone who has tried the
genpep.py simulator.  It is a pure Python module that allows you
to directly experiment with each of the proposals:

http://sourceforge.net/tracker/download.php?group_id=5470&atid=305470&file_i
d=17348&aid=513752

A test suite with demos and examples is at:

http://sourceforge.net/tracker/download.php?group_id=5470&atid=305470&file_i
d=17412&aid=513756


Raymond Hettinger




From gmcm at hypernet.com  Sun Feb 10 10:08:56 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 10 Feb 2002 15:08:56 GMT
Subject: Python + windows + problems with make.py
References: <mailman.1013342055.30975.python-list@python.org> <F2u98.17443$n4.2988116@newsc.telia.net>
Message-ID: <Xns91B16764CFE86gmcmhypernetcom@199.171.54.216>

Fredrik Lundh wrote:

> "gombosk at vas.birosag.hu" wrote:
>> 3. I've got the next:
>>
>> reading exceptions from
>> Traceback (most recent call last):
>>   File "./Make.py", line 133, in ?
>>     main()
>>   File "./Make.py", line 73, in main
>>     print "reading exceptions from", exceptions.__file__
>> AttributeError: 'exceptions' module has no attribute '__file__'
>>
>> Yes, exceptions module has no attribue '__file__'
>>
>> What can I do
> 
> remove the print statement?
> 
> blame it all on me and barry?

Sounds good to me :-).
 
> in recent versions of Python, exceptions is a built-in module
> (on all platforms).
> 
> if you really need to know where the standard library is, check
> the __file__ attribute of some other module (e.g. os or string),
> use sys.prefix, or scan the sys.path list.

For Python 1.5, Installer on Linux freezes exceptions.py
into the embedding executable. In later versions, this
is conditional, but the original poster was using a version
written when 1.5 was current.

-- Gordon
http://www.mcmillan-inc.com/


From knight at baldmt.com  Fri Feb 22 11:43:12 2002
From: knight at baldmt.com (Steven Knight)
Date: Fri, 22 Feb 2002 10:43:12 -0600 (CST)
Subject: SCons 0.05 is now available
Message-ID: <Pine.LNX.4.44.0202221042050.8318-100000@theopera.home.baldmt.com>

Version 0.05 of SCons has been released and is available for download
from the SCons web site:

        http://www.scons.org/

Or through the download link at the SCons project page at SourceForge:

        http://sourceforge.net/projects/scons/

RPM and Debian packages and a Win32 installer are all available, in
addition to the traditional .tar.gz files.


WHAT'S NEW IN THIS RELEASE?

Version 0.05 of SCons adds the following features:
  - A library can be specified as a $SOURCES file, not just in the LIBS
    construction variable.
  - Scanners can now return a list of strings.
  - .sconsign files are first written to a temporary file and renamed.
  - The list of variables exported to an SConscript file may now be
    a UserList, too.
  - Preliminary support for Unicode strings has been added.
  - The -q option has been added.
  - The -u option has been added.
  - SCons now scans files with .cc and .hh suffixes for #include lines.
  - A Builder's Action may now be a Python code object that *returns* a
    command line to be executed.

The following fixes have been added:
  - PROGSUFFIX is now set to .exe under Cygwin.
  - Work around for a bug in os.path.normpath() returning '' for './'
    on WIN32 systems.
  - White space in a PREFIX or SUFFIX creates separate command-line
    arguments as appropriate.
  - Fetching MicroSoft DevStudio information from the Windows
    registry has been fixed and made more robust.
  - sys.stdout is now flushed after print so it intermixes correctly
    with sys.stderr output.
  - Scanned dependencies are now relative to the directory of the file
    being scanned.
  - All of the directories for a list of targets are created before
    trying to build any of the targets.
  - Ignore() now ignores both indirect (scanned) and direct dependencies
  - The -c option no longer stops if a target doesn't exist to be removed.
  - The --debug=pdb option now works on WIN32 systems.
  - The scons script now returns an error code on build failures.

Performance has been improved as follows:
  - #include lines from a file are now cached once.
  - Dependencies are now scanned only once per file when walking the
    dependency tree.

The documentation has been improved:
  - Writing your own Scanners is now documented.
  - The LIBPATH construction variable is now documented.
  - The man page is now included in the Debian distribution.
  - HTML documents are cleaned up by running them through tidy.


WHAT IS SCONS?

SCons is a software construction tool (build tool, make tool) written
in Python.  Its design is based on the design which won the Software
Carpentry build tool competition in August 2000 (in turn derived from
the Perl-based Cons build tool).

Distinctive features of SCons include:

  - configuration files are Python scripts, allowing the full use of a
    real scripting language to solve build problems
  - a modular architecture allows the SCons Build Engine to be
    embedded in other Python software
  - a global view of all dependencies; no multiple passes to get
    everything built
  - the ability to scan files for implicit dependencies (#include files);
  - improved parallel build (-j) support
  - use of MD5 signatures to decide if a file has changed
  - easily extensible through user-defined Builder and Scanner objects

An scons-users mailing list has been created for those interested in
getting started using SCons.  You can subscribe at:

        http://lists.sourceforge.net/lists/listinfo/scons-users

Alternatively, we invite you to subscribe to the low-volume
scons-announce mailing list to receive notification when new versions of
SCons become available:

        http://lists.sourceforge.net/lists/listinfo/scons-announce


ACKNOWLEDGEMENTS

Special thanks to Chad Austin, Charles Crain, Steve Leblanc, and Anthony
Roach for their contributions to this release.

On behalf of the SCons team,

	--SK




From logiplexsoftware at earthlink.net  Mon Feb 18 12:34:55 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 18 Feb 2002 09:34:55 -0800
Subject: Free Software For You .
In-Reply-To: <1cef01c1b6da$944ca020$28eb42d9@supervisor>
References: <1cef01c1b6da$944ca020$28eb42d9@supervisor>
Message-ID: <20020218093455.41c56955.logiplexsoftware@earthlink.net>

On Sat, 16 Feb 2002 13:10:45 +0200
feedback at cd-house.net wrote:

> Do You Need Free Software in the fields of
> PC Games , Graphics , Applications , DVD , VCD , Music Audio , MP3 ,
Educational Software and many more ............

> JUST Visit : http://www.cd-house.net
> and don't forget to join our membership for free to get free software

Sic 'em, Oleg!

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From kp87 at lycos.com  Fri Feb 15 10:50:03 2002
From: kp87 at lycos.com (kevin parks)
Date: 15 Feb 2002 07:50:03 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net> <mailman.1013718563.21631.python-list@python.org> <5e8bd451.0202142238.580f15c1@posting.google.com> <7xzo2bnt1m.fsf@ruckus.brouhaha.com> <just-26D236.09311915022002@news1.xs4all.nl>
Message-ID: <5e8bd451.0202150750.7e9dc295@posting.google.com>

> Nah, just do this:
> 
>   yourlist.sort()
> 
> Just

so does that mean that what i want is the default behavior of the sort
method?
it sorts by column one then sorts by column 2m, then column 3.....
?

Gosh, i've got no interpreter here (@ an internet cafe). Ive got to
run home and see if that is waht really happens...

-kevin


From marklists at mceahern.com  Mon Feb 11 14:16:52 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Mon, 11 Feb 2002 11:16:52 -0800
Subject: Option Parsing Libraries
In-Reply-To: <3C67F75D.D1EF37DA@prescod.net>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEDHKCAA.marklists@mceahern.com>

[Paul Prescod]
> If you have a competitive library, or suggestions for changes to Optik,
> please forward your comments to python-dev mailing list
> (python-dev at python.org).

I love optik.  We use it for all of our option parsing.

I have one feature request related to error handling.  I've attached sample
code below that shows a common thing I end up doing:  raising an error if a
required option is missing.  I guess it's not really an option, then, is it?
<wink>  Anyway, I searched optik's documentation for some way to "inspect"
the options collection itself for the original information used when
creating the option.  In the code below, you'll notice the requiredVar
method takes a description parameter.  It would be nice to be able to do
something like this instead:

	if var is None:
		parser.error("Missing: %s" % options.var.description)

In fact, it would seem that this itself is so common it would be "built-in"
to optik itself.  So that all I have to do is declare an option as required
when I add it with add_option?

Thanks,

// mark

#! /usr/bin/env python
# testError.py

from optik import OptionParser

def requiredVar(parser, options, var, description):
    """Raise a parser error if var is None."""
    if var is None:
        # Here's where it'd be nice to have access to the attributes of the
        # options; at the very least, so I could say which option is missing
        # without having to pass in the description.
        parser.error("Missing: %s" % description)

def parseCommandLine():
    """Parse the command line options and return (options, args)."""

    usage = """usage: %prog [options]
Testing optik's error handling.
"""
    parser = OptionParser(usage)
    parser.add_option("-f", "--file", type="string", dest="filename",
                      metavar="FILE", help="read data from FILE",
default=None)

    options, args = parser.parse_args()

    requiredVar(parser, options, options.filename, "filename")

def main():
    options, args = parseCommandLine()

if __name__=="__main__":
    main()




From alexhelder at R191.net  Thu Feb 21 14:47:20 2002
From: alexhelder at R191.net (Alex)
Date: Thu, 21 Feb 2002 19:47:20 GMT
Subject: newbie floating point question
Message-ID: <pan.2002.02.21.12.47.20.378591.4031@R191.net>

Why does this happen ?

>>> .2
0.20000000000000001

Is this the correct behavior for the Python 2.2 interpreter ? (version info
is printed below )

Python 2.2 (#1, Jan 29 2002, 01:43:44)
[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-98)] on linux2


Thanks,


-- 
Alex Helder

to email: change r191 --> att


From jason at jorendorff.com  Fri Feb  8 21:04:56 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 8 Feb 2002 20:04:56 -0600
Subject: FixedPoint
In-Reply-To: <Dhx88.95486$KM2.3465594@atlpnn01.usenetserver.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEFHDBAA.jason@jorendorff.com>

Mark McEahern wrote:
> I'm not sure what the best fix is here but one that works is to 
> modify the __cmp__ method for FixedPoint to anticipate comparison
> to None.  Here's the current __cmp__:
>
>     def __cmp__(self, other):
>         xn, yn, p = _norm(self, other)
>         return cmp(xn, yn)

How about:

    def __cmp__(self, other):
        try:
            xn, yn, p = _norm(self, other)
            return cmp(xn, yn)
        except:
            return cmp(float(self), other)

## Jason Orendorff    http://www.jorendorff.com/



From ivano.delpiccolo at degroof.be  Wed Feb 20 05:44:38 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Wed, 20 Feb 2002 11:44:38 +0100
Subject: RS232 Serial Port
Message-ID: <3c737e09$0$12226$4d4efb8e@news.be.uu.net>

How can I read/write on a RS232 port ?
I'm working under windows 98.

Thanks




From Frank.Gerhardt at web.de  Sun Feb 10 15:02:23 2002
From: Frank.Gerhardt at web.de (Frank Gerhardt)
Date: Sun, 10 Feb 2002 21:02:23 +0100
Subject: Which IDE most Smalltalk-like
Message-ID: <3C66D1CF.2080603@web.de>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Which Python IDE is most Smalltalk-like? I don't mean the language, of
course. What about the debugger, inspectors, "white-box" (everything
open for modification).

Frank.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (MingW32)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAjxm0c4ACgkQ9i+Nt+YA+UuuWQCdGw3kublhe70gD5YL2facwfj3
Rk4AoJ7TFn8Zsa+Eu6TEC1GqKt3WAHd/
=z1Xt
-----END PGP SIGNATURE-----



From lac at strakt.com  Thu Feb 28 07:19:30 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 28 Feb 2002 13:19:30 +0100
Subject: wanted - platform independent way to time database accesses
Message-ID: <200202281219.g1SCJU5i029191@ratthing-b246.strakt.com>

I've read thr thread that starts here:
http://mail.python.org/pipermail/python-list/2001-July/053317.html

But I didn't find one.  Is there one?
Laura Creighton



From phr-n2002a at nightsong.com  Sat Feb 16 17:15:48 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 16 Feb 2002 14:15:48 -0800
Subject: Guido wins FSF Award
Message-ID: <7xit8xxecr.fsf@ruckus.brouhaha.com>

Congratulations to Guido for winning the fourth Free Software Foundation
Award, for inventing and implementing Python as Free Software:
 
  http://www.fsf.org/press/2002-02-16-FSF-Award.html

And to think I learned of this from a Timothy Rue post.  Sigh.

Anyway, hats off to Guido!


From store_li at sina.com  Fri Feb 22 11:36:22 2002
From: store_li at sina.com (Ricky)
Date: Sat, 23 Feb 2002 00:36:22 +0800
Subject: Why ConfigParser can't write to file?
Message-ID: <a56sr2$52ge1$1@ID-12869.news.dfncis.de>

I intend to use ConfigParser module to manage my INI file. I can use get to
retrieve data from my ini file, and use set to change option value in
Python, just in python, after I closed python, I found that all changes I
made in python lost. There is nothing altered in my ini file!

How can I make my ini file to be changed?

Thanks for any reply!




From mertz at gnosis.cx  Tue Feb  5 13:31:00 2002
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Tue, 05 Feb 2002 13:31:00 -0500
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
 <20020130073810.382$1b@news.newsreader.com>
 <b34d3835.0201310005.548ee6be@posting.google.com>
 <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com>
 <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com>
 <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu>
 <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com>
 <a3h3cn$cs6$1@newshost.accu.uu.nl>
 <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>
Message-ID: <kTCY8kKkX8Lc092yn@bellatlantic.net>                         

|"Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
|> I suppose it might be a brain thing. I use my left hand for writing and
|> drawing while I use my right hand for more coarse-grained tasks, like
|> throwing a ball and such.

"Steve Holden" <sholden at holdenweb.com> wrote
|Strange. I don't come across many other people who do this. I've always
|thought my brain must have been differently wired for some reason I'll never
|know.

Same story for me.  I don't think it is that unusual though.  Culture is
sort of geared to the right-handed majority, so us folks who are mostly
left-handed wind up learning the things they can do "well enough" with
the right hand in the common style.  Once you've learned something,
propioceptic memory is more important than underlying "handedness."

--
 mertz@   _/_/_/_/_/_/_/ THIS MESSAGE WAS BROUGHT TO YOU BY:_/_/_/_/ v i
gnosis  _/_/                    Postmodern Enterprises         _/_/  s r
.cx    _/_/  MAKERS OF CHAOS....                              _/_/   i u
      _/_/_/_/_/ LOOK FOR IT IN A NEIGHBORHOOD NEAR YOU_/_/_/_/_/    g s




From logiplexsoftware at earthlink.net  Mon Feb  4 18:22:23 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 4 Feb 2002 15:22:23 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <a3n03l$tfn6@news1.gtech.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3d46d$tfn2@news1.gtech.com>
	<Wls68.14178$om6.357057@news1.tin.it>
	<mailman.1012590723.7301.python-list@python.org>
	<a3kb3p$se44@news1.gtech.com>
	<mailman.1012843209.19773.python-list@python.org>
	<a3n03l$tfn6@news1.gtech.com>
Message-ID: <20020204152223.4f4040e3.logiplexsoftware@earthlink.net>

On Tue, 5 Feb 2002 08:38:53 +1100
Peter Milliken wrote:

> 
> Hmmm.... Sounds like we have a small "disconnect" - this sounds like what
I
> would consider design to me - now I am in the "real" world of commercial
> programming, it is becoming more and more obvious that this just doesn't
> happen (design documentation I mean :-)). As a programmer for defense
> contractors, you did preliminary designs and then detailed designs (with
> attendant reviews) so the "coding problems" were all worked out at that
> stage i.e. by the times you had sorted out design to the point of pseudo
> code (or somewhere of that level) the coding became very straight forward
-
> and you knew it was all going to hang together! :-) (assumed that
> traceability of the requirements was done correctly and no boo-boos were
> made! :-)).  Perhaps I am wrong though? Anyway, it sounds like a good,
> comfortable working relationship that could only help you both generate
> better product.

I haven't had the (mis?)fortune to work on a large enough project where
there were a lot of preliminary steps and documentation, pseudo-code, etc,
generated prior to actual coding.  Of course there were discussions with
management and customers to outline what was wanted, but usually after that
it came down to discussing (with the other programmers) what
tools/libraries (and how they would fit in a general scheme) would be used
to develop the application.  Much discussion revolved around longevity,
simplicity, and generalization of developed code.  What we usually strived
for was to develop generalized libraries that could be used throughout the
application (and elsewhere). After that, application-specific code would be
used to glue the whole thing together.  The main goal was always to reduce
the amount of application-specific code needed for any particular program.

We developed this approach based upon statements made by Richard Stallman
(and common sense ;) on the GNU approach to development: 

1. no arbitrary limits
2. embed a language in your application and use that to develop parts of
the application
3. turn parts of the program into independent libraries  <-- repeat,
repeat, repeat

To this I would add that using an n-tier design will greatly extend the
lifetime of a piece of software (especially since much "upgrading" often
involves cosmetic changes or changes that are irrelevant to the function of
the software [like supporting a different SQL backend]) and makes things
like redundancy easier to implement.  Of course this is just an extension
of the Unix "tools" philosophy.

Python is an excellent language if you plan to follow these guidelines
(especially #2 which can otherwise be very difficult).

The difficult thing with this approach is that it does slow initial
development down a bit (or it appears to).  However, that lost time is
usually quickly recovered near the end of the development cycle (not to
mention debug times).  Also, given an extensible infrastructure, it's
tempting to start extending right away (hey, you have to test it, right? ;)
and not focus on the original spec.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From mdk at mdk.com  Tue Feb  5 16:33:30 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 16:33:30 -0500
Subject: What is the command to determine an object's type?
References: <mailman.1012942149.18474.python-list@python.org>
Message-ID: <a3pj3e$1900rd$1@ID-98166.news.dfncis.de>

"Mark McEahern" <marklists at mceahern.com> wrote in message
news:mailman.1012942149.18474.python-list at python.org...
> > MDK wrote:
> > How can I tell if an object is a string, a list, or a dictionary in a
> > program using if and ifelse so that I can perform a differnent action
> > depending on the object's type?
>
> I posted code that will do the trick and then realized, I should probably
> add this...
>
> It goes without saying, of course, that you may want to investigate the
> Strategy design pattern.  Eliminating that sort of conditional type
checking
> is Strategy's strong suit.
>
> Cheers,
>
> // mark
>
>

The code that you posted made perfect sense; thanks for that.  However, I am
not clear about what you are talking about in regards to "...the Strategy
design pattern."

I would like to investigate it but I do not know what it is I am
investigating.  Can you clarify a little more for me?

Thanks.




From amuys at shortech.com.au  Thu Feb 21 19:55:59 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 21 Feb 2002 16:55:59 -0800
Subject: calling a function indirectly
References: <mailman.1014185673.27565.python-list@python.org>
Message-ID: <7934d084.0202211655.4c9d5058@posting.google.com>

Michal Wallace <sabren at manifestation.com> wrote in message news:<mailman.1014185673.27565.python-list at python.org>...
> On Tue, 19 Feb 2002, Rajarshi Guha wrote:
> 
> > Hi,
> >   is it possible to have a variable contain a function name, and use the 
> > variable to call the function? 
> > An example:
> > 
> > def func:
> >    some code
> > 
> > funcvar = 'func'
> > 
> > Now use funcvar (somehow!) to call func
> 
> 
> I also think the easiest thing to use is funcvar = func,
> but if for some reason it has to be a string, you can
> use exec, or:
> 
> locals()[funcvar]("whatever")
> 

Note that this only works if func is in the local scope, however if it
is in an enclosing scope there is no analogous function to use in
place of locals().

Quoting from PEP227 (Statically Nested Scopes)

locals() / vars()
    These functions return a dictionary containing the current scope's
    local variables.  Modifications to the dictionary do not affect
    the values of variables.  Under the current rules, the use of
    locals() and globals() allows the program to gain access to all
    the namespaces in which names are resolved.

    An analogous function will not be provided for nested scopes.
    Under this proposal, it will not be possible to gain
    dictionary-style access to all visible scopes.

== end quote ==

So you will need to revert to the eval() hack :(.

I am curious to know why an analogous function wasn't provided?  Or
more presisely why the rationale behind the current behaviour wasn't
captured in the PEP?

OTOH, you are almost certainly better off using the function directly,
if necessary creating your own dictionary to store them ie...

func = func1
func()

or

ops = { 'func1':func1, 'func2':func2, 'func3':func3 }
func = 'func1'
ops[func]()

etc.

Andrae Muys


From phr-n2002a at nightsong.com  Wed Feb 20 20:36:53 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 17:36:53 -0800
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <258fd9b8.0202201701.5f8e4e42@posting.google.com>
Message-ID: <7xpu2zsjii.fsf@ruckus.brouhaha.com>

magnus at thinkware.se (Magnus Lyck?) writes:
> Rajarshi Guha <rxg218 at psu.edu> wrote in message news:<a50umn$1dou at r02n01.cac.psu.edu>...
> > Why can't I use d.keys().sort() directly? 
> 
> Because .sort() is not a function that returns a sorted
> copy of it's input. It sorts the list in-place and returns
> none. A sort method returning a copy would use much more
> memory.

But there's no reason sort couldn't sort in place and return the list.
d.keys() already makes a temporary list with the keys in it after all.
The temporary list can be sorted in place without disturbing the
dictionary.


From logstx at bellatlantic.net  Fri Feb 22 20:33:26 2002
From: logstx at bellatlantic.net (logistix)
Date: Sat, 23 Feb 2002 01:33:26 GMT
Subject: fun with ADO
References: <2ec1bc1c.0202221202.1e0ea6aa@posting.google.com>
Message-ID: <GjCd8.2426$Hg1.1707@nwrddc02.gnilink.net>

Stupid question, but did you run MakePy against the ADO objects?  Seems like
the time saved by dropping the Value reference could have to do with
Dispatch name resolution.

Just curious.

--
-

"Larry" <wrbt at email.com> wrote in message
news:2ec1bc1c.0202221202.1e0ea6aa at posting.google.com...
> I thought I'd share some stuff I found while trying to tune an app
> that interacts heavily with a sql server database via ADO. The web
> application I've been rewriting in python was having serious problems
> with performance. I knew it wasn't IIS because things were running
> really slow from the interactive window too.
>
> After fooling around with profile for a while the pig was unveiled as
> actually getting the field values from the Sql Server database via
> win32com/ADO.
>
> a toy version of the technique I was using was:
>
>     rs=win32com.client.Dispatch(r'ADODB.Recordset')
>     rs.Open(sql,connectString)
>     while not rs.EOF:
>         name=str(rs.Fields("clientname").Value)  #oink!
>         gid=str(rs.Fields("gid").Value)
>         (- do something fun with fields -)
>         rs.MoveNext()
>     rs.Close()
>
> A put some timers on a test script that ran through a few hundred
> records, and it ran in about 1.9 seconds. That sucks. I also tried
> using mapping field values into a list and extracting from that, no
> better.
>
> So I tried removing the Value, as in: name=str(rs("clientname"))
> The average new run time was 0.7 seconds! Woah. Nice.
>
> Next I tried using the sparsely documented Collect method of the
> recordset object, as in: name=str(rs.Collect("clientname"))
> The average run time was now down to 0.24 seconds. Viva Collect!
>
> Finally I ran across an article where someone mentioned this
> technique:
>
>     rs.Open(sql,connectString)
>     f1=rs("gid")
>     f2=rs("clientname")
>     while not rs.EOF:
>         name=str(f2)
>         gid=str(f1)
>         (- do something fun with fields -)
>         rs.MoveNext()
>     rs.Close()
>
> We have a winner! Average time was 0.18 seconds for this method.
>
> So for any "ADOites" (tm) out there in python land I'd recommend
> either Collect or maintaining field references outside of recordset
> iterations. I've increased performance in many parts of my application
> by 10x or more.




From jimmy at retzlaff.com  Sat Feb 16 14:12:09 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Sat, 16 Feb 2002 11:12:09 -0800
Subject: win32com and MS Excel
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F55A2@MAYOR.retzlaff.com>

>from win32com.client import Dispatch
>ll = [['aaa', 333, 'ccc']] * 1000
>xlApp = Dispatch("Excel.Application")
>xlApp.Visible = 1
>xlApp.Workbooks.Add()
>t = xlApp.Workbooks(1).Sheets(1)
>t.Range(t.Cells(1,1), t.Cells(len(ll),3)).Value = ll

>but with 50000 elements you'll have to wait some minutes for
>Excel to show the data. And with 100000, I get a fatal error. 

If you haven't already, you can get quite a speed boost by running
Makepy on the Excel type library. In PythonWin, select "COM Makepy
utility" on the Tools menu and look for Microsoft Excel. The example
above with 50,000 takes about 1 second on my 850MHz notebook. The Python
file that makepy generates can also be quite valuable in figuring out
Python/Excel nuances. You don't need to do anything in your code with
that file, its mere existence speeds things up considerably (especially
if you make a large number of calls into Excel).

The reason for the error with 100,000 is that Excel only supports 256
columns and 65536 rows per spreadsheet.

Jimmy



From janfbr at sex.ifi.uio.no  Sat Feb 16 07:56:35 2002
From: janfbr at sex.ifi.uio.no (Jan-Fredrik Braseth)
Date: 16 Feb 2002 13:56:35 +0100
Subject: Problems with telnetlib
Message-ID: <xyuvgcxlh4s.fsf@sex.ifi.uio.no>

Hi, there.

I'm having a little problem with a bot I'm making. It uses
telnetlib to connect to a chatserver (not irc) and works
fine for a period, but after some time I get this error message:

Traceback (most recent call last):
  File "./quizbot.py", line 366, in ?
    foo = QuizBot(name,password,channel)
  File "./quizbot.py", line 28, in __init__
    line = self.tn.read_until("\n").split()
  File "/usr/lib/python2.2/telnetlib.py", line 297, in read_until
    self.process_rawq()
  File "/usr/lib/python2.2/telnetlib.py", line 424, in process_rawq
    self.msg('IAC %d not recognized' % ord(opt))
UnboundLocalError: local variable 'opt' referenced before assignment

The main loop of the bot is just a while(1) which reads a line at the
time from the connection: line =self.tn.read_until("\n").split()
and it looks like this fails after some time.

tn is initiated like this:

tn = telnetlib.Telnet(server,port)

I really hope someone can help me with this problem

-- 
Jan-Fredrik Braseth		www.ifi.uio.no/~janfbr/
2B|!2B


From mcherm at destiny.com  Fri Feb  8 09:35:18 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Fri, 08 Feb 2002 09:35:18 -0500
Subject: How did TimBL's speech at Python 10 go yesterday?
Message-ID: <3C63E226.2040808@destiny.com>

>> I agree, and apparently so does TimBL. He was careful to point out that 
>> you don't actually go out to the web each time to get module, rather 
>> that the actual code is "cached" locally and there needs to be something 
> 
> Note that there's a distinction between using a URI as a unique
> identifier and actually downloading something from it.  In the XML
> Namespaces spec, URIs are used in things like <element
> xmlns:rng="http://xml.mems-exchange.org/schema/">.  That URI can't be
> dereferenced (the hostname doesn't even exist) but it's never
> retrieved, so that doesn't matter.  I think TBL's suggestion was more
> along the lines of having unique identifiers for modules and versions,
> and not primarily importing via URL.
> 
> --amk                                                  (www.amk.ca)
> 


Well, sort of. Except that he ALSO pointed out that you probably DID 
want to have something (like the module) available at the URL. In fact, 
he surprised me by suggesting that it was probably good practice (NOT 
mandatory) to put a DTD or schema or something at the URL used to define 
XML namespaces -- surprising because I heard it from TimBL, not for the 
idea itself, which was always downright obvious.

He certainly talked about downloading the module to local storage and 
keeping it there as long as the user wants... clearly this implies that 
there's something out there to download.

-- Michael Chermside

PS: I intentionally used URL in my comments, because if the URI is a URN 

instead of a URL then it can't possibly be "downloaded" as there's 
noplace to connect to. Just trying to keep the terminology straight.




From jlh at cox.net  Mon Feb 25 19:44:36 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Tue, 26 Feb 2002 00:44:36 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu> <Vg%c8.4179$XT1.116344@news1.east.cox.net> <a5cnvo$2mdu$1@news.idiom.com>
Message-ID: <UTAe8.9921$jC3.362611@news1.east.cox.net>

"Jim Dennis" <jimd at vega.starshine.org> wrote in message
news:a5cnvo$2mdu$1 at news.idiom.com...
> In article <Vg%c8.4179$XT1.116344 at news1.east.cox.net>, Jeff Hinrichs
wrote:
>
> >see:
> >http://mail.python.org/pipermail/python-list/2001-August/060976.html
> >for some info.   Unless you have total control over what is being eval'd
you
> >are at risk.
> >-Jeff
>
> >"Rajarshi Guha" <rxg218 at psu.edu> wrote in message
> >news:a50udo$1dou at r02n01.cac.psu.edu...
> >> On Wednesday 20 February 2002 01:24 in comp.lang.python Jeff Hinrichs
> >wrote:
>
> >>> If you wanted to get away from the dangerous eval, you could put your
> >>> functions inside of a class and then,
>
> >> Why is eval dangerous?
>
>  In THIS case it's not a problem.  The string being eval'd is a
>  reference to be bit of code he wrote himself.  However, if it was
>  being combined with a string from any untrusted source (esp.
>  user input, or data from a network connection or subprocess, even
>  filenames from an os.listdir() or some such, IF that was the case
>  then eval could be quite dangerous because it could be executing
>  arbitrary Python code (which could, in turn, execute arbitrary bits
>  of system code).
>
>  Regulars on this newsgroup display a knee jerk reaction to eval().
>  This could be construed as a healthy wariness in some cases.  One
>  should always ask:
> where did this data come from?
> how did I validate it?
>    and especially,
> what are the risks of using/trusting this data in this way?
>
>  ... but those questions should not be limited merely to strings
>  that we might be passing to eval, they must be constantly applied
>  throughout our coding if we intend to write robust code which
>  works across security contexts (handles any sort of foreign data)
>  and which is capable of enforcing the most basic implicit security
>  policy (don't get subverted to executing arbitrary code or tricked
>  into removing, or corrupting our data).
This is true, but you should always consider that modifications/enhancements
in the future might expose those
"shielded" yet vulnerable sections of code.  With that in mind, you should
always code "on the defensive."
Danger lurks not only where we expect it but where we don't.  So having a
"knee jerk" reaction to eval is proper.  It has a number of uses and can be
quite handy in one shot scripts but any code with a lifetime of more than
one run...."Danger! Will Robinson.  Danger!"

-Jeff





From mpoulin at verinet.com  Mon Feb  4 14:40:16 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Mon, 04 Feb 2002 14:40:16 -0500
Subject: Calling a function or method by name
References: <mailman.1011796114.1594.python-list@python.org> <a2mjrd$2ok6$1@norfair.nerim.net> <mailman.1011824379.21237.python-list@python.org>
Message-ID: <3c5efea3$0$2197$7586b60c@news.frii.com>

In article <mailman.1011824379.21237.python-list at python.org>, "Jonathan
Gardner" <jgardn at alumni.washington.edu> wrote:

>>
> I remembered the real reason why. I want it to be modifiable with user
> input. This means eventually I'll have to convert a string to a
> function.
> 

This brings back some really bad memories. I once worked on a system that
consisted of a core run-time module that was customized through the use
of configuration tables stored in a relational database.

At start-up, the application would configure its internal function
dispatch tables by reading the *names* of the functions from the DB.

In my experience, this approach was quite brittle, since it was very easy
to misspell the name in the DB. Or sometimes a programmer would change the 
function name in the code without updating the corresponding entry in the DB.

My only advice is to think long and hard before you go down this path.


From jason at jorendorff.com  Fri Feb  1 23:08:23 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 22:08:23 -0600
Subject: question about subclassing dict
In-Reply-To: <a3fbse$gc4$1@nntp6.u.washington.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLCENOCPAA.jason@jorendorff.com>

Russell E. Owen wrote:
> I started by defining __iter__. I was then pleased to find that the 
> methods iterkeys, itervalues and iteritems all worked just fine.

They don't seem to return the objects in the right order (which
I gather is the whole point):

>>> x = OrderedDict.OrderedDict()
>>> x['a'] = 'A'
>>> x['z'] = 'Z'
>>> x[3] = 'three'
>>> x[17.5] = 'seventeen and a half'
>>> x[None] = 'nothing much'
>>> list(iter(x))
['a', 'z', 3, 17.5, None]
>>> list(x.iterkeys())
['a', 3, None, 'z', 17.5]
>>> 

Your test code in OrderedDict.py itself doesn't seem to
test the iterfoo() methods.


> Clearly dict is basing those three iterators on the
> iterator returned by __iter__.

I think it's just using its builtin definition of iterkeys().
Anyway, it's easy enough to define the extra methods:

class OrderedDict(dict):
    ...

    def iterkeys(self):
        return iter(self.__keyList)

    def itervalues(self):
        for k in self.__keyList:
            yield self[k]

    def iteritems(self):
        for k in self.__keyList:
            yield (k, self[k])

## Jason Orendorff    http://www.jorendorff.com/



From threeseas at earthlink.net  Wed Feb  6 23:28:42 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:42 GMT
Subject: Autocoding project proposal.
References: <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net> <slrna62c6f.afn.philh@comuno.freeserve.co.uk>
Message-ID: <1427.802T2899T14116077threeseas@earthlink.net>

On 06-Feb-02 08:34:07 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Wed, 06 Feb 2002 04:54:05 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>>On 03-Feb-02 19:47:37 Peter Seebach <seebs at plethora.net> wrote:
>>>In article <1173.799T632T11653818threeseas at earthlink.net>,
>>>Timothy Rue <threeseas at earthlink.net> wrote:
>>>>Don't like the idea of autocoding huh?
>>
>>>Haven't been able to perceive whether or not you have an idea.  I have to
>>>say, you're talking to the wrong crowd.  You need people less mired in
>>>conventional ideas of software engineering.  Your best bet is almost
>>>certainly the C# people; they are certainly not mired in conventional ideas
>>>of software engineering.
>>
>>Why would  be interested in a programming language by a company that has
>>been found guilty of Anti-trust? Not to mention that language is not
>>available anywhere near as much as something like Python.

>There is an open source implementation of C# as part of gcc, so I expect it
>runs on pretty much everything.

Well, you are wrong. <shrug>

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From starx at pacbell.net  Sat Feb 16 01:00:23 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 06:00:23 GMT
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk>
Message-ID: <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4>

Christopher Saunter || Wed 13 Feb 2002 06:26:53a:

> Greetings All!
> 
> I have been using Python now for arround a year, and have found it
> incredably easy to learn the language, primarily because things are
> very obvious cf other languages, and until the other day, I had never
> had a 'Gotcha' - a bug caused by me entering some code wrongly and not
> realising it.
> 
> It all stems to this, I was assigning to numbers to variables:
> 
> var1 = 1000
> var2 = 0100
> etc.

Just use spaces to balance, eg:
var1 = 1000
var2 =  200
var3 =   45

I do this all the time, in more complicated places...

glVertex3d( -(x_size + 1), -0.5,  0   )
glVertex3d( -(x_size + 1),  0.5,  0   )
glVertex3d( -(x_size + 1),  0,   -0.5 )
glVertex3d( -(x_size + 1),  0,    0.5 )

Balancing negative signs and variable sizes =)

> [snip]
> 
> Personally, I would like to see octal number entry migrated from 0---
> to something like 0o---.  This would bring things more in line with
> the entry of hex numbers (0x---)
> 
> I know it's only a minor thing, but it'd be nice if my numbers behaved
> as I expected...

Steve has mentioned that UNIX people are used to it... i think it's a 
little deeper than that, b'cause as far as i know, _every_ programming 
language that has ever supported octal does it this way...

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From eppstein at ics.uci.edu  Wed Feb  6 20:39:31 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 06 Feb 2002 17:39:31 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu>
Message-ID: <eppstein-80FFB2.17393106022002@news.service.uci.edu>

Here it is in a functional programming style:

import operator

def flatten(x):
        if operator.isSequenceType(x) and (len(x) != 1 or x[0] != x):
                return reduce(operator.concat, map(flatten, x), [])
        else:
                return [x]
                
print flatten(["cat",5,['dog',[3,3,1]],"zoo"])

...outputs...
['c', 'a', 't', 5, 'd', 'o', 'g', 3, 3, 1, 'z', 'o', 'o']

All that list copying makes it somewhat inefficient, though...
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From sjmachin at lexicon.net  Fri Feb 15 16:34:39 2002
From: sjmachin at lexicon.net (John Machin)
Date: 15 Feb 2002 13:34:39 -0800
Subject: COFF2OMF C++ Builder 4
References: <2830c89c.0202150629.5a6b711f@posting.google.com>
Message-ID: <c76ff6fc.0202151334.23294940@posting.google.com>

andrew.gregory at npl.co.uk (Andrew Gregory) wrote in message news:<2830c89c.0202150629.5a6b711f at posting.google.com>...
> Trying to create BCB4 DLLs which can be imported into Python 2.2
> COFF2OMF python22.lib new.lib gave me error message "Invalid COFF file".
> What is the easiest way to fix this?
> 
> Thanks, Andrew.

1. Check that your python22.lib is not corrupted.

2. Try using an OMF lib built using the COFF2OMF from the Borland 5.5
free-from-the-net kit -- I've e-mailed you one such.

3. Upgrade to the 5.5 freebie -- supported by distutils.

4. If (2) works and (3) is not feasible for you for some reason, you
could possibly download 5.5, grab the COFF2OMF.EXE, and jettison the
rest ...


From jeff at ccvcorp.com  Fri Feb  1 16:54:51 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 01 Feb 2002 13:54:51 -0800
Subject: Equivalent of Perl chomp?
References: <mailman.1012513005.17840.python-list@python.org> <%ok68.88316$8b.3281206@atlpnn01.usenetserver.com> <a3etov$hgu$0@216.39.172.122>
Message-ID: <3C5B0EAA.F4527F5E@ccvcorp.com>


Bengt Richter wrote:


> I think slicing is probably a winner because endswith subsumes more
> complex slicing options that have to be tested for, and the argument can be
> multicharacter. I would be surprised if single-character string handling
> wasn't special-cased a lot for speed, benefiting s[-1:] == '\n'.

Well, one of the things that Tim Peters hinted at is more likely to be the main reason
that endswith is slower...  every iteration, the endswith version has to resolve a name.
If you changed the loop to something more like:

func = s.endswith
for n in xrange(iterations):
    func('\n')

you would likely see a notable difference in the results.   :)

(But I'm too lazy/otherwise busy to actually run the tests myself....)

Jeff Shannon
Technician/Programmer
Credit International




From fredrik at pythonware.com  Sat Feb  9 19:37:11 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 10 Feb 2002 00:37:11 GMT
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com> <Xns91B074EB2FA5ahmebahhotmailcom@198.99.146.10>
Message-ID: <Xgj98.24854$l93.4542796@newsb.telia.net>

"ameoba" <ahmebah at hotmail.com> wrote:
> Actually, there's a handful of win32 calls that would handle this
> relatively simply; less than a page worth of code (well, until you
> range/error checking and a Python interface).

http://effbot.org/efflib/console/index.htm

about a 1000 lines, but that includes quite a few bells and
whistles.

the api is relatively clean; writing a curses implementation
should be trivial (did most of that once, but lost the code).

</F>




From foser at _SpamNo_sesmar.eu.org  Mon Feb 25 12:36:48 2002
From: foser at _SpamNo_sesmar.eu.org (foser)
Date: Mon, 25 Feb 2002 18:36:48 +0100
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com> <a5dqoh$1e99$1@norfair.nerim.net>
Message-ID: <3C7A7630.2060101@_SpamNo_sesmar.eu.org>

For linux (Gnome) there's Anjuta. http://anjuta.sourceforge.net .

- fos




From no at spam.com  Tue Feb 26 12:58:02 2002
From: no at spam.com (Tokomar)
Date: Tue, 26 Feb 2002 17:58:02 GMT
Subject: Tkinter list boxes, selection emphasis, and the keyboard focus
Message-ID: <K0Qe8.48006$BA5.2207030344@newssvr14.news.prodigy.com>

Python fans:

First, some definitions...

Focus - the single place on entire screen where keyboard activity will 
reflect. This displays either a text caret or an ant rectangle.

Selection - in a list box, which item is the user's favorite. This displays 
by changing the color scheme to an opposing contrast.

Now the problem. Tkinter appears to slur these definitions together. If, 
for example, I put two list boxes next to each other, they treat the focus 
and selection the same. If the keyboard focus is in the second one, we 
can't see which item in the first one is selection.

I will deal, short term, by changing some other property in the lists. (I 
will therefor do Tkinter's housekeeping for it.)

But has anyone addressed this problem directly? Is there a configuration 
option I overlooked?

-- 
Tokomar


From member at dbforums.com  Mon Feb 18 07:17:32 2002
From: member at dbforums.com (krallabandi)
Date: 18 Feb 2002 06:17:32 -0600
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net>
Message-ID: <3c70f0dc$1_1@spamkiller.newsgroups.com>

y



--
Posted via dBforums
http://dbforums.com


From peter at engcorp.com  Tue Feb 12 07:29:30 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 12 Feb 2002 07:29:30 -0500
Subject: Shell command
References: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net>
Message-ID: <3C690AAA.37D9A6C6@engcorp.com>

DelPiccolo Ivano wrote:
> 
> How can I do to launch a command in my python script ?
> example : I want to launch the command : rm test.txt

import os
os.system('rm test.txt')


From c.duncan at nlada.org  Wed Feb 13 08:48:26 2002
From: c.duncan at nlada.org (Casey Duncan)
Date: Wed, 13 Feb 2002 08:48:26 -0500
Subject: Request for IPC10 and Zope 3 Sprint Pictures
Message-ID: <E16aziX-0006q3-00@mail.python.org>

Oh, ye mailing list gods, forgive the cross-post...

I am making a photo album that will be featured in the next issue of the Zope 
news. I took several snap-shots at the Python conference and the Zope 3
Sprints, but I'm sure I missed a lot. If anyone cares to contribute some 
snapshots they took to the photo album, please send them to me!

Just email the jpegs (any size, I will crop) to me and if possible a caption 
or a list of the people in the picture (I don't know everybody 8^). And 
please make sure the pictures are suitable for "family viewing" 8^). I will 
of course give each photographer credit in the album.

Thanks a bunch!

P.S. The Zope news will be coming out pretty shortly, but I will continue to 
add pictures to the album as I get them.

/---------------------------------------------------\
  Casey Duncan, Sr. Web Developer
  National Legal Aid and Defender Association
  c.duncan at nlada.org
\---------------------------------------------------/



From andreas at mtg.co.at  Sat Feb 16 04:01:45 2002
From: andreas at mtg.co.at (Andreas Kostyrka)
Date: Sat, 16 Feb 2002 10:01:45 +0100
Subject: string compare question.Please Help!!!
In-Reply-To: <020301c18d01$c2b04460$0effa8c0@dbaseserver>
References: <001f01c18ceb$fdb53900$4301010a@sky>
	<020301c18d01$c2b04460$0effa8c0@dbaseserver>
Message-ID: <20020216100145.3be067c0.andreas@mtg.co.at>

On Tue, 25 Dec 2001 15:05:21 +1000
"Evgeny Jonson" <john at ablogic.ru> wrote:

> def ip_cmp(ip1, ip2):
>     """Compare IP addreses as strings
>     """
>     l_ip1 = ip1.split('.')
>     l_ip2 = ip2.split('.')
>     if len(l_ip1) == 4 and len(l_ip2) == 4:
>         s_ip1 = string.zfill(l_ip1[0], 3) + '.' + \
>                 string.zfill(l_ip1[1], 3) + '.' + \
>                 string.zfill(l_ip1[2], 3) + '.' + \
>                 string.zfill(l_ip1[3], 3)
>     
>         s_ip2 = string.zfill(l_ip2[0], 3) + '.' + \
>                 string.zfill(l_ip2[1], 3) + '.' + \
>                 string.zfill(l_ip2[2], 3) + '.' + \
>                 string.zfill(l_ip2[3], 3)
>         resault = cmp(s_ip1, s_ip2)
>     else:
>         print 'Wrong IP addres format!'
>         sys.exit(0)
Well, this is obviously wrong. Something like "raise ValueError,"bad ip address "+str(ip1)+" or "+str(ip2) would
appropiate. sys.exit just kills your python process.

But one can do this even easier:
def ip_cmp(a,b):
    a=[int(x) for x in a.split(".",3)]
    b=[int(x) for x in b.split(".",3)]
    return cmp(a,b)

Only caveat, that my ip_cmp doesn't check for adresses are shorter than 4 bytes. All other error cases are
caught directly by the expressions.
>>> ip_cmp("10.66.73.78","10.0")
1
>>> ip_cmp("a.66.73.78","10.0.0.0")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "t.py", line 2, in ip_cmp
    a=[int(x) for x in a.split(".",3)]
ValueError: invalid literal for int(): a>>> ip_cmp("9.66.73.78","10.0.0.0.244")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "t.py", line 3, in ip_cmp
    b=[int(x) for x in b.split(".",3)]
ValueError: invalid literal for int(): 0.244

Andreas






From info at forpctechs.com  Sat Feb  2 04:15:43 2002
From: info at forpctechs.com (info at forpctechs.com)
Date: Sat, 02 Feb 2002 09:15:43 GMT
Subject: PC Repair & Networking
Message-ID: <G988F82N.8JA282XY@forpctechs.com>

---*** Networking Tool Kits at Wholesale ***---

GREAT PRICES ON ELECTRONIC TOOL KITS, LARGE VARIETY OF KITS WITH 
PICTURES AND PRICES - Over 700 Items in Inventory

http://www.forpctechs.com - #1 in Customer Service!


** LAN CAT 5 Termination Kit
 
 Includes Punch Down Tool, LAN Tester, Ratched Crimper, UTP Stripper and Assorted Connectors and Boots
 

** Electronic Pro's Tool Kit
 
 13 Assorted Screwdrivers, 6 Assorted Precision Screwdrivers, Brush, Mini-Drill Set, Utility Knife, IC Extractor, Vice, Electrical Tape, Flashlight, Anti-Static Strap, Soldering Iron & Stand, Asoorted Pliers, Solder, 6" Stainless Steel Scissors, Folding Hex Wrench, 8" Lineman's Pliers, 6" Tweezers, Mini-Vacuum Cleaner, Desoldering Pump, 6" Adjustable Wrench, All-In-One Crimper, Voltage Test Screwdriver, Components Box, 5 pc. File Set, Removable Tool Pallets, 10 pc. Mini-Wrench Set, Heavy Duty Carrying Case and more.
 
 
** Network Maintenance Tool kit -  
 
 IC Extractor, Folding Hex Wrench, 6" Adjustable Wrench, 6" Forceps, Inspection Mirror, Mini-Flashlight, Desoldering Pump, 5" Diagonal Cutter, 5" Snipe Nose Pliers, Wire Stripper (AWG 30-20), Modular Plug Crimper, Rotary Coax Stripper, Coax Crimper, Assorted Screwdrivers, Soldering Iron, Zippered Carrying Case 
  

** Master Network Installation Tool Kit - 
 
 "Super Crimp" Frame, RJ45 Die Set, RJ11 Die Set, RG58 RG59/62 BNC/TNC Die Set, RG58 RG59/62 AU Die Set, Punch Down Tool with 110 & 66 Blades, Rotary Coax Stripper, Cable Cutter, Precision Wire Stripper, microcutters, Long Nosed Pliers, Inspection Mirror, Mini-Flashlight, Curved Forceps, 7 pc. Folding Hex Wrench Set, 30w Ceramic Solder Iron, Desoldering Pump, "Super Torque" Racheted Screwdriver Handle with Assorted Nutdriver and Screwdriver Blades, Professional Digitial Multimeter, 2 Removable Tool Pallets, ABS Tool Case, Multi-Network LAN Cable Tester, Desoldering Braid, Blowing Brush, Anti-Static Wrist Strap, Network Connector Kit with BNC, RJ11, RJ22, RJ45, AMP RJ45 Plugs & Boots 


GREAT PRICES ON ELECTRONIC TOOL KITS, LARGE VARIETY OF KITS WITH 
PICTURES AND PRICES
http://www.forpctechs.com - #1 in Customer Service!



From sholden at holdenweb.com  Fri Feb 22 07:53:34 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 07:53:34 -0500
Subject: execv giving a Fatal Error (11)
References: <a541uu$l0e@r02n01.cac.psu.edu>
Message-ID: <Nard8.78153$n73.1508731@atlpnn01.usenetserver.com>

"Rajarshi Guha" <rxg218 at psu.edu> wrote in message
news:a541uu$l0e at r02n01.cac.psu.edu...
> Hi,
>   I have a small python script which uses PyGTK. In one of the call backs
> I do:
>
> os.execv('/usr/bin/emacs',[])
>
> But as a result I get:
>
> Fatal error (11).Segmentation fault
>
> This occurs even when I run a script that contains the os.exev(...) as the
> only command.
>
> Does anybody know why this happens? Is there any other way I can implement
> this?
>
Should that be

    os.execv('/usr/bin/emacs', ['emacs'])

perhaps. Long time since I used exec, but I seem to remember it needs a
first argument in args to stand for the program name. Indeed, in cygwin
under 2.2 I get:

 >>> os.execv("/usr/bin/ls", [])
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: execv() arg 2 must not be empty

This is certainly more helpful than a segfault!

>>> os.execv("/usr/bin/ls", ['ls'])
BookProject  lib     mydatabase.fs       nonweb   sketch2.jpg  tm1.py
PyNet        mounts  mydatabase.fs.lock  rhubarb  test.cpp     tm1.pyc
bin          msqd    mydatabase.fs.tmp   share    tf           wiki

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From sholden at holdenweb.com  Wed Feb 20 08:41:31 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 08:41:31 -0500
Subject: Help with code
References: <a4votp$mmf$1@infocom.km.ua>
Message-ID: <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com>

"Dmitry Savchuk" <compedu at nh.km.ua> wrote in message
news:a4votp$mmf$1 at infocom.km.ua...
> Hi,
>
> I'd like to ask you a question. There is a code from the book of Ivan Van
> Laningham "Sams Teach Yourself Python in 24 Hours (Teach Yourself -- 24
> Hours)":
>
> import time
> class now:
>     def _init_(self):
>         self.t=time.time()
>         self.storetime()
>     def storetime(self):
>         self.year,\
>         self.month,\
>         self.day,\
>         self.hour,\
>         self.minute,\
>         self,second,\
>         self.dow,\
>         self.doy,\
>         self,dst=time.localtime(self.t)
>     def _str_(self):
>         return time.ctime(self.t)
>
> n=now()
> print "The year is ", n.year
> print n
> s=`n`
> print s
>
>
> It doesn't work properly. Appear a messge:
>
> The year is
> Traceback (most recent call last):
>   File "C:\Projects\test", line 20, in ?
>     print "The year is ", n.year
> AttributeError: now instance has no attribute 'year'
>
> Where is my mistake?
>
Besides the underscores another poster mentioned, note also that you have
replaced a period with a comma. Your line which reads

            self,dst=time.localtime(self.t)

should read

            self.dst=time.localtime(self.t)


--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jeff at ccvcorp.com  Tue Feb 12 14:05:43 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 12 Feb 2002 11:05:43 -0800
Subject: Global Variables in Modules
References: <20020211111025.63f4c956.jsc-news@bluesine.com>
Message-ID: <3C696787.8A70736C@ccvcorp.com>


Jeff Clement wrote:

> What I'm actually doing is running a code block in an exec block and declaring a bunch of globals for that exec.  The code block imports some modules and I need those to be able to see the global variables.  I can make this work by adding them to __builtins__ but that doesn't seem right somehow :)

Chances are that if you're using exec, you're not doing things in the best way possible.  In the year that I've been using Python, I don't think I've seen *any* situation that's best done with an exec.  (Admittedly, I haven't been doing anything at all esoteric, but still....)  I'd recommend seriously
rethinking your architecture, honestly--exec/eval() are nice to have around, just in case, but using them is a big red flag that something ugly is going on.  (It's good to have a fire extinguisher, too, but I wouldn't use one to put out a candle....)

However, as far as the global variables go... if you really must do this, then the best solution is to have a separate module (called globalvars or constants or whatever) that all your other modules import;  they can then read (and modify) variables in that module as needed.  (Though beware that while
global constants are useful, depending on global variables *is* a major source of bugs... do it if you need to, but be careful.  :) )

Jeff Shannon
Technician/Programmer
Credit International




From phr-n2002a at nightsong.com  Thu Feb 21 09:50:56 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 06:50:56 -0800
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org>
Message-ID: <7xbseiga7j.fsf@ruckus.brouhaha.com>

William Park <opengeometry at yahoo.ca> writes:
> On Wed, Feb 20, 2002 at 03:51:24PM -0800, Kevin Dahlhausen wrote:
> > Does anyone know of an inexpensive web host that allows compilation
> > and use of Python extensions from  your own directory?
> 
> With the money you pay to ISP, why don't you get DSL with static IP, and
> run your own server?

Usually you don't get nearly as fast a connection through ADSL as you
get at a well-connected ISP, and also ADSL is pretty unreliable.
Remember also that ADSL is usually slower in the upload direction,
which is what you care about if you're running a server.



From tdelaney at avaya.com  Sun Feb  3 18:01:15 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 4 Feb 2002 10:01:15 +1100 
Subject: Making a list-inhereted object return standard repr?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A15B@natasha.auslabs.avaya.com>

> From: Philip Swartzleonard [mailto:starx at pacbell.net]
> 
> Jason Orendorff || Fri 01 Feb 2002 08:32:57p:
> 
> > Philip Swartzleonard wrote:
> >> Ok, so i decided to make my container class inherit from 
> list, [...]
> >> I want to change __repr__ and __str__ to do something more like the
> >> first case again [...] 
> > 
> > class Foo(list):
> >     def __repr__(self):
> >         basic = object.__repr__(self)
> >         values = list.__repr__(self)
> >         return basic[:-1] + ' containing ' + values + '>'
> > 
> > ## Jason Orendorff    http://www.jorendorff.com/
> 
> Thanks! That's copy-and-paste'ly what i need. I knew it was 
> something more 
> or less obvious i was missing %D.
> 
> (although i may have to leave out the contains, as it gives 
> me things like     
> <__main__.Cell object at 0x015F0210 containing 
> [<__main__.Player object at  
> 0x015E7870 containing []>]> ... so far every object is a container :)

This brings up the interesting question of whether there should be a an
object method which returns the "canonical repr" of an object i.e. the
<module.Class object at address> or possibly even a builtin method (so it
would work with classic classes too).

I know I've wanted to use this in classic classes many times for debugging
purposes - I want to be easily able to distinguish instances from each
other. Each time I end up having to code it myself.

Tim Delaney



From quinn at upchuck.ugcs.caltech.edu  Tue Feb 12 00:57:49 2002
From: quinn at upchuck.ugcs.caltech.edu (Quinn Dunkan)
Date: 12 Feb 2002 05:57:49 GMT
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <slrna664d7.d87.philh@comuno.freeserve.co.uk> <Xns91AFC76818BBRASXnewsDFE1@130.133.1.4>
Message-ID: <slrna6hbmt.qkr.quinn@upchuck.ugcs.caltech.edu>

On 9 Feb 2002 03:36:11 GMT, Philip Swartzleonard <starx at pacbell.net> wrote:
>I may stop doing this in the future though... i think i may make a 
>module 'interface' and do 'form xxx import *'s in there, so then import 
>it as 'i'. i.wxFrame, i.glutPostRedisplay, etc (it would also be nice if 
>there was a sperate module of the GL's that didn't preface GL infront of 
>everything. gl.vertex3d looks a lot better than gl.glVertex3d... :)

def un_uGLify(mod_name):
    import new
    mod = __import__(mod_name)
    m = new.module(mod_name)
    for k, v in mod.__dict__.items():
        if k.startswith('gl'):
            k = k[2:]
        setattr(m, k, v)
    return m
gl = un_uGLify('gl')


... should do what you want... I think.

I don't see what you'd gain from an 'i' module... you'd simply shift all the
potential name clashes and the 'where did this symbol come from'
headscratching over to another module.


From loewis at informatik.hu-berlin.de  Thu Feb 14 16:45:10 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 14 Feb 2002 22:45:10 +0100
Subject: undefined symbols when embedding python
References: <slrna6mm3i.2aj.landshark@burrito.verticaladdiction.net> <m3y9hwpht2.fsf@mira.informatik.hu-berlin.de> <slrna6nkf8.2sp.landshark@burrito.verticaladdiction.net>
Message-ID: <j48z9v68m1.fsf@informatik.hu-berlin.de>

landshark at verticaladdiction.net (Dave Marotti) writes:

> I tried out the all-in-one, and it didn't.  It could have something
> to do with what SIP created, I'm not sure.  But when I attempt to
> import my wrapped stuff in python, it tells me it can't find
> libwraptestc.so

Did you adjust config.c to list your module? (it's in the same
location where libpython.a is, and you need to make a copy and edit it
to your needs).

Regards,
Martin


From ahmebah at hotmail.com  Mon Feb  4 08:17:44 2002
From: ahmebah at hotmail.com (ameoba)
Date: 4 Feb 2002 13:17:44 GMT
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net>
Message-ID: <Xns91AB3616C9C9Bahmebahhotmailcom@198.99.146.10>

"Timothy Rue" <threeseas at earthlink.net> wrote in
news:972.799T919T11694325threeseas at earthlink.net: 

> Autocoding will be a nice thing to have, it'll get rid of all the
> arrogant jerks.


Unfortunately, it will do nothing to remove people spouting off meaningless 
drivel and claiming it's revolutionary.


From wwaldo4 at cox.net  Thu Feb 21 21:31:33 2002
From: wwaldo4 at cox.net (william)
Date: Fri, 22 Feb 2002 02:31:33 GMT
Subject: Python 2.1 and PIL
Message-ID: <94id8.563$s03.36009@news2.east.cox.net>

Ok been playing with this python 2.1.2 and PIL.  The problem is when i
install the PIL it installs it for python1.5.  I don't know how to get the
PIL to install for Python2.1.

I am assuming I need to make some changes in some files before i do
./configure.  Any suggestions?

Thanks.




From jprice at cosette.nampaz.com  Fri Feb  8 09:08:22 2002
From: jprice at cosette.nampaz.com (Jason Price)
Date: Fri, 08 Feb 2002 14:08:22 GMT
Subject: readlines() a file in 2 parts.
Message-ID: <87vgd85a7h.fsf@cosette.nampaz.com>

I have several files to read that logically break into 2 parts.  It would
be much cleaner to write 2 different 'for line in file.readline() :' loops,
but the break point is different in each file, so I don't know the starting
line.  What I tried to do origionally was:

for line in file.readlines() :
   if condition :
      break
   else:
      do thing 1

for line in file.readlines()
   do thing 2

I tried both readlines and xreadlines, and neither seem to do what I'd
like.  Is there a way to do this cleanly, or should I punt, and just make
the loop more complex?

Thanks for your help;
Jason


From logiplexsoftware at earthlink.net  Thu Feb 14 11:44:09 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 14 Feb 2002 08:44:09 -0800
Subject: IPython 0.2.5
In-Reply-To: <Xns91B516D3A1E11RASXnewsDFE1@130.133.1.4>
References: <a4ainp$gts$1@peabody.colorado.edu>
	<fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com>
	<Xns91B516D3A1E11RASXnewsDFE1@130.133.1.4>
Message-ID: <20020214084409.5fb6dc45.logiplexsoftware@earthlink.net>

On 14 Feb 2002 10:14:41 GMT
Philip Swartzleonard wrote:

> Then there are those of us who refuse to upgrade to XP because of MS's 
> recent we-own-your-computer you're-a-pirate-until-proven-otherwise
paying-
> customers-come-dead-last registration BS, among other things. (And 

I wonder how much this is going to affect the size of the Windows user-base
in the long run... hopefully for the better ;)

> generally think 98 is the best thing to come out of them in general :).

Ick.  If you _must_ use Windoze, by all means, get Win2k.  

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From kp87 at lycos.com  Sun Feb 17 07:43:29 2002
From: kp87 at lycos.com (kevin parks)
Date: 17 Feb 2002 04:43:29 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <mailman.1013843905.25550.python-list@python.org>
Message-ID: <5e8bd451.0202170443.7106633a@posting.google.com>

The Wesley Chun book on Python "Core Python" programming has a bit on this.


From bairef2 at cti.ecp.fr  Mon Feb 18 10:45:24 2002
From: bairef2 at cti.ecp.fr (François Baire)
Date: 18 Feb 2002 15:45:24 GMT
Subject: Performances of Pyhton programs
Message-ID: <Xns91B9AA94FE461bairef2ctiecpfr@138.195.8.3>

Are there any benchmark or does anoyone know what are the performances of a 
program written in python (in terms of CPU and RAM) compared with other 
languages (especially C++ and compiled languages) ?
   I've met many articles which explain that python programs are slowler 
because it uses the interpreter, but none says how much slower it is ! I 
would like to know this for a rather big OO program which uses standard 
fonctions of the language.
Thanks

--
Fran?ois Baire


From jkraska at san.rr.com  Sun Feb  3 02:34:59 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 07:34:59 GMT
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012597563.30612.python-list@python.org> <v5F68.110829$8b.4204925@atlpnn01.usenetserver.com> <833d0jyu1u.fsf@panacea.canonical.org>
Message-ID: <sdqp5u49m52b1u8vq1fe1gtm7gjej7t1fj@4ax.com>

>It's true that Java objects all have declared types, but that doesn't
>save you as much as you'd think;

While you obviously know what you're talking about and don't need my
help, I'd just like to add that those folks talking about an optional
type declaration system for speeding up Python don't seem to have a
good idea of where the time in Python goes. There are more important
fish to fry, as I'm sure you realize.

C//



From jyrinxat at mindspringdot.com  Tue Feb 12 22:48:44 2002
From: jyrinxat at mindspringdot.com (Jyrinx)
Date: Tue, 12 Feb 2002 19:48:44 -0800
Subject: Overriding adding items to container
Message-ID: <pan.2002.02.12.19.48.43.817029.26601@mindspringdot.com>

I have a class derived from list (v2.2 is cool ...), and I want to
override its behavior so that a function is called whenever an object is
added to the list (in particular, I want to create a weak reference to
the parent). To wit:

>>> bob = myList()
>>> joe = myItem()
>>> bob.append(joe) # (or insert(joe), or bob[0:0]=[joe], etc.)
>>> joe.parent is bob
1

Is there any good, general way to do this? It seems I can override
__setitem__(), append(), insert(), and others, but it would seem more
elegant just to be able to change one thing. Is this possible? Thanks!

Jyrinx
jyrinxat at mindspringdot.com (spam-proofed)


From jjl at pobox.com  Thu Feb 14 16:32:25 2002
From: jjl at pobox.com (John J. Lee)
Date: Thu, 14 Feb 2002 21:32:25 +0000
Subject: Recommendations for cross-platform GUI Bindings ?
In-Reply-To: <864rkk119z.fsf@apo.hmotakef.int>
References: <xBMa8.217$B92.52511@news.xtra.co.nz> <ictn6uomd63papjtut186c67lickq0cjj9@4ax.com> <864rkk119z.fsf@apo.hmotakef.int>
Message-ID: <Pine.SOL.4.30.0202142130340.24038-100000@mimosa.csv.warwick.ac.uk>

On 14 Feb 2002, Henrik Motakef wrote:

> Courageous <jkraska at san.rr.com> writes:
[...]
> However, I doubt it runs on the Mac. But for Unix/Windows, it's definitely
> my Toolkit of choice.

It doesn't -- until somebody buys Phil Thompson a Mac.


John



From syring at email.com  Mon Feb  4 05:33:29 2002
From: syring at email.com (Karl M. Syring)
Date: Mon, 4 Feb 2002 11:33:29 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <13285ea2.0202030404.5dfe9b66@posting.google.com> <a3j9du$181gf1$1@ID-7529.news.dfncis.de> <13285ea2.0202031557.244bb70d@posting.google.com> <mailman.1012787226.16823.python-list@python.org>
Message-ID: <a3lo1p$18b48u$1@ID-7529.news.dfncis.de>

"Gerhard H?ring" <gh_pythonlist at gmx.de> schrieb
> Le 03/02/02 ? 15:57, J.Jacob ?crivit:
> > [Karl M. Syring]>
> > > This is utter nonsense. Pdcurses simply works on Windows.
>
> True, but this doesn't say anything about building the Python curses
> module againsat it.
>
> In fact, I've tried and I can tell you that this isn't plug-and-play.
> After seeing lots of undefined symbols, I gave up. It simply wasn't
> worth the effort to me.
>
> But you seem to be pretty sure that it's pretty easy, so ...
>
> > Maybe -you- can get it to work?
>
> ... I'd also be interested if you get this to work.

I will give it a try, time to udate to 2.2.

Karl M. Syring




From bc90021 at bc90021redundantdot.net  Mon Feb 18 16:00:35 2002
From: bc90021 at bc90021redundantdot.net (bc90021)
Date: Mon, 18 Feb 2002 21:00:35 GMT
Subject: Newbie question...
Message-ID: <3C716B27.3090408@bc90021redundantdot.net>

X-No-Archive: Yes
Hi all,

Sorry if this has been asked a thousand times already... I tried 
searching google with very little luck.  I'm a newbie to Python, but not 
to Linux.  (Though of course, RedHat has to make everything difficult 
with their RPMS and such... ;)  )

I am trying to install Python 2.2 RPMS on a RedHat 7.2 system.  I can 
get the docs and the devel packages to install, as there are no 
dependencies for those.  However, when I try and install the Python 2.2 
rpm itself, I get the following:

error: failed dependencies:
	rpmlib(PartialHardlinkSets) <= 4.0.4-1 is needed by python-2.2-7
	libcrypto.so.3   is needed by python-2.2-7
	libdb-4.0.so   is needed by python-2.2-7
	libssl.so.3   is needed by python-2.2-7


I have been unable to find out what rpmlib is or what the 
"PartialHardLinkSets" are.

Libcrypto.so.3 and libssl.so.3 are both related to OpenSSL, but as far 
as I can find on RPMFind.net, there is nothing above libcrypto.so.2 and 
libssl.so.2.*

* These are really just symlinks, I learned from searching google. 
Creating symlinks that point to the same thing with .so.3 didn't work.

I also have no idea what libdb-4.0.so is...  Searching for "libdb" on 
RPMFind.net yields Database independent C files that are less than 
version 1, and searching for "libdb-4.0.so" yields development files for 
the Berkeley DB (which are at least the right version).

If I get frustrated enough with this, I will just compile from source so 
that I can run my Python programs.  However, since I uninstalled all the 
old Python stuff, all of my RedHat tools are broken.  I figure it will 
be easier to fix the Python dependencies than that RedHat tools, so 
that's what I'm trying... yes, there are CLI equivalents, and I use them 
most of the time, but once in a while it's nice to use the included 
tools.  ;)

(And, yes, I know, I can always just go back to 1.5, but what's the fun 
in that?  ;)  )

Any and all help is appreciated in advance.

bc90021



From graham_guttocks at yahoo.co.nz  Wed Feb 20 13:28:53 2002
From: graham_guttocks at yahoo.co.nz (Graham Guttocks)
Date: Wed, 20 Feb 2002 10:28:53 -0800 (PST)
Subject: why does popen2 silently ignore IOError?
Message-ID: <20020220182853.88267.qmail@web10301.mail.yahoo.com>

J?rg W?lke <dvass at felis.uni-freiburg.de> wrote:

>> As you can see, no exceptions were raised despite the fact that
>> there is no such file as '/bin/nonexistent'.  Why is this?
>
> Because popen(3) returns a file descriptor or a Null-pointer?
> Try testing for None.

Greetings,

I don't quite understand, where should I test for None?

Also, this behavior really seems wrong to me.  I don't see how doing a
popen to a non-existent file, then writing to it and closing it all
succeed without raising any warnings or exceptions.  This seems very
dangerous to me.

Regards,
Graham


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From jkraai at paypal.com  Wed Feb 13 19:33:04 2002
From: jkraai at paypal.com (Jim Kraai)
Date: Wed, 13 Feb 2002 16:33:04 -0800
Subject: Container Objects Enforcing Contents
Message-ID: <7A9A4362115F384095113F27CBF8BC43F0991F@mv1-mail2>

On 13-Feb-2002 Sean 'Shaleh' Perry wrote:
> On 13-Feb-2002 jim kraai wrote:
> > Greetings,
> > 
> > I'm looking for suggestions on making container objects which must,
> > may, or must not allow particular kinds of container objects to be
> > added.
> > 
> > Any suggestions on a general approach for this?
> 
> Let's see .......
> 
> a constraints approach:
> 
> class Foo:
>     def __init__(self, item):
>         try:
>             # look for a method of item that must exist
>         except AttributeError:
>             raise ConstraintException
> 
> This way you could add only items which support a method or have a certain
> attribute.
> 
> You could also try to enforce types by ensuring the type of item matches
types
> you know about.  This often leads to a maintenance nightmare though.

Maybe a dictionary approach based on the parental enforcement
class Type1:
	def __init__(self,item):
		self.type = 'Type1'
		self.children_must = ['type2','type3']
		self.children_may  = ['type4','type5']
		}

	def addChild(self,child):
		if (
			! child.name in self.children_must or
			! child.name in self.children_may
		):
			raise exception

The problem with this is that it can accumulate children over time, 
but when will all of the constraints be finalized/enforced?  Add a 
finalize method?

	def finalize(self):
		# loop over self.children_must and verify existence
		# if constraints not met, raise exception
		# recurse and call finalize for all children

This sounds like a pain, the mandatory call to root.finalize() would 
be a good way to make enforcement _optional_, but maybe not good in a 
context where you want it to be enforced 100% of the time.

Maybe leave the enforcement responsibility on the child, but this 
could be semantically very ugly because of the need to pass the parent 
to the child for introspection.

Maybe child enforcement could be done by the parent calling a method 
in the child class before the child is actually added
in (parent) class Type 1:
	def addChild(self,child):
		if child.validateParent(self):
			self.children.append(child)

in (child) class Type 2:
	def __init__(self):
		self.valid_parents = [list of acceptable parents]
	def validateParent(self,parent):
		if ! parent.type in self.self.valid_parents:
			raise exception

It looks like leaving validation up to the child might be marginally 
easier, but it doesn't remove the need for the finalize call.

--jim



From cfelling at iae.nl  Thu Feb 28 11:29:35 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 17:29:35 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl> <a5lk0a$8enfq$1@ID-11957.news.dfncis.de>
Message-ID: <a5lltf$4ra$1@animus.fel.iae.nl>

Emile van Sebille <emile at fenx.com> wrote:
> "Remco Gerlich"
>> Compare:
>>
>> for i in 6: print i
>> for i in 6,7: print i
>> for i in 6,7,8: print i
>>
>> You wouldn't think the first line does something different, would you?

> Is that really any different from what we have now:

> for i in "Spam": print i

Yes, it clearly is!  `"Spam"' is a sequence, and iterating over a sequence
is understandable.

> for i in "Spam","and": print i

And here `"Spam", "and"' is a sequence of sequences, and again iterating
over a sequenc--even if it's a sequence of sequences--is understandable.

But iterating over an integer is even after that number theory lessons
we had last night still very hard for me to appreciate.


> I find the best argument for adding an iter method to ints is that it
> allows me to say:

>     for ii in iter(len(mylist)):

And how would this compare to

      for ii in indici(mylist):

This would read `for all ii in the indici for mylist', extremely clear
to me.  We could even consider enhancing range to accept a list in
case of a single argument.


> However, the clarity and intent conveyed by allowing this, particularly
> for newbies, more than offsets the gotcha's, which, once learned, is the
> same set of gotcha's in place for strings.

And here you lost me. How can turning a single number into a sequence of
all its preceding numbers be easy to understand for someone not versed
in modern number theory or without a crack for weird syntactic sugar.



-- 
groetjes, carel


From afp_randjohnson at yahoo.com  Sun Feb 24 05:34:51 2002
From: afp_randjohnson at yahoo.com (Mike Johnson)
Date: 24 Feb 2002 02:34:51 -0800
Subject: Extracting data from telnet session ?
References: <m3d6yuai41.fsf@marnie.henning-sauer.de>
Message-ID: <4rk74184.fsf@fido.trinitycapital.com>

Henning Sauer <hsauer at marnie.teuto.de> writes:
> 
> 
> What I would like to know is how can I extract the information from
> this data stream ?  Is this really the data from the router or might
> it be debug information ? When I'm not setting the set_debuglevel
> I'm getting no output.

That suggests that you haven't logged in yet.

Try logging into the machine and running "who" to see if your app
really did log into the server. My guess is that the "password:"
prompt is off, possibly you will need "Password: ". Otherwise, your
app may wait forever, scanning for the wrong prompt.

Also, you will need to use "\r" instead of "\n" for newline
characters.

I'm not sure if it's still needed (somebody here can correct me), but
you used to need to write:

mytelnet.write("\255\253\003") #this is the "fix" command added

after logging in.

For some odd reason I don't remember. :-)

- Mike Johnson



From skip at pobox.com  Tue Feb 12 18:42:15 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 17:42:15 -0600
Subject: Major league corruption - anyone seen anything like this?
Message-ID: <15465.43095.985391.388092@beluga.mojam.com>

I'm developing a small library and accompanying wrapper.  I sometimes get a
segmentation fault (environment is Mandrake 8.1, Python is from CVS).  Upon
my return to the shell prompt, I discover that the /usr/local/bin/python
link is gone as are a number of shared libraries (readline among them).

Obviously I have some serious problems in what was supposed to be a simple
wrapper (perhaps the problems are with the underlying library which has only
been lightly tested), but I've never seen a segfault take files with it like
this.  Has anybody else encountered this before?

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From huaiyu at gauss.almadan.ibm.com  Wed Feb 27 20:04:50 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Thu, 28 Feb 2002 01:04:50 +0000 (UTC)
Subject: map-like function on dict values?
Message-ID: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>

Is there a way to do this

for k, v in dict.items():
    dict[k] = f(v)

without involving a Python for loop?  I'd imagine that a function
implemented in C could speed things up quite a bit.  Something like
either of the following

mapvalue(f, dict)
dict.mapvalue(f)

Huaiyu


From loewis at informatik.hu-berlin.de  Thu Feb 28 13:07:12 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 19:07:12 +0100
Subject: struct_time type testing 2.1-2.2
References: <VCtEHjA9Nmf8Ewzm@jessikat.demon.co.uk>
Message-ID: <j4r8n5eazz.fsf@informatik.hu-berlin.de>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> is there a cleverer way than
> 
>     if tn in _SeqTypes or tn is getatr(time,'struct_time',None):
> 
> to get a cross version test?

I'm not sure it's very elegant, but what about

      if tn.__name__ in ['list','tuple','time.struct_time']

I actually prefer

try:
   from time import struct_time
except ImportError:
   _TimeTypes = (ListType, TupleType)
else:
   _TimeTypes = (ListType, TupleType, struct_time)


if tn in _TimeTypes

> Presumably we could extend _SeqTypes as that needs to be done only once,
> but ideally a simple check like
> 
>         arg.supports_index(3)
> 
> would be more informative than defining these rather arbitrary named
> type lists.

For that, I'd just write

  try:
    self.normalDate = int("%04d%02d%02d" % normalDate[:3])
  except TypeError:
    # else code

Regards,
Martin


From brueckd at tbye.com  Thu Feb  7 10:01:28 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 7 Feb 2002 07:01:28 -0800 (PST)
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users"
 ?
In-Reply-To: <7xr8nxy0rb.fsf@ruckus.brouhaha.com>
Message-ID: <Pine.LNX.4.33.0202070643020.4847-100000@ns0.tbye.com>

On 7 Feb 2002, Paul Rubin wrote:

> As much as I like Python, I don't see it working so well in the kind
> of large, real-world projects that Java works pretty well in:

Why? You listed some cases where you don't think it would work well, but
didn't give any reasons why you think that. Give us some insight into your
line of thought. :-)

>   - big codebase (>200K LOC) being developed/maintained by lots (say
>     20 or more) of people
>
>   - Program stays around for years but developers come and go.
>     Average turnover is maybe 1 programmer per month.  So there
>     are always people hacking on the codebase who aren't that familiar
>     with it.
>
>   - The programmers are what you get through normal industry-style
>     recruiting processes, i.e. they are mostly pretty good, but not great.
>
> In this environment, C is an unholy nightmare, C++ is somewhat better,
> but Java seems to do a much better job than either of keeping
> complexity under control and avoiding crazy application bugs.

Agreed, but to me the above 3 are examples of when Python would shine:

- Big codebase - in cases where we've had C, C++, or Java code and
equivalent Python code, the Python version has tended to be anywhere from
2 to 10 times shorter. So a huge project can be a medium project just be
virtue of the language. It's also been my experience that the code
requires fewer developers, but I don't have a good feel for how many
fewer.

- Program stays but programmers go - the smaller codebase pays of hugely
here, as does the general cleanliness of Python code. In the large Java
projects I've worked on, the biggest struggle was coming up to speed on
the jazillions of different files in million different directories, each
with very verbose code. On the most recent Java project I was involved in,
I remember that even after being there for several months I was still
doing lots of "Browse for symbol" type lookups - all the code fluff in
Java made it very hard to keep in my head.

- Good but not necessarily great programmers - to me this has less to do
with language and more to do with hiring processes and development
processes. In the specific case of Java, there seem to be (well, in my
limited experience at least) lots of mediocre-to-poor Java programmers
(possibly because they became Java programmers due to the hype surrounding
it) and I've fared better by hiring experienced C/C++ programmers and
having them learn Java than by hiring programmers who are just good at
Java. That being said, Python is still benenficial for the above reasons -
few LOC, easier to get a handle on the code, clearer code - all things you
want if you have less than stellar developers. Being able to fire up an
interpreter to try stuff out also encourages them to do more incremental
testings. Python helps bad programmers be good, and good programmers be
better.

One last comment: I haven't found Java to work all that well in those
"large, real-world projects" you mentioned. Better than C and C++ in many
cases, but as the projects got bigger Java seemed to stop solving problems
we would have had in C++ and instead began replacing them with different
ones (the proliferation of files and directories was a big one).

 -Dave




From phr-n2002a at nightsong.com  Sun Feb 17 19:18:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 16:18:04 -0800
Subject: basic books/guides on multithreading programming?
References: <mailman.1013813246.12449.python-list@python.org>
Message-ID: <7xlmdrmymb.fsf@ruckus.brouhaha.com>

Tripp Scott <tripps81 at yahoo.com> writes:
> which ones do you recommend? examples in any language is fine, though
> i prefer python, perl, or ruby. what i would like is guidance on what
> is thread safety, how to achieve it, what are the issues in
> multithreading programming, the global interpreter lock, and other
> basics.

There's a pretty good book called (I think) Java Threads.  I looked at it
ages ago so I don't remember too much specific from it.  Python's threading
module is sort of a subset of Java threads.


From logiplexsoftware at earthlink.net  Wed Feb  6 02:26:37 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 5 Feb 2002 23:26:37 -0800
Subject: Autocoding evolves from........
In-Reply-To: <811.802T1774T34536threeseas@earthlink.net>
References: <4314.793T504T6883966threeseas@earthlink.net>
	<3c5742e8$0$79557$3c090ad1@news.plethora.net>
	<slrna5fnkg.l4j.philh@comuno.freeserve.co.uk>
	<972.799T919T11694325threeseas@earthlink.net>
	<Xns91AB3616C9C9Bahmebahhotmailcom@198.99.146.10>
	<811.802T1774T34536threeseas@earthlink.net>
Message-ID: <20020205232637.39127faa.logiplexsoftware@earthlink.net>

On Wed, 06 Feb 2002 04:54:36 GMT
Timothy Rue wrote:

> On 04-Feb-02 08:17:44 ameoba <ahmebah at hotmail.com> wrote:
> >"Timothy Rue" <threeseas at earthlink.net> wrote in
> >news:972.799T919T11694325threeseas at earthlink.net:
> 
> >> Autocoding will be a nice thing to have, it'll get rid of all the
> >> arrogant jerks.
> 
> 
> >Unfortunately, it will do nothing to remove people spouting off
meaningless
> >drivel and claiming it's revolutionary.
> 
> Hmmmm, try evolution a bit.
> 
> Was GNU revolutionary or evolutionary?
> 

All that matters is that GNU was/is more than just empty words.  Perhaps
you should provide some functional code before you start comparing yourself
to Richard Stallman. 

BTW, I would point out the irony of comparing yourself to an achieved
programmer in light of your earlier statements regarding such individuals,
but I expect it would be wasted on you.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From donn at u.washington.edu  Thu Feb 14 13:36:25 2002
From: donn at u.washington.edu (Donn Cave)
Date: 14 Feb 2002 18:36:25 GMT
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com> <1013641215.358571@yasure> <a4f29e$bhm$1@animus.fel.iae.nl>
Message-ID: <a4h039$2r2q$1@nntp6.u.washington.edu>

Quoth Carel Fellinger <cfelling at iae.nl>:
| Donn Cave <donn at drizzle.com> wrote:
| ...
| > Check it out.  Not popen, but as they say, "popen2 et al."
| > It will exec the command on its own, no shell, if you give it
| > an argument list.  If you give it a string, it will use the
| > shell.
|
| How is one supposed to know this?  (excluding reading the source or
| remembering from the last time you brought this up:)
|
| I've reread the fine manuals, but can't find a clue hinting at this
| interesting behaviour.

I guess I should read the manuals some time.  I wonder if they explain
why os.popen2 reverses the order of popen2.popen2's returns.

	Donn Cave, donn at u.washington.edu


From shredwheat at attbi.com  Sat Feb 23 13:16:04 2002
From: shredwheat at attbi.com (Pete Shinners)
Date: Sat, 23 Feb 2002 18:16:04 GMT
Subject: pygame puzzler
References: <1eed326b.0202222300.4666822c@posting.google.com>
Message-ID: <3C77DDD9.8020507@attbi.com>

Gil wrote:
> Can somebody help me here? I've installed SDL_mixer-1.2.1 from a
> .tar.gz package without complaint except one:
> "SMPEG-version>=0.4.3...no"

this shouldn't be a problem. if SDL_mixer can find the SMPEG library 
when it configures, it enables MP3 playback. without the library you 
just get no MP3 support. so it's definitely optional and shouldn't stop 
things from working...


> [caltrop at localhost examples]$ python sound.py
> Traceback (most recent call last):
>   File "sound.py", line 10, in ?
>     import pygame.mixer, pygame.time
> ImportError: No module named mixer

the mixer module in pygame is also optional. that means pygame can build 
and install without it. pygame comes with a "config.py" script, and if 
it cannot find the SDL_mixer library, it disables the pygame.mixer 
module. the first thing to check is the "Setup" file. this is a simple 
text file created by the config, it stores the compile settings. open 
the file and first check to see if there are compile flags for the 
"MIXER" variable. once that seems reasonable, scroll down to the next 
section which lists the optional modules. make sure the line for "mixer" 
is not commented out. (uncomment "mixer_music" too)

if the Setup file already looks ok, then it means the mixer module 
probably already compiled ok. perhaps take a look for "mixer.so" in the 
site-packages.

if the Setup file is now looking correct. rerun the "setup.py" which 
will get your installation up to date. as long as it doesn't error out, 
i'm certain the mixer module will be installed. if the setup.py runs 
cleanly and it still doesn't work, i'm guessing it might have something 
to do with mismatched python versions?



> My sound card works and I ran test_linuxaudiodev.py successfully.

the problem here is not that the sound doesn't work, but the needed 
pygame.mixer modules isn't being built, or isn't being found by python.




From spencer at efn.org  Fri Feb  1 12:31:43 2002
From: spencer at efn.org (Spencer Ernest Doidge)
Date: Fri, 1 Feb 2002 09:31:43 -0800
Subject: Windows, RS232, and Python
Message-ID: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org>

I want to have control of every aspect of RS232 transmitting and
receiving in Win9x/NT with Python commands.

Question 1:
What such functionality is already built into Python and where is it
documented?

Whatever functionality is lacking, I will implement in Win32
API. My final product will present a complete GUI to the end user, but
behind-the-scenes folks will use extended Python from  the command line.

Question 2:
Between two approaches, which would work better:

A. Extend Python as needed, then embed this extended Python into my GUI
app. This would necessitate maintaining two applications: the extended
command-line Python for the behind-the-scenes users, and the GUI app for
the customers.

B. (I'm on shaky ground here, but from my early readings, I suspect that
this might be possible.)... Write my GUI to fire up a 2nd process, i.e.,
the extended Python, then somehow make calls to it across process
boundaries. By doing this, I would only have to maintain the GUI as
itself, and the extended Python as itself. Making a change in the extended
Python would not a priori require me to rebuild the GUI.

To anyone who takes the time to read and answer this post, I extend my
thanks in advance.

Spencer Doidge



From starx at pacbell.net  Thu Feb 14 05:33:48 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 14 Feb 2002 10:33:48 GMT
Subject: Disabling creation of .pyc files?
References: <E16b1mB-0001Ls-00@mail.python.org> <mailman.1013619835.26532.python-list@python.org>
Message-ID: <Xns91B51A11D5957RASXnewsDFE1@130.133.1.4>

Skip Montanaro || Wed 13 Feb 2002 09:03:43a:

> 
>     Dominic> ... I'm a bit annoyed by the fact that for every .py file
>     I'm Dominic> importing, Python automatically creates a
>     corresponding .pyc Dominic> file, which persists after my program
>     exits. 
> 
> It's there for your benefit (imports are faster when up-to-date .pyc
> files are present because the compilation step is avoided), but yes,
> you can disable it by simply making the directory containing the .py
> files read-only.
> 

Does this work under windows? I just set a folder read-only (it is an 
option, but FIAK the 'properties' window might lie or cheat regarging the 
underling attrib calls, but...), and it didn't seem to prevent me from 
adding files or changing names...

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From db3l at fitlinxx.com  Fri Feb 22 19:57:55 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 22 Feb 2002 19:57:55 -0500
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1>
Message-ID: <uzo21ovzg.fsf@ctwd0143.fitlinxx.com>

Jim Richardson <warlock at eskimo.com> writes:

> I have an application that does a calculation on a single byte of data,
> after breaking it into an upper and lower nibble of four bits. How can I
> do this in python? To clarify.

As others have mentioned the actual bit twiddling is fairly straight
forward (| for or, & for and, and so on) and since this is just a byte
you shouldn't run into some of the more subtle issues with 32-bit
unsigned values (since Python integers are 32-bit signed).

But one thing I haven't seen mentioned is that you will probably want
to make use of the ord() and chr() functions to convert between a
single character string and an integer value and vice versa.

That is, assume you have a string buffer that contains bytes you read
from a serial port or something, and you want to flip the two nibbles.
For example:

    >>> value = '\x12\x34\x56\x78'
    >>> print map(hex,map(ord,value))
    ['0x12', '0x34', '0x56', '0x67']
    >>> result = ''
    >>> for char in value:
    ...   byte = ord(char)
    ...   byte = ((byte & 0xF) << 4) | ((byte & 0xF0) >> 4)
    ...   result = result + chr(byte)
    ...
    >>> print map(hex,map(ord,result))
    ['0x21', '0x43', '0x65', '0x76']

Obviously this particular example isn't very efficient in building a
result string, but I went for clarity rather than performance.
Building up a list of converted values and then joining them at the
end is a clear optimization.  For larger buffers, using map(ord,value)
as in the print to first convert to a list of integers, mutate it, and
then map(chr,result) and string.join to put it back into strings would
probably be the most efficient at the expense of some extra memory.

The other suggestion of a translation table also works really well if
the mapping is fixed and amenable to generating such a table up front.

For example, here's a small excerpt from a script I used to process a
file and invert the bits on each "word" in the file (16-bit/2 byte
words by default).  I chose to use a single byte translation table and
just reverse the bytes in each word with reverse(), since that seemed
a reasonable tradeoff in terms of space (e.g., if the word size was
set to 32-bit) and performance.

    # Create a byte translation table for efficiency
    xlate = {}
    for byte in range(256):
        key = byte
        rbyte = 0
        for bit in range(8):
            rbyte = rbyte << 1
            rbyte = rbyte | (byte & 1)
            byte  = byte >> 1
        xlate[key] = rbyte

    while 1:
        buffer = input.read(wordsize)
        if not buffer: break

        bytes = map(ord,buffer)
        bytes.reverse()

        reversed = string.join(map(lambda x,xlate=xlate:chr(xlate[x]),
                                   bytes),'')

        output.write(reversed)

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


From thoa0025 at mail.usyd.edu.au  Fri Feb 15 08:50:54 2002
From: thoa0025 at mail.usyd.edu.au (Trung Hoang)
Date: Fri, 15 Feb 2002 13:50:54 GMT
Subject: certification
Message-ID: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>

Hi,

i was wondering if i can get certified with python from an internationally
recognised company?

cheers

--
Trung Hoang
Final year Undergraduate of Computer Science and Technology
University of Sydney, Australia.




From theller at python.net  Thu Feb 21 05:08:39 2002
From: theller at python.net (Thomas Heller)
Date: Thu, 21 Feb 2002 11:08:39 +0100
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <3C6D8B63.5070909@geocities.com> <mailman.1013864067.16756.python-list@python.org> <69587ukeu1m51mf46kpumjnt9qfujho7ju@4ax.com>
Message-ID: <a52h06$43hjh$1@ID-59885.news.dfncis.de>

Another interesting technology are Jean-Claude Wippler's
'Scripted Documents'. The main link seems to be

http://www.equi4.com/scripdoc/

but you should also google for others.

Thomas




From martin at v.loewis.de  Wed Feb 20 02:09:44 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 08:09:44 +0100
Subject: print and unicode strings
References: <mailman.1014174755.8438.python-list@python.org>
Message-ID: <m38z9o6347.fsf@mira.informatik.hu-berlin.de>

"Jason Orendorff" <jason at jorendorff.com> writes:

> I think IDLE should do this by default.  It's fine for
> Python to "refuse to guess" encodings, but in the case of Tk
> it's UTF-8 on all platforms... right?

Well, not exactly. Tkinter supports Unicode strings as primary data
type for text, with byte strings (e.g. UTF-8) only being the second
choice. So IDLE should set sys.stdout to a stream that accepts Unicode
objects as-is.

If you wonder how Tcl interprets byte strings: All bytes below 128 are
treated as ASCII. All byte sequences above 128 are considered as UTF-8
if possible. If there is an encoding error (because of illegal
sequences), they are treated with the platform's native encoding. This
is a pretty crude heuristics, since it means that you may end up with
a byte string that  uses different encodings at different offsets.

Also, setting the stream for the console to cp437 is not ideal; it
would be much better if Python would use the Unicode console API
(i.e. WriteConsoleW) if it detects that sys.stdout is a console.  If
you then use the Lucida Console font (instead of the raster font), you
might be even able to display COMET in a console.

Regards,
Martin


From anthony_barker at hotmail.com  Thu Feb 14 12:24:30 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 14 Feb 2002 09:24:30 -0800
Subject: URLLIB2 problems (solved)
References: <899f842.0202121213.603e6617@posting.google.com> <899f842.0202130742.2bc64e4c@posting.google.com>
Message-ID: <899f842.0202140924.454fcfb6@posting.google.com>

After a bit of frustration and trying a bunch of things (reading the
urllib2.py file) i found some documentation by Jeremy and Moshe Zadka.

ProxyHandler([proxies])'
     Cause requests to go through a proxy.  If PROXIES is given, it
     must be a dictionary mapping protocol names to URLs of proxies.
     The default is to read the list of proxies from the environment
     variables `PROTOCOL_proxy'.

So I added the environment variable to my .bash_profile in the format

export HTTP_PROXY="http://user:pass at proxy.com:8080/"

relogged in and checked my environment variables.

launched the python interpreter

import urllib2
f = urllib2.urlopen('http://www.python.org')
print f.read()

It works!!


Just a quick note - if you want to find out the realm of your proxy. 2
methods I found:
1) use netscape 4.7
2) telnet in on port (eg 8080) 
Type GET / HTTP/1.0

You will get something like this:
HTTP/1.0 407 Proxy Authentication Required
Date: Thu, 14 Feb 2002 16:57:27 GMT
Content-Length: 233
Content-Type: text/html
Server: NetCache (NetApp/5.1R2D15)
Connection: keep-alive
Proxy-Authenticate: basic realm="NETCACHE2"

Telling us that it uses basic authentication (not MD5) and the realm
is "NETCACHE2"

> > I am having problems with python and our company's proxy server, which
> > requires  authentication.
> > 
> > here is the code I am using:
> > 
> > import urllib2 
> > proxy = urllib2.ProxyHandler({"http":"http://myname:mypass at proxyserver.company.com:8080/"})
> > opener = urllib2.build_opener(proxy) 
> > f = urllib2.urlopen('http://www.python.org') 
> > 
> > results
> > Traceback (most recent call last):
> >   File "<stdin>", line 1, in ?
> >   File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
> >     return _opener.open(url, data)
> >   File "/usr/lib/python2.2/urllib2.py", line 322, in open
> >     '_open', req)
> >   File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
> >     result = func(*args)
> >   Fil
> 
> e "/usr/lib/python2.2/urllib2.py", line 785, in http_open
> >     return self.do_open(httplib.HTTP, req)
> >   File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
> >     raise URLError(err)
> > urllib2.URLError: <urlopen error (101, 'Network is unreachable')>
> > 
> > Under windows error is the same except for the last line (and line
> > numbers)
> > URLError: <urlopen error (10051, 'Network is unreachable')>
> > 
> > Both perl and wget work. wget only supports Basic Authentication, so I
> > assume it is the same. Thinking it might be a network type issue I
> > changed the domain names to IP addresses. (results follow)
> > 
> > Any help would be appreciated................
> > 
> > File "<stdin>", line 1, in ?
> >   File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
> >     return _opener.open(url, data)
> >   File "/usr/lib/python2.2/urllib2.py", line 322, in open
> >     '_open', req)
> >   File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
> >     result = func(*args)
> >   File "/usr/lib/python2.2/urllib2.py", line 785, in http_open
> >     return self.do_open(httplib.HTTP, req)
> >   File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
> >     raise URLError(err)
> > urllib2.URLError: <urlopen error (111, 'Connection refused')>


From srumbalski at prodigy.net  Mon Feb 11 21:19:15 2002
From: srumbalski at prodigy.net (Steven Rumbalski)
Date: Tue, 12 Feb 2002 02:19:15 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <eSz98.4759$zb.260473@news1.east.cox.net> <3C682A16.1D6C800D@ccvcorp.com> <Xns91B28C2B6EBE0RASXnewsDFE1@130.133.1.4>
Message-ID: <DY_98.1732$v94.371156124@newssvr17.news.prodigy.com>

Philip Swartzleonard wrote:

> Then it becomes 'Guidoista' perhaps, for following a specific leader? ;)

'Guidoista': three syllables or four? GWEE-DOY-STAH or GUI-DOH-EE-STAH?


From skywalkerpackage at hotmail.com  Thu Feb  7 13:36:29 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Thu, 07 Feb 2002 18:36:29 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org>  <3c60fb80.199334@news.online.de> <mailman.1013016376.5624.python-list@python.org>
Message-ID: <3c62b7ab.113923401@news.online.de>

Laura,

Thanks very much for the code. I think it is very helpful for me.
I have analyzed your program and it seems to contain a lot of good
tricks. However, I probably didn't understand the main principle
behind. I suggest the best is if I wrtte some commends in between your
code

On Wed, 06 Feb 2002 18:23:55 +0100, Laura Creighton <lac at strakt.com>
wrote:

...

>> This doesn't work together with an optionDict={'labelText':'hello',
>> 'padx':'6'}
>---------------------------------------------------^^^^^^^^
>
>That is your problem.  Tkinter labels want to get their text
>set with text='hello', not labelText='hello'

Yes. Then, what I wanted to do define a configure (or a config) method
that worked exactly as the one(s) in the Tkinter module.
Unfortunately, I don't think I will understand the code inside Tkinter
in limited time ;<

>
>You have to make some design decisions here.  If you want to pass one big
>dictionary, you will then have to unpack it, parse it, and generate
>the correct keys for the correct widgets.  This is do-able; it is
>what PMW does.  But I didn't like that approach.  I just use 2 dictionaries.

Why didn't you like the PMW approach?

2 dictionaries for what? Do you mean one master dictionary for your
own megawidgets and for basic (Tkinter/Pmw) widget which appear in the
higher level of your GUI, and one or more lowlevel dictionaries for
the basic widgets which are contained by the highlevel widgets? I am
not clear what kinds of the two dictionaries you mean (see below).

>
>You also mention:
>>But then, how do I handle default values, and how do I ensure that the
>>objects are created with the right parameters? Maybe I should use a
>>local dictionary.
>
>[snip]
>
>>Is there an elegant solution to this problem?
>
>Well, I don't know if you would call this elegent but ...
>
>
>from __future__ import nested_scopes
>import Tkinter
>
>class LabelledWidget(Tkinter.Frame):

Ok, this is an example of your own megawidget

>    def __init__(self, parent, widgetName=Tkinter.Entry,
>                 labelDict={}, widgetDict={}, **frameDict):
>        
>        Tkinter.Frame.__init__(self, parent, **frameDict)
>
>        """
>        let us say that you always want the label to be written on a
>        cyan background in purple letters, unless you ask differently.
>        The widget you just want on a green background.
>        """
>        
>        ldefaults={'bg': 'cyan', 'fg' : 'purple'}
>        wdefaults={'bg': 'green'}
>        # overwrite any of the defaults with the dictionary you passed
>        ldefaults.update(labelDict)
>        wdefaults.update(widgetDict)
>       
>        self.label = Tkinter.Label(self, **ldefaults)
>        self.label.grid(row=0, column=0, sticky='w')
>        self.widget = widgetName(self, **wdefaults)
>        self.widget.grid(row=0, column=1, sticky='e')
>
>########
>if __name__ == '__main__':
>    root = Tkinter.Tk()
>    widgetList = (
>        [Tkinter.Entry,
>         {'text':'An entryfield label:'},{}],
>        [Tkinter.Button,
>         {'text':'A button label:', 'bg':'peachpuff'},
>         {'text':"I'm a button"}],
>        [Tkinter.Label,
>         {'text':'A label label:'}, {'text':"I'm a label", 'fg':'honeydew'}],
>    )

This is probably one of your two kinds of dictionaries, right?
It is specific for your megawidget, so, in general, you use one
dictionary for the construction of your megawidget out of its basic
(Tkinter) widgets.

>    # for testing.
>    def doit(widgetlist, part, method, **dict):
>        for w in widgetlist:
>            object = getattr(w, part, None)
>            if object:
>                command = getattr(object, method, None)
>                if callable(command):
>                    command(**dict)

This is a very interesting routine, but you can only use it together
with a uniform widgetlist (that is, for example, a list containing
only LabelWidget objects). Also, you can only apply the function to
one subwidget (basic widget) and one subwidget method thereof.

>
>    r = 0
>    widgets=[]
>    for name, ld, wd in widgetList:
>        w = LabelledWidget(root, name, ld, wd)
>        widgets.append(w)
>        w.grid(row=r, col=0, sticky='ew')
>        r += 1
>
>    r = 0
>    for w in widgets:
>        w.grid(row=r, col=0)
>        r += 1

Is there a reason why you have called the grid manager twice?

>
>    cmdList = (
>        ['pink label', 'label', 'configure', {'fg':'pink'}],
>        ['yellow widget', 'widget', 'configure', {'bg':'yellow'}],
>        )

Ok, I guess that is the second dictionary. It contains attributes of
the megawidgets (that is, the basic widgets) and the options. Well, I
can't see much more.

>
>    c = 0
>    for txt, part, method, dict in cmdList:
>        button = Tkinter.Button(root, text=txt,
>                                command=lambda w=widgets,
>                                p=part, m=method, d=dict :
>                                doit(w, p, m, **d))
>        button.grid(row=r, col=c, sticky='w')
>        c += 1
>    button = Tkinter.Button(root, text='Exit', command=root.destroy)
>    button.grid(row=r, col=c)
>    
>    root.mainloop()
>--------------------------------------
>
>Laura Creighton
>

Ok, what I try to do is the following:

I want to create a dialog box (a GUI window). This dialog box is
divided into several panes. The panes do have widgets (basic widgets
*and* composed widgets) in each side (left and right).

The structure of my dictionary (the only as you said is the case in
Pmw) is as follows

    # lstPanes = [(pane1_text, 
[(widget_pane1_left, {option_dict_pane1_left}),
 (widget_pane1_right, {option_dict_pane1_right})
]                         ),
                          (pane2_text,

.......

For example, the following list defines a dialog box with three panes.
The first pane has two LabelEntry megawidgets (my own created) on the
left and on the right, respectively. The second pane ('Compilation
mode') consists of a radiobar (again my own widget, just a column of
several radio bars) to the left and so on.
>From the third pane it is clear that the widgets inside a pane can
change and don't have to belong to one widget class. Also, the amount
of columns is variable (one in the second pane and three in the third
pane).


    lstPanes = [('Root class definition', [
                            (LabelEntry, {'labelText':'Root class'}),
                            (LabelEntry, {'labelText':'Root
procedure'})
                            ]),
                ('Compilation mode', [
                            (Radiobar2, {'radioTexts':['boost,
'no_check', 'all_checks, 'debug_check']})
                            ]),
                ('Pane3', [
                            (Checkbox, {....}),
                            (LabelEntry, {'labelText':'label5'}),
                            (RadioButton, {...})
                            ])
                ]


This scheme already works with Tkinter widgets in the columns of the
panes, but not with my own megawidgets (neither with Pmw megawidgets).

I would just find it useful to be able to code a whole GUI window in
this way, but perhaps it is not worth the effort after all.


Johannes


From duncan at NOSPAMrcp.co.uk  Thu Feb 21 04:33:00 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Thu, 21 Feb 2002 09:33:00 +0000 (UTC)
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <Xns91BC6120D2B5duncanrcpcouk@127.0.0.1>

Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote in 
news:pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi at 4ax.com:

> 12. This is not exactly about Python, but about programming in general.
> I keep reading about continuations and coroutines and I would like to
> know, in simple terms if possible, what the heck are they and in what
> ways they differ from generators as they exist in Python.

The main differences between a coroutine and a generator are:

1) Coroutines maintain their own stack. This allows a coroutine to yield 
not just from the function you originally called, but also from any nested 
function no matter how deep.
Python's generators can only yield directly from the generator as they do 
not attempt to maintain any kind of stack.

BUT, this is misleading. If you want to split a generator function up into 
smaller functions, say to make the code clearer, you can do this provided 
all the smaller functions are also generators and the outer generator calls 
them in a for loop and immediately yields out. So although a generator 
doesn't have an explicit stack maintained by the Python runtime, you can 
nest generators so that they behave in a very similar way to coroutines.

2) Coroutines allow you to pass arguments in each time they are activated 
whereas generators only take arguments when they are created. There is a 
conceptual difference here as coroutines are more peer to peer but 
generators have more of a parent (the caller) and child (the generator) 
relationship. Probably the best way to pass values in to a generator is to 
pass it an iterator as an argument, but this may not be as convenient as 
passing a value directly.

-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From mlh at vier.idi.ntnu.no  Fri Feb  8 07:05:37 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Fri, 8 Feb 2002 12:05:37 +0000 (UTC)
Subject: range(seq)?
Message-ID: <slrna67foh.dl7.mlh@vier.idi.ntnu.no>

I just had an idea... I'm sure there are others out there who don't
particularly like the much-used (and necessary) expression
range(len(seq))... One proposed solution is to make integers iterable,
so one can do:

  for x in len(seq): ...

That is one way of doing it, of course (i'm not sure I'm all that fond
of it, but that's another matter ;)

Then it occurred to me that perhaps it could be done the _other_ way:

  for x in range(seq): ...

I.e. let range() take a sequence (or iterable) as an argument, and
generate range(len(seq)). (I guess in mathematical terms, if one sees
a sequence as a function, that domain(seq) would be more appropriate,
but what the hey ;)

Of course, we have the proposed indices(seq)... But by allowing
range() to work on iterables, we wouldn't have to add another
function. And I can't see any significant code breakages (but I'm sure
there are some... :)

So, why is this a stupid idea? (I'm sure it is...)

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From jason at jorendorff.com  Mon Feb 18 14:34:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 13:34:02 -0600
Subject: There is more than one way to do it - and for no apparent reason.
In-Reply-To: <3C70ECB1.2090709@mxm.dk>
Message-ID: <HFEKILOLEFEFMKAECNDLAEHODDAA.jason@jorendorff.com>

> Basically you can say that JScript uses the dictionary notation for 
> getting attributes. As fas I can see this is smarter than the way we do 
> it in Python where we have several notations for the same action:
> 
> In an object:
> setattr(myObj, 'name', 'max')
> getattr(myObj, 'name')
> 
> In a dict:
> myObj['name'] = 'max'
> myObj['name']

They're not really the same action in Python.

Dictionaries have methods in Python. For example:

  d = {'keys': 'a b c', 'values': 'x y z'}
  print d.keys()

Here "d.keys" is a method.  It does not refer to the string
'a b c'.  Should it?

One way around this is to say that data structures shouldn't have
methods at all.  This is the approach JavaScript takes - and there
is nothing at all wrong with it; it's just a different way of
seeing things.

I like using methods to operate on data structures.
It's convenient.  So I'm fond of Python's approach.


> I know it's just syntactic sugar, but it would be so sweet.
> Is there any reason why not to do it?

Just a design difference between JavaScript and Python, really.
In Python, attributes (and methods) are quite different in meaning
from dictionary keys.

Good question.

## Jason Orendorff    http://www.jorendorff.com/



From gcordova at hebmex.com  Tue Feb 19 11:37:46 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 19 Feb 2002 10:37:46 -0600
Subject: [Python-Dev] Rattlesnake progress
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AB1@SRM201002>

> 
> It's also *much* harder to model the cache hierarchy properly so that 
> you can make sure they'd fit in the l1 cache, than it is to make sure 
> they stay in registers where needed in the first place.
> 
> 
> Try taking a performance critical loop entirely in registers, 
> and change 
> it to save to and load from memory into a register on every iteration.
> See how much slower it gets.
> 
> --Dan
> 

You, Sir, are Absolutely Right.

(and we all know there's no absolutes, so he must be relatively
absolutely right).

This reminds me about programming a C64 in assembler. Having
only three 8-bit registers, with only one of them capable of
arithmetic operations, and the other two have their own weird
addressing schemes, makes me wonder about all those great
video games and animations.

LDA ($FE,X)

vs.

LDA ($FE),Y

anyone? :-)

PLUS, you'd have to take care about the clock delays from
crossing a page boundary, beware pushing or pulling too much
from the single-page stack, and all those "bad" cycles, and...

Heh, it was all an adventure.

-gus



From jason at jorendorff.com  Mon Feb  4 20:31:12 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 19:31:12 -0600
Subject: Backwards Compatibility of Python versions
In-Reply-To: <7xu1sx41cl.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEFKDAAA.jason@jorendorff.com>

Paul Rubin wrote:
> If I write in 2.2 or later 2.x's, I have to choose between two
> alternatives, both bad: [...]

I write int(x/y).

But note that this only works correctly for positive results,
which is mostly what I have.

## Jason Orendorff    http://www.jorendorff.com/





From tim.one at comcast.net  Thu Feb  7 22:47:33 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 07 Feb 2002 22:47:33 -0500
Subject: Please stop responding to Rue
In-Reply-To: <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGINKAA.tim.one@comcast.net>

[Donn Cave]
> So is sci.math choked with posts from people who normally seem
> to have their heads screwed on, endlessly replying this guy as
> if they sincerely think there's some purpose in it?  People who
> take time several times every day to post about how useless his
> ideas are and how they have no interest in helping?  Or are
> they saner there than we are?

On whole, no; neither in sci.physics, or even sci.logic (in fact, if you
think about it, sci.logic attracts more than its share for a reason, and on
both sides <wink>).

indulging-an-urge-to-cruelty-is-anti-pythonic-ly y'rs  - tim




From oren-py-l at hishome.net  Sat Feb  2 12:34:35 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Sat, 2 Feb 2002 19:34:35 +0200
Subject: Re-iterable generator comprehensions
Message-ID: <20020202193435.A7984@hishome.net>

On Mon, Apr 1, 2002 at 09:53:15AM -2600, J. Random Newbie wrote:
> I am using Python 2.3b1 and I am trying to use generator comprehensions 
> to generate all combinations of a letter and a digit without creating any 
> intermediate lists:
> 
> >>>from __future__ import generator_comprehensions
> >>>letters = [yield chr(i) for i in xrange(chr('a'),chr('z')+1)]
> >>>digits = [yield str(i) for i in xrange(10)]
> >>>letdig = [yield l+d for l in letters for d in digits]
>
> The generator letdig yields 'a0' through 'a9' and stops.  If I remove 
> the yield keywords and use lists it works fine. What's happening here?

According to the current proposal for generator comprehensions the letters
and digits objects are iterators. They can only be traversed once.  If you 
call iter(digits) it just returns itself.  When the letdig generator 
reaches the end of the digits source it calls iter(digits) again, expecting 
a fresh iterator for the same container.  Instead it just gets the same 
exhausted iterator and stops.

The solution is simple: generator comprehensions should not return an 
iterator - they should return an iterable object. This will make their 
behavior much more like real containers and make the code above work 
as expected.

According to the current proposal for generator comprehensions the 
letters object is equivalent to this code:

  def tmpgen():
    for i in xrange(chr('a'),chr('z')+1)]: yield chr(i)
  letters = tmpgen()

With re-iterable generators it would be something like this:

  def tmpgen():
    for i in xrange(chr('a'),chr('z')+1)]: yield chr(i)
  letters = ReGenerator(tmpgen)

where the class ReGenerator is defined as:

  class ReGenerator:
    def __init__(self, genfunc):
      self.__iter__ = genfunc

This object more closely emulates a list or other container: each time
iter(letters) is called it returns a fresh, independent iterator.  Unless
the generator or any of its sources are affected by some global variables 
it should produce the same stream of objects again and again.

The should also apply to the functions xmap, xfilter and xzip.  They 
should return an iterable object, not an iterator.  The xrange built-in 
function already works this way: it returns an iterable xrange object 
emulating a container.  Each time iter(<xrange object>) is called it 
returns a fresh, independent iterator.

   from-__future__-ly yours,

	Oren Tirosh




From news at chriswysocki.com  Sun Feb 17 23:04:10 2002
From: news at chriswysocki.com (Chris Wysocki)
Date: Mon, 18 Feb 2002 04:04:10 GMT
Subject: Python 2.2 <-> Solaris 2.6 installation errors
References: <mailman.1013616443.8156.python-list@python.org>
Message-ID: <1vs07ug1ttgnbuhsjaieclc68gkarh2h6b@4ax.com>

On Wed, 13 Feb 2002 09:52:31 -0600, Jeff Bauer <jbauer at rubic.com>
wrote:

>I think the problems Chris and I are having may be similar.
>
>Should I submit this as a build/installation bug?
>
>Jeff Bauer
>Rubicon Research

Jeff,

I checked the log you posted in sourceforge bug log and compared it to
mine.  You are clearly getting a lot further along the 'make install'
step than I am hence I must be doing something wrong.

Basically I disabled the exception handling around the module import
in the setup.py to get a more meaningful error message and got
something equivalent to the following:

> ImportError: ld.so.1: python: fatal: libgcc_s.so.1: open failed: No such file or directory

Didn't take me long to figure out my problem.  As my userid didn't
have authority to install files in /usr/local, I was running the 'make
install' with 'sudo'.  Unfortunately sudo was not recognising my
LD_LIBRARY_PATH instead it was using roots LD_LIBRARY_PATH which was
not being set to anything.  I guess this makes sense for sudo to do
this and it was my misunderstanding of how it worked that caused the
problem.

Anyway signed on as root, set the LD_LIBRARY_PATH and the make install
worked perfectly all the way through including the readline module
which seems to work fine.

Hence although our errors look the same I suspect that the cause is
different.

Something for you to try...
After you run the make, start up python interactively from your local
directory and then do 'import readline' and see if you get an error.
If you ran 'make install' then you probably need to run 'make' again
for this to be a valid test.  If you run 'make install' in a different
environment than 'make', like I was with 'sudo', then be sure to start
python in that environment.  If you get an error then that should give
you a clue where to go.

One possible scenario is that you might have a LD_LIBRARY_PATH entry
for readline that is not there when you run 'make install'.

Your posting encouraged me to sort out my own readline problems as
previously I was not able to compile readline.  I did have the
readline package from sunfreeware installed but I think it must have
had some include files missing which gave me my compile errors.
I got around the problem by uninstalling the readline package and
manually installing readline 4.2 from source.

Good luck,  Chris


From jkraska at san.rr.com  Thu Feb 21 22:35:46 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 22 Feb 2002 03:35:46 GMT
Subject: Embedding Python in C#
References: <3c75b88c$1_5@news1>
Message-ID: <jgfb7ucnlchr4qfueqkmq1lc5sd5msrroj@4ax.com>


>has this been done?

Not that I know of.

>is it even possible?

I am pretty sure, yes.

>if it has where can i obtain information on doing so?

Well, you'll have to look into the .dll interfaces for
.NET, and then likely do something like make a .NET
component for the Python interpreter and go from there.
I'm assuming you mean "embedding" in the classic style
of Python extending and embedding. IOW, you want to use
the Python interpreter from inside a C# component.

C//



From burt at dfki.de  Wed Feb 20 03:13:50 2002
From: burt at dfki.de (Alastair Burt)
Date: 20 Feb 2002 09:13:50 +0100
Subject: Python and GIS
References: <3C72F52F.5050901@sympatico.ca>
Message-ID: <vgy9hoa7up.fsf@dfki.uni-sb.de>

Ramkumar Kashyap <rkashyap at sympatico.ca> writes:

> I was wondering if there are any modules for GIS, Mapping, Layering, or
> hooks to Arcinfo, Mapinfo, etc.

Intevation[1] is a good first stop. The firm itself is responsible for
MapIt[2] and Bernhard Herzog of Intevation is the author of Sketch[3],
which has a plugin to handle ArcView shapefiles. In addition Intevation
maintain FreeGIS[4]. Searching for the string "Python" on this site will
give you more links.

--- Alastair


Footnotes:
[1] http://intevation.de/ 

[2] http://www.mapit.de/index.en.html

[3] http://sketch.sourceforge.net

[4] http://freegis.org/index.en.html


From caljason76 at yahoo.com  Tue Feb 19 17:24:43 2002
From: caljason76 at yahoo.com (Jason)
Date: 19 Feb 2002 14:24:43 -0800
Subject: This isn't recursive
Message-ID: <e9e6dd1f.0202191424.51a23ba8@posting.google.com>

# Ummm... I think the posted wanted a recursive solution.
#
# I am a Python novice so I don't really understand how things
# work, yet.  So some of this may either be wrong or not the
# Python way.  Also I have no idea why the 'if __name__ == ...'
# construct is needed instead of just the naked main statements.
#
# A few notes on the code.  There are two solutions: list_naked
# and list.
#
# list_naked is the simplies solution, it has no wrapper object
# and each node in list_naked is the head of a list.  You can see
# the recursion in list_naked.sum() and list_naked.__str_all(),
# where you will notice the if statements that test for the
# base case of the recursion.
#
# list_wrapped is a little more complex.  It has a wrapping object
# that is always used to refer to the list (unlike list_naked
# where the head of the list can change when a new value is pushed,
# hence the ll = ll.push(ll,x) assignment).  Here, the recursion
# is stopped by the_list_wrapped_empty_node and all the control
# flow is done by the interpreter.  I like this best, because I
# hate explicit control flow in programs and I think that it makes
# it easier to see the forest instead of the trees.  Looking at
# list_wrapped.sum() it is clear: add this node to the sum of the
# rest.  In list_naked.sum() you now have to look at two statements
# which do the same thing semantically.  With list_wrapped there
# is no way to forget to check for your base case because you don't
# check anything!  With just a list it seems a little too much,
# but when you start playing with various tree types and graphs
# it makes the code so much more readable.
#
# Also, can somebody help me find the documentation to the builtin
# procedures, like print().  I checked the __builting__ module docs
# but there is nothing there.  Thanks
#
# -j (nordwick at xcf.berkeley.edu)
#

class list_naked:
    def __init__(s, data, next = None):
        s.data, s.next = data, next
    def __str__(s):
        return "[%s]" % s.__str_all()
    def push(s, data):
        return list_naked(data, s)
    def sum(s):
        if s.next: return s.data + s.next.sum()
        else:      return s.data
    def __str_all(s):
        if s.next: return str(s.data) + " " + s.next.__str_all()
        else:      return str(s.data)

class list_wrapped_node:
    def __init__(s, data, next): s.data, s.next = data, next
    def __str__(s): return str(s.data) + " " + str(s.next)
    def sum(s): return s.data + s.next.sum()

class list_wrapped_empty_node(list_wrapped_node):
    def __init__(s): pass
    def __str__(s): return ""
    def sum(s): return 0

class list_wrapped:
    def __init__(s): s.head = list_wrapped_empty_node()
    def __str__(s): return "[" + str(s.head) + "]"
    def push(s, data): s.head = list_wrapped_node(data, s.head)
    def sum(s): return s.head.sum()
        
if __name__ == '__main__':
    ll = list_naked(0)
    for i in range(6): ll = ll.push(i+1)
    print ll
    print ll.sum()

    l = list_wrapped()
    for i in range(6): l.push(i)
    print l
    print l.sum()


From phr-n2002a at nightsong.com  Fri Feb  8 20:02:38 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 08 Feb 2002 17:02:38 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <e840346c.0202070321.452d405d@posting.google.com> <7xr8nxy0rb.fsf@ruckus.brouhaha.com> <slrna664ir.d87.philh@comuno.freeserve.co.uk>
Message-ID: <7xeljv79ht.fsf@ruckus.brouhaha.com>

philh at comuno.freeserve.co.uk (phil hunt) writes:
> >  - Program stays around for years but developers come and go.
> >    Average turnover is maybe 1 programmer per month.  So there
> >    are always people hacking on the codebase who aren't that familiar
> >    with it.
> 
> The resultant architecture will be a Big Ball Of Mud.

That's unfortunate but completely true.  Welcome to the real world.  
The question is whether Java programs survive turning into mud better
than Python programs do.


From pan-news at asheeshenterprises.com  Wed Feb 27 23:57:07 2002
From: pan-news at asheeshenterprises.com (Asheesh Laroia)
Date: Thu, 28 Feb 2002 04:57:07 GMT
Subject: Windows GUI - which way?
References: <a5k8t9$hrp$1@news.efn.org>
Message-ID: <DMif8.149266$dh.38838329@typhoon.nyroc.rr.com>

You can use the Win32API in Python.

See www.activestate.com for details.

It would make the most sense, then, to do the whole thing in Python,
using the Win32 API.

Pros: 
	Uses info you already know
	Easy to integreate Python with Python

Cons:
	GUI not as abstracted and cross-platform as it would be with wxWindows
or Tkinter or another cross-platform GUI toolkit.

-- Asheesh Laroia.

On Wed, 27 Feb 2002 22:45:21 -0500, Spencer Doidge wrote:

> I have a nice script running and doing useful work--in the Windows
> console mode.
> It sends RS232 commands and receives replies, punctuated by various time
> delays.
> I parked the script at http://www.spencerdoidge.com/tp1.py Kudos and
> credits to Isaac Barona for his contribution, the SerialPort class,
> which I modified slightly to suit my needs. The script tabulates and
> reports the results and writes a log.
> 
> I'd like to make this palatable for others around the company to use,
> and that means Windows GUI.
> The user would select the script from a menu, or type the filename in a
> dialog, or use File/Open, or...
> 
> (Forgive me, please if I miss the mark with my questions--I'm a bit
> fried after burying myself in Python the last couple of weeks. I'm just
> coming up for air...)
> I'm looking for suggestions and insights on which of the following
> approaches might be best:
> 
> So should I write up a program in my favorite style, Win32 API, and
> embed Python in it?
> Or should I spawn a child process and run my good old console version in
> the child process? (Not sure how I would be able to interrupt the child
> process though.)
> Or should I write the whole thing in Python?
> 
> Would somebody like to comment on the pros and cons?
> 
> Spencer Doidge


From python at chajadan.net  Mon Feb 18 03:41:20 2002
From: python at chajadan.net (chajadan)
Date: Mon, 18 Feb 2002 00:41:20 -0800
Subject: help - inserting values into database tables
In-Reply-To: <7xpu33myrc.fsf@ruckus.brouhaha.com>
References: <mailman.1013981787.10266.python-list@python.org>
Message-ID: <5.1.0.14.0.20020218003729.00a54b30@63.118.78.184>

My apologies, odbc is part of the win32all distribution, and is therefore 
available only under Windows.

However, the module you found, or other similar modules, are bound to be a 
viable solution for you, database interaction being so widespread as it is.

I know of another possible solution, by the creators of mxTime and whatnot, 
available in this general location: 
http://www.lemburg.com/files/python/eGenix-mx-Extensions.html

--chajadan

At 04:15 PM 2/17/02 -0800, Paul Rubin wrote:
>chajadan <python at chajadan.net> writes:
>
> > Yes, this is altogether possible.
> >
> > There is a module include in the main distribution called odbc, and
> > one called dbi.
>
>Whaaa?  That's the first I've heard of this.  In Python 2.2 under Linux,
>"import odbc" and "import dbi" both throw ImportError.
>
>There's an ODBC module in Vaults of Parnassus that looks pretty good, but 
>I haven't
>tried it yet.  Is that what you meant?
>--
>http://mail.python.org/mailman/listinfo/python-list




From hmartires at cultalg.pt  Mon Feb 25 08:31:59 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 13:31:59 -0000
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de>
Message-ID: <s%qe8.598$Zj1.7273540@newsserver.ip.pt>

And how can i count in a list ?
It is possible ?

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4bsedlnai.fsf at informatik.hu-berlin.de...
> "Hugo Martires" <hmartires at cultalg.pt> writes:
>
> > How can i count the number of bytes of:
> > --> List []
> > --> array('d',[])
> >
> > There is any function that return's the number of bytes ?
>
> For the list, no. For the array: len(arr.tostring()) will give a rough
> estimate. The actual memory consumption is higher, due to Python and
> malloc overheads.
>
> Regards,
> Martin
>
>




From rjones at ekit-inc.com  Thu Feb  7 18:56:26 2002
From: rjones at ekit-inc.com (Richard Jones)
Date: Fri, 8 Feb 2002 10:56:26 +1100
Subject: Anyone got diaries of the conference online?
Message-ID: <200202072356.XAA16278@crown.off.ekorp.com>

Has anyone got some good info online from the conference yet? I've read the 
zopezen log, but it's a bit scant on the info :)

/me wants a fix ;)


   Richard



From mike_lg_t at yahoo.com  Sun Feb 24 22:54:23 2002
From: mike_lg_t at yahoo.com (mike)
Date: Sun, 24 Feb 2002 22:54:23 -0500
Subject: exception popen2
Message-ID: <gzie8.1422$it5.519280@news20.bellglobal.com>

Hi:

I am trying to catch the exception from popen2

try:
  child_in.child_out=os.popen2("python dth.py")
except OSError:
  print "can not open file"

dth.py does not exist
but when i try to run this script. the message "can not open file" does not
display.
which exception I should catch?


best Regards

zhengyi




From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 20:52:19 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 19:52:19 -0600
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org>
Message-ID: <00a701c1ad1e$96340420$0101010a@local>

----- Original Message -----
From: "Spencer Doidge" <spencer at spencerdoidge.com>


> I really tried to RTM first this time, but I failed.
> Can someone give me a hint where in the doc html's it explains how to pass
> by reference?
> Specifically, I want to do the equivalent of this:

Briefly, you don't.  Or you do.  Mutable types (lists, dictionaries,
objects)
can be changed as you request, but immutable types (int, float, string)
can't.

Since Python supports returning multiple values (via tuples or lists) you
normally don't need to do this.

If you have a C function like this:

    int MyCFunction(char *buffer);

the Python equivalent would be called like this:

    intval, stringval = MyCFunctionWrapped()

or if the buffer is both input and output:

    intval, stringval = MyCFunctionWrapped(stringval)






From guy at wyrdrune.com  Sun Feb  3 23:49:32 2002
From: guy at wyrdrune.com (Guy Gascoigne - Piggford)
Date: Sun, 3 Feb 2002 20:49:32 -0800
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <mailman.1012745045.21128.python-list@python.org> <3c5d7e28_2@news.nwlink.com> <1012765918.861234@yasure>
Message-ID: <3c5e12c4_1@news.nwlink.com>

What I really want is some way to get anonymous functions that can be passed
as part of a list of paramaters and called at the appropriate time.  There
doesn't appear to be any way to get that particular effect in Python.

As for expect like functions in java - luckily I don't need all of the
features of expect, more a case of simple telnet scripting. Between our own
telnet code and pattern matching we've actually got something that works
pretty well so far.

BTW I copied my original question to the end of the message in case it rang
any bells.

Guy

"Donn Cave" <donn at drizzle.com> wrote in message
news:1012765918.861234 at yasure...
> Quoth "Guy Gascoigne - Piggford" <guy at wyrdrune.com>:
> Which I missed, but when I had thoughts like that once upon a time,
> the best I could think of was to use the exception mechanism.  Or
> abuse it, if you like.  Make pattern objects, feed them to the expect
> parser in a try block, which would raise it as an exception to dispatch
> to the appropriate except block.  Horrible.
>
> But you must have support for pseudotty devices, which is a low level
> functionality that is not so portable.  To my knowledge available only
> on UNIX and similar platforms, and not on all such for Python.  Have
> that in Java?  I don't know, but if not, you're sunk.
>
> Donn Cave, donn at drizzle.com

> Quoth "Guy Gascoigne - Piggford" <guy at wyrdrune.com>:
>I'm trying to find a way to write something like expect in a python
>module, and I'm having trouble getting the syntax the way that I want.
>In expect I can write something like this:
>
>expect {
>    -re "ast login" {
>           pass
>           }
>    -re "ogin" {
>           send  user
>           }
>    -re "sword" {
>           send password
>           }
>}
>
>etc.
>
>The nearest that I've managed to get to in Python is something like
>this:
>
>        while 1:
>            try:
>                if self.expect( ( "ast login", 0 ),
>                                ( "ogin", self.send_user ),
>                                ( "sword", self.send_password ),
>                                ( "\$", "done" ) ) == "done":
>                    break
>            except IndexError:
>                skipcount += 1
>                if skipcount == 2:
>                    raise IndexError
>
>Where send_user and send_password are separately defined functions.
>What I really want to be able to do is to define the action beside the
>pattern, calling short functions that are defined separately makes it
>harder to read and trickier to follow exactly what is happening.  I've
>also played around with just returning an index, but that also gets
>ugly as the list of possible patterns gets larger.
>
>Am I missing something?  I'm fairly new to python but I don't see a
>way to get what I want.  Does anyone have any suggestions on how to
>handle this more cleanly?
>





From wtanksley at bigfoot.com  Wed Feb 20 21:35:29 2002
From: wtanksley at bigfoot.com (William Tanksley Google)
Date: 20 Feb 2002 18:35:29 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com> <slrna784si.m30.akuchlin@dust.mems-exchange.org>
Message-ID: <de3fc1ef.0202201835.7f49f4be@posting.google.com>

"A.M. Kuchling" <akuchlin at dust.mems-exchange.org> wrote:
>    William Tanksley Google wrote:
> > My friend was griping about one battery that Basic comes with but
> > Python doesn't (at least not that I've seen): graphics.  Yes, I know

> turtle.py is cute, and requires Tkinter:

Good suggestion.  Where is it?

> When you say you'd like pixel graphics, what do you mean?  Just
> setting elements of a 2D pixel array to various colours?  Maybe
> something for that could be built on top of Tkinter, too.

I'm talking about being able to set pixels on the screen :-).  I don't
really care about the interface, as long as it's not horribly hokey. 
After all, we're talking about people who like BASIC because it
supports graphics.  We're not talking about high aesthetics here ;-). 
Seriously, an interface like QBasic's would be sufficient -- a
function named "screen" which pulls up a 'canvas', and a function
named 'pset' which sets points on it.  Bonus points for implementing
flood fill, lines, ellipses, and so on.  Extra bonus points for
designing an API which is adopted into the next version of Python, and
is thereby available on all reasonable platforms (and eventually
ported to DOS ;-).

I'm not 100% joking about that last part -- at least, it would be nice
to be able to use full-screen graphics modes where available.

> --amk                                                  (www.amk.ca)

-Billy


From skip at pobox.com  Thu Feb 14 08:45:16 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 14 Feb 2002 07:45:16 -0600
Subject: Disabling creation of .pyc files?
In-Reply-To: <Xns91B51A11D5957RASXnewsDFE1@130.133.1.4>
References: <E16b1mB-0001Ls-00@mail.python.org>
        <mailman.1013619835.26532.python-list@python.org>
        <Xns91B51A11D5957RASXnewsDFE1@130.133.1.4>
Message-ID: <15467.49004.717503.436404@12-248-41-177.client.attbi.com>

    >> It's there for your benefit (imports are faster when up-to-date .pyc
    >> files are present because the compilation step is avoided), but yes,
    >> you can disable it by simply making the directory containing the .py
    >> files read-only.

    Philip> Does this work under windows? I just set a folder read-only (it
    Philip> is an option, but FIAK the 'properties' window might lie or
    Philip> cheat regarging the underling attrib calls, but...), and it
    Philip> didn't seem to prevent me from adding files or changing names...

Can I finesse the issue and just say, "get a real OS"? ;-) Sorry, I don't
recall the original posted indicating the operating system.  On
Unix-oriented systems, making a directory read-only is sufficient.

Skip



From sdm7g at Virginia.EDU  Wed Feb 13 13:27:23 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Wed, 13 Feb 2002 13:27:23 -0500 (EST)
Subject: Python Gotcha with Octal Numbers
In-Reply-To: <yuqa8oHpvCEW092yn@visi.com>
Message-ID: <Pine.OSX.4.43.0202131324440.640-100000@d-128-61-180.bootp.virginia.edu>


On Wed, 13 Feb 2002, Richard Steiner wrote:

> Here in alt.folklore.computers,
> claird at starbase.neosoft.com (Cameron Laird) spake unto us, saying:
>
> >In article <130220020751360934%jwbaxter at spamcop.net>,
> >John W. Baxter <jwbaxter at spamcop.net> wrote:
> >			.
> >>Did the elves at AT&T pick up the "leading 0 means octal" from
> >>something earlier, or did they invent this stupidity themselves?
> >			.
> >OK, folks, how far back *can* we trace this?  My
> >mind associates it with DEC systems, going back
> >to the '60s, but I couldn't find any confirmation
> >of that in a quick search.
>
> I've seen the leading-zero-means-octal convention used rather heavily
> in documentation dated 1967-1968 in a UNIVAC 1106/1108 context.
>
> It seems to have been a fairly well-established convention at that
> point in time, at least in the UNIVAC world.
>

I'm betting that it's something that C (or B or BCPL) picked up
from some macro assembler syntax -- unfortunately, I finally
threw out all my VAX, PDP-11 and PDP-8 manuals last year -- I
had to clean out a back room that another lab is going to take
over.


-- Steve "the reformed packrat" Majewski




From gbreed at cix.compulink.co.uk  Tue Feb 19 05:57:59 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Tue, 19 Feb 2002 10:57:59 +0000 (UTC)
Subject: Adding static typing to Python
References: <ur047u8lrm308jbivi5h8g7ld4dugfb5i7@4ax.com>
Message-ID: <a4tb3n$pc0$1@thorium.cix.co.uk>

Courageous wrote:

> Well, one has been able to roll one's own version of this for a long
> time; simply override getattr and settatr and on a per instance basis
> do what you need to do. Basically, it's all syntactic sugar. For the
> C++ "family" of language, it's quite an innovation, however, as this
> is the first time one has been able to use an attribute but get a method
> as a consequence.
> 
> I don't know if 2.2 offered anything like this. It's not obvious in
> the documenation. The docs are in sorry shape, though.

I haven't worked out 2.2 either.  But Jython does present Java's getters 
and setters as attributes.  See 
<http://www.jython.org/docs/properties.html>.


                          Graham


From bosahv at netscapenospam.net  Mon Feb  4 03:12:23 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Mon, 04 Feb 2002 08:12:23 GMT
Subject: Windows CE vs PalmOS version
References: <mailman.1012697474.21632.python-list@python.org>
Message-ID: <Hnr78.316307$pa1.77732823@news3.rdc1.on.home.com>

At the small risk of being unpopular here I would suggest you really
consider whether you need to write the app in python.

You might want to consider mature developement packages for the palm if this
is a professional project. Running an interpreter on a compartively weak
processor is a lot of overhead. While you might link into a python program
on the base computer, manipulate the data etc... I would suggest you run as
close to hardware as you can on the pda. Plan it to run on the cheapest
palm.


I could be wrong of course :) but I think the python implemenations on palm
and winCE are a little young.

I hope some people have concrete experience to prove me wrong ??? :) ?!!!

Bo



"Lance Ellinghaus" <lellinghaus at yahoo.com> wrote in message
news:mailman.1012697474.21632.python-list at python.org...
> Hello everyone...
> I need to write a program in Python that requires a GUI and to run on a
> hand held device. This would be very easy if it was a desktop
> application, but it is not.
>
> So the question is:
>     Should I purchase a Windows CE or Palm compatible device and run
> Python on it?
>
>     What GUI interfaces are available for the Windows CE environment?
>
> Thank you!
>
> Lance Ellinghaus
>
>
> =====
> --
> Lance Ellinghaus
>
> __________________________________________________
> Do You Yahoo!?
> Great stuff seeking new owners in Yahoo! Auctions!
> http://auctions.yahoo.com
>




From tim.one at home.com  Sun Feb 17 03:21:55 2002
From: tim.one at home.com (Tim Peters)
Date: Sun, 17 Feb 2002 03:21:55 -0500
Subject: Using tabs instead of spaces in IDLE?
In-Reply-To: <iKOa8.102499$A51.43433850@typhoon.southeast.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEAINNAA.tim.one@home.com>

[Tom Bryan, on fiddling the the Tk text widgets default tab width]
> I did.  I changed it to 4 and 4.  I toggled tabs on in Idle.  It
> seemed to work on my Linux box.

Are you sure?  Here's what I see, laboriously replacing tabs with spaces by
hand so you'll see the same thing, and using Courier New (a fixed-width
font, just to simplify life).  It's not the initial whitespace that screws
you, so it's quite likely you don't notice at first.  Everywhere you see a
run of whitespace (one or more blanks) here, I hit TAB once followed by the
letter T:

123456789012345678901
    T   T   T   T   T
a   T   T   T   T   T
ab  T   T   T   T   T
abc T   T   T   T   T
abcd T  T   T   T   T
abcde T T   T   T   T
abcdef T T  T   T   T
abcdefg T T T   T   T

No programmer expects that.  Tk "wants to" put the character following the
first hard tab into column 5, no matter how far to the right of column 5 you
are when you hit the tab key for the first time.  Similarly it "wants to"
put the character following a line's second hard tab in column 9, and again
irrespective of where on the line you are when you hit tab the second time.

What programmers expect is:

123456789012345678901
    T   T   T   T   T
a   T   T   T   T   T
ab  T   T   T   T   T
abc T   T   T   T   T
abcd    T   T   T   T   T
abcde   T   T   T   T   T
abcdef  T   T   T   T   T
abcdefg T   T   T   T   T

It's incredibly jarring when you delete characters "on the left side" of
lines containing hard tabs after changing the Tk default, as the widget goes
thru amazing hoops trying to keep seemingly random portions of the remaining
line at fixed distances from the left margin.

> Ick.  So the tabs will depend on the font, etc.?

Not really; IDLE tries to compute the "average width" of a character in the
font then, but that's just to fool the Tk text widget; see the latter's docs
at, e.g.,

    http://tcl.activestate.com/man/tcl8.4/TkCmd/text.htm

You'll find that the only way to specify non-default tabs under the covers
is to give a list of absolute distances from the left margin (in units of
centimeters or inches or points, etc, but not in units of "characters").

> Hm.  PythonWin might be a good option.  The UNIX heads will probably just
> use vi and Emacs, which I think we can handle.  The current "approved"
> coding style for all languages in the group is supposed to be "always
> indent with leading tabs; for alignment purposes, tabs should be
> viewed as 4 spaces; set your visual display of tabs to whatever you
> prefer."

Well, everyone is entitled to live with the consequences of their
self-imposed insanities <wink>.

> I would, but then we've got spaces in some languages and tabs in others.

Most (all?) programmer's editors can be taught to use different rules based
on file extension.  The standard for the Python library is 4-space indents
and no hard tab characters anywhere.  That's what the Python developers have
learned works best.

> ...
> It's stupid, but the whitespace issue is probably going to be one of the
> biggest obtacles to introducing Jython to my co-workers.  I'll be back to
> tell everyone how it goes.

So don't tell them.  Use Jython yourself, and leave them in the dust.
They'll either beg to learn your secret, or you'll be promoted so high so
fast you won't have to endure them anymore <wink>.

let-'em-stew-in-their-own-tabs-ly y'rs  - tim




From James_Althoff at i2.com  Fri Feb  8 17:20:12 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 8 Feb 2002 14:20:12 -0800
Subject: range(seq)?
Message-ID: <OFE70192F6.273F8C47-ON88256B5A.00788B4A@i2.com>

[Magnus Lie Hetland]
>   for x in range(seq): ...
>
> I.e. let range() take a sequence (or iterable) as an argument, and
> generate range(len(seq)).

[Jeff Shannon]
>This seems more logical and intuitive to me, than the previous
>suggestions.

What Jython users will appreciate is that the above suggestion -- although
it might be handy for Pythonic structures and, possibly, might be "more
logical and intuitive" than other suggestions -- doesn't necessarily help
with the world of structures one encounters outside the arena of 100%
Python.  Java/Swing tables, for example, cannot be passed to the Python len
() function (they are two-dimensional, for starters, and are not defined as
a sequence of sequences).  So such an extension to the range() function
won't make it any easier to deal with (certain) Java structures in Jython.
PEP 276, for example, suggests a simple way to deal with index-based
structures that have been designed outside the sphere of Python practices
while also making it easier at times to deal with Pythonic designs.

Jim




From maxm at mxm.dk  Fri Feb 15 11:49:11 2002
From: maxm at mxm.dk (Max M)
Date: Fri, 15 Feb 2002 16:49:11 +0000
Subject: Using Python to generate MIDI files (Mac)
References: <5e8bd451.0202150813.20720fc5@posting.google.com>
Message-ID: <3C6D3C07.2020805@mxm.dk>

kevin parks wrote:


> Anyone have a working module to generate MIDI files? with Python? I
> have seen this asked before, but not really answered. I know that
> there is a module to read MIDI files, and a few shakey attempts to
> send MIDI data in real-time out a port, but I specifically want to
> generate a midi file that i can then further mangle in other programs.

Would you be interrested in a half-hearted attempt??

The problem is that there is the midi filespec, and then there is a 
defacto standard where some elements _must_ be present for a programme 
to read it as a standard midi file.

At some time I WILL get around to finishing it, but as it is a hobby 
project all kind of customer stuff seems to get in the way :-(

regards Max M

Well here is a starter that sort-of works:

def byteString(string):
     return [ord(letter) for letter in string]

def word(value):
     # Big endian 2 byte word
     # returns a list with 2 bytes holding the value
     lsb = value & 0xFF
     msb = value & 0xFF00
     msb = msb >> 8
     return [msb, lsb]

def byte3(value):
     # Big endian 3 byte word
     b1 = value & 0xFF
     b2 = value & 0xFF00
     b3 = value & 0xFF0000
     b2 = b2 >> 8
     b3 = b3 >> 16
     return [b3, b2, b1]

def long(value):
     # Big endian 4 byte word
     b1 = value & 0xFF
     b2 = value & 0xFF00
     b3 = value & 0xFF0000
     b4 = value & 0xFF000000
     b2 = b2 >> 8
     b3 = b3 >> 16
     b4 = b4 >> 24
     return [b4, b3, b2, b1]

def varLen(value):
     # Big endian variable byte word
     # as defined by the midi standard
     result = []
     while value:
         result.append((value & 0x7F) | 0x80)
         value = value >> 7
     if len(result) == 0:
         return [0]
     else:
         result[len(result)-1] -= 128
         result.reverse()
         return result

class midiFile:

     # a midi file to which tracks can be added

     def __init__(self, fileName):
         self.fileName = fileName
         self.file = open(self.fileName, 'w')
         self.tracks = []

     def addTrack(self, item):
         self.tracks.append(item.get())

     def _writeBytes(self, vals):
         # writes a list of values 0-255 to bytes
         bytes = [chr(i) for i in vals]
         for byte in bytes:
             self.file.write(byte)

     def save(self):
         self._writeBytes(byteString('MThd') + long(6) + word(1) + \
                 word(len(self.tracks)) + word(96))
         for track in self.tracks:
             self._writeBytes(track)
         self.file.close()


class track:

     # a Midi Track

     def __init__(self):
         self.events = []

     def addEvent(self, item):
         self.events += item

     def get(self):
         return byteString('MTrk') + long(len(self.events)) + self.events


class events:

     def note_on(self, time, channel, note, velocity):
         return varLen(time) + [0x90 + channel, note, velocity]

     def note_off(self, time, channel, note, velocity):
         return varLen(time) + [0x80 + channel, note, velocity]

     def tempo(self, time=0, bpm=120):
         tempo = int(1000000/(bpm/60.0))
         return varLen(time) + [0xFF, 0x51, 0x03] + byte3(tempo)

     def signature(self, time, nn=4, dd=2):
         # dd => 2^-dd, dd = 1/8 = 2
         return varLen(time) + [0xFF, 0x58, 0x04, nn, dd, 24, 8]

     def text(self, time, value):
         return varLen(time) + [0xFF, 0x01] + byte3(len(value)) +\
                 byteString(value)

     def seqTrackName(self, time, value):
         return varLen(time) + [0xFF, 0x03] + byte3(len(value)) + \
                 byteString(value)

     def lyric(self, time, value):
         return varLen(time) + [0xFF, 0x05] + byte3(len(value)) + \
                 byteString(value)

     def endOfTrack(self, time):
         return varLen(time) + [0xFF, 0x2F, 0x00]

if __name__ == '__main__':

     # This is a library, so test code goes here
     evt = events()

     track1 = track()
     track1.addEvent(evt.tempo(0, bpm=120))
     track1.addEvent(evt.signature(0, 4, 2))
     #track1.addEvent(evt.text(0, 'Test song'))
     track2.addEvent(evt.endOfTrack(0))

##    track2 = track()
##    track2.addEvent(evt.note_on(0, 0, 60, 127))
##    track2.addEvent(evt.note_off(96, 0, 60, 127))
##    track2.addEvent(evt.endOfTrack(96))

     fileName = 'C:/root/pythonScripts/mxm_MIDI/midiOut.mid'
     myMidi = midiFile(fileName)
     myMidi.addTrack(track1)
##    myMidi.addTrack(track2)
     myMidi.save()




From phd at phd.pp.ru  Wed Feb 13 03:59:09 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 13 Feb 2002 11:59:09 +0300
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <15465.43095.985391.388092@beluga.mojam.com>; from skip@pobox.com on Tue, Feb 12, 2002 at 05:42:15PM -0600
References: <15465.43095.985391.388092@beluga.mojam.com>
Message-ID: <20020213115909.D20054@phd.pp.ru>

On Tue, Feb 12, 2002 at 05:42:15PM -0600, Skip Montanaro wrote:
> been lightly tested), but I've never seen a segfault take files with it like
> this.  Has anybody else encountered this before?

   Looks like a hrdware error. Memory error, or disk error. For example I
have a number of Pentium computers with VIA Apollo VP3 Pro chips, and on
most of them Linux upon loading disabled DMA to disks. Once I tried to
force DMA (using hdparam) and in a minuite was forced to reboot and run
fsck - files were started to be corrupted all over the disk.

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



From bokr at oz.net  Sat Feb 23 23:01:18 2002
From: bokr at oz.net (Bengt Richter)
Date: 24 Feb 2002 04:01:18 GMT
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <a59oie$55s$0@216.39.172.122>

On Sat, 23 Feb 2002 15:34:01 +0700, Tripp Scott <tripps81 at yahoo.com> wrote:

>Suppose I want to make a function that modifies its argument:
>
>  a = -1
>  make_abs(a)
>  print a # = 1
>
>What is the Pythonic way of doing this? Wrap it in a list?
>
>  make_abs([a])
>
Well, if you really want to[**], I think you might want to use
the list wrapper a little differently:

 >>> a = [-1]
 >>> def make_abs(x):
 ...     x[0] = abs(x[0])
 ...
 >>> a
 [-1]
 >>> make_abs(a)
 >>> a
 [1]

but be careful how you bind to mutable things, as all bindings
to the same thing will show a change to the thing bound to:

 >>> a = b = [-1]
 >>> a
 [-1]
 >>> b
 [-1]
 >>> make_abs(a)
 >>> a
 [1]
 >>> b
 [1]

note the different result here:

 >>> a = [-1]
 >>> b = [-1]
 >>> a
 [-1]
 >>> b
 [-1]
 >>> make_abs(a)
 >>> a
 [1]
 >>> b
 [-1]

You can also define your own mutable object with a class,
as mentioned elsewhere.

USING A CLASS IS PROBABLY A BETTER IDEA <<<< [**]notice ;-)

a=A(-1) and a.v = -1 or a.setval(-1) are probably less misleading than
a[0] = -1 or a = [-1].

Though it might be nice Python had an object update operator, so you
wouldn't have to know method or attribute names to do a default
state update operation. E.g. (using a ':=' token for the operator),

a := -1  #read as 'a updated with -1'

could translate to a.__update__(-1), which you could define
or override as desired. Incidentally, this could be an expression
evaluating to a reference to the updated object by returning self
from the method, so it could be used flexibly.

Though it introduces ambiguities, I sometimes think I'd like
to be able to define a __getval__ method that would be called
by a simple reference to the object in an expression. E.g.,

x = y

might become x = y.__getval__() so instead of x binding to the
y object instance, it would get whatever y was defined to deliver
as a default value. It would mean that you couldn't bind to y
as an object, unless __getval__() returned self, or you provided
another method to do that, as in z = y.it_self() or such.

With __getval__, x.y and x .y would probably mean different things.
I.e., x.__getattr__('y') vs x.__getval__().__getattr__('y')

It could obviously be abused, but could be useful too.

Regards,
Bengt Richter



From mwh at python.net  Thu Feb  7 06:37:33 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 11:37:33 GMT
Subject: for item in (pydoc, docs, lib): question_regarding(item)
References: <3c615b18.47723171@news.t-online.de>
Message-ID: <uofj1a5bj.fsf@python.net>

gerson.kurz at t-online.de (Gerson Kurz) writes:

[snip pydoc woes, as I don't know much about that]

> 2) There are some minor things I would like to see changed in the
> official python documentation. e.g., the example in section "6.4 stat
> -- Interpreting stat() results" uses the line
> 
> pathname = '%s/%s' % (dir, f)
> 
> which IMHO should read
> 
> pathname = os.path.join(dir, f)
> 
> I think that IF you at last give in and RTFM, it should at least
> silently guide you to the python-way of doing things. Now, where to
> submit things like these?

I don't know which version of the docs you were looking at, but the
ones at

    http://python.sourceforge.net/devel-docs/lib/lib.html

have, at the bottom of every page, a line:

    See About this document... for information on suggesting changes.

Clicking the link tells you to email python-docs at python.org or submit
bugs to sourceforge (or patches, I guess, if you speak TeX).

> And saving the best for last:
> 
> 3) Is there something like a TODO-list for "The Python Standard
> Library"?

Don't think so.

> And how to contribute?

    http://www.python.org/patches/

> I'm thinking of tiny things like a hexdump function

What do you mean here?  I suspect it probably already exists, though
perhaps not with the most obvious name...

> or administrative functions (like "granting/denying a user access to
> a file on NT") and things like that. [If you speak out regarding "OS
> dependancy", remember "sunau" and those IRIX modules].

Use win32all, surely?  The IRIX and sunau modules date from a time
lost in the mists of creation, and I don't think they can be used as
an excuse to wedge more platform dependence into the stdlib.

> I'd also like to add bigger things like a stripped down generic
>SMTP/POP3 server.

There's smtpd, already -- maybe you could look at that and improve it
if it doesn't match your needs.  A pop3d sounds like a good idea to
me, at least.

> I know we had threads on this before and they never got anywhere,
> but hey this time it could all be totally different :)

Did we?  Must have killed them...

Cheers,
M.

-- 
  Ability to type on a computer terminal is no guarantee of sanity,
  intelligence, or common sense. 
                                 -- Gene Spafford's Axiom #2 of Usenet


From lac at strakt.com  Wed Feb 13 12:09:59 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 13 Feb 2002 18:09:59 +0100
Subject: newbie asking for info about Python 
In-Reply-To: Message from "Joseph Youssef" <leader730@hotmail.com> 
   of "Tue, 12 Feb 2002 21:21:25 EST." <xkka8.2932$VX6.245554@weber.videotron.net> 
References: <xkka8.2932$VX6.245554@weber.videotron.net> 
Message-ID: <200202131709.g1DH9xGo032077@ratthing-b246.strakt.com>

Can you please not send HTML mail out to this list.  It is very hard
on those of us that read this not as a newsgroup but as a digestified
mailing list.  

Thanks very much,
(posted not mailed since there was more than one of you)

Laura Creighton



From phr-n2002a at nightsong.com  Mon Feb 11 19:40:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 16:40:51 -0800
Subject: Problem building Python 2.2 docs
Message-ID: <7xeljrcz1o.fsf@ruckus.brouhaha.com>

I just loaded a fresh Red Hat GNU/Linux 7.2 system on a newly
formatted hard disc, and then installed Python 2.2 on it from the .tgz
file.  Python 2.2 itself built and installed just fine.  When I typed
"make" in the Doc directory though, a bunch of errors resulted, like:

     *** processing declarations ***

    Loading /usr/share/latex2html/versions/latin1.pl

     *** Fatal Error --- but easy to fix ***

    Cannot have '.' in file-name prefix, else dvips fails on images
    Change the name from  /home/phr/Python-2.2/Doc/lib/lib.tex  and try again.

    *** Session transcript and error messages are in /home/phr/Python-2.2/Doc/html/lib/lib.how.
    *** Exited with status 255.

I don't remember having this problem with Python 2.1.1 under Red Hat 7.1.

Is this a known bug, in either Python 2.2 or Red Hat 7.2?

Thanks.


From phr-n2002a at nightsong.com  Thu Feb 21 14:28:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 11:28:36 -0800
Subject: Recursively traverse linked list -- help!
References: <mailman.1014318148.16546.python-list@python.org>
Message-ID: <7xu1say6qj.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> Here is how it might work.  Python could create a new frame.
> The optimization is that the frame performing the tail call can be
> culled right away, saving some memory and shortening the stack.
> The culled frame would disappear from traceback information too,
> unfortunately.
> 
> I'm sure this has been discussed endlessly before.

I don't think it's important for Python to do tail call optimization,
as long as users know better than to code in tail recursive style.
Only some Scheme-heads actually think that way most of the time, and
they generally understand the issues well enough to not do it in Python.

I do sometimes think about how to interpret Python using a Scheme
evaluator, but the answers aren't pretty.


From janez.jere at void.si  Mon Feb 25 06:28:02 2002
From: janez.jere at void.si (Janez Jere)
Date: Mon, 25 Feb 2002 12:28:02 +0100
Subject: ipc pickle
Message-ID: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com>

Hi, I am experimenting with IPC, my current implementation is very
slow, so I am asking for some hints, how to optimize it. 
Currently I know that cStringIO would cut execution time for 30%. but
I don't want to drop unicode support

Thanks in advance, Janez

import cPickle
import StringIO
#unicode required
#import cStringIO as StringIO
import struct
import os

class Client:
    def __init__(self, file):
        self.file = file
    def read(self):
        # read length and load object
        x = os.read(self.file, 4)
        length = struct.unpack('i', x)[0]
        buf = os.read(self.file, length)
        f = StringIO.StringIO(buf)
        return cPickle.load(f)
        
class Server:
    def __init__(self, file):
        self.file = file

    def write(self, obj):
        # write length of serialozed object, write serialized object
        f = StringIO.StringIO()
        cPickle.dump(obj, f, 1)
        val = f.getvalue()
        length = len(val)
        plen = struct.pack("i", length)
        os.write(self.file, plen)
        os.write(self.file,val)

if __name__=='__main__':
    rp, wp = os.pipe()
    c = Client(rp)
    s = Server(wp)
    s.write(u'\u0161\u010durek')
    print repr(c.read())

    import time
    start = time.clock()
    for i in xrange(3000):
        s.write('a')
        c.read()
    print time.clock() - start




From thomas at gatsoft.no  Wed Feb 27 08:52:15 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Wed, 27 Feb 2002 13:52:15 GMT
Subject: SSL Server Sockets?
References: <a5ijos$1565$1@news.idiom.com>
Message-ID: <jw5f8.129$5vh.170404352@news.telia.no>

I got the same problem. Wish there were more precompiled binary packages of
M2Crypto. Why can't Python be shipped with SSL-support compiled into it?
Could M2Crypto be included in the distro? If it's dependant on OpenSSL
that's open source too, so why not join forces and create a real
security-focused programming-package?

The ability to compile from source is great, but when it has a lot of
dependencies it would be easier to include the package in the standard
distro so people don't have to hassle with it.

To make life alot easier for many of us, either :
a) get more packages into the standard distro
or
b) offer some precompiled and/or easy to install add-on package containing
the most widely used modules ( PIL, M2Crypto, BSDDB 3.x-wrappers,
StandaloneZODB, PyXML ).

Just my two cents.

Thomas Weholt

"Jim Dennis" <jimd at vega.starshine.org> wrote in message
news:a5ijos$1565$1 at news.idiom.com...
> Is the native Python2.2 socket.ssl support still client only,
>  or is it possible to use it for simple ssl servers, yet?
>
>  I know everyone recommends m2crypto --- but m2crypto-0.07-snap2
>  doesn't build on my Debian box.  I get errors like:
>
> Generating wrappers for Python
> _bio.i : Line 30. Syntax error in input.
> _bio.i : Line 31. Syntax error in input.
> _bio.i : Line 32. Variable %constant multiply defined (2nd definition
> ignored).
> _bio.i : Line 32. Syntax error in input.
> _bio.i : Line 33. Variable %constant multiply defined (2nd definition
> ignored).
> _bio.i : Line 33. Syntax error in input.
>
>  ... meanwhile I can simply apt-get install python2.2-ssl
>
>




From neil at anyisle.org  Thu Feb  7 19:08:06 2002
From: neil at anyisle.org (Neil Edwards)
Date: Fri, 8 Feb 2002 00:08:06 -0000
Subject: SocketServer - Interrupting get_request()
Message-ID: <u665n564lv9v8b@corp.supernews.com>

Hi there, I am currently writing a piece of code based on
some code I saw here (in the group). Basically I am
trying to write a simple MUD in python using the SocketServer
class and am having trouble when it comes to shutting down
the server from a client.

Here is the code so far:

#! /usr/bin/python -O

import sys, SocketServer, socket
import string

true  = 1
false = 0
PORT = 2001
serverUp = false
connections = [] #list of all client sockets

class RequestServer(SocketServer.ThreadingTCPServer):
   allow_reuse_address = 1

class RequestHandler(SocketServer.StreamRequestHandler):
   def handle(self):
      global serverUp, connections
      host, port = self.client_address
      print "Connected by", host, port
      connections.append(self.connection)
      # get data
      while true:
         request = self.rfile.readline()
         self.server.request = request
         request = string.rstrip(request)
         print request
         if(request == "shutdown"):
            for eachConnection in connections:
               eachConnection.send("Shutting down NOW!\n")
            serverUp = false
            #FIXME - Need someway of stopping server tasks

def main():
   global serverUp
   # get port
   if len(sys.argv) > 1:
      port = int(eval(sys.argv[1]))
   else:
      port = PORT
   print "Waiting..."
   server = RequestServer(('', port), RequestHandler)
   serverUp = true
   while(serverUp):
      server.handle_request() # do not server forever - serve just 1 request
   print "Closing all connections..."
   for eachConnection in connections:
      eachConnection.close()
main()

Wondered if anyone had any ideas, basically the server seems to block
waiting
for the next connection and so I can't end the program while it is blocked
(I even tried a sys.exit(0) where the FIXME is and it didn't exit)
Sorry if my code looks a bit poor this is one of the first progs I've done
in python)

Full credit to the original author of the code by the way!!
Thanks for making it public!


Thanks for the help
---
Neil Edwards (neil at anyisle.org)
http://anyisle.org




From donn at u.washington.edu  Fri Feb 22 17:40:41 2002
From: donn at u.washington.edu (Donn Cave)
Date: 22 Feb 2002 22:40:41 GMT
Subject: hep needed with pipes (popen4)
References: <iUyd8.18492$X2.205012@nnrp1.uunet.ca>
Message-ID: <a56hd9$18ea$1@nntp6.u.washington.edu>

Quoth "Darcy Kahle" <dkahle at interactdirect.com>:
...
| I am running a comand using os.popen4().  The command produces output until
| the user hits control-C, at which time, it prints a summary, and exits.  I
| want to send that character down the pipe to it without the user doing so
| (control-c kills the script, by default).  I tried sending "\x003", which to
| the best of my knowledge, is the hex code for it, but it didn't work.

Once the command is behind a pipe, control-C is just data.  You got the
data right, but if you want the same effect as the terminal driver gives
you, deliver a SIGINT to the process (os.kill(pid, signal.SIGINT).)

| If this is not possible, then I could do the same by redirecting the stdin
| to /dev/null, but if I do that at the start, the command will stop
| immediately.

What happens if you close its input pipe?  That's really the ideal way to
tell a process its job is done.

	Donn Cave, donn at u.washington.edu


From jkraai at paypal.com  Wed Feb 13 18:28:33 2002
From: jkraai at paypal.com (jim kraai)
Date: 13 Feb 2002 15:28:33 -0800
Subject: Container Objects Enforcing Contents
Message-ID: <ff8171ee.0202131528.1a9aabfe@posting.google.com>

Greetings,

I'm looking for suggestions on making container objects which must,
may, or must not allow particular kinds of container objects to be
added.

Any suggestions on a general approach for this?

The target app is not very exciting--an HTML generator that should be
able to guarantee that if no exceptions are raised, the HTML is valid.

TIA,

--jim


From jason at jorendorff.com  Fri Feb  1 00:07:36 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 31 Jan 2002 23:07:36 -0600
Subject: Naming higher order functions
In-Reply-To: <a3cch0$8j9$07$1@news.t-online.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEJJCPAA.jason@jorendorff.com>

> is the following right: [...]

Yes.

## Jason Orendorff    http://www.jorendorff.com/



From cesar.douady at asim.lip6.fr  Tue Feb  5 13:33:06 2002
From: cesar.douady at asim.lip6.fr (Cesar Douady)
Date: Tue, 05 Feb 2002 19:33:06 +0100
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de> <a3p2fg$49g$1@mail1.wg.waii.com>
Message-ID: <20020205.193304.1335939811.1571@asim.lip6.fr>

In article <a3p2fg$49g$1 at mail1.wg.waii.com>, "Martin Franklin"
<martin.franklin at westgeo.com> wrote:

> MDK wrote:
> 
> 
>> "Emile van Sebille" <emile at fenx.com> wrote in message
>> news:a3p0ss$19ob7s$1 at ID-11957.news.dfncis.de...
>>>
>>> "MDK" <mdk at mdk.com> wrote in message
>>> news:a3p01b$186eq8$1 at ID-98166.news.dfncis.de...
>>> > Hello,
>>> >
>>> > I am trying to create my own special type of list object.
>>> >
>>> > However, when I do x.append('yo') it uses Python's append instead of
>>> the one
>>> > from my class.
>>> >
>>> > How can I get it to use my append method?  I have tried def
>>> > __append__
>>> and
>>> > then called my own class' function but that did not work.  I've
>>> > played
>>> with
>>> > __getattr__ but that did not help.
>>> >
...
>> 
> Post some code.  perhaps the answer is obvious...
> 
> 
> for example
> 
> 
>>>> class MyList:
> ...     def __init__(self):
> ...             self.list=[]
> ...     def append(self, item):
> ...             self.list.append(item) ...
>>>> ml=MyList()
>>>> ml.append(1)
>>>> ml.list
> [1]
>>>> 
>>>> 
> would apear to work?

as does inheriting from list (python 2.2) :

>>> class l(list):
...     def append(self,x):
...             print "hello"
...             list.append(self,x)
... 
>>> a=l()
>>> a
[]
>>> a.append(2)
hello
>>> a
[2]


From eric at enthought.com  Thu Feb 14 17:16:09 2002
From: eric at enthought.com (eric jones)
Date: Thu, 14 Feb 2002 22:16:09 GMT
Subject: help!: optimization of search problem (example code)
References: <3c6bcf70.16509609@news.easynews.net>
Message-ID: <JGWa8.159434$XZ1.6306345@typhoon.austin.rr.com>

> Hi,
>
> after some days of playing around I am not getting
> any further and have to ask for some more help.
>
> Here is my problem. I have two large lists ("la" and "lb") of tuples.
> Each tuple represents a point in space and has four entries:
> x, y, z, pointID
>
> Now I have to find out for each point of list "la" which point of
> "lb" is closest. The result should be for all points of "la":
>
> Point 123 of "lb" is the closest one to Point 5932 of "la"
> where 123 and 5932 is the pointID.
>

SciPy has a vector quantization module that does what your looking for and
it is pretty fast.

C:\temp>python vq_ex.py
500 points ran in 0.037537 seconds

#vq_ex.py
from scipy import *
from scipy.cluster.vq import vq
from RandomArray import random
import time

N = 500

a = random((N,3))*100
b = random((N,3))*100

index = arange(N)[:,NewAxis]

la=hstack((a,index))
lb=hstack((b,index))

t1 = time.clock()
indices,distortion = vq(la[:,:3],lb[:,:3])
t2 = time.clock()
print '%d points ran in %f seconds' % (N,t2 - t1)

#print 'points in a are closet to the following points in b:'
#print indices

Hope that helps,
eric




From gimbo at ftech.net  Tue Feb 26 12:07:51 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 17:07:51 +0000
Subject: Precompiled regular expressions slower?
In-Reply-To: <1014742048.11444.12.camel@hunsmac.mit.edu>
References: <31575A892FF6D1118F5800600846864DCBCD08@intrepid> <1014742048.11444.12.camel@hunsmac.mit.edu>
Message-ID: <20020226170751.GB28508@andy.tynant.ftech.net>

On Tue, Feb 26, 2002 at 11:47:27AM -0500, Peter Bienstman wrote:
> Sticking with readlines rather than xreadlines for the moment to compare
> apples to apples, these are the timings:
> 
> precompiled as global var: 4.7 sec
> precompiled as local var : 4.5 sec
> not precompiled          : 3.9 sec
> 
> It helps somewhat, but doesn't solve the problem.

Do you get similar results when you ramp up the number of searches
being executed?

Disclaimer: I don't know if what I'm about to suggest is true, I
haven't looked at the code, tried this out or measured anything, I'm
guessing, I should be ignored, etc.  ;-)

Possible mechanism for what's happening:

    - Setting up a compiled re object takes time X

    - Executing one search using compiled re object takes time x

    - Thus, n searches using compiled object takes X + (n * x)

    - Executing search using re.search() takes time y

    - "Setup" is included in that time (basically consists of
      following reference to re.search())

    - Thus, n searches using re.search() takes n * y

I'm thinking maybe n is small enough and X is large enough that even
though x _is_ smaller than y, X + (n * x) is still larger than n * y

The question is, how many searches do you need to perform to make
compiled regular expressions worthwhile?

Just a thought...  As I said, I could be completely barking up the
wrong tree.  If you _do_ get similar results for much bigger values of
n, I'd guess I'm wrong.  But if that's the case then yeah, what the
heck are precompiled regular expressions for? :-)

-Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From kfarmer at thuban.org  Mon Feb 25 22:13:09 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Tue, 26 Feb 2002 03:13:09 GMT
Subject: Time to rename Stackless?
References: <mailman.1014690581.12732.python-list@python.org>
Message-ID: <93De8.43299$jl6.2013070358@newssvr14.news.prodigy.com>

PILOT?

I remember that language!

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:mailman.1014690581.12732.python-list at python.org...

> Pylet (Pilot).





From jgardn at alumni.washington.edu  Mon Feb  4 11:50:25 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Tue, 5 Feb 2002 01:50:25 +0900
Subject: OT: Common Courtesy
Message-ID: <200202041655.g14Gtqi29796@my.knctv.co.kr>

Go do a google search about something called "netiquette" and while you're at 
it, do a cross search for "mailing list" and "usenet". 

Here's a few links for the lazy or ignorant:

http://www.google.com/search?q=netiquette
http://www.google.com/search?q=netiquette+mailing+list
http://www.google.com/search?q=netiquette+usenet

Did you actually read it? They all say pretty much the same thing. Let me 
summarize:

1) Don't feed the TROLLS.
http://www.google.com/search?q=netiquette+troll

2) Don't FLAME.
http://www.google.com/search?q=netiquette+flame

3) Do some other nice stuff to make everyone else's life a little easier.

Jonathan

PS: Don't bother replying to my message and telling everyone in the Python 
community about it. If you think I made a good point, then you can tell me 
personally.






From jeff at ccvcorp.com  Thu Feb 21 17:25:06 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 14:25:06 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <7xg03vsidi.fsf@ruckus.brouhaha.com> <e9e6dd1f.0202211235.4e7be9aa@posting.google.com>
Message-ID: <3C7573C1.8C588E1E@ccvcorp.com>


Jason wrote:

> I still think that the convenience of having sort() return the list
> far outweighs the cost.  People would only get bit once, maybe twice,
> before they learned the behavior of sort.

Except that it's a very subtle bug, very difficult to spot that first time, and even if you've seen it
before it'd be easy to miss.  And you won't get bit by it every time, so it'll lull you with a false
sense of security.


> I like Python, but the biggest problem I have with the language right
> now is what low code density it has at times.

>cough<  Try comparing the code examples in, say the wxWindows documentation -- a page or more of code
in C++, which can be replaced with 3-5 lines of Python...

Jeff Shannon
Technician/Programmer
Credit International




From jjl at pobox.com  Sat Feb  2 17:04:10 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 22:04:10 +0000
Subject: Standing Objects ... possible?
In-Reply-To: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>
References: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>
Message-ID: <Pine.SOL.4.30.0202022157430.597-100000@mimosa.csv.warwick.ac.uk>

On 2 Feb 2002, Philip Swartzleonard wrote:
[...]
> the interperter. What i want is a construct like:
[...]
>     	or more likely:
>
> class Foo( ancestor, standing_mixin ):
>     	pass
>
>     	to be more or less equivelent to:
>
> class _Foo( ancestor ):
>     	pass
> Foo = _Foo()
[...]
> I can make it self-initiate... In any case, is there a prefered way of
> writing hidden stuff like the _Foo for doing it the more possible way?
[...]

What is your motivation here?  I really can't imagine why you'd want those
two pieces of code to be equivalent!  BTW, _name is only a convention: it
is treated by the interpreter in the same way as any other symbol not
starting with a *double* underscore.  Double underscore __names are hidden
to some extent.


John



From dalke at dalkescientific.com  Sun Feb 17 12:22:03 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Sun, 17 Feb 2002 10:22:03 -0700
Subject: Convert String to Dictionary question
References: <mailman.1013870847.12516.python-list@python.org>
Message-ID: <a4op0l$cjp$1@nntp9.atl.mindspring.net>

Jason Orendorff:
>Whereas there are no known security holes in pickle.

Nope.  Don't ever unpickle untrusted data without a *lot*
of care.

In
http://groups.google.com/groups?selm=9slgqh%24ffm%241%40slb2.atl.mindspring.
net&output=gplain
I show how to delete an arbitrary file using pickle (note:
doesn't work with cPickle, but there are any other attacks I
didn't try).

                    Andrew
                    dalke at dalkescientific.com





From joshm at taconic.net  Wed Feb 27 01:06:28 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Wed, 27 Feb 2002 01:06:28 -0500
Subject: Parse variable into variables >Re: Apology
References: <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <2240.820T2407T12845765threeseas@earthlink.net> <r29j7uo1ckvhesgc7po0thcevmdg0jeukg@4ax.com> <3C7AF404.6BC628F0@cosc.canterbury.ac.nz> <1121.822T529T14383507threeseas@earthlink.net>
Message-ID: <3c7c7503_2@corp.newsgroups.com>

> testfile = open('test','r')
> varset = testfile.readline()
> tple = varset.split()
> vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = tple
> I assume this will work???

Almost.

vic = {}
testfile = open('test','r')
varset  = testfile.readline()
tple = varset.split()
vic[('AI','1')], vic[('AI','2')], vic[('AI','3')] = tple
testfile.close() # don't forget to clean up after yourself

and yes, you can shorten the entire thing down to:

vic[('AI','1')], vic[('AI','2')], vic[('AI','3')] =
open('test','r').readline().split()

You don't have to close the file in this case, because the file handle goes
out of scope after the line ends, and the file closes automatically.  Of
course, if your file doesn't have exactly three fields separated by white
space on the first line, this will probably throw an exception.
--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From eppstein at ics.uci.edu  Wed Feb 27 17:46:05 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 27 Feb 2002 14:46:05 -0800
Subject: Status of PEP's?
References: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com> <eppstein-546E75.11465727022002@news.service.uci.edu> <ckmq7uo8l48cq9ig5qpgciu96gfc6sqb4r@4ax.com>
Message-ID: <eppstein-029172.14460527022002@news.service.uci.edu>

In article <ckmq7uo8l48cq9ig5qpgciu96gfc6sqb4r at 4ax.com>,
 Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote:

> >> PEP 276: Iterator for ints.
> >> (I really love this one - as simple and "obvious" as it gets)
> >
> >I don't know an answer to your question, but in contrast to your opinion, 
> >I really dislike this one.  It gives a very nonintuitive meaning for
> >"x in y" when both are integers (equivalent to "x <= 0 < y"), and that 
> 
> Now you have me lost. As far as I understand the PEP only talks about
> providing an iterator for integers. It does not talk about supporting
> the in method for integers, which, as you state, would make up for a
> confusing syntax.

The "in" operation takes as arguments an object and an iterator.
If integers can be iterators, they can be second arguments to "in".
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From jimmy at retzlaff.com  Mon Feb 18 14:48:01 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Mon, 18 Feb 2002 11:48:01 -0800
Subject: ActiveX Questions
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F55A6@MAYOR.retzlaff.com>

Rony says:
> Well i did try it with the same error !

I took a very quick look at this OCX. It's hard to tell from its
voluminous documentation :) but it's possible that it's only intended to
run as an ActiveX Control embedded in a form (i.e., dialog). Notice that
both the VB and VC++ samples have a little rectangle at the top of the
form that represents the control in design mode. It might not work
through COM Automation alone (which is what win32com.client.Dispatch
gives you access to).

Hosting ActiveX Controls is a good bit more work. I don't know how
complete it is, but Mark Hammond had started on support for this in
win32comext.axcontrol (I've never tried it).

wxPython works with win32all to support this in
wxPython.lib.activexwrapper.


Jimmy



From phr-n2002a at nightsong.com  Sun Feb 17 19:15:03 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 16:15:03 -0800
Subject: help - inserting values into database tables
References: <mailman.1013981787.10266.python-list@python.org>
Message-ID: <7xpu33myrc.fsf@ruckus.brouhaha.com>

chajadan <python at chajadan.net> writes:

> Yes, this is altogether possible.
> 
> There is a module include in the main distribution called odbc, and
> one called dbi.

Whaaa?  That's the first I've heard of this.  In Python 2.2 under Linux,
"import odbc" and "import dbi" both throw ImportError.

There's an ODBC module in Vaults of Parnassus that looks pretty good, but I haven't
tried it yet.  Is that what you meant?


From jkraska at san.rr.com  Thu Feb 28 01:47:16 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 28 Feb 2002 06:47:16 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <slrna7rdna.5m1.QnickQm@localhost.localdomain> <3c7dcc76.261969897@mammoth.usenet-access.com>
Message-ID: <htkr7ukldjug0vh33jacu2rbf0i52u93sr@4ax.com>

>Yeah.  I've designed dozens of ciphers before, all of which fail
>horribly the second I think about frequency analysis.  On the other
>hand, I've never tried actual encryption... I don't quite grok modular
>mathematics yet enough to do anything but monkey-see-monkey-do copies
>of other opensource ciphers.

Start with basic Linear hashing algorithms. Move your way up
the ladder to CRC calculations and then MD5. While you may
never be a whacky mathematician, you should be able to create
your own novel ciphers in short order after that. They'll all
be breakable probably, but that's a different issue.

Hashing is a baby step towards encryption: it's an attempt
to create a non random thing which is undecypherable from random.

Of course, the randomness of hashing is used for some purpose
other than to hide things, usually. :-)

C//



From mcherm at destiny.com  Tue Feb 19 11:35:45 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 19 Feb 2002 11:35:45 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
Message-ID: <3C727EE1.1090801@destiny.com>

Thank you Aahz for stating it so nicely. Most conversation in c.l.py 
goes on in English, and if it's in anything else *I* am not likely to be 
able to follow it, but I certainly don't begrudge others the use of 
their own languages here!!!

I-n-t-e-r-n-a-t-i-o-n-a-l-i-z-a-t-i-o-n-lly yours,

-- Michael Chermside




From bugcreator at gmx.de  Thu Feb 14 09:52:49 2002
From: bugcreator at gmx.de (Dominic =?iso-8859-1?q?Sacr=E9?=)
Date: Thu, 14 Feb 2002 15:52:49 +0100
Subject: Disabling creation of .pyc files?
In-Reply-To: <15466.40047.754661.201615@beluga.mojam.com>
References: <E16b1mB-0001Ls-00@mail.python.org> <15466.40047.754661.201615@beluga.mojam.com>
Message-ID: <E16bNJ4-0000Wx-00@mail.python.org>

On Wednesday 13 February 2002 18:03, Skip Montanaro wrote:
>
> It's there for your benefit (imports are faster when up-to-date .pyc
> files are present because the compilation step is avoided), but yes, you
> can disable it by simply making the directory containing the .py files
> read-only.

I know that it's generally a useful feature, but sometimes it's annoying, 
because I'd like to keep the directories containing my scripts clean...
And making the directories read-only is not possible when I'm still 
working on the scripts.

Well, looks like I'll have to remember running
find -name "*.pyc" | xargs rm
every now and then...

    Dominic




From majamete at yahoo.com  Fri Feb 15 04:40:13 2002
From: majamete at yahoo.com (majamete at yahoo.com)
Date: Fri, 15 Feb 2002 04:40:13 -0500
Subject: Elegant - Unique Last Minute Valentine Gift Idea
Message-ID: <09029H8C.7XR89S78505XT.majamete@yahoo.com>

A VALENTINE GIFT

A TIMELESS UNIQUE PRESENT FOR YOUR SWEETHEART FROM 
  
  
J. HAMBLETON, LTD. 
Order today to guarantee delivery by February 14th. 
  
http://www.jhambleton.com 
  

Store Locations: 
Century City Shopping Center, Los Angeles, CA 
Biltmore Fashion Center, Phoenix, AZ 
  
800-337-1995

 
======================================================= 
 
This email was sent to you by the NetWizards Marketing, 1601 NW 97th Ave, SJO3016, Miami FL 33102.  011-506-267-7139.  This email cannot be considered spam as it was sent in compliance with all existing and proposed email legislation.  If you wish to be removed from this mailing, please send an email here carsonsitos at yahoo.com with remove in the subject line 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020215/6a47f86f/attachment.html>

From quinn at hork.ugcs.caltech.edu  Sat Feb  9 21:41:41 2002
From: quinn at hork.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 02:41:41 GMT
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4>
Message-ID: <slrna6bnf4.9tv.quinn@hork.ugcs.caltech.edu>

On 9 Feb 2002 18:21:31 GMT, Philip Swartzleonard <starx at pacbell.net> wrote:
>So i've been poking around with ways of doing this messaging thing, and 
>i think up this (sorry for the inanity of some of the names, i was sick 
>and tired when i worte this :):    	
>
>
>class Message:
>    type = "none"
>class FightoMessage(Message):
>    type = "fighto"
>class SpelloMessage(Message):
>    type = "spello"
>
>class Reciever:
>    def message( self, message ):
>        methname = 'rec_' + message.type
>        if hasattr(self, methname):
>            if callable(getattr(self, methname)):
>                getattr(self, methname)(message)

I recommend leaving the 'type' attributes out.

class Message: pass
class Fighto(Message): pass
class Spello(Message): pass

class Receiver: # i before e except after c!
    def message(self, msg):
        msgname = 'rec_' + msg.__class__.__name__
        if hasattr(self, msgno):
            # recommend leaving out the callable(), why let people define
            # non callable rec_ attrs?
            getattr(self, msgname)(message)

To the other poster who had the idea of using dicts to alias messages, you
can do the same with:

class F(Receiver):
    def rec_fighto(self, message):
        print "FIGHTO!!"
    rec_hacko = rec_fighto

>I dont think a better solution i could ever have seen. It does all the 
>work of a big ugly switch statment, with none of the uglyness OR the 
>hard-codedness. I can define new message types at will and only update 
>things that i WANT to recieve them.

Yes, you shouldn't be reinventing a type system with messages and receivers,
since the language already implements a perfectly good type systom with
messages and receivers (switching to smalltalk-speak since you're using it, but
the usual python terms are methods/attributes and objects).

>Also, is there any real reason to have the message classes inherit when 
>the only thing that seems to identify them is an attribute?

No, unless you wanted messages to have some messages of their own, which I
assume you did since otherwise you could have made them strings.  If you
do want all Messages to respond to a few common messages (small 'm'), like
'sender' or 'time_sent', then inheritance is natural.


From ivan at kupol.ru  Mon Feb 25 23:44:49 2002
From: ivan at kupol.ru (Ivan Bykov)
Date: Tue, 26 Feb 2002 08:44:49 +0400
Subject: socket in python 2.2 vs 2.1
In-Reply-To: <j48z9mq1r0.fsf@informatik.hu-berlin.de>
References: <mailman.1014206732.25525.python-list@python.org>
 <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de>
 <mailman.1014290914.10912.python-list@python.org>
 <j48z9mq1r0.fsf@informatik.hu-berlin.de>
Message-ID: <442684269.20020226084449@kupol.ru>

Hello!

Thursday, February 21, 2002, 19:43:31, Martin wrote:
>> Python 2.1: '255.255.255.255' - valid address. And my application
>> working is correct.
MvL> 255.255.255.255 is not a valid IP number (even though Python 2.1
MvL> accepts it - it actually looks like a bug in the system's IP stack
MvL> which fails to reject this address). Your application works just
MvL> because the system fails to check your address correctly. Python 2.2
MvL> uses different code to check, thus detects the error.
IMHO, '255.255.255.255' is a valid IP number. This is number mask, use
for broadcast packet.
MvL> Just remove the connect call.
Then program to work not will.

-- 
Best regards,
 Ivan                            mailto:ivan at kupol.ru





From python at chajadan.net  Tue Feb 26 15:19:53 2002
From: python at chajadan.net (chajadan)
Date: Tue, 26 Feb 2002 12:19:53 -0800
Subject: gnuplot woes, a second story
In-Reply-To: <f2730b27.0202260345.2d984578@posting.google.com>
References: <mailman.1014696821.3069.python-list@python.org>
Message-ID: <5.1.0.14.0.20020226121826.00a7a680@63.118.78.184>

At 03:45 AM 2/26/02 -0800, Jannie Hofmeyr wrote:
>chajadan <python at chajadan.net> wrote in message 
>news:<mailman.1014696821.3069.python-list at python.org>...
> > When I try to run gnuplot from my Python2.2 installation, I get an error
> > (listed below)
> >
> > I'm on windows 98, I installed NumPy, I unzipped gp371w32.zip into my main
> > python directory, and I tried changing the command to the gnuplot program
> > FROM gnuplot_command = 'pgnuplot.exe' TO gnuplot_command =
> > 'c:\Python20\gp371w32\pgnuplot.exe'
>
>Are you sure your path to gnuplot is correct? Shouldn't it be
>'c:\Python22\...


something so simple--indeed, should be 22, not 20. incidentally, it works 
fine using the win32pipe module for me--I haven't tried the os one yet.

Thanks!

--chajadan




From usenet at thinkspot.net  Sat Feb  9 18:06:53 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 09 Feb 2002 15:06:53 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com>
Message-ID: <a43due.3vv9a5r.1@kserver.org>

[posted and emailed]

On 9 Feb 2002 14:32:35 -0800, wookiz at hotmail.com (wooks) wrote in
comp.lang.python in article
<8852d33c.0202091432.452ad6f9 at posting.google.com>:

> Haven't got that far. The book is the most appropriate for my needs -
> scripting glue, interacting with WSH and COM rather than building
> applications. I met the prerequisites for readership but I am finding
> out that the devil is in the detail.

I think Python is extremely appropriate for your needs. The problem is,
you are trying to jump straight into high-level stuff like that before
you are really comfortable with the language itself.

I recommend:
(1) The Python tutorial that comes with the standard distribution:
http://www.python.org/doc/current/tut/tut.html
(2) Core Python Programming by Wesley Chun
(not for beginning programmers...aimed at someone with some programming
experience...addresses the OO points and many features of the language
quite well)
(3) For an advanced online tutorial, try Dive Into Python:
http://diveintopython.org/

I also have the Programming Python book, and I liked it very much. It
isn't a reference, though, but has many excellent examples. Good for
experienced programmers who already know Python. I learned a lot from
it. 

I've seen good comments about the Learning Python, also from O'Reilly.
Sounds to me like a possibility for someone in your shoes. Anyhow, I
can't recommend that one, since I'm not familiar with it.

I have a friend I'm working with right now, that we are putting together
a project in Python and PHP (he's doing all of the PHP part). He has
only just been learning Python for a few weeks. He might have some good
suggestions for you, as to how he approached learning Python (as an
experienced programmer in other languages). If you'd like me to put you
in touch with him, send me an email.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From thomas.heller at ion-tof.com  Fri Feb  1 09:30:04 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Feb 2002 15:30:04 +0100
Subject: ANN: py2exe 0.3.1 uploaded (was py2exe version 0.3.0 - does now windows NT services)
References: <a3dp5u$179scg$1@ID-59885.news.dfncis.de> <a3dsee$17f557$1@ID-59885.news.dfncis.de>
Message-ID: <a3e8pk$17pqfh$1@ID-59885.news.dfncis.de>

> Thomas Heller wrote...
> > py2exe is a distutils extension to convert python scripts
> > into windows exe-files (plus maybe a few dlls), able to run
> > on computers without requiring a python installation.
> >
> >
> > New in release 0.3.0:
> >
> >   The first release which can create windows NT services:
> >   A sample service with detailed instructions is provided.
> >   Builds version resources into the executables.
> >   Other small bug fixes as well.
> >
> >
> > Download from the usual location:
> > http://starship.python.net/crew/theller/py2exe/
> >
>
> Peter Goode found a bug in this release which prevents
> using the -O flag. Many thanks to him!
>
> I will hide the 0.3.0 release, and prepare a 0.3.1 release
> which fixes this bug. Sorry for the inconvenience.

I've uploaded 0.3.1, which fixes this bug.

Thomas




From fperez528 at yahoo.com  Mon Feb 11 20:09:31 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 12 Feb 2002 01:09:31 +0000
Subject: ANN: IPython 0.2.5
Message-ID: <a4ainp$gts$1@peabody.colorado.edu>

Announcing an update to IPython, an enhanced interactive Python shell. 

WHERE: http://www-hep.colorado.edu/~fperez/ipython

NEW since last public release:

 - Ported to Windows (only tested XP, should be ok elsewhere), including a 
point-n-click installer (text-mode, but functional).

 - Completed user manual (html and pdf).

 - Added ability to run full programs from inside IPython (its tracebacks are 
much better than the standard python's), with command-line args.

 - Macro system for re-executing previous input (multi-line) with named 
macros.

 - Enhanced object introspection system to report class constructors 
docstrings and function signatures.

 - Updates for small changes with Python 2.2 and unicode.

 - Many small enhancements and bugfixes.


What is IPython? (this is just for context. See the webpage for full 
details). IPython tries to:

1. Provide an interactive interpreter superior to Python's default. IPython 
has many features for object introspection, shell access, and its own special 
command system for adding functionality when working interactively.

2. Serve as an embeddable, ready to use interpreter for your own programs. 
IPython can be started with a single call from inside another program, 
providing access to the current namespace. This can be very useful both for 
debugging purposes and for situations where a blend of batch-processing and 
interactive exploration are needed.

3. Offer a flexible framework which can be used as the base environment for 
other systems with Python as the underlying language. Specifically scientific 
environments like Mathematica, IDL and Mathcad inspired its design, but 
similar ideas can be useful in many fields.


Portability: Linux (and other unices, including Mac OSX), Windows XP. Should 
run fine on all WinNT, and probably also on Win9x (I can't test that).

Python version: requires 2.1 or newer.

License: LGPL (a few files from third parties carry MIT licenses).

Cheers,

Fernando Perez.



From h.jehadalwan at student.kun.nl  Sat Feb 23 04:33:08 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Sat, 23 Feb 2002 10:33:08 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl> <a56o39$7dg$1@animus.fel.iae.nl>
Message-ID: <3C7761D4.7050100@student.kun.nl>

Carel Fellinger wrote:

> husam <h.jehadalwan at student.kun.nl> wrote:
> 
> 
>>hi Gustaveo,
>>You'r solution did not work either. The code hangs at the first loop, 
>>specifically at "log.write(o.read())".
>>
> 
> You don't seem to be having a Python problem, so what about trying to
> get some output from the program?  Instead of reading the log into a



> var, what about trying it with `os.popen("/home/programs/profit", "w)'

Off course, it works this way. But, then I have to execute another 
script to process the output. Therefor I chosen for "popen2" so that I 
can process the output in the same script.



> and have the output directed at the terminal for direct inspection.
> Maybe you're just missing error messages:)
> 
> An other problem that might buck you (but I don't see how this would
> be the case as your program seems to work for a few files) is that
> pipes are buffered.  So uses `i.flush()' to be sure that what you


I'v tried "flush()" already, no go.


> wrote really is transmitted to the other end of the pipe.
> 
> And as a last resort you could try reading the `ls' output a line at a
> time.
> 
>     i = os.popen("/home/programs/profit ", "w")
> 
> Should this ---------------------------^ space be there?


The space is needed here, otherwise the command and it's arguments 
passed to the progam, get passed like this: "CommandArguments" instead 
of " Command Arguments"


> 
>     i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") 
>     for PDB in os.popen('ls /home/data/pdbs'):
>         i.write(" mobile /home/data/pdbs/" + PDB + "\n")
> 
> Should this -----^ space be there?


The same explanation as above!

> 
>         i.write("fit\n")
>         i.flush()
>     i.close()
> 




From mhammond at skippinet.com.au  Fri Feb 15 20:22:45 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 16 Feb 2002 01:22:45 GMT
Subject: win32 DeleteObject ??
References: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk>
Message-ID: <3C6DB4D0.1020601@skippinet.com.au>

Robin Becker wrote:
> In Mark Hammond's excellent win32 extensions I can see how to obtain a
> handle to a bitmap, but how does one remove from memory? I can't seem to
> find DeleteObject.
> 
> Similarly I can't seem to find things like CreateCompatibleDC, BitBlt
> etc which are part of win32. It almost seems as though I'm missing
> win32gdi.pyd :)
> 
> Do I really have to lug along the whole of MFC to do simple things?

These should be exposed by a module, I agree.  But they aren't :(.  They 
could be added to win32gui, or, as you suggest, a new win32gdi module 
created.

These few functions would be trivial to add to win32gui, but it is not 
clear if that is all you need.

Mark.



From jeff at ccvcorp.com  Wed Feb  6 13:19:07 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 06 Feb 2002 10:19:07 -0800
Subject: Binary Python file.
References: <3C5FA9E1.DF4AC37F@lucent.com> <Xns91ADDDAD65A6RASXnewsDFE1@130.133.1.4>
Message-ID: <3C61739B.E4701402@ccvcorp.com>


Philip Swartzleonard wrote:

> Lukasz Pawelczyk || Tue 05 Feb 2002 01:46:09a:
>
> > Hello,
> >
> > Is there a possibility to make a binary python file?
> > I would like to hide the code in unix environment.
>
> Well, the compile function listed here will give you a pyc file...
> http://www.python.org/doc/current/lib/module-pycompile.html#l2h-3816
>
> ...but from what i've heard it's only a little bit better then say, rot-13,
> for keeping people out of your source. It will probably keep out the
> slightly curious, the incapable, and those of us who just don't care what's
> in there :)

And really, *nothing* will keep out the capable who *do* care, so...  >shrug<

Hiding code has proved to be of only marginal benefit.  (How many hidden-code
commercial projects can you find cracks for on scores of warez sites?)  If
someone wants to reverse-engineer your code badly enough, they will.    :)

(A well-written license does more to prevent unauthorized copying, than any
amount of code-hiding and disc-key-based systems.  Just tell people how you
want them to treat it, and ask them to play nice.)

Jeff Shannon
Technician/Programmer
Credit International




From altis at semi-retired.com  Sun Feb  3 15:21:28 2002
From: altis at semi-retired.com (Kevin Altis)
Date: Sun, 3 Feb 2002 12:21:28 -0800
Subject: Python 10 conference blogs and news?
Message-ID: <IXg78.53$HX4.83750@news.uswest.net>

Is anyone is going to be blogging the Python 10 conference? If your site
will be providing regular updates, news, and comments during the conference,
please respond to this thread with the URL of your site.

It is unclear whether there will be any coverage at http://www.python10.org/
or even direct links off the main http://www.python.org/ page.

Thanks,

ka




From brian at sweetapp.com  Mon Feb 18 21:01:00 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 18 Feb 2002 18:01:00 -0800
Subject: Adding static typing to Python
In-Reply-To: <0DFA7478F1E5D311B90900508B1208CF565712@ausmail02.mis.gtech.com>
Message-ID: <000901c1b8e9$474b4310$445d4540@Dell2>

Peter wrote:
> You have drawn a false assumption there Brian, you assume that 95% of
my
> bugs are due to misnamed variables - I wished! :-) That wasn't what I
was
> trying to imply or state with that statement.
> 
> Strong typing requires that you design before you code (Extreme
> Programmers beware! :-)) - therefore by the time a clean compile 
> is achieved 95% of the bugs *are* out :-) That is one of the 
> primary advantages to strong typing, it *forces* you to do the design 
> up front otherwise you will never get a clean compile.
> 
> You should try it one day :-)

I'm a reasonably competent C/C++ developer so I have experience working
with statically typed languages.

I don't see how they force you to invest more design effort before you
code. I have written a fair bit of C code, which (eventually) compiled,
without doing any design up front (the UTF-7 codec in core Python is one
such example :-). 

However, I often do more design work with C++ before coding, but that is
because the weaknesses in C++ makes it more complex to achieve the
desired semantics. I can hardly credit that as a good thing.

Cheers,
Brian




From ol1 at v10a.com  Tue Feb 19 03:16:38 2002
From: ol1 at v10a.com (O Polite)
Date: 19 Feb 2002 09:16:38 +0100
Subject: Recursively traverse linked list -- help!
In-Reply-To: <3C71B585.90F644EB@kontact-no-spam-please.com>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
Message-ID: <1014106598.4102.20.camel@matijek>

On Tue, 2002-02-19 at 03:16, bob wrote:

Hello -- I searched the archives to no avail. 

Does anyone have a python code snippet showing how to recursively
traverse a linked list? 
  

thanks 

Bob at kontactNOspamPLEASE.com -- 

Well, first you'd want to have a way of representing a linked list.
Here's one straighforward way of doing it, but unless your taking
Datastructures 101 you'll never be interested in it. The builtin list
type will always be faster than anything you implement in python, and
now in python2.2 you can subclass the builtin types as any other class.


class LinkedList:
    def __init__(self):
        self.start = None
        
    def push(self, value = None):
        new = ListElement(value)
        if self.start:
            new.next = self.start
        self.start = new

    def pop(self):
        result = None
        if self.start:
            result = self.start.data
            self.start = self.start.next
        return result
    
    def __str__(self):
        result = ""
        current = self.start
        while current:
            result += "%s " % (str(current))
            current = current.next
        return result

class ListElement:
    def __init__(self, value = None):
        self.data = value
        self.next = None

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

if __name__ == '__main__':
    mylist = LinkedList()

    for i in range(10):
        mylist.push(i)
        
    for i in range(11):
        print mylist
        mylist.pop()



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





From s.schwarzer at ndh.net  Thu Feb 14 15:04:43 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 14 Feb 2002 21:04:43 +0100
Subject: preference tree like in jdk 1.4
References: <mailman.1013704956.3992.python-list@python.org>
Message-ID: <3C6C185B.168D3F10@ndh.net>

Hello Tripp

Tripp Scott wrote:
> i was thinking of an empty function:
> [...]
> to store or load this pref object, just use the pickle mechanism.
> 
> my question is, is there a simpler, more beautiful, elegant way?

If you don't need this more than two levels deep, look at the
ConfigParser module. You can modify its instances also after having
read them and write them to disk, again as a configuration file. :-)

Stefan


From emile at fenx.com  Sun Feb 17 20:33:58 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 17 Feb 2002 17:33:58 -0800
Subject: execute file interactively
References: <e9e6dd1f.0202171731.6cf2a5e2@posting.google.com>
Message-ID: <a4pm48$260t3$1@ID-11957.news.dfncis.de>

"Jason" <caljason76 at yahoo.com> wrote in message
news:e9e6dd1f.0202171731.6cf2a5e2 at posting.google.com...
> When using the Python interpreter interactively, how can I execute a
> file, so I can use the function definitions and data defined in it
> interactively.

If you've got yourfile.py on your python path, simply

import yourfile

--

Emile van Sebille
emile at fenx.com

---------



From logiplexsoftware at earthlink.net  Wed Feb 27 14:37:20 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 27 Feb 2002 11:37:20 -0800
Subject: wxPython blocks in thread?
In-Reply-To: <rk63czmu79v.fsf@pegase.inria.fr>
References: <rk63czmu79v.fsf@pegase.inria.fr>
Message-ID: <20020227113720.0867df6b.logiplexsoftware@earthlink.net>

On 27 Feb 2002 19:08:44 +0100
Arnaud de La Fortelle wrote:

> 
> I am a beginner with wxPython and tried to embed a GUI within a
> thread.  So I came with a simple implementation, but it does not work.
> I do not understand why the following application blocks. What the
> matter?

> import threading
> class GUIThread(threading.Thread,object):
>   def __init__(self):
>     threading.Thread.__init__(self,target=self.loop)
>     self.start()
>   def loop(self):
>     print gui_thr
>     self.app = MyApp(0)
>     self.app.MainLoop() # if I comment out this line, it works
>     print "Ending wxLoop", self.isAlive()

wxApp.MainLoop() doesn't return until the last wxWindow is destroyed.  

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From gh_pythonlist at gmx.de  Sat Feb  2 20:44:56 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Sun, 3 Feb 2002 02:44:56 +0100
Subject: curses module
In-Reply-To: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>
Message-ID: <20020203014455.GB1251@lilith.hqd-internal>

Le 02/02/02 ? 16:59, Lawrence Oluyede ?crivit:
> 
> Why in the std distribution of Python 2.2 there isn't the curses
> module?

If you're using Windows, that explains it ;-) There is indeed no curses
module included in the Python binary distribution for Windows. IIRC the
reason is that there is no ncurses library readily available on win32 to
build the curses support against.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.8 ?C      Wind: 0.6 m/s



From pawelczyk at lucent.com  Thu Feb 14 05:22:57 2002
From: pawelczyk at lucent.com (Lukasz Pawelczyk)
Date: Thu, 14 Feb 2002 11:22:57 +0100
Subject: Layout management in tkinter.
Message-ID: <3C6B9001.E5568D2C@lucent.com>

Hello,

I try to change the geometry manager pack into grid in the code
and I have problem.
Why it does not work?

prz1=Button(main,text="Otw?rz",cursor="question_arrow",command=otworz)
prz1.grid(row=0)

prz2=Button(main,text="Zachowaj",cursor="dot",command=zachowaj)
prz2.grid(row=0,column=1)

prz3=Button(main,text="Zamknij",cursor="pirate",command=wyjscie)
prz3.grid(row=0,column=2)

plik=ScrolledText(main,background="grey")
plik.grid(row=1,column=0,columnspan=3)		- there the code stop


Regards,
Lukasz


From mwh at python.net  Fri Feb 22 05:56:02 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 22 Feb 2002 10:56:02 GMT
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014319058.22717.python-list@python.org>
Message-ID: <u1yfdlr4i.fsf@python.net>

"Jason Orendorff" <jason at jorendorff.com> writes:

> David Abrahams wrote:
> > Look at the site documentation. They only claim about 85% coverage of C++.
> 
> Cint limitations:
>   http://root.cern.ch/root/Cint.phtml?limitations
> 
> An interesting read.  Apparently their parser isn't very
> good.  (I would think just parsing the code would be the
> easy part.  Shows what I know.)

Parsing C++ is not much easier than compiling it, I believe.
Templates!

Cheers,
M.

-- 
 "An infinite number of monkeys at an infinite number of keyboards
  could produce something like Usenet."
 "They could do a better job of it."
              -- the corollaries to Gene Spafford's Axiom #2 of Usenet


From hamish_lawson at yahoo.co.uk  Mon Feb 11 09:47:04 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 11 Feb 2002 06:47:04 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com> <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu> <8852d33c.0202110117.374f062@posting.google.com>
Message-ID: <915a998f.0202110647.2f5a3b01@posting.google.com>

wooks wrote:

> Further if I get magic methods right, they are special functions which
> if present enable the object to apply a unique interpretation of a
> python operator.

> So + has standard behaviour but will  adopt whatever is specified in
> the __add__ function if one is present for the class.

If an object doesn't have __add__ defined (by itself, its class, or an
inherited class), then it won't have an interpretation *at all* for +.
There's no 'standard behaviour' that is defaulted to. This holds true
for the built-in types too: 4.0 + 3 has an interpretation only because
the float type has __add__ predefined for it.

Thus

    4.0 + 3

effectively results in a call to

    (4.0).__add__(3)

The second form could actually be directly called that way in Python
2.2 (prior to 2.2, magic methods for built-in types weren't exposed to
the programmer); the brackets around the 4.0 are necessary only to
disambiguate the number from the following dot.

So there is no standard behaviour for + in the sense of some default
action that gets performed in the absense of a defined __add__ method.
However the __add__ method is already defined for most built-in types.

But the operator symbols provide more than just a prettier notation.
When the operand types are different the interpreter will inspect the
operands and will decide which operand to ask to handle the operation.

Thus 

    4 + 3.0

will effectively result in a call to

    (3.0).__radd__(4)

rather than

    (4).__add__(3.0)

since integers don't know how to add floats.


Hamish Lawson


From mdk at mdk.com  Tue Feb  5 15:27:19 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 15:27:19 -0500
Subject: What is the command to determine an object's type?
Message-ID: <a3pf7k$1a0fij$1@ID-98166.news.dfncis.de>

How can I tell if an object is a string, a list, or a dictionary in a
program using if and ifelse so that I can perform a differnent action
depending on the object's type?




From op73418 at mail.telepac.pt  Mon Feb 25 14:34:11 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 19:34:11 +0000
Subject: How to call COM constants and global functions?
References: <a50631f6.0202251126.467ef9d7@posting.google.com>
Message-ID: <ub4l7usglcai95f8sus48b4jvnv7vdk65q@4ax.com>

On 25 Feb 2002 11:26:26 -0800, newgene at bigfoot.com (newgene) wrote:

>Hi, group,
>     Recently, I tried to integrate python with excel by using
>win32com module, but I don't know how to call a constants in VB for
>Excel, like 'xlCSV' for 'XlFileFormat', 'xlBackgroundAutomatic' for
>'XlBackground',etc. Also, I don't know how to call a global function
>like "RGB".
>     Hope you can help me out.
>
>PS. When I follow the code in "Python programming on win32" Page
>206-207(Using COM constants), it doesn't work for me. The code is :
>
>>>> from win32com.client import constants, Dispatch
>>>> xl=Dispatch("Excel.Application")
>>>> constants.xlAscending
>Traceback (most recent call last):
>  File "<interactive input>", line 1, in ?
>  File "C:\Python22\win32com\client\__init__.py", line 133, in
>__getattr__
>    raise AttributeError, a
>AttributeError: xlAscending
>
>My platform is ActivePython 2.2a2+ on Win2K.
>

As far as i remember from the same book you have to run makepy on the
excel library to have acess to the constants.

Hope it helps,
Gon?alo Rodrigues

>
>Thanks.
>
>CL



From sorr at rightnow.com  Tue Feb  5 14:48:29 2002
From: sorr at rightnow.com (Orr, Steve)
Date: Tue, 5 Feb 2002 12:48:29 -0700 
Subject: Saving Browser State without Cookies
Message-ID: <D408B6AC060DB24790622E1A54204DB00823AA81@xchange.rightnowtech.com>

Python Newbie here.
 
I need to migrate a web application from PHP to Python and I have a
requirement to save state info WITHOUT using cookies where much of the state
data MUST not be visible to the end users. The PHP application allows users
to logon to one of several databases using any of a number of Oracle
user/passwords. The logon info needs to be saved for queries on other pages
in the app. With PHP it's easy to avoid cookies by saving this info using
its session handling functions. Is there an EASY way to do this in Python
without having to adopt the Mother of all frameworks?
 
AtDhVaAnNkCsE,
Steve Orr
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020205/0e6a2c71/attachment.html>

From emile at fenx.com  Tue Feb 26 08:48:27 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 05:48:27 -0800
Subject: ask help about raise, thank u
References: <mailman.1014686626.31452.python-list@python.org> <c76ff6fc.0202260144.70e52642@posting.google.com>
Message-ID: <a5g41c$70hbr$1@ID-11957.news.dfncis.de>

"John Machin" 
>       # re-raise the exception, with traceback giving the *original*
>       # source of the problem (somewhere inside the 'try' suite),
>       # not the 'raise' statement
>       raise exc, data, tb
> 

Isn't this what raise by itself does?

>>> def test(d=None):
...     try:
...             1/d
...     except TypeError:
...             return 0
...     except:
...             raise
...
>>> test()
0
>>> test(0)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in test
ZeroDivisionError: integer division or modulo by zero
>>>

-- 

Emile van Sebille
emile at fenx.com

---------



From peter at engcorp.com  Thu Feb 28 09:39:36 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 28 Feb 2002 09:39:36 -0500
Subject: Want to monitor process..
References: <87550ef1.0202272315.15779bbd@posting.google.com> <f2lf8.385$Az6.1854@news.hananet.net>
Message-ID: <3C7E4128.3B839BD3@engcorp.com>

Jonathan Gardner wrote:
> 
> Haeyoung Kim wrote:
> 
> > I want to run or kill a process(program?) using python.
> 
> from os import *
> # Start a process
> ...
> pid = fork()
> if not pid:
>         # I'm a child!
>         exec*(what you want to do)
> # I'm a parent

Whoa!  Don't do that.  I mean the "import *" part 
with the os module.

Module os includes the open() function, and if you
import the namespace that way you'll be masking the
builtin open() function and have all kinds of grief.

Always "import os" and use os.fork(), os.kill() etc.
(Or at least do a "from os import fork, kill".)

-Peter


From claird at starbase.neosoft.com  Fri Feb  8 16:06:51 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 8 Feb 2002 15:06:51 -0600
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <41914a.537.ln@wombat.san-francisco.ca.us> <a40e2r.3vvegs9.1@kserver.org>
Message-ID: <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net>

In article <a40e2r.3vvegs9.1 at kserver.org>,
Sheila King  <usenet at thinkspot.net> wrote:
>On Fri, 8 Feb 2002 11:30:44 -0800, kkeller at speakeasy.net (Keith Keller)
>wrote in comp.lang.python in article
><41914a.537.ln at wombat.san-francisco.ca.us>:
>
>> In article <a417rn$ge8$1 at panix2.panix.com>,
>> 	aahz at panix.com (Aahz Maruch) writes:
>> > I know some of y'all still keep roots to the Perl community, so I was
>> > wondering if there's any kind of definitive nickname for Perl
>> > programmers, the way we call ourselves Pythonistas?
>> 
>> I don't think so--having one nickname would contradict TMTOWTDI.  :)
>
>I thought it was Perl Monks.
			.
			.
			.
"Mongers" is the closest answer.

I generally write "Perlites", though.
-- 

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


From rxg218 at psu.edu  Thu Feb 21 19:05:37 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Thu, 21 Feb 2002 19:05:37 -0500
Subject: execv giving a Fatal Error (11)
Message-ID: <a541uu$l0e@r02n01.cac.psu.edu>

Hi,
  I have a small python script which uses PyGTK. In one of the call backs  
I do:

os.execv('/usr/bin/emacs',[])

But as a result I get:

Fatal error (11).Segmentation fault

This occurs even when I run a script that contains the os.exev(...) as the 
only command.

Does anybody know why this happens? Is there any other way I can implement 
this?

TIA,
-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
Dept. Of Chemistry             | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
Alone, adj.:
        In bad company.
                -- Ambrose Bierce, "The Devil's Dictionary"



From jason at jorendorff.com  Wed Feb 27 17:36:50 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 27 Feb 2002 16:36:50 -0600
Subject: string title() method does not work on Danish characters
In-Reply-To: <3C7CFF19.5070703@mxm.dk>
Message-ID: <HFEKILOLEFEFMKAECNDLEENCDFAA.jason@jorendorff.com>

> I am writing strings from a database, and am using the title() method to
> format the look of the strings.

First: Unicode strings will solve your problem.

Second: I'm on US Windows, so my console won't display the
character ? properly; IDLE also doesn't really like it, etc.
So below, I'll use \xd8 to refer to ?.

Okay, here's the example.

>>> s = u'this is a name: MAX M\xd8LLER'
>>> s.title()
u'This Is A Name: Max M\xf8ller'

Here, Python properly converts \xd8 to \xf8.
To output it, though, you must encode() it properly...
I imagine you probably want Latin-1, so it would be

>>> s2 = s.title()
>>> print s2.encode('latin-1')
...or...
>>> file.write(s2.encode('latin-1')

## Jason Orendorff    http://www.jorendorff.com/




From spencer at spencerdoidge.com  Sun Feb 10 16:26:02 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sun, 10 Feb 2002 13:26:02 -0800
Subject: importing/reloading in Pythonwin
References: <a46l7r$57e$1@news.efn.org>
Message-ID: <a46o9r$69c$1@news.efn.org>

I should add that if I quit and restart Pythonwin, then import all from Foo,
I *will* see the changes I made.

"Spencer Doidge" <spencer at spencerdoidge.com> wrote in message
news:a46l7r$57e$1 at news.efn.org...
> Running Python22's Pythonwin.
> I import all from a script Foo, run some routines, and then edit Foo in
the
> edit window. I check syntax and save, then click the Import/reload button.
> Returning to the interactive window, I run Foo again, but I don't see the
> expected evidence of the changes I made.
> What am I doing wrong?
>
>




From logiplexsoftware at earthlink.net  Tue Feb  5 20:31:36 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 5 Feb 2002 17:31:36 -0800
Subject: Which GUI toolkit to use?
In-Reply-To: <3c607ffd_3@corp.newsgroups.com>
References: <3c607ffd_3@corp.newsgroups.com>
Message-ID: <20020205173136.476c527c.logiplexsoftware@earthlink.net>

On Tue, 5 Feb 2002 20:07:41 -0500
Joshua Muskovitz wrote:

> Specifics:
> 
> 1.  Needs to run on Windows platform.
> 2.  Needs to easily support completely custom paint (buttons, scrollers,
> listboxes, etc).
> 3.  Needs to handle simple input (editboxes, onLeftClick events).
> 
> The goal is to produce an app that runs full screen on a Win platform,
and
> is interacted with via a touchscreen.  I would like to just paint the
whole
> damn thing myself, but a widget manager will make input handling easier.
> There will be a lot of bitmaps making the whole thing pretty.
> 
> Tk seems like it is a pain to use, wxWindows seems like it doesn't make
> painting easy, pythonwin (win32ui) might be the way to go, ...

What do you mean "custom paint"?  I highly recommend wxPython.  If you are
referring to putting bitmaps on windows, buttons, etc, this is very doable
in wxPython.

> Decisions, decisions.  There wouldn't be anyone around here who has an
> opinion, is there?  :-)
> 
> --
> # Joshua Muskovitz
> # joshm at taconic.net
> def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
> lyyrs('Hire me!  I need the work!')
> 
> 
> 
> 
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----
> -- 
> http://mail.python.org/mailman/listinfo/python-list


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From wrbt at email.com  Mon Feb  4 10:52:30 2002
From: wrbt at email.com (Larry)
Date: 4 Feb 2002 07:52:30 -0800
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net> <MPG.16c760cdaa8ee3639896d5@newnews.dircon.co.uk> <PSi78.6815$3E5.522096@newsread2.prod.itd.earthlink.net> <slrna5s35r.h88.grante@isis.visi.com>
Message-ID: <2ec1bc1c.0202040752.60dc0bc4@posting.google.com>

believe it or not the big money is in tracking golf carts :)


From tdelaney at avaya.com  Thu Feb 14 18:39:34 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 15 Feb 2002 10:39:34 +1100
Subject: JAR equivalent in Gordon McMillan's Installer
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A188@natasha.auslabs.avaya.com>

> Hmm. Looks more messy than I'd hoped. Needs a bootstrap 
> module, and then
> extra hacking to handle open/read. And I guess the fact that I'm doing
> this manually means that there's not much chance of something like a
> wxPython toolbar "just working" with icons stored in the archive.
> 
> I guess it's not worth the hassle.
> 
> The reason I liked the Java approach, BTW, is that a complete
> application comes in a single file, which you can just double click on
> and it works. Nothing to unpack, put it anywhere you like, run off a
> floppy/CD/zip disk, no dependencies (other than needing Java 
> installed).
> But it's just a convenience thing - not enough to endure 
> writing all my
> code in Java :-)

http://www.jython.org/

Python implemented in Java. Has a standard system (jythonc) for compiling to
java classes and packaging them up as a jar file which does everything you
want.

In most cases you don't need to produce any java code at all - just code in
Python. It makes it *really* easy to use Swing too - much easier than in
Java.

Tim Delaney



From gh_pythonlist at gmx.de  Sat Feb 16 04:08:48 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Sat, 16 Feb 2002 10:08:48 +0100
Subject: Has anyone used Zorp????
In-Reply-To: <VNob8.383622$pa1.90623055@news3.rdc1.on.home.com>
References: <VNob8.383622$pa1.90623055@news3.rdc1.on.home.com>
Message-ID: <20020216090845.GA9937@lilith.hqd-internal>

Le 16/02/02 ? 08:31, Bo Vandenberg ?crivit:
> I'm looking to replace a firewall and I thought that a python configurable
> wall would be better than not.
> 
> I haven't heard anything about Zorp, which could be a good thing, so I was
> wondering if anyone here has played with it???

I've only very briefly played with it, when I was using Linux on the
gateway. I really liked the idea of an XML-configurable firewall, but in
the end, it wasn't worth it to me switching my relatively complex
iptables configuration over to .... oh wait. In this paragraph I talked
about DAXFi, not Zorp.

Yes, I've tried Zorp, too. IMHO it rocks. I also once tried DeleGate
(http://www.delegate.org/delegate/), which is also very nice.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 0.9 ?C      Wind: 2.4 m/s



From jeff at ccvcorp.com  Tue Feb 26 15:10:56 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 26 Feb 2002 12:10:56 -0800
Subject: MSWord and Python
References: <3C7A4EF8.4010404@thomas-guettler.de> <a5f00p$6udv4$1@ID-12869.news.dfncis.de> <3C7B4477.2040909@thomas-guettler.de>
Message-ID: <3C7BEBD0.ED91F25F@ccvcorp.com>


Thomas Guettler wrote:

> Ricky wrote:
>
> > The client side document is include in Python online documents.
>
> Sorry I didn't understand you. Is there are library for creating word
> documts without COM? This would be nice since I could create the
> documents on unix, too. What's its's name?

No, what he meant was that the pythoncom documentation includes some
examples of automating Word/Excel.  They're COM examples, though, and
will only work on Windows, just as Word and Excel will only work on
Windows.  (Oh, wait, there's Mac versions too -- do those have any
COM-like functionality?  I don't know...)  Anyhow, you will need to use
Windows/COM to do what you want.

If this is anything more than a simple one-off throwaway script, do
yourself a favor and buy "Python Programming on Win32", by Mark Hammond
and Andy Robinson.  It goes into a lot of detail on using COM from
Python, including extended examples of using Excel, and to a lesser
extent Word.  If you're going to be doing this often, or are doing a
non-trivial-sized project, it will be well worth the investment.

For details of Word's API/object model, you'll need to either see the
included help files, or MSDN.  This will be aimed at VisualBasic users,
but it's mostly pretty easy to translate.  There's also a COM browser
built in to PythonWin, which can give you some help, but it's a little
sparse (it just lists objects, methods, and parameters, you have to
figure out when/how to use them yourself, but it's good for
double-checking that method that you almost remember...)

Jeff Shannon
Technician/Programmer
Credit International




From piet at cs.uu.nl  Fri Feb  1 06:33:17 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 01 Feb 2002 12:33:17 +0100
Subject: Opening and reading randomly text file with umlauts and accents truncates output - unicode how?
References: <3C5A1564.9070202@uol.com.br>
Message-ID: <wzzo2t77ya.fsf@nono.cs.uu.nl>

>>>>> synthespian <synthespian at uol.com.br> (S) writes:

S> das Appartement, Appartements
S> das Auge, Augen
S> das Bad, B?der
S> das Bein, Beine
S> das Beispiel, Beispiele
S> das Buch, B?cher
S> das B?ro, B?ros
S> das Caf?, Caf?s

S> p = re.compile('^(der|die|das(\s\w+))')

If \w doesn't match the accented letters then probably your locale isn't
set properly.
A simple locale-independent solution would be to enumerate the characters
yourself, like [A-Za-z????...]
Or just use the fact that the words are terminated by a comma (or maybe
space)

p = re.compile('^(der|die|das(\s[^ ,]+))')

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From Mitch.Chapman at bioreason.com  Fri Feb  8 18:12:24 2002
From: Mitch.Chapman at bioreason.com (Mitch Chapman)
Date: Fri, 08 Feb 2002 16:12:24 -0700
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org> <00W88.108659$KM2.4246695@atlpnn01.usenetserver.com>
Message-ID: <3C645B58.7274695A@bioreason.com>

Steve Holden wrote:
>     http://www.holdenweb.com/python/py10keynote.pdf

Dangit.  This URL gives me 'HTTP/1.1 404 Object Not Found'.

-- 
Mitch Chapman
Mitch.Chapman at bioreason.com

Stuck on NT, without PPT


From jkraska at san.rr.com  Mon Feb  4 01:18:21 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 04 Feb 2002 06:18:21 GMT
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <mailman.1012745045.21128.python-list@python.org> <3c5d7e28_2@news.nwlink.com> <1012765918.861234@yasure> <3c5e12c4_1@news.nwlink.com>
Message-ID: <889s5u08er1gf6mbdn01a0ibndbr6qimrp@4ax.com>

>What I really want is some way to get anonymous functions that can be passed
>as part of a list of paramaters and called at the appropriate time.  There
>doesn't appear to be any way to get that particular effect in Python.

Well, there is lambda. It's just not fully featured, and this because of
complications in the way that Python manages whitespace and how that interacts
with the parser. Attempt the following in Python:

	for i in range(0,3): j=i; k=i

Now attempt this:

	for i in range(0,3): k=i; for j in range(0,3): m=j

Python won't let you do it. And for good reason. The statement is completely
ambiguous. There are similar problems with anonymous function definition in
Python, although they're not insurmountable.

I'm working on something similar in a play language, using ANTLR's lexer
_multiplexing_ capability. Certain kinds of blocks know how to multiplex
back and forth between whitespace-sensitive and insensitive forms. Using
this technique, I'm able to easily do exactly what you describe.

IIRC, it looks like this:

	f ( y,
		code( x ):
			return x
	)

Like Python (){}[] always imply no indent/dedent mechanics. Unlike Python,
the ':' block introduction multiplexes back to indent/dedent mode.

It works really well!

A more fun thing that I'm working on is a macro processor that is such that
the following could be a macros in the language:

	do:
		[statements]
	while [condition];

... or even...

	case [expression]:
	switch [expression]:
	switch [expression]:

C//



From dale at riverhall.NOTHANKS.co.uk  Mon Feb 25 11:17:35 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Mon, 25 Feb 2002 16:17:35 +0000
Subject: The good "import" convention?
References: <a50631f6.0202221235.5858bdcd@posting.google.com>
Message-ID: <2pok7u011kop7rd309133ho096ou65k6po@4ax.com>

Generally, I import everything at the top. Exceptions are when a
module is only used in special cases and it's large. Then I will
import it where it's needed but I added a comment at the top to note
that. 
--
Dale Strickland-Clark
Riverhall Systems Ltd


From lac at strakt.com  Fri Feb  8 09:27:52 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 08 Feb 2002 15:27:52 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary? 
In-Reply-To: Message from skywalkerpackage@hotmail.com (Johannes Eble) 
   of "Thu, 07 Feb 2002 18:36:29 GMT." <3c62b7ab.113923401@news.online.de> 
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org> <3c60fb80.199334@news.online.de> <mailman.1013016376.5624.python-list@python.org>  <3c62b7ab.113923401@news.online.de> 
Message-ID: <200202081427.g18ERqGo020985@ratthing-b246.strakt.com>

> Laura,
> 
> Thanks very much for the code. I think it is very helpful for me.

You are most welcome.

> I have analyzed your program and it seems to contain a lot of good
> tricks. 

Isn't python cool?

>However, I probably didn't understand the main principle
> behind. I suggest the best is if I wrtte some commends in between your
> code
> 
> On Wed, 06 Feb 2002 18:23:55 +0100, Laura Creighton <lac at strakt.com>
> wrote:
> 
> ...
> 
> >> This doesn't work together with an optionDict={'labelText':'hello',
> >> 'padx':'6'}
> >---------------------------------------------------^^^^^^^^
> >
> >That is your problem.  Tkinter labels want to get their text
> >set with text='hello', not labelText='hello'
> 
> Yes. Then, what I wanted to do define a configure (or a config) method
> that worked exactly as the one(s) in the Tkinter module.
> Unfortunately, I don't think I will understand the code inside Tkinter
> in limited time ;<

I think you are correct there.  
> 
> >
> >You have to make some design decisions here.  If you want to pass one big
> >dictionary, you will then have to unpack it, parse it, and generate
> >the correct keys for the correct widgets.  This is do-able; it is
> >what PMW does.  But I didn't like that approach.  I just use 2 dictionaries.
> 
> Why didn't you like the PMW approach?

If I did the pmw approach my code would be idiosyncratic, personal,
tailored for me, different by using names that I pick for the
dictionary.  Every time I got a new cool idea of how i might want
to configure a widget I would have to think up a new name and
write more code.  My approach just lets me use the native tkinter
way of doing things. 

One virtue of it was that I could haul this code out of a directory,
make 17 changes, test it, and post it here so you could have it.
This means that my code is relatively reusable, more so than you
would get if you made your less anonymous.

But I do this stuff without conscious thought.  So sometimes
when I do things like that its just the wrong approach.  But I
still like this.  Might be better to make it a factory function,
though.

What I lose is the ability to use keyword arguments fg='cyan' and the
like.  I thought it was worth it.

> 
> 2 dictionaries for what? Do you mean one master dictionary for your
> own megawidgets and for basic (Tkinter/Pmw) widget which appear in the
> higher level of your GUI, and one or more lowlevel dictionaries for
> the basic widgets which are contained by the highlevel widgets? I am
> not clear what kinds of the two dictionaries you mean (see below).

I mean  labelDict and widgetDict

> >Well, I don't know if you would call this elegent but ...
> >
> >
> >from __future__ import nested_scopes
> >import Tkinter
> >
> >class LabelledWidget(Tkinter.Frame):
> 
> Ok, this is an example of your own megawidget

ah, it _is_ the widget.
> 
> >    def __init__(self, parent, widgetName=Tkinter.Entry,
> >                 labelDict={}, widgetDict={}, **frameDict):
> >        
> >        Tkinter.Frame.__init__(self, parent, **frameDict)
> >
> >        """
> >        let us say that you always want the label to be written on a
> >        cyan background in purple letters, unless you ask differently.
> >        The widget you just want on a green background.
> >        """
> >        
> >        ldefaults={'bg': 'cyan', 'fg' : 'purple'}
> >        wdefaults={'bg': 'green'}
> >        # overwrite any of the defaults with the dictionary you passed
> >        ldefaults.update(labelDict)
> >        wdefaults.update(widgetDict)
> >       
> >        self.label = Tkinter.Label(self, **ldefaults)
> >        self.label.grid(row=0, column=0, sticky='w')
> >        self.widget = widgetName(self, **wdefaults)
> >        self.widget.grid(row=0, column=1, sticky='e')
> >
> >########
> >if __name__ == '__main__':
> >    root = Tkinter.Tk()
> >    widgetList = (
> >        [Tkinter.Entry,
> >         {'text':'An entryfield label:'},{}],
> >        [Tkinter.Button,
> >         {'text':'A button label:', 'bg':'peachpuff'},
> >         {'text':"I'm a button"}],
> >        [Tkinter.Label,
> >         {'text':'A label label:'}, {'text':"I'm a label", 'fg':'honeydew'}],
> >    )
> 
> This is probably one of your two kinds of dictionaries, right?
> It is specific for your megawidget, so, in general, you use one
> dictionary for the construction of your megawidget out of its basic
> (Tkinter) widgets.

{'text':'A label label:'}  -- that is dictionary 1
{'text':"I'm a label", 'fg':'honeydew'} -- that is dictionary 2

> 
> >    # for testing.
> >    def doit(widgetlist, part, method, **dict):
> >        for w in widgetlist:
> >            object = getattr(w, part, None)
> >            if object:
> >                command = getattr(object, method, None)
> >                if callable(command):
> >                    command(**dict)
> 
> This is a very interesting routine, but you can only use it together
> with a uniform widgetlist (that is, for example, a list containing
> only LabelWidget objects). Also, you can only apply the function to
> one subwidget (basic widget) and one subwidget method thereof.

What are you talking about? I used it with an Entry, a Button and a Label,
and it worked great.

This is just a quick bit of code for testing purposes, as the comment
said.  The name 'doit' is sort of a give away as well.

> 
> >
> >    r = 0
> >    widgets=[]
> >    for name, ld, wd in widgetList:
> >        w = LabelledWidget(root, name, ld, wd)
> >        widgets.append(w)
> >        w.grid(row=r, col=0, sticky='ew')
> >        r += 1
> >
> >    r = 0
> >    for w in widgets:
> >        w.grid(row=r, col=0)
> >        r += 1
> 
> Is there a reason why you have called the grid manager twice?

Sloppiness when I did a cut and paste. and later when I proofread
the article.

> 
> >
> >    cmdList = (
> >        ['pink label', 'label', 'configure', {'fg':'pink'}],
> >        ['yellow widget', 'widget', 'configure', {'bg':'yellow'}],
> >        )
> 
> Ok, I guess that is the second dictionary. It contains attributes of
> the megawidgets (that is, the basic widgets) and the options. Well, I
> can't see much more.

No. this is just standard boilerplate testing junk that I use for
testing widgets I make all the time.  There is nothing remarkable
about this, it is just that I write the tests first, and the code
later.  So I wanted to make sure that I sent you a widget that
was capable of doing this, so these were the tests I made first.

> 
> >
> >    c = 0
> >    for txt, part, method, dict in cmdList:
> >        button = Tkinter.Button(root, text=txt,
> >                                command=lambda w=widgets,
> >                                p=part, m=method, d=dict :
> >                                doit(w, p, m, **d))
> >        button.grid(row=r, col=c, sticky='w')
> >        c += 1
> >    button = Tkinter.Button(root, text='Exit', command=root.destroy)
> >    button.grid(row=r, col=c)
> >    
> >    root.mainloop()
> >--------------------------------------
> >
> >Laura Creighton
> >
> 
> Ok, what I try to do is the following:
> 
> I want to create a dialog box (a GUI window). This dialog box is
> divided into several panes. The panes do have widgets (basic widgets
> *and* composed widgets) in each side (left and right).
> 
> The structure of my dictionary (the only as you said is the case in
> Pmw) is as follows
> 
>     # lstPanes = [(pane1_text, 
> [(widget_pane1_left, {option_dict_pane1_left}),
>  (widget_pane1_right, {option_dict_pane1_right})
> ]                         ),
>                           (pane2_text,
> 
> .......
> 
> For example, the following list defines a dialog box with three panes.
> The first pane has two LabelEntry megawidgets (my own created) on the
> left and on the right, respectively. The second pane ('Compilation
> mode') consists of a radiobar (again my own widget, just a column of
> several radio bars) to the left and so on.
> >From the third pane it is clear that the widgets inside a pane can
> change and don't have to belong to one widget class. Also, the amount
> of columns is variable (one in the second pane and three in the third
> pane).
> 
> 
>     lstPanes = [('Root class definition', [
>                             (LabelEntry, {'labelText':'Root class'}),
>                             (LabelEntry, {'labelText':'Root
> procedure'})
>                             ]),
>                 ('Compilation mode', [
>                             (Radiobar2, {'radioTexts':['boost,
> 'no_check', 'all_checks, 'debug_check']})
>                             ]),
>                 ('Pane3', [
>                             (Checkbox, {....}),
>                             (LabelEntry, {'labelText':'label5'}),
>                             (RadioButton, {...})
>                             ])
>                 ]
> 
> 
> This scheme already works with Tkinter widgets in the columns of the
> panes, but not with my own megawidgets (neither with Pmw megawidgets).
> 
> I would just find it useful to be able to code a whole GUI window in
> this way, but perhaps it is not worth the effort after all.
> 
> 
> Johannes

I'm afraid I am not following you here.  If you post all the code and
the test cases you are using to not generate what you want I suspect
we could fix this in pretty short order.

Laura Creighton



From writeson at earthlink.net  Wed Feb 13 12:07:09 2002
From: writeson at earthlink.net (Doug Farrell)
Date: 13 Feb 2002 09:07:09 -0800
Subject: wxTextCtrl and handling button down even in OnChar()
Message-ID: <88bc63c6.0202130907.31a13d52@posting.google.com>

Hi,

I've written an application that has a control I've derived from
wxButton so I can provide my own EVT_BUTTON(self, ID, self.OnClick)
event handling. In the method OnClick() I'm getting the button press,
but I see both the button down and button up events, which causes my
code to run twice. I'd like to filter the button press and only react
to the button down event. The trouble is, I can't seem to find the
method of event or whatever that will allow me to do this. If any one
has some ideas how I can process only moust button down events, I'd
really appreciate it.

Thanks,
Doug Farrell


From jeremy at alum.mit.edu  Mon Feb  4 23:59:39 2002
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: 4 Feb 2002 20:59:39 -0800
Subject: RELEASED - StandaloneZODB 1.0 release candidate 1
References: <mailman.1011990406.22651.python-list@python.org> <7876a8ea.0201290640.5f194252@posting.google.com>
Message-ID: <b0f083db.0202042059.3f9b47e6@posting.google.com>

woodsplitter at rocketmail.com (stalin) wrote in message news:<7876a8ea.0201290640.5f194252 at posting.google.com>...
> > Today we are releasing StandaloneZODB 1.0 release candidate 1.  We
> > expect to release the final version on 1-Feb-2002.
> 
> StandaloneZODB-1.0-rc1 + Standard Python 2.2 + Visual C++ 6.0 + WinNT 4.0-SP6a:
> 
> "python setup.py build" -- works fine
> "python setup.py test" -- throws this exception
> 
> Traceback (most recent call last):
>   File "lib.win32-2.2\ZEO\tests\winserver.py", line 61, in ?
>     main(sys.argv[1], sys.argv[2], sys.argv[3:])
>   File "lib.win32-2.2\ZEO\tests\winserver.py", line 50, in main
>     klass = load_storage_class(storage_name)
>   File "lib.win32-2.2\ZEO\tests\winserver.py", line 45, in load_storage_class
>     package = __import__("ZODB." + name)
>   File "lib.win32-2.2\ZODB\FileStorage.py", line 133, in ?
>     from fsIndex import fsIndex
>   File "lib.win32-2.2\ZODB\fsIndex.py", line 36, in ?
>     from BTrees._fsBTree import fsBTree as _fsBTree
>   File "lib.win32-2.2\BTrees\__init__.py", line 11, in ?
>     import Interfaces
>   File "lib.win32-2.2\BTrees\Interfaces.py", line 98, in ?
>     class IDictionaryIsh(IKeyed, Interface.Standard.MinimalDictionary):
> AttributeError: 'module' object has no attribute 'MinimalDictionary'

We don't expect StandaloneZODB users to execute the code in
BTrees.Interfaces.
The BTrees package __init__ imports BTrees.Interfaces and the
Interface module.  We don't ship the Interface module with
StandaloneZODB, and expect that the import of Interface will fail. 
__init__ is prepared to catch the ImportError and move on.

Do you have an Interface module installed somewhere on your
PYTHONPATH?  In site-packages?

Jeremy


From kp87 at lycos.com  Mon Feb 25 12:24:27 2002
From: kp87 at lycos.com (kevin parks)
Date: 25 Feb 2002 09:24:27 -0800
Subject: Iterators and generators for the masses
Message-ID: <5e8bd451.0202250924.70f47df1@posting.google.com>

A note in my Python 2.2 folder says that:

Iterators (PEP 234) and generators (PEP 255) were added.

When I ask questions on the tutor list and also this list i get
answers that sometimes start:

...We can do this with Python's new generators...
...Python's new iterators make this easy...

Numerous searches turned up many complex PEPs, rewritten PEPs and 
wild drag-down knock out USENET geek fights over whether or not these
were
good and should be added to Python. Nothing I could make heads or
tails of.

Myself I am just getting used to seeing list comprehensions...

1> any chance that there will be tutorials on these important new
features added to www.python.org?(a *real* tutorial on list
comprehensions would be nice too, but at least there is documentation
on them on www.python.org)

2> Anyone up to explaining them here? (hopefully in a way that someone
other than Tim Peters and Guido understands! Insult me if you must --
talk to me like I am a 12 year old! Why? Because there are 12 year
olds out there learning Python and those of us with the
math/science/programming chops of a 12 year old who are also trying to
understand this stuff).

what do they do? what can they do, what tasks do they make easier,
what is the simplest possible example, what is a more complex
example...etc. & co.

cheers,
kevin parks
seoul, korea


From magnus at thinkware.se  Fri Feb 22 10:59:00 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 16:59:00 +0100
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org> <7xit8rhebq.fsf@ruckus.brouhaha.com> <slrna7a6sv.ekg.scarblac@pino.selwerd.nl> <7xvgcqpz29.fsf@ruckus.brouhaha.com>
Message-ID: <3C766AC4.3040500@thinkware.se>

Paul Rubin wrote:

> Oh come on, the most common errors in C are probably pointer errors.

Ok, so it's just the second most common error. I've
certainly written "if a = b" when I meant == in C++.
Especially when I'm switching between C++ ans SQL...

Also, making expressions return their value to the
left like that breaks: 4 < x < 8

In python that returns true for x in [5, 6, 7]. If
it worked like you want it would be either (4 < x) < 8
or 4 < (x < 8), and youd have to write interval tests
as (4 < x) and (x < 8)



From logiplexsoftware at earthlink.net  Tue Feb 19 20:36:33 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 19 Feb 2002 17:36:33 -0800
Subject: Python and GIS
In-Reply-To: <3C72F52F.5050901@sympatico.ca>
References: <3C72F52F.5050901@sympatico.ca>
Message-ID: <20020219173633.6da14f22.logiplexsoftware@earthlink.net>

On Tue, 19 Feb 2002 20:00:31 -0500
Ramkumar Kashyap wrote:

> Hi,
> 
> I was wondering if there are any modules for GIS, Mapping, Layering, or 
> hooks to Arcinfo, Mapinfo, etc.

A quick search on Google turned up:

http://www.geocities.com/brucedodson.rm/avpython.htm

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From joshm at taconic.net  Mon Feb 11 17:21:55 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 11 Feb 2002 17:21:55 -0500
Subject: Standing Objects ... possible?
References: <mailman.1013420066.23115.python-list@python.org>
Message-ID: <3c6841b6_4@corp.newsgroups.com>

*pop*  (head explodes)

After I scoop up all the extraneous grey matter, I'll be amazed and
impressed.  Cool!

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From aahz at panix.com  Fri Feb  8 18:55:45 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 15:55:45 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org>
Message-ID: <a41oi1$a22$1@panix2.panix.com>

In article <mailman.1013196740.22073.python-list at python.org>,
 <James_Althoff at i2.com> wrote:
>[Aahz]
>>
>>Haven't figured out a good/cheap way to put more than two monitors
>>on a desk, particularly when both are 17" or larger.  It's hard
>>enough stuffing *one* monitor on most crapola office desks.
>
>Fry's Electronics has been selling a *really* nice 18-inch flat panel
>display by NEC.  Same amount of actual display area as many 20 or 21 inch
>video monitors and totally flat, lightweight, space saving, etc.  The
>prices have come down remarkably over the past year.  I bought 4 of them
>for my house over the holidays (yes, I *have* been accused of going
>overboard on occasion, thank you. <wink>).

Two point:

* Horizontal space counts for a lot -- you still can't easily set up
more than two flat-panel displays on most desktops.  Heck, the desk I'm
currently using has a 19" monitor, but it certainly couldn't handle two
15" flat-panel monitors.

* Yes, I'm aware that prices have come down a lot, but the equivalent
flat-panel display is still twice as expensive.  And I *won't* buy
anything at Fry's.

I have to admit that I'm seriously thinking about an iMac, though.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From jkraska at san.rr.com  Sun Feb 10 13:03:00 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 10 Feb 2002 18:03:00 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com> <slrna67k0u.for.philh@comuno.freeserve.co.uk> <br186u8qg3bu2hpsf0dcgs1ldih7gir4rg@4ax.com> <slrna6bfie.o7n.philh@comuno.freeserve.co.uk>
Message-ID: <jndd6ugqesuqibt2vssrfg6l368mj64mtk@4ax.com>

>So I wouldn't be able to do that macro expansion. This means the macro
>processor is conceptually more complex, and therefore more difficult
>to learn and probably has more bugs in it too.

Potentially. However, hygienic macro systems are better known for
introducing less comprehensional complexity. Likewise, it's usually
easier to integrate them with the debugger. Often, they are implemented
by doing something like inserting into the parse-tree directly, which
affords the opportunity to record both expansion-point as well as
source-origin line numbers and line positions. So while these things
need to be unit tested like any other element of the language, once
done, they are fare less likely to _introduce_ hard-to-find bugs into
the user's code.

And note that in any case, the question of whether or not any macro
system is strictly _hygienic_ is distinct from its first-class
implementation.

C//



From tim.one at comcast.net  Thu Feb 28 20:36:48 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 28 Feb 2002 20:36:48 -0500
Subject: Status of PEP's?
In-Reply-To: <slrna7tis4.1m1j.Gareth.McCaughan@g.local>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEGFOAAA.tim.one@comcast.net>

[Gareth McCaughan]
> ...
> (My own preferences, not that anyone asked for them:
> I always rather liked the range literal proposal, but
> GvR rejected that.

I did too -- until I applied the patch and used it for a couple hours.
That's part of how it ended up getting rejected; another part was that the
patch author confessed he didn't really care for it either.  One annoyance
was that

    for i in [0:len(a)]:

was actually more clumsy to type than

    for i in range(len(a)):

and even after two intense hours I just couldn't stop from writing it

    for i in [len(a)]:

instead -- which also has a meaning, but not one I had in mind.  Negative
bang for the buck isn't Pythonic, at least not when we can help it.

> I also like the "for 3 <= i < 10" one, which I think is *beautiful*
> but maybe slightly too cute.)

I'd rather iterate over i than over 3 <wink>.




From kp87 at lycos.com  Wed Feb 13 11:27:50 2002
From: kp87 at lycos.com (kevin parks)
Date: 13 Feb 2002 08:27:50 -0800
Subject: filtering random values (sieves?)
Message-ID: <5e8bd451.0202130827.2df6f50b@posting.google.com>

I am not sure I am even using the right terminology here, but I
looking for a way to filter random values. Let's say that I am using
random and randrange to generate values from 0-11, but i only want to
accept values 1, 3, 5, 7, 11, (that is just off the top of my head,
could be anything) how would i filter out values that i don't want?
The only way i can think of is just asking the program to choose
again, but that may not be the best way since, at least
hypothetically, you could keep choosing values that are rejected...

Any help would be appreciated.

-kevin


From richard at NOstarfighterSPAM.freeuk.com  Mon Feb 25 14:08:39 2002
From: richard at NOstarfighterSPAM.freeuk.com (Richard Townsend)
Date: Mon, 25 Feb 2002 19:08:39 -0000
Subject: installing and finding PMW
References: <3C79A250.B24F6CC5@alum.mit.edu>
Message-ID: <1014663916.3564.0@eos.uk.clara.net>

> I've installed pwm in c:\program
> files\gary\python22\lib\site-packages\pmw
> sys.path seems to include this directory.  I've added this directory to
> PATH and PYTHONPATH.
> I notice that there is no file called pmw.py --- how does import work in
> this case?
>
> Where should I put PMW????   How should the environment variables be
> set???

I just copied the Pmw directory to the top level python directory
(C:\Python21\Pmw in my case) - it was not necessary to set any environment
variables.

regards,
Richard







From usenet at thinkspot.net  Mon Feb 25 18:25:55 2002
From: usenet at thinkspot.net (Sheila King)
Date: Mon, 25 Feb 2002 15:25:55 -0800
Subject: Website to post code for peer review
References: <mailman.1014676907.27964.python-list@python.org>
Message-ID: <a5dl24.3vs4847.1@kserver.org>

On Mon, 25 Feb 2002 17:42:43 -0500, Ben Jacobs <dooberwah at yahoo.com> wrote
in comp.lang.python in article
<mailman.1014676907.27964.python-list at python.org>:

> I looking for a good web site where I can submit the code I
> write to be looked at and reviewed by other programmers. What I have in mind
> is something like http://www.perlmonks.org, but for python related stuff
> (please don't flame me because I like perl as well as python).
> 	PerlMonks is a large web site where perl hackers can hang our, swap
> code, and chat about perl or programming in general. It allows people to
> reply and vote on code you write. People can gain experience points and
> different levels depending on how many votes their code gets.
> 	comp.lang.python offers some of these of thing, but from my brief
> stay here I've felt that it's mostly a place to get your programming
> questions answered.

There is no "community" site in the sense of the perlmonks.org site (AFAIK)
Most Python discussion takes place on mailing lists or in newsgroups rather
than on web sites.

The closest you may be able to come to what you're looking for at this
time, is a combination of
Useless Python: http://www.lowerstandard.com/python/whatsthepoint.html
and a subscription to the Python-Tutor mailing list:
http://mail.python.org/mailman/listinfo/tutor

I don't know about any "voting" or "experience points", but you can get
your code reviewed and discussed.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From mwh at python.net  Thu Feb 21 12:10:26 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 17:10:26 GMT
Subject: Simple regexp problem
References: <3C75289B.8050504@nyu.edu>
Message-ID: <ueljelpvz.fsf@python.net>

Shane Hoversten <srh232 at nyu.edu> writes:

> Hi all,
> 
> 	This question has gotta be real simple.  I want to remove
> preceeding whitespace from a string.  Seems like I should be able to
> say (using another example to make it easier to see):
> 
> re.sub("^1", "", "11111junk")

re.sub("^1*", "", "11111junk")

Maybe you should have another look at the docs...

But you do know about .lstrip(), right?

Cheers,
M.

-- 
  ARTHUR:  Don't ask me how it works or I'll start to whimper.
                   -- The Hitch-Hikers Guide to the Galaxy, Episode 11


From donn at u.washington.edu  Fri Feb  1 13:09:43 2002
From: donn at u.washington.edu (Donn Cave)
Date: 1 Feb 2002 18:09:43 GMT
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com>
Message-ID: <a3ell7$29bm$1@nntp6.u.washington.edu>

Quoth Jonathan Hogg <jonathan at onegoodidea.com>:
| On 1/2/2002 9:37, in article 4b4c5d99.0202010136.8e6ff51 at posting.google.com,
| "pawn" <NOSpawnPAM at lightspawn.org> wrote:
|> I need my Python CGIs to run scripts under my own uid, since they read
|> and update .html files which are core to the site.
|> 
| [...]
|> 
|> * Switch back to perl which respects 04755 permissions.
|
| I don't think the lack of respect for setuid permissions is anything to do
| with Python I'm afraid. Quick test on my machine:
|
| -----
| % cat > test.py
| #! /usr/bin/python
|
| import os
|
| print os.geteuid(), os.getegid()
| ^D
| % chmod 755 test.py
| % ./test.py
| 501 20
| % sudo chown 71:71 test.py
| % ./test.py 
| 501 20
| % sudo chmod 4775 test.py
| % ./test.py 
| 71 20
| % sudo chmod 6775 test.py
| % ./test.py 
| 71 71
| -----

Can I ask what operating system that experiment was performed on?

...
| Note that setting an executable's setuid bit will change the "effective" uid
| of the resulting process, not the uid (see 'os.get*e*uid()' above). This
| means that the process may use that user's permissions.

That's true, but "executable" in this context usually means the binary,
i.e., python itself.  Years ago, some UNIX platforms introduced a hack
that made setuid effective for script files, but today that is universally
recognized to have been misguided.  That's why I'm curious about yours.

Perl is subject to the same limitation, but has optional support for
setuid.  How can that be?  I'm no Perl whiz, but I assume you install
perl itself setuid root, and then it checks its input file for setuid
and either sets its ID accordingly or reverts to the invoker's ID.
This option is not the default and is rather discouraged as I recall it.

	Donn Cave, donn at u.washington.edu


From mhammond at skippinet.com.au  Thu Feb 21 23:23:10 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 22 Feb 2002 04:23:10 GMT
Subject: Embedding Python in C#
References: <3c75b88c$1_5@news1>
Message-ID: <3C75C828.9000306@skippinet.com.au>

Adonis Vargas wrote:
> has this been done? is it even possible?
> if it has where can i obtain information on doing so?
> 

The simplest way is probably to create a Python COM object, then have C# 
reference it like any other "legacy" COM object.

As //C comments, other techniques may also work.

Mark.



From emile at fenx.com  Wed Feb 13 11:50:23 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Feb 2002 08:50:23 -0800
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <a4e44s$tli$1@netnews.upenn.edu>
Message-ID: <a4e5ql$1e6fkl$1@ID-11957.news.dfncis.de>

"Paul Hughett"
> It could be worse.  In one language I worked in, "10" was an octal
integer,
> "10." was a decimal integer, and "10.0" was floating point.  Nice
language
> otherwise, though.
>

Was "9" invalid?

--

Emile van Sebille
emile at fenx.com

---------



From boyle5 at llnl.gov  Wed Feb 13 11:09:32 2002
From: boyle5 at llnl.gov (jboyle)
Date: 13 Feb 2002 08:09:32 -0800
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com>
Message-ID: <9871078e.0202130809.3bd60a46@posting.google.com>

WEC7108 <WEC7108 at boeing.com> wrote in message news:<3C6968A4.1CA9C62E at boeing.com>...
> I would like to develop a GUI program similar to the ever-popular BOIDS
> using Python instead of Java.  Are there URL references or books on this
> kind of development?  :o)  Bill

There is a Boids implementation in python. You can find it on the
Vaults of Parnassus, it is called pyboids. It uses threads to control
multiple flocks. I used it to learn about Tk and threads.


From tismer at tismer.com  Mon Feb 25 10:15:35 2002
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 25 Feb 2002 16:15:35 +0100
Subject: Time to rename Stackless? (Re: Stackless Platform Independence?)
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org> <7xd6yt6btq.fsf@ruckus.brouhaha.com>
Message-ID: <3C7A5517.7050306@tismer.com>

Paul Rubin wrote:

> Christian Tismer <tismer at tismer.com> writes:
> 
>>>Maybe something like "Microthreaded Python" or
>>>"Multitasking Python" would be better?
>>>
>>We have now tasklets, proclets, and the stack slices
>>are stacklets. :)
>>
> 
> What about generators (yield statement)?  Since you have microthreads,
> is it now ok to yield through multiple levels of function calls?


Sure.
I can't support the yield statement, since this is occupied
by the new built-in generators.
But it is easily possible to use tasklets as generators.
They can be used like the ICON coexpressions, from anywhere.
Switching to and from such a generator is slightly more
expensive, probably like 4 calls to a builtin primitive.

ciao - chris
-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From djrassoc01 at mindspring.com  Wed Feb 20 11:24:41 2002
From: djrassoc01 at mindspring.com (Dr. David J. Ritchie, Sr.)
Date: Wed, 20 Feb 2002 10:24:41 -0600
Subject: Tkinter and grid: strange alignment with and without sticky option
References: <9ff6b66.0202120802.3f20bcaf@posting.google.com>
Message-ID: <3C73CDC6.F110D0C@mindspring.com>

Yes, I see the same thing.  My Python version is:

Python 2.1.1 (#97, Aug  2 2001, 21:53:31)  [CW PPC GUSI2 THREADS] on mac

The Tkinter is the one that came with the Python on the CDROM for
O'Reilly's "Learning Python"
as is the copy of Python.

Running a slightly modified version of the original poster's example
listed below,
I find that the non-sticky red Listbox is horizontally
misaligned by a small amount (probably 1 pixel)
from the following non-sticky
Entry box, the sticky blue Listbox
and the sticky Entry box.
---

from Tkinter import *

def main():
    root = Tk()
    Listbox(root, width = 20, relief = RIDGE, font = "Courier",
        bg = "red").grid(row = 0)
    Entry(root, width = 20, relief = RIDGE, font = "Courier") \
        .grid(row = 1)
    Listbox(root, width = 20, relief = RIDGE, font = "Courier",
        bg = "blue").grid(row = 2, sticky = W)
    Entry(root, width = 20, relief = RIDGE, font = "Courier") \
       .grid(row = 3, sticky = W)
    root.mainloop()
    return
if __name__=="__main__":
    main()

---

--D.

vtail wrote:

> Hi, guys,
>
> I have very odd problem with using grid() with and without sticky
> option. Here is my python version:
>
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>
> I'm using Tkinter that ships with standard distribution.
>
> Consider the following snipet:
> ----

>
> snip

> ---
>
> The code in if..else.. block only differs in whether or not we are
> using a "sticky = W" option
>
>

--
Dr. David J. Ritchie, Sr.
djrassoc01 at mindspring.com
http://home.mindspring.com/~djrassoc01/




From bkelley at wi.mit.edu  Tue Feb 26 09:15:51 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Tue, 26 Feb 2002 09:15:51 -0500
Subject: Dictionary type access to list of instances
References: <a5dcgh$b22$1@ctb-nnrp1.saix.net> <3C7A5AD4.9090404@wi.mit.edu> <a5fedu$qe1$1@ctb-nnrp1.saix.net>
Message-ID: <3C7B9897.4020507@wi.mit.edu>

Grant Beasley wrote:

> What I'm ultimately aiming for is a database-like view of a list of
> instances, so that I can use SQL commands to get instances. So in the
> analogy, a list of instances would be a table, with each instance a record,
> and each property a field.
> 
> eg.
> class User:
>     def __init__(self, username, accesslevel):
>         self.username = username
>         self.accesslevel = accesslevel
> 
> list = [User('Bob', 1), User('Jim', 2), User('Pete', 2)]
> 
> Then using my imaginary function ListSQL:
> 
> ListSQL("select * from list where accesslevel = 2")
> would return a list consisting of the Jim and Pete instances.
> 
> or ListSQL("select username, accesslevel from list where accesslevel = 2")
> would return [('Jim', 2), ('Pete',2)] - i.e. a List of tuples.
>


> Is this an interesting / useful idea? Has it been done before? Any comments?

You might want to check out database objects 
http://www.xs4all.nl/~bsarempt/python/dbobj.html

or gadly, a python relational database
http://www.chordate.com/kwParsing/gadfly.html#table

I have had a great deal of success with metakit which I endorse heavily
http://www.equi4.com/metakit/python.html

and I willing to bet Zope has a solution in this realm as well...
http://www.zope.org/

These might help from not reinventing the wheel.

Brian Kelley



From tiago_castro_henriques at yahoo.com  Mon Feb 25 15:29:22 2002
From: tiago_castro_henriques at yahoo.com (Tiago Castro Henriques)
Date: 25 Feb 2002 12:29:22 -0800
Subject: Question on Python path
References: <3qnk7ug977hk4peorufobtlo38nh8j9v87@4ax.com>
Message-ID: <43901295.0202251229.760a4e46@posting.google.com>

Gon?lo Rodrigues <op73418 at mail.telepac.pt> wrote in message news:<3qnk7ug977hk4peorufobtlo38nh8j9v87 at 4ax.com>...
> I have put all my own modules in a subdirectory of site-packages. How
> can I instruct Python (I'm using Python 2.2 on win2k) to also look up
> this directory when a
> 
> import <whatever>
> 
> is found?
> 

Try using the following or similar:
sys.path.insert(0, "C:/Program Files/Apache Group/Apache/cgi-bin")


From nas at python.ca  Sun Feb 10 11:51:30 2002
From: nas at python.ca (Neil Schemenauer)
Date: Sun, 10 Feb 2002 08:51:30 -0800
Subject: Standalone ZODB + BTree question
In-Reply-To: <cc498.2133$0c.92062@news2.east.cox.net>; from cgale1@cox.net on Sat, Feb 09, 2002 at 07:28:08AM +0000
References: <cc498.2133$0c.92062@news2.east.cox.net>
Message-ID: <20020210085130.A24953@glacier.arctrix.com>

Van Gale wrote:
> I've been able to figure out FileStorage and DbmStorage easy enough from the
> docs, and I can create and use an OOBTree... I just can't seem to tie it
> into persistent storage.

Is your BTree reachable from the root object?

  Neil



From tanzer at swing.co.at  Wed Feb 13 02:49:05 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 13 Feb 2002 08:49:05 +0100
Subject: Defending the Python lanuage...
In-Reply-To: Your message of "Tue, 12 Feb 2002 20:02:48 +0100."
             <200202121902.g1CJ2mGo029604@ratthing-b246.strakt.com>
Message-ID: <E16auA5-0002KC-00@tswings.swing.cluster>

> > >Have you read 'Engineering and The Minds Eye'?  I just finished it.
> > >Wonderful.
> >
> > How about _Things That Make Us Smart_?
> > --
> >                       --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)
>
> Do you mean Donald Norman's book?  I haven't.  I have read
> 'The Design of Everyday Things' though.

If you liked that one, you should definitely get _Things That Make Us
Smart_. I'd also recommend _The Invisible Computer_.

Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From lac at strakt.com  Tue Feb 12 14:02:48 2002
From: lac at strakt.com (Laura Creighton)
Date: Tue, 12 Feb 2002 20:02:48 +0100
Subject: Defending the Python lanuage... 
In-Reply-To: Message from aahz@panix.com (Aahz Maruch) 
   of "11 Feb 2002 14:23:37 PST." <a49g99$bt4$1@panix2.panix.com> 
References: <3C62EE4A.5090305@destiny.com> <200202081311.g18DBQGo020260@ratthing-b246.strakt.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <mailman.1013420657.25827.python-list@python.org>  <a49g99$bt4$1@panix2.panix.com> 
Message-ID: <200202121902.g1CJ2mGo029604@ratthing-b246.strakt.com>

> >me:
> >
> >Have you read 'Engineering and The Minds Eye'?  I just finished it.
> >Wonderful.  
> 
> How about _Things That Make Us Smart_?
> -- 
>                       --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

Do you mean Donald Norman's book?  I haven't.  I have read
'The Design of Everyday Things' though.  

Laura



From aliocha_z at caramail.com  Sun Feb 17 05:13:04 2002
From: aliocha_z at caramail.com (Aliocha)
Date: Sun, 17 Feb 2002 11:13:04 +0100
Subject: Apache windows server : python module
References: <a4m865$80v$1@wanadoo.fr> <a4mq7d$th2$1@news.doit.wisc.edu>
Message-ID: <a4nvnd$28j$1@wanadoo.fr>

Ok , thanks.




From michael.tu at reachasia.net  Thu Feb 14 11:43:21 2002
From: michael.tu at reachasia.net (Michael Tu)
Date: 14 Feb 2002 08:43:21 -0800
Subject: Beginner question please help!
Message-ID: <28eed654.0202140843.510948d5@posting.google.com>

Hello,

I am trying to do the following exercise but after spending couple of
hours without any luck, I think it is time to get some helps. Thank
you in advance.

Write a compare function that return 1 if x > y, 0 if x == y, and -1
if x < y.

Mike


From vtail at yandex.ru  Wed Feb 27 12:01:27 2002
From: vtail at yandex.ru (vtail)
Date: 27 Feb 2002 09:01:27 -0800
Subject: proxy authentication: how to hide password
References: <9ff6b66.0201250109.1c68a130@posting.google.com> <899f842.0202120818.11412d4b@posting.google.com>
Message-ID: <9ff6b66.0202270901.3837af8b@posting.google.com>

anthony_barker at hotmail.com (Anthony_Barker) wrote in message news:<899f842.0202120818.11412d4b at posting.google.com>...
> 
> But don't feel bad - Perl has a similar issue - but uses an
> environment variable instead. I also use wget - this uses the .wgetrc
> (yet another location for the password).

I _feel_ bad - Perl didn't ask me any password when I run similar
script. Neither did I defined any environment variables.

> prompt for the password if need be.

That's so painful to enter your login&password every time the script
is running... ;-(

Anyway, thanks for you ideas.

Vtail


From klaatu at evertek.net  Tue Feb 26 14:47:54 2002
From: klaatu at evertek.net (Mike Dean)
Date: Tue, 26 Feb 2002 13:47:54 -0600
Subject: Editors for Python
Message-ID: <20020226134754.B9558@evertek.net>

* Andy Todd <andy47 at halfcooked.com> [2002-25-02 23:22]:
> I can't believe anyone could hate vim, its the best editor in the world ;-)

He didn't say he hated *vim*, just *vi* (I suppose there isn't much
difference, but...).  It's possible to use Vim without being a vi
person.

So, check out EasyVim (haven't used it much, and I personally don't like
it, but it does allow one to use Vim's GUI without being exposed to its
VI near heritage as much; you can use Vim much as you would a 'normal'
GUI editor).  Vim has syntax highlighting, and there is an add-on script
available at vim.sf.net that creates a menu of the class and function
definitions in the Python script you are working on (it uses ctags to do
this).

My $.02 for what it's worth...
Mike



From henrik.motakef at web.de  Sun Feb 10 09:43:11 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 10 Feb 2002 15:43:11 +0100
Subject: Python IDEs
References: <zfC08.15504$Hg7.1623775@news11-gui.server.ntli.net> <a1uv39$o79$1@rex.ip-plus.net> <3c664c2b@nntp.server.uni-frankfurt.de> <867kplznbt.fsf@apo.hmotakef.int> <NTu98.19162$H9.2428975@news1.cableinet.net>
Message-ID: <863d09z9c0.fsf@apo.hmotakef.int>

John Ochiltree <jo at johnochiltree.uklinux.net> writes:

> > AFAIK, there is Python (and other non-C++ languages) support in the
> > CVS version of Gideon (the successor of KDevelop). So no need to take
> > the time yoursef.
> Is there a URL for Gideon, the Gideon on SourceForge isn't it.

Sorry, AFAIK there isn't anything except some threads on various
KDE(velop)-related mailing lists (http://lists.kde.org). So would have
to try it - it's the kdevelop module in KDE's CVS. Unfortunatly, it
depends on QT 3 (and probably on KDE 3 also, but I didn't go further).

However, after looking at the sources (I didn't bother to compile it
yet), it looks like it would support projects in C++, Java, Python,
Perl and PHP, and is scriptable using Python (yay!). The only
documentation yet is on writing plugins, so one could expect more
languages to follow.

BTW: Anybody tried Eclipse yet? Is there Python support already?



From andy47 at halfcooked.com  Fri Feb  1 04:31:43 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Fri, 01 Feb 2002 20:31:43 +1100
Subject: locale on cygwin
References: <Xns91A1A678DEEE3andy47halfcookedcom@203.109.250.24> <20020125140805.GC1252@dothill.com>
Message-ID: <3C5A607F.6060107@halfcooked.com>

Jason Tishler wrote:

> Andy,
> 
> On Fri, Jan 25, 2002 at 05:22:02AM +0000, Andy Todd wrote:
> 
>>Jason Tishler probably knows the answer to this off the top of his head,
>>but ... 
>>
> 
> Not really, being locale challenged...
> 
> 
>>I'm having problems getting 'locale' to work with Python under cygwin.
>>
>>[snip]
>>
>>But on the same machine with Cygwin python I get;
>>
>>Python 2.2 (#1, Dec 31 2001, 15:21:18)
>>[GCC 2.95.3-5 (cygwin special)] on cygwin
>>Type "help", "copyright", "credits" or "license" for more information.
>>
>>>>>import locale
>>>>>locale.getdefaultlocale()
>>>>>
>>(None, None)
>>
>>>>>locale.setlocale(locale.LC_ALL, '')
>>>>>
>>'C'
>>
>>I presume this means that the _locale module is not present on Cygwin, or
>>compiled in with Python.
>>
> 
> No, it's there:
> 
>     /usr/lib/python2.2/lib-dynload/_locale.dll
> 
> but...
> 
> 
>>Is there any particular reason why? 
>>
> 
> Yes, IIRC, Cygwin does not currently support locale.  Searching the
> Cygwin mailing list I have found the following:
> 
>     http://sources.redhat.com/ml/cygwin/2001-11/msg00458.html
> 
> Which seems to concur.  Sorry, this is the best that I can do.  Maybe
> others know for sure.
> 
> Jason
> 
> 
> 

Thanks for the input Jason, but the plot thickens ;-)

I've expanded my research to Linux and am getting some funny results 
there. I've got Python 2.1.1 on Debian testing (2.2) and with KDE 
everything is hunky dory;

"""
angua:/home/andy47# python
Python 2.1.1+ (#1, Jan  8 2002, 00:37:12)
[GCC 2.95.4 20011006 (Debian prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
 >>> import locale
 >>> dir(locale)
['CHAR_MAX', 'Error', 'LC_ALL', 'LC_COLLATE', 'LC_CTYPE', 'LC_MESSAGES', 
'LC_MONETARY', 'LC_NUMERIC', 'LC_TIME', '__all__', '__builtins__', 
'__doc__', '__file__', '__name__', '_build_localename', '_group', 
'_parse_localename', '_print_locale', '_setlocale', '_test', 'atof', 
'atoi', 'encoding_alias', 'format', 'getdefaultlocale', 'getlocale', 
'locale_alias', 'localeconv', 'normalize', 'resetlocale', 'setlocale', 
'str', 'strcoll', 'strxfrm', 'sys', 'windows_locale']
 >>> locale.LC_ALL
6
 >>> locale.setlocale(locale.LC_ALL, '')
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
   File "/usr/lib/python2.1/locale.py", line 374, in setlocale
     return _setlocale(category, locale)
locale.Error: locale setting not supported
 >>>
angua:/home/andy47# echo $LANG
en_AU
angua:/home/andy47#
"""

But with gnome I get very similar results to Cygwin;

"""
angua:/home/andy47# python
Python 2.1.1+ (#1, Jan  8 2002, 00:37:12)
[GCC 2.95.4 20011006 (Debian prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
 >>> import locale
 >>> dir(locale)
['CHAR_MAX', 'Error', 'LC_ALL', 'LC_COLLATE', 'LC_CTYPE', 'LC_MESSAGES', 
'LC_MONETARY', 'LC_NUMERIC', 'LC_TIME', '__all__', '__builtins__', 
'__doc__', '__file__', '__name__', '_build_localename', '_group', 
'_parse_localename', '_print_locale', '_setlocale', '_test', 'atof', 
'atoi', 'encoding_alias', 'format', 'getdefaultlocale', 'getlocale', 
'locale_alias', 'localeconv', 'normalize', 'resetlocale', 'setlocale', 
'str', 'strcoll', 'strxfrm', 'sys', 'windows_locale']
 >>> locale.LC_ALL
6
 >>> locale.setlocale(locale.LC_ALL, '')
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
   File "/usr/lib/python2.1/locale.py", line 374, in setlocale
     return _setlocale(category, locale)
locale.Error: locale setting not supported
 >>>
angua:/home/andy47# echo $LANG
en_AU
angua:/home/andy47#
"""

I'm still confused, and the exception raised by Gnome is causing me a 
right pain. Anyone else like to chip in?

Regards,
Andy
-- 
-----------------------------------------------------------------------
 From the desk of Andrew J Todd esq.
"Another year older, still no wiser." - Me, on my birthday




From peter at engcorp.com  Tue Feb 26 00:21:19 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 26 Feb 2002 00:21:19 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu> <mailman.1014687041.32731.python-list@python.org> <a5eqvu$79a$1@news.cis.ohio-state.edu>
Message-ID: <3C7B1B4F.400E35C2@engcorp.com>

Locke wrote:
> 
> so you are saying it sends the data in groups? how do i know how many calls
> to recv i need to do before the server is done sending everything it needs
> to send? because if i call recv after the server is done sending everything,
> the program will just deadlock waiting. any suggestions? anybody know
> anything about pop3?

You need a protocol between client and server with termination sequences
demarking the end of a block.  Things like POP3 (which you can readily find
lots of information on via google... try "POP3 RFC" or something like that)
use a command/response interaction with each part of the exchange terminated
with CR/LF.  Other protocols might use something different.  You could
try sending the length ahead of time and waiting only for the appropriate
amount of data (harder, perhaps more efficient or lower latency).

Also, if you are worried about "deadlock" (it's really just 'blocking') 
you can turn blocking off and put your client recv() in a loop with
a timeout or something... there are many ways of handling these things.
Looking at example code will teach you more.

-Peter


From david.abrahams at rcn.com  Tue Feb 26 19:44:05 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Tue, 26 Feb 2002 19:44:05 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net> <6bf54244.0202260920.5920efc6@posting.google.com> <3C7BCCA5.604@physics.ucsb.edu>
Message-ID: <a5h9qn$a3h$1@bob.news.rcn.net>

"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
news:3C7BCCA5.604 at physics.ucsb.edu...

> Actually, after doing a little more digging, I found this:
>
> http://public.kitware.com/Cable/
> http://public.kitware.com/GCC_XML/
>
> Does anyone know anything about this project?

Another promising but incomplete entry, last time I looked (it still
couldn't represent the full C++ language in its XML). The mailing list
archives indicate that's still the case, though things may be changing:
http://public.kitware.com/pipermail/gccxml/2002-February/000029.html

> This is what I would have guessed would be the "right" way to go about
> things (having known nothing about the EDG front-end).  Presumably gcc
> gets the parsing right if it can compile ;)

Even the EDG parser has bugs. However, if your code compiles with GCC 3.0,
gccxml should be able to handle it (note that GCC is up to 3.0.3 and 3.1
will be released soon, so things have improved further since gccxml was
developed).

-Dave





From aleax at aleax.it  Fri Feb  1 17:36:20 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 22:36:20 GMT
Subject: Conventions for iterators
References: <mailman.1012555082.6918.python-list@python.org> <U2A68.17631$3J1.540244@news2.tin.it> <a3emcv$17igt2$1@ID-11957.news.dfncis.de>
Message-ID: <ELE68.18874$3J1.587028@news2.tin.it>

Emile van Sebille wrote:

> "Alex Martelli" <aleax at aleax.it> wrote in message
> news:U2A68.17631$3J1.540244 at news2.tin.it...
>> _timbot_ in a typo -- dies albo signanda lapillo...!-)].
> 
> Translation please.  You know we can only read dutch here... ;-)

Literally "A day with a white to be marked stone" (Latin's
freedom in word-order is at times amazing...:-) -- reordering
for our smaller modern minds, "A day to be marked with a
white stone" (=="a peculiarly memorable occasion").


Alex



From usenet at thinkspot.net  Fri Feb  8 14:50:50 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 08 Feb 2002 11:50:50 -0800
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <41914a.537.ln@wombat.san-francisco.ca.us>
Message-ID: <a40e2r.3vvegs9.1@kserver.org>

On Fri, 8 Feb 2002 11:30:44 -0800, kkeller at speakeasy.net (Keith Keller)
wrote in comp.lang.python in article
<41914a.537.ln at wombat.san-francisco.ca.us>:

> In article <a417rn$ge8$1 at panix2.panix.com>,
> 	aahz at panix.com (Aahz Maruch) writes:
> > I know some of y'all still keep roots to the Perl community, so I was
> > wondering if there's any kind of definitive nickname for Perl
> > programmers, the way we call ourselves Pythonistas?
> 
> I don't think so--having one nickname would contradict TMTOWTDI.  :)

I thought it was Perl Monks.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From phd at phd.pp.ru  Tue Feb  5 08:54:38 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 5 Feb 2002 16:54:38 +0300
Subject: directly executing a python script in Unix
In-Reply-To: <60018f02.0202050547.4836c5ff@posting.google.com>; from laymon@crd.ge.com on Tue, Feb 05, 2002 at 05:47:47AM -0800
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com> <60018f02.0202050547.4836c5ff@posting.google.com>
Message-ID: <20020205165438.D8004@phd.pp.ru>

On Tue, Feb 05, 2002 at 05:47:47AM -0800, Marc Laymon wrote:
> (laymon) [29] ls -l retest.py
> -rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*
> 
> I can duplicate the error message with a perl script by changing
> "#!/usr/bin/perl" to "#!/usr/bin/foo", so it seems that for some
> reason, Unix can't find python, but since /usr/bin is in my path
> and I am explicitly giving the command "usr/bin/python",
> I can't figure out why.  Anybody have any other ideas ?

   Check, if python binary is executable by you. Check that the line
"#!/usr/bin/python" contains only these symbols and nothing else. For
example, UNIX python does not like DOS line endings (CR+LF). UNIX python
prefers only UNIX line endings (LF).

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



From jon+usenet at unequivocal.co.uk  Mon Feb 25 18:03:35 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Mon, 25 Feb 2002 23:03:35 -0000
Subject: help: Problem with cgi form
References: <Xns91BE6246576BAAusYourStandingInIt@130.133.1.4> <3C7AAF31.8040201@mxm.dk>
Message-ID: <slrna7lgm7.1kf.jon+usenet@snowy.squish.net>

In article <3C7AAF31.8040201 at mxm.dk>, Max M wrote:
> And then print it to the form like:
> 
> print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % last_byte_pos

Beware that, in general, that is a bug. You are open to Cross Site
Scripting attacks. Even in the absence of these, if the value happens
to contain characters significant to HTML (such as '"', '&', '>', etc)
then your program will go wrong.


From writeson at earthlink.net  Mon Feb  4 08:09:49 2002
From: writeson at earthlink.net (Doug Farrell)
Date: 4 Feb 2002 05:09:49 -0800
Subject: wxPython and Boa-Constructor
Message-ID: <88bc63c6.0202040509.74258657@posting.google.com>

Hi all,

I have Python 2.2 and wxPython 2.3 installed on my machine. I've also
been curious about using Boa-Constructor to generate some code and
look at what it generates as a tutorial to me on what to do. I've
downloaded the latest Boa-Constructor, but it doesn't like the version
of Python/wxPython that I've got installed. Does anyone know what I
might have to do to get it running or when it might be upgraded again?

Thanks
Doug Farrell


From johnroth at ameritech.net  Thu Feb 14 07:58:31 2002
From: johnroth at ameritech.net (John Roth)
Date: Thu, 14 Feb 2002 07:58:31 -0500
Subject: Recommendations for cross-platform GUI Bindings ?
References: <xBMa8.217$B92.52511@news.xtra.co.nz>
Message-ID: <u6ncsqng8e2c4c@news.supernews.com>

"Steven M. Castellotti" <SteveC at innocent.com> wrote in message
news:xBMa8.217$B92.52511 at news.xtra.co.nz...
> Greeting all--
>
> I'm writing an Open Source software application that allows children
to
> create their own video games, and would like to hear anyone's
> recommendations for GUI bindings. The key things I'm looking for are:
>
>  - Cross Platfrom (I need to run on at least Linux and Windows, though
> Mac would be great if possible.)

Have you looked at either tkinter or anygui? Tkinter is availible and
mature on all platforms you suggest, and anygui has the stated
purpose of being both cross-platform and simple.

>
>  - Compiles to something easily executable (Ideally can be run from
the
> web, or can at least be wrapped by an installer)

Anygui is completely python, so it should execute anywhere
that python will. It will use the native GUI in most cases (I believe
they have put using the Mac Gui on the back burner for now, however).

Likewise, Tkinter has wide distribution, and should run most places
that python does.

>  - Has good Image/Graphics routines
>
>
>  - Is well supported (Online documentation/examples are available, or
at
> least an active mailing list. I don't want to get stuck coding for a
> target the dissapears after the developer gets bored)
>
>  - Uses Python or python-esqe syntax.
>
>
>
> So far I'm looking at:
>
> Java 1.2 (Using Jython)
> Java 1.4 (Using Jython)
> pyGTK
>
>
> Java 1.2 is nice because most current web browsers support it, and
Swing
> has some nice features, but saving and loading games to the hard drive
is
> tough without a paid-for signed applet.
>
> Java 1.4 is the most current version of Java, and
> I'm told includes some nice ImageIO routines, but isn't in web
browsers
> yet, and I don't want to have to ask kids to upgrade their software.
> There's still the problem of saving and loading to the hard drive.
>
> pyGTK is what I'm currently using, but GnomeCanvas, the most usful way
> to manipulate on-screen images is not available under win32. In theory
> a pyGTK program can be compiled into a .exe for win32 distribution,
but I
> haven't worked out the details yet.
>
>
>
> If anyone has any recommendations as to technologies I should look at,
I
> would be very interested to hear your thoughts. If Mono was further
along
> on the Linux side, I might consider targetting .Net for win32
development
> ride now, but I like to code under linux and am willing to wait to
move
> onto Mono when its ready.
>
>
> Cheers,
>
> -Steve Castellotti
>  SteveC at Innocent.com
>  http://cogengine.sourceforge.net/




From bkelley at wi.mit.edu  Wed Feb 20 17:15:50 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Wed, 20 Feb 2002 17:15:50 -0500
Subject: Editable strings anyone?
References: <mailman.1014178114.23674.python-list@python.org>
Message-ID: <3C742016.9030507@wi.mit.edu>

Jason Orendorff wrote:

> Brian Kelley wrote:
> 
>>What would it take to make array.array or array.ArrayType an acceptable 
>>base type for subclassing?
>>
> 
> Wow, you're right, this is a mess.  I will put together a patch
> for it.  Quite a few little things are involved, but it's mostly
> just a modernization of the way this module uses the Python C API.
> 
> That's not the kind of thing that's likely to be really
> well-documented anywhere, unfortunately.  :P
> 
> ## Jason Orendorff    http://www.jorendorff.com/


Cool.  I'll buy you a beer at the next python conference :)

Brian Kelley




From jacobkm at cats.ucsc.edu  Fri Feb  1 03:35:31 2002
From: jacobkm at cats.ucsc.edu (Jacob Kaplan-Moss)
Date: Fri, 01 Feb 2002 08:35:31 GMT
Subject: Python and TK possible with MacOS X ?
References: <3C59A411.9020103@uni-muenster.de>
Message-ID: <jacobkm-0102020035380001@192.168.0.120>

Hendrik --

One way to do it is by installing XDarwin (XFree86 for OS X) and the *nix
versions of Python and Tkinter.  You can do this quite easily using Fink
<http://fink.sf.net>.  I've writen a quick tutorial on how to get a
working Tkinter on OS X.  You can find it at:

<http://people.ucsc.edu/~jacobkm/tkinter_osx_howto.html>

If you have any questions, please feel free to email me.

Jacob

In article <3C59A411.9020103 at uni-muenster.de>, Hendrik Maeder
<maederh at uni-muenster.de> wrote:

> Hi everyone,
> 
> I`d love to use python to create some GUI apps for MacOS X.
> 
> I downloaded a standard python 2.2 and compiled it, however, using 
> Tkinter is not possible under OS X, it needs an X server like XDarwin. 
> Unfortunately this is not an option, I need native-OS-X-TK... :-(
> 
> MacPython unfortunately has Tkinter support only under MacOS <9.x. The 
> Carbon version (OS X) just says "no module named Tkinter".
> 
> However, there IS a native recent TK port available from Apple.
> Seems to consists of some dynamic libraries, nothing more.
> Does anyone know if it is possible to get this to work with OS X?
> 
> Or does anybody know some workaround or other portable modules for GUI 
> programming that work with OS X?
> 
> Thank you very much,
> Hendrik


From gyromagnetic at excite.com  Wed Feb 13 15:32:09 2002
From: gyromagnetic at excite.com (gyromagnetic)
Date: 13 Feb 2002 12:32:09 -0800
Subject: files, temp files, and file-like objects
Message-ID: <4620daca.0202131232.4754675@posting.google.com>

Hi,
I am building an application in which many of the properties of
objects are computed based on results produced by a number of external
applications. Each of these compiled (C or Fortran) applications is
usually run from the command line with a generic syntax like
   command -i input file -o outputfile_basename

In the process of 'doing their stuff', each application creates a slew
of temporary and subsidiary output files. My objects' properties
(numbers) are usually taken from one or two of the output files, while
the rest of the files are 'clutter' for my purposes.

My question is 'how should I handle the many, many files that are
involved?'
I will be creating hundreds of objects, with each one potentially
creating a dozen files in the process. Is there some kind of file-like
objects that can be used instead of real files that could 'fool' the
applications?

I would appreciate any suggestions that you can give me.

Thanks.

-gyro


From threeseas at earthlink.net  Sat Feb  9 11:24:57 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 09 Feb 2002 16:24:57 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1372.801T1070T13994028threeseas@earthlink.net> <3c60c05c$0$79564$3c090ad1@news.plethora.net> <2106.802T789T14243611threeseas@earthlink.net> <3c647f02$0$22477$3c090ad1@news.plethora.net>
Message-ID: <3665.805T682T7313596threeseas@earthlink.net>

On 08-Feb-02 20:44:34 Peter Seebach <seebs at plethora.net> wrote:
>In article <2106.802T789T14243611threeseas at earthlink.net>,
>Timothy Rue <threeseas at earthlink.net> wrote:
>>C# is a subtopic of .NET / Mono, as is many other programing languages,
>>including Python.

>Unless, of course, you use that to build *your* framework.  They need
>options.  Give them some.

Who needs options?

>>The community you speak of is not inclusive of the end user/consumer, but
>>rather a community of programmers that are likley to follow MS mindset
>>programming.

>I think it's safe to say they're the community most likely to accept your
>vision.

Or the community most likely to mimic it and try and shoot it down. MS has
all the bucks to bullshit the public into believing what ever it is they
want the public to believe. Embrase and extend right?

>>If you have never done it all the way right, how will you know when you
>>do?

>How would you know?  You never really know; you simply observe that your
>current model covers everything you've seen.

???? You assume what I've presented is a model that I've come up with and
based on only what I've seen? You are greatly mistaken and in such, you
also expose a bit of C# mindset. If you are to be an example of the
Acceptance of what I present, then you are only accepting what you want
to believe, not what is.

I don't like MS and I don't have a C# compiler for any of my systems and I
am not going to sign any damn NDA or anti-gpl or such nonsense just so I
can spend more of my time and money supporting some another damn overly
complicated programming language that the typical consumer doesn't have
the time or resources or interest to deal with.

There are over three thousand programming language. The VIC is not
programming language dependant and as such I don't see any reason for me
to learn another language, because there is always going to be someone
that says what you do about some other language. I'm sure the VIC could be
written in many different languages, so? What is important is having it
written in a language that is widely (cross platform) available and freely
available. C# doesn't fit that spec. but it doesn't mean the VIC can't be
used to generate C# code.

I'm in/on usenet where there is an unbiased date time stamped archiving
machinery attachability (important in legal matters). I'm not hiding from
anyone, not even C# and MS. And I'm fully aware that People such as Steve
Jobs and Bill Gates do in fact access such, though usually behind a mask
for more than for the reason of celebraty issues.

BUT YOU HAVEN'T NAMED A USENET C# NEWSGROUP, HAVE YOU?

I have no fear of going into such a newsgroup. But then that is not what
you are suggesting, is it? You really want me to just leave here and maybe
go behind some damned closed door mailing list that has no inherent
assurence that it can provided any unbiased date time stamped public
accessible archive.

Name your favorite C# usenet newsgroup. If you pick a moderated one, then
you pick a censored one, and you will not know the truth from it.

I have no reason to believe what I present will have any different effect
on such a group of people as it apparently has here. You should think
about that.

But lets find out! Name a unmoderated Usenet C# newsgroup.



---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jason at jorendorff.com  Sun Feb 24 22:33:41 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 24 Feb 2002 21:33:41 -0600
Subject: Suggestion for best intro to Python
In-Reply-To: <c510a321.0202210608.1acc4d29@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOENIDEAA.jason@jorendorff.com>

> What book/web site/online document/something else would be best as a
> source to learn Python for an experienced programmer?

If you already know a few languages, start here:
  http://www.python.org/doc/current/tut/tut.html

This is how I learned.
Four hours with the tutorial = instant expert.
You can't learn that fast with a book.

Enjoy Python.

## Jason Orendorff    http://www.jorendorff.com/



From jimd at vega.starshine.org  Wed Feb 27 05:30:01 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 27 Feb 2002 10:30:01 GMT
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de> <3C7ADE12.C6D895A0@engcorp.com>
Message-ID: <a5icf9$165$1@news.idiom.com>

In article <3C7ADE12.C6D895A0 at engcorp.com>, Peter Hansen wrote:
> Emile van Sebille wrote:

>> "Joshua Muskovitz" <joshm at taconic.net> wrote in message
>> news:3c7ab974_4 at corp.newsgroups.com...
>>> Hey all,

>>> If you had to give a 15 minute introduction to Python, what would you
>> cover?
>>> What is a good example of a simple, but useful, text-based app for a
>> Wintel environment, that has good "wow" quality?

>> Hmm... text-based... "wow" appeal... oxymoron?  ;-)
>> Otherwise, what Sheila said.

> Something text-based that retrieves some useful information
> from the web (a weather site?) and parses the result and
> displays something useful with a few lines of Python code
> using the standard modules would have enough "wow" appeal
> for anyone used to other languages...

> At least, it "wow"ed me when I did this sort of thing
> from the interactive prompt for the first time... :)

> -Peter

 I agree with the first comments that ask: 

  	What is the purpose of your demonstration?
	What is the background of the audience?

 From what you've said, you're giving a presentation to a 
 group of (high school?) teachers.  Is this to assess your
 skills as an instructor?  Are they considering the suitability
 of Python for introductory programming classes? Are these 
 teachers familiar with modern programming (or are they stuck
 in the academic traditions of BASIC, and Pascal)?  Are they 
 likely to know about the Fibonacci series and it's traditional
 role as a very basic first term programming assignment?


 One cool "wow" demo script that I just whipped up is a simple
 threaded URL checker in about 50 lines of code.  This isn't fancy, 
 and I'm sure the local NG wizards can improve it immensely, but it 
 works and it makes for an impressive demonstration of how simply
 one can use Python's networking, HTTP/HTML and threading modules,
 and basic exception handling though it shows only the most rudimentary 
 data types (just a couple of lists).

 Here's the code:

#!/usr/bin/env python2.2
## Based on code from ASPN Python Cookbook (by Andy McKay)
##   http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/101276
## JimD <jimd at starshine.org>: Wed Feb 27 02:26:08 PST 2002
##	- hacked to take command line args and be multi-threaded

from httplib import HTTP 
from urlparse import urlparse 
import threading 

class URLCheckResults:
	def __init__(self):
		self.lock=threading.Lock()
		self.failure=[]
		self.success=[]
	def succeed(self,url):
		self.lock.acquire()
		self.success.append(url)
		self.lock.release()
	def fail(self,url):
		self.lock.acquire()
		self.failure.append(url)
		self.lock.release()

class URLChecker(threading.Thread):
	def __init__(self, results, url):
		threading.Thread.__init__(self)
		self.url=url
		self.results=results
	def run(self): 
		print "start:\t%s" % repr(self)
		# Gratuitous tuple assignment/unpacking
		# <scheme>://<netloc>/<path>;<params>?<query>#<fragment>
		(x, host, path, x, x, x) = urlparse(self.url)
		try:
			h = HTTP(host) 
			h.putrequest('HEAD', path) 
			h.endheaders() 
			if h.getreply()[0] == 200: self.results.succeed(self.url)
			else: self.results.fail(self.url)
		## except (IOError):
		except: 	# Warning! catchall!
			self.results.fail(self.url)
		print "end:  \t%s" % repr(self)

if __name__ == '__main__': 
	import sys
	results = URLCheckResults()
	threads = []
	for each in sys.argv[1:]:
		threads.append(URLChecker(results, each))
		threads[-1].start()
	for each in threads:
		each.join()
	for each in results.success:
		print "Reached: " + each
	for each in results.failure:
		print "Problem: " + each
	sys.exit(len(results.failure))

 
 It's pretty simplistic, but I was able to demonstrate it, and
 explain it to my wife (a programmer and webmaster, with no
 Python experience and just a little armchair book knowledge of
 its syntax) in under ten minutes.
 
 I think it's "cool" because it mixes a few hot buzzy programming
 needs into a small, fairly short space.  It shouldn't take much
 imagination for your audience to think of ways that a few hundred
 lines of additional code could classify the connection results
 into more categories based on other h.getresults() and more
 specific exceptions, use GET requests instead of HEADs, parse 
 results, URL encode arguments (form data) for GETs or POSTs, 
 etc.  You can spur those notions by mentioning these other module
 features in passing as you walk through the code example.

 This code also has some deliberate flaws, things that would be 
 worth discussing in a class as targets for refactoring.




From just at xs4all.nl  Thu Feb 28 05:36:59 2002
From: just at xs4all.nl (Just van Rossum)
Date: Thu, 28 Feb 2002 11:36:59 +0100
Subject: dirwalk.py generator version of os.path.walk
References: <a5d36e$1daf$1@news.idiom.com> <ac677656.0202271621.5a134d44@posting.google.com> <a5kv49$ae8$2@news.idiom.com>
Message-ID: <just-BED145.11365928022002@news1.xs4all.nl>

In article <a5kv49$ae8$2 at news.idiom.com>,
 jimd at vega.starshine.org (Jim Dennis) wrote:

> > I wrote a different implementation of this general concept at:
> 
> > http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/105873
> 
> > You don't really need to keep a stack of directories and push/pop
> > things, because with generators you can recurse instead.
> 
> >Tom
> 
>  But recursion is likely to cost more.  The only state I need
>  to keep is my current "todo" list of directories.  A recursion 
>  would store functional state (unless Python supported tail-end
>  recursion).  So the append/pop (total cost, 3 lines of code)
>  seems like the lightest weight way to do this.

Why don't you try it out? Setting up a generator is as expensive as 
calling a function, but resuming or yielding a generator is _very_ cheap.

I personally find the recursive version clearer.

Just


From jimd at vega.starshine.org  Mon Feb 25 05:21:02 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 25 Feb 2002 10:21:02 GMT
Subject: dirwalk.py generator version of os.path.walk
Message-ID: <a5d36e$1daf$1@news.idiom.com>

 This function could probably use a bit of polishing,
 and it certainly could use some enhancement (some options to
 control if, and how we follow symlinks, to how to handle 
 exceptions on listdir(), whether to be depth first, and an 
 option to avoid crossing mount boundaries with os.path.ismount(), 
 etc).

 However, it seems to work.  

 dirwalk() simply takes an optional top level directory/path name
 as an argument and instantiates a generator which will walk down
 that tree and return every filename that it can access.  

 It's late and I need sleep.  So I'm just going to post this in
 it's rough (and probably buggy) form and let y'all thrash on it
 a bit.

 I guess there's some sort of statcache module that might let me
 cache the stat() tuples.  I guess I'm implicitly incurring a stat()
 system call for each node by checking islink() and isdir() on it
 so it seems like I ought to cache that and make it available to 
 my caller (without forcing them to make an additional stat system
 call).

 I hope that something like this (a simple dirwalk() or other 
 greatly simplified alternative to os.path.walk()) makes it into 
 Python 2.3 or later.

#!/usr/bin/env python2.2
from __future__ import generators 
import os

def dirwalk(startdir=None):
	if not startdir:
		startdir="."
	if not os.path.isdir(startdir):
		raise ValueError ## Is this the right exception?
	stack = [startdir]
	while stack:
		cwd = stack.pop(0)
		try:
			current = os.listdir(cwd)
		except (OSError):
			continue	# Skip it if we don't have access
		for each in current:
			each = os.path.join(cwd,each)
			if os.path.islink(each): 
				pass
			elif os.path.isdir(each):
				stack.append(each)
			yield(each)

if __name__ == "__main__":
	# import unittest?
	# test suite should consist of:
	# 	dirwalk() vs. os.listdir()
	# 	dirwalk("/") vs. os.path.walk()
	# 	dirwalk("/etc/passwd") (should raise exception)
	import sys
	for i in sys.argv[1:]:
		for j in dirwalk(i):
			print j
	# should compare this to os.popen("find ....") and
	# or to os.path.walk(...)



From logiplexsoftware at earthlink.net  Fri Feb 22 16:10:28 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 22 Feb 2002 13:10:28 -0800
Subject: wxlayoutconstraints inside a wxDialog ?
In-Reply-To: <3c76b920.39590390@news.easynews.net>
References: <3c76666e.18420093@news.easynews.net>
	<Xns91BDC97DAC9B9rcamesz@amesz.demon.nl>
	<3c76b920.39590390@news.easynews.net>
Message-ID: <20020222131028.2f143217.logiplexsoftware@earthlink.net>

On Fri, 22 Feb 2002 21:41:28 GMT
Marcus Stojek wrote:

> I did what you said, but still the main effect is there. When the
> dialog pops up, all elements are crowded at the upper left corner.
> If I resize the dialog or add the line 
> 
> self.SetSize((355,205))
> 
> everything jumps in place. The Doc says if I don't use 
> SetAutoLayout i have to overwrite the OnSize() function
> and call Layout() myself. Fine. But I AM using SetAutoLayout
> here.

Try calling Layout() explicitly at the end of your __init__ method.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From phr-n2002a at nightsong.com  Tue Feb 26 02:50:54 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 23:50:54 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org>
Message-ID: <7x3czobs0x.fsf@ruckus.brouhaha.com>

"Brian Quinlan" <brian at sweetapp.com> writes:

> Paul Rubin wrote: 
> 
> [snipped]
> > Python has grown from a tiny acorn to a healthy young tree.  I believe
> > any changes and additions to its design should be made with the idea
> > that it may become a mighty oak one day, i.e. that it will have
> serious
> > native-code optimizing compilers.   Language design changes should
> work
> > toward that goal, or at least not impede it.
> 
> Is optimization the only purpose of your proposals? If that is the case,
> they are probably not needed. The Python compiler already identifies
> variables that have local-only scope. Presumably other compilers could
> do the same.

Optimization is one of the goals of type advice.  Error checking is
another: in an appropriate runtime mode, it's similar to having an
assert statement every place the variable is set.

Local variable declarations are for code cleanliness.  Currently, using
some name for a temp variable leaves detritus around after you're done
with it.  To figure out what the program is doing, you have to check
whether further references to 'temp' really wanted the value left from
someplace far away, or were holding just a different temporary value.
When the usefulness of a variable ends, its scope should also end.


From gh_pythonlist at gmx.de  Wed Feb  6 15:37:46 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Wed, 6 Feb 2002 21:37:46 +0100
Subject: get all paths
In-Reply-To: <d5534716.0202061219.60f2af7e@posting.google.com>
References: <d5534716.0202061219.60f2af7e@posting.google.com>
Message-ID: <20020206203745.GA2482@lilith.hqd-internal>

Le 06/02/02 ? 12:19, Ed ?crivit:
> Hi
> I'm trying to create a treeview widget, but i am having problems
> getting alll directories and files from a specified directory
> 
> i.e
> 
> getList("C:/") should return all directories and files from C:/
> onwards whereas
> getList("C:/Python") should return all from that onwards

You can use the following functions: os.listdir(), os.path.isdir(),
os.path.isfile().

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.4 ?C      Wind: 3.7 m/s



From martin at v.loewis.de  Tue Feb 12 19:47:29 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 01:47:29 +0100
Subject: SystemError: unknown opcode
References: <3C696D1C.AE3F87BE@snakefarm.org>
Message-ID: <m3heomw6la.fsf@mira.informatik.hu-berlin.de>

Carsten Gaebler <clpy at snakefarm.org> writes:

> I just got an exception from a script that is regularly run via cron.
> The script is supposed to choose a random signature file, which it
> does perfectly well. Just every now an then I receive the following
> error. I'd be glad if anyone could explain what's happening.

That's hard to explain; this is not supposed to happen.

One theory is that there is a hardware problem with your computer's
memory (or your hard disk), making it present the byte code
differently in memory than it is on disk. I've also seen file system
drivers stomp on the buffers they read from disk (Linux reiserfs in
particular).

It could be a bug in Python also, but that is unlikely :-)

Regards,
Martin



From cmedcoff at home.com  Thu Feb 21 17:59:28 2002
From: cmedcoff at home.com (Charles Medcoff)
Date: 21 Feb 2002 14:59:28 -0800
Subject: Problems building DCOracle 1.3.2 under Win2K/Cygwin
Message-ID: <1a241b9d.0202211459.394c3e64@posting.google.com>

I'm just following the instructions in the src/readme.txt file.  It
appears that a file is missing from the distribution.  Any advice
appreciated.

$ make
gcc -DUSE_DL_IMPORT -g -O2 -Wall -Wstrict-prototypes
-I/usr/local/include/python
2.1 -I/usr/local/include/python2.1 -DHAVE_CONFIG_H  -I/rdbms/demo
-I/rdbms/publi
c -I/network/public -I/plsql/public -DDCORACLE8 -c ././oci_.c -o
./oci_.o
././oci_.c:565: ociapr.h: No such file or directory
In file included from ././oci_.c:697:
oci_8.c:67: oci.h: No such file or directory
make: *** [oci_.o] Error 1

Regards,
Chuck


From mal at lemburg.com  Sat Feb 23 05:20:10 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 23 Feb 2002 11:20:10 +0100
Subject: FILEDSN
References: <c13db1aa.0202220643.37578e1c@posting.google.com> <Ujtd8.84991$n73.1594161@atlpnn01.usenetserver.com>
Message-ID: <3C776CDA.70197FED@lemburg.com>

Steve Holden wrote:
> 
> "ilgattodelvicino" <ilgattodelvicino at hotmail.com> wrote in message
> news:c13db1aa.0202220643.37578e1c at posting.google.com...
> > Can i use a file DSN with odbc module??
> >
> Sorry, you can anly use file DSNs with ADO components, as far as I know.
> ODBC database systems need to pick up the appropriate ODBC driver for the
> particular database instance you have to work with.

You can use FILEDSNs with ODBC as well. See the mxODBC documentation
for how this is done.

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From daves_spam_dodging_account at yahoo.com  Thu Feb 28 14:05:02 2002
From: daves_spam_dodging_account at yahoo.com (David Brady)
Date: Thu, 28 Feb 2002 11:05:02 -0800 (PST)
Subject: FW: python-list, FOR YOUR PRIVATE EYES ONLY !!!!
In-Reply-To: <AFE3FBA300B6F04282BE07949DB2E12303F3C776@TORINO.corp.es.com>
Message-ID: <20020228190502.40177.qmail@web21101.mail.yahoo.com>

> -----Original Message-----
> From: Tim Peters [mailto:tim.one at comcast.net]
> 
> Do you think an organization as shadowy as the PSU
> would just come out and
> say "Please join the PSU!"?  No.  In fact, 90% of
> the "porn spam" on the web
> is actually

Tim,

I'm worried that the PSU may be censoring some of the

=====
David Brady
daves_spam_dodging_account at yahoo.com
I'm feeling very surreal today... or *AM* I?

__________________________________________________
Do You Yahoo!?
Yahoo! Greetings - Send FREE e-cards for every occasion!
http://greetings.yahoo.com



From robin at jessikat.fsnet.co.uk  Wed Feb 27 16:53:07 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 27 Feb 2002 21:53:07 +0000
Subject: PIL vs eps
References: <mailman.1014835309.1869.python-list@python.org>
 <just-B1FF7B.22204827022002@newszilla.xs4all.nl>
Message-ID: <HRCwzIADVVf8EwB0@jessikat.fsnet.co.uk>

In article <just-B1FF7B.22204827022002 at newszilla.xs4all.nl>, Just van
Rossum <just at xs4all.nl> writes
>In article <mailman.1014835309.1869.python-list at python.org>,
> <brueckd at tbye.com> wrote:
>
>> Forgive me if I'm totally off my rocker, but I think by definition eps
>> files should *not* have a showpage and it is technically correct that
>> nothing gets rendered unless you have a showpage command. EPS is
>> "encapsulated postscript" - postscript that you intend to encapsulate or
>> embed in another document, so showpage would muck things up.
>
>The EPS spec says showpage is optional, and that the surrounding ps code 
>should redefine showpage as a no-op.
>
>  http://partners.adobe.com/asn/developer/pdfs/tn/5002.EPSF_Spec.pdf
>
>Just
so I guess that EPSPlugin isn't going to be able to handle some eps
files. Indeed the sample .eps that came with ghostscript didn't have a
showpage.
-- 
Robin Becker


From joachim.kaeber at imk.fraunhofer.de  Tue Feb 19 05:09:14 2002
From: joachim.kaeber at imk.fraunhofer.de (Joachim Kaeber)
Date: Tue, 19 Feb 2002 11:09:14 +0100
Subject: confusion regarding os.path.walk()
References: <mailman.1014078032.30409.python-list@python.org> <Xns91BA65E93C0ECandrewdarwinwarscom@194.168.222.19>
Message-ID: <3C72244A.5704081A@imk.fraunhofer.de>

Hi,

Andrew Brown wrote:
> The practical question I was left with was "how do you identify broken
> symlinks with python?" I'd have liked a script that rm-ed broken and only
> broken symlinks, and I can't figure out how to test for them.
> os.path.islink(file) will tell me whether it's a symlink. But is there a
> call to say what it's supposed to point at? Then I can test whether that
> exists.

Maybe os.stat is your friend:

% ln -s /dev/null a
% ln -s /dev/xxxx b

Python 2.1 (#1, May 17 2001, 11:31:53)
[GCC 2.95.3 19991030 (prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import os
>>> os.stat("/tmp/a")
(8630, 8, 6L, 1, 0, 0, 0, 0, 0, 0)
>>> os.stat("/tmp/b")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OSError: [Errno 2] No such file or directory: '/tmp/b'

HTH
-- 
Joachim Kaeber                Fraunhofer Institut Medienkommunikation
Schloss Birlinghoven                   phone: +49 (0) 2241 - 14 25 46
D-53754 Sankt Augustin                   fax: +49 (0) 2241 - 14 24 49
Germany                                  http://www.imk.fraunhofer.de


From robin at jessikat.fsnet.co.uk  Fri Feb 15 04:08:08 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 15 Feb 2002 09:08:08 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk>
 <3C6C4C37.6020209@skippinet.com.au>
Message-ID: <t4oqZKA4$Mb8Ew3k@jessikat.fsnet.co.uk>

In article <3C6C4C37.6020209 at skippinet.com.au>, Mark Hammond
<mhammond at skippinet.com.au> writes
>Robin Becker wrote:
>> I am having bad problems with zope not starting up properly. I think
>> this is caused by having Python-2.1.2 + win32all installed. It seems
>> that paths set up in the registry by win32all are getting into sys.path
>> when zope runs. This is causing massive confusion between modules
>> installed in the base python and those I wish to test with zope.
>
>IMO, the zope install should be fixed.  The Zope installer should decide 
>if it wants to work with an installed Python OR it wants its own, 
>discrete Python.
......
>If the latter (which is more professional) it should just take steps to 
>ensure that it is partitioned correctly.  Generally, this should only 
>mean changing *one* string in the compiled Python DLL, which will 
>completely change the registry key used by Python to load sys.path. 
>Obviously the installer itself then changes to reflect this new key, but 
>that should be all that is necessary.
sounds right to me, they would also have to fix pythonservice as well.
Incidentally I notice that the path used by pythonservice is the 8.3 dos
mangled version. Inside the Zope startup script they're using the normal
long filename version to possibly correct the path. Will that not cause
problems? 
-- 
Robin Becker


From quinn at vomit.ugcs.caltech.edu  Mon Feb 25 13:44:23 2002
From: quinn at vomit.ugcs.caltech.edu (Quinn Dunkan)
Date: 25 Feb 2002 18:44:23 GMT
Subject: problem with exception output
References: <a5d01k$69es1$1@ID-79019.news.dfncis.de>
Message-ID: <slrna7l1g7.3al.quinn@vomit.ugcs.caltech.edu>

On Mon, 25 Feb 2002 10:30:25 +0100, Felix Seeger <seeger at sitewaerts.de> wrote:
>Hi
>
>I can't get the hole exception if my script (cgi) fails.
>I use this to get every exception:
>
>except:
>        print '<table width="100%">'

python2.2 has 'cgitb', which may already do what you want.


From fredrik at pythonware.com  Sun Feb 17 03:24:13 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 17 Feb 2002 08:24:13 GMT
Subject: Have to try
References: <mailman.1013916386.10113.python-list@python.org> <3c6f47dc.3932421@news.t-online.de>
Message-ID: <NMJb8.26247$l93.4947379@newsb.telia.net>

Gerson Kurz wrote:
> def strcmp(a,b):
>     "Compare two strings, case-sensitive"
>     return a == b

the real "strcmp" returns zero if the strings match; try
this one instead:

    strcmp = cmp

</F>




From rjones at ekit-inc.com  Thu Feb 14 17:10:58 2002
From: rjones at ekit-inc.com (Richard Jones)
Date: Fri, 15 Feb 2002 09:10:58 +1100
Subject: Java vs Python ... ???
In-Reply-To: <Xns91B5CBEF0A103Scirious@140.99.99.131>
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <200202142211.WAA08035@crown.off.ekorp.com>

On Fri, 15 Feb 2002 09:00, Scirious wrote:
> I've been learning JAVA for a while, but now I've heard avout Python.
> Any coments concerning these two languages?

Please read:
  http://www.python.org/doc/Comparisons.html

And come back when you have specific questions to ask about Python.


   Richard



From borcis at geneva-link.ch  Sat Feb  9 19:07:14 2002
From: borcis at geneva-link.ch (Boris^2)
Date: Sun, 10 Feb 2002 01:07:14 +0100
Subject: OT: Perl programmers?
References: <mailman.1013225776.2733.python-list@python.org>
Message-ID: <3C65B9B2.6FC19A1D@geneva-link.ch>

hungjunglu wrote:
> 
> --- In python-list at y..., aahz at p... (Aahz Maruch) wrote:
> > ... the way we call ourselves Pythonistas?
> 
> Never liked that name. You may call yourself Pythonista, but I prefer
> the term Pythoneer.

Pythoness :)

<filter(lambda w : w not in "illiterate","bullshit")-ly yours>,

Boris


From phr-n2002a at nightsong.com  Mon Feb 25 00:54:47 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 24 Feb 2002 21:54:47 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com> <a5cfpe$67voi$1@ID-11957.news.dfncis.de>
Message-ID: <7xbseeayxk.fsf@ruckus.brouhaha.com>

"Emile van Sebille" <emile at fenx.com> writes:

> "Paul Rubin" 
> > Both of those build up a new list of the results, instead of
> > discarding the values.  If the f function takes an integer and
> > computes a 20-megabyte structure, you've got a problem.  
> 
> If you know it's coming, throw it away first:
> 
> [x.f() and None for x in items]

OK, but that buidls a list of None as long as the item list.  Not
so bad, since it gets reclaimed right away, but still unsatisfying.

> filter(None,[x.f() and None for x in items])
> 
> returns an empty list vs 0 for the reduce.  

Again there's this intermediate list that's as long as the original list.

Do you see a way around that?

Thanks


From bokr at oz.net  Mon Feb 18 02:13:06 2002
From: bokr at oz.net (Bengt Richter)
Date: 18 Feb 2002 07:13:06 GMT
Subject: Problems with telnetlib
References: <mailman.1014010108.5663.python-list@python.org>
Message-ID: <a4q9i2$rvq$0@216.39.172.122>

On Mon, 18 Feb 2002 00:27:02 -0500, Tim Peters <tim.one at comcast.net> wrote:

>[Bengt Richter]
>> ...
>> Otherwise maybe a standalong unified diff GUI viewer with buttons to see
>> old, new, or both?
>
>On Windows, that's called windiff.[...
I actually meant a viewer for the output of plain old "diff -u f1 f2"
to show it in alternative modes, so someone could accept diffs in unified
format and not feel inconvenienced ;-) Of course, grep plus less will get
you pretty close, if you're not fussy.

>...] That was the first app I ever used that
>made me suspect a GUI might actually be better for something than working in
>an Emacs process buffer.  Luckily, it was also the last <wink>.
>
Funny, I have windiff, but I ported plain old diff and made an NT console app
out if it, and I use the latter almost 100%.

Regards,
Bengt Richter



From guido at python.org  Thu Feb  7 22:51:14 2002
From: guido at python.org (Guido van Rossum)
Date: Thu, 07 Feb 2002 22:51:14 -0500
Subject: Got Python10 pictures or other memorabilia online?
Message-ID: <200202080351.g183pEw23822@pcp742651pcs.reston01.va.comcast.net>

The conference was a huge hit!

If you're putting your photos, slides or other memorabilia from
Python10 on the web, please drop a note with a link to
webmaster at python.org and I'll create a page of links on python.org.

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




From amuys at shortech.com.au  Mon Feb 18 19:16:54 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 18 Feb 2002 16:16:54 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>	<mailman.1013816158.26552.python-list@python.org>	<7934d084.0202172009.75324c2a@posting.google.com> <mailman.1014058862.32020.python-list@python.org>
Message-ID: <7934d084.0202181616.7d333836@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1014058862.32020.python-list at python.org>...
> Obviously you were aiming this statement at me, not Tripp ;)  I won't argue
> much (because you are correct), I will simply point out that these
> situations, while real, aren't that common in typical multithreaded
> applications (in my experience) unless the program is poorly thought out to
> start with.  They do however, make multithreaded programming seem
> _extremely_ daunting to a beginner.
>
> Speaking for myself, I've written several long-running multithreaded
> applications where I never _consciously_ considered these things.  I think
> once someone has developed a feel for multithreaded programming, these
> issues are handled at a different level in the designer's mind and explicit
> consideration of them usually isn't necessary (unless perhaps I'm just
> applying my own experience too broadly).  Maybe I'm just smarter than I
> give myself credit for <big wink>, as I've never run into the "inevitable
> trouble" you mention. 

Fair enough.  I just hate to think how many times I've seen:

pthread_mutex_lock(mutex);
/*!!! BUG: needs to be while loop or mut-ex is violated in critical
section. */
if (!condition) {  
    pthread_cond_wait(cond_var, mutex);
}
<critical section>
pthread_mutex_unlock(mutex);

let alone other more subtle bugs.

I had an email discussion with a beginner the other day, during which
I quickly sketched out a simple master-slave thread-pool class (C++
using pthreads).  The guy thought I had copied it out of a book, you
are correct with time you develop a feel for concurrency and
internalise the solutions to the various problems involved. 
(Ironically my sketch was text-book quality, ie. suitable for
illustration only, completely inadequate for production use ;)

> Considering all this, I'll take a different position: For beginners: don't
> worry about these issues now, but after becoming familiar with the basics
> of multithreaded programming (for which the standard Python docs should be
> sufficient), find a good book to deepen your knowledge.  That is to say,
> I'll qualify my earlier statement "end of story" with "for now".

And so we prove that reasonable people can differ :).

> > Everyone: Does anyone know of an accessable concurrency text?  One
> > that moves from discussing the basics (data-races, critical sections,
> > mutural exclusion, lock-order based deadlock, resource based deadlock,
> > etc) to the various solutions (mutex/cond-var, semaphore, monitor,
> > rendezvous) to the various design approaches (master-slave, pipeline,
> > work-pile, etc).  Possibly even a discussion of various performance
> > considerations (convoy's, wait-free synchronisation, cache-aliasing,
> > spin vs. blocking locks, etc).
> 
> I would certainly like to see a good discussion of these.  Do you have a
> URL for the tutorial you wrote?

I will be rewriting it over the course of the next couple of months. 
It was written as a short-form talk for our local home-unix user
group.  I never expected it to be distributed as far as it has. 
(apart from anything, the attached code is very poor quality, and I
really should write some production quality code to supliment the
tutorial examples).

http://www.humbug.org.au/talks/pthreads/

Andrae Muys


From peter.milliken at gtech.com  Mon Feb  4 16:53:27 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Tue, 5 Feb 2002 08:53:27 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com><Wls68.14178$om6.357057@news1.tin.it> <mailman.1012590723.7301.python-list@python.org> <a3kb3p$se44@news1.gtech.com> <3C5EE7C2.4DF4A885@ccvcorp.com>
Message-ID: <a3n03m$tfn7@news1.gtech.com>

"Jeff Shannon" <jeff at ccvcorp.com> wrote in message
news:3C5EE7C2.4DF4A885 at ccvcorp.com...
>
>
> Peter Milliken wrote:
>
> > G'Day Cliff,
> >
> > "Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
> > news:mailman.1012590723.7301.python-list at python.org...
> > > On Fri, 01 Feb 2002 08:29:42 GMT
> >
> > > I much prefer the "continuous ongoing code review" idea better than a
> > > formalized procedure (if I understand what you mean by it correctly).
> > > Working in pairs provides this as a side-effect rather than having a
> > formal
> > > procedure for code-review and seems less time-consuming overall.
> > >
> >
> > Not sure what you mean by "continuous ongoing code review". I used to be
a
> > "team leader",  .....
>
> I hesitate to speak for others, but...   I'm going to stomp right on in
anyhow.
> :)
>
> I believe that the "continuous ongoing code review" being mentioned is in
> reference to the Extreme Programming practice of pair programming -- all
code is
> written jointly by two people, sitting at the same computer.  Thus,
everything
> is, in effect, being reviewed by the other person as it's being written.
This
> should (in principle ;) ) greatly reduce the need for formal group
reviews...
> though I expect they'd probably still be desirable regardless.
>
>
Really? What an unusual idea - I didn't realise we had such a surplus of
programmers to advocate something like this :-)

Actually, it is just another attempt at fixing the same old problem. I have
no problems with trying new techniques such as these but at the end of the
day, if good old fashioned reviews are done properly (and in a timely
manner! :-)) then you would achieve the desired effect i.e. solid, robust
code. If the individuals involved approach the problem with the correct
frame of mind then I have never, ever seen inspections/reviews fail! It is
only when it is not tackled with the right sort of commitment that it can
appear not to work.

I get somewhat frustrated sometimes when I look at yet more attempts to
solve the same problem when I believe there are already workable solutions
(this will, I am sure be disputed :-) but is only a statement based upon
personal experience :-)). I think it is most likely the lack of
self-discipline (in programmers, not themselves) that caused the authors to
look for alternative methods - also, the fact that they needed something new
to write about otherwise they wouldn't be able to publish a book! :-) I must
admit that in a weak moment, I purchased a copy of Extreme Programming, I
read it on the train one evening (not the whole book! :-)) and saw enough to
make me put it where it is now, on my shelf gathering dust :-).

> > ..... So I assume that "continuous ongoing code review" is similar to
> > my idea of "early and often"? :-)
>
> That's more or less my interpretation of it, too.  :)
>
> Jeff Shannon
> Technician/Programmer
> Credit International
>
>




From lars at larsshack.org  Tue Feb  5 00:50:42 2002
From: lars at larsshack.org (Lars Kellogg-Stedman)
Date: Tue, 05 Feb 2002 05:50:42 GMT
Subject: Inconsistence syntax in os module?
Message-ID: <3C5F731D.4060500@larsshack.org>

Howdy all,

I'm growing increasingly frustrated with the os module, in particular 
with its implementation of the system() and popen() calls.  The 
traditional (e.g., libc) implementation of both of these calls spawns a 
/bin/sh process to run the command, with the associated problems of 
shell metacharacter interpretation, the cost of an extra fork(), and so 
forth.

On the other hand, the popen2 module (and thus, at least on unix, the 
os.popen2(), os.popen3(), etc. functions) will also accept a *list* as 
the command argument.  In this situation, the command is exec()'d 
directly, without the intervening shell (since we no longer need the 
shell to perform tokenization for us).

Rewriting system() and popen() to be consistent with popen2() and 
friends is trivial, especially if one writes the implementations in python.

Is there a reason why the stock implementation differs in this fashion?

Thanks,

-- Lars



From bellman at lysator.liu.se  Fri Feb  1 18:12:49 2002
From: bellman at lysator.liu.se (Thomas Bellman)
Date: Fri, 1 Feb 2002 23:12:49 +0000 (UTC)
Subject: Equivalent of Perl chomp?
References: <mailman.1012403361.2370.python-list@python.org> <u5i8b8jtee2r13@corp.supernews.com> <7xbsfag023.fsf@ruckus.brouhaha.com> <3C59D2EE.3937ABBB@ndh.net>
Message-ID: <a3f7dh$lq9$1@news.island.liu.se>

Stefan Schwarzer <s.schwarzer at ndh.net> writes:

> It might be interesting to give some examples of code that "wants" chomp.
> We then could see if there is a pythonic way to make a chomp-like function
> unnecessary. Maybe there are better ways.

I have typically needed to strip newlines when I'm processing a
text file where each line is a record with fields separated by
some character.  If you just .split() on the separator, you end
up with an unwanted newline in the last field.

/etc/passwd is a good example of such a file under Unix.  Sure,
there's the pwd module, but 1) it only handles *that* format, not
other, formats, and 2) it really is an interface to the current
machines passwd database; if you want to process some random file
in /etc/passwd format, you can't use it.


Another occasion I have wanted a chomp(), is when I want to check
if the line ends in some special character (like \).  I can't
then just do

    for line in fp.xreadlines():
	if line[-1:] == "\\":
	    blahonga(line)

but instead have to do

    for line in fp.xreadlines():
	if line[-2:] == "\\\n":
	    blahonga(line)

and I find that less readable.


-- 
Thomas Bellman,   Lysator Computer Club,   Link?ping University,  Sweden
"Adde parvum parvo magnus acervus erit"       ! bellman @ lysator.liu.se
          (From The Mythical Man-Month)       ! Make Love -- Nicht Wahr!


From ykingma at accessforall.nl  Mon Feb 25 14:33:12 2002
From: ykingma at accessforall.nl (Ype Kingma)
Date: Mon, 25 Feb 2002 20:33:12 +0100
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com> <3c7a70d2_4@news.bluewin.ch> <gherman-A86A1B.18413225022002@news.t-online.com>
Message-ID: <3C7A916B.234D6690@accessforall.nl>

Dinu,

Dinu Gherman wrote:
> 
> In article <3c7a70d2_4 at news.bluewin.ch>,
>  "Samuele Pedroni" <pedronis at bluewin.ch> wrote:
> 
> > Yes they are bugs. Please report them.
> > Since 2.1 release (two month ago) there have been approximately  10000
> > download.
> >
> > How are such bugs possible?
> >
> > 1. the typical idiom for list(l) is l[:] (mildly irrelevant I know)
> > 2. if you check the test suite for CPython 2.1 and Jython there is no test
> > for list(l) behavior
> > 3. the new reset method is also not tested by the test suite and is not
> > reported in the NEWS file
> >
> > (These are the result of some grepping, maybe I'm wrong but given the bugs
> > probably I'm right).
> >
> > We try to follow the big picture and the PEPs and check the NEWS file but
> > for the rest the test suites
> > are our best hope vs. delusional friend.
> > If a feature is old, under-used, or through usage the bug does not show
> > through, and un-tested things become tricky.
> >
> > It is open source: Jython is as conforming as its community and CPython
> > community ACTIVELY want it to be.
> 
> I'm pretty surprised! I knew the Python test suite is *very* far
> from complete, but list() is an *extremely* crucial function,
> isn't it?

Samuele already gave the workaround for list() on a list, (ie. [:])
so it's not very extremely crucial.

> My impression is that Jython claims to be an implementation of
> Python in Java. Everybody understands the existance of bugs, but
> if functions are missing I'm not sure there is sufficient quality
> control, leave alone a useful test suite. My bold guess is that
> it should be very easy to check automatically for each module in
> the std.lib. at least if the same classes and methods do exist
> in CPython and JPython.

That requires dumping the std lib interface in CPython and checking
this dump against the Jython standard libs. It could be straightforward
but I don't see how.

> Honestly, I don't think it makes much sense to maintain two
> code bases without some degree of automatic testing...

There are test suites for Jython but they cannot be completely 
the same as for CPython, eg. because the implementations
differ in support for the python language, ie. jython is normally
a bit behind.
It's almost straightforward to run the CPython test suite in
Jython when you have both of them installed. I did this when I
started in jython just over a year ago to see how wel it Jython
conforms to CPython.
At that time most of the difference was in unicode support.
You might try the same now, I guess you'll be pleasantly surprised.

> 
> When seeing such bugs, my immediate reaction (like that of most
> others) is to think that not many people can be using this serious-
> ly.

Well, for every bug solved in jython the developers add a test case
these days. The bug reports at jython.sourceforge.net will give
you an impression of what kinds of things people run into nowadays
when they use Jython. I thought I'd refer you to bugs link at
www.jython.org, but you've been there already. Thanks.

> Regards,
> 
> Dinu
> 
> PS: BTW, how about this one:
> 
> [localhost:~] dinu% jython
> Jython 2.1 on java1.3.1 (JIT: null)
> Type "copyright", "credits" or "license" for more information.
> >>> import os
> >>> os.chdir('..')
> Traceback (innermost last):
>   File "<console>", line 1, in ?
>   File ".../Jython-2.1/Lib/javaos.py", line 56, in chdir
> OSError: [Errno 0] chdir not supported in Java: ..

Since Jython is based on Java there is no way to directly
support chdir.
There has been some work going on on the jython's
os module lately, but this may not be in the latest release.
I don't know whether a workaround was made for chdir.
Ask on jython-dev if you want to know more...

Have fun,
Ype

P.S. Although jython used to be known for finding bugs in JIT
compilers, I can recommend to use a JIT if you have one...


From mertz at gnosis.cx  Thu Feb 14 14:56:47 2002
From: mertz at gnosis.cx (Lulu of the Lotus-Eaters)
Date: Thu, 14 Feb 2002 14:56:47 -0500
Subject: Finding path to __init__.py in package
Message-ID: </ZBb8kKkXYWJ092yn@bellatlantic.net>                

I think this is one where the answer will be so stunningly obvious that
I'll know myself on the head.  But it is not coming to me right away (I
haven't worked with packages much, mostly modules, so that might be the
problem).

Anyway, I have created a package that has various subpackages.  One
subpackage (and more to come) is documentation files.  These ASCII
documents can be read in a plain text editor, but I also want to import
their content into some namespaces.  For example, I'd like a script to
be able to include:

    from gnosis.xml.pickle.doc import history

Specifically, some other namespaces want to import some names this way.

My current setup works, but looks kludged:

    $PYTHONPATH/gnosis/xml/pickle/doc/__init__.py:
    --------------------------------------------------------------
    from os import sep
    import gnosis
    gdir = gnosis.__file__.split(sep)[:-1]
    ddir = sep.join(gdir+['xml','pickle','doc',''])
    _= lambda f: open(ddir+f).read()
    l= lambda f: _(f).rstrip().split('\n')

    howto     = _('HOWTO')
    copyright = _('COPYRIGHT')
    history   = _('HISTORY')
    security  = _('SECURITY')
    todo      = _('TODO')
    version   = _('VERSION')
    author    = l('AUTHOR')
    thanks_to = l('THANKS_TO')

That is, I *know* that each (ALLCAPS) documentation file is contained in
the same directory as the __init__.py file.  But os.getcwd() just reports
the directory I am working from, and sys.path[0] reports the directory
of the script that -imports- gnosis.xml.pickle.doc, not the directory of
the __init__.py.

What I would like is a function that asks "what directory is this
current __init__.py script running from?"

Yours, Lulu...

--
mertz@  | The specter of free information is haunting the `Net!  All the
gnosis  | powers of IP- and crypto-tyranny have entered into an unholy
.cx     | alliance...ideas have nothing to lose but their chains.  Unite
        | against "intellectual property" and anti-privacy regimes!
-------------------------------------------------------------------------




From gh_pythonlist at gmx.de  Thu Feb 14 23:24:28 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Fri, 15 Feb 2002 05:24:28 +0100
Subject: Curses & Colors
In-Reply-To: <3C6C06FF.DAFFBB43@zumnet.com.br>
References: <3C6C06FF.DAFFBB43@zumnet.com.br>
Message-ID: <20020215042424.GA450@lilith.hqd-internal>

Ok, I've checked my FreeBSD installations :-)

Le 14/02/02 ? 16:50, Rafael Mentz Aquino ?crivit:
> Hi, there...
> 
> I'm creating some simple interfaces with curses with python under FreeBSD,
> but I'm having some problems to change colors. I tried to do something in
> /etc/termcap, but when I run curses.can_change_color(), I always obtain 0.
> My env term is cons25, but I don't know if python uses the same terminal.

TERM=cons25 definitely works on my machines - in xterm *and* system console.

But in order to make it work on the system console, I had to change the default
console settings. Try adding this to your /etc/rc.conf:

saver="daemon"
blanktime="60"
font8x8="iso-8x8"
font8x14="iso-8x14"
font8x16="iso-8x16"
scrnmap="NO"

As you can guess, the screensaver ist the most important part ;-)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: -1.0 ?C      Wind: 4.8 m/s



From mal at lemburg.com  Mon Feb 25 06:02:06 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 25 Feb 2002 12:02:06 +0100
Subject: ANN: py2html.py -- Version 0.7
Message-ID: <3C7A19AE.AFC84AE7@lemburg.com>

________________________________________________________________________

ANNOUNCING:
                      py2html.py -- Version 0.7

                A Python Source Code to XHTML converter

________________________________________________________________________

WHAT IS IT ?:

py2html.py is a script which converts Python source files
to highlighted XHTML. It can be run as command line application
or used as CGI script.

________________________________________________________________________

WHAT'S NEW ?

Version 0.7 adds support for XHTML thanks to Ville Skytt?.

________________________________________________________________________

WHERE CAN I GET IT ?

You can download the script from the usual place...

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

________________________________________________________________________

WHAT DOES IT COST ?

The script comes with the eGenix.com Public License, which means
that you can use them in both commercial and non-commercial settings
without fee or charge.

________________________________________________________________________

WHERE CAN I GET SUPPORT ?

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

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

for details about the eGenix support offerings.

Free support is available through the eGenix.com User Mailing List
which provides a forum for eGenix.com Software Users:

 http://lists.egenix.com/mailman/listinfo/egenix-users/

________________________________________________________________________

Enjoy,
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/




From skywalkerpackage at hotmail.com  Wed Feb  6 07:25:33 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Wed, 06 Feb 2002 12:25:33 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org> <3c60fb80.199334@news.online.de> <a3qvmn$6o2$1@norfair.nerim.net>
Message-ID: <3c6118c0.7688521@news.online.de>

Hello Eric,

thanks very much for your quick response. Now I have get the error
message:

Traceback (most recent call last):
  File
"D:\python\activestate\python21\Pythonwin\pywin\framework\scriptutils.py",
line 301, in RunScript
    exec codeObject in __main__.__dict__
  File "D:\python\activestate\python21\myscripts\labelentry_dyn.py",
line 9, in ?
    apply(bt.config, (), optionDict)
  File "D:\PYTHON\ACTIVESTATE\PYTHON21\lib\lib-tk\Tkinter.py", line
1085, in configure
    self.tk.call((self._w, 'configure')
TclError: unknown option "-labelText"


although bt is an intstance of LabelEntry and it has the labelText
option.

I think the problem is that I haven't defined a config method in my
Class LabelEntry. It thought that this is not necessary.

When I define a config method, I have the problem that some of the
LabelEntry options (padx and pady) are options for the packer of the
widgets inside (that is, Label and Entry). So, I can't just write
something like

    def config(self, kw):
        self.label.config(text=kw.get('labelText',''))
        self.entry.config(padx=kw.get('padx',4),
pady=kw.get('pady',4))

Puh, wired problem for me :<

I think I won't use apply, but use the call by dictionary option
instead.

Johannes



On Wed, 6 Feb 2002 11:18:18 +0000, Eric Brunel
<eric.brunel at pragmadev.com> wrote:

><posted & mailed>
>
>Hi Johannes,
>
>> def __init__(self, master=None, **kw):
>>       Frame.__init__(self, master)
>>      ...
>>       self.label = Label(self, text=kw['labelText'])
>> ...
>> 
>> But then, how do I handle default values, and how do I ensure that the
>> objects are created with the right parameters?
>
>Do not use kw['labelText'], but kw.get('labelText', ''). The get method on 
>a dictionary does just what you want. Please refer to:
>http://www.python.org/doc/current/lib/typesmapping.html
>
>HTH
> - eric -



From peter at engcorp.com  Sun Feb 10 12:50:40 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 10 Feb 2002 12:50:40 -0500
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <7x665cz2xl.fsf@ruckus.brouhaha.com> <3C5FABB1.5010703@thomas-guettler.de> <7xbsf4cjdp.fsf@ruckus.brouhaha.com>
Message-ID: <3C66B2F0.1D70D9DF@engcorp.com>

Paul Rubin wrote:
> 
> Thomas Guettler <st-newsgroups at thomas-guettler.de> writes:
> > The module random contains a lot of functions. I think getting a
> > random 32 random number is needed often.
> >
> > randrange(0,2147483647) is not a 32 Bit random number because
> >   2**32-1 is 4294967295.
> 
> Woops, sorry, I misread 2**32-1 as 2**31-1 which is the largest
> 32-bit positive integer.  2**32-1 is an overflow if you use 32 bit
> ints.  I guess you could try random(-2147483648, 2147483647)
> if you want all 32 bits random.  Alternatively you could try
>   (random(0,65536) << 16 | random(0,65536))

Unfortunately, random.randrange(-2147483648, 2147483647) would
exclude 2147483647.  randrange(1) produces only zero....


From usenet at thinkspot.net  Thu Feb 14 17:27:16 2002
From: usenet at thinkspot.net (Sheila King)
Date: Thu, 14 Feb 2002 14:27:16 -0800
Subject: Mutex a file?
References: <a4h98k$ao2f$1@ID-98166.news.dfncis.de>
Message-ID: <a4ghg4.3vv79at.1@kserver.org>

On Thu, 14 Feb 2002 16:12:51 -0500, "MDK" <mdk at mdk.com> wrote in comp.lang.python in
article <a4h98k$ao2f$1 at ID-98166.news.dfncis.de>:

> I have a file that is to be read to and written from by many Python scripts
> running at the same time.
> 
> Is there a way to put a Mutex on it so that the scripts wait their turn
> before tyring to open it?

This is a problem I worked on some time back. On most Unices, it is not 
hard to lock the file (which is what you need to do). I guess there are 
some problems if you are working across certain types of networks. On 
WinNT and Win2000 (and probably XP) there is also a great deal of control 
over files and file access. It's the Win9x where you loose a lot of this 
ability. I'm not sure what platform you are running on...if Windows, 
you might want to look in more detail at the win32all libraries from 
Mark Hammond.

Anyhow, here are some links to the discussion I initiated on this topic 
some time back. Read the threads. There are many useful comments.
http://groups.google.com/groups?as_umsgid=t7uqot8n8gjm03r2m03md8kfsjuauchqhu%404ax.com
http://groups.google.com/groups?as_umsgid=djsfotomogdeba70gp1cuogjbt5ii3chrb%404ax.com
http://groups.google.com/groups?as_umsgid=bo2ootg61vmposfvsg8gvrks6eqiuss940%404ax.com
http://mail.python.org/pipermail/tutor/2001-August/008366.html


Here is the module I wrote, and am using in a production environment 
on a Linux server for file locking. Advice given to me re: the 
Windows component of this modules is...well, that it wouldn't be 
recommended to use in a production environment. But who would run 
a production server on a Win98 machine anyhow? So, for me it let's 
me test my scripts on my home Windows machine and then upload it 
to my Linux server and expect the same results.

FWIW, I've pasted the code I'm currently running below my sig.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_


'''MutexFile.py

wraps WinMutexFile.py and posixMutexFile.py to provide
a single interface for working with mutex files
across win32 and posix platforms.
Does not support fine-grained locking.
Recommended for getting a lock on a sentinel file
before trying to read/write data to a separate data file.

public interface:
obj = MutexFile.MutexFile('somefile.txt')
obj.flock(flag)

where flag is one of the following:
LOCK_SH - get a shared (or read) lock
LOCK_EX - get an exclusive (or write) lock
LOCK_UN - release a lock
'''
import os

if os.name == 'nt':
     from winMutexFile import mutexfile
elif os.name == 'posix':
     from posixMutexFile import mutexfile
else:
    raise ImportError, "MutexFile is not supported on your platform."


class MutexFile(mutexfile):
    pass



''' posixMutexFile.py

class lockfile
supports the same function calls as
winLock.lockfile
export to MutexFile.py: a wrapper module
around the win and posix lockfiles
'''

import os, fcntl

class mutexfile:
    def __init__(self, filename):
        if os.access(filename, os.F_OK):
            self.filename = filename
            self.fd = None
        else:
            errmssg = filename + " does not exist. Can't lock non-existent file."
            raise IOError, errmssg

    def __del__(self):
        try:
            self.unlock()
        except:
            pass
        try:
            self.f.close()
        except:
            pass

    def getReadLock(self):
        self.f = open(self.filename, 'r')
        self.fd = self.f.fileno()
        fcntl.lockf(self.fd, fcntl.LOCK_SH)


    def getWriteLock(self):
        self.f = open(self.filename, 'r+')
        self.fd = self.f.fileno()
        fcntl.lockf(self.fd, fcntl.LOCK_EX)


    def unlock(self):
        fcntl.lockf(self.fd, fcntl.LOCK_UN)
        self.f.close()
        self.fd = None

    def flock(self, flag):
        '''flags are:
        LOCK_UN - unlock
        LOCK_SH - acquire a shared (or read) lock
        LOCK_EX - acquire an exclusive (or write) lock
        '''
        if flag == 'LOCK_SH':
            self.getReadLock()
        elif flag == 'LOCK_EX':
            self.getWriteLock()
        elif flag == 'LOCK_UN':
            self.unlock()
        else:
            errmssg = "The flag " + flag + " is not implemented for flock"
            raise NotImplementedError, errmssg



''' winMutexFile.py

class lockfile
supports the same function calls as
posixLock.lockobject
export to lockobject.py: a wrapper module
around the win and posix lockobjects
'''
import os
from time import time, sleep

try:
    import win32file
    from win32con import GENERIC_READ, GENERIC_WRITE,\
        FILE_SHARE_READ, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL
except ImportError, e:
    print e
    print "winMutexFile.py requires the ActiveState.com Python win32 extensions"
    raise

MAXTIME = 8  # number of secs to retry for a lock before timing out
sleepinterval = 0.1 # number of secs to sleep between retries for lock

class mutexfile:
    def __init__(self, filename):
        if os.access(filename, os.F_OK):
            self.filename = filename
            self.fd = None
        else:
            errmssg = filename + " does not exist. Can't lock non-existent file."
            raise IOError, errmssg

    def __del__(self):
        try:
            self.unlock()
        except:
            pass
        try:
            self.f.close()
        except:
            pass

    def getReadLock(self):
        start_time = time()
        while time() - start_time < MAXTIME:
            try:
                self.fd = win32file.CreateFile(self.filename, GENERIC_READ,\
                                              FILE_SHARE_READ, None, OPEN_EXISTING,\
                                              FILE_ATTRIBUTE_NORMAL, 0)
                return 1
            except:
                pass
            sleep(sleepinterval)
        if not self.fd:
            errmssg = self.filename + " temporarily unavailable"
            raise IOError, errmssg

    def getWriteLock(self):
        start_time = time()
        while time() - start_time < MAXTIME:
            try:
                self.fd = win32file.CreateFile(self.filename, GENERIC_READ,\
                                              0, None, OPEN_EXISTING,\
                                              FILE_ATTRIBUTE_NORMAL, 0)
                return 1
            except:
                pass
            sleep(sleepinterval)
        if not self.fd:
            errmssg = self.filename + " temporarily unavailable"
            raise IOError, errmssg

    def unlock(self):
        win32file.CloseHandle(self.fd)
        self.fd = None

    def flock(self, flag):
        '''flags are:
        LOCK_UN - unlock
        LOCK_SH - acquire a shared (or read) lock
        LOCK_EX - acquire an exclusive (or write) lock
        '''
        if flag == 'LOCK_SH':
            self.getReadLock()
        elif flag == 'LOCK_EX':
            self.getWriteLock()
        elif flag == 'LOCK_UN':
            self.unlock()
        else:
            errmssg = "The flag " + flag + " is not implemented for flock"
            raise NotImplementedError, errmssg



From lac at strakt.com  Fri Feb 22 07:10:47 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 22 Feb 2002 13:10:47 +0100
Subject: Adding GUI to application 
In-Reply-To: Message from Mark Gash <mgash@trisystems.co.uk> 
   of "Fri, 22 Feb 2002 09:24:21 GMT."
Message-ID: <200202221210.g1MCAltB005137@ratthing-b246.strakt.com>

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/82965

Laura Creighton



From stojek at part-gmbh.de  Mon Feb 18 05:26:35 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Mon, 18 Feb 2002 10:26:35 GMT
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net> <3C6C1298.505C578B@ccvcorp.com>
Message-ID: <3c70d6b0.2091734@news.easynews.net>

>
>But this tends to be pretty slow.  For large lists, it's usually
>considered that the "decorate-sort-undecorate" pattern is the most
>effective way.  Something like this:
>
>def SortOnItem(mylist, index):
>    templist = [ (line[index], line) for line in mylist ]
>    templist.sort()
>    return [ line[1:] for line in templist ]
>
yapp. That's what I was looking for.
Thanks Marcus


From compedu at nh.km.ua  Wed Feb 20 04:01:36 2002
From: compedu at nh.km.ua (Dmitry Savchuk)
Date: Wed, 20 Feb 2002 11:01:36 +0200
Subject: Help with code
Message-ID: <a4votp$mmf$1@infocom.km.ua>

Hi,

I'd like to ask you a question. There is a code from the book of Ivan Van
Laningham "Sams Teach Yourself Python in 24 Hours (Teach Yourself -- 24
Hours)":

import time
class now:
    def _init_(self):
        self.t=time.time()
        self.storetime()
    def storetime(self):
        self.year,\
        self.month,\
        self.day,\
        self.hour,\
        self.minute,\
        self,second,\
        self.dow,\
        self.doy,\
        self,dst=time.localtime(self.t)
    def _str_(self):
        return time.ctime(self.t)

n=now()
print "The year is ", n.year
print n
s=`n`
print s


It doesn't work properly. Appear a messge:

The year is
Traceback (most recent call last):
  File "C:\Projects\test", line 20, in ?
    print "The year is ", n.year
AttributeError: now instance has no attribute 'year'

Where is my mistake?

Thak you in advance.
Dmitry Savchuk.




From skip at pobox.com  Wed Feb 13 09:45:11 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 13 Feb 2002 08:45:11 -0600
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <15466.27652.704528.795698@12-248-41-177.client.attbi.com>
References: <mailman.1013557302.9749.python-list@python.org>
        <7xzo2ekv5g.fsf@ruckus.brouhaha.com>
        <mailman.1013572042.18876.python-list@python.org>
        <3C69ED33.CE0100DF@pizero.colorado.edu>
        <15465.63503.142011.246239@12-248-41-177.client.attbi.com>
        <Pine.LNX.4.21.0202122224100.12912-100000@higgs.colorado.edu>
        <m33d05u2dn.fsf@ferret.object-craft.com.au>
        <15466.27652.704528.795698@12-248-41-177.client.attbi.com>
Message-ID: <15466.31735.232727.980577@12-248-41-177.client.attbi.com>

    Dave> Use strace to track system calls.

Thanks to Dave's suggestion I found the culprit.  I was using an unsigned
int as a for loop counter that counted down to zero:

    unsigned int i;
    ...
    for (i=nchunks; i>=0; i--) {
        ...
    }

This is (of course) an infinite loop.  The strace output showed 28,000+
calls to unlink.  Obviously I was forgetting to check its return value as
well. :-(

Thanks for all the help...

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From phr-n2002a at nightsong.com  Mon Feb 18 01:53:54 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 22:53:54 -0800
Subject: Editable strings anyone?
References: <mailman.1014011067.11712.python-list@python.org>
Message-ID: <7xd6z3xou5.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> > that has multiple segments and implements the buffer interface.
> > Then the re module can be extended to handle the buffer interface,
> > if it doesn't handle it already.
> 
> Someone may have mentioned this before, but I think you would have
> to sneakily close the gap whenever the re module asks you for your
> buffer pointer, because re needs a contiguous buffer.  See:

Yes, that's why the re module would have to be extended--to handle
noncontiguous segments.  The GNU regexp library already has an API
for that and it shouldn't be that hard to add something similar
to Python's re library.


From m.faassen at vet.uu.nl  Sat Feb  2 16:59:25 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Feb 2002 21:59:25 GMT
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com>
Message-ID: <a3hnft$cgu$1@newshost.accu.uu.nl>

Rony <rony.steelandt at bucodi.com> wrote:
> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1012274265.29834.python-list at python.org>...
>> Guys, I know we have the reputation as the nicest bunch of nerds on
>> the 'Net, but Rue is consuming more of our time and bandwidth than spam.
>> Despite his comments to the contrary, no one on this list gives him
>> any credence.

> Yes it's true that there is a nice bunch here. BUT what keeps me
> amazing here, is  :

> 1. How you keep defending the Python language over and over again.
> Don't understand me wrong, Python is a VERY good language and we are
> using it for years now, like a lot of you. But each time some guy asks
> here a question like 'Should i use Python of Perl' or 'Python or C for
> a real world application' evrybody here starts here to defend Python.

I more usually list a number of factors that may influence this decision.
Of course those factors would be frequently tilted towards the use
of Python, as I do believe it is in many circumstances a good choice.

Anyway, I've seen others go along those lines too.
 
> IMHO this is useless, since allready the question is stupid...

It may be somewhat ignorant to ask 'should I use Python or X' without
much qualifying context, but one has to start exploring somewhere.
I'd agree some context is useful ('for games?' 'for accounting 
applications?' 'for web applications?' etc).

Anyway, asking such a question in a newsgroup may get you some interesting
perspectives that could help you explore the issues involved better than you 
could before.

Of course, if you ask 'Should I use X or Y' in comp.lang.X you're most likely
going to get people in favor of X speaking up. But even that can help
you. Here for instance you'll often hear: 'X is good except if you want
extreme speed, but then you can do this and that', while over on 
comp.lang.c you might hear things more like 'X is good, some say some other
language is better for higher level abstractions but you can do this and
that'. You'll have learned something.

Then again, one could've done a simple web search and learned the same. :)

> The language you want to choose depends on a lot of items and not only
> wether you have to use brackets of indents to structure code block, or
> to have to declare variables or not. Even the developmentspeed issues
> like you see here often depends on a lot of things, we have as an
> example 2 C goeroes in house that develop at the speed of other python
> applications, only the applications will be different and THAT is the
> reason that some application is written in C and others in Python...
> So if someone asks if he should use Python , the only real answer
> could be 'Have a look at it, try it and decide... you're a big boy'.

Yes, that, and set them on the way. You can list strong points of Python
(clarity and expressiveness), and weaker points (speed).

> 2. How you keep answerring the same questions over and over again.
> Style 'WIch IDE to use to develop with python' or worse 'Does python
> have a compiler'. If you think this is just a small part of the
> question here, take google and do a search on IDE or compiler on this
> group.

Hey, but these are fun holy war questions! Besides from usefulness I 
suspect many here *like* the whole advocacy aspects of this. And of course
in comp.lang.python it's very frequent some thread on some boring issue
that's been doing over and over again will grow some interesting
outcroppings on how morality evolved or how there could be life on
other planets and why the capslock ended up in the bizarre place it is
today (why, anyway?).

I'd like to thank this thread for pointing me at Tim Rue by the way. I vaguely
recall having seen him post somewhere (or perhaps his web page) many years
ago; I wonder where..  Seems to be active in the amiga community (but
I never was in that community, so that doesn't help me much :).

Regards,

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


From weeks at vitus.scs.agilent.com  Sat Feb 23 19:02:41 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Sun, 24 Feb 2002 00:02:41 +0000 (UTC)
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <1014508961.884758@cswreg.cos.agilent.com>

Tripp Scott (tripps81 at yahoo.com) wrote:

: Suppose I want to make a function that modifies its argument:  ..
: What is the Pythonic way of doing this?

The Python way is to train yourself not to want such things.

As is often the case, the Python way is the correct way.  It is easier to
read code if you don't have to wonder whether f(x) rebinds x.

Greg


From martin at v.loewis.de  Tue Feb 26 04:01:34 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 10:01:34 +0100
Subject: Convert String to Dictionary question
References: <mailman.1014686268.29712.python-list@python.org>
Message-ID: <m3y9hgfwgh.fsf@mira.informatik.hu-berlin.de>

"Jason Orendorff" <jason at jorendorff.com> writes:

> For what it's worth, *this* particular hole seems to have been
> patched.  But pickle can still call class constructors and
> __setstate__ methods and so forth, and it seems to me that
> plenty of standard lib constructors do at least a little disk
> access and socket stuff.  So it's still not safe.

If you want to restrict it to allow only restauration of a selected
number of classes, you need to perform unpickling in a restricted
environment. There you have explicit control over what builtins are
available; 

import rexec, pickle

l = pickle.dumps(["1",2,3])

class RExec(rexec.RExec):
  ok_builtin_modules = rexec.RExec.ok_builtin_modules + ('cPickle',)

r=RExec()
r.r_exec("import cPickle")
print r.r_eval("cPickle.loads(%s)" % repr(l))

works fine for me.

Regards,
Martin


From skip at pobox.com  Thu Feb  7 23:23:51 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 7 Feb 2002 22:23:51 -0600
Subject: threads killing other threads
In-Reply-To: <3C62E0D5.99C586DF@ccvcorp.com>
References: <1013105970.135322@cswreg.cos.agilent.com>
        <3C62E0D5.99C586DF@ccvcorp.com>
Message-ID: <15459.21207.175329.367778@12-248-41-177.client.attbi.com>

    Jeff> Create a global threading.Event object, named something
    Jeff> descriptive like Quit, and have every thread peek at it
    Jeff> occasionally.  If the Quit event is set, then the thread aborts.
    Jeff> If you need to be able to kill specific threads individually, then
    Jeff> you have a separate event object for each thread.

    Jeff> Of course, this *does* rely on the threads being well-behaved, and
    Jeff> not getting stuck in an infinite loop elsewhere, and so on....
    Jeff> but then, that's pretty much a requirement anyhow.

This note from the threading module docs suggests to me a different way to
do things:

    A thread can be flagged as a ``daemon thread''.  The significance of
    this flag is that the entire Python program exits when only daemon
    threads are left.

If all your worker threads are daemon threads, when your main thread exits,
the program should exit as well.  I haven't tried this, but may well try it
shortly.  I do have trouble getting a multi-threaded server to exit.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From tjreedy at home.com  Wed Feb 27 15:37:15 2002
From: tjreedy at home.com (Terry Reedy)
Date: Wed, 27 Feb 2002 20:37:15 GMT
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com>
Message-ID: <%rbf8.71116$X64.25961788@news1.rdc2.pa.home.com>

"Michael P. Soulier" <msoulier at nortelnetworks.com_.nospam> wrote in
message news:slrna7q9m0.4pb.msoulier at pmerd071.ca.nortel.com...
>     While I know about bound methods in Python, I'm wondering if a
full
> closure can be implemented. Can it?
>
>     Example: I want to pass an arbitrary code reference into a
closure and
> have all of its variables preserved.
>
>     ie.
>
>     print "the current count is %d" % counter; counter += 1

Default arg works as long as function is called properly:

>>> def prcnt(counter = [0]):
...   print "The current count is %d" % counter[0]
...   counter[0] = counter[0] + 1
...
>>> prcnt()
The current count is 0
>>> prcnt()
The current count is 1

Terry J. Reedy





From bokr at oz.net  Mon Feb 18 22:46:42 2002
From: bokr at oz.net (Bengt Richter)
Date: 19 Feb 2002 03:46:42 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl> <xyuvgcujsyl.fsf@sex.ifi.uio.no>
Message-ID: <a4shr2$49s$0@216.39.172.122>

On 19 Feb 2002 00:00:50 +0100, Jan-Fredrik Braseth <janfbr at sex.ifi.uio.no> wrote:

>"Patrick Vrijlandt" <p.vrijlandt at aig.azn.nl> writes:
>
>> Hi,
>> 
>> Previous discussions have narrowed down the line in telnetlib that is buggy.
>> The problem is probably that the server is sending you a special (Telnet)
>> command.
>
>Even though I haven't looked very closely it looks like they just forgot
>to initiate the 'opt' variable in one part of the if-else branch
>
It's not that the opt variable was not initialized, it's that opt is the
wrong variable. It has no legitimate value to bind to at that point. If it had
happened to be initialized, there would be the same error, but it would have been
silent, or would have indicated the wrong value if you turned on the program's
debug messages.

The fix is not to do anything with opt in that path. The
fix is to use the right variable in its place (i.e., c).

But I think Patrick is trying to point out that you could be receiving any of

      NAME               CODE              MEANING

      SE                  240    End of subnegotiation parameters.
      NOP                 241    No operation.
      Data Mark           242    The data stream portion of a Synch.
                                 This should always be accompanied
                                 by a TCP Urgent notification.
      Break               243    NVT character BRK.
      Interrupt Process   244    The function IP.
      Abort output        245    The function AO.
      Are You There       246    The function AYT.
      Erase character     247    The function EC.
      Erase Line          248    The function EL.
      Go ahead            249    The GA signal.
      SB                  250    Indicates that what follows is
                                 subnegotiation of the indicated
                                 option.

and, other than ignoring nulls and ^Q, the telnetlib.py code appears only to be
looking looking for IAC followed by

      WILL (option code)  251    Indicates the desire to begin
                                 performing, or confirmation that
                                 you are now performing, the
                                 indicated option.
      WON'T (option code) 252    Indicates the refusal to perform,
                                 or continue performing, the
                                 indicated option.
      DO (option code)    253    Indicates the request that the
                                 other party perform, or
                                 confirmation that you are expecting
                                 the other party to perform, the
                                 indicated option.
      DON'T (option code) 254    Indicates the demand that the
                                 other party stop performing,
                                 or confirmation that you are no
                                 longer expecting the other party
                                 to perform, the indicated option.

followed by an option that it rejects with proper protocol, or IAC followed by

      IAC                 255    Data Byte 255.
which when following itself is passed as a plain single character
with that value.

Have you tried enabling the debug print?
It looks like after
    tn = telnetlib.Telnet(server,port)
you should be able to turn on the debug messages by
    tn.debuglevel = 1
or
    tn.set_debuglevel(1)

Then you can see if any of the above are arriving.
Make the the opt => c change first.

It's not clear to me that it's generally advisable to ignore
all the command codes telnetlib.py ignores (which is the apparent
effect of the self.msg() call), but that will depend on what
you are communicating with, and what you're doing with it, I suppose.

Where is the mystery telnetlib.py author? ;-)

Regards,
Bengt Richter



From logiplexsoftware at earthlink.net  Mon Feb 11 19:50:49 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 11 Feb 2002 16:50:49 -0800
Subject: Apology
In-Reply-To: <00a001c1b35d$865173e0$0101010a@local>
References: <mailman.1013120898.32043.python-list@python.org>
	<a40urp$o10$1@peabody.colorado.edu>
	<558.804T2090T11654864threeseas@earthlink.net>
	<3C68274C.F6F702F5@ccvcorp.com>
	<00a001c1b35d$865173e0$0101010a@local>
Message-ID: <20020211165049.7c658abe.logiplexsoftware@earthlink.net>

On Mon, 11 Feb 2002 18:37:51 -0600
Chris Gonnerman wrote:

> Indeed.  He uses a nonstandard definition of "language" and applies
> "physics"
> to programming.  He can't program, you know; ten years on this project
and
> nothing to show for it.  If I had tried to develop something like this
for
> ten years and had nothing to show for it but a website, I'd give up.

But if you had to fight hoards of conspirators who were actively
suppressing the programs you were trying to write, you might be a little
distracted too ;)

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From peter at engcorp.com  Thu Feb 28 19:16:21 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 28 Feb 2002 19:16:21 -0500
Subject: Windows GUI - which way?
References: <a5k8t9$hrp$1@news.efn.org>
Message-ID: <3C7EC855.5844B19F@engcorp.com>

Spencer Doidge wrote:
> 
> I have a nice script running and doing useful work--in the Windows console
> mode. [...]  I'd like to make this palatable for others around the company 
> to use, and that means Windows GUI.
> So should I write up a program in my favorite style, Win32 API, and embed
> Python in it?
> Or should I write the whole thing in Python?
> 
> Would somebody like to comment on the pros and cons?

Write the whole thing in Python (my preference would be with wxPython)
and package it up with py2exe or equivalent.  Use a standard 
installer to distribute it and no one in the company will ever
know it was written with Python.  Except for the higher reliability.
And your greater responsiveness in implementing feature requests
quickly.  And how quickly you got it to them.  Etc. :-)

(We've written several utilities with wxPython and serial ports
running under Windows.  So far the wxPython side of things
is quite primitive since we're still learning, but even in 
their current state they're quite acceptable, and have no
real issues the way a more traditionally developed program
(i.e. Windows GUI with C/C++ or whatever) would have.

-Peter


From bokr at oz.net  Sat Feb 16 21:52:05 2002
From: bokr at oz.net (Bengt Richter)
Date: 17 Feb 2002 02:52:05 GMT
Subject: How to get CNAME with sockets ?
References: <a4lpg1.3vvdg2v.1@kserver.org>
Message-ID: <a4n5sl$9uj$0@216.39.172.122>

On Sat, 16 Feb 2002 14:14:25 -0800, Sheila King <usenet at thinkspot.net> wrote:

>I'm working on using Vipul's Razor (http://razor.sourceforge.net) in
>conjunction with a Python mail script that I'm working on.
>
>I'm studying some of the Perl code (and I don't really know Perl...makes
>it a bit tough). Anyhow, I've gathered that in the sub discover what I
>basically need to do, is query each subdomain of razor.vipul.net where
>the subdomains are a..z,1..9999 until I find one whose CNAME (or
>Canonical Name) is last.terminator.
>
I don't know what you are really doing, but that sounds like
systematically dialing all phone numbers in various area codes
until you get Arnold Schwarzenegger ;-)

Can't you get a copy of the 'phone directory and find him yourself?
Or do you need to visit with all those others?

Or is the metaphor totally inapplicable? ;-)

Regards,
Bengt Richter



From tchur at optushome.com.au  Thu Feb  7 15:07:17 2002
From: tchur at optushome.com.au (Tim Churches)
Date: Fri, 08 Feb 2002 07:07:17 +1100
Subject: Saving Browser State without Cookies
References: <D408B6AC060DB24790622E1A54204DB00823AA89@xchange.rightnowtech.com>
Message-ID: <3C62DE75.12039220@optushome.com.au>

"Orr, Steve" wrote:
> In this regard I'm looking into Albatross. As a welcome departure from the
> norm with frameworks, the number of lines of code is actually exceeded by
> the number of lines of documentation. ;-)

I'd be happy for that to be known as Tim's (not Timothy's) Axiom...

> In the book you mention Albatross along with many other Python solutions.
> (There's so many it's overwhelming to the relative newbie.) Preferring
> brevity and simplicity, Albatross looks like a very well engineered
> solution.
> 
> It would be great if someone could share their experience with using
> Albatross in a real world Python/Web implementation.
> 
> Anyone???????

I have been working on an Albatross-based app. Admittedly, the Object
Craft guys (the authors of Albatross) did the first version of the
front-end (i.e. the Web pages and associated logic) under contract, but
after spending a few hours reading the excellent Albatross docs, I have
been happily tweaking and extending the application. The application is
fairly complex: it is an epidemiological mortality analysis system, with
separate pages to collect fairly detailed user-specified parameters for
each of the basic epidemiological questions: who (died i.e. age and
sex), when (did they die), where (did they live), why (did they die i.e.
cause of death, not a higher metaphysical question...), and also what
(does the user want - tables, graphs etc). Selections on one page may
limit the valid choices on another page in complex ways, and the 5
parameter pages can all be accessed in random order. Albatross makes it
quite easy to implement the logic for all of this by transparently
saving state, so the Python code is relatively straighforward, and is
all neatly contained in Python modules, not scattered about in the HTML
templates. This also allows a unittest suite to be written to direcctly
exercise all of the application logic without having to sit and click on
Web pages ad nauseum. The line count for the application is only about
500 lines. The way in which Albatross strongly encourages the separation
of Python code and the HTML templates also means I can give the HTML
templates which define the "presentation layer" to someone with a better
developed aesthetic faculty than I for prettification, without fear that
they will come back with all the code mangled, because all the code is
elsewhere. I should also mention that one of the app's pages allows the
user to browse trees of ICD-8, ICD-9 and ICD-10 cause-of-death codes and
make selections. Each of these code sets contains between 5000 and 10000
elements (nodes on the tree, each with a code and a description).
Albatross successfully manages the display of these trees through its
<al-tree> tag and associated classes - in folded form first, showing
only the top levels of the ICD code hierarchies - the user can then
drill down to find the codes they require and select them.

The app is still in pre-production form, but it works perfectly and the
performance is excellent - I haven't needed to think about mod_python
but that is also an option with Albatross if you need more performance.
I encountered one bug in the Albatross code but the Object Craft guys
(there are three of them, two in Melbourne, one in Sydney) issued a fix
within 12 hours.

I believe that Object Craft plan to host some live demos of Albatross
apps on their Web server in the not too distant future, as well as
provide some more sample apps as starting points for would-be Albatross
developers.

Tim C



From erno-news at erno.iki.fi  Mon Feb  4 23:38:32 2002
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 05 Feb 2002 06:38:32 +0200
Subject: Scopes
References: <mailman.1012872608.6113.python-list@python.org>
Message-ID: <ku1yg04k6v.fsf@lasipalatsi.fi>

In article <mailman.1012872608.6113.python-list at python.org>,
<dude at deletia.com> writes:

| I am not sure of the advantage here, and would appreciate a little help

use http://groups.google.com/ to search the newsgroup archives for
mentions of the "default argument hack" from the era before nested
scopes. most of the joy of nested scopes is eliminating the need for
that.

  -- erno





From lienard.bruno at free.fr  Sat Feb  2 04:51:56 2002
From: lienard.bruno at free.fr (Bruno Liénard)
Date: Sat, 2 Feb 2002 10:51:56 +0100
Subject: Xmi parser
Message-ID: <3c5bb698$0$4609$626a54ce@news.free.fr>

I am looking for a xmi parser. Does anybody here have already written one ?

Thanking you by advance.

Bruno Lienard




From neal at metaslash.com  Fri Feb  8 14:55:37 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Fri, 08 Feb 2002 14:55:37 -0500
Subject: What is the most efficient way to do this?
References: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
Message-ID: <3C642D39.869614D0@metaslash.com>

MDK wrote:
> 
> What is the most efficient way to remove empty items from this list?
> 
> ['', '1', '2', '3', '', '456', '789', 'wow']

Assuming python 2.0 or higher:

seq = ['', '1', '2', '3', '', '456', '789', 'wow']
new_list = [ x for x in seq if x ]

Neal


From gte743n at cad.gatech.edu  Tue Feb 12 14:42:04 2002
From: gte743n at cad.gatech.edu (Joseph Holland King)
Date: Tue, 12 Feb 2002 19:42:04 +0000 (UTC)
Subject: catching error output
Message-ID: <a4br6c$lt1$1@news-int.gatech.edu>

in a small script i am using the popen command to run a command 
and then parse the output:
fs = os.popen('foo some_data')
fs.readline()
...
however when foo fails i need to catch that error and be able to parse
it. is there anyway to do this? fs = .. does not catch it just the 
normal output. thanks.

--
Holland King        
gte743n at cad.gatech.edu


From husam at macrozon.net  Wed Feb 20 08:10:59 2002
From: husam at macrozon.net (husam)
Date: Wed, 20 Feb 2002 14:10:59 +0100
Subject: Anybody can help on this?
Message-ID: <3C73A063.73AEB333@macrozon.net>

Dear friends,
I hvae the following situation:
I have 10000 files that needs to be processed by a program called ProFit
( Biochemistry program).  What the program does is it reads one molecule
and then reads another one and performs calculations to see how much
these molecules are similar. Sinse I have 10000 molecules It would takes
me a month typing all these file names and the corresponding commands to
make the calculations.  Since I'm a newbie to python, I thought it must
be possible to automate this process. I wrote this little script which
does the job nicely:

listit=os.popen('ls *pdb')
pdbs=listit.readlines()
listit.close()
passit=os.popen('/home/chemistry/profit ','w')
passit.write('reference Complex_1g.pdb ')
for PDB in pdbs:
     passit.write('\n')
     passit.write(' mobile ')
     passit.write(PDB)
     passit.write(' fit ')
passit.close()

As you see the "reference" molecule is the molecule to which the
"mobile" one should be compared. This is the output of the program :

Reading reference structure...
   Reading mobile structure...
   Reading mobile structure...
   Fitting structures...
   RMS: 8.765
   Fitting structures...
   RMS: 9.588

The problem is how can I save the output to a file? the only way to do
this at the moment is to run the scrip like this:
earth>$ myscript.py > output
What I want to do is to incorporate a code into the script to save the
results to a file. I tried with "os.popen() with "r" option but it
doesn't work because I don't know where to put it in the code. What I
don't understand is the following:
I have a file object of "os.popen()" open to send some commands to. When
can I retrieve the output of each command?
and last but not least, how can I print some stuff to the lines of the
output of the program so that the output will be like this:

Fitting structures...
RMS:10.343 ===>> My stuff
Fitting structures...
RMS: 6.389====>> My stuff


any help will be very much appreciated.

PS: I run python 1.52 on redhat 7.2




From anthony_allen at yahoo.com  Tue Feb  5 18:09:14 2002
From: anthony_allen at yahoo.com (Anthony Allen)
Date: 5 Feb 2002 15:09:14 -0800
Subject: Python 2.2 dynamic library problem on Compaq Tru64
References: <mailman.1012925289.31030.python-list@python.org>
Message-ID: <eecbdf57.0202051509.f3fd8b5@posting.google.com>

Glenn Stauffer <stauffer at swarthmore.edu> wrote in message news:<mailman.1012925289.31030.python-list at python.org>...

> I am using cx Oracle to access an Oracle 8.1.6.2 database.  Running my 
> scripts at the unix prompt, all is well.  When I run them through cron,
> I get this error:
> 
> > Traceback (most recent call last):
> > ? File "/usr/users/stauffer/python/dbamon/dbchk.py", line 3, in ?
> > ? ? import cx Oracle, os, sys, smtplib
> > ImportError: dlopen: cannot load
> > /usr/local/lib/python2.2/site-packages/cx Oracle.so

One difference between running a command from cron versus running it
from an interactive shell is that cron will not execute your
login/profile script.  As a result, you may be missing some
environment variables when running from cron.  In particualar, check
your LD_LIBRARY_PATH variable, as this may be needed to locate
libraries required by cx_Oracle.


From pythonlist at zumnet.com.br  Fri Feb 15 12:20:00 2002
From: pythonlist at zumnet.com.br (Rafael Mentz Aquino)
Date: Fri, 15 Feb 2002 15:20:00 -0200
Subject: Curses & Colors
References: <3C6C06FF.DAFFBB43@zumnet.com.br> <20020215050206.GB2867@lilith.hqd-internal>
Message-ID: <3C6D4340.F76600C5@zumnet.com.br>

Thanks a lot, Gerhard,

I found yesterday night how to do it. There wasn't any problem at all,
it' s just that the default terminal file brings terminals without support
for colors, but this is changeable. And I did it! Now I am exploring
everything I can with curses (backgounds, attrs, ...)

Thanks a lot for your answer...

Rafael Mentz Aquino

Gerhard H?ring wrote:

> Le 14/02/02 ? 16:50, Rafael Mentz Aquino ?crivit:
> > Hi, there...
> >
> > I'm creating some simple interfaces with curses with python under
> > FreeBSD, but I'm having some problems to change colors. I tried to do
> > something in /etc/termcap, but when I run curses.can_change_color(), I
> > always obtain 0.  My env term is cons25, but I don't know if python
> > uses the same terminal.  I'm running python 2.1.1. Can anybody help
> > me...? ;-))
>
> I recently fiddled with this, too. And it took me a while to get the
> curses code right. What I *can* do is give you some source that will
> work if color works at all:
>
> import curses
>
> def mymain(stdscr):
>     curses.init_pair(1, curses.COLOR_RED, curses.COLOR_WHITE)
>     stdscr.clear()
>
>     stdscr.addstr(0, 0, "This text is red on white.", curses.color_pair(1))
>
>     stdscr.refresh()
>     stdscr.getkey()
>
> curses.wrapper(mymain)
>
> I've no solution for how to activate color on your FreeBSD console,
> though. Maybe best ask on a FreeBSD group.
>
> Gerhard
> --
> This sig powered by Python!
> Au?entemperatur in M?nchen: -0.7 ?C      Wind: 4.1 m/s
>
> --
> http://mail.python.org/mailman/listinfo/python-list




From etoffi at bigfoot.com  Fri Feb 22 20:06:28 2002
From: etoffi at bigfoot.com (e toffi)
Date: 22 Feb 2002 17:06:28 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com> <mailman.1014085471.28317.python-list@python.org> <slrna73h0o.18i.grante@tuxtop.visi.com> <mailman.1014112771.5778.python-list@python.org>
Message-ID: <655abd4a.0202221706.280feee1@posting.google.com>

Gerhard =?unknown-8bit?Q?H=E4ring?= <gh_pythonlist at gmx.de> wrote in message news:<mailman.1014112771.5778.python-list at python.org>...

> So I'm still leaning towards Ada 95 and C++ for statically typed
> languages.
> 
> Gerhard

C++ sucks -- eiffel (http://smalleiffel.loria.fr/) is actively maintained.

--
etoffi


From threeseas at earthlink.net  Sat Feb 16 17:02:05 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 22:02:05 GMT
Subject: FSF award to...
Message-ID: <498.812T2576T10686134threeseas@earthlink.net>

http://www.fsf.org/press/2002-02-16-FSF-Award.html

Given what I've seen here.....

Well I guess Usenet ain't part of .....

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From matt at lorfeld.com  Fri Feb  1 01:08:25 2002
From: matt at lorfeld.com (mlorfeld)
Date: 31 Jan 2002 22:08:25 -0800
Subject: While loop error
Message-ID: <32c382ad.0201312208.166d5de3@posting.google.com>

I continue to get the error:
Traceback (most recent call last):
  File "Columbus:Desktop Folder:gaus.py", line 19, in ?
    n=list[i]
IndexError: list index out of range

my code is:
i=0
j=0
sum=0
list=[1,2,3,4,5,6,7,8,9,10,50,100]
list_length=len(list)
#G(n)=sum of first n natural numbers
while i < list_length:
	n=list[i]
	while j <= n:
		sum=sum+j
		j=j+1
	print sum
	j=0
	sum=0
	i=i+1

any ideas??


From gh_pythonlist at gmx.de  Wed Feb  6 13:59:10 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Wed, 6 Feb 2002 19:59:10 +0100
Subject: Autocoding project proposal.
In-Reply-To: <slrna62c6f.afn.philh@comuno.freeserve.co.uk>
References: <un0yxkupb.fsf@synopsys.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net> <slrna62c6f.afn.philh@comuno.freeserve.co.uk>
Message-ID: <20020206185906.GA1259@lilith.hqd-internal>

Le 06/02/02 ? 13:34, phil hunt ?crivit:
> On Wed, 06 Feb 2002 04:54:05 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
> >On 03-Feb-02 19:47:37 Peter Seebach <seebs at plethora.net> wrote:
> >>In article <1173.799T632T11653818threeseas at earthlink.net>,
> >>Timothy Rue <threeseas at earthlink.net> wrote:
> >>>Don't like the idea of autocoding huh?
> >
> >>Haven't been able to perceive whether or not you have an idea.  I have to
> >>say, you're talking to the wrong crowd.  You need people less mired in
> >>conventional ideas of software engineering.  Your best bet is almost
> >>certainly the C# people; they are certainly not mired in conventional ideas
> >>of software engineering.
> >
> >Why would  be interested in a programming language by a company that has
> >been found guilty of Anti-trust? Not to mention that language is not
> >available anywhere near as much as something like Python.
> 
> There is an open source implementation of C# as part of gcc, so I expect it runs
> on pretty much everything.

If you're speaking about the Mono project, then this is not true:

,----[ http://www.go-mono.com/c-sharp.html ]
| Questions and Answers
|
| Question 1: Why not write a C# front-end for GCC?
|
| I wanted to learn about C#, and this was an exercise in this task. The
| resulting compiler is highly object-oriented, which has lead to a very
| nice, easy to follow and simple implementation of the compiler.
|
| I found that the design of this compiler is very similar to Guavac's
| implementation.
|
| Targeting the CIL/MSIL byte codes would require to re-architecting GCC,
| as GCC is mostly designed to be used for register machines. The GCC Java
| engine that generates Java byte codes cheats: it does not use the GCC
| backend; it has a special backend just for Java, so you can not really
| generate Java bytecodes from the other languages supported by GCC.
`----

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.0 ?C      Wind: 4.1 m/s



From syring at email.com  Thu Feb 21 08:48:13 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 14:48:13 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <3b091a1c.0202210409.5beccbb7@posting.google.com>
Message-ID: <a52tte$4abm8$1@ID-7529.news.dfncis.de>

"cmkl" <cmkleffner at gmx.de> schrieb im Newsbeitrag
news:3b091a1c.0202210409.5beccbb7 at posting.google.com...
> One of the dilettantic NASA amateurs (David Bailey) once wrote the
ultimate

I meant of course the present director, who is essentially an accountant.

> fast and accurate MP-package 'mpfun' in Fortran 77 - later F90.
> A C++ implementation is available at:
> http://www.cs.unc.edu/Research/HARPOON/mpfun++/
> I havn't tested this C++ version and I don't think there is a a python
> wrapper available (should not be to hard ??) but it may worth trying.

Floating point simply is not meaningful in many problems of computational
geometry and a rational data type is not more difficult to use. I think,
some old mainframes had them even build in.

Karl M. Syring





From syring at email.com  Mon Feb  4 14:17:48 2002
From: syring at email.com (Karl M. Syring)
Date: Mon, 4 Feb 2002 20:17:48 +0100
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>  <b34d3835.0202040802.2cc7d0c0@posting.google.com> <mailman.1012840811.6930.python-list@python.org>
Message-ID: <a3mmov$19ebhk$1@ID-7529.news.dfncis.de>

"Laura Creighton" <lac at strakt.com> schrieb
> The e-book I want is made of electronic reusable paper.
> See: http://www.parc.xerox.com/dhl/projects/gyricon

AOL, AOL!
Btw: Is it water proof?

Karl M. Syring



From lac at strakt.com  Thu Feb 21 02:50:03 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 21 Feb 2002 08:50:03 +0100
Subject: Adding static typing to Python 
In-Reply-To: Message from ajeru@vknn.org (Alexander Jerusalem) 
   of "20 Feb 2002 13:00:29 PST." <24c39b2c.0202201300.3dfe7dbc@posting.google.com> 
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com> <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com> <24c39b2c.0202192354.765da777@posting.google.com> <mailman.1014210154.9297.python-list@python.org>  <24c39b2c.0202201300.3dfe7dbc@posting.google.com> 
Message-ID: <200202210750.g1L7o3Go032360@ratthing-b246.strakt.com>

> Alexander Jerusalem:
> The proposals of the types-SIG say that static typing is used on a per
> module basis. That means that weather there is or is not a performance
> improvement it's only going to affect you if you choose to use the
> static types.
> 
> Alexander

The problem is that static typing effects everybody in terms of code
reusability.  Let us say you make a terrific function which works on 
a string.  And in some way you specify that, yes, the first and only
argument to this function is a string.  

Now, let us assume I love your function and want to use it a lot.
But I want to use it on my own objects which are in some way 'string
like'.  Right now I have to read your code, and determine what
were the properties of a string you used, to determine whether my
stringlike object can already be used with your function.  I would
like to automate this, and be able to interrogate both your function
and my object and get a result  'will work' or 'won't work: reason
object does not have an islower method which function requires' and 
the like.

What I don't want to have to do is rewrite the code, because
Alexander Jerusalem was only thinking about strings when he wrote it.
Static type checking makes you think that 'strings' and 'ints' and the
like a real meaningful things rather than a convenient shorthand for
'is a sequence and has the following methods':

['capitalize', 'center', 'count', 'encode', 'endswith', 'expandtabs', 'find',
 'index', 'isalnum', 'isalpha', 'isdigit', 'islower', 'isspace', 'istitle',
 'isupper', 'join', 'ljust', 'lower', 'lstrip', 'replace', 'rfind', 'rindex',
 'rjust', 'rstrip', 'split', 'splitlines', 'startswith', 'strip', 'swapcase',
 'title', 'translate', 'upper']

See 'PEP 246: Object Adaptation - (Clark C. Evans)'
http://python.sourceforge.net/peps/pep-0246.html  for one attempt to
get the safety you want without the restriction I don't.

Laura Creighton



From sjmachin at lexicon.net  Sun Feb 24 01:02:05 2002
From: sjmachin at lexicon.net (John Machin)
Date: 23 Feb 2002 22:02:05 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com> <mailman.1014496298.16453.python-list@python.org>
Message-ID: <c76ff6fc.0202232202.6f4b3ded@posting.google.com>

Mike Dean <klaatu at evertek.net> wrote in message news:<mailman.1014496298.16453.python-list at python.org>...
> * Bob Roberts <bobnotbob at byu.edu> [2002-23-02 11:52]:
> > What would be a good way to replace every one or more spaces (" ") in
> > a string with just one space?  Or replace any number of newlines with
> > just one?
> 
> This is a perfect job for Python's regular expression (RE) library. For
> example:
> 
> import re
> 
> # Replace all occurances of one or more spaces with a single space
> re.sub(' +', ' ', mystring)

Not using raw strings with re? Not a good habit ...

> # Ditto for newlines
> re.sub('\n+', '\n', mystring)
> # And, to combine the two into one operation:
> re.sub('(\n+| +)', '\1', mystring)

and *whammo* here's the result.

>>> mystring = "aaaa\n\n\n\nbbbbbbb\n\n\nbb\nbbbb"
>>> re.sub('(\n+| +)', '\1', mystring)
'aaaa\x01bbbbbbb\x01bb\x01bbbb'
>>>


From caljason76 at yahoo.com  Wed Feb 20 20:35:34 2002
From: caljason76 at yahoo.com (Jason)
Date: 20 Feb 2002 17:35:34 -0800
Subject: strings and sort()
Message-ID: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com>

How do I sort the characters in a string, so far I use this:
        a="qwerasdfzxcv"
        b=[x for x in a]
        b.sort()

Why doesn't sort() return the sorted list.  I would like to chain it
to other operations:
        b=[x for x in a].sort()

-j


From aldo at nullcube.com  Fri Feb  8 19:50:40 2002
From: aldo at nullcube.com (Aldo Cortesi)
Date: Sat, 9 Feb 2002 11:50:40 +1100
Subject: readlines() a file in 2 parts.
In-Reply-To: <87vgd85a7h.fsf@cosette.nampaz.com>
References: <87vgd85a7h.fsf@cosette.nampaz.com>
Message-ID: <20020209005040.GB5395@nullcube.com>

Thus spake Jason Price (jprice at cosette.nampaz.com):

> I have several files to read that logically break into 2
> parts.  It would be much cleaner to write 2 different 'for
> line in file.readline() :' loops, but the break point is
> different in each file, so I don't know the starting line.
> What I tried to do origionally was:
> 
> for line in file.readlines() :
>    if condition :
>       break
>    else:
>       do thing 1
> 
> for line in file.readlines()
>    do thing 2
> 

> I tried both readlines and xreadlines, and neither seem to
> do what I'd like.  Is there a way to do this cleanly, or
> should I punt, and just make the loop more complex?



Jason,

Every time you run file.readlines() a brand new list of
lines is created from the file. What you really want to do
is to create an iterator - an object you can loop over, and
which will keep your "place" in the file. This is excactly
what xreadlines() does.

The general syntax you give above is correct, with a small
modification:

itr = open("file").xreadlines()

for line in itr:
    if condition:
        break
    else:    
        print line

for line in itr:
    print line  

If you are using Python 2.2, it is probably in better taste
to replace the first line with:

itr = iter(open("file"))

For more info on iterators in Python 2.2, see this article
by David Mertz:

http://www-106.ibm.com/developerworks/linux/library/l-pycon?t=grl,l=252,p=iterators



Cheers,



Aldo



---
Aldo Cortesi
aldo at nullcube.com




From jason at jorendorff.com  Sat Feb 16 23:42:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 22:42:28 -0600
Subject: Help! IIS and Python
In-Reply-To: <oqcu6uc4poo2akj76sg1lh76n8tp7kkdcu@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCECCDDAA.jason@jorendorff.com>

> ...
> for i in range(5):
> %>
>   <%= formatter % i %>
> 
> Perhaps the issue is that I don't know how the <%= form is actually
> implemented in Python/ASP.  If it is just a textual substitition
> [...], then I understand why it is not possible.

It is a textual substitution.

I'd be tempted to try something like:

...
for i in range(5):
   %>
   <p>Foo...
   <%= formatter % i %>
      ...bar</p>
   <%
%>

Note that I have the '%>'s and '<%'s rather carefully
lined up as if they were Python statements.

## Jason Orendorff    http://www.jorendorff.com/




From sdm7g at Virginia.EDU  Wed Feb 13 15:56:32 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Wed, 13 Feb 2002 15:56:32 -0500 (EST)
Subject: Python Gotcha with Octal Numbers
In-Reply-To: <a4efi4$1ebg78$1@ID-11957.news.dfncis.de>
Message-ID: <Pine.OSX.4.43.0202131521550.640-100000@d-128-61-180.bootp.virginia.edu>


On Wed, 13 Feb 2002, Emile van Sebille wrote:

> > -- Steve "the reformed packrat" Majewski
> >
>
> Is there a step program to reform?  ;-)
>
> desperately-seeking-anything-ly y'rs,


Step 1: Move to a smaller house. While sorting, packing and triaging
        get a significant other to keep repeating: "Where are you
	going to put it all!" ( If you can't find anyone, I'll ask
	my wife if she'll hire out. )

Step 2:	Keep repeating to yourself, as you throw out piles of magazines
	and journals: "I wonder if I can get this all on CD-ROM?"

Step 3: Remember: the Web, Google/deja-news, the internet wayback
	machine, ...  You can probably find it on the web faster
	than you can find it on paper!  Visit the library and
	remind yourself that there are people who get *paid* to
	keep all this junk. (But STAY AWAY from any books by
	Nicholson Baker!)

Step 4: Some used book stores (at least here in C'ville!) will come
	to your house with a truck, sort through your books for
	what they want, haul it away, and WRITE YOU A CHECK!
	I got almost $1,000 ... AND I STILL HAVE TOO MANY BOOKS!
	(and now I get to visit them at the used bookstore, and
	 sometimes, ransome a dearly departed one back again! )

Step 5:	Find people who actually want the stuff to give it away to --
	it feels better if you think it's going to a good home.

Step 6:	Rather than just dumping it all in the trash, I encourage
	you to make at least one trip to the dump or landfill
	yourself -- the impressive pile of STUFF is inspiring!

Step 7:	Tie some of your possessions to yourself with rope and/or
	chain and drag them around with you for a couple of days.
	Start using your car as an office.

Step 8:	As soon as you get confortable enough in the new house to
	start accumulating and filling it up again ... MOVE TO A
	NEW HOUSE ( actually, we're planning on moving back to the
	old house again next year. )

Step 9:	If all else fails, consider moving into a van or mobile home,
	or just taking to the road with everything on your back.
	Moving *really* far away might also be a good idea -- I gazed
	longingly at the jobs in Sweden on the bulletin board at
	Python-10! (although my secret dream is to move to Kabul and
	open the first skate park in Afghanistan, but I've been
	unable to talk my 14 year old son -- who's the one who actually
	knows something about running one -- into the adventure.)

Step 10: Did I mention moving ? If you can't move to a different
	house, just try swapping rooms: Piles grow when left to sit.

Step 11: There is no step 11!

Step 12: Memorize and delete this message! (Your computer's disk
	is filling up at this very moment. )


-- Steve






From gbeasley at tsa.ac.za  Tue Feb 26 02:44:33 2002
From: gbeasley at tsa.ac.za (Grant Beasley)
Date: Tue, 26 Feb 2002 09:44:33 +0200
Subject: Dictionary type access to list of instances
References: <a5dcgh$b22$1@ctb-nnrp1.saix.net> <3C7A5AD4.9090404@wi.mit.edu>
Message-ID: <a5fedu$qe1$1@ctb-nnrp1.saix.net>

Thanks for the response

Firstly, I really like your NotEqual class, it solves a bunch of other
problems I'm experiencing, because as you point out, None can sometimes be a
valid value.

The only part I don't like about your solution is the fact that the
properties array will be generated for all instances, whereas, using a for
loop that returns the instance when it finds it, not all instances need to
be inspected for the property. So this might result in better efficiency (In
the best case where the instance required is the first in the list).

Obviously catching the attribute exception is a better solution though, and
I'll probably integrate that into my solution. Currently I'm using my filter
solution just because I tend to prefer things I can do on one line instead
of having to write a complete new subroutine (which is why I enjoy Python so
much), but I'll probably adapt that soon.

What I'm ultimately aiming for is a database-like view of a list of
instances, so that I can use SQL commands to get instances. So in the
analogy, a list of instances would be a table, with each instance a record,
and each property a field.

eg.
class User:
    def __init__(self, username, accesslevel):
        self.username = username
        self.accesslevel = accesslevel

list = [User('Bob', 1), User('Jim', 2), User('Pete', 2)]

Then using my imaginary function ListSQL:

ListSQL("select * from list where accesslevel = 2")
would return a list consisting of the Jim and Pete instances.

or ListSQL("select username, accesslevel from list where accesslevel = 2")
would return [('Jim', 2), ('Pete',2)] - i.e. a List of tuples.

Is this an interesting / useful idea? Has it been done before? Any comments?

Thanks
Grant

"Brian Kelley" <bkelley at wi.mit.edu> wrote in message
news:3C7A5AD4.9090404 at wi.mit.edu...
> Grant Beasley wrote:
>
> > Hi
> >
> > If I have a list of instances, each with a property which I want to use
as a
> > key to access the list in dictionary-like manner, what would be the best
 way
> > of doing this?
> >
> > eg.
> > func get(list, property, value):
> >     for x in list:
> >         if getattr(x, property) == value:
> >             return x
> >
>
> In this usage, getattr will raise an Attribute error if property doesn't
> exist which might not be what you want.
>
> Here is an attempt using list comprehensions and the built-in list.index
> function.  I have created a NotEqual class so that getattr can return a
> default value is never equal to any other instance.  This is because
> None might be a valid value for a property.
>
> class NotEqual:
>      """instances of this class are never equal to anything"""
>      def __eq__(self, other):
>          return 0
>
> def get(inputList, property, value, default=NotEqual()):
>      try:
>          properties = [getattr(x, property, default) for x in inputList]
>          return inputList[properties.index(value)]
>      except ValueError:
>          raise ValueError, "attribute %s with value %s not found"%(
>              `property`, `value`)
>
> if __name__ == "__main__":
>      class Foo:
>          def __init__(self):
>              self.dog = 1
>
>      foo = Foo()
>      print get([foo], 'dog', 1)
>      try:
>          get([foo], 'dog', 2)
>      except ValueError, msg:
>          print 'error caught:', msg
>
>      try:
>          get([foo], 'cat', 'raisin')
>      except ValueError, msg:
>          print 'error caught:', msg
>




From Aristotle_00 at yahoo.com  Sun Feb  3 15:11:08 2002
From: Aristotle_00 at yahoo.com (Mark)
Date: 3 Feb 2002 12:11:08 -0800
Subject: Elegant solution needed: Data manipulation
References: <mailman.1012663144.30202.python-list@python.org>
Message-ID: <12e3779b.0202031211.617e8d7e@posting.google.com>

"Jason Orendorff" <jason at jorendorff.com> wrote in message  

> I'm a bit surprised you didn't take Steve's code and
> say "Yay, I'm done!".  I'm extra surprised you didn't
> reply to any of the responses to your message or
> ask any followup questions.  Your approach might be
> faster than Steve's (I'm not sure), but it seems a
> bit more fragile.

It had nothing to do with being unappreciative, that's for sure :). 
No, the only real problem with Steve's code was a need to have infix
expressions.  Also, I think I didn't get his message right away b/c of
usenet server update problems.

Anyway, the other motivation for my attempt is that I'm both _using_
and _learning_ python right now.  So, I'm happy to spend a little more
time on a problem so that I can learn some more.

I didn't ask followups b/c I think I saw how the solutions were
working and could figure out how they would work for me (of course, I
could be delusional).

Regards,
Mark


From starx at pacbell.net  Wed Feb  6 16:14:30 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 21:14:30 GMT
Subject: What is the command to determine an object's type?
References: <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com> <mailman.1013019261.19443.python-list@python.org>
Message-ID: <Xns91AD86B05A35CRASXnewsDFE1@130.133.1.4>

Laura Creighton || Wed 06 Feb 2002 10:12:20a:

> I think that everybody should read Design Patterns because it is the
> single best way I know of to improve your code.  (And your life.
> Life is more fun when your code is more maintainable.)  However,
> for a Python programmer, there is one problem with the original
> text:  Design Patterns.

In my opinion (;), 'Code Complete' is up there around it too. It has 
nothing to do with objects, but it has a lot of material on designing at 
the routine level, how to make good loops and stuff like that, and you 
still need to be able to do that kind of thing in your objects methods. It 
also discusses general programming/intelectual honesty/metahphors and 
similar, and has some basic information on highel level design of 
functional programs. But Design Patters is definatly a great book if you 
want to do anything with objects :). (For c++, i'd reccomend those two, 
'accelerated c++', and stroushtrup-whatsit's bible :)
 
> It is written for C++.
> 
> This means that while a lot of the text is about the really good stuff,
> the patterns, at least half of it could be entitled 'my struggles with
> the C++ type system'.  And since you came to this discussion because
> you wanted to test the type of something, which is something that
> Mark McEahern and I both think that you probably ought not to be doing
> at all, you may get the wrong lesson from reading that book now.  
> 
> There is another book, The Design Patterns Smalltalk Companion,
> (Alpert Brown Woolf), which is meant to be read at the same time
> as the Design Patterns.  Smalltalk does not have declared types
> either.  As a result, patterns in Python end up looking a lot 
> more like Patterns in Smalltalk.  I find the Design Patterns
> Smalltalk Companion to be a lot easier to understand than its
> parent, as well.  But when you get a look at the Smalltalk syntax,
> you will see one of the reasons why Smalltalk has not taken over
> the world by now, mores the pity.

Hm, i may have to take a look at this one. Thanks for pointing out that it 
exists :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From carifio.nospam at nospam.usys.com  Mon Feb  4 19:07:40 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Tue, 05 Feb 2002 00:07:40 GMT
Subject: What's the current state of the "interfaces" proposal?
Message-ID: <gnF78.18266$zJ3.1417982@typhoon.maine.rr.com>

I'm not sure if this is the right place to ask this question. Is there a
better forum?
I started at http://python.sourceforge.net/peps/pep-0245.html and got a
little lost
following some of the links. It looks like "interfaces" has been "accepted"
as a future
part of python. Is that accurate? What's the time frame? Pl. advise. Thanks.




From marknenadov at winningteam.com  Fri Feb 22 14:27:16 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Fri, 22 Feb 2002 14:27:16 -0500
Subject: XML RPC Woes
Message-ID: <20020222.142716.1960709859.1525@d57-47-79.home.cgocable.net>

I am in a bit of a weird situation with the 'xmlrpclib' and
'SimpleXMLRPCServer' modules.  I have a simple server that registers one
simple function and then serves 'forever'. The client I made simply
connects and triesto execute that function via XML RPC (also passing one
parameter ot it) and print the result. Both the server and the client are using
Python 2.2.

That was working just fine. However, all of a sudden I have not been
able to get it to work at all. The server just keeps on returning "500
Internal Error" now. I have investigated my own code as well as parts of 
the SimpleXMLRPCServer and xmlrpclib (printing out values of variables
that may be causing problems, etc).  The XML that is being sent by XML 
RPC to the server seems to be good. Does anybody have a clue what may be
causing this?

Any help is greatly appreciated,
~Mark Nenadov


From mcherm at destiny.com  Tue Feb 12 12:51:53 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 12 Feb 2002 12:51:53 -0500
Subject: I wish.............
Message-ID: <3C695639.5010400@destiny.com>

Mark Hammond writes:

> Just to be clear, I believe that Pippy is a port rather than an 
> implementation.  It is a port that has ripped the guts out of the 
> implementation (for size purposes) but that is still a port :)


Thanks for the correction.
-- Michael Chermside






From jason at jorendorff.com  Sun Feb 24 23:52:35 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 24 Feb 2002 22:52:35 -0600
Subject: The good "import" convention?
In-Reply-To: <a50631f6.0202221235.5858bdcd@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGENNDEAA.jason@jorendorff.com>

newgene writes:
> Which one is better or they are the same for the performance?

A single set of imports, at the top of each file, is better,
because that's the convention.  "Prefer the standard to the
offbeat."

Which is faster doesn't really matter --which is lucky,
because there's no simple answer to that question anyway.

If you need help improving performance in an actual program,
post some sample code.  Otherwise, relax.  :)

## Jason Orendorff    http://www.jorendorff.com/




From h.jehadalwan at student.kun.nl  Fri Feb 22 10:28:31 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Fri, 22 Feb 2002 16:28:31 +0100
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com> <3C74AAA0.D3A050EE@macrozon.net> <3C7571A2.41D4E9E2@ccvcorp.com> <3C76551A.7060107@student.kun.nl>
Message-ID: <3C76639F.8070700@student.kun.nl>

  nevermind, I found it.





From laz at strakt.com  Fri Feb  8 11:47:59 2002
From: laz at strakt.com (Fredrik Juhlin)
Date: Fri, 8 Feb 2002 17:47:59 +0100
Subject: Label width
In-Reply-To: <a40ump$d9c$1@netnews.upenn.edu>
References: <a40ump$d9c$1@netnews.upenn.edu>
Message-ID: <20020208164759.GH3733@strakt.com>

On Fri, Feb 08, 2002 at 04:34:33PM +0000, Salim Zayat wrote:
> Is there a way to determine the width and height of a Label widget (in 
> pixels) before it is actually drawn to screen?
label.winfo_width() and label.winfo_height() should do it. You might need
to call label.update_idletasks() first though, to make sure you get the
correct values.

//FJ



From sholden at holdenweb.com  Tue Feb 19 18:11:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 19 Feb 2002 18:11:04 -0500
Subject: Safe to use asynchat?
References: <slrna6t7f8.n2k.mlh@vier.idi.ntnu.no>
Message-ID: <QXAc8.11346$Zp2.1133224@atlpnn01.usenetserver.com>

I  believe the lack of documentation is accidental, simply due to high
workload at the documentary coal-face. You'll just have to flog Fred Drake
harder!

I was a bit perturbed about its absence when I wrote <shameless plug>
"Python Web Programming", but I bit the bullet and wrote it up there
</shameless plug>. It's a very well thought-out system.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py

"Magnus Lie Hetland" <mlh at vier.idi.ntnu.no> wrote in message
news:slrna6t7f8.n2k.mlh at vier.idi.ntnu.no...
> I just wondered... Is asynchat "here to stay"? The fact that it is
> undocumented made me worry that maybe it was deprecated or that it
> might suddenly disappear from the distribution... What's its status?
>
> --
> Magnus Lie Hetland                                  The Anygui Project
> http://hetland.org                                  http://anygui.org




From maxm at mxm.dk  Fri Feb 22 05:05:57 2002
From: maxm at mxm.dk (Max M)
Date: Fri, 22 Feb 2002 10:05:57 +0000
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <3C761805.7040507@mxm.dk>

Rajarshi Guha wrote:

> Why can't I use d.keys().sort() directly? And is there any other more 
> concise (elegent?) way to loop over sorted dictionary keys?

You could just create your own dict that returns sorted keys:

regards Max M

#################################################

import UserDict

class sDict(UserDict.UserDict):

     def keys(self):
         keys = self.data.keys()
         keys.sort()
         return keys

     def values(self):
         return [self.data[key] for key in self.keys()]

     def items(self):
         return [(key, self.data[key]) for key in self.keys()]


sDict = sDict()
sDict[1] = 10
sDict[2] = 20
sDict[3] = 30
sDict[4] = 40
sDict[5] = 50

print sDict
for key in sDict.keys():
     print sDict[key]

for key, val in sDict.items():
     print '%s: %s' % (key, val)

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

{5: 50, 4: 40, 3: 30, 2: 20, 1: 10}

10
20
30
40
50

1: 10
2: 20
3: 30
4: 40
5: 50



From fperez528 at yahoo.com  Sun Feb 17 15:28:17 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 17 Feb 2002 20:28:17 +0000
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <5d283efe.0202162142.53e3b402@posting.google.com> <roy-05958C.07303117022002@reader2.panix.com> <ubsenehjm.fsf@online.no>
Message-ID: <a4psgi$5jo$3@peabody.colorado.edu>

Syver Enstad wrote:

> mlh at vier.idi.ntnu.no (Magnus Lie Hetland) writes:
>> 
>> Jorg: Du kan l?ne denne av meg :)
>> 
> 
> Begynner det ? bli s? mange norskinger p? comp.lang.python at vi kan
> begynne ? skrive postingene v?re p? norsk ;-)

Would you folks mind sticking to English in public discussions? Use private 
email for other languages, please.

Thanks,

f.


From jkraska at san.rr.com  Fri Feb  8 12:03:44 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 17:03:44 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com>
Message-ID: <kk186ugk4hpo7m0eca81bv2lm2d0smnjiv@4ax.com>

>1. Nobody or nothing gives you that right.

That's the _one_ thing you're mistaken about. I can
entertain the idea of everything else you said, but
the notion that people don't have the right to do this
is just plain wrong. Plainly is the right not only held,
but it can be exercised.

C//



From arturs at iidea.pl  Sat Feb 16 07:41:45 2002
From: arturs at iidea.pl (Artur Skura)
Date: Sat, 16 Feb 2002 12:41:45 +0000 (UTC)
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net> <20020215225431.7bdef34e.risnes@alfanett.no>
Message-ID: <slrna6sovm.i3o.arturs@aph.waw.pdi.net>

Andr? Larsen Risnes wrote:
>         else:
>             Label(root, text=part).pack(side=LEFT)
>         Label(f, ...)
>         f.pack(side=TOP, anchor=W, fill=X, expand=YES)
> 
> (haven't tested this last part, though)

Works like a miracle (with small modifications):


from Tkinter import *

root = Tk()

file=open("README","r")
for line in file.readlines():
    f = Frame(root)
    for part in line.split("%"):
        if part == "text":
            Text(f, width=8, height=1).pack(side=LEFT)
        else:
            Label(f, text=part).pack(side=LEFT)
        f.pack(side=TOP, anchor=W, fill=X, expand=YES)

root.mainloop()

Some things don't line up still (http://7bulls.com/~askura/tk1.png), 
but it's easy sailing now.

Thanks to everybody!

Regards,
Artur
-- 
http://www.wolneprogramy.org


From maxm at mxm.dk  Mon Feb 18 08:36:11 2002
From: maxm at mxm.dk (Max M)
Date: Mon, 18 Feb 2002 13:36:11 +0000
Subject: There is more than one way to do it - and for no apparent reason.
References: <3C70ECB1.2090709@mxm.dk> <po6c8.82758$3J1.2641405@news2.tin.it>
Message-ID: <3C71034B.20508@mxm.dk>

Alex Martelli wrote:
> Max M wrote:
>         ...
> 
>>Basically you can say that JScript uses the dictionary notation for
>>getting attributes. As fas I can see this is smarter than the way we do
>>
> Maybe you can't see very far....?-)

That's probably true (-6,-6.5)  on left and right eye ;-)

> adict = dict()
> somerandomwordfromsomewhere = 'get'
> adict[somerandomwordfromsomewhere] = 23
> 
> Now what should adict.get be?  The number 23, or a bound method that you
> can save somewhere and later call with an argument or two?

Well it should be 23 because you have assigned it to that.

> If items and attributes were confused, you basically couldn't use
> bound methods any more -- you could never be sure if by some
> mischance somebody has set an item called get, or keys, or items,
> or setdefault, or ... whatever other name you might like to use.

Yes I see your point here. The reason that it can be used in JScript is 
because you allways itterate it by key like::

     for (var key in myObj){
         val = myObj[i];
         print(key, val);
     }

And if we did the same in Python we would risk overwriting some of the 
bound methods by accident::

     dirtyList = ['keys', 'values', 'items']

     newDict = {}
     for item in dirtyList:
         newDict[item] = 42 # this would indeed make the dict hard to 
use :-)

So I agree that unifying object and dict would be a bad idea.

But we could still use the dict like notation on an objects without any 
greater risk than that imposed by getattr/setattr. And that was in fact 
my main idea behind the question.

well I can allways just mix-in my own class

class IWILLHaveItMyWay:

     def __getitem__(self, key):
         return getattr(self, key)

     def __setitem__(self, key, value):
         setattr(self, key, value)

regards Max M



From marklists at mceahern.com  Fri Feb  8 01:01:43 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 22:01:43 -0800
Subject: FixedPoint
In-Reply-To: <000601c1b064$29d3a940$445d4540@Dell2>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEJFKBAA.marklists@mceahern.com>

[Tim Peters]
> Note that I said "non-numeric object".  FixedPoint supports a rich set
> of auto-coercions from types that "make sense".  I'm not a fan of
> semantically senseless comparisons, e.g.
>
> >>> 5 < '5'
> 1

[Brian Quinlan]
> Even if you disagree with that example, I would argue that equivalence
> is a special case. If you cannot convert a => b in a meaningful way then
> a != b.

I've already solved my problem.  I appreciate the discussion--I've learned a
lot from it.

I can see Tim's point about how comparing a FixedPoint instance to None like
this doesn't make sense:

	if f == None:

However, the code that's doing this isn't under my control.

As a practical matter, I was faced with these options:

1) Change a third party source that we use heavily (PyPgSql, where the
comparison to None is made).
2) Change a third party source that is not maintained (FixedPoint) where the
fix is not likely to break anything.
3) Subclass the third party source and make the fix there.

I choose 2.  Since we use CVS, I just did cvs import on a folder with
FixedPoint, which creates a branch where the source lives.  Then I made my
simple modification to __cmp__ and checked it into the trunk.

Thanks,

// mark




From gavrilov at attbi.com  Wed Feb 27 17:16:17 2002
From: gavrilov at attbi.com (Alexander Gavrilov)
Date: Wed, 27 Feb 2002 22:16:17 GMT
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com> <29e28c51.0202270014.61dbd1d0@posting.google.com>
Message-ID: <RUcf8.13561$TB2.72289@rwcrnsc52.ops.asp.att.net>

Are you working in Microsoft, aren't you?

Alexander

"Cimarron Taylor" <cimarron+google at taylors.org> wrote in message
news:29e28c51.0202270014.61dbd1d0 at posting.google.com...
> I have used the following tricks before with Java code, some of them
> should be useable with python.
>
> 1. Use a source-to-source translator to convert all your class, method
>    and variable names into gibberish.
>
> 2. Introduce special "dispatch" classes to obscure the communications
>    paths.  E.g instead of calling foo.bar(), call dispatcher.bar(foo)
>    where dispatcher is a dynamically instanciated object.  Have lots
>    of dispatchers and periodically switch them around in meaningless ways.
>
> 3. Add lots of decoy code.  Put bar() methods in lots of classes which
>    are never used.   Include the gibberish versions of a few previous
>    versions of your code with logic deliberately reversed.  Intermix
>    meaningless logic with the code which does the actual work.  Bulk up
>    your product with a few dozen distributions of other redistributable
>    software.
>
> 4. Hide some initialization logic outside your program.  Have your
>    program step reload classes.  If your program manages external files
>    (e.g. userids or passwords), have it hide some critical initialization
>    logic there.  Encode some of your program state in records in a
database
>    which requires a separate initialization and/or installation step.
>
> 5. Distribute your program in separate pieces. Have your program partially
>    depend on external libraries.
>
> 6. Partition your program's operation into separate phases and/or
processes.
>    Have one process generate code on the fly to be executed by the other.
>    Hide important information in otherwise meaningless status messages
>    exchanged between the two.
>
> Most of these things will make your program more difficult to reverse
engineer.
> Heaven help you if you ever need to fix a bug in it!
>
> Cim




From drgergl at mindspring.com  Thu Feb 14 09:14:31 2002
From: drgergl at mindspring.com (Greg Landrum)
Date: Thu, 14 Feb 2002 06:14:31 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net>
Message-ID: <a4gglj$ic1$1@slb7.atl.mindspring.net>

"Marcus Stojek" <stojek at part-gmbh.de> wrote in message
news:3c6bbe96.12195968 at news.easynews.net...
> Hi,
>
> I am still struggling with tuple-lists.
>
> How can I sort a list of tuples (all tuples have 4 float items)
> by the second (or third or n-th) tuple item.

Here's the basic incantation:
>>> foo = [(1,2,3),(1,3,3),(1,0,3)]
>>> foo.sort(lambda x,y:cmp(x[1],y[1]))  # sort on the second element
>>> foo
[(1, 0, 3), (1, 2, 3), (1, 3, 3)]

> And I have to
> do it fast.

well now, that's a different problem. :-)
Seriously, the default python sort is pretty fast, so give that a try.

-greg




From graham_guttocks at yahoo.co.nz  Tue Feb 19 20:12:25 2002
From: graham_guttocks at yahoo.co.nz (Graham Guttocks)
Date: Tue, 19 Feb 2002 17:12:25 -0800 (PST)
Subject: why does popen2 silently ignore IOError?
Message-ID: <20020220011225.59044.qmail@web10308.mail.yahoo.com>

Greetings,

Can anyone explain why the popen2 module silently ignores an IOError?
I'm trying to use it to send mail instead of os.popen() to avoid
invoking /bin/sh.  

If all the command-line arguments are proper, things work fine, but if
I open a bogus command-line, popen2 doesn't complain, neither during
the write, nor close.  Take the following simple example:

% python2
Python 2.1.1 (#1, Aug 13 2001, 19:37:40) 
[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-96)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import popen2
>>> def sendmail(s):
...     args = ['/bin/nonexistent', '-f', 'graham', 'testacct']
...     pp = popen2.popen2(args)[1]
...     pp.write(s)   
...     pp.close()
... 
>>> sendmail('testing')
>>> 

As you can see, no exceptions were raised despite the fact that there
is no such file as '/bin/nonexistent'.  Why is this?

Regards,
Graham


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From usenet at thinkspot.net  Sat Feb  9 08:31:57 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 09 Feb 2002 05:31:57 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com>
Message-ID: <a42c8e.3vvbc1l.1@kserver.org>

[posted and mailed]

On 9 Feb 2002 02:22:44 -0800, wookiz at hotmail.com (wooks) wrote in
comp.lang.python in article
<8852d33c.0202090222.34be7da2 at posting.google.com>:

> New to the language, new to OO but I do understand the basic concepts,
> not at all new to programming, trying to do the right thing by
> searching archives before asking questions. Yet I feel I need to be
> doing a university degree course in philosophy or mathematics to
> follow almost any discussion. 

...<snippage>...

You might want to try the Tutor mailing list:
http://mail.python.org/mailman/listinfo/tutor

I highly recommend it.
 
> Bought Python for Win32 book and already been caught out (Part 1 was a
> breeze part 2 and I shuddered to a grinding halt). For perspective I
> am a tester who wants to learn Python because of it's excellent unit
> test framework and it's scripting glue like abilities (not interested
> in building applications). If vbunit was as easy to figure (or had as
> good a tutorial) as PyUnit (thank you Mark Pilgrim) I probably never
> would have come here.  As it is perhaps getting to grips with vbunit
> might be the lesser evil.

Ah, well, that's Windows for you. You might want to dabble and get your
feet wet in the language a bit before diving into the win32 module(s).
They are more complicated to work with, but that's just because it's
Windows you're dealing with. Python can't fix that interface.

Have you been to this page?
http://www.python.org/doc/Newbies.html
Some of the tutorials listed there might be more useful in the "language
learning stage" than that win32 book. Or not. I'm not sure where you are
with the whole thing.

You might also want to try the win32 mailing list:
http://mail.python.org/mailman/listinfo/python-win32

> The benefits of using of Python as a scripting glue like language are
> widely touted but don't seem (I am probably wrong but it's a 1st
> impression) to be well catered for. As I said I am not interested in
> building applications. Is there a forum for the dumber less demanding
> Python user?

See above.

Hope this helps,

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From brian at sweetapp.com  Tue Feb 12 03:23:44 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 12 Feb 2002 00:23:44 -0800
Subject: how to compare two strings?
In-Reply-To: <7xofivazj1.fsf@ruckus.brouhaha.com>
Message-ID: <000501c1b39e$9799faa0$445d4540@Dell2>

>>> print cmp.__doc__
cmp(x, y) -> integer

Return negative if x<y, zero if x==y, positive if x>y.
>>>

> Is there any built-in function in Python for comparing two strings
> alphabetically?  Something suitable for passing as the optional
> comparison function for the list "sort" method.  I could code up
> something like
> 
>   def scmp(a,b):
>     if a<b: return -1
>     if b<a: return 1
>     return 0
> 
> but it's pretty ugly to need to do that.  I've looked in the docs but
> don't see any obvious counterpart to perl's lozenge (<=>) operator.
> Is there one?
> 
> Thanks.
> --
> http://mail.python.org/mailman/listinfo/python-list




From sjoshi at ingr.com  Thu Feb  7 16:44:02 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 7 Feb 2002 13:44:02 -0800
Subject: Help! Using Boa Constructor
Message-ID: <8f8ffe67.0202071344.6673ed5d@posting.google.com>

Hello
I'm trying to run boa constructor and I keep getting this error
message when the application loads up
03:45:01 PM: Traceback (most recent call last):
03:45:01 PM:   File "C:\Python21\wxPython\tools\boa\Editor.py", line
1168, in poll_for_open
03:45:01 PM:     except Queue.Empty:
03:45:01 PM:     AttributeError: 'Queue' module has no attribute
'Empty'
03:45:01 PM:     AttributeError: 'Queue' module has no attribute
'Empty'Traceback (most recent call last):
03:45:01 PM:     AttributeError: 'Queue' module has no attribute
'Empty'  File "C:\Python21\wxPython\tools\boa\Editor.py", line 1168,
in poll_for_open

Any ideas what might be wrong here. I'm using Python 2.1, Wxpython
2.3.1

thanks
Sunit


From tany at sympatico.ca  Thu Feb  7 01:10:55 2002
From: tany at sympatico.ca (Nikolay Mladenov)
Date: Thu, 07 Feb 2002 01:10:55 -0500
Subject: Tcl problem
Message-ID: <3C621A6F.87A446D9@sympatico.ca>

Hi all,

This has probably been asked before but I could not find answer for my
question.
I am embedding python (2.1) into a third party application through a
plugin.

I have a simple console for testing and I am getting this:
>>>import Tkinter
>>>tk = Tkinter.Tk()
Traceback (most recent call last):
  File "<Console>", line 1, in ?
  File "E:\PROGRA~1\Python\Lib\lib-tk\Tkinter.py", line 1480, in
__init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} E:/PROGRA~1/Python/lib/tcl8.3 e:/lib/tcl8.3 e:/lib/tcl8.3/library
e:/library e:/../tcl8.3/library ../tcl8.3/library
This probably means that Tcl wasn't installed properly.

I have no problems with this from the standard python console.
Any help is appreciated!

Thanks



From emile at fenx.com  Mon Feb 11 06:49:25 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 11 Feb 2002 03:49:25 -0800
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com> <69c619074b.tim@worthy.demon.co.uk>
Message-ID: <a48ben$1d5ma8$1@ID-11957.news.dfncis.de>

"Tim Howarth":
> 0xFFFFFFFFL
> to become
> -1

Numeric has:

>>> Numeric.fromstring('\377\377\377\377',1,'i')
array([-1],'i')

-- 

Emile van Sebille
emile at fenx.com

---------



From jamesk at ifm-services.com  Mon Feb 25 01:43:07 2002
From: jamesk at ifm-services.com (James Knowles)
Date: Sun, 24 Feb 2002 23:43:07 -0700
Subject: Newbie Q: Import code at run-time?
Message-ID: <3C79DCFB.D216144A@ifm-services.com>

To learn Python, I dove in and wrote a program that takes a series of
dictionaries that define a complete database structure and spits out the SQL
necessary to build that database in either MySQL or Interbase. It was quite
enjoyable, and the program has been *very* useful for managing a single
database structure on two database platforms. No more errors from hacking
two incompatible files. 

ANYHOW, to the question:

In testing, I put the database structure dictionaries in a file called
dbstruct.py, and in the program I put:

from dbstruct import *

Now I want to generalize the program so that I can pass in the name of the
Python module with the dictionary definitions along with the database type,
e.g.

$ python builddb dbA mysql >dbA-mysql.sql
$ python builddb dbB ib >dbB-ib.sql

Is there some preferred way to do get Python to then do 

from dbA import * 

and 

from dbB import *

?

Thanks!



-- 
Even if you're on the right track, you'll get run over if you just sit
there.
- Will Rogers


From dale at riverhall.NOTHANKS.co.uk  Fri Feb 15 14:23:18 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Fri, 15 Feb 2002 19:23:18 +0000
Subject: Python on Windows CE
References: <a4ev4g012jn@enews4.newsguy.com> <3C6D587B.5040203@wi.mit.edu>
Message-ID: <cvnq6u83o72bhqr4a0v7f6i8djdn27k8mf@4ax.com>

There is PythonCE which is Python 2.2.

I have it on my iPAQ and it seems to work.
--
Dale Strickland-Clark
Riverhall Systems Ltd


From hamish_lawson at yahoo.co.uk  Thu Feb 14 04:05:27 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 14 Feb 2002 01:05:27 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <slrna6lbrp.guj.mlh@vier.idi.ntnu.no> <d7d5ebdf.0202131429.16a21cca@posting.google.com>
Message-ID: <915a998f.0202140105.1df2e70@posting.google.com>

Matthias Janes wrote:

> I tried the link to get more info but it just displayed page not found.
> Please could you let me know where to get more info for EASYGUI

There was a typo (a missing y) in the URL:

    http://www.ferg.org/easygui/index.html
                           ^

Hamish Lawson


From starx at pacbell.net  Thu Feb 14 02:13:55 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 14 Feb 2002 07:13:55 GMT
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com> <b16e4ef7.0202130950.67f4004d@posting.google.com>
Message-ID: <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4>

Stephen Ferg || Wed 13 Feb 2002 09:50:00a:

>> Is there a simple and easy way to allow the end-user to select a
>> directory?  
> 
> I'm working on a package that I'd like to polish some more, called
> easygui.  Go to www.ferg.org/easygui.  The function you want is called
> dirbox().
> 
> If you try it, let me know how you get on... :-)
> -- Steve Ferg

This doesn't really have anything to do with this thread, but i'd like to 
make this comment: =PLEASE= don't use the (apparently) standard directory 
selcetion dialog in windows if you can avoid it. It can be truly 
aggrivating when you know exactly how to type your way there, but -have- to 
use the box... and on a reasonably crowded file system, this means 'open 
node, scroll down, open node, scroll down' repeat-until-insane. 

Ok, enough random ranting =)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From gmuller at worldonline.nl  Fri Feb  8 14:37:11 2002
From: gmuller at worldonline.nl (GerritM)
Date: Fri, 8 Feb 2002 20:37:11 +0100
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <a4193i$1b9$1@reader1.tiscali.nl>

"Skip Montanaro" <skip at pobox.com> schreef in bericht
news:mailman.1013181915.301.python-list at python.org...
<..snip...>
> What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT
slides
> to other, more accessible, formats
<..snip..>

If someone can convert them to pdf with distiller or jaws, then the problem
is solved too.

Regards Gerrit
--
www.extra.research.philips.com/natlab/sysarch





From grante at visi.com  Fri Feb  1 12:02:51 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 01 Feb 2002 17:02:51 GMT
Subject: What is a "bus error"?
References: <mailman.1012579203.9499.python-list@python.org>
Message-ID: <%Sz68.13779$Wf1.4178922@ruti.visi.com>

In article <mailman.1012579203.9499.python-list at python.org>, Steven Majewski wrote:

>> | 	However, some of these programs now give a "bus error" message -
>> | what does this mean?
>>
>> it means python got a SIGBUS signal from the kernel, ie it crashed.
>> SIGBUS commonly results from attempting to access unaligned memory.
> 
> ... or chasing a nil pointer or useing something that's not a pointer
> as a pointer.

On many systems (e.g. Linux/Intel) that's a different signal.
On Linux/Intel SIGSEGV is used to indicate you've tried to
access memory which you're not allowed to access (e.g.
de-referencing a null pointer).

In my experience SIGBUS has always been an alignment error.
I've seen it on SPARCs running both SunOS and Linux.  I don't
think I've ever seena SIGBUS on Intel hardware.

It's not uncommon for SW developed on Intel hardware to have
SIGBUS faults when it's run on SPARC, ARM, or other hardware
with alignment requirements.  I remember finding such a bug in
the Linux ppp daemon when I tried to run it on a SPARC system.

-- 
Grant Edwards                   grante             Yow!  Here I am at the flea
                                  at               market but nobody is buying
                               visi.com            my urine sample bottles...


From mhammond at skippinet.com.au  Mon Feb 25 04:41:36 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 25 Feb 2002 09:41:36 GMT
Subject: problem with exception output
References: <a5d01k$69es1$1@ID-79019.news.dfncis.de>
Message-ID: <3C7A06D2.9080702@skippinet.com.au>

Felix Seeger wrote:
> Hi
> 
> I can't get the hole exception if my script (cgi) fails.
> I use this to get every exception:
> 
> except:
>         print '<table width="100%">'
>         trace = traceback.extract_tb(sys.exc_traceback)
> 
>         print '<tr><td>Exception Type:</td><td>', sys.exc_type, '</td></tr>'
>         print '<tr><td>Exception Value:</td><td>', sys.exc_value, '</td></tr>'
>         for filename, lineno, function, message in trace:
>                 print '<tr><td>Filename:</td><td>', filename, '</td></tr>'
>                 print '<tr><td>Error message:</td><td>', message, '</td></tr>'
>                 print '<tr><td>Line:</td><td>', lineno, '</td></tr>'
>                 print '<tr><td>Function:</td><td>', function, '</td></tr>'
> 
>         print '<tr><td>Complete Stack:</td></tr>'
>         print '<tr><td><hr></td></tr>'
>         print '<tr><td colspan="2">', traceback.print_stack(), '</td></tr>'
>         print '<tr><td><hr></td></tr>'
> 
>         print '<tr><td>Complete Traceback:</td></tr>'
>         print '<tr><td><hr></td></tr>'
>         print '<tr><td colspan="2">', traceback.print_exc(), '</td></tr>'
>         print '<tr><td><hr></td></tr>'
>         print '</table>'
> 
> Everything works fine but the print_stack() and print_exc() prints None.

These functions print the information to sys.stderr/out, rather than 
returning the value as a string.  You probably want something like:

f = StringIO.StringIO()
f.write("<tr><etc>")
traceback.print_exc(file=f)
f.write("</etc></tr>")

etc.

Mark.



From jkraska at san.rr.com  Tue Feb  5 00:36:47 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 05 Feb 2002 05:36:47 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
Message-ID: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>

NOTE! This is _not_ a python enhancement proposal. Please don't
respond as if it were. The Python steering committee rightly
worries themselves with the idea that macros can seriously
worsen the maintainability and comprehensibility of code when
misused; further Guido has historically been adamantly opposed
to macros in Python. I am not addressing any of those political
issues at this moment in time.

Onward...

I am currently experimenting with an implementation of hygienic macros
in a custom Python parser as part of a larger experiment. One of the
things I wish to accomplish in this experiment is a practical idea
of what it means to have a macro in a whitespace-sensitive language,
to what useful ends it can be put, lexing/parsing issues, how to
express declaration and invocation thereof, and so on.

I've come up with this concept that I'm calling "lingos": the name
was chosen to evoke the idea of defining one's own lingo as a subset
of the language. I'm still very much in the concept definition stage.

I'm certain of little, but I'm certain that I want lingos to have
at least the following properties.

1. Lingos must be understood natively by the parser. No preprocessor
front-ends or any other hack.

2. Lingos must be "hygienic". For those unfamiliar with the term,
this means that names and symbols in the macro definition must not
permute or modify the symbols in the namespace of the caller except
where such a possibility is explicitly expressed: i.e., only the
arguments to the macro itself can be permuted.

3. Lingos should be able to modify arguments as if the code in the
lingo were used in place of the lingo itself. I suppose that this
is a fancy way of saying that one additional level of referentiality
is supported. I'm a bit fuzzy about this one; what I do know is that
I don't necessarily mean that it will behave strictly like a macro
and be _expanded_ at the caller. That may or may not be a solution;
perhaps note.

4. Lingos should be able do deal with _code itself_ as a parameter.
This affords the opportunity to have lingos masquerade as proper
Python block forms.

5. Lingos might be able to support serial lingo forms, where one
lingo knows how to aggregate multiple lingos into one lingo
invocation. What I mean by this last is easier to describe by
example.

I've actually been able to get some of what I want to parse in my
environment, although it's not clear to me that I'll be able to
avoid special symbols, or even whether or not I _should_ avoid
special symbols (one might argue that the use of lingo implies a
special behavior and using a symbol to mark this might be appropriate;
if so I suppose "$" is the logical choice).

Discarding the issue of special symbols for the moment, here are
some possible lingos by example:

1. Here "invariant" is a lingo macro which has 0 expression arguments
   and 1 code argument:

	invariant:
		dosomething()

2. Here "unless" is a lingo macro taking 1 expression argument and 1
   code argument:

	unless i < 3:
		dosomething()

3. Here "factor" is a lingo macro taking 1 expression argument and
   no code argument:

	result = factor 17

4. Here "do" is a lingo macro taking 0 expression arguments, a code
   argument, and acting as an aggregator for exactly 1 "while" lingo
   form:

	do:
		dosomething()
	while i <10

So. That, at last, brings me to the point of this message. Ideas?
Feedback? This has been an interesting experiment so far, I'm just
wondering how many folks find this sort of thing interesting and if
anyone has any input.


Joe Kraska
San Diego
CA



From loewis at informatik.hu-berlin.de  Thu Feb 21 10:43:31 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 21 Feb 2002 16:43:31 +0100
Subject: socket in python 2.2 vs 2.1
References: <mailman.1014206732.25525.python-list@python.org> <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de> <mailman.1014290914.10912.python-list@python.org>
Message-ID: <j48z9mq1r0.fsf@informatik.hu-berlin.de>

Ivan Bykov <ivan at kupol.ru> writes:

> Python 2.1: '255.255.255.255' - valid address. And my application
> working is correct.

255.255.255.255 is not a valid IP number (even though Python 2.1
accepts it - it actually looks like a bug in the system's IP stack
which fails to reject this address). Your application works just
because the system fails to check your address correctly. Python 2.2
uses different code to check, thus detects the error.

Just remove the connect call.

Regards,
Martin


From phr-n2002a at nightsong.com  Wed Feb 20 19:24:25 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 16:24:25 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org>
Message-ID: <7xit8rhebq.fsf@ruckus.brouhaha.com>

Gustavo Cordova <gcordova at hebmex.com> writes:
> > Next question: why is muddying the statement/expression distinction a
> > detriment?  As opposed to the alternate view, that the 
> > statement/expression
> > distinction is itself a detriment?
> > -- 
> 
> Or, if I may rephrase the question:
> 
> Why is the statement/expression distinction a detriment,
> as opposed to the alternate view, that the statement/expression
> distinction is a virtue?

Because we've already seen, assignments in expressions are useful.
And blurring the statement/expression distinction has proven quite
successful in languages like C and Lisp.

Anyway, language attempts to impose "morality" on programmers tend to
annoy programmers more than they makes code any cleaner.

> I, at least, prefer the legibility and explicitness
> of the current situation, where there can be no
> misunderstanding, versus the common "Oops, I typed '='
> instead of '==' mistakes".

But we just got through agreeing that allowing += inside expressions
doesn't lead to =/== mistakes.

Where this is leading is the proposal from a month or so ago, that :=
(rather than =) be allowed for ordinary assignments in expressions.

=/== confusion is really a weak reason to not allow assignments in
expressions.  Do you have some other reasons?  If not, prohibiting
them is just imposing a subjective opinion of which method makes
code more readable.  I certainly find

   while (m := re.match(pat,x)):  blah()

more readable than some multi-statement contortion for doing the same
thing, or worse, using some kludgy special-purpose class that's
defined far away in the code and maybe done differently from one
program to the next.  It's possible that some other people find the
contorted versions more readable, but they're not the ones who have to
maintain my code.

Maybe there's a principle to be found here: "design by FAQ".  If newbies
keep arriving and asking over and over again "how do I do X?", maybe
there's a good reason for having a way to do X.  We've seen that already
for augmented assignments.


From lac at strakt.com  Thu Feb  7 11:57:45 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 07 Feb 2002 17:57:45 +0100
Subject: FixedPoint 
In-Reply-To: Message from "Mark McEahern" <marklists@mceahern.com> 
   of "Thu, 07 Feb 2002 08:22:48 PST." <NCBBLFCOHHDIKCAFGCFBMEGKKBAA.marklists@mceahern.com> 
References: <NCBBLFCOHHDIKCAFGCFBMEGKKBAA.marklists@mceahern.com> 
Message-ID: <200202071657.g17GvjGo018700@ratthing-b246.strakt.com>

Mark McEahern wrote:
> Of course, I could ask the PyPgSql folks not to test for None like that.
> But then they're going to say, "how's the bug actually happening?"  And I'm
> going to have to tell them I'm sending a FixedPoint instance as a param to
> cursor.execute().  And *that* only works because I modified FixedPoint.py's
> __repr__ to return str(self).  As you can tell, at the moment, I'm just
> flying by the seat of my pants trying to make something work.

I would tell the PyPgSql folks to please change the test so that their
code will work when passed an argument which redefines the __cmp__
function.  Its just something they need to learn -- which Andrew Dalke
taught me hmm, 2 weeks ago now.

Laura Creighton



From jason at jvoegele.com  Fri Feb  8 11:17:12 2002
From: jason at jvoegele.com (Jason Voegele)
Date: 8 Feb 2002 08:17:12 -0800
Subject: Defending the Python lanuage...
References: <3C62EE4A.5090305@destiny.com> <mailman.1013120294.28856.python-list@python.org>
Message-ID: <91acf731.0202080817.73f1d644@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1013120294.28856.python-list at python.org>...
> On Thu, 07 Feb 2002 16:14:50 -0500
> Michael Chermside wrote:
> > Peter Milliken writes that he didn't know we had such a surplus of 
> > programmers -- this is a common misconception. Suppose that, working on 
> > your own, you get X amount of coding done in a day. If we could spare a 
> > second programmer to stand there and just help you out -- someone to 
> > bounce ideas off of, to watch for typos, to suggest things that you had 
> > forgotten -- then you'd get MORE done... say X * k.
> 
> I can get this by having someone in close proximity who I can interrupt on
> occasion.  He doesn't have to watch the _whole_ time, just when there is
> something interesting or puzzling or when there are design issues to be
> discussed.  I do the same for him and since we're able to both code
> independently _and_ collaborate, I suspect k is even higher than in
> pair-programming.  Not unlike open-source development, but in this case the
> developers have a faster, better means of communication (face-to-face
> versus the net).

(First, take everything I say with a big grain of salt, because I've
never done pair programming, only read about it.)

I doubt that k in the above scenario would be as high as it is with
pair programming.  The idea is that one person is "driving", typing in
the code, and the other is observing.  The person driving is usually
concerned with the low-level details of whatever it is she's working
on, such as the syntax for the for loop in the language, while the
observer is freed to think about higher-level issues such as design,
and how the code can most clearly express its intent.  There is a lot
of communication and feedback between the two individuals in the pair
that leads to code that is simple, relatively bug-free, and expresses
its intention.
 
> > If k is 1.05, then you are getting a 5% speedup by having someone around 
> > to help out. If k were 3, then you'd get THREE TIMES as much done. If 
> > THIS were the case, then it'd be worth it to your company to tell 
> > someone else to stop working on their own and just help you out! The 
> > break-even point is somewhere around k=2.
> > The exact value of k depends on LOTS of things, including the 
> > personalities of the programmers involved, the kind of work involved, 
> > etc, but I have heard several different annecdotal estimates that place 
> > k at just-less-than 2. (Maybe k=1.9 +/- 0.2 for typical situations.)
> 
> If if.  No doubt things such as personality are going to be a huge factor. 
> I guess that sums up my assertion: the type of personality attracted to
> programming is not the type to endure watching someone else code.  My
> fingers would twitch themselves numb.

With pair programming, the roles switch back and forth frequently and
rapidly.  No one person is driving for more than a half-hour, and
typically you'll switch drivers much more often than that (say 10
minutes or so).  Anytime you, as an observer, want to communicate
something about the code that you can't express in a simple sentence
two, you take the keyboard and express it directly.

> To have pair-programming work you
> would need two programmers who: 
> 
> 1.  have similar skill levels

This one is not true.  Part of the point of pair programming is the
feedback and mentoring process that's involved.  A junior programmer
can learn a lot from a senior programmer.  Even the senior programmer
will occasionally learn something from the junior programmer.  And at
least in XP, pairs are not fixed.  You pair with many different people
during a day, some of whom will at a lower skill level, and probably
some of whom will be at a higher skill level.

> 2.  like and respect each other

This certainly helps ;-)

> 3.  have compatible programming philosophies

I'm not sure about that one.  Maybe.  XP alleviates this with some of
its other practices (TestFirst,
DoTheSimplestThingThatCouldPossiblyWork, RefactorMercilessly,
CodingStandard).



> Did I mention you need two programmers?  My current employer has only one -
> yours truly.

That does indeed make it difficult ;-)

> 
> Besides, if I understand correctly, k = 1 = normal productivity for a
> single programmer.  Then your own "anecdotal" estimates place
> pair-programming just below individual programming (for 2 individual
> programmers, k = 2, for 2 pair-programmers, k = 1.9), so what's the point?

Well, assuming k < 2, the point is code that is clearer, more
maintainable, well-designed, and relatively bug-free.  It's like doing
code inspections...constantly, all the time.  Except it works better. 
In addition, two people now readily understand the code instead of
just one.  Combine this with XP's CollectiveCodeOwnership, and
frequent pair swapping, everybody understands a lot more about the
system than they would otherwise.  Perhaps these factors could result
in a k > 2?

> > If k is close to 2, then there it might NOT be a stupid idea to put 
> > programmers in pairs, even without a surplus. Particularly because we 
> > expect some additional benefits in addition to the coding speed, like 
> > fewer bugs and better sharing of the knowledge among the coding team.
> 
> No doubt, assuming k is close to 2.  That would seem difficult to measure
> with any degree of accuracy and so seems like a big assumption.

There's been at least one study (can't recall who or where, sorry)
that showed k to be fairly close to 2.  I recall some discussion of
other studies currently being undertaken.

> > As for Cliff Wells' contention that he doesn't know many programmers who 
> > would want to do this, I have to differ. There are clearly some 
> > programmers who absolutely hate working in pairs, but there are also 
> > many others who LOVE it.
> I wonder: would you be the person doing the typing or the person doing the
> watching?  For the person doing the actual coding, I can see where this
> wouldn't be so bad (it'd probably be great), but I find it difficult to
> believe that watching someone else code would be something any programmer
> could love for very long.

You do both! Pair programming encourages frequent role change as well
as the frequent pair swaps.
 
> Interestingly, your last paragraph brings to mind previous posts on this
> thread where code reviews were discussed and I asserted that I didn't
> consider them useful, not because they wouldn't be effective, but because
> programmers won't do them.  In this case, it's the managers who won't do
> it.  Even if pair-programming is very effective and programmers love it, I
> doubt that managers and employers are ever going to be very receptive to
> the idea, which IMHO makes it somewhat less than useful.

Well, some managers will be receptive, and perhaps those that aren't
receptive right now can learn from their successes (or failures).

Jason Voegele


From ajeru at vknn.org  Mon Feb 18 18:31:47 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 18 Feb 2002 15:31:47 -0800
Subject: Adding static typing to Python
Message-ID: <24c39b2c.0202181531.187fad4c@posting.google.com>

Has anyone heard of plans to add static type checking to Python? I
like Python very much but I don't use it for bigger projects because I
fear the testing effort. For example I just refactored a rather large
Java codebase. And when I change a method name for example, all I have
to do is run everything thru the compiler and it will show me all
places in other classes where this method is called, so I can go there
and chenge the method call as well.

In Python I'd have to test the whole software suite including each and
every branch in each and every if statement to be sure that I didn't
forget something. To build such a test suite would take almost as long
as it takes to build the whole software and I'd have to maintain the
test suite as well. That's just unafordable.

On the other hand, Python is the best language that currently exists
and I would like to use it on bigger projects not just for prototyping
as some suggest.

Regards,
Alexander Jerusalem


From DavidA at ActiveState.com  Fri Feb  1 23:59:47 2002
From: DavidA at ActiveState.com (David Ascher)
Date: Fri, 01 Feb 2002 20:59:47 -0800
Subject: Authentification
References: <3C5A840D.85C673A0@luc.ac.be>
Message-ID: <3C5B7243.7BDE14DC@ActiveState.com>

Useless answer:

   It's 'authentication', no 'fi' involved.

Sorry, it's a pet peeve. =)

--david



From greg at cosc.canterbury.ac.nz  Mon Feb 25 22:01:31 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 26 Feb 2002 16:01:31 +1300
Subject: functional programming with map()
References: <mailman.1014667482.23800.python-list@python.org> <eppstein-FEA312.13144225022002@news.service.uci.edu>
Message-ID: <3C7AFA8B.47A23F10@cosc.canterbury.ac.nz>

David Eppstein wrote:
> 
> No side effects?  Try this:
> 
> L = [x for x in range(10)]
> print x

Well, that's more of an "accidental" side effect.
Ideally, Guido would use his time machine to go back
and make for-loop control variables local to the
loop, so that the same could then be done for LCs.

Rumour has it that he tried to do just that, but
it turned out that some critical code in the time
machine's control software (which is written in
Python, of course) relies on side effects from
for-loop variables. So when he tried to make the
change, the time machine stopped working, preventing
him from having attempted to make the change in
the first place.

An even less substantiate rumour has it that
he is now working on translating all the time
machine software into Perl (or Ruby, depending
on who you talk to) so that this sort of problem
never happens again. Not that he will ever admit
to this, of course.

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


From tom at 8dynamics9.com  Fri Feb  8 12:42:34 2002
From: tom at 8dynamics9.com (Tom F.)
Date: Fri, 08 Feb 2002 17:42:34 GMT
Subject: dependency hell
References: <xIw88.500$9%6.51706@bgtnsc04-news.ops.worldnet.att.net> <pan.2002.02.08.17.01.51.564647.1752@optushome.com.au>
Message-ID: <e6U88.3621$9%6.877174@bgtnsc04-news.ops.worldnet.att.net>

antskip wrote:

> Python 2.2.2 doesn't appear to exist yet!

It does. Check out http://www.python.org/
You'll see the link at the top of the page.
(It is a fairly new release. December 21, 2001)

You know what, I'm an idiot! In looking up the link to tell you about it, I 
just realized I downloaded the wrong RPM. I got the redhat distribution 
instead of source files.

Hopefully this will solve the problem...

Sorry guys.
-- 
Tom F.
Please remove the "9"
to reply to this message.


From jacobkm at cats.ucsc.edu  Mon Feb  4 01:19:25 2002
From: jacobkm at cats.ucsc.edu (Jacob Kaplan-Moss)
Date: Mon, 04 Feb 2002 06:19:25 GMT
Subject: Defending the rodent (was Re: Defending the Python lanuage...)
References: <mailman.1012606383.28987.python-list@python.org>
Message-ID: <jacobkm-0302022219430001@192.168.0.120>

In article <mailman.1012606383.28987.python-list at python.org>,
James_Althoff at i2.com wrote:
 
> Of course, if Douglas Englebart had had his way, this would all be moot.
> Englebart -- who invented the computer mouse -- didn't like moving his hand
> off the keyboard either.  So for his famous NLS system he devised a
> one-handed "chording" keypad on which he typed (all characters) using his
> left hand while manipulating the mouse with his right.  Decades later he
> still maintained that this is *the* way to do it.  :-)

Have you ever tried a chording keyboard?  It's absolutely amazing; I can
type about twice as fast on a chording keyboard as I can on a QWERTY one
(although I probably couldn't now -- I haven't used one for years).

Jacob (who now wants to get his circa 1990 serial chording keyboard to
work with his palm for taking notes in class)


From nanotech at europa.com  Mon Feb 25 12:45:52 2002
From: nanotech at europa.com (czrpb)
Date: Mon, 25 Feb 2002 17:45:52 -0000
Subject: Compiling for HPUX and AIX
Message-ID: <a5dt8g+93hb@eGroups.com>

All:

I am having the following issues compiling:

HPUX 11.0
=========

/opt/ansic/bin/cc -Ae -c -DNDEBUG -O -I. -I./Include -DHAVE_CONFIG_H 
-o
Modules/python.o Modules/python.c
cpp: "./Include/pyport.h", line 437: error 4062: "LONG_BIT definition
appears wrong for platform (bad gcc/glibc config?)."
gmake: *** [Modules/python.o] Error 1

So, I took out the check and continued:

/opt/ansic/bin/cc -Ae -DNDEBUG -O  Parser/acceler.o Parser/grammar1.o
Parser/listnode.o Parser/node.o Parser/parser.o Parser/parsetok.o
Parser/tokenizer.o Parser/bitset.o Parser/metagrammar.o
Python/mysnprintf.o Parser/firstsets.o Parser/grammar.o Parser/pgen.o
Parser/printgrammar.o Parser/pgenmain.o -lnsl -ldld -ldl  -lpthread -o
Parser/pgen
/usr/ccs/bin/ld: Can't find library: "dl"

No good. I do have the said library at: /usr/lib/pa20_64/libdl.1
but adding it to the Make file did not work. (Also, I had to removed
the required lib: ldl.)

AIX 4.1.4
=========

./Modules/makexp_aix Modules/python.exp "" libpython2.2.a;
-Wl,-bE:Modules/python.exp -lld -o python \
                Modules/python.o \
                libpython2.2.a    -lm
/bin/sh: -Wl,-bE:Modules/python.exp:  not found


As you can see, I am not having much luck (except Linux which was
trivial!)

  -- Quentin




From cjw at sympatico.ca  Mon Feb 18 11:47:01 2002
From: cjw at sympatico.ca (Colin J. Williams)
Date: Mon, 18 Feb 2002 11:47:01 -0500
Subject: [Fwd:2nd Workshop on Open Source Software Engineering]
Message-ID: <3C713005.70394C05@sympatico.ca>

Call for papers.
-----Original Message-----
From: ISWORLD Information Systems World Network
[mailto:ISWORLD at LISTSERV.HEANET.IE]On Behalf Of Feller, Joe
Sent: February 6, 2002 6:14 AM
To: ISWORLD at LISTSERV.HEANET.IE
Subject: CFP: 2nd Workshop on Open Source Software Engineering


CALL FOR PAPERS

Meeting Challenges and Surviving Success: The 2nd Workshop on Open
Source
Software Engineering
http://opensource.ucc.ie/icse2002

The 2nd Workshop on Open Source Software Engineering is an official
event
(http://www.icse-conferences.org/2002/info/workshops.html) of the 24th
International Conference on Software Engineering (ICSE 2002), which will
be
held in Orlando, FL, USA from May 19-25, 2002.

The workshop is currently scheduled for Saturday May 25, and is a
one-day
event. The workshop is limited to 35 participants.

In May 2001, "Making Sense of the Bazaar: The 1st Workshop on Open
Source
Software Engineering" was held at the 23rd International Conference on
Software Engineering (ICSE 2001) in Toronto, Ontario, Canada.  Since
then,
interest in Open Source Software (OSS) has continued to gain momentum,
both
in the practitioner and research communities.  The goal of this year's
workshop is to bring together researchers and practitioners to discuss
both
the short and long term sustainability of OSS. Specifically, the
workshop
will seek to address the quality and maintainability of OSS products and
tools, the replicability and portability of the OSS software engineering
process and tool kit, the stability and sustainability of OSS developer
and
user communities, and the viability and profitability of OSS business
models.

The discussion will be organized around four key themes:

1. The quality and maintainability of OSS products.  This includes, but
is
not limited to, the reliability, availability, security, and usability
of
OSS products and tools; the architectural characteristics of OSS
products
and tools; and the current and future roles played by OSS products and
tools
in different application spaces. Of particular interest is the
phenomenon of
"boot-strapping", i.e., the role of OSS software (both existing and
potential) in the development of OSS software.

2. The replicability and portability of the OSS software engineering
process
and tool kit. This includes, but is not limited to, dialogue and
knowledge
transfer between OSS and traditional software engineering approaches and
tool use; comparisons between OSS and other methods of software
engineering;
and the extent to which traditional software developers are able to
effectively adopt OSS tools and techniques.

3. The stability and sustainability of OSS developer and user
communities.
This includes, but is not limited to, the community-based management of
knowledge, reputation and trust; mechanisms and practices for effective
collaboration; and comparisons between company-led (or company-enabled)
OSS
development teams and emergent, volunteer communities.

4. The viability and profitability of OSS business models. This
includes,
but is not limited to, the sustainability of pure-play OSS business
models;
the emergence of hybrid proprietary-OSS business models; and the impact
of
intellectual property law on OSS.

Position Papers:
All participants are strongly encouraged to submit a position paper of
no
more than 3 pages, addressing one or more of the workshop themes
described
above, or related issues. Position Papers should follow the guidelines
at
http://www.acm.org/sigs/pubs/proceed/template.html.

Position papers should be submitted by March 15, 2002 using one of the
following formats: plain text, rich text, html, or MS-Word (yes, we
appreciate the irony).  Send position papers and inquiries to Joseph
Feller
at jfeller at afis.ucc.ie.

Position papers will be reviewed by the Organizing Committee (see below)
and
accepted papers will be published in the official workshop proceedings
and
on the workshop website.  Four position papers will be selected to be
delivered as mini-presentations on the day of the workshop.

Organizing Committee:
Joseph Feller, University College Cork, Ireland (Chair)
Brian Fitzgerald, University of Limerick, Ireland
Frank Hecker, CollabNet, Inc., USA
Scott A. Hissam, Software Engineering Institute, USA
Karim Lakhani, MIT Sloan School of Management, USA
Andr? van der Hoek, University of California - Irvine, USA.

________________________
Joseph Feller
jfeller at afis.ucc.ie
http://opensource.ucc.ie
http://afis.ucc.ie/jfeller

___________________________________________________________________
|The Sender assumes responsibility that this message conforms to the
|ISWorld list policy and conditions of use available at:
|http://www.isworld.org/isworldlist
|----------------------- ISWorld Net Footer ------------------------
| Need help with ISWorld? -> http://www.isworld.org/isworldlist
| If you can't find an answer contact isworld_manager at cornell.edu
|___________________________________________________________________

___________________________________________________________
hosted by Vancouver Community Network  http://www.vcn.bc.ca




From chris.gonnerman at newcenturycomputers.net  Mon Feb 18 23:48:16 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 18 Feb 2002 22:48:16 -0600
Subject: Adding static typing to Python
References: <Pine.LNX.4.33.0202181937180.4907-100000@ns0.tbye.com>
Message-ID: <011401c1b900$a6f817e0$0101010a@local>

----- Original Message -----
From: <brueckd at tbye.com>


> On Mon, 18 Feb 2002, Milliken, Peter wrote:
>
> > Python also lacks features of "data hiding", it attempts to provide
> > some support for it but only in a desultory "procedural" way - try
> > this with 40 programmers and you won't get far! :-)
>
> This is silly - let me guess, your same 40 programmers also can't remember
> to document their code, they follow no naming conventions, they use global
> variables everywhere, etc, right? :) Surely, if you can somehow convince
> them to follow basic coding standards then Python's support for data
> hiding won't cause you any problems. It keeps the honest people honest,
> that's all. OTOH, if you have to babysit your developers then you're
> right, Python is not for your project. Also, you should probably fire most
> of your development staff.

The harder you try to hide the data, the more likely the programmer will
find it "necessary" to work his/her butt off to reach it.  Python's data
hiding, which might be called "by suggestion," is IMHO better because the
programmer *knows* when he/she has crossed the line, and can be prepared
to take responsiblity for it.

(If your programmers don't take responsibility, get new programmers.)

> > and the compiler won't let you do sloppy coding
>
> That's one mighty intelligent compiler! <good-natured snicker>

No kidding.  I've yet to see an automated tool to correct code; I think
I remember reading it is impossible to prove correctness in that sense.

> > Finding such bugs (misnamed variables) requires multiple runs through
your
> > test suite i.e. find a bug, fix it and re-run the tests. You *can't*
find
> > all such bugs in a single pass!
>
> Yes and, all told, that process took a whopping 2 seconds. Meanwhile, the
> compiler for those other languages has finally moved on to the link stage
> and is busy printing out 100 lines of "errors" because of your one
> misnamed variable.

Been there, done that, and the T-shirt is only available in XXXXXL (to allow
space for the error messages).  Give me Python tracebacks any day; my errors
are usually obvious.

> > True, you obviously have a large investment in your test suite. My point
is
> > that up front development time is reduced through the features of strong
> > typing.
>
> Can you give a concrete example showing this?

I'd like to see your metrics also.  Strong typing (as in "contract" oriented
design) has never been effective for me.  I learned Ada years ago (when it
was still new) and I remember fondly/have nightmares about the verbose,
detailed error dumps from the compiler.  Sure, it was nice to know exactly
where in the manual I could find an explanation of my stupidity, but doing
common activities in Ada was just too much work... I saw way too many of
those compiler dumps, and since learning Python I have found that a logical,
simple language design is the best for me.

I can put out working code on the first try many times with Python; with C,
Ada, and other more antique languages, I have always produced working code
on about the fourth or later compile.  I am pretty experienced with C, and
before Python it was my most productive language; those error-removal
iterations are pretty short for me in C.  Given a job done equally well in
either language I am probably about as efficient in either, but that is the
result of many years of C experience stacked against about a year of light-
duty Python experience... and many jobs are much simpler in Python than C.
Few if any buffer-overruns to deal with, for example.

Strong typing (in function prototypes) is for control freaks IMHO.  I used
to be one, but it's just too much work.





From chris.gonnerman at newcenturycomputers.net  Wed Feb  6 08:52:01 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Wed, 6 Feb 2002 07:52:01 -0600
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de>
Message-ID: <006f01c1af15$8a8c2960$0101010a@local>

----- Original Message ----- 
From: "MDK" <mdk at mdk.com>


> I need to convert a list into a list of characters.
> 
> For example:
> 
> ("cat",5,['dog',[3,3,1]],"zoo")

That's not a "list", it's a tuple containing strings, ints, and lists;
further, you have nested lists.  Ouch.

> Would become:
> 
> ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
> 
> Any information would be appreciated.

This is doable, but there is no Python builtin I am aware of that can
do it.  Why do you need to do this?  If we understand what you are trying
to do, we may be able to steer you away from this mess.





From a-steinhoff at web_.de  Mon Feb 11 16:38:38 2002
From: a-steinhoff at web_.de (Armin Steinhoff)
Date: 11 Feb 2002 13:38:38 -0800
Subject: Best python way to do 'object messaging'
References: <Xns91AFB91DF83A2RASXnewsDFE1@130.133.1.4>
Message-ID: <a49dku030ne@drn.newsguy.com>

Hi,

sounds that Python 2.2 under QNX6.1 could be a good platform for your
application.

QNX6.1 provides a very fast message passing inter process communication ... at
http://www.sf.net/projects/pyqnx is a Python 2.2 port and a 
message passing extension module.

Have fun

Armin



In article <Xns91AFB91DF83A2RASXnewsDFE1 at 130.133.1.4>, Philip says...
>
>
>Ok, so i have this pre-prototype (embrionic? :) game system, rogueish, 
>that  basically involves a bunch of different objects running around a 
>map, containing each other, beating each other up, that kind of thing 
>:). What i'm trying to do is set up a general purpouse system for 
>passing messages between the interface, the objects, the AI's, the game 
>world and timing system, and whatever else comes up. Here are my current 
>ideas/contarints:
>
>1. Basic interface: anything that can accept a message has a .message
>(msg) function attaced to it. If the player hits a critter, some part of 
>the player (i have the body and control seperated) sends the body of the 
>critter a damage/type/amount/source type message.
>
>2. It has to be able to be passed easily through various receptor 
>objects that don't really care about what the message is. For example, i 
>do something along the lines of 'map.messageradius( 
>msg,x,y,radius,inv_arsqrd(1.2) ), which only affects a 'percentage of 
>effect' segment of the message while distrubiting it to all of the 
>affected cells, which do nothing but ditrubite it to whatever objects or 
>creatures happen to be standing there. 
>
>3. It has to be storable. My idea for a global timekeeping system is 
>something that stores messages to send to objects at various times. For 
>example, an AI object would always set a ai-tick message to itself at 
>some time in the future at the end of its run, and this is the way that 
>I would get keep 'live' objects updated properly.
>
>I am currently thinking of representing the message as an object with 
>various properties, and having each object that recieves one switch on 
>the type of message, or even better, have a routine in a base class that 
>takes some element of the message object, does a little string mangling, 
>and calls a 'Rec<MESSAGETYPE>(message)' function similar to how TADS 
>uses the VerDoXXX, VerIoXXXWith kind of functions for dealing with verbs 
>(if anyone knows enough about TADS to understand this analogy ... it's 
>one of my favorite languages :).
>
>This is -probably- what i'm going to go with, but i'd like to see if 
>anyone has any ideas/comments/insperations/pointers-to-relevant-
>information, because i've never really made a system like this in any 
>language, and i'd like to get it right before i build most of the rest 
>of the system over it (i alredy have to refactor(i think that's how you 
>use it) all of my display stuff out of the game data stuff, and remake 
>the resource system -again- :)
>
>Thanks for anything :)
>
>-- 
>Philip Sw "Starweaver" [rasx] :: www.rubydragon.com



From aahz at panix.com  Mon Feb 18 11:33:23 2002
From: aahz at panix.com (Aahz Maruch)
Date: 18 Feb 2002 08:33:23 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <roy-05958C.07303117022002@reader2.panix.com> <ubsenehjm.fsf@online.no> <a4psgi$5jo$3@peabody.colorado.edu>
Message-ID: <a4racj$bm2$1@panix2.panix.com>

In article <a4psgi$5jo$3 at peabody.colorado.edu>,
Fernando =?ISO-8859-1?Q?P=E9rez?=  <fperez528 at yahoo.com> wrote:
>
>Would you folks mind sticking to English in public discussions? Use private 
>email for other languages, please.

Why?  While I would be bothered if English ceased to be the main language
in c.l.py (because I'ma typical rude American who knows no other
languages ;-), I think the occasional non-English post simply reflects
the international nature of Python.

I heartily encourage people to partly ignore Mr. =?ISO-8859-1?Q?P=E9rez?=

(The above sarcasm directed at the glass house of someone who uses
something other than ASCII as the $NAME.)
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From emile at fenx.com  Fri Feb  1 19:07:09 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 16:07:09 -0800
Subject: Windows 2000
References: <mailman.1012607823.393.python-list@python.org>
Message-ID: <a3fat5$17o13b$1@ID-11957.news.dfncis.de>

"Brian Mitchell" <bmitch at copper.net> wrote in message
news:mailman.1012607823.393.python-list at python.org...
> I tried an install of Python 2.2 on a Windows2000 box today. IDLE
> seemed to function okay but I couldn't close the command mode
> with control-Z.  Is there anything special about using Python under
> Windows2000 as compared to Windows98?  TIA.
>
If you're getting something like:

Exception in Tkinter callback
Traceback (most recent call last):
  File "F:\Python22\lib\lib-tk\Tkinter.py", line 1292, in __call__
    return apply(self.func, args)
  File "F:\PYTHON21\Tools\idle\PyShell.py", line 514, in eof_callback
    self.close()
  File "F:\PYTHON21\Tools\idle\PyShell.py", line 421, in close
    return PyShellEditorWindow.close(self)
TypeError: unbound method close() must be called with
PyShellEditorWindow instance as first argument (got PyShell instance
instead)

>>>


Take a look at sys.path for a non-python22 directory.  In my case I
found it at 'F:\\PYTHON21\\Tools\\idle' which I suspect is the problem.

I haven't fixed it though so let us know what you find out.

If you aren't getting something like the above then don't read this...
;-)


--

Emile van Sebille
emile at fenx.com

---------



From feedback at cd-house.net  Sat Feb 16 06:10:45 2002
From: feedback at cd-house.net (feedback at cd-house.net)
Date: Sat, 16 Feb 2002 13:10:45 +0200
Subject: Free Software For You .
Message-ID: <1cef01c1b6da$944ca020$28eb42d9@supervisor>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020216/2f0a253f/attachment.html>

From just at xs4all.nl  Thu Feb 14 05:45:52 2002
From: just at xs4all.nl (Just van Rossum)
Date: Thu, 14 Feb 2002 11:45:52 +0100
Subject: How to generate XML
References: <3c21e330@news.opennet.it> <j44rmko5s6.fsf@informatik.hu-berlin.de> <a0212d$57k$1@news1.xs4all.nl> <1013618891.574031@rexx.com> <3C6B80D8.C80DF0ED@unice.fr>
Message-ID: <just-E01E2B.11455214022002@news1.xs4all.nl>

In article <3C6B80D8.C80DF0ED at unice.fr>, Jerome Alet <alet at unice.fr> 
wrote:

> > >> print "<foo/>"
> 
> I suggest you to try my jaxml module available under the GPL at
> http://cortex.unice.fr/~jerome/jaxml/
> 
> it is alsmost as easy to use as print statements but takes care of the
> closing tags for you.

Wow. That's the complete opposite of what I needed when I wrote this:

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/fonttools/fontt
ools/Lib/xmlWriter.py?rev=1.2&content-type=text/plain

I needed a relatively simple interface to create (potentially) large xml 
files, so it writes everything to disk right away. It's otherwise pretty 
braindead...

Just


From kfarmer at thuban.org  Tue Feb 19 19:31:25 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Wed, 20 Feb 2002 00:31:25 GMT
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com>
Message-ID: <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com>

The same behavior was seen with wxPython's PyCrust shell, so I suspect it's
not (just) activepython, but rather the way the interpreter module's
written.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org




From andre.risnes at tenpipes.no  Fri Feb 15 07:28:13 2002
From: andre.risnes at tenpipes.no (André Risnes)
Date: Fri, 15 Feb 2002 13:28:13 +0100
Subject: VB code -> Python
References: <3C6CDDC8.4ED3B329@stop.spam.volvo.com>
Message-ID: <a4iv0k$al9$1@lise.netcom.no>

"Mikael Lex?n" <mikael.lexen at stop.spam.volvo.com> wrote in message
news:3C6CDDC8.4ED3B329 at stop.spam.volvo.com...
[...]
> So my questin is how do I translate the GetObject(...) thing in to
> Python code.

I don't know, but could you do this instead: Make a COM object in VB
with a function that returns the result of GetObject, and call that function
from Python?

--
Regards
Andr? Risnes




From jkraska at san.rr.com  Wed Feb  6 19:38:14 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 00:38:14 GMT
Subject: RFC -- Hygienic Macros in Python
References: <HFEKILOLEFEFMKAECNDLCEMPDAAA.jason@jorendorff.com> <mailman.1013027051.23790.python-list@python.org>
Message-ID: <2hj36ugtjpbivf6ce8ltp3je0fh9rgiakb@4ax.com>

>I only meant this to apply to a theoretical "macros in Python"
>feature.  In other more Lisp-like languages, the number may be
>much better.

Oh, I don't think so. They're _powerful_, but it's quite easy
to giantly obfuscate with macros in Lisp. They're a wonderful
tool, but that's a different issue.

C//



From ryanmorillo at hotmail.com  Mon Feb 11 14:48:02 2002
From: ryanmorillo at hotmail.com (Ryan)
Date: 11 Feb 2002 11:48:02 -0800
Subject: Deep override
Message-ID: <4a7fde99.0202111148.542f8fce@posting.google.com>

I have a module I'm writing for work that has a class that inherits
from websucker.Sucker  I'm trying to override websuckers' inheritance
of webchecker.Checker.checkforhtml so it will just return 1(I want it
to download some files, not html).
I've tried different ways to name in and even tried to def it (Python
didn't like all those dotted names in a def).  Does anyone know how to
get this to work?
Thanks in advance for any insights.


From kragen at pobox.com  Sun Feb  3 01:41:44 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 03 Feb 2002 01:41:44 -0500
Subject: Defending the Python lanuage...
References: <mailman.1012432610.4390.python-list@python.org>
Message-ID: <83vgdfxe1j.fsf@panacea.canonical.org>

"Tim Peters" <tim.one at home.com> writes:
> Open source *is* like that, to the extent that free support is
> relatively easy to get for as long as people like it enough to work on it,
> and even after that support for hire remains a possibility forever.  Would
> that it were so for my copy of, e.g., Macsyma.

One version of Macsyma is now open source:
http://www.ma.utexas.edu/maxima.html

> Software bascially sucks, and you've got to come to terms with that.  The
> reality is that no software you write is going to have a useful lifetime
> spanning even a decade without continual rewriting.

I sure wish some parts of TeX, Emacs, Xlib, and CMUCL had been
rewritten sometime in the last decade.  But they haven't, and I still
manage to use them somehow...



From ryan.phillips at csus.edu  Wed Feb  6 19:19:13 2002
From: ryan.phillips at csus.edu (Ryan Phillips)
Date: 07 Feb 2002 00:19:13 GMT
Subject: Embedded Python Woes
References: <a3se8h$bnm$1@siamese.noc.ucla.edu>
Message-ID: <a3sh61$3br@dispatch.concentric.net>

You are not linking to the python library.  -lpython22 might be the correct
switch.  Winbloze user here :/

-ryan

"Salvador Peralta" <admin at salvador.venice.ca.us> wrote in message
news:a3se8h$bnm$1 at siamese.noc.ucla.edu...
> Hi,
>
> I'm trying to compile the following hello sir robin with gcc without
> much success.  Can someone please tell me where am I going wrong?
>
> #include<Python.h>
> #include<pythonrun.h>
>
> int main() {
>
> Py_Initialize();
> PyRun_SimpleString("x = brave + sir + robin");
>
> }
>
>
> $ gcc pytest.c -I/usr/include/python2.1/
>
> /home/foo/tmp/ccpO93io.o: In function `main':
> /home/foo/tmp/ccpO93io.o(.text+0x7): undefined reference to
> `Py_Initialize'/home/foo/tmp/ccpO93io.o(.text+0x14): undefined
> reference to `PyRun_SimpleString'
> collect2: ld returned 1 exit status
>
> $ pwd
> /usr/include/python2.1
> $ grep "PyRun_SimpleString\|Py_Initialize" *
> pythonrun.h:DL_IMPORT(void) Py_Initialize(void);
> pythonrun.h:DL_IMPORT(int) PyRun_SimpleString(char *);




From p.mayers at ic.ac.uk  Tue Feb 26 13:54:28 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Tue, 26 Feb 2002 18:54:28 +0000
Subject: Coredumps with Python 2.1.1-3
Message-ID: <20020226.185425.1884661237.5746@ic.ac.uk>

I'm using the RPM for Python 2.1.1-3 from the Python website, and have
just had a *really* weird problem. I was editing some code, and did a
test run. It all worked. I turned to chat to a colleague again - turned
back a few minutes later, scrolled one line up in the shell history, BOOM
core dump with the following traceback:

#0  0x806f9e1 in Py_GetProgramFullPath ()
#1  0x80928d2 in PyTuple_GetSlice ()
#2  0x806fa12 in Py_GetProgramFullPath ()
#3  0x806fe0c in Py_GetProgramFullPath ()
#4  0x807009a in Py_GetProgramFullPath ()
#5  0x80700e9 in _PyGC_Insert ()
#6  0x807ea82 in PyMethod_New ()
#7  0x807c640 in PyInstance_New ()
#8  0x807c2ae in PyInstance_New ()
#9  0x8058c70 in PyEval_CallObjectWithKeywords ()
#10 0x805939b in PyEval_CallObjectWithKeywords ()
#11 0x8057830 in PyEval_EvalCode ()
#12 0x8054e22 in PyEval_EvalCode ()
#13 0x806c1c5 in PyRun_FileExFlags ()
#14 0x806b5a5 in PyRun_SimpleFileExFlags ()
#15 0x8051f90 in Py_Main ()
#16 0x40077f31 in __libc_start_main (main=0x8051a50 <main>, argc=7, ubp_av=0xbffffa64, init=0x8050eb8 <_init>, fini=0x809eb60 <_fini>, 
    rtld_fini=0x4000e274 <_dl_fini>, stack_end=0xbffffa5c) at ../sysdeps/generic/libc-start.c:129

No debugging symbols in the binary I have (grr!) - will attempt to get a
version compiled with them in an repro it.

Funnies - I'm using the psycopg database adaptor, *BUT* the crash is
definitely happening in pure python code (isolated using sys.stderr.write
and some sleeps). It seems to happen (mostly) in a line:

de = null()

Where null is a class:

# Stupid psycopg doesn't turn None into null
# when quoting
class null:
        def __init__(self):
                pass

        def __str__(self):
                return 'null'

WTF? Even weirder, if I remove the line of input on which the crash
happens, it will get a lot further, but die somewhere else... Very
strange...

I can't post the exact code of the script, due to the fact it's filled
with internal information, but will try to isolate a test case. The
various versions involved are:

[hdb at hdb DoC]$ python2
Python 2.1.1 (#1, Oct 19 2001, 17:19:16) 
[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-85)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.version
'2.1.1 (#1, Oct 19 2001, 17:19:16) \n[GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-85)]'

psycopg 0.99.5

Unhelpful suggestions would be to "upgrade" any component of the system
at all. This is a heavy production system, and I've nearly killed it once
because of Python dependencies on weird things in the thread libraries...

Suggestions welcome - help!

Cheers,
Phil


From gerson.kurz at t-online.de  Sun Feb  3 09:55:16 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 03 Feb 2002 14:55:16 GMT
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org> <j47kpuhbuu.fsf@informatik.hu-berlin.de>
Message-ID: <3c5d4c21.38617375@news.t-online.de>

On 03 Feb 2002 15:35:37 +0100, Martin von Loewis
<loewis at informatik.hu-berlin.de> wrote:

>I think they got introduced because some people found them convenient,
>especially those coming from a functional background. To them, it is a
>religious matter: If you have map and filter, you also ought to have
>lamba.
>
>Later, they found that, even though they got what they asked for, they
>did not get what they wanted: inside a lambda expression, you can only
>use expressions, but they also wanted anonymous code blocks.
>
>It is unlikely that they will ever get that, since nested functions
>can now refer to local variables conveniently.

Well, you can do that right now (given 2.2)! For example, take this
very useless sample code
-------------------------
a, b = 0, 10
while a < b:
	print a
	a += 1
-------------------------
Of course, "print range(10)" would make much more sense right here;
but for the sake of this example lets look at this small loop. OK, you
first need some auxiliary definitions:
-------------------------
import sys

# print is a statement, not a function, so we define our own
print-like function
OUT = lambda *args:sys.stdout.write(" ".join(map(str,args))+"\n")

# This will make an integer expression return either 0 or 1
BOOL = lambda x: x and 1

# Will always return 0
FALSE = lambda x: 0

# WHILE x() evaluates to true, execute y()
WHILE = lambda x,y: BOOL(x()) and (FALSE(y()) or WHILE(x,y))
-------------------------
Now, we can write the above code simply as
-------------------------
w = lambda x=[10,0]:
WHILE(lambda:(x[1]<x[0]),lambda:(OUT(x[1]),x.__setitem__(1,x[1]+1)))

-------------------------
There are some tricks involved; here are a few hints:

- a loop can be written as a recursive function (see WHILE above). You
can use this to get the REPEAT_UNTIL function I'm still missing in
python. for() is best solved by map(), and if() is just a boolean
expression. 
- You need to know and love your boolean functions and, or, and not.
- if you have a block of statements you want to execute, say:

	instruction #1
	instruction #2
	...
	instruction #n

you can write that in a tuple, and (given the nature of the current
CPython) have all instructions execute in order. That is how the two
statements "print a" and "a += 1" are mapped to lambda:
(OUT(x[1]),x.__setitem__(1,x[1]+1)).

So, I see few limitations in the current implementation of lambda for
the DEDICATED lambda-user. 



From debl2nonospammywhammy at bellatlantic.net  Sat Feb  2 12:23:36 2002
From: debl2nonospammywhammy at bellatlantic.net (David Lees)
Date: Sat, 02 Feb 2002 17:23:36 GMT
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <3C5C20AB.5CE12EA1@bellatlantic.net>

You might take a look at a very nice page at python.org which has links
to a number of tutorials.  With 6 different beginner tutorials there
should be something that suits you.  The link is:

http://www.python.org/doc/Newbies.html

David Lees

Fred wrote:
> 
> Hello, I am just starting out with Python. I have a very basic book
> that is not helping me much. I have done some simple C programming in
> the past and I think if I could see the below C code done in Python it
> would get me over the beginners hump. I tried to do it in Python but
> got stuck right at the start, Python did not like it when I tried to
> say: for i = 1
> 
> Can anybody help here? Thanks!
> 
> Fred  c at tech.usd253.org
> 
> //Simple C code to create a times table.
> 
> #include <stdio.h>
> int main (void)
> 
> {
> 
> int i;
> int j;
> 
> for (i = 1; i < 13; i++)
>         {
>         for (j = 1; j < 13; j++)
>           printf("%4d", i*j);
>                 printf("\n");
>    }
> return 0;
> }


From dmaster at synopsys.com  Thu Feb  7 11:39:14 2002
From: dmaster at synopsys.com (David Masterson)
Date: 07 Feb 2002 08:39:14 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net>
Message-ID: <u3d0dcklp.fsf@synopsys.com>

>>>>> Timothy Rue writes:

> On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:
>>>>>>> Timothy Rue writes:

>>> There is only the creation of a configuration of very common code
>>> functionality, carried out to it's natural conclusion in support
>>> of versatility.

>> That has already been done.  It's been done over and over for the
>> last 50 years of computer programming.  Without a clear definition
>> of what you mean by "common code functionality" to differentiate
>> your approach versus what's already out there, I see nothing
>> worthwhile in your project.

> I've defined the nine commands, so show me where this configuration
> you claim has been done over and over again.

I didn't say they did it in "nine commands".  I said that everyone has
been working on a definition of "common code" since the very beginning
of computer programming (for proof, read some of the history of many
computer programming languages).  The problem is that everyone has a
different definition of *exactly* what works best.  Oh, there are many
generally accepted principles but everyone "colors" the definition of
those principles within their particular implementation (its only
natural for that to occur).  You might be the first to make the claim
that those principles can be represented in "nine commands".

However, so far, your definition of the "nine commands" isn't clear
enough to determine whether your claim is valid.

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From cliechti at gmx.net  Sun Feb 17 11:25:45 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 17 Feb 2002 17:25:45 +0100
Subject: Redefining "\n"
References: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com> <m3bseouu8b.fsf@mira.informatik.hu-berlin.de>
Message-ID: <Xns91B8B1D9A767Fcliechtigmxnet@62.2.16.82>

martin at v.loewis.de (Martin v. Loewis) wrote in 
news:m3bseouu8b.fsf at mira.informatik.hu-berlin.de:
> Dale Strickland-Clark <dale at riverhall.NOTHANKS.co.uk> writes:
>> Is there some way to redefine the "\n" escape sequence?
> The only way to do achieve that is to change the Python interpreter.

or do it on each string manualy:
>>>'abcd\n'.replace('\n','\r\n')


-- 
Chris <cliechti at gmx.net>



From opengeometry at yahoo.ca  Wed Feb 20 19:08:59 2002
From: opengeometry at yahoo.ca (William Park)
Date: Wed, 20 Feb 2002 19:08:59 -0500
Subject: Hosting
In-Reply-To: <283adf56.0202201551.7026d548@posting.google.com>; from kdahlhaus@yahoo.com on Wed, Feb 20, 2002 at 03:51:24PM -0800
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <20020220190859.A3284@node0.opengeometry.ca>

On Wed, Feb 20, 2002 at 03:51:24PM -0800, Kevin Dahlhausen wrote:
> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

With the money you pay to ISP, why don't you get DSL with static IP, and
run your own server?

-- 
William Park, Open Geometry Consulting, <opengeometry at yahoo.ca>
8 CPU cluster, NAS, (Slackware) Linux, Python, LaTeX, Vim, Mutt, Tin



From kent at goatnospamhill.org  Sat Feb 16 16:48:42 2002
From: kent at goatnospamhill.org (kent)
Date: 16 Feb 2002 21:48:42 GMT
Subject: photo id ?!?!
References: <mailman.1013549360.16845.python-list@python.org>
Message-ID: <1013895846.890880@fezzik.endicor.com>

On Tue, 12 Feb 2002 14:30:26 -0700, Andrew Dalke wrote:
> Dear Guido, Foretec, and the Python community,
> 
> Last week I was almost unable to attend the Python conference.  Not
> because I was sick, or unable to travel, but because I refused to
> present photo id during registration.  [1]
> 
> I was told at this point that the reason for the photo id
> requirement was 1) to prevent fraud, and 2) in this day and age
> we cannot be too careful.
> 
> I understand the fraud part.  I know of two relevant forms of
> fraud - illegal use of a credit card and identity theft.  If I
> had known beforehand I would have sent a check a month early so
> that it would clear before I attended, or I would have brought
> cash with me to pay at the conference itself.  As to the identity
> theft, I had with me the printout from my registration, which
> contained the confirmation code.  That should be a unique and
> relatively well hidden information.
> 
> Therefore I assert that there are other ways to handle the fraud
> issue than to require photo id.
> 
> As to the "in this day and age" claim.  First off, I find it
> outrageous to believe that showing photo id for the conference has
> any bearing on the state of terrorism in the world.  After all,

Correct. This belief that a photo-id can prevent crime or terrorism,
by determining intent is simply ridiculous, but unfortunately is
very widespread.  There is no evidence that ID's have ever prevented
any act of terrorism, yet they vastly increase the false-positive
rate while increasing the false-negative rate also. So much for
intelligence.

> everyone on those planes showed photo id.  Israel has mandantory
> id yet also has many acts of political violence.  The UK had
> mandantory id but that didn't eliminate with violent protests related
> to Ireland, while it did enforce discriminitory practices.  As
> far as I can research, nationwide mandantory identification papers
> does not reduce crime nor does it prevent acts of violence.

Exactly. The intent of multi-purpose id's is not to prevent crime.
In fact they do the opposite as they greatly increase it. What they
do provide is Foretec, etc. with a 'due diligence' clause that
takes the the burden of liability off of their shoulders and places
it on yours. And by requiring a government multi-purpose id, if
any problems arise from their misuse of your personal information,
they can be absolved by hiding behind Mommy Government's skirts.
I have no idea if Foretec traffics in personal information, but
your personal information is now a very valuable commodity to most
large companies that obtain it.

> But in a deeper, more philosophical sense, I will not show photo
> id for a conference because it represents an undue and completely
> inappropriate government influence on my activities.  I have
> two forms of photo id - a driver's license and a passport.  The
> first gives me license to drive and the second one gives me the
> right to enter and leave national boundaries.  Both are revocable
> at will by the government.  Neither has any relevance to the Python
> community.
>
> Why should revocation of those two forms of id give just cause
> to prevent my from attending the Python conference?  For that matter,
> it means that having my wallet stolen has an all too serious and
> drastic effect on my life.  (Had it been stolen in DC I would have
> been unable to show ID; and I don't know how I would have gotten home!
> For fear of that, I left my driver's license at my friends' house
> while I played tourist in DC after the conference.)

Correct again.

> In addition, stretching more into my personal history, I come from
> a family of non-denominational Protestant missionaries, who have
> worked in Baptista-era Cuba, Velasco and Lara-era Ecuador, and
> Duvalier-era Haiti.  These were all autocratic governments.  One

A (likely) majority of US citizens have no personal knowledge of
what it is like to be threatened or assaulted by their government.
They have placed a tremendously large trust in the government which
minorities, etc. have learned is quite foolish to do. You evidently
have a somewhat implicit understanding of some of the dangers but
haven't methodically thought this thing through. You will begin to
do so now.

> of the things I took to heart from family stories was that freedom
> and liberty includes the right to travel and conduct business without
> having to get government permission for everything.

> I am proud that is it possible even now in the US for people to live
> and prosper without needing any sort of photo id and suffer only
> relatively few difficulties.  I could have taken a bus, train,
> hitchhiked, biked, chartered a plane, or even canoed to get to DC,
> without needing ID along the way.

Yes, these are benefits of freedom. Note that they are almost
entirely gone. You very likely need at least an implicit approval
for travel these days. You say you can travel without needing photo
id, but look around and actually try them out lately. Most of them
now will result in someone requiring your photo-id somewhere along
the path.

Historically, many slaves could come and go as they wished, provided
they had some sort of badge providing explict or implicit
authentication/approval from their master. Since they were not
free, they needed to be authenticated by someone. The badge often
provided that third-party authentication of self or role. A driver's
license or photo-id attempts to provide an implicit third-party
approval or authentication of self.  What is so different?

> (Let me caution that I am not claiming the US is a police state or
> other sorts of rash claims.  I am stating one of several reasons for
> why I have such strongly held beliefs on this matter.  I happen to
> also believe this matter can be resolved such that Foretec and I are
> both happy with the result.)

You don't need to claim it. I will. State governments are now
withholding licenses, arresting and threatening to arrest people
who simply request the legal authority for requiring SSN and
fingerprint for licenses. By those actions, they commit at least
three felonies per person. And that is simply the starting point.

One cannot use a third party authenticataion mechanism to authenticate
a 'self'. They can only authenticate a role.  When you attempt to
authenticate a self, you simply invalidate your authentication
system and assault the individuals who have submitted themselves
to that authentication mechanism.

> When I complained about this to Michel Sanner, he kidded me that I
> was acting like a real American - complaining about government
> control and living in the Southwest where life is more free.  I take

It was once more free here because the inhabitants knew first-hand
what the irrational, absolute power of the government was like.
They inherently knew what they often couldn't explain, that:

Freedom is the right of self.authentication()

When you lose the ability to live without having to authenticate
yourself (not the same as authenticating a role), you have lost
your freedom.

> I would like to attend my 6th Python conference next year, but
> cannot do so without compromising my strongly held beliefs unless
> Foretec changes their policies.  What I ask as a result of this
> complaint is one of two possibilities:

Unfortunately, I expect they won't. I expect they will again assault
their supposedly fellow Pythoneers next year.

Congratulations Andrew. This was a very well written article. While
your fellow Python conference attendees likely haven't a clue of
what you are talking about and likely will make great fun of you
over this, you just stood up for your's and their liberty.
Unfortunately, most of them will likely soon learn, too late, that
they have given away the most precious thing they ever had.

Thanks for taking the time and energy to write and post this.



From bit_bucket5 at hotmail.com  Wed Feb 27 15:56:28 2002
From: bit_bucket5 at hotmail.com (Chris)
Date: 27 Feb 2002 12:56:28 -0800
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de> <3C7ADE12.C6D895A0@engcorp.com>
Message-ID: <fa7108b0.0202271256.57e27d82@posting.google.com>

Peter Hansen <peter at engcorp.com> wrote in message news:<3C7ADE12.C6D895A0 at engcorp.com>...
> 
> Something text-based that retrieves some useful information
> from the web (a weather site?) and parses the result and
> displays something useful with a few lines of Python code
> using the standard modules would have enough "wow" appeal
> for anyone used to other languages...
> 
> At least, it "wow"ed me when I did this sort of thing
> from the interactive prompt for the first time... :)
> 
> -Peter

I agree--if you have Internet access, do something that grabs
something from a web page.  I also did the above (based on an example
in Learning Python to grab the temp for a city from a weather site)
and it wowed me.

Or, create a simple CGI script and generate a web page.

Good luck,
Chris


From peter at engcorp.com  Sun Feb  3 17:07:07 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 03 Feb 2002 17:07:07 -0500
Subject: Compiling from Pythonwin?
References: <a3k0cf$dko$1@news.efn.org>
Message-ID: <3C5DB48B.3585A9CE@engcorp.com>

Spencer Doidge wrote:
> 
> Is there a way to compile .py scripts from the Pythonwin GUI screen other
> than using compileFile(...)?

What is it about compileFile() that doesn't work for you?


From bas.vangils at home.nl  Sun Feb  3 11:14:55 2002
From: bas.vangils at home.nl (Bas van Gils)
Date: Sun, 3 Feb 2002 17:14:55 +0100
Subject: lambda expressions
In-Reply-To: <3c5d4c21.38617375@news.t-online.de>
References: <mailman.1012737185.19345.python-list@python.org> <j47kpuhbuu.fsf@informatik.hu-berlin.de> <3c5d4c21.38617375@news.t-online.de>
Message-ID: <20020203161455.GA16171@kub.nl>

On Sun, Feb 03, 2002 at 02:55:16PM +0000, Gerson Kurz wrote:
> [snip]
>
> Well, you can do that right now (given 2.2)! For example, take this
> very useless sample code
> -------------------------
> a, b = 0, 10
> while a < b:
> 	print a
> 	a += 1
> -------------------------
>
> [snip]
>
> Now, we can write the above code simply as
> -------------------------
> w = lambda x=[10,0]:
> WHILE(lambda:(x[1]<x[0]),lambda:(OUT(x[1]),x.__setitem__(1,x[1]+1)))
> 
> -------------------------
>
> [snip]


Thanks for the explanations so far... I think the lambda-version is more
difficult to understand then the `original' version. From the examples
that I've seen, this is (almost) always true. Being only a `mediocre'
programmer, I value readability a lot. Guess I won't be using lamba's a
lot.

Having read the recent thread on timing how fast code exectes, though, I
wonder: are lambda-statments _faster_ ?

    cheers

        Bas

-- 
Bas van Gils <bas.vangils at home.nl>   -    http://members.home.nl/bas.vangils
Build a system that even a fool can use, and only a fool will want to use it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20020203/5d3fc459/attachment.sig>

From spencer at spencerdoidge.com  Sun Feb 10 15:33:46 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sun, 10 Feb 2002 12:33:46 -0800
Subject: importing/reloading in Pythonwin
Message-ID: <a46l7r$57e$1@news.efn.org>

Running Python22's Pythonwin.
I import all from a script Foo, run some routines, and then edit Foo in the
edit window. I check syntax and save, then click the Import/reload button.
Returning to the interactive window, I run Foo again, but I don't see the
expected evidence of the changes I made.
What am I doing wrong?




From jason at jorendorff.com  Fri Feb  1 01:31:38 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 00:31:38 -0600
Subject: confused by isinstance()
In-Reply-To: <3c5a3352$0$4604$626a54ce@news.free.fr>
Message-ID: <HFEKILOLEFEFMKAECNDLMEJNCPAA.jason@jorendorff.com>

Francois:

Try reading and running this little program
and see if it clears anything up for you.

# === foo.py
class X:
    pass

if __name__ == '__main__':
    import foo
    import __main__

    print foo is __main__
    print foo.X is __main__.X

## Jason Orendorff    http://www.jorendorff.com/



From warlock at eskimo.com  Tue Feb 12 18:38:02 2002
From: warlock at eskimo.com (Jim Richardson)
Date: Tue, 12 Feb 2002 15:38:02 -0800
Subject: Defending the Python lanuage...
References: <mailman.1013116393.10397.python-list@python.org>
Message-ID: <q09c4a.ag3.ln@127.0.0.1>

On Thu, 07 Feb 2002 16:14:50 -0500,
 Michael Chermside <mcherm at destiny.com> wrote:
>>> Really? What an unusual idea - I didn't realise we had such a surplus of
>>> programmers to advocate something like this :-)
>> 
>> This is a little more "extreme" than I was referring to as well.  The idea
>> of two programmers sitting at the same computer doesn't really go over well
>> with me... I was referring having two programmers sit side-by-side at
>> separate computers, working on different parts of the same program versus
>> having one programmer work in isolation.  Indeed, I don't know many
>> programmers with the patience to sit and watch someone else type ;) 
> 
> 
> Ermmm.... No.
> 
> When XP or Agile Programming advocates talk about "pair programming" 
> they really DO mean one keyboard, two programmers.
> 
> Peter Milliken writes that he didn't know we had such a surplus of 
> programmers -- this is a common misconception. Suppose that, working on 
> your own, you get X amount of coding done in a day. If we could spare a 
> second programmer to stand there and just help you out -- someone to 
> bounce ideas off of, to watch for typos, to suggest things that you had 
> forgotten -- then you'd get MORE done... say X * k.
> 
> If k is 1.05, then you are getting a 5% speedup by having someone around 
> to help out. If k were 3, then you'd get THREE TIMES as much done. If 
> THIS were the case, then it'd be worth it to your company to tell 
> someone else to stop working on their own and just help you out! The 
> break-even point is somewhere around k=2.
> 
> The exact value of k depends on LOTS of things, including the 
> personalities of the programmers involved, the kind of work involved, 
> etc, but I have heard several different annecdotal estimates that place 
> k at just-less-than 2. (Maybe k=1.9 +/- 0.2 for typical situations.)
> 
> If k is close to 2, then there it might NOT be a stupid idea to put 
> programmers in pairs, even without a surplus. Particularly because we 
> expect some additional benefits in addition to the coding speed, like 
> fewer bugs and better sharing of the knowledge among the coding team.
> 
> As for Cliff Wells' contention that he doesn't know many programmers who 
> would want to do this, I have to differ. There are clearly some 
> programmers who absolutely hate working in pairs, but there are also 
> many others who LOVE it.
> 
> Now I just wish I could convince my company to consider doing pair 
> programming. Because for *ME* personally, k is around 3 or 4.
> 
> -- Michael Chermside
> 
> 
> 
> 
> 
> 



I would suspect that K is also strongly affected by the relative
abilities of the programmers, If the "newbie" (Say for example, moi)
were at the keyboard, with Guru sitting beside me, K=10+ :) but what
would K= if we swapped seats? 
 This sounds to me, like one of those things that can only be measured,
not accurately modeled, due to the variables. But the idea is
intriguing. It certainly is used in a lot of non-programming jobs.
Sounds like a great way to learn too :)

-- 
Jim Richardson
	Anarchist, pagan and proud of it
www.eskimo.com/~warlock
    Step by step, day by day, the penguins march forward. 


From niemeyer at conectiva.com  Mon Feb 18 10:17:26 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Mon, 18 Feb 2002 12:17:26 -0300
Subject: super(obj)?
Message-ID: <20020218121726.B2045@ibook.distro.conectiva>

Hello everyone!

AFAICS, this seems to be the most useful version of super:

super(obj.__class__, obj)

So, I'm just wondering why this isn't supported as well:

super(obj)

(yes, I know mysuper() could be easily implemented)

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From jason at jorendorff.com  Fri Feb  1 23:51:32 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 22:51:32 -0600
Subject: converting between data types
In-Reply-To: <rwJ68.2444$Ri2.7632@rwcrnsc54>
Message-ID: <HFEKILOLEFEFMKAECNDLOEOACPAA.jason@jorendorff.com>

> I would like to know how to convert a string into a number.

Use int() or float().





From piet at cs.uu.nl  Wed Feb 27 05:53:08 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 27 Feb 2002 11:53:08 +0100
Subject: PEP 263 comments
References: <mailman.1014762060.30054.python-list@python.org>
Message-ID: <wzk7sz6vsb.fsf@nono.cs.uu.nl>

>>>>> "Jason Orendorff" <jason at jorendorff.com> (JO) writes:

JO> Martin v. Loewis wrote:
>> To make some progress on PEP 263, I suggest that some of the open issues
>> are resolved as follows:

JO> Counter-proposal:

JO>  - Comment syntax: none.
JO>  - UTF-8 file signature: not supported.
JO>  - Python source code encoding: must always be UTF-8.

There are still non utf-8 files around, and not everyone has a utf-8
editor.

JO>  - Implementation: within the parser, everything's just
JO>    ordinary UTF-8 bytes.
JO>  - IDLE: always save UTF-8 unless otherwise directed.

JO> Advantage: simple, universal, easy, similar to what Java does.

Java does accept iso-latin-1 files as input. In fact on my machine (Mac
OSX) it doesn't even accept utf-8 files with the utf-8 signature. And
strings containing utf-8 are interpreted as just 8-bit characters, meaning
every byte is a character.

JO> No confusion about embedded 0x22 bytes in strings.  Also,
JO> stylistically I prefer not to have a document specify its own
JO> encoding, or for comments to affect the meaning of a source
JO> file.

Which 0x22 bytes?

To a certain extend it is possible to autodetect if a file with 8-bit
characters (highest bit set) could be utf-8, but it is error-prone.

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From paul at prescod.net  Mon Feb 11 17:20:18 2002
From: paul at prescod.net (Paul Prescod)
Date: Mon, 11 Feb 2002 14:20:18 -0800
Subject: Coming of age...
Message-ID: <3C6843A2.CBDE3E60@prescod.net>

Someone asks in Slashdot (to paraphrase): "What's the best language." Of
course it's a pretty stupid question. In fact the way he asks it is even
worse, but anyhow, the interesting part is the number of references to
Python in the followups. There are lots. And they are moderated up,
despite being (in at least one case) even more vacuous than the
question. That means that the moderators are in favour of Python also.

This is one more datapoint over a few I've collected in the last few
weeks that somehow Python has crossed some kind of line from being a
"like to try it someday" language to an "everybody cool is using it
these days" language. Despite waiting for this moment for several years
I am somewhat, er, nervous about what this means for the future...

Be careful what you wish for.

http://slashdot.org/article.pl?sid=02/02/11/1857203&mode=nested&tid=156

 Paul Prescod



From emile at fenx.com  Wed Feb 27 18:39:04 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 15:39:04 -0800
Subject: Thank you Python!!!!!
References: <BC74F8F7E87DD511990C000103E2FEFA2F36@XCHG> <mailman.1014850554.26360.python-list@python.org>
Message-ID: <a5jr0o$7q4nl$1@ID-11957.news.dfncis.de>

"Cliff Wells"
> Yes, but your solution, while it may work and do exactly what you
want, is
> obviously inferior since it wasn't written in a "real" progamming
language.
>
> <wink>
>

I'm surprised his bid was even considered!  The problem was clearly
_not_ understood by witness of the low bid amount, and therefore the
proposed solution couldn't possibly be right.

<wink><wink>

--

Emile van Sebille
emile at fenx.com

---------



From mhammond at skippinet.com.au  Mon Feb  4 07:33:10 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 12:33:10 GMT
Subject: Implementing PythonWin console from a Visual Basic application!
References: <mailman.1012742705.13047.python-list@python.org>
Message-ID: <3C5E7FD7.6050106@skippinet.com.au>

codetube wrote:
> Hello Everyone,
> 
> I wonder if any one knows some thing about implementing the PythonWin 
> console and other capabilities from with in a Visual Basic 
> application.
> I am writing an application in which I propose to have a console 
> similar to PythonWin and also be able to run python files etc.
> This might be a "big" question, so I would appreciate any clues or 
> leads in that direction.
> I had looked into some examples given by Mark Hammond in his 
> book "Programming Python" but then I am looking for a more tight 
> integration into the VB application.

You can't really implement the GUI in Python.  You will probably need to 
implement the edit control in VB, hooking the ENTER etc keystrokes.

You then create a sub-class of code.py's main class and make it a COM 
object with the necessary public methods for your VB GUI code to use.

Mark.



From Christopher.Saunter at durham.ac.uk  Wed Feb 13 09:26:53 2002
From: Christopher.Saunter at durham.ac.uk (Christopher Saunter)
Date: Wed, 13 Feb 2002 14:26:53 +0000 (UTC)
Subject: Python Gotcha with Octal Numbers
Message-ID: <a4dt3d$6us$1@sirius.dur.ac.uk>

Greetings All!

I have been using Python now for arround a year, and have found it
incredably easy to learn the language, primarily because things are very
obvious cf other languages, and until the other day, I had never had a
'Gotcha' - a bug caused by me entering some code wrongly and not realising
it.

It all stems to this, I was assigning to numbers to variables:

var1 = 1000
var2 = 0100
etc.

When I ran the code, things did not go as they sould.  Some investigation
(running the code line by line in the interpreter...) revealed the problem
to be with asigning a value of 0100 to a variable.  (I just put the 0 in
there to ballance the screen.  Daft I know...)

>From the interpreter:
>>> 0100
64

I missed the bit in the docs where prefacing a number with 0 means you are
entering in octal...

Personally, I would like to see octal number entry migrated from 0--- to
something like 0o---.  This would bring things more in line with the entry
of hex numbers (0x---)

I know it's only a minor thing, but it'd be nice if my numbers behaved as
I expected...

Opinions?

Regards,
	Chris Saunter


From justin at iago.org  Thu Feb 14 15:25:57 2002
From: justin at iago.org (Justin Sheehy)
Date: Thu, 14 Feb 2002 15:25:57 -0500
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net>
Message-ID: <x7eljnstd6.fsf@ra.iago.org>

stojek at part-gmbh.de (Marcus Stojek) writes:

> How can I sort a list of tuples (all tuples have 4 float items)
> by the second (or third or n-th) tuple item. And I have to
> do it fast.

Python's builtin sort is fairly fast, all you need to handle is the indexing.

The straightforward way would be:

lst.sort(lambda a,b: cmp(a[2], b[2]))

or if you don't like lambda:

def cmp3(a, b):
    return cmp(a[2], b[2])
lst.sort(cmp3)

If you want more speed, you may want to decorate the list first:

newlst = [(x[2], x) for x in lst]
newlst.sort()
lst = [x[1] for x in newlst]

-Justin

 




From rdacker at pacbell.net  Wed Feb  6 14:25:50 2002
From: rdacker at pacbell.net (bob ackerman)
Date: Wed, 06 Feb 2002 11:25:50 -0800
Subject: [osx] milter problem
In-Reply-To:  <Pine.OSX.4.43.0202042131160.317-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <5431FA3A-1B37-11D6-BF58-003065428126@pacbell.net>

i guess you mean rebuilding python. i added pthread to my lib list for the 
milter module and nothing changed. i am going to try installing python 2.2

On Monday, February 4, 2002, at 06:32  PM, Steven Majewski wrote:

>
>
> On 4 Feb 2002, rdack wrote:
>
>> osx 10.1.2, python2.1, milter-0.3.9
>> i have installed the milter ok, but when i do:
>> python sample.py
>> i get:
>> # python sample.py
>> dyld: python Undefined symbols:
>> _pthread_sigmask
>> _sigwait
>>
>> any ideas?
>
> Maybe you need:  "-lpthread"  added to the cc switches. (?)
>
> -- Steve Majewski
>
>




From aleax at aleax.it  Sun Feb 17 12:18:48 2002
From: aleax at aleax.it (Alex Martelli)
Date: Sun, 17 Feb 2002 17:18:48 GMT
Subject: Why Python is like BASIC ...  java "vs" python
References: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph> <mailman.1013826279.13649.python-list@python.org> <i4iu6ug73jr2lmpblj2e70e1tci043st0l@4ax.com>
Message-ID: <YBRb8.80754$3J1.2551057@news2.tin.it>

Tim Roberts wrote:
        ...
> and compiled languages is important; it is perfectly valid to say that
> "BASIC, Python and Perl are usually implemented with interpreters, whereas
> Java, C and C++ are usually implemented with compilers".  In that one way,

It may be "valid" for some definition of the word "valid" (it seems to
respect the rules of English syntax, for example).  It's definitely
peculiar if it's meant to assert truths about the real world, though.

The typical implementations of Java process source code into bytecode.
Later, the bytecode is interpreted by a 'virtual machine' (a bytecode
interpreter).

The typical implementations of Python process source code into bytecode.
Later, the bytecode is interpreted by a 'virtual machine' (a bytecode
interpreter).

It seems to me that this is a very good example of why "the distinction
between interpreted and compiled *languages*", far from being "important",
is utter, total, and absolute bunk.  It shows reasonably clearly why
the issue is strictly with implementations and absolutely not with
languages (Java as handled by gcc 3.whatever is [meant to be] just
the same language, but, due exclusively to an implementation specific
choice, gets compiled down to machine-code; Python uses very different
bytecodes -- one of them, as it happens, exactly that which is most
often associated with Java -- in the two most widespread current
implementations, Jython and Classic Python; ...).

I'm not too sure what's being discussed on this thread (assuming
for the sake of argument that anything at all IS being discussed),
as I'm mostly off c.l.p these days, but this is the sort of assertion
that it hurts my liver to leave unchallenged (and my liver is in a
frail enough state these days that I can't well risk subjecting
it to such further stresses).


Alex



From sab at NOSPAM.freeuk.com  Mon Feb 11 15:23:50 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Mon, 11 Feb 2002 20:23:50 -0000
Subject: Is it possible to call a function from a Tkinter widget.bind and include args???
Message-ID: <a4995n$gbg$1@news5.svr.pol.co.uk>

Is it possible to call a function from a Tkinter widget.bind and include
args???
i.e. i want to do something like this:

def printButtonAndCoords(event, button):
    print "you rolled over %s and coords are %d:%d" % button, event.x,
eventy

Button1.bind("<Enter>", printButtonAndCoords("button1"))
Button2.bind("<Enter>", printButtonAndCoords("button2"))

is this possible???

i have tried this: but it seems a bit crap!:

[snip]
...
  self.newFileButton.bind("<Enter>", self.getMouseCoords)
  self.newFileButton.bind("<Leave>", self.deleteToolTip)
...
 def getMouseCoords(self, event):
  self.xPos=event.x_root-5
  self.yPos=event.y_root+5
  self.showToolTip(str(event.widget), self.xPos, self.yPos)

 def showToolTip(self, id, xPos, yPos):
  if id==".8387596.7951612": #this looks very breakable!
   self.toolTip=tkToolTip.toolTip(xPos, yPos, liftAbove=self.master,
text="Create a new file")

 def deleteToolTip(self, event):
  self.toolTip.destroyToolTip()
  del self.toolTip
[/snip]

help...  :'-(

G. Willoughby






From skip at pobox.com  Fri Feb  1 18:13:13 2002
From: skip at pobox.com (Skip Montanaro)
Date: Fri, 1 Feb 2002 17:13:13 -0600
Subject: comparing DateTime instances
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBKEFPKAAA.mark@mceahern.com>
References: <NCBBLFCOHHDIKCAFGCFBKEFPKAAA.mark@mceahern.com>
Message-ID: <15451.8457.657961.274671@12-248-41-177.client.attbi.com>

    >>> e = DateFrom(expected)
    >>> a = DateFrom(actual)
    >>> cmp(a, e)

Why go to the trouble of calling cmp() explicitly?  "==" seems to work for
me:

    >>> import mx.DateTime
    >>> e = mx.DateTime.DateTimeFrom("2003-01-31 14:13:13.65")
    >>> a = mx.DateTime.DateTimeFrom("2003-01-31 14:13:13.65")
    >>> e == a
    1
    >>> cmp(e,a)
    0
    >>> b = mx.DateTime.DateTimeFrom("2003-01-31 14:13:13.64")
    >>> e == b
    0
    >>> cmp(e,b)
    1
    >>> e < b
    0
    >>> e > b
    1
    >>> c = mx.DateTime.DateTimeFrom("2003-01-31 14:13:13.66")
    >>> e > c
    0
    >>> e < c
    1

See?  No need to stretch your brain out of shape after all... ;-)

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From brueckd at tbye.com  Sun Feb 10 22:45:29 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Sun, 10 Feb 2002 19:45:29 -0800 (PST)
Subject: Pythonistas? <- Re: OT: Perl programmers?
In-Reply-To: <a46uqm$gc7$1@peabody.colorado.edu>
Message-ID: <Pine.LNX.4.33.0202101938190.17089-100000@ns0.tbye.com>

On Sun, 10 Feb 2002, Fernando P?rez wrote:

> David Eppstein wrote:
>
> >> Pythonista - Pie thun ee sta
> >> I took it as a unintended play on Sandanista, which is a group of  "freedom
> >> fighting" rebels in Nicaragua.
> >
> > I think it's a normal Spanish suffix, but it seems to be moving into
> > English -- I've also seen "Gapatista" used by some Calif. activists
> > <http://www.ics.uci.edu/~eppstein/pix/sv/j4p/Gapatista.html>
>
> Indeed, it's just the Spanish form of the 'ist' English suffix (pianist,
> violinist, etc.) It simply means 'who performs an activity'.

Could be from a number of languages; I always assumed quasi-Esperanto:

"-ist" - an individual professionally or avocationally occupied with the
idea or activity defined by the root [1].

Of course, in either Spanish or Esperanto it would more commonly be
"Pythonisto" anyway, so who knows.

-Dave

[1] http://www.webcom.com/~donh/esperanto.html




From tim.one at comcast.net  Mon Feb 11 20:41:13 2002
From: tim.one at comcast.net (Tim Peters)
Date: Mon, 11 Feb 2002 20:41:13 -0500
Subject: Compiling Python 2.2 in Microsoft VC7 (.net)
In-Reply-To: <a85e8a65.0202111643.5f5fda89@posting.google.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEHINLAA.tim.one@comcast.net>

[Joe Shochet]
> I'm trying to get Python 2.2 to compile in VC7 (part of Visual Studio
> .net), but I'm running into all sorts of troubles. Before digging into
> the problems, has anybody else already done this?

David Ascher, at ActiveState, has, but it was "on his own time" so I don't
know whether he'll be able to make time to talk about it.  I don't know of
anyone else.




From jason at jorendorff.com  Sun Feb 10 23:06:22 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 22:06:22 -0600
Subject: A newbie question about class
In-Reply-To: <3c66e1b6_4@corp.newsgroups.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEKCDBAA.jason@jorendorff.com>

Joshua Muskovitz writes:
> And yet, unless you do use things like "!= None", bugs in packages like
> FixedPoint would propogate.  It's a double edged sword.  Being able to use
> "is not None" is a good way to be able to work around a bug like this (if
> you don't have access to the source, for example), but I would hesitate to
> say that it is "better".

The wording "x is not None" is preferred stylistically.
It is done this way throughout the standard library.
(The score is 445 to 2.)

It also runs faster and will not throw surprise exceptions.

So, don't hesitate.  :)

## Jason Orendorff    http://www.jorendorff.com/




From jkraska at san.rr.com  Fri Feb 22 22:30:22 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 23 Feb 2002 03:30:22 GMT
Subject: HELP! Obscure Python Reentrancy Question
Message-ID: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com>

This is complicated, but I'll ellide this to the minimum
necessary for describing the problem.

1. Suppose you begin a C++ program.

2. This C++ program contains at its heart a scheduler which
manages cooperative thread states (specifically, Windows
Fibers, but just think "cooperative threads").

3. The C++ program occasionally dispatches to a bit of Python
code.

4. The Python code calls an external C++ function which
triggers the thread switch back to the scheduler.

5. The scheduler starts or resumes another thread, which
very likely will contain more Python, some of which has
been left in this intermediate state like this.

My test shows this going to hell in a handbasket. That
doesn't surprise me.

What now?

C//


From mtaylor at lowebroadway.com  Mon Feb 25 11:18:39 2002
From: mtaylor at lowebroadway.com (Moray Taylor)
Date: 25 Feb 2002 08:18:39 -0800
Subject: Editors for Python
Message-ID: <d906b2c1.0202250818.57dcfcc7@posting.google.com>

Hi,

I am desperatly seeking a good python editor, so far JEdit is the best
I have found, but since it is written in Java, I find it painfully
slow, and PyBrowse, although good, a little flaky.

Can anyone reccommend a good python editor for UNIX/linux/IRIX. I
don't mind paying for it, but $300 for SlickEdit is a little steep.

Before all you Emacs/vi fanatics start ranting at me, I have tried,
and hate, them both.

I have only two requirements for my editor, syntax coloring, and a
Class/function list.

Somebody please help, before I have to buy a faster computer for
JEdit.

Thanks in advance

Moray


From jimd at vega.starshine.org  Wed Feb 27 07:34:36 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 27 Feb 2002 12:34:36 GMT
Subject: SSL Server Sockets?
Message-ID: <a5ijos$1565$1@news.idiom.com>

 Is the native Python2.2 socket.ssl support still client only,
 or is it possible to use it for simple ssl servers, yet?

 I know everyone recommends m2crypto --- but m2crypto-0.07-snap2
 doesn't build on my Debian box.  I get errors like:

Generating wrappers for Python
_bio.i : Line 30. Syntax error in input.
_bio.i : Line 31. Syntax error in input.
_bio.i : Line 32. Variable %constant multiply defined (2nd definition
ignored).
_bio.i : Line 32. Syntax error in input.
_bio.i : Line 33. Variable %constant multiply defined (2nd definition
ignored).
_bio.i : Line 33. Syntax error in input.

 ... meanwhile I can simply apt-get install python2.2-ssl 




From m.faassen at vet.uu.nl  Sat Feb  2 11:16:23 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Feb 2002 16:16:23 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com>
Message-ID: <a3h3cn$cs6$1@newshost.accu.uu.nl>

David Bolen <db3l at fitlinxx.com> wrote:
> Alex Martelli <aleax at aleax.it> writes:

>> If you have to *think* about it, as "memorization" would suggest, then
>> you'll probably be happier going with the rodent.  The power of vi, for
>> me (and most other vi-enthusiasts I've known) is that the *fingers*
>> learn the patterns -- sort of like in needlepoint; your conscious brain
>> can focus on higher-abstraction stuff.  (...)

> As an Emacs-die-hard, I'll also agree strongly with this. The way I
> like to put it is that I touch type Emacs editing commands nearly as
> quickly as I touch type normal text.  Most of the time, I'm rarely
> even conscious of the mechanics of achieving an edit - rather I know I
> want to "move that line over there" or something, and my fingers just
> perform the right operations.  And whether it's vi or Emacs, I find
> not having to leave the home keys and reach for a mouse (with all the
> associated eye hand coordination involved) is an enormous time saver
> for most editing tasks.

As a vi and emacs user (is this rare?) and a rather fast typer I agree that
it's very much worth it not having to leave the home row so often. Pretty 
much the same argument applies to the commandline; the commandline to me is
really easier on my hands than a lot of mouse usage. Even though I do
use the mouse fairly commonly, mostly it's more for 'browsing' types of
activity (switching and moving windows, clicking links, etc) than for
actually *doing* stuff, like writing code, or copying files. The one
thing that I do use a mouse for (in inconsistent X..) is for copy and
paste, but usually I only revert to that only when there's long chunks of
text involved; retyping is often easier if it's a short piece of text.

Perhaps I can use emacs and vi because I use them in different settings
always. I use vi for writing mail and news, and quick file editing tasks. 
I use emacs when I write code or longer texts. In effect the whole thing
is modal and I only rarely get deluded into thinking I'm in vi while 
I'm in emacs and vice versa as my mind knows what mode it is in.

I suppose it might be a brain thing. I use my left hand for writing and
drawing while I use my right hand for more coarse-grained tasks, like
throwing a ball and such. I wouldn't count myself as ambidextrous though.

Regards,

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


From dsavitsk at e-coli.net  Mon Feb 11 01:49:22 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 06:49:22 GMT
Subject: xor on strings
References: <mailman.1013406617.24389.python-list@python.org>
Message-ID: <SPJ98.3346$pC1.674619@newssrv26.news.prodigy.com>

thanks, i appreciate the help.

-doug


"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1013406617.24389.python-list at python.org...
> dsavitsk wrote:
> > >>> red = hex(56)[2:4]
> > >>> green = hex(156)[2:4]
> > >>> blue = hex(12)[2:4]
> > >>> thestring = '&H' + blue + green + red
> >
> > but here, since the blue string is only 1 character it throws the
> > rest off.
>
> >>> colors = (56, 156, 12)
> >>> print '&H%02x%02x%02x' % colors
> &H389c0c
>
> If you would prefer to have a function instead,
> you can use string.zfill().
>
> ## Jason Orendorff    http://www.jorendorff.com/
>




From jeff at ccvcorp.com  Thu Feb 21 14:05:37 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 11:05:37 -0800
Subject: Adding GUI to application
References: <mailman.1014283420.30343.python-list@python.org>
Message-ID: <3C754501.5C770790@ccvcorp.com>

Simon Brunning wrote:

> Starting the threads up isn't too complicated - see the effbot's "Python
> Standard Library" for an example. (You can look at my copy - I'm only up the
> stairs, after all. <wink>)

Yes, the threading module is pretty simple to use.


> But then you need to get your two threads to communicate. I don't know how
> to do this, but I'd have a look at the Queue module if I were you.

Queue makes inter-thread communication a snap.  In order for two threads to send
messages back and forth, you simply set up two Queues, one in each direction.
Drop 'message objects' (which can be anything that makes sense for your
application) into one end, pull them out the other in a different thread, and
everything pretty much just works--the Queue takes care of almost all of the
synchronization issues.

Jeff Shannon
Technician/Programmer
Credit International




From sjmachin at lexicon.net  Sat Feb 16 06:18:34 2002
From: sjmachin at lexicon.net (John Machin)
Date: 16 Feb 2002 03:18:34 -0800
Subject: line break confusion
References: <mailman.1013827165.15861.python-list@python.org>
Message-ID: <c76ff6fc.0202160318.4411defb@posting.google.com>

Michael Mell <mike at nthwave.net> wrote in message news:<mailman.1013827165.15861.python-list at python.org>...
[snip] 
> In the string module, whitespace = ' \t\n\r\v\f' (5 characters here).

1. space
2. tab
3. line feed
4. carriage return 
5. vertical tab
6. form feed

... looks like 6 to me

> Yet, when I run this:
> >>> import string
> >>> for c in string.whitespace:
> >>>     print '.'+ c + '.'
> 
> I get 7 results.
> What's going on here?
> 

You are counting the interpreter prompting output of "..." because it
looks similar to the print output. 6 + 1 = 7.

Try print '-' + repr(c) + '-'


From chad.maine at broadlink.com  Tue Feb 26 16:15:25 2002
From: chad.maine at broadlink.com (Chad Maine)
Date: Tue, 26 Feb 2002 21:15:25 GMT
Subject: Class functions
Message-ID: <NVSe8.151$en5.1429@typhoon.sonic.net>

I'd like to access the data updated in real time via a class' function.
I've got something like this:

the class:

def CLASS:
    def __init__...

    def __del__...

    def FUNC(self):
        while 1:
            self.DATA = StringIO.StringIO(UPDATE)
            if self.DATA == '': break
        self.DATA.close()

in my client script:

import theclass
X = theclass.CLASS
X.FUNC()
while 1:
    try:
        DATA = self.DATA
        if not DATA: break
        print DATA.read()
    except ValueError:
        break





From henrik_joakim.sjoholm at telia.com  Sun Feb  3 06:43:53 2002
From: henrik_joakim.sjoholm at telia.com (Henrik Sjöholm)
Date: Sun, 03 Feb 2002 11:43:53 GMT
Subject: Installing GDChart
Message-ID: <Zn978.15837$n4.2527083@newsc.telia.net>

After a few hours of trying to get GDchart running on a win 98 machine using
python 2.2, I'm looking for some advice. I've put the following files in the
DLLs directory:
* zlib.dll
* libpng.dll
* libgd.dll (to be on the safe side)
* gdchart.pyd
However, I couldn't find a dll for the jpeg option (jpeg-6?)

Now, trying to import gdchart, I get an import error: dll not found.

Can anyone please help me?





From kfarmer at thuban.org  Sat Feb  2 03:21:54 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Sat, 02 Feb 2002 08:21:54 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com> <3C5B90F3.3020503@skippinet.com.au>
Message-ID: <CkN68.6145$F6.3643599033@newssvr14.news.prodigy.com>

My mistake:

lr = win32gui.DefWindowProc(self._GetAppBarWnd().GetSafeHwnd(),
    WM_NCHITTEST,
    0,
    self._LowHighWordsToLparam(win32api.GetCursorPos())
   )

I'm catching the message with:
    self._appbarwnd.HookMessage(self._OnWMNCHitTest, WM_NCHITTEST)

And yet though I return HTBORDER, the edges are still resizeable.  This is
the C-code I'm trying to replicate:

 case WM_NCHITTEST:
    {
        LRESULT lr = DefWindowProc (hwnd, message, wParam, lParam) ;

        if (lr == HTBOTTOM || lr == HTBOTTOMLEFT || lr == HTBOTTOMRIGHT)
            return HTBORDER;
        else
            return lr;
    }

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C5B90F3.3020503 at skippinet.com.au...

> code.  Are you sure _GetAppBatWnd() returns an integer?





From jkraska at san.rr.com  Sun Feb 24 02:14:45 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 24 Feb 2002 07:14:45 GMT
Subject: HELP! Obscure Python Reentrancy Question
References: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com> <mailman.1014528880.19882.python-list@python.org>
Message-ID: <u45h7us5qqnk9bj6ltt7kqgomoojpjhger@4ax.com>

On Sat, 23 Feb 2002 23:27:53 -0600, "Jason Orendorff" <jason at jorendorff.com> wrote:

>Courageous wrote:
>> What now?
>
>Are you using any actual threads?

Well, as it turned out, there was this timing issue of a
thread that transiently came into existence just prior
to converting itself to a fiber. The interplay between
the parent thread (starting in Python) and the child
thread/fiber was causing the problem.

Thanks to Christian Tismer for figuring this one out.

So, to answer you're question: no, I don't have any
actual thread, err, I take that back, I had a badly
behaved one and fixed it. :-)

C//



From jimmy at retzlaff.com  Tue Feb  5 15:25:25 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Tue, 5 Feb 2002 12:25:25 -0800
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F557D@MAYOR.retzlaff.com>

> But about the only hand skill relevant to Python is typing, which I 
> (and probably most of us) do both-handed...

Not me. I just had surgery on my dominant right hand last Thursday so
I'm temporarily non-dexterous. :(

Since I haven't been able to write much Python since, c.l.py has been a
welcome diversion!

Maybe ol' Tim Rue can come up with an autocoder concept that only uses
keys on the left half of the keyboard. I'd volunteer to help implement
it, but I type too slowly with this bloody stump. ;)

Jimmy



From oren-py-l at hishome.net  Sun Feb  3 03:25:31 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Sun, 3 Feb 2002 10:25:31 +0200
Subject: Enhanced Generators - reiterable x-functions
In-Reply-To: <837kpvz2mw.fsf@panacea.canonical.org>; from kragen@pobox.com on Sat, Feb 02, 2002 at 10:05:11PM -0500
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org> <83g04kzoy2.fsf@panacea.canonical.org> <mailman.1012677065.886.python-list@python.org> <837kpvz2mw.fsf@panacea.canonical.org>
Message-ID: <20020203102531.A16902@hishome.net>

On Sat, Feb 02, 2002 at 10:05:11PM -0500, Kragen Sitaker wrote:
> In any case, it isn't clear to me that it's more useful to be able to
> say:
>     mymap = xmap(f, xs)
>     for ii in zs:
> 	for jj in mymap:
> 	    do_something(ii, jj)
> than to have to say:
>     for ii in zs:
> 	for jj in xmap(f, xs):
> 	    do_something(ii, jj)

Bingo. Your example clearly shows how re-iterable x-functions will behave 
as expected in cases where non-reiterable xmap will produce a surprising 
result.  My definition of surprising in this case is "behaves differently 
if you remove the x".

Here's an even better example:

>>>alphabet = xmap(chr, xrange(ord('a'), ord('z')+1))
>>>twoletterwords = [a+b for a in alphabet for b in alphabet]
>>>print list(twoletterwords)

This example will produce strange results with non-reiterable xmap.  If 
you remove the x from xmap it suddenly starts to work.  

The x-functions are already lazy in reading from their sources - why
should they be eager in asking their sources for an iterator?.  This 
half-lazy half-eager chimera is the true source of this inconsistency.  

> So the upside of restartable xmap() etc. is that it saves you a lambda
> and a lambda call when you need it; the downside is that it gives you
> a function whose semantics subtly break when it's applied to
> nonrestartable iterators.

There's nothing subtle about that.  The iter() function, by definition, 
returns an object that changes each time you reference it.  It's about
as surprising as the fact that the expression 'itr.next()' returns a 
different result each time you evaluate it.  

The iter() function is useful in code blocks with the appropriate loop
structures and try/except clauses.  

Inside expressions, iterators should be kept as invisible plumbing.  This
ensures that expressions of arbitrary complexity using x-functions can be 
re-iterated as many times as you like.

	Oren




From phr-n2002a at nightsong.com  Wed Feb 20 12:29:35 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 09:29:35 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org>
Message-ID: <7xr8ng83k0.fsf@ruckus.brouhaha.com>

Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet
> 
> Since the session ID comes _before_ the name of the servlet, all relative
> links work just fine.  WebKit knows how to parse the URL and extract the
> session ID.

Hey, that's clever.  You don't even have to put the session id in the
URL at all.  Just shoot something like

<base href=http://foo.bar/cgi-bin/_SID_=36987348732/>

into the HTML at the top of each page, and all the links will
automatically have the session ID in them.  Use mod_rewrite to map the
SID path component to a query parameter and also to redirect the
request to a URL without the ugly SID in it.


From mcherm at destiny.com  Mon Feb 11 11:12:55 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 11 Feb 2002 11:12:55 -0500
Subject: I wish.............
Message-ID: <3C67ED87.8050901@destiny.com>

 > .....that we quit calling Python as CPython.

I wish that we would use the term "Python" for the LANGUAGE that Guido 
invented (with some help from other folks). We could then use the term 
"CPython" to refer to the most popular implementation (in C), and 
"Jython" for the next-most-popular implementation. For perl there is no 
need for such a distinction -- the language is so poorly defined that 
the implementation effectively defines the language and makes it 
practically impossible to create another implementation.

However, most of the time when people say "Python" meaning "CPython" 
there isn't any confusion, so I don't really mind. But using the terms 
properly helps people keep in mind that things like Jython (and Pippy, 
and a few other implementations out there) exist, and are important 
parts of the Python community.

-- Michael Chermside




From phr-n2002a at nightsong.com  Tue Feb 19 08:01:27 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 05:01:27 -0800
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi> <23891c90.0202190434.147ee6d1@posting.google.com>
Message-ID: <7x7kp9zkuw.fsf@ruckus.brouhaha.com>

paul at boddie.net (Paul Boddie) writes:
> You could always make a SWIG wrapper for iODBC - I started one to see
> how hard it might be - but in the end it's probably cheaper (time is
> money, so they say) to license mxODBC.

I hadn't heard of iODBC before, but just found the home page www.iodbc.org
and it mentions that a Python bridge (whatever that is) already exists.

There are also some Perl DBI ODBC implementations, I think.  Maybe one
of those could be connected to Python somehow.


From SBrunning at trisystems.co.uk  Thu Feb 21 09:14:42 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 14:14:42 -0000
Subject: Suggestion for best intro to Python
Message-ID: <31575A892FF6D1118F5800600846864DCBCCDC@intrepid>

> From:	hack at socket.net [SMTP:hack at socket.net]
> What book/web site/online document/something else would be best as a
> source to learn Python for an experienced programmer?
 
<http://diveintopython.org/> is very good.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From jeff at ccvcorp.com  Mon Feb  4 20:48:55 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 17:48:55 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com> <7xheow3hl2.fsf@ruckus.brouhaha.com>
Message-ID: <3C5F3A07.45126DAF@ccvcorp.com>

Paul Rubin wrote:

> Would anyone actually WANT to use a 1.5.3 release if it existed?
> (Not that I'm volunteering).  I assume 2.1.2 was done because someone
> wanted to use it rather than switching to 2.2.  But 1.5.anything
> is crufty enough that I'd use 2.1.2 or 2.2 in any new installation.

I suspect that this is the crux of the whole issue.  My personal feeling is that
the only reason anybody *cares* at all about 1.5.2 is because of Red Hat, and
thus, indirectly, the large number of ISP's using it because it's what came with
RH.  Once RH starts shipping with a reasonable Python version (which *ought* to be
with 8.0, whenever *that* comes out, but hopefully *soon*), then 1.5.2 will begin
fading as people upgrade their RH installations.  It probably won't completely
disappear by the time that Python 3.0 is released, but it should be a negligible
concern by then.  At that point, the division change won't be an issue any
more--you'll be able to use the __future__ directive and reasonably expect people
to upgrade if they're running <= 2.2.  In the meantime, you can either use
explicit castings with float() and int(), or plan on having a <= 2.1 version and a
>= 2.2 version.  If you're explicit enough in your documentation/installers/etc
what version is required, and how to get the other version, this shouldn't be too
horrible of a burden.

Jeff Shannon
Technician/Programmer
Credit International




From usenet at ixokai.net  Sun Feb 10 22:33:06 2002
From: usenet at ixokai.net (Ixokai)
Date: Sun, 10 Feb 2002 19:33:06 -0800
Subject: PYTHONPATH on the Windows Registry?
Message-ID: <3c673b3e$1_2@nopics.sjc>

    Question-- if Python on Windows (w/o win32all) is running on a machine,
without any PYTHONPATH/etc set in the registry, but with an environment
variable set, will it set its 'current' python path into the registry? Or
set anything at all? I want a zero-impact run of Python on the system. :)

Thanks.

--S






From phr-n2002a at nightsong.com  Tue Feb 19 02:06:43 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 18 Feb 2002 23:06:43 -0800
Subject: Adding static typing to Python
References: <Pine.LNX.4.33.0202181937180.4907-100000@ns0.tbye.com> <mailman.1014094230.2291.python-list@python.org>
Message-ID: <7xofimx858.fsf@ruckus.brouhaha.com>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> writes:
> The harder you try to hide the data, the more likely the programmer will
> find it "necessary" to work his/her butt off to reach it. 

That may happen in C++ but it doesn't happen appreciably in Java.
Java is designed to resist even malicious attempts to reach into
another object's data.

> Python's data hiding, which might be called "by suggestion," is IMHO
> better because the programmer *knows* when he/she has crossed the
> line, and can be prepared to take responsiblity for it.

I'm not that worked up about type declarations as a bug-avoidance
tool; type declarations mostly help optimizing compilers generate
efficient object code, which current Python implementations don't
attempt at all.  However, lack of declarations for local variables and
for object slots are shortcomings in Python (2.2 adds a slots
declaration).  For example, in pre-2.2 Python it's fairly easy to
subclass something and then create some instance variables in the
subclass that accidentally clobber instance variables from the parent
class.  2.2's slots declaration might fix that--I'm not sure, since
it's not documented in the manual and I'm not sure any more where the
description is.


From lurker_member at newsguy.com  Thu Feb  7 12:53:56 2002
From: lurker_member at newsguy.com (lurker)
Date: 7 Feb 2002 09:53:56 -0800
Subject: Autocoding project proposal.
References: <5422.785T1897T13084671threeseas@earthlink.net> <slrna4o45g.4ge.philh@comuno.freeserve.co.uk> <1911.788T2678T8325744threeseas@earthlink.net>
Message-ID: <a3uevk01lav@drn.newsguy.com>

In article <1911.788T2678T8325744threeseas at earthlink.net>, "Timothy says...
>
>On 21-Jan-02 07:59:29 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>On Mon, 21 Jan 2002 02:01:27 GMT, Timothy Rue <threeseas at earthlink.net>
>>wrote:
>>>This posting is to introduce or suggest an open source software project
>>>to produce a core tool that will allow an autocoding environment to be
>>>developed, allowed to evolve in the open source software community and
>>>spirit of. I believe we can overcome many of the problems that proprietary
>>>autocoding systems inherently have (not to mention that existing
>>>autocoding systems appear to be field/domain specific limited and that
>>>it may be possible to beat this too.)
>
>>When you say autocoding, what do you mean?
>
>Programming is the act of automating the use of complexity that is made up
>of simpler things. Done so that such complexity is easier to use. As such
>the complexity of programming itself can be automated.

How long do you pray this? Why did you not realize such a tool by yourself??

>There is something of an inherent resistance to the project by those who
>would feel threatened by such a tool.

That's always the same killer phrase do use since years :)

Seems so you have probably lost the contact to the reality ... looks a little
bit manic, IMHO.





>
>The project I present is a tool that allows easier automation. It's more a
>general automation tool than specific to autocoding, but inherently it
>can be used to create an auocoding environment that even the typical user
>could use.
>
>It amounts to what would be a small shell and/or library containing nine
>commands w/command options. It's really quite logical as it makes use of
>existing integrated documentation/example code of what is in the format of
>typical unix command, library quick reference, etc, manual pages. But can
>handle alot more.
>
>
>---
>*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
>   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
>Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
>Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
>Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<
>



From paul.moore at atosorigin.com  Thu Feb 14 11:56:21 2002
From: paul.moore at atosorigin.com (Paul Moore)
Date: Thu, 14 Feb 2002 17:56:21 +0100
Subject: JAR equivalent in Gordon McMillan's Installer
Message-ID: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com>

One nice feature of Java (sorry!) is the ability to ship an application
as a single JAR file - containing all code, resources, etc, etc. Is
there any equivalent feature available in Python? The obvious place
where it would be is in Gordon McMillan's Installer package, with its
Python Archive format, but I don't think (from looking at the docs) that
does quite what I'm after.

I have no requirement to include a Python interpreter, or C extensions,
in the archive. The code is all pure Python. But, I do want to be able
to include support files (text files, image files for use in wxPython
toolbars, graphics for display on screen, sound files, etc etc) which
can be picked up from the same archive.

I'm looking for something which works a bit like Java's
ClassLoader.getResourceAsStream() (along with the necessary stuff to
allow the stream to then be *used* as an icon/image/whatever).

It seems like the sort of thing which should be possible with Python,
but I can't find it. Did I miss something?

Thanks in advance,
Paul.



From fredrik at pythonware.com  Mon Feb 11 16:54:36 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 21:54:36 GMT
Subject: Add iterator to gzip files?
References: <a493fa$11ei$1@nntp6.u.washington.edu>
Message-ID: <w4X98.17818$n4.3074909@newsc.telia.net>

Russell E. Owen wrote:
> Similarly, I'm wondering if sockets will, or do, support iteration to
> read lines of data. (I've also wondered why they don't directly support
> read and write as unix sockets do

they sure do: it's just that the methods are spelled recv and send.

> one can create a file from a socket, but that seems a bit indirect.

the makefile method creates a *buffered* stream interface to the
raw socket.  if you're going to use readline, that's probably exactly
what you want.

</F>




From dieter at handshake.de  Sat Feb 23 13:37:21 2002
From: dieter at handshake.de (Dieter Maurer)
Date: 23 Feb 2002 19:37:21 +0100
Subject: PIL jpeg problems
References: <mailman.1014236017.29342.python-list@python.org>
Message-ID: <x7vgcohwny.fsf@linux.local>

Jonathan Miller <notjonathan at yahoo.com> writes on Wed, 20 Feb 2002 12:12:41 -0800 (PST):
> I'm having trouble using jpegs with PIL on my ISP's server--FreeBSD
> 4.1.1-STABLE, Python 2.0 (#4, Nov 22 2000), GCC 2.95.2 19991024 (release), at
> pair.com--a good box.
> ...
> IOError: decoder jpeg not available
I once read (either here or in the Zope mailing lists) that
JPEG support must be enabled explicitly.

Please search Google to find out the details.


Dieter


From danielk at aracnet.com  Tue Feb 19 21:43:35 2002
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 19 Feb 2002 18:43:35 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com>
Message-ID: <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com>

The point is that it ceases to be the same language with the same attractions
(or flaws) depending on how you look at it.

Dan

On 19 Feb 2002 16:46:00 -0800, ajeru at vknn.org (Alexander Jerusalem) wrote:

>Daniel Klein <danielk at aracnet.com> wrote in message
news:<ghp47u02vceavgohdod9c8vbs9u5po91bt at 4ax.com>...
>> Let me reverse the question: What do you think the response would be if
Sun
>> was asked to drop static type checking (as well as checked exceptions).
>> 
>> When you can answer this, then you'll have _your_ answer.
>> 
>> Daniel Klein
>
>Well, my hope is that open source projects are more flexible than
>bureaucracies like Sun.
>
>
>Alexander



From pobrien at orbtech.com  Mon Feb 11 14:55:33 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Mon, 11 Feb 2002 19:55:33 GMT
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <mailman.1013255897.29941.python-list@python.org>
Message-ID: <VkV98.2452$yV3.182935@newsread2.prod.itd.earthlink.net>

You might find this useful:

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/87056

This module, dispatcher.py, provides global signal dispatching services
suitable for a wide variety of purposes, similar to Model-View-Controller or
Model-View-Presenter patterns. This particular implementation allows a
looser coupling than most Observer patterns. It also does transparent
cleanup through the use of weak references and weak reference callbacks.
This version defaults to using weak references, but provides an option to
not use weak references for those cases where weak references are
problematic (lambdas and such).

--
Patrick K. O'Brien
Orbtech


"Mike C. Fletcher" <mcfletch at geocities.com> wrote in message
news:mailman.1013255897.29941.python-list at python.org...
> Sat down to work out a model-view framework for a little application and
> had one of those "oh, darn, didn't think that out very carefully"
> moments.  You see, I was planning on doing a fairly straightforward
> callback-passing mechanism:
>
[snip]
>
> Yes, it's slightly different semantics than weakref, but it's actually
> useful in the real world ;) .  Anyway, thought I'd share the code so
> that people can object that it's a tool of the devil :) .  If the powers
> that be want to point out any reasons why I shouldn't do this, would
> love to hear it before my house of cards goes *phoosh*.
>
> Enjoy all,
> Mike
> _______________________________________
>    Mike C. Fletcher
>    http://members.rogers.com/mcfletch/
>
>
>
>




From Rave_Dj at btopenworld.com  Sun Feb  3 08:35:40 2002
From: Rave_Dj at btopenworld.com (Callum Golding)
Date: Sun, 3 Feb 2002 13:35:40 +0000 (UTC)
Subject: HELP?
Message-ID: <a3jebb$cg8$1@knossos.btinternet.com>

I'm sorry to bother you, but i was wondering if you could give me a little
advice. I've recently downloaded Python 2.2. Ive read a begginers guide to
programming and tried creating a small program, I used the input function.
Now as soon as I open the program it goes into dos and asks me the question,
and when i type in an anwser, dos shuts down. Now its really starting to bug
me. If you could give me some advice I would be real grateful. The program I
wrote is below; -


tempetature = input("What is the temperature of the spam?")

if temperature > 50:
    print "The salad is properly cooked"
else:
    print "Cook the salad some more"

    Mat.G

        Cheers




From jason at jorendorff.com  Thu Feb 14 17:27:45 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 16:27:45 -0600
Subject: Newbie: Object assignment return value
In-Reply-To: <e81be8b2.0202141051.157160d6@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEKGDCAA.jason@jorendorff.com>

Dean Goodmanson wrote:
> (I think) I'm trying to understand if object assignment has a 
> return value.

Nope, it doesn't.  Assignment is a statement, not an expression,
in Python.

## Jason Orendorff    http://www.jorendorff.com/



From martin at v.loewis.de  Mon Feb  4 01:45:01 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 04 Feb 2002 07:45:01 +0100
Subject: parsing an xml document with funky ascii characters
References: <be58dbe.0202031714.3ebee2df@posting.google.com>
Message-ID: <m3ofj5kaoi.fsf@mira.informatik.hu-berlin.de>

ayinger1 at pacbell.net (andrew) writes:

> I am using sax parser in python 2.1.
> How do I deal with xml documents with characters like '?'?

Depends on how this character is encoded.

> 
> I have tried:
> 	- setting encoding="ISO-8859-1 in the xml doc itself

If you use that encoding, this is the right way (assuming you put it
into the xml header, and assuming you added the closing double-quote).

> 	- setting the InputSource encoding via:
> source.setEncoding('ISO-8859-1')

That has no effect, I believe (although it probably should)

> 	- escaping the character in the doc: ('\x84')

That will give you the string '\x84' in the output. BTW, in what
encoding is LATIN SMALL LETTER A WITH DIAERESIS encoded as \x84?
Certainly not in iso-8859-1

> 	- and, finally, encoding the parsed strings that have this
> character: myString.encode("ISO-8859-1")

That also has no effect, and should not have. Invoking .encode
performs unicode conversion of the string, followed by encoding it as
Latin-1, again. Unless you get a UnicodeError, this will give you the
original string back.

> What I have found is that the default parser (appears to be expat,
> retrieved from sax.make_parser) seems to store every element as
> unicode strings.  

That is proper behaviour. XML document pieces are always reported as
Unicode strings to the application, as per the XML spec.

> It appears to store them incorrectly (so, '?' appears in the unicode
> string as '\xe4' instead of '\x84').

If you are using ISO-8859-1, then \xe4 *is* LATIN SMALL LETTER A WITH
DIAERESIS (more precisely, it is \u00e4).

> The result is that if I try to encode the unicode string that i get
> back from the parser, the character in question incorrectly appears
> as 'E' (sum).

Depends on how you are encoding it. If you encode it as ISO-8859-1,
you get the byte \xe4, which does *not* mean 'E (sum)', but indeed
means LATIN SMALL LETTER A WITH DIAERESIS.

That gives a clue as to what encoding you are using, though: In IBM
code page 437, we indeed have the assignments

<U00E4>     /x84         LATIN SMALL LETTER A WITH DIAERESIS
<U03A3>     /xe4         GREEK CAPITAL LETTER SIGMA

(where the latter looks somewhat like a 'E (sum)' indeed).

> Any ideas?  Am I doing something wrong here?

I'm still not sure what you are trying to achieve, and why you think
anything is not working properly. Your main problem seems to be that
you don't know what encoding you are using. 

Since you get \xe4 reported from the parser, it appears that the input
encoding is indeed ISO-8859-1. It is somewhat surprising, though, that
you then expect that the output encoding is CP437. Perhaps you are
viewing the output in a Windows command.com/cmd.exe window? Try
redirecting the output into a file, and view the file with notepad.

If you really need to output as CP437, invoke .encode("cp437") on the
Unicode strings that the parser is giving you before printing them.

If you want to learn more about these things, please read in the
Microsoft documentation about the "OEM" and "ANSI" code pages.

Regards,
Martin


From brian at sweetapp.com  Mon Feb 25 13:11:04 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 25 Feb 2002 10:11:04 -0800
Subject: XML RPC Woes
In-Reply-To: <1916920022125173556140@winningteam.com>
Message-ID: <004301c1be27$ca5c1980$445d4540@Dell2>

Mark J Nenadov wrote:
> I am attempting to send an mcrypt'ed string to
> an xmlrpc function. The server is returning the
> same "Internal Error" that I got before.  I
> assume that it is because it doesn't recognize
> the type, though the type() function identifies
> the result of the mcrypt as a string.

I don't know what the output of mcrypt is but I would assume that it can
include characters that aren't valid XML characters. That is likely
causing your problem.

> I assume that I need to do some sort of encoding
> since the mcrypt'ed string is probably not  a normal
> string. From the documentation I have read, it
> seemed that I needed to call the xmlrpclib.binary()
> method, which gives me an "incorrect padding"
> error. Is xmlrpclib.binary() required to get an
> mcrypt'ed string to be accepted by the XMLRPC
> server?

Close. What you actually want to do return/pass a xmlrpclib.Binary
object. You would use it this way:

# Client usage
s = xmlrpclib.Server(...)
s.method(xmlrpclib.Binary(mcrypted_string))

# Server usage
def do_something():
	return xmlrpclib.Binary(mcrypted_string)

s = SimpleXMLRPCServer(...)
s.register_function(do_something)

> I realize that you don't maintain xmlrpclib, but
> I thought you might have some insight that I
> couldn't find in the documentation.

No problem. Please CC these questions to the Python list though so the
question and answer will be archived.

Cheers,
Brian




From d_w_lee at yahoo.com  Tue Feb  5 17:00:15 2002
From: d_w_lee at yahoo.com (dustin lee)
Date: 5 Feb 2002 14:00:15 -0800
Subject: Saving Tkinter images?
Message-ID: <f88b21fc.0202051400.50a694ad@posting.google.com>

I'm pretty new to Tkinter programming so bear with me.  I'm interested
if there is some way to save an image (eg. to a jpeg or png) that has
been drawn to a Tkinter canvas.  I'm guessing there isn't but I
thought I'd at least ask.  I know I can do a screen print if a get
desparate but I'd like to be able to create them programatically.  I
also know about (but haven't used) PIL and gdmodule, I just thought it
would be cool to go directly to an image from Tkinter.

thanks

dustin


From tim.one at comcast.net  Mon Feb  4 21:46:35 2002
From: tim.one at comcast.net (Tim Peters)
Date: Mon, 04 Feb 2002 21:46:35 -0500
Subject: Floating-point multiplication
In-Reply-To: <OJEJKEOFJEGJNCEHNGHKGEFOCBAA.xander@yifan.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEMPNJAA.tim.one@comcast.net>

[Alex Holkner]
> I'm experiencing discrepencies in floating point multiplication
> between the Python interpreter and embedded python.

I don't know what "embedded python" means to you, and the answer is probably
crucial.

> Running Python 2.2 under Win2k, compiling with VC++.  The interpreter
> gives:
>
>   >>> print sys.version
>   '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
>   >>> print 0.785398163397 * 4.0
>   3.14159265359
>   >>> print math.pi
>   3.14159265359
>
> Which is expected.  Under my embeddeded python, however:
>
>   print sys.version
>     '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
>   print 0.785398163397 * 4.0
>     3.14159274101
>   print math.pi
>     3.14159274101
>
> Which is quite wrong,

It's actually correct, provided that fp arithmetic is being done in IEEE
single precision (instead of the IEEE double precision Python normally
uses).  Somehow or other your Pentium FPU has gotten wedged into 24-bit
"precision control" mode.  Core Python doesn't muck with the FPU control
registers, so something outside of core Python you're doing, or calling, is
doing the damage.  Scour the sources for calls to control87, _control87,
controlfp or _controlfp.  If you find exactly one such call, it's wrong
<wink>.

Sometimes using a buggy device driver can do this to you too; for some
unknown reason, printer drivers are notorious for screwing up the FPU state
(they change it internally for a legitimate purpose, but neglect to restore
the FPU control state upon exit).




From hfoffani at yahoo.com  Sat Feb  2 18:07:57 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Sun, 3 Feb 2002 00:07:57 +0100
Subject: Defending the Python language...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <20020202134649.468$Gg@news.newsreader.com>
Message-ID: <20020202180808.403$mg@news.newsreader.com>

[me]
> I fully agree that when a manager evaluates techs alternatives
> the decision may be affected by non-tech factors. I don't think
> many readers here would deny such claim and c.l.py is a great
> forum to look for answers on TECH's issues of Python.

[orig. poster]
> Wait just a moment here ... what are you saying here? c.l.py is
> a great place to tooks for tech's issues But we don't consider
> managment issues ...
> Well in the ***real*** world where i live all descission are
> taken by managers !

[And I replied again with almost nonsenses]
...

Being too late to make an effective usenet cancel, I'll try to
make my statements more clear.

In my first post, I was addressing your worries on the fact that
"very often wrong questions are asked here."

I claim that your legitimate concerns aren't specific to Python,
and the way this thread turned out is a little proof of it :-)

Therefore you'll find here much more technical info to help you
in your decision processes that on pure software engineering or
business strategies regardless of the high quality of the
postings.

Of course, my personal desires about c.l.py's content were out
of place and I'd never expected to drive it, much less try to
censor you in any way.

Regards,
-Hernan

Hope it's clear now. If so, I wholeheartly recommend to read
usenet with pizza and beer aside!

[posted and mailed to the O.Poster]





From gerson.kurz at t-online.de  Thu Feb 21 04:51:39 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 09:51:39 GMT
Subject: Stupid Lambda Tricks [LONG]
Message-ID: <3c74c211.20012718@news.isar.de>

Note: the official URL for the following is
http://www.p-nand-q.com/lambda.htm

-----------------------------------------------------
Stupid Lambda Tricks

This document describes some stupid tricks for using lambda
expressions in Python. It is not intended for the "functional
programming style" crowd, but for the "wacky geek fun" hackers.

To understand the need for such a page, the most common limitations on
lambda expressions are:

- You can only use expressions, but not statements. 
- You can only use one expression. 
- You cannot declare and use local variables. 

* Adding Statements to Lambda

The official python documentation says, that you can only use
expressions in lambda, but not statements. For example, the tutorial
says in section 4.7.4: "They (lambda expressions) are syntactically
restricted to a single expression". You can also see this from the
syntax definition in section 5.10 of the "Python Reference Manual":

expression:     or_test | lambda_form
or_test:        and_test | or_test "or" and_test
and_test:       not_test | and_test "and" not_test
not_test:       comparison | "not" not_test
lambda_form:    "lambda" [parameter_list]: expression

* Using print in lambda

The best solution is to define your own print function:

>>> out=lambda *x:sys.stdout.write(" ".join(map(str,x)))
>>> out("test",42)
test 42

Note that this function doesn't do string formatting - you can use the
normal %-style syntax for that.

* Using data assignment in lambda

In Python, assignments are statements (see "6.3 Assignment statements"
in the "Python Reference Manual"), so you cannot easily use them in
lambdas.

* Data assignment for lists

To set an item in a list, you can use the member function __setitem__.
For example, rather than writing

data[4] = 42

you can write

data.__setitem__(4,42)

Example: Here is a function that swaps two elements in a given list:

def swap(a,x,y):
    a[x] = (a[x], a[y])
    a[y] = a[x][0]
    a[x] = a[x][1]

You can write this as a lambda expression like this:

swap = lambda a,x,y:(lambda
f=a.__setitem__:(f(x,(a[x],a[y])),f(y,a[x][0]),f(x,a[x][1])))()

* Data assignment for locals

Sometimes you have local variables, and want to assign values to them.
The best solution is to store all local variables in a list, and use
list assignment as described above.

* Using IF statements

Here is a quote from section "4.16." from "The Whole Python FAQ:"

4.16. Is there an equivalent of C's "?:" ternary operator?
Not directly. In many cases you can mimic a?b:c with "a and b or c",
but there's a flaw: if b is zero (or empty, or None -- anything that
tests false) then c will be selected instead. In many cases you can
prove by looking at the code that this can't happen (e.g. because b is
a constant or has a type that can never be false), but in general this
can be a problem. Tim Peters (who wishes it was Steve Majewski)
suggested the following solution: (a and [b] or [c])[0]. Because [b]
is a singleton list it is never false, so the wrong path is never
taken; then applying [0] to the whole thing gets the b or c that you
really wanted. Ugly, but it gets you there in the rare cases where it
is really inconvenient to rewrite your code using 'if'. 

This method is of course trivially defined as lambda:

IF = lambda a,b,c:(a and [b] or [c])[0]

* Using WHILE statements

Here is a generic while loop:

while expression():
    function()

So, given expression and function, you can write this as:

WHILE = lambda e,f:(e() and (f(),WHILE(e,f)) or 0)

Which is sort-of tail recursive, but uses a lot of stack.

* Using FOR statements

Here is a generic for loop:

for x in items:
    f(x)

Use map() instead. You can rewrite this as:

map(f,items)

Which has the benefit of being faster in most cases, too.

* Using Multiple expressions in Lambda

There are several possible solutions to the problem of how to use
several distinct expressions in one lambda statement. The general form
of the problem is: How to use

    f1()
    f2()
    ...
    fn()

in one lambda statement. The basic idea used in most of the following
is to rewrite that as

for function in (f1, f2, f3):
    function()

* Use tuples and rely on evaluation order

If the functions are not related, you can say:

(f1(), f2(), ..., fn())

The current evaluation order is from left to right, so that should
work.

* Use tuples, and implement your own evaluation order

If you are a coward and fear that evaluation order will change in a
future release of python, you can use eval

map(eval,("f1()","f2()",...,"fn()"))

You can also use apply

map(lambda x:apply(x,()),(f1,f2,f3))

* Using local variables

The best idea is to define a helper lambda function that has the local
variables as argument. For example, say you have this function:

def f(x):
    result = range(20)
    return do_something(result,x)

You can rewrite this as

f = lambda x: (lambda result=range(20):do_something(result,x))()

As mentioned above, local variables are best used in a list, because
that way you can use [].__setitem__ for data assignments.

* Misc topics

Here are some other topics that don't seem to fit in any of the above.

* Wrappers for list functions that return None

Many list functions return None, so you cannot easily use them in a
stacked expression inside lambda. Examples of this are [].sort() and
[].reverse(). Instead of writing:

list.some_function_that_returns_None()
function(list)

you can write

function((list.some_function_that_returns_None(),list)[1])

Here is an example:

import random

sort = lambda x: (x.sort(),x)[1]
reverse  = lambda x: (x.reverse(),x)[1]

test = range(20)
print test
random.shuffle(test)
print sort(test)
print reverse(sort(test))




From James_Althoff at i2.com  Mon Feb  4 14:34:05 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Mon, 4 Feb 2002 11:34:05 -0800
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
Message-ID: <OF66B4D0F9.B651EC9E-ON88256B56.00664B8B@i2.com>

[Tim Hammerquist]
>Imagine a "Lisp-style" indent!
>
>if ( some_condition() ) {
>    i = first_i();
>    for (i = first_i(); i <= 10; i++) {
>        initialize_some_stuff();
>        while ( some_other_condition() ) {
>            do_some_more_stuff() } } }

Although one rarely encounters the above indentation style in C/C++/C#/Java
code -- and, in fact, one occasionally sees the suggestion of said style
accompanied by giggling ridicule -- it is amusing to reminisce about the
path taken by the Smalltalk-80 development team.  Smalltalk-80 has block
delimiters "[" and "]".  When we first started implementing the
Smalltalk-80 system (all of which was written in Smalltalk itself, save the
VM, object memory, and primitives) we had the usual debates about delimiter
style.  As with C coders since the dawn of time, each style had strong
advocates.  And each developer used his/her style of choice.  Until ...
Dan Ingalls, who was the lead architect for Smalltalk-80, had a very strong
preference for the above "indentation-based" style.  And he was clever
enough to write a pretty-print routine that converted code to his preferred
style *and* to add a call to his pretty-printer to the "compile" menu in
the editor.  Voila.  No more inconsistency and no more debates.  Although
not everyone immediately appreciated this direct approach to "solving" the
"inconsistent style" problem, the end result -- interestingly -- was that
*all* early Smalltalk-80 code had a very nice "Pythonic"
indentation-oriented look and feel (assuming that you ignored a little bit
of cruft at the end of each line <wink>).  Note that finding matching
brackets was not an issue because the editor had a feature whereby if you
clicked inside a left bracket it would highlight all text to the matching
right bracket.

(And, I suppose, the next step should have been to put the pretty-print/tab
stuff in the compiler itself <wink>)

Jim




From alien275 at libertysurf.fr  Wed Feb 27 13:14:22 2002
From: alien275 at libertysurf.fr (Riss Nicolas)
Date: Wed, 27 Feb 2002 19:14:22 +0100
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de> <a5j53i$oms$1@wanadoo.fr> <8z9f2rks.fsf@fido.trinitycapital.com>
Message-ID: <a5j7fb$fb$1@wanadoo.fr>

No, the port is correct. The telnet server is cingured to listen of the port
32.
Another suggestions ?

Thanks
"Mike Johnson" <afp_randjohnson at yahoo.com> a ?crit dans le message de news:
8z9f2rks.fsf at fido.trinitycapital.com...
> "Riss Nicolas" <alien275 at libertysurf.fr> writes:
>
> > I jave used it but when y tried to connect it makes an error:
> >    File "c:\test.py", line 9, in ?
> >     tn = telnetlib.Telnet("192.168.0.9",32)
> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 190, in __init__
> >     self.open(host, port)
> >   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 219, in open
> >     raise socket.error, msg
> > socket.error: (10061, 'Connection refused')
> >
> > Programm:
> >
> > import getpass
> > import sys
> > import telnetlib
> > password = getpass.getpass()
> > tn = telnetlib.Telnet("192.168.0.9",32)
> > tn.read_until("Administrator Password : ")
> > tn.write(password + "\n")
> > tn.write("session\n")
> >
> >
> > Where is the mistake ?
>
> Wrong port. Try:
>
> tn = telnetlib.Telnet("192.168.0.9", 23)
>
>
> - Mike Johnson
>




From ol1 at v10a.com  Fri Feb 22 04:12:48 2002
From: ol1 at v10a.com (O Polite)
Date: 22 Feb 2002 10:12:48 +0100
Subject: Hosting
In-Reply-To: <3C7504E0.2050602@thinkware.se>
References: <283adf56.0202201551.7026d548@posting.google.com>
	<mailman.1014250235.21771.python-list@python.org> 
	<3C7504E0.2050602@thinkware.se>
Message-ID: <1014369169.829.16.camel@matijek>

On Thu, 2002-02-21 at 15:32, Magnus Lyck? wrote:

> ADSL here is about $30 in the cheapest version. That will not
> include a static IP, maybe you won't even be allowed to use
> a server. Don't really know.
> 
Actually Bostream, who operates in Stockholm and Gothenburg, will give
you a static IP number. I don't remember what the contract says but I
run all the services I like and they haven't complained. They'd probably
come down on me if traffic loads got high. They monthly fee is 250 SEK
(25 USD)






From rdsteph at earthlink.net  Wed Feb 27 21:18:28 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Thu, 28 Feb 2002 02:18:28 GMT
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org> <3C7D8AC4.3F9A74D0@engcorp.com> <yMff8.10349$pC1.1875911@newssrv26.news.prodigy.com>
Message-ID: <3C7D9410.A7841097@earthlink.net>

Yes! Sign up Mark Hammond to do an "Advanced Python on Win32" and you
will have a winner indeed, Ms. Capparelle!

Look, I hope you publish dozens of Python books (anything to help the
cause, besides, I gotta keep reviewing 'em), but I know what Peter
means: it would really be more helpful at this point to have more
advanced Python books, and books exploring various Python specialties. 

How about a Python and Bioinformatics, a Python Scientific Programming,
a Python and AI, an Advanced Pythonic Extensions and Embedding book, 
Python Game Programming, Advamced Pythonic GUI's, Python Functional
Programming and re-Factoring? 

Best of all, how about a "On the Meaning of Life" by Tim Peters? 

And all the authors you'll ever need for all these great advanced books
are right here on the newsgroup Ms. Capparelle, you needn't look any
further. 

Hmm, better leave a few of them alone or we'll miss their open source
contributions as they all churn out more and more books... ;-)))

Ron Stephens, Python City
http://www.awaretek.com/plf.html


From gerson.kurz at t-online.de  Sun Feb 24 00:51:46 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 24 Feb 2002 05:51:46 GMT
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com>
Message-ID: <3c787c1f.2523859@news.t-online.de>

On 23 Feb 2002 11:52:10 -0800, bobnotbob at byu.edu (Bob Roberts) wrote:

>What would be a good way to replace every one or more spaces (" ") in
>a string with just one space?  Or replace any number of newlines with
>just one?

Lets see. There were four solutions mentioned:

-------------(cut here)---------------
def test1(newstring):
    while newstring.find('  ') > -1:
        newstring = newstring.replace('  ', ' ')
    return newstring

def test2(newstring):
    return " ".join(filter(None,newstring.split(' ')))

def test3(newstring):
    return re.sub(' +', ' ', newstring)

def test4(newstring):
    return ' '.join(newstring.split())
-------------(cut here)---------------

Note that test4() does also split newlines (and tabs), this is why
test2() explicitly splits for *blanks only*. Given these, you can test
which one is fastest

-------------(cut here)---------------
def measure(count,func):
    start = time.clock()    
    for i in xrange(count):
        func("This  is a    test\n isn't it?")
    print "%s took %.2f for %d elements" % (str(func),time.clock() -
start, count)

for count in [pow(10,i) for i in range(5,7)]:
    for func in (test1, test2, test3, test4):
        measure(count,func)
    print
-------------(cut here)---------------

results in

test1 took 1.82 for 100000 elements
test2 took 2.12 for 100000 elements
test3 took 3.75 for 100000 elements
test4 took 1.38 for 100000 elements

test1 took 19.06 for 1000000 elements
test2 took 22.17 for 1000000 elements
test3 took 38.20 for 1000000 elements
test4 took 12.17 for 1000000 elements

So, the simple "while"-solution (test1) is actually the fastest if you
really only want to replace spaces (and not newlines and tabs, also).
I find that quite interesting because the other three solutions seem
so much more sophisticated.


From mikael.lexen at stop.spam.volvo.com  Fri Feb 15 10:40:30 2002
From: mikael.lexen at stop.spam.volvo.com (Mikael =?iso-8859-1?Q?Lex=E9n?=)
Date: Fri, 15 Feb 2002 16:40:30 +0100
Subject: VB code -> Python
References: <3C6CDDC8.4ED3B329@stop.spam.volvo.com> <a4iv0k$al9$1@lise.netcom.no>
Message-ID: <3C6D2BEE.A494B1D6@stop.spam.volvo.com>

That's what I'm doing to day but I want to get ride of the "home made" COM
object and just do it with "native" python. Basically I want to list all
NT-groups a user have in a domain.

/Mikael

"Andr? Risnes" wrote:

> "Mikael Lex?n" <mikael.lexen at stop.spam.volvo.com> wrote in message
> news:3C6CDDC8.4ED3B329 at stop.spam.volvo.com...
> [...]
> > So my questin is how do I translate the GetObject(...) thing in to
> > Python code.
>
> I don't know, but could you do this instead: Make a COM object in VB
> with a function that returns the result of GetObject, and call that function
> from Python?
>
> --
> Regards
> Andr? Risnes

--
/Mikael

_________________________________________________
Mikael Lex?n
Volvo Information Technology AB
Dept 9514, HD2N
SE-405 08 G?teborg, Sweden

Telephone: +46 31 3215757
E-mail: mikael.lexen at volvo.com




From skip at pobox.com  Fri Feb  8 15:13:37 2002
From: skip at pobox.com (Skip Montanaro)
Date: Fri, 8 Feb 2002 14:13:37 -0600
Subject: What is the most efficient way to do this?
In-Reply-To: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
References: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
Message-ID: <15460.12657.784894.865356@12-248-41-177.client.attbi.com>

    mdk> What is the most efficient way to remove empty items from this
    mdk> list? 

    mdk> ['', '1', '2', '3', '', '456', '789', 'wow']

I'll give you a couple options and let you worry about efficiency...

    >>> l = ['', '1', '2', '3', '', '456', '789', 'wow']
    >>> filter(None, l)
    ['1', '2', '3', '456', '789', 'wow']
    >>> [x for x in l if x != '']
    ['1', '2', '3', '456', '789', 'wow']
    >>> while '' in l:
    ...   l.remove('')
    ... 
    >>> l
    ['1', '2', '3', '456', '789', 'wow']

Note that the first two ways create new lists and leave l untouched.  The
third way modifies l in place.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From sholden at holdenweb.com  Thu Feb 21 16:55:20 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 21 Feb 2002 16:55:20 -0500
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com> <258fd9b8.0202201733.3888e711@posting.google.com> <mailman.1014287322.17283.python-list@python.org> <3C74E31A.1000203@thinkware.se>
Message-ID: <K0ed8.53729$n73.986413@atlpnn01.usenetserver.com>

"Magnus Lyck?" <magnus at thinkware.se> wrote in message
news:3C74E31A.1000203 at thinkware.se...
> M.-A. Lemburg wrote:
>
> > The egenix-mx-* packages should compile out of the box
> > using cygwin, not sure about the other extensions.
>
>
> That's great. I guess what I was trying to point out was
> that normal windows installs for python modules are unlikely
> to work with cygwin. And this will cause some inconveniences
> (and confusion if you aren't aware of the problem).
>
This is true to an extent. But certainly Marc Andre's modules do compile on
cygwin - licensing conditions mean they can't be distributed in pre-compiled
binary installable form. And MySQLdb also compiles. I'd like to see every
Windows-capable module run under both: it turns out you don't usually have
much to do to make the same extension install under Windows and cygwin.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From gh_pythonlist at gmx.de  Wed Feb  6 02:59:08 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Wed, 6 Feb 2002 08:59:08 +0100
Subject: cx_Oracle on cygwin
In-Reply-To: <Xns91AD8D95B2B69andy47halfcookedcom@203.109.250.24>
References: <Xns91AD8D95B2B69andy47halfcookedcom@203.109.250.24>
Message-ID: <20020206075907.GA1182@lilith.hqd-internal>

Le 06/02/02 ? 02:55, Andy Todd ?crivit:
> [...]
> Now everything is fine until we reach this stage;
> 
> gcc -shared -Wl,--enable-auto-image-base build/temp.cygwin-1.3.9-i686-
> 2.2/cx_Oracle.o -L/Tools/Lib -L/usr/lib/python2.2/config -loci -
> lpython2.2 -o build/lib.cygwin-1.3.9-i686-2.2/cx_Oracle.dll -s
> /usr/lib/gcc-lib/i686-pc-cygwin/2.95.3-5/../../../../i686-pc-
> cygwin/bin/ld: cannot find -loci
> collect2: ld returned 1 exit status
> error: command 'gcc' failed with exit status 1
> 
> Reading 'man' for gcc this *seems* to be a problem because the linker is 
> looking for a file called liboci.a which, naturally, doesn't exist under 
> windows. I think I should be linking to oci.dll, but I don't know enough 
> to be sure.

I've little experience with Cygwin, but I have some with mingw32.

You need to create an import library for the oci.dll fle. The import
library will have a name liboci.a. This is a two-step task.

1) Create a .def file for the DLL
2) Create the import library

Assuming the mingw approach also works with Cygwin (which I'm 95 % sure
it does), you can look here: http://www.mingw.org/docs.shtml in the
section "How to create a def file for a dll".

This should work (more options at the URL above):

    echo EXPORTS > oci.def
    nm oci.dll | grep ' T _' | sed 's/.* T _//' >> oci.def
    dlltool --def oci.def --dllname oci.dll --output-lib liboci.a

Be sure to put liboci.a in a place where Python finds it when linking.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 8.2 ?C      Wind: 0.3 m/s



From h.jehadalwan at student.kun.nl  Fri Feb 22 17:48:25 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Fri, 22 Feb 2002 23:48:25 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org>
Message-ID: <3C76CAB9.50509@student.kun.nl>

Gustavo Cordova wrote:

>>  pdbdir = os.popen('ls /home/data/pdbs') # if this directory 
>>contains 
>>10000 files, the code does not work.
>>
>>files=pdbdir.readlines()
>>pdbdir.close()
>>(o,i) = popen2.popen2("/home/programs/profit ")
>>
>>i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") 
>># This the 
>>reference molecule.
>>
>>for PDB in files:
>>     i.write(" mobile /home/data/pdbs/" + PDB + "\n") # A new 
>>molecule.
>>     i.write("fit\n")		# compares reference and new molecules.
>>
>>i.close()
>>result= o.read()
>>o.close()
>>
>>
> 
> You read from the input pipe only upon writing all your data,
> but maybe the pipe is full?
> 
> How about changing the loop with this:
> 
> 
> log = StringIO.StringIO()
> 
> for PDB in files:
>   i.write("mobile /home/data/pdbs/%s\n" % PDB)
>   i.write("fit\n")
>   log.write(o.read())
> 
> i.close()
> o.close()
> 
> 
> So now your results are in the log object.
> 
> -gus
> 
> 

hi Gustaveo,
You'r solution did not work either. The code hangs at the first loop, 
specifically at "log.write(o.read())".



From sholden at holdenweb.com  Fri Feb  1 13:14:47 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 13:14:47 -0500
Subject: Equivalent of Perl chomp?
References: <mailman.1012513005.17840.python-list@python.org> <Hxk68.4423$Xk4.215834@news1.east.cox.net> <u5js1gd8eecda8@corp.supernews.com>
Message-ID: <UYA68.107130$8b.4038386@atlpnn01.usenetserver.com>

"Paul Watson" <pwatson at mail.com> wrote ...
> Perl chomp only removes -all- traliing newlines when in paragraph mode.
> Paragraph mode is when $/ is set to "".  The default mode is not paragraph
> mode, so chomp removes only one (1) newline.
>
> === code
> $s = "now\n\n\n";
> print "length($s) = ".length($s)."\n";
> chomp($s);
> print "length($s) = ".length($s)."\n";
>
> === output
> length(now
>
>
> ) = 6
> length(now
>
> ) = 5
>
Maybe we shoud ask the guys on c.l.perl about this [ducks and runs]

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From wayne at mishre.com  Tue Feb 12 00:37:53 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Mon, 11 Feb 2002 21:37:53 -0800
Subject: Possible dumb query
References: <000801c1b385$97747aa0$0100000a@ahimsa>
Message-ID: <3C68AA31.8040402@mishre.com>

ahimsa consulting wrote:
> Greeting Pythonites
> 
> This might qualify as one of those "really dumb questions", but:
> 
> Under Linux RH 7.2 when doing a search for which python ("$ which python") I
> have discovered that I have one entry for python as /usr/bin/python and one
> entry for python2.2 as /usr/bin/python2.2. My question is this: if I were to
> delete the directory /usr/bin/python what impact - if any - will it have on
> any of the libraries/ etc. that I would require to use python2.2

If they are separate installs it shouldn't cause a problem, which is 
usually the case.  RH usually ships both versions of something like that 
to allow a transition period for developers.  If you want to verify this 
check the lib directory to see if any of the .py or .pyc files are 
linked to the other directory.

Although you really shouldn't just delete the directory, check to see if 
a python RPM has been installed.  This way you will see if any other 
programs have a dependency (sp?) on that version of Python.  Even if 
there is no other program relying on it, removing the RPM is usually 
cleaner than just rm-ing the directory.

> Once again, my apologies if this is a dumb query, but gotta learn somehow.

You could always just remove it and see what happens. >:]

HTH,

-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From bokr at oz.net  Sat Feb 16 02:41:59 2002
From: bokr at oz.net (Bengt Richter)
Date: 16 Feb 2002 07:41:59 GMT
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au> <a4j5hd$l3j2$1@ID-11957.news.dfncis.de> <mailman.1013800706.14636.python-list@python.org> <a4jsk2$q4e$0@216.39.172.122> <mailman.1013807725.17084.python-list@python.org>
Message-ID: <a4l2g7$f5r$0@216.39.172.122>

On Fri, 15 Feb 2002 15:59:16 -0500, "Dr. David Mertz" <mertz at gnosis.cx> wrote:

>|>In other words, I don't believe that Brainbench's Python exam
>|>is a *good* testing tool.
>
>|What would your thoughts be re an automated exam as a PSF - licensed free
>|collaborative project? Maybe even the exam definition itself could somehow
>|feed off a wiki-like process, if you constrained formats appropriately
>|for defining test elements?
>
>I don't necessarily have anything against testing in general.  It is
>possible--at least to a first approximation--to learn something about
>people's skill levels by means of tests.  And I wouldn't even go so far
Actually, I was thinking more in terms of test as a self-assessment/tutorial
tool than an employer's screening thing.

>as to suppose that a complete novice would get the same result as Alex
>Martelli on the Brainbench exam.  It's not as good as it *should* be, but
>it's not completely random either.
>
I'd guess it would be more effective used privately to assess one's own
mastery (or lack thereof), since one is privy to one's own internal
reactions, where an employer can only see external signs and end results.

>Then again, I don't have so much faith in testing to really think making
>an exam is the best direction for PSF resources (just think what they
I didn't mean using PSF resources (except by the usual way of side effects
from Python jocks and peacocks doing their thing). I just meant an open
project under PSF-style license, with copyrights assigned to the PSF.
Maybe it could be useful in the CP4E context also? (Maybe there is even
a grant to be pursued?)

>could accomplish if Tim Peters were to sell his second hand copy of _How
>to Form a Non-Profit_, and spend the proceeds on something important).

>Nonetheless, if someone wants to pay me more money than Brainbench did,
>I would be happy to prepare a better exam than Brainbench let me do :-).
>
I haven't seen the Brainbench tests. I was musing in terms of a kind of
topic-mastery validation suite, run as a script with opportunity to type
in actual code snippets as well as T/F and multiple choice answers for
evaluation and comparison to expected results.

Dynamic evaluation could allow some interesting test/tutorial questions,
and modules could have try-again/skip options. And the whole test/exercise
could be automatically configured to take into account correct prior results,
so you could concentrate on filling in holes. Like some (natural) language
learning programs.

Collaborative development of the material is an idea to get around the
pre-conditions imposed by a direct profit motivation and its likely
narrowing effect on focus and ideas. (So here is my OTTOMH preconceptions ;-)
There would just have to be some pattern for a test module so that it could
be easily integrated into the whole. (The test suite that comes in the Python
distribution probably has a lot of useable stuff)?

IOW, people who had trouble (or particularly enjoyed an aha moment,
as the case may be) might be motivated to contribute a tutorial/test
module that would make sure the user knew that there was something to know,
and whether s/he knew it or not.

Eventually it might be cross-indexed with the regular documentation
(e.g., results in HTML with links from wrongly answered questions
to relevant docs).

Just musing...

Regards,
Bengt Richter




From mcfletch at geocities.com  Sat Feb 23 16:20:54 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Sat, 23 Feb 2002 16:20:54 -0500
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com> <a582kl.3vsjht5.1@kserver.org>
Message-ID: <3C7807B6.3090601@geocities.com>

Another approach to the original problem (replace X instances of Y with 
a single instance of Y)...:

 >>> import string
 >>> data = 'this\n\nthose\n\n\nthem\ntheir\nthose'
 >>> string.join(filter( None, string.split( data, '\n')), '\n')
'this\nthose\nthem\ntheir\nthose'
 >>>

Which has the virtue of letting you do the same thing with _any_ 
character or multi-character string just by parameterising '\n' in the 
above.

For your particular code (replace \n _and_ ' ' with single spaces), why not:

 >>> data = 'this\n\nthose\n\n  \nt s a s t o v hem\ntheir\nthose'
 >>> string.join(string.split(data ), ' ')
'this those t s a s t o v hem their those'

Which replaces all whitespace runs with the single space...

Enjoy,
Mike


Sheila King wrote:
> [posted and mailed]
> 
> On 23 Feb 2002 11:52:10 -0800, bobnotbob at byu.edu (Bob Roberts) wrote in
> comp.lang.python in article
> <c4e6b17d.0202231152.5f87765d at posting.google.com>:
> 
> 
>>What would be a good way to replace every one or more spaces (" ") in
>>a string with just one space?  Or replace any number of newlines with
>>just one?
>>
> 
> This function, which I'm currently using in an email filter, replaces
> all '\n' characters with a space, and then replaces all occurrences of
> more than one space with a single space.
> 
> def remove_new_lines(searchstring):
>     """ Returns a string without newlines or double spaces. """
>     newline = '\n'
> 
> 	# Replace newlines with spaces
>     newstring = searchstring.replace(newline, ' ')
> 
> 	# Replace multiple spaces with single spaces 
>     while newstring.find('  ') > -1:
>         newstring = newstring.replace('  ', ' ') 
>     return newstring
> 
> You could also do this with regular expressions, but this method only
> requires regular string operations.
> 
> 


-- 
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From eppstein at ics.uci.edu  Sun Feb 24 22:34:34 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Sun, 24 Feb 2002 19:34:34 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu>
Message-ID: <eppstein-69765A.19343424022002@news.service.uci.edu>

In article <a5cb54$27c2$1 at agate.berkeley.edu>,
 Daniel Yoo <dyoo at hkn.eecs.berkeley.edu> wrote:

> : But what is the functional equvalent of:
> 
> : for x in items:
> :     x.f()
> 
> 
> 
> Here's one way to do it:
> 
> ###
> map(lambda x: x.f(), items)
> ###

I'd prefer
[x.f() for x in items]

It's not functional syntax, but so what?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From gbreed at cix.compulink.co.uk  Fri Feb 15 12:38:29 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Fri, 15 Feb 2002 17:38:29 +0000 (UTC)
Subject: Using Python to generate MIDI files (Mac)
References: <5e8bd451.0202150813.20720fc5@posting.google.com>
Message-ID: <a4jh2l$jid$1@thorium.cix.co.uk>

kevin parks wrote:

> Just curious. I would have thought that someone might have done this
> buy now. I am surpised there is not much call for MIDI modules in
> Python. MIDI (though in general total crap for some musical
> application (like anything not equal-tempered, which pretty much kills
> anything i am doing) (ironic isn't it? MIDI was made for music!, see
> skinny and other protocols for what should have been) MIDI is ok for
> some music applications and for controlling (music, lighting, etc.)

It's interesting you mention not equal-tempered in that rant.  What are 
you actually working on?  I've got C++ MIDI software, see 
<http://www.microtonal.co.uk/software.htm> you might get some inspiration 
from.  Now you have a basic library, perhaps I could assist you on 
improving it.  Contact me privately if this is getting to specialist.


                  Graham


From fdrake at zope.com  Mon Feb 18 11:06:44 2002
From: fdrake at zope.com (Fred Drake)
Date: Mon, 18 Feb 2002 11:06:44 -0500
Subject: [Idle-dev] IDLE detection code snippet
In-Reply-To: <3C702E59.7396.FDF741@localhost>
Message-ID: <web-3559587@digicool.com>

On Sun, 17 Feb 2002 22:27:37 -0800
 "Jesse W" <jessw at loop.com> wrote:
> detecting when IDLE is running.  I think this code is a
> solution to 
> that problem.  Please try it out and see if it works
...
> def IDLEtest():
>     """Returns 1 when IDLE is running, 0 else.
>     Please let me know (through the IDLE-devl list) if
> you find
>     a situation where this gives the wrong answer."""
>     import sys
>     try:
>         if sys.stdin.__module__=='PyShell':
>             return 1
>         else:
>             return 0
>     except AttributeError:
>         return 0

Simplify:

def IDLEtest():
    import sys
    try:
        return sys.stdin.__module__ == 'PyShell'
    except AttributeError:
        return 0



  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Zope Corporation



From cfelling at iae.nl  Fri Feb 22 19:34:49 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 23 Feb 2002 01:34:49 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl>
Message-ID: <a56o39$7dg$1@animus.fel.iae.nl>

husam <h.jehadalwan at student.kun.nl> wrote:

> hi Gustaveo,
> You'r solution did not work either. The code hangs at the first loop, 
> specifically at "log.write(o.read())".

You don't seem to be having a Python problem, so what about trying to
get some output from the program?  Instead of reading the log into a
var, what about trying it with `os.popen("/home/programs/profit", "w)'
and have the output directed at the terminal for direct inspection.
Maybe you're just missing error messages:)

An other problem that might buck you (but I don't see how this would
be the case as your program seems to work for a few files) is that
pipes are buffered.  So uses `i.flush()' to be sure that what you
wrote really is transmitted to the other end of the pipe.

And as a last resort you could try reading the `ls' output a line at a
time.

    i = os.popen("/home/programs/profit ", "w")

Should this ---------------------------^ space be there?

    i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") 
    for PDB in os.popen('ls /home/data/pdbs'):
        i.write(" mobile /home/data/pdbs/" + PDB + "\n")

Should this -----^ space be there?

        i.write("fit\n")
        i.flush()
    i.close()
-- 
groetjes, carel


From gerson.kurz at t-online.de  Thu Feb 21 08:04:47 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 13:04:47 GMT
Subject: Stupid Lambda Tricks [LONG]
References: <mailman.1014295775.30436.python-list@python.org>
Message-ID: <3c74effc.290906@news.t-online.de>

On Thu, 21 Feb 2002 14:05:46 +0000, Luigi Ballabio <ballabio at mac.com>
wrote:

>I wouldn't advise using anything of the sort of your IF thing, because it 
>doesn't short-circuit. Try this:
>
>def a():
>     return 1
>def b():
>     print "lengthy b() calculation, possibly with side effects"
>def c():
>     print "lengthy c() calculation, possibly with side effects"
>
># Use the ternary trick:
> >>> (a() and [b()] or [c()])[0]
>lengthy b() calculation, possibly with side effects
>
># Use you IF thing instead:
> >>> IF(a(),b(),c())
>lengthy b() calculation, possibly with side effects
>lengthy c() calculation, possibly with side effects

ok, it should read

IF = lambda a,b,c:(a() and [b()] or [c()])[0]
IF(a,b,c)

if a, b or c need arguments, store them in a list and do

IF = lambda a,b,c,x:(a(x) and [b(x)] or [c(x)])[0]




From akakakhel at attbi.com  Thu Feb 28 21:56:23 2002
From: akakakhel at attbi.com (Ali K)
Date: Fri, 01 Mar 2002 02:56:23 GMT
Subject: images
References: <iGhe8.20623$vP.68360@rwcrnsc51.ops.asp.att.net> <a5f093$6mqk7$1@ID-12869.news.dfncis.de>
Message-ID: <r5Cf8.24338$TB2.108678@rwcrnsc52.ops.asp.att.net>

How do i do this with Tkinter?


"Ricky" <store_li at sina.com> wrote in message
news:a5f093$6mqk7$1 at ID-12869.news.dfncis.de...
> If you want to display a pic, you must use one GUI toolkit. If your
program
> is not very explicit, you may use PIL+Tkinter.
>
> "Ali K" <akakakhel at attbi.com> wrote in message
> news:iGhe8.20623$vP.68360 at rwcrnsc51.ops.asp.att.net...
> > I would like to know how to show a picture in Python so that it shows up
> on
> > my window.
> >
> > Please Reply.
> >
> > Thank You.
> >
> > Ali K.
> >
> >
>
>




From mwh at python.net  Fri Feb  1 05:30:51 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 1 Feb 2002 10:30:51 GMT
Subject: Building Python 2.2 - curses module test failure
References: <a370or$230$1@leviathan.uklinux.net> <u4rl41626.fsf@python.net> <a39043$18v$1@leviathan.uklinux.net> <u8zafzwlu.fsf@python.net> <u4rl3zvya.fsf@python.net> <a39amn$1om$1@leviathan.uklinux.net> <a3c55h$2ac$1@leviathan.uklinux.net>
Message-ID: <uvgdhxzj6.fsf@python.net>

Nick Chalk <nick at leviathan.uklinux.net> writes:

> [ Apologies if this screws up the threading. My
> news server machine chose last night to have a
> crisis, and all of comp.lang.python ended up in
> lost+found. :-/ ]
> 
> I pulled the three updated files from CVS, and
> rebuilt with the same debugging, no optimization
> CFLAGS as before. The curses tests now run fine -
> no segfaults, no warnings of invalid pointers.

Good.

> One operator trap to watch for when running the
> tests - don't try to redirect stdout, ncurses
> doesn't like it. Unfortunately, if you do
> something like...
> 
>    make test 2>&1 | tee Test.log
> 
> ...which is what I habitually do, ncurses doesn't
> warn you - it just fails hard. :-(

I can believe that.  But if you run it exactly like that, test_curses
shoudln't run...

> Problem solved! Thanks for your help.

No problem.  Thanks for the report!

Cheers,
M.

-- 
40. There are two ways to write error-free programs; only the third
    one works.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From phr-n2002a at nightsong.com  Tue Feb 12 22:46:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 19:46:36 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com>
Message-ID: <7x1yfq2gdf.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> Paul, this thread's probably now old enough for you to tell us what the real
> problem is! Why exactly do you need to search backwards from the 50,000th
> character to find the beginning of an HTML tag?

Suppose I'm parsing the file and I see a </table> tag and I want to
find the matching <table> tag.  It could be pretty far back in the file.
That's what I was doing when I encountered this question.  But searching
backwards is a normal thing to want to do in general--for example it's
a standard command in any decent text editor.

Anyway, I just entered a sourceforge bug about it being missing from
Python's re module.

Thanks


From slhath at charter.net.nospam  Sat Feb 23 10:16:26 2002
From: slhath at charter.net.nospam (Scott Hathaway)
Date: Sat, 23 Feb 2002 09:16:26 -0600
Subject: call me
Message-ID: <u7fcig3gep8k40@corp.supernews.com>

Mom,

Give me a call when you have a chance.

Thanks,
Scott




From jkraska at san.rr.com  Fri Feb  1 01:15:16 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 01 Feb 2002 06:15:16 GMT
Subject: Java vs Python Benchmarks: Java is faster
References: <20020201015223.GC4394@dman.ddts.net> <mailman.1012540142.24714.python-list@python.org>
Message-ID: <a0dk5u4g7t8de7hifl0a8ljob1li3v9p4o@4ax.com>

>1.  You can buy a real obfuscator for a few hundred bucks.

Give the man a dollar. :)

>3.  If you must do this yourself:  ANTLR (for example) comes
>    with everything you need to tokenize Java source code ...

I have to say, I heartily recommend ANTLR all the way around.
It's _far and above_ the best lexer/compiler/ast/walker generator
I've ever encountered. That it combines all of these duties into
one integrated whole and does so very well is simply amazing.

http://www.antlr.org

C//



From oren-py-l at hishome.net  Mon Feb 18 05:34:28 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Mon, 18 Feb 2002 05:34:28 -0500
Subject: Warning about "for line in file:"
In-Reply-To: <3C6D73DE.3090201@wi.mit.edu>
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu>
Message-ID: <20020218103428.GA52916@hishome.net>

On Fri, Feb 15, 2002 at 03:47:26PM -0500, Brian Kelley wrote:
> count = 0
> for line in file.xreadlines():
>     if count > 10: break
>     print line
>     count = count + 1
> 
> for line in file.xreadlines():
>     print line
> 
> So what is REALLY happening is that you are creating two seperate 
> iterators in the above examples.  Writing "for line in file" instead of 
> "for line in file.xreadlines()" simply hides and confuses this.

If you trace the problem to its true source you will see that file objects 
are not really containers that can be iterated - they are already iterators.
The container is the file on the disk.  A file iterator object is not a
real independent object, just a different protocol to access the file object 
using next() and StopIteration instead of readline() and an empty string.
The buffering problem that started this thread is just a side-effect of this 
case of mistaken identity: iterators pretending to be containers.

There is no need for a separate object to implement another protocol. A
single object can expose both the iterator and file protocols:

class file_(file):
    def __iter__(self):
        return self

    def xreadlines(self):
        return self

    def next():
        s = self.readline()
        if s:
            return self
        else:
            raise StopIteration

I believe it is fully backward compatible with existing sources that use 
file iteration. Implementing file iterators using xreadlines objects was just 
the quickest way to do it by reusing an existing piece code that was written 
long before the iterator protocol.

	Oren




From david.abrahams at rcn.com  Wed Feb 20 00:08:16 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Wed, 20 Feb 2002 00:08:16 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <a4vaqi$j8e$1@bob.news.rcn.net>

"Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
news:3C7283E4.4050604 at physics.ucsb.edu...
>
> Hi all.
>
> I'm doing some scientific computing and have
> amassed a moderate set of classes.  Due to the
> heavy computational demands, I don't think I
> can justify rewriting everything in an interperated
> language; nice as that would be now that I'm
> playing around with what I've created.
>
> I'm in a situation now where I've written about
> a dozen of quickie, throw away, utility routines
> that I call from the command line, and in essence,
> BASH has become my scripting environment.
>
> Unfortunately, the number of throw-away routines
> is exploding, and I feel like I'm loosing the advantage
> of having constructed a nice set of classes with
> which to play.
>
> I would, of course, like to have some sort of scripted
> environment (e.g. Python) in which to "play"
> with my objects.
>
> That said, I'd like to solicit opinions as to the relative
> merits of the various possible approaches.

There is a comparison of some of the choices at:

http://www.boost.org/libs/python/doc/comparisons.html

> I'm making essential use of templates and polymorphism.

One of the design goals for Boost.Python was to avoid the language support
issues that tend to plague systems which attempt to parse C++. Parsing C++
is hard; AFAIK all of the wrapping systems which try to do it break down in
some areas. Boost.Python uses the builtin type inference abilities of your
C++ compiler instead of parsing the C++ code separately. Another issue is
that the interfaces to your C++ code doesn't typically carry enough semantic
information to tell a system how it should be wrapped: if I pass you a
pointer, are you taking ownership? Are you going to hold onto it?, etc.
Boost.Python V2 currently under development addresses those issues.

Another design goal was to allow you to wrap your C++ code without intruding
on its design. One example of a relatively large scientific project using
Boost.Python is http://cctbx.sourceforge.net/index.html,


> I have no experience with Boost.Python, but from what I
> understand, it would not involve writing wrapper definitions
> (ala SWIG or IDL), right?  Are there other options?

Since I'm not intimately familiar with SWIG or IDL, I'm not sure what
wrapper definitions you're referring to. You do have to write a small amount
of additional code to export your C++ to Python using Boost.Python. You can
see examples of what's required to make Python modules from your C++ code
starting here: http://www.boost.org/libs/python/doc/example1.html

You might also consider posting questions to the C++ sig
(www.python.org/sigs/c++-sig).

HTH,

Dave






From sholden at holdenweb.com  Wed Feb 20 07:53:45 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 07:53:45 -0500
Subject: Help! IIS and Python
References: <eqlm6ugbbtsu4i1d61pek0c4g49as7j119@4ax.com> <a4g77r$ngb$1@thorium.cix.co.uk> <oqcu6uc4poo2akj76sg1lh76n8tp7kkdcu@4ax.com>
Message-ID: <3%Mc8.16921$Zp2.1738068@atlpnn01.usenetserver.com>

"Tim Roberts" <timr at probo.com> wrote ...
> gbreed at cix.compulink.co.uk wrote:
> >
> >> Since we're on the subject of Python in ASP programming, how do you use
> >> the <%=xxx%> trick with indentation?  This fails with an indentation
> >> error, for example:
> >>
> >> <%
> >> for i in range(5):
> >> %>
> >>   <%="aaa%d" % i %>
> >
> ><%
> >formatter = """
> >  aaa%d
> >"""
> >
> >for i in range(5):
> >  Response.Write(formatter%i)
> >%>
> >
> >Why not?
>
> We're diving deeply into religion and philosophy here.  Partly, it's just
> the principle of the thing; I assumed there was a magic method to do this
> with Python, but I'm coming to the conclusion that there is not.  The
> "Response.Write" call seems extraneous to me; VBScript and JScript forms
> can omit it with the <%= trick, as if we could write this:
>
> ...
> for i in range(5):
> %>
>   <%= formatter % i %>
>
> Perhaps the issue is that I don't know how the <%= form is actually
> implemented in Python/ASP.  If it is just a textual substitition, where
the
> <%= ... %> is replaced with Response.Write(...) by simple textual
> substitution, then I understand why it is not possible.

<%= ... %> is indeed a simple macro for Response.Write(...).

The real problem is that ASP has no knowledge of indentation-sensitivity in
a language, and tends (as far as I can see) to pass blocks of code
individually to the interpreter.

It makes a lot more sense in Python to generate the whole HTML string before
emitting it with a single Response.Write() call. This anyway suits the
philosophy of modules such as HTMLGen rather better, and allows much simpler
exception handling than if you've already emitted half of your HTML!


regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From ragu at daylight.com  Thu Feb  7 17:18:53 2002
From: ragu at daylight.com (Ragu Bharadwaj)
Date: Thu, 07 Feb 2002 15:18:53 -0700
Subject: encapsulation, typing & other questions
References: <mailman.1013107933.30563.python-list@python.org>
Message-ID: <3C62FD4D.2B908BA9@daylight.com>

brueckd at tbye.com wrote:
> 
> On Thu, 7 Feb 2002, Ragu Bharadwaj wrote:
> 
> I don't know of any plans to do this, and the benefits are probably less
> than you'd imagine. This issue tends to come up as a theoretical problem
> much more often (if ever) than it does in practice.

Beg to disagree completely. I've seen too many folks encounter problems
to believe this.

-Ragu
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ragu.vcf
Type: text/x-vcard
Size: 422 bytes
Desc: Card for Ragu Bharadwaj
URL: <http://mail.python.org/pipermail/python-list/attachments/20020207/ae76288a/attachment.vcf>

From nhodgson at bigpond.net.au  Tue Feb  5 18:54:15 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Tue, 05 Feb 2002 23:54:15 GMT
Subject: distutils, C extensions, Borland compiler, wxPython
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk> <mailman.1012939329.5851.python-list@python.org> <C3Y78.38682$Ni2.219590@news-server.bigpond.net.au> <mailman.1012945695.5205.python-list@python.org>
Message-ID: <Hg_78.87$Zu6.190@news-server.bigpond.net.au>

Gerhard H?ring:
> So it would be possible to get rid of the python{major}{minor}.dll and
> still be able to dynamically load .pyd extension modules?

   I'd expect so. While DLL -> EXE linkage is possible, it should not be
done unless there is a net advantage. It is unexpected, linking becomes
messier and it may require more effort to support new compilers.

   Neil





From fredrik at pythonware.com  Tue Feb 26 17:22:11 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 26 Feb 2002 22:22:11 GMT
Subject: All-on, all-off button for checkbuttons?
References: <mailman.1014664062.9011.python-list@python.org>
Message-ID: <nUTe8.28500$l93.5471728@newsb.telia.net>

Nick Arnett wrote:
> The checkbuttons are created in __init__.  I'm thinking that there must be a
> way to filter or iterate through the objects in __init__, and for each one,
> if it's a button, select or deselect as appropriate.  I see how to get the
> namespace as a dictionary, but that hasn't helped.

a better approach is to loop over the children of the
master widget, and use isinstance.

something like this could work:

from Tkinter import *

root = Tk()

for x in range(10):
    var = IntVar()
    w = Checkbutton(root, text=str(x), variable=var)
    w.pack()
    w.var = var # keep a reference

def set_all(value):
    for w in root.children.values():
        if isinstance(w, Checkbutton):
            w.var.set(value)

w = Button(root, text="on", command=lambda: set_all(1))
w.pack()

w = Button(root, text="off", command=lambda: set_all(0))
w.pack()

mainloop()

</F>

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




From cmaloney at physics.ucsb.edu  Mon Feb 25 12:32:13 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: Mon, 25 Feb 2002 09:32:13 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org>
Message-ID: <3C7A751D.1080703@physics.ucsb.edu>

Hi Craig.

Since your site specifically lists g++ 2.95 and NOT 3.x, am I to infer
that there will be problems with the new ABI?  -- Or any problems that 
will, in practice or principle, cause conflicts with g++ 3.0?

*Holds breath... and crosses fingers...*

--Craig

rasmussn at lanl.gov wrote:

> On Wednesday, February 20, 2002, at 04:48 PM, David Abrahams wrote:
> 
>>> As I understand it, the SILOON project uses a commercial parser known as
>>> PDT from uoregon.  Have you, or has anyone else had any experience with
>>> the SILOON/PDT project?  On the face of it, my guess would be that you
>>> would be somewhat skeptical of their claims?
>>
>>
>> I haven't had any experience with it. I just found that site. My 
>> skepticism
>> just dropped by a factor of 8. If they are really using the EDG 
>> front-end as
>> they claim to, then I am sure the parser is excellent. I do have to 
>> wonder
>> who is paying for the FE. The issues of dealing with the semantics of 
>> what
>> they've parsed remain, but these guys at least seem to be doing the 
>> parsing
>> part right.
> 
> 
> It is my understanding that an educational or research license to the EDG
> front end is fairly easy to obtain.  The University of Oregon has obtained
> the right to distribute the EDG front-end as binaries and does so for 
> several
> platforms in the Program Database Toolkit (PDT).  So for anyone wanting
> interface level information from C++ or Fortran 90 source files, the PDT
> is an excellent tool.  We use it in several projects.
> 
> See http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/ and
> http://www.acl.lanl.gov/siloon/ for more information.
> 
> Craig
> 
> 




From AdrianLeu at kelseus.com  Wed Feb 20 04:18:31 2002
From: AdrianLeu at kelseus.com (Adrian Leu)
Date: 20 Feb 2002 01:18:31 -0800
Subject: Python crash
References: <993ae4ca.0202190754.6ccb79d2@posting.google.com> <u664tcv2n.fsf@python.net>
Message-ID: <993ae4ca.0202200118.30724909@posting.google.com>

I am using Python 2.0. Thanks.

Adrian.

Michael Hudson <mwh at python.net> wrote in message news:<u664tcv2n.fsf at python.net>...
> AdrianLeu at kelseus.com (Adrian Leu) writes:
> 
> > Hi !
> > I have written a script which I am calling from inside a C++
> > application using:
> > 
> >  module = PyImport_ImportModule( ... ) and 
> >  method = PyObject_CallMethod( ... ) 
> > 
> > methods. 
> 
> I think knowing which version of Python you have would be useful.
> 
> Cheers,
> M.


From robin at jessikat.fsnet.co.uk  Sat Feb  2 21:31:05 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 3 Feb 2002 02:31:05 +0000
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
 <313g3a.d22.ln@wombat.san-francisco.ca.us>
 <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk> <37d27e024b.tim@worthy.demon.co.uk>
 <PLPSljAQX$W8Ewpx@jessikat.fsnet.co.uk> <ac5588024b.tim@worthy.demon.co.uk>
 <ClZxOKANQBX8Ewis@jessikat.fsnet.co.uk>
Message-ID: <RugFcIApDKX8Ewx7@jessikat.fsnet.co.uk>

In article <ClZxOKANQBX8Ewis at jessikat.fsnet.co.uk>, Robin Becker <robin at jessikat.fsnet.co.uk>
writes
>In article <ac5588024b.tim at worthy.demon.co.uk>, Tim Howarth
><tim at worthy.demon.co.uk> writes
>.....
>>> but if it's easy with win32net what's the easy way?
>>
>>import win32net
>>
>>userinfo=win32net.NetUserGetInfo('','tim',2)
>>userinfo['password']='secret'
>>win32net.NetUserSetInfo('','tim',2,userinfo)
>>
>>
>>'' is server name, in this case server it is run on (no need for name)
>>2 is data level - what to get/set
>....
....
looking at win32net I see

win32net.NetUserChangePassword( server , username , oldPassword , newPassword )

which should do exactly what I want.
-- 
Robin Becker


From cmkleffner at gmx.de  Thu Feb 21 07:10:02 2002
From: cmkleffner at gmx.de (cmkl)
Date: 21 Feb 2002 04:10:02 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de>
Message-ID: <3b091a1c.0202210409.5beccbb7@posting.google.com>

One of the dilettantic NASA amateurs (David Bailey) once wrote the ultimate 
fast and accurate MP-package 'mpfun' in Fortran 77 - later F90.
A C++ implementation is available at:
http://www.cs.unc.edu/Research/HARPOON/mpfun++/
I havn't tested this C++ version and I don't think there is a a python 
wrapper available (should not be to hard ??) but it may worth trying.

Carl

"Karl M. Syring" <syring at email.com> wrote in message > 
> But this is really dilettantic work and NASA today is an organization run by
> amateurs. If you look at real computational geometry programs like CGAL, you
> can see, how it is done.
> Anyway, it is not necessary to appeal to authorities to get a grasp of the
> problem.: Imagine, you have a polygon in the plane and you do a series of
> rotations and translations on it. Now, you want to know whether the position
> is exactly identical to another polygon. If you do this with floating point,
> you will never get an exact result, while in rational coordinates you always
> will.
> 
> is-the-loss-of-knowledge-a-form-of-postmodernism?
> Karl M. Syring


From jjl at pobox.com  Sat Feb  2 17:06:41 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 22:06:41 +0000
Subject: curses module
In-Reply-To: <kjho5uc7ut6tt12898b62sthsq5hdofvqu@4ax.com>
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012671184.6510.python-list@python.org> <kjho5uc7ut6tt12898b62sthsq5hdofvqu@4ax.com>
Message-ID: <Pine.SOL.4.30.0202022205420.597-100000@mimosa.csv.warwick.ac.uk>

On Sat, 2 Feb 2002, Lawrence Oluyede wrote:

> On Sat, 2 Feb 2002 20:31:16 +0300, Oleg Broytmann <phd at phd.pp.ru>
> wrote:
>
> > It *is* in the distribution. It is not a single module, but a package.
>
> >>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named curses
[...]

You probably didn't compile it in (or the setup.py didn't do it for you).
Check Modules/Setup.


John



From bevank at dstc.edu.au  Mon Feb 25 00:26:15 2002
From: bevank at dstc.edu.au (Bevan Koopman)
Date: Mon, 25 Feb 2002 15:26:15 +1000
Subject: HTTP_COOKIE variable not set
Message-ID: <3C79CAF7.97EA935F@dstc.edu.au>

Hi there,

When I run the script below the 'HTTP_COOKIE' environment variable is
not set, i.e. "HTTP_COOKIE" does not appear in the scripts output:

#####################
import cgi
import os
import Cookie

print "Content-Type:text/html\n\n"

cookie = Cookie.SimpleCookie()
cookie["user"] = the_user

print os.environ
#####################

Hence when I run the following script the key is not found

#####################
import cgi
import os
import Cookie

c = Cookie.SimpleCookie()
if(os.environ.has_key('HTTP_COOKIE')):
	c.load(os.environ['HTTP_COOKIE'])
	print c["user"].value
else:
	print "no user"
#####################

How do I correctly create the cookie so I can retrieve in the second
script.

Thanks,
-- 
|  _     _
|\|_\  //_\|\ | bevan koopman
|/|_ \/ | || \| bevan at dstc.edu.au
Distributed Systems Technology Centre
http://www.dstc.edu.au


From fperez528 at yahoo.com  Tue Feb 26 10:48:00 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 26 Feb 2002 15:48:00 +0000
Subject: [ANN] JPE 2.0 The Java-python Extension
References: <mailman.1014758097.11472.python-list@python.org>
Message-ID: <a5h3eh$2df$1@peabody.colorado.edu>

Frederic Giacometti wrote:

> 
> -?JPE?is?distributed?in?source?form?from?CVS.?Substantial?efforts?have
> been invested to provide a config and build environment that makes it
> straightforward to configure, build from source, and test
> multi-component, multi-platform software systems. Hopefully, that can be
> one step ahead from the current autoconf or distutil systems.
> Easily extensible to support arbitrary software components and tools,
> libplus currently supports C, C++, Python, Java, and swig wrappers.
> Libplus generates configuration-specific makefiles that chain software
> component dependencies together, and provides runtime software/tool
> configuration information.
> Litteratly, this is an 'open-source' project :))

I'm assuming this paragraph is in reference to the libplus part of your 
system. I happen to be interested, and grabbed the current cvs, but 
unfortunately there's not _one_ piece of documentation that comes with it. I 
realize you're still in development, but learning the entire structure of 
your code just to see if it's useful to me is a bit much, I'm sorry to say :)

Would it be possible for you to include a simple README in libplus which 
outlines its usefulness as a 

> config and build environment that makes it
> straightforward to configure, build from source, and test
> multi-component, multi-platform software systems ?

That's exactly one area with which the scipy project has struggled and it 
would be great if your project solved the issue. But I honestly can't really 
say, as there is zero documentation included, and most modules have very 
terse docstrings.

I'm not trying to disparage your work, but unfortunately without some 
documentation it's just very difficult for an outsider to even judge whether 
it would be useful for me or not.

Best regards,

f. 


From jeff at ccvcorp.com  Tue Feb 12 14:22:28 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 12 Feb 2002 11:22:28 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de> <mailman.1013521581.26252.python-list@python.org>
Message-ID: <3C696B74.AF696405@ccvcorp.com>


Chris Gonnerman wrote:

> > Looks like it is missing a DLL, probably one of MFC's (If it is on a bare
> > bones Win95, there are more interesting modes of failure). Can't you
> > inspect, what your OCX is trying to import?
>
> Ur.  What's the Win32 equivalent of ldd (dumps linked dll's)?

There's a program called Dependency Walker that does this.  I don't know a URL
for it offhand, but I'm sure you can google for it.  (It also comes with Visual
Studio 6, but I know I downloaded it independently.)

Jeff Shannon
Technician/Programmer
Credit International




From cmueller at onebox.com  Thu Feb 21 19:32:31 2002
From: cmueller at onebox.com (rockmuelle)
Date: 21 Feb 2002 16:32:31 -0800
Subject: wxPython MDI question
Message-ID: <aaa059f1.0202211632.1f9e9d98@posting.google.com>

I am trying to create an MDI application that has a list control on
the left side of the main MDI frame and the MDI client area on the
right side, similar to the way Visual Studio has other windows
surrounding its client area.

I've tried creating the list control and then adding it and the client
area to a sizer, but the client area always takes up the whole main
frame. The client area and the list control are situated at (0, 0), as
if there was no sizer.  If I add more controls to the sizer, the
non-client area windows get layed out correctly, but the sizer and
client area still overlap.

Does anyone know how to fix this?

I'm using wxPython 2.3.1 with Python 2.1.1 on Windows 2K.  Code
demonstrating what I've tried is appended to this message.

Thanks.

-Chris


from wxPython.wx import *

class MyMDIFrame(wxMDIParentFrame):
  """
  Main frame that has list control and MDI client area.
  """

  def __init__(self, parent, id=-1, title='MDI Example'):
    wxMDIParentFrame.__init__(self, parent, id, title=title)

    sizer = wxBoxSizer(wxHORIZONTAL)
    sizer.Add(wxListCtrl(self, -1))
    sizer.Add(wxListCtrl(self, -1))  # add another list
    sizer.Add(self.GetClientWindow())
    sizer.Layout()

    self.SetSizer(sizer)
    self.Fit()
    
    return


if __name__=="__main__":

  class App(wxApp):

    def OnInit(self):
      frame = MyMDIFrame(NULL)
      frame.Show(true)
      self.SetTopWindow(frame)

      return true

  app = App(0)
  app.MainLoop()


From mcherm at destiny.com  Fri Feb  8 10:19:49 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Fri, 08 Feb 2002 10:19:49 -0500
Subject: To all Higher Level....A True Beginner
Message-ID: <3C63EC95.5080408@destiny.com>

> ... Though it must not be 
> obvoius who's using usenet from the email system... over here, mail-list 
> people's responses don't seem to thread the same way as normal messages.


Yeah... I notice that too. I wish I could access via usenet instead of 
having to use the email interface and a viewer that I coded up myself 
(in python of course), but unfortunately, I don't have access to a 
usenet feed of c.l.py :-(.

-- Michael Chermside






From Scirious at Scirious.com  Thu Feb 14 17:02:06 2002
From: Scirious at Scirious.com (Scirious)
Date: Thu, 14 Feb 2002 22:02:06 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <Xns91B5CC2A8926BScirious@140.99.99.131>

Sorry, please consider avout = about.
Scirious.


From ajeru at vknn.org  Tue Feb 19 19:37:08 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 16:37:08 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <slrna758ng.d1u.quinn@yak.ugcs.caltech.edu>
Message-ID: <24c39b2c.0202191637.34508374@posting.google.com>

quinn at yak.ugcs.caltech.edu (Quinn Dunkan) wrote in message news:<slrna758ng.d1u.quinn at yak.ugcs.caltech.edu>...
> >function like def computeSalary(person)... I don't know at first sight
> >if person is the id of person, a person object and if it's an object
> >what its exact class in a Person class hierarchy is. I have to find
> >the place where the constructor is called to find that out. That's not
> 
> Or you can just look and see how the 'person' is used in the function.  If the
> first statement is a 'o = db.by_id[person]' then you know it's an id.

No I can't do that if there's an inheritance hierarchy involved and
the function uses only methods that are already included in a base
class. If I want to add some piece of code to the function that uses
the more specialized properties of the parameter class, then I have to
go look at the constructor. There's no other way in this case.

> If it's an object, why should you care at what exact position it lives in
> some class hierarchy?  Isn't the point of dynamic dispatch that you don't
> have to know?

Sometimes I don't want to know but sometimes I have to know. That
depends on if I want to use it polymorphically or if I want to use the
specific features of some class in the hierarchy.

> 
> >And the final argument I have for static type checking is that it
> >enables method dispatching based on parameter types. In a statically
> >typed language you can create two methods that have the same name but
> >differ on the parameter types. The correct method will be called for
> 
> Yeah, python has this too:
> 
> class A:
>     def m(self): print 'm'
> class B(A):
>     def m(self): print 'bm'
> o = B()
> o.m() #  B.m is called
> p = A()
> p.m() # A.m is called
> 

I didn't mean inheritance based dispatching.

> It's only of the first argument, but that's what mono-dispatch means.
> Static typing having anything to do with multi-dispatch is incorrect,
> though.  The best-known object system with multi-dispatch is CLOS, which
> is certainly dynamic.  In fact, I don't know any static languages that do
> what you want (I'm not counting c++-ish "overloading" because it doesn't
> dispatch based on the type of the variable, but the type of the declaration
> (which, since you're forced to write out the declaration, you're not
> gaining any flexibility or having to type less---so that's the point?)).
> 

In Python, I can't have two functions with the same name that are
selected on the basis of their parameter types. I don't know CLOS very
well but if you can't declare the function parameters, I wouldn't know
how a dynamic type system at runtime would choose to which of the
existing functions it should dispatch. How do you define multimethods
in CLOS?

> When you don't have multi-methods you get by with single dispatch.
> 
> >you depending on the type of the argument you pass in your call. That
> >makes for a quite flexible way of extending a program. You can just
> >add another method with the same name and another type to handle a
> >special case without touching the existing methods.
> 
> Yep, it's called inheritance, and OO people go on and on about it :)

No. I think inheritance based dispatching is rather limited. It makes
systems inflexible because you have to have just one type hierarchy
and there's no complex system that can be expressed in a single class
tree. That's something the philosophers recognised a long time ago.

> 
> >I think type checking could be added as an optional feature, like in
> >Dylan, without hurting the character of Python.
> 
> Oh yes, and dylan, another dynamically typed multi-method language.

Yes I like some features in Dylan including multimethods but Dylan
isn't in very broad use and there are many features I like better in
Python. Why not add  the good features of other languages to Python if
they don't break anything?

> 
> >I'm not, however, completely satisfied with static type checking as
> >well because I don't see why the compiler forces me to have a
> >parameter comply with a static interface that has 10 methods when the
> >method that uses the parameter acutally only uses one of those
> >methods. I want the compiler to check if the parameter type I'm
> >passing has what the method needs but nothing more.
> 
> Yeah, that sort of thing is nice.  I want more though... I want
> to check if the parameter has what the symbols method needs, *and* that
> those symbols do what the method wants.  For that, you need contracts
> or something, which work at run time.  As long as you're working at
> run time, you'll have to test it, so leave the contracts out and see
> if it throws an error :)

Ok, I get what your opinion is. Check everything at runtime. I
disagree :-)

Alexander


From jorgelr at stud.ntnu.no  Sat Feb 16 18:11:43 2002
From: jorgelr at stud.ntnu.no (Jorg Rødsjø)
Date: Sun, 17 Feb 2002 00:11:43 +0100
Subject: win32gui
Message-ID: <a4monr$hmg$1@tyfon.itea.ntnu.no>

Hi

I am currently involved in a project at my university (www.ntnu.no),
developing a shared workspace system running on ipaqs and stationary
computers.

We are at the moment looking at different languages, and we have tried
python on the ipaq running WinCE(2002 to be presice.) Which seems to work
ok.
Our current problem is the lack of gui. We see win32gui is there, but the
documentation seems to be skimpy at best. And our searches for more have
been to no avail.

We tried to use anygui, which should be able to run using win23gui, but we
are unable to import it without errors.

Can anyone give us, or point us to some resources when it comes to using
win32gui?

Any and all help would be greatly appreciated, as we are hoping to avoid
using embedded C++ or VB.

-jorg





From bp at sam.localdomain  Wed Feb 27 22:35:22 2002
From: bp at sam.localdomain (Bob Parnes)
Date: Thu, 28 Feb 2002 03:35:22 -0000
Subject: Mod-python problem
Message-ID: <slrna7rdq5.2vf.bp@sam.localdomain>

Hello,

I am having difficulty with the FieldStorage class in 
the mod-python module for apache. I cannot get 
util.FieldStorage() to return anything. If someone could 
point me to an example, I would be grateful.

Just in case I am missing something obvious, the following 
is my test code, for a linux system:
---------------------------------------------

#!/usr/bin/env python

from mod_python import util

class clTest:
    def cltest(self, req, formDict):
        res = '<html><head></head><body>'
        res += '<table align=center>'
        res += '<TR><TH colspan=2>Welcome to the \
            mod-python class test'
        res += '<TR><TD colspan=2>keys: %s' \
		    % formDict.keys()
        res += '<form method=post>'
        res += '<TR><TD>Name <TD><input type=text \
            name="name">'
        res += '<TR><TD>Phone <TD><input type=text \
            name="phone">'
        res += '<TR><TD colspan=2 align=center> \
            <input type=submit name=submit>'
        res += '</form></table></body></html>'
        return res

def main(req):
    formDict = util.FieldStorage(req)
    cl = clTest()
    return cl.cltest(req, formDict)
--------------------------------------------

The form displays as I expect it to, but when I enter 
data and press the submit button, formDict remains 
empty

In contrast, the following code in the cgi-bin directory 
works fine:
---------------------------------------------

#!/usr/bin/env python

import cgi

class clTest:
    def cltest(self, formDict):
        res = 'Content-Type: text/html\n\n'
        res += '<html><head></head><body>'
        res += '<table align=center>'
        res += '<TR><TH colspan=2>Welcome to the \
            mod-python class test'
        res += '<TR><TD colspan=2>keys: %s' \
		    % formDict.keys()
        res += '<form method=post>'
        res += '<TR><TD>Name <TD><input type=text \
            name="name">'
        res += '<TR><TD>Phone <TD><input type=text \
            name="phone">'
        res += '<TR><TD colspan=2 align=center> \
            <input type=submit name=submit>'
        res += '</form></table></body></html>'
        return res

def main():
    formDict = cgi.FieldStorage()
    cl = clTest()
    print cl.cltest(formDict)

if __name__ == '__main__':
    main()
--------------------------------------------	

Thanks in advance for any help or references.

Bob Parnes

-- 
Bob Parnes
rparnes at megalink.net


From tdelaney at avaya.com  Mon Feb 11 18:29:15 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 12 Feb 2002 10:29:15 +1100
Subject: New PEP (281): Loop Counter Iteration with range and xrange
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A178@natasha.auslabs.avaya.com>

> From: mlh at vier.idi.ntnu.no [mailto:mlh at vier.idi.ntnu.no]
> 
> I've submitted a draft PEP describing the idea of letting range and
> xrange accept sequences (or possibly, for xrange, any iterable) and
> return/yield one index per element. It's closely related to PEP 212,
> but instead of proposing new functions/syntax, it proposes extending
> the capabilities of existing functions.
> 
> The PEP is available at:
> 
>   http://www.python.org/peps/pep-0281.html

I've proposed this in the past and was told by the timbot IIRC that it
wouldn't get off the ground.

Tim Delaney



From jeff at ccvcorp.com  Thu Feb 14 14:40:08 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 11:40:08 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net>
Message-ID: <3C6C1298.505C578B@ccvcorp.com>


Marcus Stojek wrote:

> Hi,
>
> I am still struggling with tuple-lists.
>
> How can I sort a list of tuples (all tuples have 4 float items)
> by the second (or third or n-th) tuple item. And I have to
> do it fast.

You can supply a custom comparison function to sort(), such as
this:

def mycmp(first, second):
    return cmp( first[2], second[2] )

mylist.sort(mycmp)

But this tends to be pretty slow.  For large lists, it's usually
considered that the "decorate-sort-undecorate" pattern is the most
effective way.  Something like this:

def SortOnItem(mylist, index):
    templist = [ (line[index], line) for line in mylist ]
    templist.sort()
    return [ line[1:] for line in templist ]

What this does is build a separate list containing a tuple of the
element that you want to sort on, and the entire line, sorts that
list (by the first element, of course), and then strips that first
element off.  This can use a lot of memory (you're creating two
copies of your original list, but they are only references, not
deep copies), but until you hit memory constraints this is likely
to be faster than a custom cmp() function.  (The custom cmp() makes
an extra function call for *each* line of the list, and function
calls are relatively expensive in Python.)

(Of course, the list comprehensions require Python 2.x; if you're
stuck on 1.5.2, the equivalent can be done with map/filter/lambda.)

Jeff Shannon
Technician/Programmer
Credit International




From paul at prescod.net  Sat Feb 16 13:06:07 2002
From: paul at prescod.net (Paul Prescod)
Date: Sat, 16 Feb 2002 10:06:07 -0800
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <m34rkhbvw3.fsf@mira.informatik.hu-berlin.de> <3C6E943B.BC7CCF9@earthlink.net>
Message-ID: <3C6E9F8F.AAE42150@prescod.net>

Hans Nowak wrote:
> 
>...
> The multiline XML string is in roughly the same format as the
> raw XML the server returns. That is, it does contain newlines
> after 'one', 'two', and 'three' (assume these were in the
> original multiline string). As soon as the XML is parsed,
> these newlines disappear. Maybe this is normal and to be
> expected when using xml.dom.minidom.parseString; I don't know.

That's what the XML spec says to do with attribute values. 

http://www.w3.org/TR/REC-xml#AVNormalize

If you don't control the server then I think you're kind of out of luck!

 Paul Prescod



From syring at email.com  Tue Feb 12 08:02:11 2002
From: syring at email.com (Karl M. Syring)
Date: Tue, 12 Feb 2002 14:02:11 +0100
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org>
Message-ID: <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> schrieb
> I am developing a headache messing with an OCX I picked up from the
> Internet.  It seems that my forays into ActiveX never pan out; I should
> stick with Linux.
>
> The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> have to be such a pain to find?)  I can instantiate it:
>
> >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> >>>
>
> (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
>
> >>> o.AboutBox()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py",
line
> 29, in AboutBox
>     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
> com_error: (-2147418113, 'Unexpected failure', None, None)
> >>>

Looks like it is missing a DLL, probably one of MFC's (If it is on a bare
bones Win95, there are more interesting modes of failure). Can't you
inspect, what your OCX is trying to import?

Karl M. Syring




From lsmithso at NOhare.SPAM.demon.co.uk  Fri Feb  8 07:35:00 2002
From: lsmithso at NOhare.SPAM.demon.co.uk (Les Smithson)
Date: 08 Feb 2002 12:35:00 +0000
Subject: threads killing other threads
References: <1013105970.135322@cswreg.cos.agilent.com>
Message-ID: <a40h4g$kl5$1@newsreader.mailgate.org>

>>>>> "Greg" == Greg Weeks <weeks at vitus.scs.agilent.com> writes:

    Greg> I've occasionally wanted thread objects to have an exit()
    Greg> method, so that one thread could interrrupt and kill
    Greg> another.  This doesn't seem to be available, perhaps because
    Greg> POSIX threads don't support it.

    Greg> Is that correct?

    Greg> Is there some reason why threads killing threads would be
    Greg> undesirable or hard to implement?


Posix defines a pthread_cancel that lets one thread cancel another
thread. From the man page:

       Cancellation is the mechanism by which a thread can termi?
       nate  the  execution  of another thread. More precisely, a
       thread can send a cancellation request to another  thread.
       Depending  on  its  settings,  the  target thread can then
       either ignore the request, honor it immediately, or  defer
       it till it reaches a cancellation point.


I'm sure there's a good reason for Python not implementing
this. Anyone know what it is?





From aleax at aleax.it  Fri Feb  1 03:29:42 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 08:29:42 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com>
Message-ID: <Wls68.14178$om6.357057@news1.tin.it>

Peter Milliken wrote:
        ...
> Actually, if you are *good* at what you do, then code review, whilst still
> beneficial, is not nearly *that* beneficial. Things like code review (IMO
> :-)) is there for the 25-50% of programmers who just plain shouldn't be in
> the industry - what they generate is just absolute garbage - code review
> will hopefully stop the code from getting into Unit test and beyond the
> point of redemption :-).

I disagree.  Code review, or even better the "continuous ongoing code
review" that's such a substantial part of the benefits of pair programming,
is in my experience just as beneficial with the best coders: without it, 
besides bugs, it's far too much a temptation to pepper the code with
clever tricks, individual style quirks, unexamined (subconscious) 
assumptions about inputs and environmental conditions, and the like.

The code coming out of a good review, or a good-synergy pair, is
thus of much higher quality and more valuable than the results of a lone
coding session of even a very good coder.


> Agreed with the "Art" - many programmers give one the impression that they
> perceive themselves as a Prima Donna of the software industry :-) Ask
> someone to look at their code (as in code review) and they look offended -
> like you'd just asked for their first born or something :-) A severe lack
> of comprehension is the general response - "You want to look at *my* code?
> What one earth for! It's perfect (by definition)" :-).

I fully agree with you.  But that implies my stance in the above paragraphs,
not yours:-).  Actually the root of the problem may be in 'my' vs 'yours',
the "*my*" in your quote being correctly emphasized:-).


Alex



From tim at vegeta.ath.cx  Thu Feb 28 17:58:08 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Thu, 28 Feb 2002 22:58:08 GMT
Subject: Can use python to play MP3 file?
References: <a5llkf$8a75f$1@ID-12869.news.dfncis.de>
Message-ID: <slrna7tdgp.113.tim@vegeta.ath.cx>

Ricky <store_li at sina.com> graced us by uttering:
> I hear that PHP can online play MP3 file. Is there any method that
> can make me play MP3 by using Python CGI?

If you mean an MP3 stream server written in Python, try edna
<http://edna.sf.net/>.  It will serve MP3s in a stream to the client
computer. (NB: it's up to the client to "play" the MP3, often through
MIME associations in the browser.)

If you mean you want an MP3 player written in Python, well, I don't
suggest it. Processing MP3 data is very processor intensive and is
often thought best accomplished with a compiled C program.  However,
cymbaline <http://www.silmarill.org/cymbaline/> is a Python frontend
to the (very pervasive) mpg123 MP3 player.

You don't mention which OS, so I should mention, this is from a linux
perspective. edna might very well work on Windows.

HTH
Tim Hammerquist
-- 
$ echo pizza > /dev/oven


From gh_pythonlist at gmx.de  Mon Feb 25 01:01:39 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Mon, 25 Feb 2002 07:01:39 +0100
Subject: Correction: Re: string substitutions
In-Reply-To: <20020224222228.B16970@evertek.net>
References: <mailman.1014504639.21135.python-list@python.org> <c76ff6fc.0202232228.201f829e@posting.google.com> <mailman.1014583431.3100.python-list@python.org> <c76ff6fc.0202242012.48b17a85@posting.google.com> <20020224222228.B16970@evertek.net>
Message-ID: <20020225060138.GA5411@lilith.hqd-internal>

Le 24/02/02 ? 22:22, Mike Dean ?crivit:
> * John Machin <sjmachin at lexicon.net> [2002-24-02 20:12]:
> > <snipped text about RE engine handling \n natively>
> > This important fact is often not documented well or at all.
> 
> Is there an appropriate place to request that this be added to the next
> incarnation of the documentation?

http://sourceforge.net/tracker/?group_id=5470&atid=105470

If you haven't got  one, please get an account at Sourceforge before
submitting a bug report ("New user via SSL") - anonymous submissions are
exhausting for the developers.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 3.0 ?C      Wind: 8.5 m/s



From littlejohn.75 at free.fr  Fri Feb  1 16:39:31 2002
From: littlejohn.75 at free.fr (Francois Petitjean)
Date: Fri, 1 Feb 2002 22:39:31 +0100
Subject: confused by isinstance()
References: <3c5717bb$0$23518$626a54ce@news.free.fr> <a3b9ib$61c$1@panix2.panix.com> <3c5a3352$0$4604$626a54ce@news.free.fr> <Xns91A85C660FC3Bduncanrcpcouk@127.0.0.1>
Message-ID: <3c5b0a9f$0$23532$626a54ce@news.free.fr>

>> (1) It is trivial as it has :
>> only one module
>Two modules, Foo and __main__
yes, technically, but superficially a newbie can see only one.
>> only one class
>Two classes, one in each module. Foo.Foo and __main__.Foo
as a consequence of above

>When you run a script, it executes the code in the context of the module
>called __main__. When you import a module it has a separate context even if
>the script and the module came out of the same source file. Importing a
>module executes each line in the file, even if you already ran most of them
>as a script.

I suppose that this context corresponds to a namespace, which is implemented
internally by a dict.
Where in the documentation is explained what means "Importing a module
executes each line in the file". The lines are syntaxically checked, sure,
but most of the work is not done at this stage:
def curve(x):
    return return 10.0*math.sin(x)/x
conversion to byte code.

>Every time you execute a class statement, even if the class name is the
>same, you create a new class.
This is the crux of the problem, but after the first import, it seems that
successive import of a module give the same class, due to the sys.module
caching. Right?

So, if I want to be on the safe side and to test a module Foo as it appears
from outside, I can begin the __main__ part with import Foo. In this case,
instanciation is of the form x = Foo.Foo(), all is good except for the
doubling of the memory to store the class.

Thank you for your insightful post.




From maxm at mxm.dk  Thu Feb 14 04:14:54 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 14 Feb 2002 09:14:54 +0000
Subject: Help! IIS and Python
References: <8f8ffe67.0202131245.375fe55d@posting.google.com> <eqlm6ugbbtsu4i1d61pek0c4g49as7j119@4ax.com>
Message-ID: <3C6B800E.7060901@mxm.dk>

Tim Roberts wrote:

 > Since we're on the subject of Python in ASP programming, how do you 
use the
 > <%=xxx%> trick with indentation?  This fails with an indentation 
error, for
 > example:
 >
 > <%
 > for i in range(5):
 > %>
 >   <%="aaa%d" % i %>

Generally I would say "don't do it". every codeblock is seen as a new 
block, so your code looks to Python like:

for i in range(5):
Response.Write("aaa%d" % i)

Perhaps this would work, I don't know:

<%
for i in range(5):
%>
<%=    "aaa%d" % i %>

You need to put all your logic before the html code, and then insert 
simple strings. This approach is also a faster for lots of content as 
joining a long list is a LOT faster than concatenating many strings.

Like:

<%
a3Nums = []
for i in range(5):
     a3Nums.append("aaa%d" % i)
a3Nums = '\n'.join(a3Nums)
%>

<%= a3Nums%>

It also makes your code much more readable. Mixing logic and 
presentation is really only a good idea for very simple pages.

regards Max M



From eric.brunel at pragmadev.com  Tue Feb  5 04:46:08 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 5 Feb 2002 09:46:08 +0000
Subject: [Fwd: directly executing a python script in Unix]
References: <3C5F1638.1D9B527C@ix.netcom.com> <a3ncv6$fgs$1@peabody.colorado.edu>
Message-ID: <a3o5tt$1o3s$2@norfair.nerim.net>

> Here's my python fix for that moronic problem (moronic of Dos designers,
> who thought that two characters per line ending were a good idea when
> there already was a perfectly viable convention to follow, namely that of
> Unix. Perhaps they were following a VMS convention or somesuch, I don't
> know).

Nope: VMS accepted every known text format in the whole universe, including 
the most funky ones... They just over-used their creativity...



From matthias_j at lycos.com  Tue Feb 26 01:12:08 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 25 Feb 2002 22:12:08 -0800
Subject: Trouble with getting remote Computer Name
References: <d7d5ebdf.0202251024.236b2c1b@posting.google.com> <3c7a75ea.16613828@news.t-online.de>
Message-ID: <d7d5ebdf.0202252212.9e77c32@posting.google.com>

gerson.kurz at t-online.de (Gerson Kurz) wrote in message news:<3c7a75ea.16613828 at news.t-online.de>...

> Example:
> 
> >>> socket.gethostbyname("www.google.com")
>  '216.239.37.101'
> >>> socket.gethostbyaddr('216.239.37.101')[0]
> 'www.google.com'


Thank's a lot this does it for me.


From tismer at tismer.com  Fri Feb 15 03:27:03 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 15 Feb 2002 09:27:03 +0100
Subject: Is Stackless Python DEAD?
References: <184fbd02.0110262039.74799675@posting.google.com> <j4d72xasx0.fsf@informatik.hu-berlin.de> <uady09pkm.fsf@python.net> <mailman.1009806401.11232.python-list@python.org> <110A0579A87BA846.8CEE70EF24F483BD.740CFC77F4D54FA3@lp.airnews.net>
Message-ID: <3C6CC657.40405@tismer.com>

Cameron Laird wrote:

 > In article <mailman.1009806401.11232.python-list at python.org>,
 > Christian Tismer  <tismer at tismer.com> wrote:
 >
 >>Well, it is a little late to answer this, but...
 >>
 >>Michael Hudson wrote:
 >>
 >>
 >>>Martin von Loewis <loewis at informatik.hu-berlin.de> writes:
 >>>
 >>...
 >>
 >>...
 >>
 >>
 >>>>- It adds a field nesting_level to the thread state, without
 >>>> ever checking its value (it just counts the nesting level)
 >>>>
 >>>>
 >>>I imagine this was just for bragging with :)
 >>>
 >>
 >>Yes. I needed a way to track when and why there are still
 >>recursions. At sme pahse I also had the idea to drive
 >>decisions on this when I'm allowed to switch things, but
 >>later I learned that this is the wrong way.
 >>
 > Along with bragging rights, one of the not-obvious-but-
 > easy-to-explain reasons to introspect on recursion depth
 > is as a defense against denial-of-service (or more un-
 > intentional) security hazards.  If you're executing
 > something dynamic enough to give a user a wedge into the
 > execution stack, it's prudent, however crude it might
 > seem, to impose resource limits.  One engineering discovery
 > is that recursion depth can be a useful resource to limit.


The current working copy of Stackless is already running in
a total of 2 KB of C stack on Intel machines.
This limits the area of memory to be monitored for
attacks reasonably.

 >>I'm at a redesign for Stackless 2.2. I hope to make it simpler,
 >>split apart Stackless and optimization, and continuations are
 >>no longer my primary target, but built-in microthreads.
 >>
 > Cool!


The new solution is cool and uncool at the same time.
All my difficult changes to the core have been abandoned.
All magic about Stackless' implementation is gone (sigh).
It is now orthogonal to the Python engine. The interaction
is down to the bare minimum.
The implementation has become ridiculously small.
Portability has changes substantially.
While Stackless is now much less dependant from
Python, it is now very much platform dependant, since
it involves a couple of assembly instructions.
I think this is an advantage, since Python develops
much more quickly than hardware platforms.

watch out for the soon-to-come release candidate - ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
       where do you want to jump today?   http://www.stackless.com/






From warlock at eskimo.com  Thu Feb 21 20:59:14 2002
From: warlock at eskimo.com (Jim Richardson)
Date: Thu, 21 Feb 2002 17:59:14 -0800
Subject: Pythonic bit twiddling
Message-ID: <il845a.7qa.ln@127.0.0.1>

Pythonic bit shuffling question

I have an application that does a calculation on a single byte of data,
after breaking it into an upper and lower nibble of four bits. How can I
do this in python? To clarify.
Take a single byte, break off the upper 4 bits, and do some stuff, then
do similar stuff with the lower nibble, and then munge them back
together to create a byte again. This is for a checksum to inteface with
a bit of embedded stuff. I know how I can do this in C, but as that
would be the sole bit of copiled code in the app, I'd prefer to avoid
it if possible. 


Thanks all

-- 
Jim Richardson
	Anarchist, pagan and proud of it
www.eskimo.com/~warlock
    "New MS-Wheel, now with 8 sides for a smoother ride!"


From SKuzminski at FairIsaac.com  Tue Feb 26 13:32:53 2002
From: SKuzminski at FairIsaac.com (Stefan Kuzminski)
Date: Tue, 26 Feb 2002 10:32:53 -0800
Subject: advocacy help ( Python vs VB )
Message-ID: <FF1A87368802D41199CD009027DE5C7904715632@srfmsg01>

I suppose the fact that I get 20 emails in response to this question is a
sign, I know you couldn't get 20 positive third party comparisions of VB to
Python.  I've tried to sell Python in several organizations over the last
few years with varying degrees of success.  The *key* sticking point ( other
than general fear of the unknown ) is enevitably the 'well I can just go an
hire VB programmers but nobody knows this Python thing'.  Which of course is
absurd since the python learning curve is so small.  But the managers aren't
in any position to relate since all they control are resources with resumes
that say VB, C++ or Java...  ~sigh~  So my strategy so far has to been to
build the thing without ever mentioning Python until it's done..

Stefan  



-----Original Message-----
From: Matthew Sherborne [mailto:miracle at paradise.net.nz]
Sent: Monday, February 25, 2002 5:21 PM
To: Stefan Kuzminski; python-list at python.org;
activepython at listserv.ActiveState.com; python-help at python.org
Subject: Re: advocacy help ( Python vs VB )


Google uses python to script their cluster(s).

http://www.google.com/jobs/openings.html

(Search for python on that page, I counted 10)

Also check out these URLs:

http://home.rmi.net/~lutz/advocacy.html

http://py-howto.sourceforge.net/advocacy/advocacy.html

GBU
Matthew Sherborne

----- Original Message -----
From: "Stefan Kuzminski" <SKuzminski at FairIsaac.com>
To: <python-list at python.org>; <activepython at listserv.ActiveState.com>;
<python-help at python.org>
Sent: Tuesday, February 26, 2002 1:25 PM
Subject: advocacy help ( Python vs VB )


> I need to convince my co workers and erstwhile managers to build a
> production system in Python.  They're coming from a VB world and are
fearful
> to detach from the Redmond mothership.  I need hard facts, not warm and
> fuzzies about syntax but names of big companies using Python and things VB
> just *cannot* do ( threads, regular expressions.. ? ).
>
> Thanks,
> Stefan
> _______________________________________________
> ActivePython mailing list
> ActivePython at listserv.ActiveState.com
> To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
>



From emile at fenx.com  Tue Feb 26 08:58:37 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 05:58:37 -0800
Subject: Calling baseclass method from a derivated class
References: <slrna7n2u6.f53.patois@ns.sicfa.org>
Message-ID: <a5g4s5$70ool$1@ID-11957.news.dfncis.de>

"Yannick Patois" 
> class toto2(toto1):
>    def titi(self):
>         print "titi2"

Do you mean like this?

class toto2(toto1):
    def titi(self):
        toto1.titi(self)
        print "titi2"


-- 

Emile van Sebille
emile at fenx.com

---------



From sholden at holdenweb.com  Fri Feb  1 10:54:44 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 10:54:44 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com>	<20020130073810.382$1b@news.newsreader.com>	<b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <b34d3835.0202010038.6a907bff@posting.google.com>
Message-ID: <CVy68.104504$8b.3960936@atlpnn01.usenetserver.com>

"Rony" <rony.steelandt at bucodi.com> wrote ...

> I'm sorry but i definitifly don't agree with that, and here i'm
> speaking trough personnal experience !
> Somewhere beginning 90'ths my development team came to me saying it is
> time to abandon language X, that we were using, because now you have
> languge Y who has all the possible benefits; better librarys, faster
> development, blablabla...
> And i trusted my chief development engineer and said yes.
> To make a long story very short, the end of it was that the new
> language din't have all the necessary function we needed, they never
> found a solution (even not hired outside experts on that language) and
> we trowed away one man year of development. And that is reality...
> I know all the arguments, like your chief engineer didn't do his job
> and how comes that it took a year and .... but since i'm the owner of
> the company the problem  was mine at the end. So i'm convinced that
> managment has to be involved in such decissions.
>
Rony, I feel you are making the mistake of generalizing from your own
experience. This is perfectly understandable, and something we all do.
However, my own feeling is you should solve this problem not by becoming
better technically, but by becoming a better recruiter.

Ultimately your mistake ws not in trusting your chief engineer, but in
hiring an untrustworthy chief engineer.

Fortunately it seems that you do have a natural aptitude for technical
tasks, and so you are avoiding the problem in future by developing your
technical knowledge and, effectively, acting as chief technical officer as
well as manager.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From akuchlin at ute.mems-exchange.org  Sat Feb  9 10:55:21 2002
From: akuchlin at ute.mems-exchange.org (A.M. Kuchling)
Date: 09 Feb 2002 15:55:21 GMT
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org> <slrna689o0.cn1.akuchlin@ute.mems-exchange.org> <a41o1m$8uv$1@panix2.panix.com> <mailman.1013214678.4137.python-list@python.org>
Message-ID: <slrna6ahjc.rih.akuchlin@ute.mems-exchange.org>

In article <mailman.1013214678.4137.python-list at python.org>, 
	Skip Montanaro wrote:
>     Aahz> font size.  Was anyone who was using HTML doing so under Linux?
> Yes, both Jeremy and I were.  I think Jeremy used Netscape.  I used Opera.

Me too, using Opera 6.0TP3 for Linux.  I did have to spend some of
Monday following the instructions in the Font-Deuglification HOWTO and
reconfiguring Opera to use various TrueType fonts.  Without doing
that, the large lettering looks really blocky due to poor font
rendering under X.

--amk                                                  (www.amk.ca)
And now let us hasten to the station. I have commanded the rain to
fall at exactly one-fifteen and I would hate to get my shoes wet.
    -- Lord Lavender, in SEBASTIAN O #2



From phr-n2002a at nightsong.com  Fri Feb 22 20:50:40 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 22 Feb 2002 17:50:40 -0800
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie> <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk>
Message-ID: <7xheo9uftb.fsf@ruckus.brouhaha.com>

philh at comuno.freeserve.co.uk (phil hunt) writes:
> What do you consider the other 2 methods to be? I can think of 
> encoding it in the URL, and the server remembering the browser's IP 
> address. Are there others?

You CANNOT identify sessions by browser IP addresses.  That's been
discussed here several times.  Think of proxy servers sending requests
from many browsers through the same IP address--for example, AOL
proxies millions of browsers through a few hundred addresses.

A number of other means of identifying sessions have been discussed
here in the past few weeks though.


From thomas at gatsoft.no  Wed Feb 20 04:58:28 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Wed, 20 Feb 2002 09:58:28 GMT
Subject: XML-RPC over HTTPS/SSL
Message-ID: <8rKc8.4$5vh.135400448@news.telia.no>

I got a xmlrpc-server running using M2Crypto, making it encrypted with SSL.
I cannot get any xmlrpc-client to connect though, no matter what.

Looked at resources found searching google, for instance
http://lists.zope.org/pipermail/zope/2001-May/089457.html. but with no luck.
Does anybody have any code showing how to connect a client to a
xml-rpc-server using SSL/HTTPS , preferrably with user authentication also?
( to make it perfect, it would accept gzip-encoding to speed up transfers
:-) )

I'm using xmlrpclib from pythonware.com.

I keep seeing stuff about SSL/HTTPS-support in standard python, but ... ??
Do you compile the interpreter with OpenSSL etc ?? My clients and servers
will most probably run 60% on Linux and 40% windows, and re-compilation
isn't always feasable for windows-users.

Any clues or hints appreciated

Best regards,
Thomas Weholt




From fperez528 at yahoo.com  Fri Feb 15 23:08:40 2002
From: fperez528 at yahoo.com (Fernando Perez)
Date: Fri, 15 Feb 2002 21:08:40 -0700
Subject: Why Python is like BASIC ...  java "vs" python
References: <mailman.1013826279.13649.python-list@python.org>
Message-ID: <a4km09$6d9$1@peabody.colorado.edu>

john coppola wrote:

> OK fellas, I'm gonna break it down for you.  Python is
> nothing like BASIC or perl.  Neither is Java like C++.
>  I will admit that I used BASIC.  A friend turned me
> on to Python.  I never looked back.  You cannot
> compare things on basis of how they appear or the
> syntax of the language or certain similarities.  You
> have to know what problem the language is trying to
> solve.  What is its philosophy.

[snip...]
 
> You will begin to see that basic and python are
> nothing alike even though they are both interpreted.
> In what way is basic interpreted.  In what way is perl
> interpreted?  Thats like saying a fish is a whale
> because they both swim in water.  Java is not at all
> similar to C++.  Different.

Sheesh, thanks! For a moment this newsgroup was about to close while we all 
moved to BASIC. I had already started uninstalling Python from all my 
machines and looking for good BASIC modules for 3d data visualization. I 
heard Guido was hastily applying for special discounts for VisualBasic 
books, too.

But now that you've saved the day and clarified things for us, we can 
happily return to our cozy little world of clean, beautiful, expressive 
ophidians. (And by the way, a whale is a mammal, not a fish).
 
> Sorry to be so harsh.  Long day..

Sorry to be so harsh.  Long day..

Cheers,

f.


From quinn at yak.ugcs.caltech.edu  Tue Feb 19 14:09:37 2002
From: quinn at yak.ugcs.caltech.edu (Quinn Dunkan)
Date: 19 Feb 2002 19:09:37 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com>
Message-ID: <slrna758ng.d1u.quinn@yak.ugcs.caltech.edu>

On 19 Feb 2002 00:59:42 -0800, Alexander Jerusalem <ajeru at vknn.org> wrote:
>There's another thing that bugs me with the missing static typing and
>that is that it makes code less readable. I don't know what I can do
>with a function parameter when I see it. For example, when I see a
>function like def computeSalary(person)... I don't know at first sight
>if person is the id of person, a person object and if it's an object
>what its exact class in a Person class hierarchy is. I have to find
>the place where the constructor is called to find that out. That's not

Or you can just look and see how the 'person' is used in the function.  If the
first statement is a 'o = db.by_id[person]' then you know it's an id.  If it
takes 20 lines to get to that point, maybe the function is too long.  If the
function merely passes 'person' to someone else, then the type of 'person'
doesn't matter (and the declaration that a static language (except ones infer
types automatically) would force you to make would merely reduce the
flexibility of the function).

If it's an object, why should you care at what exact position it lives in
some class hierarchy?  Isn't the point of dynamic dispatch that you don't
have to know?

>And the final argument I have for static type checking is that it
>enables method dispatching based on parameter types. In a statically
>typed language you can create two methods that have the same name but
>differ on the parameter types. The correct method will be called for

Yeah, python has this too:

class A:
    def m(self): print 'm'
class B(A):
    def m(self): print 'bm'
o = B()
o.m() #  B.m is called
p = A()
p.m() # A.m is called

It's only of the first argument, but that's what mono-dispatch means.
Static typing having anything to do with multi-dispatch is incorrect,
though.  The best-known object system with multi-dispatch is CLOS, which
is certainly dynamic.  In fact, I don't know any static languages that do
what you want (I'm not counting c++-ish "overloading" because it doesn't
dispatch based on the type of the variable, but the type of the declaration
(which, since you're forced to write out the declaration, you're not
gaining any flexibility or having to type less---so that's the point?)).

When you don't have multi-methods you get by with single dispatch.

>you depending on the type of the argument you pass in your call. That
>makes for a quite flexible way of extending a program. You can just
>add another method with the same name and another type to handle a
>special case without touching the existing methods.

Yep, it's called inheritance, and OO people go on and on about it :)

>I think type checking could be added as an optional feature, like in
>Dylan, without hurting the character of Python.

Oh yes, and dylan, another dynamically typed multi-method language.

>I'm not, however, completely satisfied with static type checking as
>well because I don't see why the compiler forces me to have a
>parameter comply with a static interface that has 10 methods when the
>method that uses the parameter acutally only uses one of those
>methods. I want the compiler to check if the parameter type I'm
>passing has what the method needs but nothing more.

Yeah, that sort of thing is nice.  I want more though... I want
to check if the parameter has what the symbols method needs, *and* that
those symbols do what the method wants.  For that, you need contracts
or something, which work at run time.  As long as you're working at
run time, you'll have to test it, so leave the contracts out and see
if it throws an error :)


From bokr at oz.net  Sun Feb 17 17:08:33 2002
From: bokr at oz.net (Bengt Richter)
Date: 17 Feb 2002 22:08:33 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122> <xyuk7tdku0w.fsf@sex.ifi.uio.no> <a4mn4k$gd2$0@216.39.172.122> <GUJb8.26252$l93.4948148@newsb.telia.net>
Message-ID: <a4p9l1$cg4$0@216.39.172.122>

On Sun, 17 Feb 2002 08:32:38 GMT, "Fredrik Lundh" <fredrik at pythonware.com> wrote:

>Bengt Richter wrote:
>>
>> >But if there really is something wrong with that, shouldn't someone
>> >send a bug report or something to the creators of Python?
>>
>> No, not to the creators of Python per se. This is not a bug
>> in Python itself. Telnetlib is just a piece of the distribution.
>
>the creators of Python gladly accepts bug reports for everything
>in the standard distribution:
>
>    http://python.sourceforge.net/sf-faq.html#bugs
>
Ok, through the proper channel. I was meaning don't email Guido
et al personally[1], guessing that he wouldn't want to be swamped with
unconfirmed bug suspicions not relating to his immediate work.

BTW, why are context diffs preferred over unified? (I assume this
refers to e.g. GNU diff 2.5's -c and -u options respectively?)

Regards,
Bengt Richter



From greg at cosc.canterbury.ac.nz  Wed Feb 13 21:39:26 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 14 Feb 2002 15:39:26 +1300
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net> <MPG.16c760cdaa8ee3639896d5@newnews.dircon.co.uk> <PSi78.6815$3E5.522096@newsread2.prod.itd.earthlink.net> <slrna5s35r.h88.grante@isis.visi.com>
Message-ID: <3C6B235C.D4C5B15F@cosc.canterbury.ac.nz>

Grant Edwards wrote:
> 
> I keep hearing about "on ramps" for the 'net, but I've never been
> able to get my car to go up one of them.  How _do_ you get a vehicle up the
> ramp onto the 'net?

I don't know, but if anyone ever finds a way, I'm sure
it'll be much easier to implement in Python than
anything else. :-)

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


From rthack at eatel.net  Mon Feb  4 23:32:32 2002
From: rthack at eatel.net (jason thackrey)
Date: Mon, 4 Feb 2002 22:32:32 -0600 (CST)
Subject: Instant Internet Empire
Message-ID: <200202050432.g1544ZqX002305@ens1.eatel.net>



    Its official, the Internet is currently creating more millionaires than any other industry in history. A vastly growing international marketplace, with seemingly no boundaries, the income potential is AMAZING. Are you taking advantage?
We can help you get 5 money making websites up and running with a HALF HOUR, and show you exactly how to produce traffic and process sales.. no expertise required!
The best part: Success is GUARANTEED. If you're not making at least $500 pure profit in your first month, or if you're not satisfied for any reason, we will refund every penny of your purchase price... there is no risk involved, you cant lose!

Click below for more info:

http://internetempire.hopto.org



Internet Empire



From jwjung at tclab.kaist.ac.kr  Sat Feb  2 19:50:28 2002
From: jwjung at tclab.kaist.ac.kr (jwjung at tclab.kaist.ac.kr)
Date: Sun, 3 Feb 2002 00:50:28 +0000 (UTC)
Subject: Shaped window on both X/*nix and win32
Message-ID: <a3i1gk$2m5$1@news.kreonet.re.kr>

Hello,

Is there any way to create shaped (not rectangular) window
using tkinter or something else? I want to make a small utility
which should run on both X and Windoze.

I want to do this using only standard packages if possible.

Thanks in advance.

-Sanori

-- 
Jung Joowon				| E-mail: jwjung at tclab.kaist.ac.kr
Dept. of Computer Science		|     /\^	Always being
Korea Advanced Inst. of Sci. and Tech.	|    /   \/\		with Network
Taejon, Republic of Korea, 305-701	|   /	  \ \	Mountain Duck (Quack)


From theller at python.net  Tue Feb 19 13:58:12 2002
From: theller at python.net (Thomas Heller)
Date: Tue, 19 Feb 2002 19:58:12 +0100
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com> <a4rsa2$2grrk$1@ID-59885.news.dfncis.de> <u1yfiz5yu.fsf@ctwd0143.fitlinxx.com>
Message-ID: <a4u78u$376an$1@ID-59885.news.dfncis.de>

"David Bolen" <db3l at fitlinxx.com> wrote in message news:u1yfiz5yu.fsf at ctwd0143.fitlinxx.com...
> "Thomas Heller" <theller at python.net> writes:
> > I'm currently working on an bdist_wininst enhancement which takes an
> > additional command line option (--install-script <script.py>).  This
> > script would then be run (with sys.argv set to [script_name,
> > '--installed']) after all the files have been copied, and it would
> > again be run (with sys.argv set to [script_name, '--removed']) just
> > before the deinstaller is about to remove all the files and
> > directories.
>
> Sounds like just the ticket.  I suppose it could still be a little
> tricky to arrange the OPs request of shortcuts unless you could assume
> the Win32 extensions would also be present, but still plenty of
> flexibility.

The prototype I have just now runs the (post)installation/(pre)deinstallation
script in an environment where another buildin function is available:

create_shortcut(path, description, filename[, arguments[, workdir[, iconpath, iconindex]]])

This function is implemented in the bdist_wininst runtime - win32all
extensions not required (for this case).

Thomas




From tbryan at python.net  Thu Feb 14 08:13:50 2002
From: tbryan at python.net (Tom Bryan)
Date: Thu, 14 Feb 2002 13:13:50 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <mailman.1013661023.30875.python-list@python.org>
Message-ID: <iKOa8.102499$A51.43433850@typhoon.southeast.rr.com>

Tim Peters wrote:

> [Tom Bryan]
>> Is there a way to use tabs exclusively for indentation in IDLE?
> 
> Yes, provided you're happy with indentwidth==8.

:-P
 
>> Basically, I'd like tabwidth to equal identwidth while permitting
>> developers to adjust the indentwidth.
> 
> You're out of luck there, if they want anything other than 8 for both.

That's what I was afraid of.  I think that Jython could do good things for 
my group at work, but I really don't look forward to the stupid whitespace 
discussions.  It's bad enough on c.l.py.  I can't wait to hear it in person.
 
>>  I see that AutoIndent.py says
[...snip...]
>> But I don't quite understand that explanation.
> 
> Don't be so timid <wink>:  change the code and see what happens.  

I did.  I changed it to 4 and 4.  I toggled tabs on in Idle.  It seemed to 
work on my Linux box. 

> meaning of the comment will soon become painfully clear.  Have you ever
> used a word processor (as opposed to a text editor)?  Tk switches to "word
> processor" tabs when you change its default tab setting.

Ick.  So the tabs will depend on the font, etc.?

>> ...
>> It doesn't look like IDLE is friendly to that type of customization.
> 
> Not exactly:  the Tk text widget is the hangup here.  IDLE's autoindent
> code is shared by PythonWin, which latter, being based on the Scintilla
> text widget, doesn't suffer the same tab surprises.

Hm.  PythonWin might be a good option.  The UNIX heads will probably just 
use vi and Emacs, which I think we can handle.  The current "approved" 
coding style for all languages in the group is supposed to be "always 
indent with leading tabs; for alignment purposes, tabs should be viewed as 
4 spaces; set your visual display of tabs to whatever you prefer."  
 
>> I know that emacs is.  I'm currently investigating the other editors
>> that my team uses, such as CodeWright.
> 
> CodeWright won't be a problem.  But stick to IDLE's default (== the Emacs
> python-mode default), and no editor anywhere will be a real problem.

I would, but then we've got spaces in some languages and tabs in others.  
And I dread the discussion of how many spaces one level of indentation 
should be even more than I dread the whitespace-delimited-blocks whining.

I've already been coding some samples with 

(defun my-python-mode-hook ()
  (setq tab-width 4)
  (setq indent-tabs-mode t)
  )
(add-hook 'python-mode-hook 'my-python-mode-hook)  

It's stupid, but the whitespace issue is probably going to be one of the 
biggest obtacles to introducing Jython to my co-workers.  I'll be back to 
tell everyone how it goes.

I-just-hope-no-one-mentions-the-alien-whitespace-eating-nanovirus-ly y'rs,
---Tom



From NOSpawnPAM at lightspawn.org  Wed Feb  6 03:03:05 2002
From: NOSpawnPAM at lightspawn.org (pawn)
Date: 6 Feb 2002 00:03:05 -0800
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu> <B8809EB2.411E%jonathan@onegoodidea.com>
Message-ID: <4b4c5d99.0202060003.4fc72845@posting.google.com>

Thank you! I got it to work, kinda. Here's what I eventually did:

#include <stdlib.h>
#include <string.h>

int main( int argc, char* argv[] ) {
    int len;
    char* args[3];
    args[0]="python";
    args[1]=argv[0];
    len = strlen(args[1]);
    args[1][len-3]='p';
    args[1][len-2]='y';
    args[1][len-1]='\0';
    args[2]=NULL;
    execv( "/usr/bin/python", args );
}

Now instead of running, say, test.cgi, which is written in python...

$ mv test.cgi test.py
cp wrapper.cgi test.cgi

Repeat once for every script that needs to run with my (your) uid.

At least I hope somebody else will come across this while having the same problem.


From phd at phd.pp.ru  Thu Feb 21 07:18:25 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 21 Feb 2002 15:18:25 +0300
Subject: Reloading modules
In-Reply-To: <u1yffm45s.fsf@python.net>; from mwh@python.net on Thu, Feb 21, 2002 at 12:02:08PM +0000
References: <a51tt1$msq$1@news1.ucsd.edu> <u664rm510.fsf@python.net> <u1yffm45s.fsf@python.net>
Message-ID: <20020221151825.J15829@phd.pp.ru>

On Thu, Feb 21, 2002 at 12:02:08PM +0000, Michael Hudson wrote:
>     if file.endswith("c"):
>         file = file[:-1]

if file.endswith("c"):
   file = file[:-1]
elif file.endswith("o"):
   file = file[:-1]

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



From stephen.tubbs at paretopartners.com  Mon Feb 25 17:19:11 2002
From: stephen.tubbs at paretopartners.com (Stephen)
Date: 25 Feb 2002 14:19:11 -0800
Subject: Help using ODBC.pyd to retrieve ntext data from SQL server
Message-ID: <24420294.0202251419.82d3373@posting.google.com>

I need help with a work around to read data from an ntext field on
Microsoft SQL Server v7.0.  Python only returns None.  I'm able to get
at the data in the field using a variety of other applications
(Access, Query analyzer, etc...)  I have been through the odbc.pyd
source and don't find any obvious exclusions for ntext data types. 
Any help would be greatly appreciated

Stephen


From phd at phd.pp.ru  Sun Feb  3 15:28:16 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sun, 3 Feb 2002 23:28:16 +0300
Subject: Python 2.2 Modules/Setup lies?
In-Reply-To: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net>; from jon+usenet@unequivocal.co.uk on Sun, Feb 03, 2002 at 08:21:28PM +0000
References: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net>
Message-ID: <20020203232816.A15561@phd.pp.ru>

Because Python 2.1+ uses Disutils, not Setup for configuring itself. This
allows better testing current OS/libraries configuration and selection of
modules.

On Sun, Feb 03, 2002 at 08:21:28PM +0000, Jon Ribbens wrote:
> I am installing Python 2.2 from source on an OpenBSD machine and a
> Solaris machine.
> 
> The build instructions indicate that you should:
> 
>   ./configure
>   vi Modules/Setup
>   make
> 
> When I come to edit Modules/Setup, I am somewhat confused, because
> everything in it is commented out, including standard built-in modules
> that are explicitly commented "on by default" but which appear to be
> off by default - e.g. the following section of the file:
> 
>   # Modules that should always be present (non UNIX dependent):
> 
>   #array arraymodule.c    # array objects
>   #cmath cmathmodule.c # -lm # complex math library functions
>   #math mathmodule.c # -lm # math library functions, e.g. sin()
>   #struct structmodule.c  # binary structure packing/unpacking
> 
> Why are these (and everything else in the file) commented out?

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



From bokr at oz.net  Fri Feb 15 15:44:34 2002
From: bokr at oz.net (Bengt Richter)
Date: 15 Feb 2002 20:44:34 GMT
Subject: certification
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au> <mailman.1013783246.14100.python-list@python.org>
Message-ID: <a4jrvi$njd$0@216.39.172.122>

On Fri, 15 Feb 2002 17:25:27 +0300, Oleg Broytmann <phd at phd.pp.ru> wrote:

>On Fri, Feb 15, 2002 at 01:50:54PM +0000, Trung Hoang wrote:
>> i was wondering if i can get certified with python from an internationally
>> recognised company?
>
>   http://www.brainbench.com
>   Python 1.5 - $50.
>
Any GPL or PSL scripted self-assessment tests exist?
For any language or subject, for that matter?

Regards,
Bengt Richter



From akuchlin at dust.mems-exchange.org  Wed Feb 20 16:22:26 2002
From: akuchlin at dust.mems-exchange.org (A.M. Kuchling)
Date: 20 Feb 2002 21:22:26 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com>
Message-ID: <slrna784si.m30.akuchlin@dust.mems-exchange.org>

In article <de3fc1ef.0202201233.2228e5b1 at posting.google.com>, 
   William Tanksley Google wrote:
> My friend was griping about one battery that Basic comes with but
> Python doesn't (at least not that I've seen): graphics.  Yes, I know

turtle.py is cute, and requires Tkinter:

>>> import turtle
>>> t = turtle.Pen()
>>> t.down()
>>> t.forward(10)
>>> t.right(45)
>>> t.forward(35)
>>> for i in range(4):
...    t.forward(50) ; t.right(90)
...
>>>

When you say you'd like pixel graphics, what do you mean?  Just
setting elements of a 2D pixel array to various colours?  Maybe
something for that could be built on top of Tkinter, too.

--amk                                                  (www.amk.ca)
Ace, I'm getting too old for this sort of thing. He's all yours from
now on... I'm going home to Doris.
    -- The Brigadier, in "Battlefield"


From sholden at holdenweb.com  Wed Feb  6 13:29:06 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 6 Feb 2002 13:29:06 -0500
Subject: Twisted Selector
References: <u5rsqqb9tamaad@corp.supernews.com>
Message-ID: <LBe88.87264$KM2.2829250@atlpnn01.usenetserver.com>

"Lee Harr" <missive at frontiernet.net> wrote ...
> 
> I am looking at this Twisted tutorial:
> http://twistedmatrix.com/page.epy/philosophy.html
> and having trouble with the very first code.
> 
> >>> from twisted import threadable
> ImportError: cannot import name threadable
> >>> from twisted.python import threadable
> >>>
> >>> from twisted import internet
> >>> selector = internet.Selector()
> AttributeError: 'twisted.internet' module has no attribute 'Selector'
> 
> Ok, I was able to get past the first problem. Looks like since
> Twisted-0.9.0 when the tutorial was written, they refactored
> threadable into a subdirectory, but I grepped the whole distribution
> and see nothing called Selector anywhere.
> 
> This is with Twisted-0.15.0, but I went back in time to 
> Twisted-0.12.3 and still nothing called Selector.
> 
> The ChangeLog does not mention Selector either.
> 
> Anyone have any ideas?
> 
Moshe? Glyph?

regards
 Steve
-- 
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 22:15:51 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 21:15:51 -0600
Subject: defining functions
References: <20020210214159.13516.00000450@mb-cq.aol.com>
Message-ID: <015501c1b2aa$6af96c00$0101010a@local>

----- Original Message ----- 
From: "Patio87" <patio87 at aol.com>


> I am pretty new to python, and I dont understand the parameters when 
> defining functions. Whenever I see a function definition with a 
> argument It seems like the argument already is a variable, but it 
> goes threw the function and it makes no sense. I dont know if what I 
> just said made any sence but if you can help me please reply
> 
> def factorial(n):
>     if n == 1:
>         return 1
>     else:
>         return n * factorial(n-1)
> What the hell does 'n' have assingned to it?

At this point, you are just *defining* the function.  Though it
is inaccurate, you may think of 'n' as a placeholder:

Python 2.1.2 (#1, Jan 16 2002, 17:04:23)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> def factorial(n):
...     if n == 1:
...         return 1
...     else:
...         return n * factorial(n-1)
...
>>> print factorial(5)
120
>>>

When you define the function, the question of "what is assigned to
'n'" is irrelevant.  When you *call* the function, *that's* when it
matters, and *that's* when it's obvious (5 in the case above).






From bh at intevation.de  Thu Feb 21 07:30:56 2002
From: bh at intevation.de (Bernhard Herzog)
Date: 21 Feb 2002 13:30:56 +0100
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <slrna79a8r.6ot.scarblac@pino.selwerd.nl>
Message-ID: <6q1yff10fz.fsf@abnoba.intevation.de>

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

> Jason <caljason76 at yahoo.com> wrote in comp.lang.python:
> > In my experimentation 0, None, "", [], and () all seem to be
> > considered false.  Just to confound me none of them are == to each
> > other.  Coming from a Scheme world this is confusing.  Are there any
> > other values that are false?
> 
> {}, at least. Any classes that return the right results on certain special
> functions (__nonzero__(), if I remember correctly).

or __len__ if __nonzero__ isn't defined.

> > Why aren't () and "" the same thing?
> 
> Because one is an empty tuple and one is an empty string.

Plus they're not equal or identical in scheme either, at least not in
guile:

guile> (equal? "" ())
#f
guile> (eq? "" ())
#f


   Bernhard

-- 
Intevation GmbH                                 http://intevation.de/
Sketch                                 http://sketch.sourceforge.net/
MapIt!                                               http://mapit.de/


From dale at riverhall.NOTHANKS.co.uk  Fri Feb 15 13:58:05 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Fri, 15 Feb 2002 18:58:05 +0000
Subject: Object copying itself
Message-ID: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com>

I have an object which is just data and an __init__ to set up some
defaults.

Now I need to allow __init__ to get the defaults from another instance
of itself if passed as an argument.

Most of the values are simple numerics and a few object references but
there are also a couple of lists. 

I want object references copied (shallow copy) but the lists need to
be copied entirely (deep copy).

Do I have to do this by hand or is there a neat way of doing it?

Ta very much.
--
Dale Strickland-Clark
Riverhall Systems Ltd


From grante at visi.com  Fri Feb 22 17:01:54 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 22 Feb 2002 22:01:54 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1014404689.32639.python-list@python.org>
Message-ID: <slrna7dfuh.gtf.grante@isis.visi.com>

In article <mailman.1014404689.32639.python-list at python.org>, Jason Orendorff wrote:
> Grant Edwards wrote:
>> I just got bit by the leading-zero thing again about 10 minutes
>> ago when I changed the minor version number of something from 07
>> to 08.  Had to skip from 1.07 to 1.10...
> 
> Serves you right.  Version numbers like 1.07 are silly.
> 1.0.7 or 1.7 would be more sensible.

Probably so, but the "official" method enshrined in the ECO database design
and procedures and whatnot is M.NN.

> A tuple is particularly swell for version numbers.
> See sys.version_info.

-- 
Grant Edwards                   grante             Yow!  The LOGARITHM of
                                  at               an ISOCELES TRIANGLE is
                               visi.com            TUESDAY WELD!!


From phr-n2002a at nightsong.com  Fri Feb  1 06:17:40 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 01 Feb 2002 03:17:40 -0800
Subject: Please comment on PEP XXX: Enhanced Generators
References: <a3at9p$2he$1@bob.news.rcn.net> <20020131092440.GA19083@hishome.net> <000b01c1aa3a$1207a300$4dd8accf@othello> <mailman.1012556582.17931.python-list@python.org>
Message-ID: <7x7kpx1mej.fsf@ruckus.brouhaha.com>

Oren Tirosh <oren-py-l at hishome.net> writes:
> > 2.  Generator comprehensions:  [yield int(x) for x in floatgen()]
> 
> The only problem I see with this syntax is that the square brackets are
> misleading - the resulting object is not a list.  How about using parens
> instead?  They are not strictly required, but just like for tuples, they
> help disambiguate the syntax.

Actually there's sort of a worse problem with either version, which is
how it interacts with the proposal that yield can return a value.

If I can say "x = yield 3", then can't I say

  x = [yield 3, yield 4, yield 5]
  
which should suspend execution 3 times before actually assigning to x?

And if I can do that, can't I say the same thing as a list (not
generator!) comprehension

  x = [yield i for i in range(3,5)]

I realize that in the original proposal, yield can't necessarily take
an arbitrary argument when it returns a value sent by .submit(), but
the option of doing that (to have fully general coroutines) should be
left open.


From wrbt at email.com  Fri Feb 22 15:02:43 2002
From: wrbt at email.com (Larry)
Date: 22 Feb 2002 12:02:43 -0800
Subject: fun with ADO
Message-ID: <2ec1bc1c.0202221202.1e0ea6aa@posting.google.com>

I thought I'd share some stuff I found while trying to tune an app
that interacts heavily with a sql server database via ADO. The web
application I've been rewriting in python was having serious problems
with performance. I knew it wasn't IIS because things were running
really slow from the interactive window too.

After fooling around with profile for a while the pig was unveiled as
actually getting the field values from the Sql Server database via
win32com/ADO.

a toy version of the technique I was using was:

    rs=win32com.client.Dispatch(r'ADODB.Recordset')
    rs.Open(sql,connectString)
    while not rs.EOF:
        name=str(rs.Fields("clientname").Value)  #oink!
        gid=str(rs.Fields("gid").Value)
        (- do something fun with fields -)
        rs.MoveNext()
    rs.Close()

A put some timers on a test script that ran through a few hundred
records, and it ran in about 1.9 seconds. That sucks. I also tried
using mapping field values into a list and extracting from that, no
better.

So I tried removing the Value, as in: name=str(rs("clientname"))
The average new run time was 0.7 seconds! Woah. Nice.

Next I tried using the sparsely documented Collect method of the
recordset object, as in: name=str(rs.Collect("clientname"))
The average run time was now down to 0.24 seconds. Viva Collect!

Finally I ran across an article where someone mentioned this
technique:

    rs.Open(sql,connectString)
    f1=rs("gid")
    f2=rs("clientname")
    while not rs.EOF:
        name=str(f2)
        gid=str(f1)
        (- do something fun with fields -)
        rs.MoveNext()
    rs.Close()

We have a winner! Average time was 0.18 seconds for this method.

So for any "ADOites" (tm) out there in python land I'd recommend
either Collect or maintaining field references outside of recordset
iterations. I've increased performance in many parts of my application
by 10x or more.


From gerson.kurz at t-online.de  Mon Feb 25 09:52:36 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 14:52:36 GMT
Subject: images
References: <iGhe8.20623$vP.68360@rwcrnsc51.ops.asp.att.net>
Message-ID: <3c7a4f22.6686125@news.t-online.de>

On Mon, 25 Feb 2002 02:53:02 GMT, "Ali K" <akakakhel at attbi.com> wrote:

>I would like to know how to show a picture in Python so that it shows up on
>my window.

I assume you want to use that on Windows (since you're posting with
OE). What GUI do you use - TK, wxPython, win32? 

For wxPython, there is a simple demo that loads a splash screen;
(Sorry, have wxPython not installed, but take a look at the samples).



From tatebll at aol.com  Fri Feb  1 08:43:20 2002
From: tatebll at aol.com (Bill Tate)
Date: 1 Feb 2002 05:43:20 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com>	<20020130073810.382$1b@news.newsreader.com>	<b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org>
Message-ID: <cb4ba455.0202010543.7d4edd1a@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message > 
> However, your last statement that managers make _all_ the decisions was an
> unfortunate one.  It's better if the technicians make the decisions and
> then managers _approve_ them.  This may be splitting hairs, but when it...
I find little to disagree with what you're saying, but managers also
need to know that their development team is being unbiased and
objective in their analysis :>).


From daniels at mindspring.com  Sun Feb 17 00:42:13 2002
From: daniels at mindspring.com (Alan Daniels)
Date: 16 Feb 2002 21:42:13 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <861yflwo4s.fsf@apo.hmotakef.int>
Message-ID: <5d283efe.0202162142.53e3b402@posting.google.com>

Henrik Motakef <henrik.motakef at web.de> wrote in message news:

> Or, in the words of Edsger Dijkstra:
> 
> "It is practically impossible to teach good programming style to
> students that have had prior exposure to BASIC: as potential
> programmers they are mentally mutilated beyond hope of
> regeneration."

Dikstra also said:

"The use of COBOL cripples the mind; its teaching should, therefore,
be regarded as a criminal offense."

After seeing that, I learned not to take the original BASIC quote
too seriously. COBOL, like BASIC, was never designed to be a true
general purpose language as far as I know; COBOL was meant to allow
accountant types to express business logic to a computer, and BASIC
was meant as a teaching language, and not ever meant to scale to real
programming tasks (sorry Microsoft, but it's true). 

So, whether these two languages either serve as springboards to "real"
languages later on, or as illusory dead-ends that stunt the growth of
a programmer's potential before it even gets started, I don't know. I
*do* know several very capable programmers who started with GW-Basic
back in the day, and they now work in C++/Java/Perl/My Favorite Language
just fine. They just need to unlearn the habits of using globals and
variants everywhere. Relatively small conceptual hurdles, I believe,
compared to the more difficult concepts such as learning stack vs. heap
memory management (or why even such a thing exists), choice of the best
data structure for a given task, etc etc. But then again, I'm no Dijkstra,
and I tend to be less judgemental of such things.


From jwk at mrao.cam.ac.uk  Thu Feb  7 10:29:35 2002
From: jwk at mrao.cam.ac.uk (James Keen)
Date: 7 Feb 2002 07:29:35 -0800
Subject: Serial port Input/Output conflicts
Message-ID: <ef77152a.0202070729.3a6a5d2a@posting.google.com>

I have been trying to write software to control a device via the
serial port on a Linux machine. I can send data sucessfully but when I
send a status request (a command line that asks my hardware to send
back some information) it disables my ability to send anything else.

The status data gets sent to the port from my Hardware, which I can
read, but the serial port refuses to send any more commands to my
hardware until I close and then reopen the port.

Any ideas?

If you're interested I've included the pertenent bits of my code...

import sys,os, fcntl, FCNTL, termios, TERMIOS
serial = os.open('/dev/ttyS1', FCNTL.O_RDWR|os.O_NDELAY,0)

def SendToMotor(axis, message):
   global serial
   os.write(serial,str(axis)+' '+message+'\n')

def ReadMotor():
   global serial
   a=''
   while (a != '\n'):
      a=os.read(serial,1)
      print '['+a+']'

def quit(self):
   os.close(serial)

Thanks in advance

James Keen


From cfelling at iae.nl  Tue Feb 26 19:54:15 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 27 Feb 2002 01:54:15 +0100
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <a5hann$j73$1@animus.fel.iae.nl>

HankC <hankc at nospam.com> wrote:
> Greetings:

> I'm 2 days new to Python and hope I can get some pointers here...

Welcome and enjoy.

> I have a sorted, delimited text file such as:

> Arty,1000
> Bobby,2000
> Charlie,3000

> Knowing a name, I want to return the entire line from the file.  My
> approach/problems (any comments appreciated):

> # use the filter method
> # filter fxn
> def Match(line): return line[:toklen] == tok

this should be (assuming a reasanably up-to-date Python):

    def Match(line, tok): return line.startswith(tok + ",")


> # find match fxn
> def FindMatch:
> 	fp = open(fname)
> 	flist = fp.readlines()
> 	dataline = filter(Match, eodflist)

I don't see how this could work (what is eodflist?), but an
alternative might be:

    def FindMatch(fname, tok):
        key = tok + ","
        for line in open(fname):
            if line.startswith(key):
                return line

The loop will end on the first match found.  If this is not what you
wanted, but instead you're after *all* matching lines, how about:

    key = tok + ","
    matches = [ x for x in open(fname) if x.startswith(key) ]


> More general...  does filter() have any performance advantage over
> just using a for loop?  The datafile list has 10,000+ items.  My

Not if you use a non builtin function.

> background is in object pascal where I would read the items into a
> list then use a binary search to find the target.  I'm hoping there's
> an efficient python method to perform a search easily on a sorted
> list.

Depending on the length of the list you might be better of with a linear
search where you stop reading from the file the moment the name maches.
Otherwise you could write your own binary search in python, but I leave
that as an excercise to you:)

> Suppose I want to make tok and toklen global to the module.  Is this
> possible?

Yes ofcourse, but I would strongly advice against it:)
-- 
groetjes, carel


From emile at fenx.com  Tue Feb  5 22:21:00 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 5 Feb 2002 19:21:00 -0800
Subject: To all Higher Level....A True Beginner
References: <ba59f11e.0202051821.5c41221e@posting.google.com>
Message-ID: <a3q7oi$1a26gu$1@ID-11957.news.dfncis.de>

"David Pham" <azndizzy at asia.com> wrote in message
news:ba59f11e.0202051821.5c41221e at posting.google.com...
> Hey! Everyone!, I like to know if any of you programmres recommend any
> books, online tutourial for myself, or anyone else that is a beginner.


http://python.org/doc/Newbies.html

--

Emile van Sebille
emile at fenx.com

---------



From claird at starbase.neosoft.com  Thu Feb 28 10:47:47 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Feb 2002 09:47:47 -0600
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org> <3C7E198C.DF6E7566@nospam.linux3dgraphicsprogramming.org.nospam> <3C7DC3EA.1827DB65@bellatlantic.net>
Message-ID: <47E28190415AE387.DCB58FC35DE503C4.2EF0C21A21479844@lp.airnews.net>

In article <3C7DC3EA.1827DB65 at bellatlantic.net>,
David Lees  <debl2nonospammy at bellatlantic.net> wrote:
>They published 'The Quick Python Book', which I have found to be quite
>useful and accurate for a beginner like me.
>
>David Lees
>
>
>nlin wrote:
>> 
>> Susan Capparelle wrote:
>> 
>> > We're a publisher of high quality computer books currently seeking
>> > authors for a number of Python related books.
>> 
>> As a published book author with two well-selling titles to my name, I'd
>> just
>> like to warn potential authors about my negative experience with
>> Manning.
			.
			.
			.
Me, too.  I've had plenty to do with Manning, some of
it disappointing, mostly OK.  They're among my pre-
ferred publishers; I also understand their limits.
-- 

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


From gbreed at cix.compulink.co.uk  Mon Feb 18 10:00:56 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Mon, 18 Feb 2002 15:00:56 +0000 (UTC)
Subject: There is more than one way to do it - and for no apparent reason.
References: <3C71034B.20508@mxm.dk>
Message-ID: <a4r4v8$ira$1@thorium.cix.co.uk>

Max M wrote:

> well I can allways just mix-in my own class
> 
> class IWILLHaveItMyWay:
> 
>      def __getitem__(self, key):
>          return getattr(self, key)
> 
>      def __setitem__(self, key, value):
>          setattr(self, key, value)

In 2.2, you can do better

class HaveItYourWay:
  def __getitem__(self, key):
    return getattr(self, key)
  def __setitem__(self, key, value):
    setattr(self, key, value)
  def __iter__(self):
    return self.__dict__.__iter__()

>>> dict = HaveItYourWay()
>>> dirtyList = ['keys', 'values', 'items']
>>> for item in dirtyList:
...   dict[item]=42
...
>>> for key in dict:
...   print key, dict[key]
...
keys 42
items 42
values 42
>>> dict.keys
42
>>> dict.values
42
>>> dict.items
42
>>> dict.items=31
>>> dict['items']
31


              Graham


From SBrunning at trisystems.co.uk  Thu Feb 21 04:22:14 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 09:22:14 -0000
Subject: Adding GUI to application
Message-ID: <31575A892FF6D1118F5800600846864DCBCCCE@intrepid>

> From:	mgash at trisystems.co.uk [SMTP:mgash at trisystems.co.uk]
> I feel that this change will have to be multi-threaded, but as I have
> said, I am a newbie to programming and am 'drowning' whilst trying to
> complete this.
 
Mark,
One way of doing this would be to run the thing as a service, as Andrae
pointed out. Another would be to run the GUI element and the scan element in
separate threads.

Starting the threads up isn't too complicated - see the effbot's "Python
Standard Library" for an example. (You can look at my copy - I'm only up the
stairs, after all. <wink>)

But then you need to get your two threads to communicate. I don't know how
to do this, but I'd have a look at the Queue module if I were you.

Good hunting!

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From pedronis at bluewin.ch  Thu Feb 28 20:25:28 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Fri, 1 Mar 2002 02:25:28 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <slrna7tis4.1m1j.Gareth.McCaughan@g.local>
Message-ID: <3c7edb7f_3@news.bluewin.ch>

> von Neumann didn't invent his construction of the ordinals
> because it was obviously the right thing, or because it
> was natural. He did it because it was simple. When what
> you're doing is laying foundations for mathematics,
> that counts for a lot. But that's not what Python is about.
>
> Go and look at a set theory textbook that describes that
> construction of the natural numbers. It's a good guess that
> within a few pages they will also have defined integers
> as equivalence classes of ordered pairs of natural numbers,
> rationals as equivalence classes of ordered pairs of integers,
> and real numbers as Dedekind sections or equivalence classes
> of Cauchy sequences or something. Should we take inspiration
> from those too? Then, for instance, we could do
>

No. But anyway I don't  want to enter a debate around
mathematical taste (because btw it would be irrelevant here
<wink>).

I'm guilty, especially because I have used and stretched
 the word natural
that for mathematicians have very peculiar meanings,
 but as I said in another post, my insight
was not particularly against or in favor.
Just a datapoint.

regards, Samuele Pedroni.

PS: I have a Master in Mathematics.






From sholden at holdenweb.com  Tue Feb 12 11:31:03 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Feb 2002 11:31:03 -0500
Subject: 19 y/o blonde want to know something about Python working with IPX protocol
References: <mailman.1013529085.695.python-list@python.org>
Message-ID: <drba8.175471$KM2.7737827@atlpnn01.usenetserver.com>

"Mariano Corsunsky" <core.lists.python at core-sdi.com> wrote in message
news:mailman.1013529085.695.python-list at python.org...
DISCLAIMER:

Hi!, I tried to get this same question answered before, but nobody did. So
I'm using the "19 y/o blonde strategy"


----

Hi boys!

I want some examples of Python dealing with the IPX protocol.
I do only need a simple, so so simple example which sends and receive
strings between 2 computers using IPX.

Could you help me?

I need someting like ........

(PC number 1)
SEND "I'm gonna aply some sun-tan oil over my californian body"

(PC number 2)
[Received-->] "I'm gonna aply some sun-tan oil over my californian body"

SEND "...but you are only pretending to be a 10aged girl just to get this
question answered!, STOP that, or everybody will start thinking you're gay!"

(PC number 1)
[Received-->]
"...but you are only pretending to be a 10aged girl just to get this
question answered!, STOP that, or everybody will start thinking you're gay!"



-----

That's it.
Thank you all!

[sigh] It's posts like this that get blondes a (quite unjustified) bad name
...

Well, perhaps if the zillion or so of us who don't know anything about IPX
all reply with "sorry, I don't know anything about IPX" will you feel any
better? I know TCP/IP is pretty well integrated into Python via the sockets
module, but haven't come across anyone using outdated Novell protocols at
all. Is there some reason why it *must* be IPX?

at-least-we-aren't-ignoring-you-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From nhodgson at bigpond.net.au  Sun Feb 10 17:27:04 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Sun, 10 Feb 2002 22:27:04 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>	<3c62a989.3448437947@news>	<a3ucsk$12h8$1@nntp6.u.washington.edu>	<a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com> <5659.806T1220T5243918threeseas@earthlink.net>
Message-ID: <YsC98.16409$Zu6.75381@news-server.bigpond.net.au>

Timothy Rue:

> MS DOS began with a Command Line Interface, then GUI but never did it
> evolve it's Command Line Interface to have what is considered standard
> functionality like command line history

   Command line history became available with the DOSKEY command in MSDOS
5.0 released 1991.

> The Mac began it's user interfaces with the GUI, never really having a
> Command Line Interface - Though Now it might have one in OSX.

   There was the MPW Shell available to programmers. Even back in 1984 there
were primitive command line shells for the Mac although the OS did not
support many of the concepts, such as redirectable streams, that make
command lines useful.

> Its third
> user interface is limited or constrained to applescript.

   It can be used from C or Python or UserLand Frontier or any other OSA
(Open Scripting Architecture) supporting language.

   Neil




From WEC7108 at boeing.com  Tue Feb 12 14:10:28 2002
From: WEC7108 at boeing.com (WEC7108)
Date: Tue, 12 Feb 2002 19:10:28 GMT
Subject: Developing a GUI program like BOIDS
Message-ID: <3C6968A4.1CA9C62E@boeing.com>

I would like to develop a GUI program similar to the ever-popular BOIDS
using Python instead of Java.  Are there URL references or books on this
kind of development?  :o)  Bill


From mhuening at zedat.fu-berlin.de  Sun Feb 17 05:47:01 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 17 Feb 2002 10:47:01 GMT
Subject: win32com and MS Excel
References: <mailman.1013886806.19234.python-list@python.org>
Message-ID: <Xns91B877D7B6FD6mhueningzedatfuberli@130.133.1.4>

"Jimmy Retzlaff" <jimmy at retzlaff.com> wrote in 
news:mailman.1013886806.19234.python-list at python.org:

> If you haven't already, you can get quite a speed boost by running
> Makepy on the Excel type library. 

Well, I have no idea what is happening when running Makepy, but you are 
definitely right: the result is a 'speed boost'. Thanks!!

One more question: what is going to happen when I build a standalone 
(using py2exe)? Will the generated file be included and affect the speed 
on another machine too?

Matthias


From phr-n2002a at nightsong.com  Thu Feb 21 06:50:10 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 03:50:10 -0800
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de>
Message-ID: <7xu1sbjbpp.fsf@ruckus.brouhaha.com>

gerson.kurz at t-online.de (Gerson Kurz) writes:
> I found an easier solution:
> 
> sort = lambda x:(x.sort(),x)[1]

Cute!


From roy at panix.com  Wed Feb 20 23:20:45 2002
From: roy at panix.com (Roy Smith)
Date: Wed, 20 Feb 2002 23:20:45 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com>
Message-ID: <roy-C72B26.23204520022002@reader1.panix.com>

Roy Smith <roy at panix.com> wrote:
> PS -- In the lists of false things in python that have been mentioned on 
> this thread, I don't think I've seen anybody mention {}, the empty 
> dictionary.

Along those same lines, I half-expected the empty function to be false, 
i.e.:

def foo():
   pass

if (foo):
   print "it's true"
else:
   print "it's false"

should have printed "it's false".  Alas, it says it's true.  Seems kind of 
irregular to me, not that I would imagine there's much practical 
consequence one way or the other.


From maxm at mxm.dk  Thu Feb  7 16:04:42 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 07 Feb 2002 21:04:42 +0000
Subject: IIS ASP Python 2.2 script works first time only
References: <f36760cb.0202071137.61bc8027@posting.google.com>
Message-ID: <3C62EBEA.3060806@mxm.dk>

Stephen Arndt wrote:

 > After upgrading from Python 2.1 to 2.2 with win32all-146.exe my
 > hello.asp script runs first time OK. Then just a blank page is
 > displayed in IE browser window next Refresh or activation.  I am
 > running Python 2.2 on WindowsNT 4.0 server and IIS 4.0.
 >
 > Any ideas why the hello.asp script would generate HTML code only the
 > first time a new hello.asp file is copied to the server?

Problem and solution here (one line url):

http://groups.google.com/groups?hl=da&threadm=vtk68.20338%248d1.7162721%40news1.rdc1.md.home.com&rnum=1&prev=/groups%3Fq%3Dpython%2Biis%2Bwin32all-146%26hl%3Dda%26selm%3Dvtk68.20338%25248d1.7162721%2540news1.rdc1.md.home.com%26rnum%3D1

regards Max M



From nospam at bigfoot.com  Tue Feb  5 12:42:52 2002
From: nospam at bigfoot.com (Gillou)
Date: Tue, 5 Feb 2002 18:42:52 +0100
Subject: SOAP and python
References: <3C600810.4080805@wanadoo.fr>
Message-ID: <a3p55j$17e9$1@norfair.nerim.net>

Kondor,

It seems that (according the traceback) this is a transport/network problem,
not directly a SOAPlib problem.
Try to connect your url using directly httplib and see what happens.

HTH
"kondor" <kondor at wanadoo.fr> a ?crit dans le message news:
3C600810.4080805 at wanadoo.fr...
> I've download the SOAP package for python and their is an exemple of a
client
> and a server (echoServer.py and echoClient.py)
> It function well on my machine (localhost)
> but when I launch the server on my machine and try the client on an other
> machine (in my network), it dosen't work.
> yet I've change localhost to the server ip adress in the clientEcho.py.
> The client say :
>
> PythonWin 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
win32.
> Portions Copyright 1994-2001 Mark Hammond (MarkH at ActiveState.com) - see
> 'Help/About PythonWin' for further copyright information.
> Traceback (most recent call last):
>    File "D:\Python21\Pythonwin\pywin\framework\scriptutils.py", line 301,
in
> RunScript
>      exec codeObject in __main__.__dict__
>    File "C:\projet\echoClient.py", line 23, in ?
>      print server.echo_ino("moo")
>    File "D:\Python21\lib\SOAP.py", line 3603, in __r_call
>      self.__hd, self.__ma)
>    File "D:\Python21\lib\SOAP.py", line 3521, in __call
>      config = self.config)
>    File "D:\Python21\lib\SOAP.py", line 3397, in call
>      r.putrequest("POST", real_path)
>    File "D:\Python21\lib\httplib.py", line 437, in putrequest
>      self.send(str)
>    File "D:\Python21\lib\httplib.py", line 379, in send
>      self.connect()
>    File "D:\Python21\lib\httplib.py", line 363, in connect
>      self.sock.connect((self.host, self.port))
>    File "<string>", line 1, in connect
> error: (10061, 'Connection refused')
>  >>>
>
>
> Can you help me ?
>




From shalehperry at attbi.com  Tue Feb 26 14:56:10 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Tue, 26 Feb 2002 11:56:10 -0800 (PST)
Subject: Precompiled regular expressions slower?
In-Reply-To: <slrna7nm7u.11s.huaiyu@gauss.almadan.ibm.com>
Message-ID: <XFMail.20020226115610.shalehperry@attbi.com>

> 
> The times of the following three tests are:
> 1.94322395325   # search(string, string)
> 2.62431204319   # search(re_obj, string)
> 0.667925000191  # re_obj.search(string)
> 
> Someone care to explain why the second ends up slower than the first?
> 

 1 def _compile(*key):
 2    # internal: compile pattern
 3    p = _cache.get(key)
 4    if p is not None:
 5        return p
 6    pattern, flags = key
 7    if type(pattern) not in sre_compile.STRING_TYPES:
 8        return pattern
 9    try:
10        p = sre_compile.compile(pattern, flags)
11    except error, v:
12        raise error, v # invalid expression
13    if len(_cache) >= _MAXCACHE:
14        _cache.clear()
15    _cache[key] = p
16    return p

My guess is #2 is slower because of the if type() check on line 7.



From phr-n2002a at nightsong.com  Thu Feb  7 20:16:05 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 07 Feb 2002 17:16:05 -0800
Subject: threads killing other threads
References: <1013105970.135322@cswreg.cos.agilent.com>
Message-ID: <7xk7toixii.fsf@ruckus.brouhaha.com>

weeks at vitus.scs.agilent.com (Greg Weeks) writes:
> I've occasionally wanted thread objects to have an exit() method, so that
> one thread could interrrupt and kill another.  This doesn't seem to be 
> available, perhaps because POSIX threads don't support it.
> 
> Is that correct?
> 
> Is there some reason why threads killing threads would be undesirable or
> hard to implement?

What you really want is a way to let threads raise asynchronous
exceptions in other threads.  There's some discussion of this in

http://sf.net/tracker/?func=detail&aid=502236&group_id=5470&atid=105470


From jason at jorendorff.com  Wed Feb 13 16:30:51 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 13 Feb 2002 15:30:51 -0600
Subject: searching a list of tuples
In-Reply-To: <3c6aabeb.6302984@news.easynews.net>
Message-ID: <HFEKILOLEFEFMKAECNDLIEGLDCAA.jason@jorendorff.com>

Marcus Stojek writes:
> I have a (huge) list of tuples :
> list=(float 1,float 2,float 3,int).
> 
> The list has been sorted with list.sort()
> Now I have to slice the list in a way
> that:
> 
>   a <= float 1 <= b

Well, on my computer I can make a float that
represents "infinity".  So this works:

import bisect

INF = 1e600
left = bisect.bisect_right(mylist, (a, -INF, -INF, -INF))
right = bisect.bisect_left(mylist, (b, INF, INF, INF))
myslice = mylist[left:right]

## Jason Orendorff    http://www.jorendorff.com/



From syring at email.com  Sun Feb 10 15:50:08 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 10 Feb 2002 21:50:08 +0100
Subject: Is Parrot dead?
References: <B88C9617.532F%Roman.Fischer@teleweb.at>
Message-ID: <a46me3$1ch6pm$1@ID-7529.news.dfncis.de>

"Ing. Roman Fischer" <Roman.Fischer at teleweb.at> schrieb im Newsbeitrag
news:B88C9617.532F%Roman.Fischer at teleweb.at...
> For a long time I do not find informations about Parrot (the programming
> language). O Reilly still estimates July 2001 for the "Programming in
Parrot
> in a Nutshell"-book and even Amazon (which is not very shy to offer books
> long before they are available) does not have it in its offer. I also did
> not find any real complete definitions of this language.
>
> So is this language dead?

Uhm, I think you have been the victim of last year's April joke
(http://www.python.org/parrot.html). To my knowledge, Parrot is the name of
the Perl VM incarnation.

Karl M. Syring



From writeson at earthlink.net  Thu Feb 14 12:49:46 2002
From: writeson at earthlink.net (Doug Farrell)
Date: 14 Feb 2002 09:49:46 -0800
Subject: wxTextCtrl and handling button down even in OnChar()
References: <88bc63c6.0202130907.31a13d52@posting.google.com> <ynxa8.209099$KM2.8640420@atlpnn01.usenetserver.com>
Message-ID: <88bc63c6.0202140949.69a08adf@posting.google.com>

Steve,

Thanks for the advice, a good idea that I will implement. 

And you're right, the wxHeads did help me out. The problem was that I
had an event.Skip() call in my event handler, so the event wasn't
'eaten' by my event handler and got handled by the default handler as
well. Essentially causing a button to look like it was hit twice.

Thanks again,
Doug

"Steve Holden" <sholden at holdenweb.com> wrote in message news:<ynxa8.209099$KM2.8640420 at atlpnn01.usenetserver.com>...
> "Doug Farrell" <writeson at earthlink.net> wrote in message
> news:88bc63c6.0202130907.31a13d52 at posting.google.com...
> > Hi,
> >
> > I've written an application that has a control I've derived from
> > wxButton so I can provide my own EVT_BUTTON(self, ID, self.OnClick)
> > event handling. In the method OnClick() I'm getting the button press,
> > but I see both the button down and button up events, which causes my
> > code to run twice. I'd like to filter the button press and only react
> > to the button down event. The trouble is, I can't seem to find the
> > method of event or whatever that will allow me to do this. If any one
> > has some ideas how I can process only moust button down events, I'd
> > really appreciate it.
> >
> Doug:
> 
> I'm sure the wxHeads will reply giving you the answer you seek, this is just
> a cautionary point about user interfaces in general -- it may not apply in
> the specific case you have in mind.
> 
> If you are going to choose a single event to action on a button, it's better
> to use the button UP rather than button Down event. There's nothing worse
> than an interface that won't let you change your mind, so that a little
> unintentional extra pressure on the mouse button is all you need for
> disaster. On the better (more "comfortable") user interfaces you should find
> you can click on a button then change your mind, and move the cursor off the
> button before releasing the mouse button, without selecting any action.
> 
> still-looking-for-dwim-mode-ly y'rs  - steve


From buzzard at urubu.freeserve.co.uk  Mon Feb 25 14:58:42 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Mon, 25 Feb 2002 19:58:42 -0000
Subject: dlg.Destroy() behaviour
Message-ID: <a5e561$q8v$1@news5.svr.pol.co.uk>

Just a quick question about whether or not I need to destroy dialogues
(wxPython) with dlg.Destroy().  When creating a number of wxTextEntryDialogs
in succession only the first dialogue gets the focus.  I need a mouse click
to set the focus on subsequent dialogues.  However, if I remove the usual
'dlg.Destroy()', then subsequent dialogues do get the focus.  Is this
advisable?  If not, what is the neatest way of setting the focus on the
subsequent dialogues?  Thanks in advance.

Duncan Smith




From vlindberg at verio.net  Fri Feb 22 19:33:51 2002
From: vlindberg at verio.net (VanL)
Date: Fri, 22 Feb 2002 17:33:51 -0700
Subject: Request for comments: Search and replace script
References: <XFMail.20020222152043.shalehperry@attbi.com>
Message-ID: <3C76E36F.3030501@verio.net>

Hello, and thanks for looking.

 > tabs, ick, tabs .....

Fixed.  Everything now uses 4 spaces instead.

 > please try to keep a line down to 78 chars

Fixed, at least for most of the code.  I don't know how to deal with 
long strings, tho:
For example, self.writeLog('This is a very long line that has this %s 
substitution and will go longer than 78 chars")

 > exiting in a constructor is not nice, especially when you are 
planning on having
 > others use your code. Just throw an exception.

Fixed.  I only saw that in the FileChecker.  If I did that anywhere 
else, please hit me upside the head.

 > check() makes some of the same checks over and over, perhaps you 
should look at
 > the code's flow here and see if it can be done differently

I am not sure what you mean here.  What I tried to do was to define each 
separate check (binary, openable, etc) in its own function and then have 
one function that would call them all in the correct order.  Is this not 
the best way to do it?

 > globals are also icky

I agree, but I wasn't sure how to have different classes accessing the 
same variable without having the variable be global.  Is there another 
way to do it?

 > why is usage a) a list (it could have been a tuple) and b) defined at 
all. You
 > could just read __doc__ and make it the file's doc string

usage was originally the doc string.  It was so long, however, that I 
broke it up so that only an appoximate screenful would appear at a time. 
 Sort of a poor man's 'more' for when this is run on a windows system.

As for why it is a list, no particular reason.  I just happened to give 
it square brackets instead of parenthesis.

Thanks,

Van






From timr at probo.com  Sun Feb 10 01:23:09 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 09 Feb 2002 22:23:09 -0800
Subject: Help! Using Boa Constructor
References: <8f8ffe67.0202071344.6673ed5d@posting.google.com> <3k276u87qkibb6p773sg318chvh0mupusj@4ax.com>
Message-ID: <ad4c6u8ei16r9n35fssq7clrul9pt3r7qb@4ax.com>

Wolfgang Strobl <ws at mystrobl.de> wrote:
>
>I successfully run it with wxPython 2.3.2.1 + Python21 (works well) and
>wxPython 2.3.2.1 + Python22 (works, but it throws a GPF in python
>during shutdown). 

I also get GPF on shutdown with the same setup.  It doesn't seem to affect
anything.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From skip at pobox.com  Wed Feb 27 20:57:21 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 27 Feb 2002 19:57:21 -0600
Subject: map-like function on dict values?
In-Reply-To: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
Message-ID: <15485.36481.903229.853522@12-248-41-177.client.attbi.com>

    Huaiyu> Is there a way to do this
    Huaiyu> for k, v in dict.items():
    Huaiyu>     dict[k] = f(v)

    Huaiyu> without involving a Python for loop?

Check out the operator module.  In particular, operator.setitem will
probably do what you want:

    import operator
    keys = dict.keys()
    vals = map(f, dict.values())
    map(operator.setitem, [dict]*len(keys), keys, vals)

    Huaiyu> I'd imagine that a function implemented in C could speed things
    Huaiyu> up quite a bit.  

It all depends on how expensive f() is related to the overhead of a for loop
and the cost of getting map ramped up.  Don't rely on your instincts.
Benchmark it.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From robin at jessikat.fsnet.co.uk  Thu Feb 14 07:41:09 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 12:41:09 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
Message-ID: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk>

I am having bad problems with zope not starting up properly. I think
this is caused by having Python-2.1.2 + win32all installed. It seems
that paths set up in the registry by win32all are getting into sys.path
when zope runs. This is causing massive confusion between modules
installed in the base python and those I wish to test with zope.

As an example

I have Python-2.1.2 installed in C:\Python and zope installed in
c:\Python\ZOPE_2_5_0

when I just run the zope binary and test the path I see

C:\Python\ZOPE_2_5_0>bin\python
Python 2.1.2 (#31, Jan 15 2002, 17:28:11) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', 'C:\\Python\\Pythonwin\\pythonwin', 'C:\\Python\\win32',
'C:\\Python\\win32\\lib', 'C:\\Python', 'C:\\Python\\ZOPE_
2_5_0\\bin\\DLLs', 'C:\\Python\\ZOPE_2_5_0\\bin\\lib', 'C:\\Python\\ZOPE
_2_5_0\\bin\\lib\\plat-win', 'C:\\Python\\ZOPE_2
_5_0\\bin\\lib\\lib-tk', 'C:\\Python\\ZOPE_2_5_0\\bin']
>>> sys.executable
'C:\\Python\\ZOPE_2_5_0\\bin\\python.exe'
>>>

the paths 'C:\\Python\\Pythonwin\\pythonwin', 'C:\\Python\\win32',
'C:\\Python\\win32\\lib', 'C:\\Python' come from the registry. and
clearly can cause confusion.


Clearly I want to test Zope, but unless I remove these registry entries
I'm having problems. The registry entries are useful for COM objects so
I don't particularly want to remove them.
-- 
Robin Becker


From no at spam.com  Tue Feb 26 14:21:48 2002
From: no at spam.com (Tokomar)
Date: Tue, 26 Feb 2002 19:21:48 GMT
Subject: Tkinter list boxes, selection emphasis, and the keyboard focus
References: <K0Qe8.48006$BA5.2207030344@newssvr14.news.prodigy.com>
Message-ID: <gfRe8.6453$c15.3356020763@newssvr21.news.prodigy.com>

Tokomar wrote:

> Python fans:
> 
> First, some definitions...
> 
> Focus - the single place on entire screen where keyboard activity will
> reflect. This displays either a text caret or an ant rectangle.
> 
> Selection - in a list box, which item is the user's favorite. This
> displays by changing the color scheme to an opposing contrast.
> 
> Now the problem. Tkinter appears to slur these definitions together. If,
> for example, I put two list boxes next to each other, they treat the focus
> and selection the same. If the keyboard focus is in the second one, we
> can't see which item in the first one is selection.
> 
> I will deal, short term, by changing some other property in the lists. (I
> will therefor do Tkinter's housekeeping for it.)
> 
> But has anyone addressed this problem directly? Is there a configuration
> option I overlooked?

It appears that "activate" does not work, and there are list server posts 
lamenting it out there.

Further, though there is an "itemconfigure" for Listbox, my Python 2.0 does 
not have it.

I am open to any suggestion for a workaround!

-- 
Tokomar


From jwilhelm at outsourcefinancial.com  Wed Feb 27 10:56:20 2002
From: jwilhelm at outsourcefinancial.com (Joseph Wilhelm)
Date: 27 Feb 2002 08:56:20 -0700
Subject: Editors for Python
In-Reply-To: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <1014825380.17135.2.camel@jwilhelm.ofsloans.com>

On Mon, 2002-02-25 at 09:18, Moray Taylor wrote:
> Hi,
> 
> I am desperatly seeking a good python editor, so far JEdit is the best
> I have found, but since it is written in Java, I find it painfully
> slow, and PyBrowse, although good, a little flaky.
<snip>

Out of curiosity, have you tried jEdit 4.0? I just installed this myself
after suffering through the crippling speed of 3.x for too long, and
I've found it to be incredibly faster. It's using Java 2 which improves
the speed, and has been much improved itself so that it's actually quite
a lot more usable now. If you haven't tried it out, I would highly
recommend it.

--Joseph Wilhelm




From jkraska at san.rr.com  Tue Feb  5 19:04:24 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 06 Feb 2002 00:04:24 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk>
Message-ID: <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com>

>Why do you want this? Isn't it just an implementation detail?
>I can see advantages in having it implemented as a preprocessor --
>you could look at the preprocessor output and see what Python is 
>generated. This might help with debugging complex macros.

Well, in systems where the macro system is native, they are
often even easier to debug. C++ templates, for example. While
early renditions were as difficult to debug as macros, macros
had to be hand pre processed. By your macro system really being
a first-class form which implements limited macro semantics,
you end up with first-class error handling of the macro itself.
Errors in macro definition are reported as the appropriate
kinds of exceptions, instead of through secondary effects, like
bizarre cascades of syntax errors.

I believe I can do this natively and do it correctly, which has
lots of value points.

>What if a macro needs to create local variables in its expansion?

This is perfectly legitimate. What's not legitimate is...

	deflingo myterm (i, codeblock):
		j=3

...where j=3 modifies the variable "j" in the context of the caller
if this variable is preexisting in the caller. That's non hygienic
behavior. Past experience with macros has shown this to be bad
behavior; hence there is a design in the community for good "hygiene,"
as it were.

>This would probably make more sense to me if you gave an example.

Yes; will observe now that I'm guilty of having selected deliberately
trivial macros. The point wasn't to demonstrate _utility_, I wanted to
show syntax. This appears to have been a mistake. :)

>>	invariant:
>>		dosomething()

Entire block of code delimited by ':' INDENT codeblock DEDENT is
evaluated in some heretofor undefined way by the invocation of the
"invariant" lingo. Note well that the block of code can be any
series of valid python expressions. I'm not sure how this particular
form would be made to be _useful_.

(GEEZE! PESTER ME WITH DETAILS WILL YA?!!) :) :)

>Why not factor(17) ?

This was a bad example. It would have been more useful to show...

	incf x

Details of the python lexer/parser might force it to be...

	$incf x

...however.

>IMO it is more Pythonic to say:
>
>   do:
>      dosomething()
>      while i <10                                                                                        

Correct. I'm not quite sure how to handle that one.

But of course macros themselves significantly erode simplicity, making
them _per se_ unpythonic, wouldn't you say? :)

>I think Python could benefit from a macro system. But the devil's
>in the details.

One of those nasty details would be collecting _sub_ macro forms separately
from parent macro forms. Note that this is distinctly different than _peer_
macro forms. Although I suppose that tree walking the ast at a later date
could do the proper manipulation. I'm having acid with that, but it's doable.

I see some difficulty (but not impossibility) of processing:

	do:
		dox()
		until i < 10
		doy()

Further, a moment of stupidity illustrated very clearly why macro systems
are dangerous. It would be _impossible_ to cleanly implement a macro which
overrode a preexisting keyword like that; at least not without a special
delimiter. In the above example I corrected the form to "until".

Egg on my face. :)

C//



From rsteiner at visi.com  Wed Feb 13 15:08:32 2002
From: rsteiner at visi.com (Richard Steiner)
Date: Wed, 13 Feb 2002 14:08:32 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org>
 <130220020751360934%jwbaxter@spamcop.net>
 <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net>
 <yuqa8oHpvCEW092yn@visi.com>
 <_dya8.209628$KM2.8685845@atlpnn01.usenetserver.com>
Message-ID: <Afsa8oHpvyub092yn@visi.com>

Here in alt.folklore.computers,
"Steve Holden" <sholden at holdenweb.com> spake unto us, saying:

>Ah me, ah my, FastRand drums.

I've never seen one, but I know of a web page.  :-)

  http://www.fourmilab.ch/documents/univac/fastrand.html

-- 
 -Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Eden Prairie, MN
   OS/2 + BeOS + Linux + Solaris + Win95 + WinNT4 + FreeBSD + DOS + PC/GEOS
            + Fusion + vMac + Executor = PC Hobbyist Heaven! :-)
              Ever notice how poems in pig-latin always rhyme?


From cimarron+google at taylors.org  Wed Feb 20 14:32:04 2002
From: cimarron+google at taylors.org (Cimarron Taylor)
Date: 20 Feb 2002 11:32:04 -0800
Subject: Question on if, while, for
References: <e959e7b.0202110545.3dff102@posting.google.com> <7xd6zcw19t.fsf@ruckus.brouhaha.com> <3C73C1E0.D05F7AC0@inenco.no>
Message-ID: <29e28c51.0202201132.273e1652@posting.google.com>

> And to spoil all the fun, here is a solution
> 
> >>> def gcd(a, b):
> ...  while a > 0:
> ...   print 'a = %d, b = %d' % (a, b)
> ...   rest = b % a
> ...   b = a
> ...   a = rest
> ...  return b
> 
> - hans kristian -

I prefer the definition found in Python22/Lib/test/test_binop.py:

def gcd(a, b):
    """Greatest common divisor using Euclid's algorithm."""
    while a:
        a, b = b%a, a
    return b

Cim


From ahmebah at hotmail.com  Mon Feb  4 08:26:21 2002
From: ahmebah at hotmail.com (ameoba)
Date: 4 Feb 2002 13:26:21 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
Message-ID: <Xns91AB378D110ABahmebahhotmailcom@198.99.146.10>

"Timothy Rue" <threeseas at earthlink.net> wrote in
news:922.799T2387T11875593threeseas at earthlink.net: 

> Its damn clear that there is a need to remove the sort of people posing
> to be professional who are in essence just good liars, abstraction
> manipulators, from the industry.

WTF are you talking about?  The only thing that I've managed to come up 
with about your project is that it -heavily- abstracts the process of 
computing to the point where joe-6pack should be able to make a computro do 
anything.

If you feel there is something wrong with being an 'abstraction 
manipulator' perhaps you shouldn't be talking to computer programmers.  
Computing is nothing but the building and manipulation of abstractions.  If 
it weren't for abstractions, then we'd still be swapping vaccuum tubes to 
program a computer.



Don't even get me started on the hipocrasy of -you- using 'abstraction 
manipulator' as an insult.


From fgeiger at datec.at  Mon Feb 25 07:36:34 2002
From: fgeiger at datec.at (F. GEIGER)
Date: Mon, 25 Feb 2002 13:36:34 +0100
Subject: Remote debugging ?
References: <mailman.1014622419.18967.python-list@python.org>
Message-ID: <a5db3k$1p0$1@rex.ip-plus.net>

> > It must not be freeware.
>
> What does that mean?

In German it reads "Es muss nicht Freeware sein". In English this should
read "It does not need to be freeware".

Cheers
Franz


"Brian Quinlan" <brian at sweetapp.com> schrieb im Newsbeitrag
news:mailman.1014622419.18967.python-list at python.org...
> Lothar wrote:
> > Is there any debugger for python that allows me to do remote debugging
> > from a windows 2000 client to a python script running on a Linux
> > Server.
>
> ActiveState's Komodo IDE features remote Python debugging:
> http://www.activestate.com/Products/Komodo/
>
> > It must not be freeware.
>
> What does that mean?
>
> Cheers,
> Brian
>
>




From eppstein at ics.uci.edu  Tue Feb 12 23:23:41 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Tue, 12 Feb 2002 20:23:41 -0800
Subject: Why no documentation of basic data structure performance?
References: <mailman.1013571621.14946.python-list@python.org>
Message-ID: <eppstein-AAD3B4.20234112022002@news.service.uci.edu>

In article <mailman.1013571621.14946.python-list at python.org>,
 <brueckd at tbye.com> wrote:

> While the info you're talking about can be valuable in some cases, in most
> cases it just doesn't matter, for example, which list operation you use -
> and putting that sort of info in the main documentation encourages us to
> think about optimization when we shouldn't. For example, if I'm not paying
> attention I catch myself doing this too much:
> 
> meth = obj.method
> for item in seq:
>   meth(item)
> 
> Why? Because it saves a lookup to obj.method. But unless I'm trying to
> remove a specific bottleneck I'm solving a non-problem and cluttering my
> code at the same time. Shame on me!

I completely agree about not micro-optimizing until it's necessary,
I was more concerned with situations where the difference between two ways 
of doing things could be much larger (in the case I mentioned, constant 
time vs linear time) and nonobvious (identical wording for the two tutorial 
examples).

> Somebody had a Python optimization page I found useful (was it Skip?) -
> maybe you could report your findings to him/her and have the page updated.
> IMO, that's where this sort of info belongs - on a web page or document
> that we can find when we need it but just far enough out of the way that
> we don't spend too much time looking at it.

Separate documentation per-implementation, somewhere easy to find, sounds 
like a reasonable choice -- another would be to have a few paragraphs here 
and there in the tutorial discussing these issues in the context of the 
major implementations.  Anyway, though, I'm not sure calling it 
optimization matches what I'm thinking of.  My feeling is that choosing the 
right algorithms and data structures is better done before you code, at 
design time, rather than as an optimization after you code -- otherwise you 
may find yourself locked into a solution that's too inefficient or 
inflexible and only able to make minor changes such as the method lookup 
lifting you discuss above.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From phr-n2002a at nightsong.com  Wed Feb 20 00:49:58 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 21:49:58 -0800
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org> <7xd6z07odl.fsf@ruckus.brouhaha.com> <e6c67uonjofbhk711hqs58lhd1ocnqf669@4ax.com>
Message-ID: <7xofik8zy1.fsf@ruckus.brouhaha.com>

Courageous <jkraska at san.rr.com> writes:
> >They're spending quite a lot of $$ running those banner ads and I
> >doubt if they have that many openings, so they must be looking for
> >"perfect" candidates.  I'm sure they get quite a few responses.
> 
> Maybe, maybe not. There are a lot of highly imperfect ways to get to
> a candidate's programming skills. Any kind of in- house testing runs
> them afoul of federal labor laws, so turning it upside down into a
> programming contest has appeal.  Likewise, you're pulling from a
> crowd of self-styled enthusiasts.  The approach has a big
> upside. They're likely awash in good candidates. "Perfect"? One
> doesn't have to be perfect if one is a motivated hacker. One is
> class-A material in this case, a good deal of the time.

Well, I'm sure they're getting some very good candidates through that
contest, and that they've turned away people that any project I've
worked on in the past would have been happy to have.  It's quite
clever, and the puzzles were fun to work on.  


From logiplexsoftware at earthlink.net  Thu Feb 14 12:09:40 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 14 Feb 2002 09:09:40 -0800
Subject: Recommendations for cross-platform GUI Bindings ?
In-Reply-To: <xBMa8.217$B92.52511@news.xtra.co.nz>
References: <xBMa8.217$B92.52511@news.xtra.co.nz>
Message-ID: <20020214090940.601e46d8.logiplexsoftware@earthlink.net>

On Thu, 14 Feb 2002 23:47:40 +1300
Steven M. Castellotti wrote:

> Greeting all--
> 
> 	I'm writing an Open Source software application that allows children to
> create their own video games, and would like to hear anyone's
> recommendations for GUI bindings. The key things I'm looking for are:

wxPython.  You can freeze it using Gordon McMillan's Installer.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From jeff at ccvcorp.com  Fri Feb 15 14:24:24 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 15 Feb 2002 11:24:24 -0800
Subject: help with silly algorhythm
References: <5e8bd451.0202142155.143ca4d9@posting.google.com> <3c6ce21d.190762859@news.isar.de>
Message-ID: <3C6D6067.F4AC05A4@ccvcorp.com>


Gerson Kurz wrote:

> On 14 Feb 2002 21:55:27 -0800, kp87 at lycos.com (kevin parks) wrote:
>
> >What I want to do is generate a new list of items based on
> >an input list and a some criteria (list of acceptable values, for
> >example).
>
> An ideal job for lambda! Lets start with the data.
>
> > * list of input values =[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
> > * a list of acceptable new 'partner values' = [1, 3, 7, 10]
>
> input_list = range(12)
> partner_values = [1,3,7,10]
>
> > so we might get (in mode 1: all closest higher value)
> > [ [0, 1], [1, 3], [2, 3], [3, 7], [4, 7], [5, 7], [6, 7], [7, 11]....
>
> GetUpper = lambda i,p:filter(lambda x:x[1],[(y,(filter(lambda
> x:x>y,p)+[None])[0]) for y in i])
>
> In typical Python fashion the code is obvious.

Heh.  I positively cannot make heads or tails of that lambda--perhaps its a
shortcoming of mine, but extended lambdas like that make my eyes glaze and my
brain shut off....

The trick here, Kevin, is that you want to select from a list, based on a
parameter and a function.  If you can define functions that'll do what you
want, then you can use a simple list comprehension (or map() ) to process it--

inputlist = range(12)
partnerlist = [1,3,7,10]
mixed = [ (i, func(i, partnerlist)) for i in inputlist ]

This generates a pair (i, f(i) ) for each of your input values.  Now, you just
need to determine how to write f(i) for each of your "modes".  So, for
instance:

def NextHigher(key, choices):
    templist = choices[:] #make a copy
    templist.sort()    # make sure it's sorted
    for value in templist:
        if value > key:
            return value
    # If key >= max(choices), then you
    # want it to wrap around?
    return templist[0]

Similarly for next lower...

def NextLower(key, choices):
    templist = choices[:]  # make a copy
    templist.sort()    # sort it
    templist.reverse()    # then reverse it
    for value in templist:
        if value < key:
            return value
    # or, if key <= min(choices)
    return templist[0]

Given these functions, you can plug them into that list comp I gave above:

>>> higher = [ (i, NextHigher(i, partners)) for i in inputlist ]
>>> higher
[(0, 1), (1, 3), (2, 3), (3, 7), (4, 7), (5, 7), (6, 7), (7, 10),
(8, 10), (9, 10), (10, 1), (11, 1)]
>>> lower = [ (i, NextLower(i, partners)) for i in inputlist ]
>>> lower
[(0, 10), (1, 10), (2, 1), (3, 1), (4, 3), (5, 3), (6, 3),
(7, 3), (8, 7), (9, 7), (10, 7), (11, 10)]
>>>

As you can see, all you need to do is define a function that will find the
appropriate value of your partnerlist, for a given single input.  You *do* have
to test against each member of the partnerlist, until you find a match, though.

Hope this helps.  :)

Jeff Shannon
Technician/Programmer
Credit International






From alexw at fast.no.nospam  Thu Feb 14 05:34:42 2002
From: alexw at fast.no.nospam (Alexander Wilkens)
Date: Thu, 14 Feb 2002 11:34:42 +0100
Subject: Problem freezing Win32 services
Message-ID: <a4g3s3$k0s$1@news.ost.eltele.no>

Hi,

I'm attempting to freeze a Windows NT service (in Win2000) written in Python
2.2 using win32all 142 and the freeze util bundled with Python in the tools
directory, and I've encountered some problems. Freezing console apps works
fine.

The following is the link error I get when using the '-s service' option to
freeze:

frozen.obj : error LNK2001: unresolved external symbol _PythonService_main
NodeControllerWin32Service.exe : fatal error LNK1120: 1 unresolved externals
NMAKE : fatal error U1077: '"E:\Apps\Microsoft Visual
Studio\VC98\bin\link.exe"' : return code '0x460'
Stop.

I've tried freezing it without the '-s service' option set in which case I
get an executable that runs, but doesn't work as a service. I can do
'myservice.exe install' on it which seems to work, but if I then do
'myservice.exe debug' I get an importerror on 'myservice'...

Returning to the '-s service' option; During freeze I get a lot of errors
about DSP files not found. I found that these could be ignored when freezing
console apps as I just use the corresponding DLLs instead. When freezing the
service I get an additional error:

e:\apps\python\tools\freeze\%PYTHONEX%\win32\cvs\PythonService EXE.dsp:
[Errno 2] No such file or directory: 'e:\\apps\\
python\\tools\\freeze\\%PYTHONEX%\\win32\\cvs\\PythonService EXE.dsp'
No definition of module servicemanager in any specified map file.

If I fix this by setting the correct path to the DSP file I get:

LINK : fatal error LNK1104: cannot open file "pywintypes.lib"
NMAKE : fatal error U1077: '"E:\Apps\Microsoft Visual
Studio\VC98\bin\link.exe"' : return code '0x450'
Stop.

I now compile pywintypes using the it's DSP and correct the path to the
pywintypes DSP file as well. I now end up with the following mix of errors
and warnings:

PyACL.cpp
e:\Apps\Python\win32\src\PyACL.cpp(13) : warning C4273: 'PyWinMethod_NewACL'
: inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(22) : warning C4273: 'PyWinObject_AsACL'
: inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(39) : warning C4273: 'Initialize' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(51) : warning C4273:
'AddAccessAllowedAce' : inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(70) : warning C4273: 'AddAccessDeniedAce'
: inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(89) : warning C4273: 'GetAclSize' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(102) : warning C4273: 'GetAceCount' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(116) : warning C4273: 'GetAce' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(149) : error C2491: 'PyACLType' :
definition of dllimport data not allowed
e:\Apps\Python\win32\src\PyACL.cpp(172) : warning C4273: 'PyACL::PyACL' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(183) : warning C4273: 'PyACL::PyACL' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(192) : warning C4273: 'PyACL::~PyACL' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(197) : warning C4273: 'getattr' :
inconsistent dll linkage.  dllexport assumed.
e:\Apps\Python\win32\src\PyACL.cpp(202) : warning C4273: 'deallocFunc' :
inconsistent dll linkage.  dllexport assumed.
NMAKE : fatal error U1077: 'cl' : return code '0x2'
Stop.

What now? :) Am I on the right track at all here? If anyone could shed some
light on this I would really appreciate it. Thanks!

--
Alex





From quinn at schilling.ugcs.caltech.edu  Wed Feb 20 13:36:01 2002
From: quinn at schilling.ugcs.caltech.edu (Quinn Dunkan)
Date: 20 Feb 2002 18:36:01 GMT
Subject: Idiots guide to code objects, frame objects and settrace?
References: <VoAc8.7320$k91.367319@wards>
Message-ID: <slrna77r4h.dg2.quinn@schilling.ugcs.caltech.edu>

On Tue, 19 Feb 2002 22:28:50 -0000, Tim Rowe <tim at eltig.freeserve.co.uk> wrote:
>Is there an idiots guide to code objects, frame objects and settrace
>anywhere? I want one program to monitor the execution of another. I've
>looked at coverage.py from the Vaults of Parnassus, but can't make enough
>sense of it to adapt it to do what I want.

A combination of the langref and interactive mucking around worked just fine
for me.  They're not very complicated.  The only possibly confusing bit is that
tb_next points downstream and f_back points upstream... or is it the other way
around?

It might not be an "idiot's" guide... good thing you're not an idiot, huh?


From jkraska at san.rr.com  Thu Feb  7 02:34:15 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 07:34:15 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org>
Message-ID: <asb46u83u9fcut1pdioagpl7479ang8m37@4ax.com>

>All in favor, post AYE.  All opposed, post NAY.

Yes. Rue needs to meet the real world. Perhaps he
should code and shutup for a good year.

C//



From martin.franklin at westgeo.com  Mon Feb 18 10:04:36 2002
From: martin.franklin at westgeo.com (Martin Franklin)
Date: Mon, 18 Feb 2002 15:04:36 +0000
Subject: tab order for Tkinter widgets
References: <mailman.1014040229.14186.python-list@python.org>
Message-ID: <a4r54b$2dq$1@mail1.wg.waii.com>

Laura Creighton wrote:

> I thought that the order in which you tabbed through widgets was the
> order in which you gridded them (assuming you are using the grid manager).
> I have a bunch of widgets that know what row and what column they are
> to be gridded in, but I cannot guarantee what order I can get them out
> of the database.
> 
> So in the middle of a loop I have:
> 
> decorated.append((d['managerArgs']['row'],
>                           d['managerArgs']['column'],
>                           d['managerArgs'], widget ))
> 
> then at the bottom I can do
> 
>         decorated.sort()
>         for junk, junk, args, widget in unordered:
>             widget.grid(args)
> 
> They are all gridding nicely.  The sort works great as well.  And
> tabbing gets you randomly flopping from one entry field to another
> same as always.  Clearly there is something else I need to do to get
> my widgets to tab in the correct order.  What?
> 
> Thanks very much,
> Laura Creighton
> 
> 

Laura,

This simple example may help??

from Tkinter import *
root=Tk()

edict={}


top_simple=Toplevel()
top_simple.title('Simple.....')

for row in range(10):
    for col in range(10):
        edict[row, col]=(row, col)
        e=Entry(top_simple)
        e.grid(row=row, col=col)
    
    
# now get the coords out 'randomly'.....
root.title('Random')    

for row, col in edict.keys():
    e=Entry(root)
    e.grid(row=row, col=col)
    
    
# this time sort the 'keys' then grid em!
top_sorted=Toplevel()
top_sorted.title('Sorted')

keys=edict.keys()
keys.sort()
for row, col in keys:
    e=Entry(top_sorted)
    e.grid(row=row, col=col)
    
    
    
    
root.mainloop()




























From sameer_ at email.com  Sat Feb 16 05:55:51 2002
From: sameer_ at email.com (sameer)
Date: 16 Feb 2002 02:55:51 -0800
Subject: distutils/package installers on windows
Message-ID: <6bd9f01b.0202160255.77244c9e@posting.google.com>

Here's the situation:
I have a python package (or two) to be installed on a windows machine
(for now) which has python installed.  I am using distutils to create
a windows installer which installs my python packages just fine.  Now,
I want to also create some shortcuts on the Desktop as well as the
Start Menu.  As of now, distutils doesn't allow me to do either of
those things.  What are my options here to do either or both of these
things.  Can the distutils package be modified easily to provide this
kind of support on windows?  Or is there another python package
available that would allow me to create the desktop and Start Menu
shortcuts?


From jason at jorendorff.com  Sat Feb 16 23:07:31 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 22:07:31 -0600
Subject: Have to try
In-Reply-To: <004501c1b763$78bca9c0$b278c53f@pacbell.net>
Message-ID: <HFEKILOLEFEFMKAECNDLMEBPDDAA.jason@jorendorff.com>

Rick Hamilton writes:
> I want to search that string for the occurrence of a sub string.
> I have tried 
> Find(s,sub)
> Rfind
> MyString.find(sub)

It looks like you're using an extra-old version of Python,
because mystring.find(sub) really should work.
  http://www.python.org/doc/current/lib/string-methods.html

>>> x = "lovely spam, wonderful spam!"
>>> x.find('spam')
7
>>> x.find('food')
-1

So you must be using Python 1.6 or earlier.  Try this.
  http://www.python.org/doc/current/lib/module-string.html

>>> import string
>>> string.find(x, 'spam')
7

## Jason Orendorff    http://www.jorendorff.com/



From wurmy at earthlink.net  Thu Feb 14 14:45:34 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Thu, 14 Feb 2002 19:45:34 GMT
Subject: Newbie: Object assignment return value
References: <e81be8b2.0202141051.157160d6@posting.google.com>
Message-ID: <3C6C1404.22615AC4@earthlink.net>

Dean Goodmanson wrote:
> 
> (I think) I'm trying to understand if object assignment has a return value.
> 
> Scenario A:
> >>> str( spam= 5 )
> ''

I don't know how you got this to work. In 2.1 I get:

>>> str(spam=5)
Traceback (most recent call last):
  File "<pyshell#0>", line 1, in ?
    str(spam=5)
TypeError: str() takes no keyword arguments
>>> 

and in 2.2:

>>> str(spam=5)
Traceback (most recent call last):
  File "<pyshell#0>", line 1, in ?
    str(spam=5)
TypeError: 'spam' is an invalid keyword argument for this function
>>> 

> >>> str((spam=5))
>   File "<stdin>", line 1
>     str((spam=5))
>             ^
> SyntaxError: invalid syntax

f(a=1) for a function f, is treated like f is called with
keyword argument a. That's why str(spam=5) didn't yield a
*syntax* error. When you put parentheses around it, it
cannot be mistaken for a keyword argument anymore, and 
you end up with passing a binding (or "assignment" as
some call it, although this isn't really correct) as a
parameter to the function, which is simply not valid
syntax.

> >>> #  continued : Scenario B:
> >>> print spam
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> NameError: name 'spam' is not defined

Even if your str(spam=5) succeeds, it does not do what
you expect. This is not an "assignment" passed to a function.
It's a keyword argument, and no new variable/name is
created (unless the function happens to do some nasty
hackery, but let's not go there).

To answer your original question: no, "assignment" does
not return a value. You can chain them, but that's as good
as it gets:

>>> a = b = c = 42
>>> a, b, c
(42, 42, 42)
>>> 

but any other things you might be familiar with (from,
for example, C) won't work:

>>> a = (b = 12)
SyntaxError: invalid syntax

>>> if x = "foo": print "bar"
SyntaxError: invalid syntax

Note that this does prevent you from the common mistake
made in C, writing if (x = y) where you really meant
if (x == y). <smile>

> B: What happend to my spam?
> Espcially when this works:
> >>> str(int())
> '0'
> >>> str((int()))
> '0'
> >>>

These are different cases... int() is an expression and
can be passed to str just fine.

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From Alessandro  Thu Feb 21 03:58:36 2002
From: Alessandro (Alessandro)
Date: Thu, 21 Feb 2002 08:58:36 GMT
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>
Message-ID: <3C74B595.30905@nokia.com>

Tere!

have you seen the stuff at http://www.python.org/doc/Comparisons.html ?

Cheers,

mac



From Frederik.VandeCasteele at NOSPAM.yucom.be  Tue Feb  5 05:39:21 2002
From: Frederik.VandeCasteele at NOSPAM.yucom.be (Frederik VdC)
Date: Tue, 05 Feb 2002 11:39:21 +0100
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com>
Message-ID: <3C5FB659.22B78476@NOSPAM.yucom.be>

I have to call my pythonscripts name.cgi in stead of name.py
on the (unix)server. Maybe that can help you

Marc Laymon wrote:

> I am just learning python.  I have written a few simple scripts which
> run fine when invoked using "python myscript.py".  The Python manual
> says you can execute a Python script directly In Unix by making the
> first line "#!/usr/bin/python", then giving the command
> "./myscript.py".  However, when I try this, I get an error "Command
> not found.".  I get the same error under both Linux and Solaris.
> Using the same format for perl works fine.  Why doesn't the
> "#!/usr/bin/python" line work for Python ?  (perl and python are both
> in /usr/bin on my Linux system.)
>
> Thanks.
>
> Marc Laymon



From peter.milliken at gtech.com  Sun Feb  3 16:42:58 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 4 Feb 2002 08:42:58 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com><Wls68.14178$om6.357057@news1.tin.it> <mailman.1012590723.7301.python-list@python.org>
Message-ID: <a3kb3p$se44@news1.gtech.com>

G'Day Cliff,

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
news:mailman.1012590723.7301.python-list at python.org...
> On Fri, 01 Feb 2002 08:29:42 GMT
> Alex Martelli wrote:
>
> > Peter Milliken wrote:
> >         ...
> > > Actually, if you are *good* at what you do, then code review, whilst
> still
> > > beneficial, is not nearly *that* beneficial. Things like code review
> (IMO
> > > :-)) is there for the 25-50% of programmers who just plain shouldn't
be
> in
> > > the industry - what they generate is just absolute garbage - code
> review
> > > will hopefully stop the code from getting into Unit test and beyond
the
> > > point of redemption :-).
> >
> > I disagree.  Code review, or even better the "continuous ongoing code
> > review" that's such a substantial part of the benefits of pair
> programming,
> > is in my experience just as beneficial with the best coders: without it,
> > besides bugs, it's far too much a temptation to pepper the code with
> > clever tricks, individual style quirks, unexamined (subconscious)
> > assumptions about inputs and environmental conditions, and the like.
> > The code coming out of a good review, or a good-synergy pair, is
> > thus of much higher quality and more valuable than the results of a lone
> > coding session of even a very good coder.
>
> I couldn't agree more.  I worked for a while with another programmer who
> was very good and I know while we were working together it seemed as if
the
> whole of our efforts exceeded the sum of the parts.  This was very likely
> due to the fact that while we both agreed on general principles, our skill
> sets and backgrounds in programming were very different, so we tended to
> each see things the other did not.
> Additionally, because programming is a pure mental excercise (if you
> discount the typing), psychology and morale play a very large role in
> productivity for a programmer and having a second like-mind (even if only
> to bounce ideas off of) can have a huge positive effect on that aspect.
>
> I much prefer the "continuous ongoing code review" idea better than a
> formalized procedure (if I understand what you mean by it correctly).
> Working in pairs provides this as a side-effect rather than having a
formal
> procedure for code-review and seems less time-consuming overall.
>

Not sure what you mean by "continuous ongoing code review". I used to be a
"team leader", I have "looked after" groups of programmers between 1985 and
2000, bitter experience taught me to "review early and often" i.e. diivide
the programmer's work in small units and as each unit is completed review
it - never, ever, ever let the stuff build up to the point where after an
unfavourble review you are faced with the decision to either miss schedule
because it should be re-written or forge ahead regardless and hope that it
will come good through some miracle of extra effort on some ones part :-).
There are so many what I will class as "poor" programmers in the industry,
that this is a necessary technique when you have to "manage" other
programmers. So I assume that "continuous ongoing code review" is similar to
my idea of "early and often"? :-)

<snip>

>
> Actually, when I say "art", I am using it in the sense of "skill", as in
> "martial art" or "art of design" versus the broader sense that includes
> unrestrained personal interpretation.  In this sense "art" means taking
> basic, established techniques and internalizing them to the point where
> they are second-nature. At that point, personal expression becomes an
> extension of those techniques - not a deviation from them.  My experience
> has been that programmers at this level feel a sense of pride and personal
> accomplishment and _appreciate_ code review (otherwise, who will ever see
> and appreciate their accomplishment?).  I suspect this to be a major
> driving force behind open-source software - it's the equivalent of a
public
> demonstration of one's skills.
>

I think you are correct about the open source motivation etc, but in any
typical project/company I have worked for, it is *relatively* rare to find
people who welcome a review of their code. The people you find in Open
Source stuff are the people who view programming as a "joy" (hobby) rather
than a "job". These people, in my experience, tend to be better programmers
(probably because they really do "care" and have pride in their work? :-)).
But the majority of people I have worked with over the years forget about
programming the minute they step out the corporate doors, it is nothing more
than a job to them - very sad :-). These people tend to be the people who
view their work as "Art" rather than increasing their skills as an "art" :-)
They do no external reading, they are not interested in seeing how other
people have solved similar problems etc etc.

Peter


> This is not to say that radical departures can never be beneficial, but
> these departures are rare and do require intense peer review.
>
> --
> Cliff Wells
> Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308
> (800) 735-0555 x308
>
> "Then with your new power you'll accomplish all sorts of cool stuff
>  in no time, and We'll All Be Sorry.  At that point you can either
>  gloat a bit, and then relent, or go ahead and send the robot army
>  after us." - Quinn Dunkan
>




From llothar at mailandnews.de  Mon Feb 25 03:59:38 2002
From: llothar at mailandnews.de (Lothar Scholz)
Date: Mon, 25 Feb 2002 09:59:38 +0100
Subject: Remote debugging ?
References: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com>
Message-ID: <j6vj7us5nq60du0o1niljv4abjlpcod0cj@4ax.com>

On Mon, 25 Feb 2002 08:34:22 +0100, Lothar Scholz
<llothar at mailandnews.de> wrote:

>Is there any debugger for python that allows me to do remote debugging
>from a windows 2000 client to a python script running on a Linux
>Server.
>
>It must not be freeware. 

And i forgot to say that the debugger should support debugging
threads.


From wolfgang.grafen at gmx.de  Thu Feb 28 06:05:51 2002
From: wolfgang.grafen at gmx.de (Wolfgang Grafen)
Date: 28 Feb 2002 03:05:51 -0800
Subject: Regular expression problem
References: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com> <3C7D85BC.71550E5F@gmx.de> <nxhf8.149057$dh.38723170@typhoon.nyroc.rr.com>
Message-ID: <2e8c61fb.0202280305.3cf3db9d@posting.google.com>

Asheesh Laroia <pan-news at asheeshenterprises.com> wrote in message news:<nxhf8.149057$dh.38723170 at typhoon.nyroc.rr.com>...
> This is great, thanks!
> 
> Only one problem.  I'm having trouble (I did give it a try) making the
> following work:
> 
> 	<@Trap Body text:>Useful Text
> 
> I need to still be able to extract "Useful Text", not delete it.
> 
> Thanks again!
Hope this is it:

rc=re.compile("<@Trap Body text\s*"
                          "(?:(?P<assigned>=)|(?P<unassigned>))\s*"
                          "(?P<UselessText>(?:<.*?>[^<>]*)*[^>]*>)\s*"
                          "(?P<UsefulText>.*?)?\s*\Z",
                          re.MULTILINE|re.DOTALL).match

t1='<@Trap Body text>'

t2='<@Trap Body text=<FONT "Times">'

t3="""<FONT "Times"><CCOLOR\n   "Black"><SIZE
11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n
58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST
19.2><G+BEFORE
0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
"ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
25>>"""

t4='<@Trap Body text=<FONT "Times"> hello <otto>> and the useful rest'

and you get
>>> rc(t1).groups()
(None, '', '')

>>> rc(t2).groups()
('=', None, '<FONT "Times">')

>>> rc(t3).groups()
('=', None, '<FONT "Times"><CCOLOR\012  
"Black"><SIZE\01211><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE
70><C+SIZE\012\01258.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE
0><CNOBREAK\0120><CLEADING -0.05\012  ><GGRID 0><GLEFT 0><GRIGHT
0><GFIRST 19.2><G+BEFORE\0120><G+AFTER 0><GALIGNMENT \012 
"justify\012  "><GMETHOD "proportional"><G&\012"ENGLISH"><GPAIRS 4><G%
120><GKNEXT 0><GKWIDOW \012  1><GKORPHAN\012\0121><GTABS
$><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0\01225>')

and just to get useful text:
t4 = t1 + t1
t5 = t2 + t2
t6 = t3 + t3

>>> rc(t4).groups()
(None, '', '>', '<@Trap Body text>')

>>> rc(t5).groups()
('=', None, '<FONT "Times">>', '<@Trap Body text=<FONT "Times">>')

>>> rc(t5).groups()
...


def parse_it(text):
    t = rc(text)
    assert t,"Text does not match!:\n%r" % text
    return t.group('UsefulText')

>>> print parse_it(t5)
<@Trap Body text=<FONT "Times">>

wolfgang.

> -- Asheesh.
> 
> On Wed, 27 Feb 2002 20:19:56 -0500, Wolfgang Grafen wrote:
> 
> > import re
> > 
> > rc=re.compile("<@Trap Body text\s*"
> >                           "(?:(?P<assigned>=)|(?P<unassigned>>))\s*"
> >                           "(?P<rest>.*?)\s*\Z",
> >                          re.MULTILINE|re.DOTALL).match
> > 
> > t1='<@Trap Body text>'
> > t2='<@Trap Body text=<FONT "Times">'
> > t3="""<@Trap Body text=<FONT "Times"><CCOLOR\n   "Black"><SIZE
> > 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n
> > 58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
> > 0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE
> > 0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
> > "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
> > 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
> > 25>>"""
> > 
> > rc(t1).groups()
> > (None, '>', '')
> > 
> > rc(t2).groups()
> > ('=', None, '<FONT "Times">')
> > 
> > rc(t3).groups()
> > ('=', None, '<FONT "Times"><CCOLOR\n   "Black"><SIZE 11><HORIZONTAL
> > 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n  58.3><C-POSITION
> > 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK 0><CLEADING -0.05\n
> > ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER
> > 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
> > "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
> > 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
> > 25>>')
> > 
> > cheers
> > 
> > wolfgang
> > 
> > 
> > Asheesh Laroia schrieb:
> > 
> >> I have some SGML input (PageMaker 6.5 tagged text), and I want to be
> >> able to recognize (and delete) a tag.  That tag looks like:
> >>
> >>         <@Trap Body text:>
> >>
> >> It may also look like <@Trap Body text: useless-data>.
> >>
> >> So, I tried the regular expression r"<@.?>".  That doesn't match the
> >> above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
> >> using, and why doesn't this work?
> >>
> >> Thanks in advance!
> >>
> >> -- Asheesh Laroia.
> >>
> >> PS: An example of the tag "in the wild" is the following string:
> >>
> >> <@Trap Body text=<FONT "Times"><CCOLOR
> >>  "Black"><SIZE 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE
> >>  70><C+SIZE
> >> 58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
> >> 0><CLEADING -0.05
> >> ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER
> >> >0><GALIGNMENT "justify
> >> "><GMETHOD "proportional"><G& "ENGLISH"><GPAIRS 4><G% 120><GKNEXT
> >> 0><GKWIDOW 1><GKORPHAN 1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75
> >> 100 150><GSPACE -5 0 25>>


From llothar at mailandnews.de  Mon Feb 25 02:34:22 2002
From: llothar at mailandnews.de (Lothar Scholz)
Date: Mon, 25 Feb 2002 08:34:22 +0100
Subject: Remote debugging ?
Message-ID: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com>

Is there any debugger for python that allows me to do remote debugging
from a windows 2000 client to a python script running on a Linux
Server.

It must not be freeware. 



From jkraska at san.rr.com  Tue Feb  5 21:03:31 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 06 Feb 2002 02:03:31 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c6011e3$1_5@news.bluewin.ch> <tor06u0j1ce99ibgd98uflckcms89apfci@4ax.com> <3c607be8$1_1@news.bluewin.ch>
Message-ID: <9k116u4r58thf9lp69cso89ubtvvhhsm7f@4ax.com>

>> If I understand you properly, I'm considering making the the
>> macro call itself _generate_ what amounts to a poor man's
>> closure block. I have a few implementation details for this,
>> but the trick to understanding it is to understand that no
>> _first class_ form needs to exist to create the block, only
>> the backend facility.

>Yup you don't need first class forms.
>
>Q: are you working on modifying the C Python parser
>and compiler? What do you mean by poor man's closure?

At the moment, no. I have my own lexer, parser, validator
and emitter. The first two are fairly close to done, missing
a few notables, like comprehensions. The last are very much
a work in progress. Lotto work there; still in design on those.

C//



From mark at mceahern.com  Fri Feb  1 18:22:19 2002
From: mark at mceahern.com (Mark McEahern)
Date: Fri, 1 Feb 2002 15:22:19 -0800
Subject: comparing DateTime instances
In-Reply-To: <15451.8457.657961.274671@12-248-41-177.client.attbi.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBGEGEKAAA.mark@mceahern.com>

Skip Montanaro [mailto:skip at pobox.com]:
> Why go to the trouble of calling cmp() explicitly?  "==" seems to work for
> me:
[sample code]
> See?  No need to stretch your brain out of shape after all... ;-)

That's an excellent question and all I can say is that I got all paranoid
after reading this:

"The internal representation of date/times behaves much like floats do in
Python, i.e. rounding errors can occur when doing calculations. There is a
special compare function included (cmp()) in the package that allows you to
compare two date/time values using a given accuracy, e.g.
cmp(date1,date2,0.5) will allow 12:00:00.5 and 12:00:01.0 to compare equal."

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

Because prior to the comparison test, I basically do all kinds of dateObj2 =
dateObj1 + RelativeDateTime(whatever).  So I actually use cmp(a, b, 0.5) in
my tests and that's good enough equality for me.  I didn't want the tests to
fail just because of rounding errors.

I think I really just needed to post that in case someone else came along
with the same missing neurons as me.  It helped me to work through it too.

Cheers,

// mark




From gcordova at hebmex.com  Thu Feb 14 10:09:22 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Thu, 14 Feb 2002 09:09:22 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A72@SRM201002>

Er,

> 
> No es importante. Es importante que Ud. continue aprender. Puede
> aprender Espa?ol, Ingles, Japon?s, Perl, Python, o Ruby.
> 
> Cheers,
> Tim Hammerquist

"Es importante que continue APRENDIENDO."

Have a nice Feb/14 :-)

Japon?s es muy interesante, me encant? la caligraf?a.

Espa?ol, ciertamente, es pr?cticamente predecible en
cuanto a su pronunciaci?n, eso lo hace agradable.

Por otro lado, tiene chorromil reglas para su uso correcto,
porque Espa?ol correcto es una modalidad de escritura
que pocos aprenden.
-- 
Gustavo C?rdova Avila
*  gcordova at sismex.com <mailto:gcordova at sismex.com> 
* 8351-3861 | 8351-3862



From frankm at hiwaay.net  Fri Feb 15 00:22:12 2002
From: frankm at hiwaay.net (frankm at hiwaay.net)
Date: Thu, 14 Feb 2002 23:22:12 -0600 (CST)
Subject: [Announce] Major update to the xml_pickle package
In-Reply-To: <E16bWt1-00051w-00@mail.python.org>
Message-ID: <Pine.LNX.4.44.0202142318450.9071-100000@slacker.hiwaay.net>

Hi, I just wanted to send a note to announce a rather large
update to the xml_pickle package. I started with xml_pickle-0.51
and (together with David Mertz) added a bunch of things to make
xml_pickle a functional superset of the standard pickle module.

[Background blurb: xml_pickle is a package that looks, internally
and externally, like the standard pickle module, but produces XML
output using a minimalistic encoding.]

New features, compared to xml_pickle-0.51:

  o Can now fully restore class functionality, just like the
    standard pickle module.

  o Full internal compatibility with the pickle protocol
    (objects with __getstate__,__setstate__, and __getinitargs__
     are pickled/unpickled exactly as pickle does).

  o A pickle-compatible API layer
    (xml_pickle.dump()/dumps()/load()/loads() are drop-in compatible
     with the standard pickle module.)

  o A type-extension "plugin" system allows you to create "pretty
    printed" XML representations of objects if so desired.

  o Virtually all objects that can be pickled can be XML-pickled.
    (The only exception is that builtin types cannot be pickled as
     *toplevel* objects, since they are not true InstanceTypes.
     This restriction may be removed in a future version.)

  o Python 2.2 compatibility (Remains compatible with Python 2.0+)

  o Fine-grained security model added. I consider this an improvement
    over regular pickle, since it allows you to control how classes
    are created during unpickling. This means you can load pickles
    from "untrusted" sources and not worry about what you're getting
    (ie. no one can trick you into loading an EvilList where you
     expected a UserList). See "pydoc gnosis.xml.pickle.doc" for more
     information about the security model.

  o DEEPCOPY option added, so you can choose whether or not to have
    references inserted in the XML stream. References are great for
    saving space, but deepcopying allows third-party/non-Python parsers
    to get the entire data set without having to understand references.

  o References now keep the "pointed-to" type in the XML tag.
    This allows third-party/non-Python parsers to easily tell what the
    referenced object contains without reconstructing the object tree.

  o A few other bugfixes/improvements:
      - Corrected refchecking of temporary objects and key/val/items.
      - Numerical.arrays are now refchecked
      - SREs are now pickled

Given that there are alternative XML encoding schemes (such as XML-RPC
and WDDX), a few words to explain the particular purpose and goals of
xml_pickle seem to be in order:

   o It looks & feels just like the standard pickle module, both
     internally and externally, so all extension classes that
     speak the pickle protocol will "just work".

   o Has a minimalistic encoding scheme. All objects are represented
     using only five tags (<attr>,<entry>,<item>,<key>,and <val>) to
     simplify parsing by third-party/non-Python code.

   o The XML format is rich enough to enable xml_pickle to restore
     objects "fully alive", yet simple enough for "dumb" parsers
     to grab the complete data structures without needing to understand
     anything about Python classes, pickling, etc.

Despite the numerous changes, xml_pickle remains backward compatible
with the xml_pickle-0.51 API (and with XML documents created by older
versions of xml_pickle). The only source-level change you should have
to make would be to change the "import" line, since xml_pickle is now
part of a larger package.

For example:
    import xml_pickle
Could be rewritten:
    import gnosis.xml.pickle as xml_pickle

The complete package (Gnosis_XML_Utils), containing xml_pickle and
other useful components like xml_objectify and xml_indexer, is
released as public domain software, and is available from:

   http://gnosis.cx/download

Specifically (as of this writing):
   http://gnosis.cx/download/Gnosis_XML_Utils-0.9.tar.gz

Given the numerous changes, it'd be great to have as many people as
possible test it. The new package includes a good set of test cases
that also demonstrate the many new features of xml_pickle. Unpack the
sources, and look in gnosis/xml/pickle/test. Run "test_all.py" for a
complete set of tests.

Please send feedback, bug reports, etc., to either myself
(frankm at hiwaay.net) or David Mertz (mertz at gnosis.cx).

(Development was done on Linux and OS/2, so feedback about successes
or failures on other platforms would be greatly appreciated!)

Thanks!
frank

Summary:
<P><A HREF="http://gnosis.cx/download/Gnosis_XML_Util.ANNOUNCE">
Gnosis_XML_Util 0.9</A>
xml_pickle (pickle to XML), xml_objectify and xml_indexer (15-Feb-02)




From gh_pythonlist at gmx.de  Thu Feb  7 16:16:14 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Thu, 7 Feb 2002 22:16:14 +0100
Subject: FixedPoint
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBMEGKKBAA.marklists@mceahern.com>
References: <Dhx88.95486$KM2.3465594@atlpnn01.usenetserver.com> <NCBBLFCOHHDIKCAFGCFBMEGKKBAA.marklists@mceahern.com>
Message-ID: <20020207211613.GA1622@lilith.hqd-internal>

Le 07/02/02 ? 08:22, Mark McEahern ?crivit:
> Steve Holden wrote:
> > Since the *intent* for None is that there should only ever be one
> > object of type None, the better test would be
> >
> >     if something is None
> 
> Thanks, that's how I generally test for None--I didn't know the bit about
> "there should only ever be one."
> 
> However, you wrote:
> 
> > which can easily be used without requiring any changes to FixedPoint.
> 
> Well, sure, I could easily change my local copy of PyPgSql's PgSQL.py.

Nah, that sucks. We can do better than that :-)

> [...] Of course, I could ask the PyPgSql folks not to test for None
> like that.

Ok, I'm listening :-)

This is a quick shot, but what about changing the _quote function in
PgSQL.py to this:

##########################################################################
def _quote(value):
    """
_quote(value) -> string
    This function transforms the Python value into a string suitable to send
    to the PostgreSQL database in a insert or update statement.  This function
    is automatically applied to all parameter sent vis an execute() call.
    Because of this an update/insert statement string in an execute() call
    should only use '%s' [or '%(name)s'] for variable subsitution without any
    quoting."""

    if hasattr(value, '_quote'):
	return value._quote()

    if value is None:
	return 'NULL'

    if type(value) in [ListType, TupleType]:
	return _handleArray(list(value))

    if type(value) in [DateTimeType, DateTimeDeltaType]:
	return "'%s'" % value

    if type(value) == StringType:
	return PgQuoteString(value)

    return repr(value)
##########################################################################

There are two changes:

1) it really tests for "is None", which should be ok for all cases I can
currently imagine.

2) It also tests for a _quote() method first, so users can more easily
override the default quoting behaviour.

I can also can't see where this would break existing code.  But I would
probably ask Bill Allie before committing a subtle change like this one.

Hmm., while we're at it, we should probably change this test too: "in
[ListType, TupleType]" Does anybody know how I can reliably test if
something is a Sequence or is iterable?

(Also will have to change the StringType test anyway for Unicode,
anyway)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 3.2 ?C      Wind: 2.4 m/s



From shalehperry at attbi.com  Thu Feb 21 01:58:35 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 20 Feb 2002 22:58:35 -0800 (PST)
Subject: strings and sort()
In-Reply-To: <7xsn7vbixg.fsf@ruckus.brouhaha.com>
Message-ID: <XFMail.20020220225835.shalehperry@attbi.com>

On 21-Feb-2002 Paul Rubin wrote:
> "Sean 'Shaleh' Perry" <shalehperry at attbi.com> writes:
>> > Why doesn't sort() return the sorted list.  I would like to chain it
>> > to other operations:
>> >         b=[x for x in a].sort()
>> > 
>> efficiency.  Why return the sequence when we usually just need to sort the
>> list.  Sort is also a meothod of an object, why should it return a new
>> object
>> instead of just affecting the one it belongs to?
> 
> I don't understand this reply.  The sort method would be no less efficient
> if it returned the sequence:
> 
>   def nsort(a):
>      a.sort()
>      return a
> 

of course it would.  returning has a cost.  Most uses of sort() just need to
in place update the list.  If there was actually a copy made as some people
expect that would be even more costly.

When programming in a language, you learn the idioms of that language.



From sjmachin at lexicon.net  Sat Feb 23 20:10:08 2002
From: sjmachin at lexicon.net (John Machin)
Date: 23 Feb 2002 17:10:08 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com>
Message-ID: <c76ff6fc.0202231710.19734226@posting.google.com>

bobnotbob at byu.edu (Bob Roberts) wrote in message news:<c4e6b17d.0202231152.5f87765d at posting.google.com>...
> What would be a good way to replace every one or more spaces (" ") in
> a string with just one space?  Or replace any number of newlines with
> just one?

Same answer as in most other languages: What is a "good" way depends
on where you want to be on the speed curve -- fast development or fast
running?. Here are some points on the curve:

1. One way is to use the regular expression substitution facility  --
replace all occurrences of "  +" (*two* or more spaces) with " "
(single space). Same for newlines. Find re.sub() in the Python
manuals.

Doing *one* or more is semantically equivalent but will be slower
because of useless substitutions --- effect would be much worse if you
are using a version of Python before 2.2, as the substitution methods
were coded in Python (now in C).

2. If what you want is to replace any contiguous chunk of whitespace
(including newlines, see example below) with a single space (and trim
off *any* leading and trailing whitespace), you could use the
split-join hack:

>>> " ".join(" b    b   \n\n    x ".split())
'b b x'

3. The collapse() function in the mxTextTools package *may* be what
you want. You'd have to find it (google), RTFM, download it, install
it, ...

4. In extremis, read the Extending/Embedding manual and reach for your
C compiler -- not such a silly idea if you are habitually reading
million-line files which need their non-newline whitespace normalised
in a fashion similar to (2) above.

Once you are over the learning curve/hump and have a C extension
module with useful fast gadgets in it, adding another is minimal
bother.


From trentm at ActiveState.com  Mon Feb 18 12:55:02 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Mon, 18 Feb 2002 09:55:02 -0800
Subject: [Python-Dev] PEP 282: A Logging System  --  comments please
In-Reply-To: <200202160247.g1G2lkJ5029529@email.nist.gov>; from mclay@nist.gov on Fri, Feb 15, 2002 at 09:43:33PM -0500
References: <20020215164333.A31903@ActiveState.com> <200202160247.g1G2lkJ5029529@email.nist.gov>
Message-ID: <20020218095502.A12534@ActiveState.com>

[Michael McLay]
> I scanned the PEP and didn't find a reference to the logging package 
> supporting logging over a network. 


[Greg Ward]
> I think the right way to handle that would be to pass a file-like object
> to the logging framework, which it then write()'s to.  That should work
> just fine for stream (TCP) sockets; I *think* it would work for datagram
> (UDP) sockets too, but I'd want to test it first.


[Fred Drake]
> It seems to me that it should be trivial to map from the
> logging API to syslog; it already provides for logging to
> remote systems as well as filtering.
> 
> I'm sure this is already hashed out in the PEP though, so I
> should be reading that instead of commenting here.  ;-)


The current plan is that you can subclass class Handler for this.
Generally a subclass will have this basic API to work with:
    class Handler:
        def __init__(self):
            # initialize you socket or whatever here
        def publish(self, logRecord):
            # write your log record here
        def flush(self):
            # flush file or socket, flush buffered output to syslog backend,
            # etc.
        def close(self):
            # cleanup
c.f. http://java.sun.com/j2se/1.4/docs/api/java/util/logging/Handler.html

>From the PEP:
>     More standard Handlers may be implemented if deemed desirable and
>     feasible.  Other interesting candidates:
> 
>         - SocketHandler: A handler for writing to remote TCP ports.
>         - CreosoteHandler: A handler for writing to UDP packets, for
>           low-cost logging.  Jeff Bauer already had such a system [5].
>         - MemoryHandler: A handler that buffers log records in memory
>           (JSR047).
>         - SMTPHandler: Akin to log4j's SMTPAppender.
>         - SyslogHandler: Akin to log4j's SyslogAppender.
>         - NTEventLogHandler: Akin to log4j's NTEventLogAppender.



Trent

-- 
Trent Mick
TrentM at ActiveState.com



From amuys at shortech.com.au  Wed Feb 27 20:41:19 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 27 Feb 2002 17:41:19 -0800
Subject: Threads in python?
References: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com> <3C7AD7F1.AFCADA42@engcorp.com>
Message-ID: <7934d084.0202271741.70c81e3e@posting.google.com>

Peter Hansen <peter at engcorp.com> wrote in message news:<3C7AD7F1.AFCADA42 at engcorp.com>...
> "Fernando Rodr?guez" wrote:
> > 
> > I'm going to start coding a little utility that needs to retrieve webpages and
> > process them.
> > 
> > However, I worried that the whole app might stall if one server takes too long
> > to respond or is offline.
> > 
> > Does Python support threads? "Learning Python" doesn't mention it...
> 
> See http://www.python.org/doc/current/lib/module-threading.html .
> 
> (Yes, Python supports threads.  There may, however, be simpler and/or
> more robust approaches to what you are doing, although I personally
> would just jump on threads since I grok them better than, for example,
> asyncore.)
> 

And I personally wouldn't bother with asyncore either.  As you are
wanting to download the webpages, not just confirm their availability
(as in linkchecker), I would split the application in two, one that
does the downloading, and one that does the processing.  Under an OS
with a lightweight fork(), I would be inclined to use multi-processing
for the first stage.  For OS's with heavier weight fork() (ie. MS),
the front end will probably require threading.  OTOH as none of the
fetch threads would need to interact in anyway this is effectively
multi-processing without the safety of seperate processes.

Similar considerations apply to the processing stage, but precisely
where the balance falls depends on what processing you're doing.

Good Luck,

Andrae Muys


From nas at python.ca  Fri Feb 15 15:59:35 2002
From: nas at python.ca (Neil Schemenauer)
Date: Fri, 15 Feb 2002 12:59:35 -0800
Subject: Warning about "for line in file:"
In-Reply-To: <Pine.OSX.4.43.0202151547390.318-100000@d-128-61-180.bootp.virginia.edu>; from sdm7g@Virginia.EDU on Fri, Feb 15, 2002 at 03:51:01PM -0500
References: <20020215110159.B12454@glacier.arctrix.com> <Pine.OSX.4.43.0202151547390.318-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <20020215125935.A13061@glacier.arctrix.com>

Steven Majewski wrote:
> I also get the same behavior on any example I can contrive on
> file iterators that I get with with either readlines or xreadlines.

That's not surprising since xreadlines calls readlines.

  Neil



From mkelly2002NOSPAM at earthlink.net  Tue Feb 19 15:59:21 2002
From: mkelly2002NOSPAM at earthlink.net (Michael Kelly)
Date: Tue, 19 Feb 2002 20:59:21 GMT
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com> <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com> <b34d3835.0202190034.24337a3e@posting.google.com>
Message-ID: <d1f57u08ahpijee3ukjadfi3hle834ifmq@4ax.com>

On 19 Feb 2002 00:34:41 -0800, rony.steelandt at bucodi.com (Rony) wrote:

I don't know what's up with your OCX but the interface
names look unconventional.  As you see in mine below,
it has IGradDlgX as the Dispatch interface(and
IGradDlgXEvents as the event interface).  Anyway,
I haven't seen anything like that before so I'm at a
loss as to why they did it that way.

from win32com.client import DispatchBaseClass
class IGradDlgX(DispatchBaseClass):
||||||||||||||||||||||||




--

"I don't want to belong to any club that would have me as a member."
    -- Groucho Marx


From sjoshi at ingr.com  Mon Feb 25 09:31:03 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 25 Feb 2002 06:31:03 -0800
Subject: Problems with POST -Solved
References: <8f8ffe67.0202230614.7ea56c9c@posting.google.com> <a5bjib$l1c$1@news1.ucsd.edu>
Message-ID: <8f8ffe67.0202250631.23bf548d@posting.google.com>

This line was missing (below). The moment I put this, everything wroked fine

req.putheader("Content-type", "application/x-www-form-urlencoded")

thanks
Sunit

Jeff Davis <jdavis at empires.org> wrote in message news:<a5bjib$l1c$1 at news1.ucsd.edu>...
> Can you run the second script as a cgi and load it with a web browser to 
> determine which script is not behaving correctly? I have a suspicion that 
> the first script is somehow not posting quite correctly.
> 
> Regards,
>         Jeff
> 
> Sunit Joshi wrote:
> 
> > Hello
> > Somehow I can't seem to get any data with POST although the same
> > script (with modifications in the first one) works for GET . Here's
> > the scripts: First one is to POST and next one to recieve the POSTED
> > data.
> > 
> > #!C:\Python21\Python.exe -u
> > #To POST and output the data - postTest.py
> > 
> > from httplib import HTTP
> > from urllib import quote
> > 
> > quoteData = "Sunit Joshi"
> > postData = 'txtName=' + quote(quoteData)
> > 
> > print 'Will POST ', len(postData), 'bytes:', postData
> > 
> > # begin HTTP request
> > req = HTTP('sunitw2k:6080')
> > req.putrequest("POST" ,'/PyCgi/formTest.py')
> > req.putheader("Accept", "text/html")
> > req.putheader("User-Agent", "postTest.py")
> > 
> > # Set Content-length to length of PostData
> > req.putheader("Content-length", str(len(postData)))
> > req.endheaders()
> > 
> > # Send POST data after ending headers
> > req.send(postData)
> > 
> > replycode, message, headers = req.getreply()
> > print "HTTP RESPONSE: ", replycode, message
> > print 'HEADERS: ', str(headers)
> > 
> > # Get file-like object from HTTP response
> > # and print received HTML to screen.
> > fd = req.getfile()
> > txtlines = fd.read()
> > fd.close()
> > print '\nReceived following HTML:\n'
> > print txtlines
> > 
> > #************************
> > #!c:\Python21\Python.exe -u
> > # To send the data posted - formTest.py
> > import sys
> > import cgi
> > 
> > sys.stderr = sys.stdout
> > 
> > form = cgi.FieldStorage()
> > print "Content-type: text/html"
> > print
> > if not form:
> > print "<h2>Field 'txtName' not filled in.</h2>"
> > print str(form)
> > else:
> > print '<h3>You entered <font color="#FF6600">%s</font></h3>' %
> > form.getvalue('txtName','none')


From bldrake1 at yahoo.com  Thu Feb  7 10:19:30 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 7 Feb 2002 07:19:30 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <fb072181.0202060723.e400eaa@posting.google.com> <ug04da4k1.fsf@python.net>
Message-ID: <fb072181.0202070719.12082362@posting.google.com>

Michael Hudson <mwh at python.net> wrote in message news:<ug04da4k1.fsf at python.net>...
> bldrake1 at yahoo.com (Barry Drake) writes:
> 
> > Maybe I'm missing something, but here is what I get:
> 
> Well, I'm missing your point.
> 
> > Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
> > win32
> > Type "copyright", "credits" or "license" for more information.
> > IDLE 0.8 -- press F1 for help
> > >>> import math
> > >>> n = 10
> > >>> nf = 10.0
> > >>> a = 1.0
> > >>> b = 0.5
> > >>> a/n
>  0.10000000000000001
> > >>> b + a/n
>  0.59999999999999998
> > >>> a/nf
>  0.10000000000000001
> > >>> b + a/nf
>  0.59999999999999998
> > >>> 
> > 
> > Python does double precision floats only (see Python Essential
> > Reference 2nd ed. by David Beazley, p.23).
> 
> But you can have arrays (as in, objects of type array.ArrayType) that
> store C single floats.  Which was Jason's point.
> 
> > Python implements IEEE 754: 17 digits of precision with exponent in
> > -308 to 308.
> 
> Oh No It Doesn't.  Python uses (for the objects of type
> types.FloatType) whatever the C compiler that compiled it called
> "double".  This may be IEEE 754 doubles on the platforms you've used,
> but there's certainly no guarantee of that (e.g. Crays, IBM boxes (I
> think)).
> 
> >          float     double
> > python   64 bits   64 bits
> > C        32 bits   64 bits
> 
> "python double" is meaningless.
> 
> > BTW, you might want to check out the Numeric module in the NumPy
> > distribution at http://numpy.sourceforge.net.
> 
> That lets you store floats or doubles too.
> 
> Cheers,
> M.

Michael,
Your message refers to my first post, which was missing the last line
of the output.  That being the actual point of my message I will
include it here:

Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import math
>>> n = 10
>>> nf = 10.0
>>> a = 1.0
>>> b = 0.5
>>> a/n
0.10000000000000001
>>> b + a/n
0.59999999999999998
>>> a/nf
0.10000000000000001
>>> b + a/nf
0.59999999999999998
>>> b
0.5
>>> 1.0 + (0.6 - (b + a/n))
1.0

This shows that the computer "sees" Jason's output as 0.6 within the
numerical precision of the machine.  Perhaps I was answering the wrong
question, which I understood to be that Jason was seeing ("suposing
that n=10 we get: myArray[0] = 0.600000023 ("something like this")")
different results than he thought he should ("but if i do: myArray[i]
+ (1.0/n) it give's me 0.6").  If I was wrong, I stand corrected.

Also, regarding the double vs float (C types), I was talking about the
default behavior of Python, which I thought Jason was using.  You are
correct, and I learned something new, that NumPy has a variety of
Floatnn types, which is very cool.  They also say in the paragraph
following the discussion of Floatnn that this is an add-on to the
default in Python which is Float64.


From joshm at taconic.net  Sun Feb 24 22:32:53 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sun, 24 Feb 2002 22:32:53 -0500
Subject: Busy with better things to do??? >Re: Guido wins FSF Award
References: <mailman.1014220773.23411.python-list@python.org> <3c73e173.10676151@127.0.0.1> <1263.820T1799T12905345threeseas@earthlink.net>
Message-ID: <3c79adfc$1_2@corp.newsgroups.com>

> You are not being productive or helpful to the project.

We were merely following your example.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From sholden at holdenweb.com  Thu Feb 21 17:09:25 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 21 Feb 2002 17:09:25 -0500
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <Zded8.54109$n73.994836@atlpnn01.usenetserver.com>

"Kevin Dahlhausen" <kdahlhaus at yahoo.com> wrote in message
news:283adf56.0202201551.7026d548 at posting.google.com...
> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

The real crux, of course, is what you mean by "inexpensive".

www.cornerhost.com (a supplier and a customer) starts at $5 per month, but
to build extensions you'd be up at $20 per month for a shell account.
However, they are very helpful (Michal Wallace, who runs the company, posts
on this group from time to time), and understand Python well.

If it were just a matter of installing certain packages rather than
developing your own you might well find they'd do that for you even if you
were using one of the cheaper accounts.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From logiplexsoftware at earthlink.net  Mon Feb 18 13:49:23 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 18 Feb 2002 10:49:23 -0800
Subject: basic books/guides on multithreading programming?
In-Reply-To: <7934d084.0202172009.75324c2a@posting.google.com>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
	<mailman.1013816158.26552.python-list@python.org>
	<7934d084.0202172009.75324c2a@posting.google.com>
Message-ID: <20020218104923.6951c12c.logiplexsoftware@earthlink.net>

On 17 Feb 2002 20:09:43 -0800
Andrae Muys wrote:

> Cliff Wells <logiplexsoftware at earthlink.net> wrote in message
news:<mailman.1013816158.26552.python-list at python.org>...
> > Thread-safety boils down to this: lock any /shared/ resource that might
be
> > /modified/ by a thread.  End of story.  You can get into all sorts of
> > debates about when these locks are necessary, but 99% of the time they
are,
> > and the other 1% isn't worth the effort to decide whether they really
are
> > or not.
> > 
> 
> "End of Story" is a rather interesting way to summarise the issues of
> Deadlock, Lock-ordering, Livelock, Starvation, Convoying, and
> Lock-Contention to name a few of the problems you may face moving from
> serial to parallel algorthims (although the last two only affect
> perfomance, not correctness).
> 
> I don't have the time to discuss these further atm, and I am not
> personally aware of a readily accessable text on concurrent
> programming.  Ben-Ari "Principles of Concurrent Programming" PH '82
> was the text used when I studied this back in university, however it
> is a little dry.  OTOH, I am very reluctant to recommend any of the
> language/library specific texts to someone without an existing
> knowledge of concurrency issues.  Even Kleiman, Shah, and Smaalders
> "Programming with Threads" (an excelent book, covers libpthread based
> programming in C/C++) gives rather preemptory coverage of the problems
> involved.
> 
> Unfortunately without an understanding of the problems, it is hard to
> understand/apply the solutions provided by the various
> languages/libraries described.
> 
> Cliff: Sorry I can't answer your question straight away, maybe check
> your local library (possibly uni-library) for a copy of Ben-Avi.  If
> you have any problems mapping the concepts he discusses to specific
> Python code I (and I imagine the rest of c.l.py) will do what I can to
> help.  Other then that, WHY do you feel you need threads?  Or is this
> just a "Something new to learn" question?
> 
> Tripp: If I sound a bit harsh, I'm sorry.  Understand that as someone
> with their name attached to a unexpectedly widely distributed tutorial
> on threading, I get *alot* of help requests from new programmers who
> think that threading is "just another library to learn", and approach
> it as such (with the inevitable trouble).  Your "End-of-story"
> statement was just more then I would reasonably let pass without
> comment.

Obviously you were aiming this statement at me, not Tripp ;)  I won't argue
much (because you are correct), I will simply point out that these
situations, while real, aren't that common in typical multithreaded
applications (in my experience) unless the program is poorly thought out to
start with.  They do however, make multithreaded programming seem
_extremely_ daunting to a beginner.

Speaking for myself, I've written several long-running multithreaded
applications where I never _consciously_ considered these things.  I think
once someone has developed a feel for multithreaded programming, these
issues are handled at a different level in the designer's mind and explicit
consideration of them usually isn't necessary (unless perhaps I'm just
applying my own experience too broadly).  Maybe I'm just smarter than I
give myself credit for <big wink>, as I've never run into the "inevitable
trouble" you mention. 

Considering all this, I'll take a different position: For beginners: don't
worry about these issues now, but after becoming familiar with the basics
of multithreaded programming (for which the standard Python docs should be
sufficient), find a good book to deepen your knowledge.  That is to say,
I'll qualify my earlier statement "end of story" with "for now".

> Everyone: Does anyone know of an accessable concurrency text?  One
> that moves from discussing the basics (data-races, critical sections,
> mutural exclusion, lock-order based deadlock, resource based deadlock,
> etc) to the various solutions (mutex/cond-var, semaphore, monitor,
> rendezvous) to the various design approaches (master-slave, pipeline,
> work-pile, etc).  Possibly even a discussion of various performance
> considerations (convoy's, wait-free synchronisation, cache-aliasing,
> spin vs. blocking locks, etc).

I would certainly like to see a good discussion of these.  Do you have a
URL for the tutorial you wrote?

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From gcordova at hebmex.com  Fri Feb 22 17:48:11 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 16:48:11 -0600
Subject: comp.lang.python in English? (was Re: Why Python is like BASI
	C)
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE7@SRM201002>

> 
> On Wed, 20 Feb 2002 17:23:22 +0100, Anders Dahlberg 
> <andda715 at student.liu.se>
> wrote:
> >
> >"Aahz Maruch" <aahz at panix.com> skrev i meddelandet
> >news:a50hd2$6vt$1 at panix2.panix.com...
> >> In article <m2wux85hud.fsf at desk.crynwr.com>,
> >> Russell Nelson  <nelson at crynwr.com> wrote:
> >> >Henrik Motakef <henrik.motakef at web.de> writes:
> >> >> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> >> >>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> >> >>>>
> >> >>>> Blig matro en cordiolios, plur tanto lestanto 
> forgolumnia.  Plachero
> >> >>>> czignat?  Classifero en bragna!
> >> >>> Woas da Deifi wos des hoasn mog.
> >> >> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> >> >
> >> >Even better: we could switch languages within threads, to *really*
> >> >keep the monolinguists confused.
> >>
> >> B'vakasha
> >
> >Ja, varf?r inte?
> 
> My s?ster was bitten by a m??se once.
> 

Das Computermaschine is not for your fingerpoken?
.... blinkenlighten?

(can't remember nor find the whole thing)

-gus



From gherman at europemail.com  Mon Feb 25 12:41:32 2002
From: gherman at europemail.com (Dinu Gherman)
Date: Mon, 25 Feb 2002 18:41:32 +0100
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com> <3c7a70d2_4@news.bluewin.ch>
Message-ID: <gherman-A86A1B.18413225022002@news.t-online.com>

In article <3c7a70d2_4 at news.bluewin.ch>,
 "Samuele Pedroni" <pedronis at bluewin.ch> wrote:

> Yes they are bugs. Please report them.
> Since 2.1 release (two month ago) there have been approximately  10000
> download.
> 
> How are such bugs possible?
> 
> 1. the typical idiom for list(l) is l[:] (mildly irrelevant I know)
> 2. if you check the test suite for CPython 2.1 and Jython there is no test
> for list(l) behavior
> 3. the new reset method is also not tested by the test suite and is not
> reported in the NEWS file
> 
> (These are the result of some grepping, maybe I'm wrong but given the bugs
> probably I'm right).
> 
> We try to follow the big picture and the PEPs and check the NEWS file but
> for the rest the test suites
> are our best hope vs. delusional friend.
> If a feature is old, under-used, or through usage the bug does not show
> through, and un-tested things become tricky.
> 
> It is open source: Jython is as conforming as its community and CPython
> community ACTIVELY want it to be.


I'm pretty surprised! I knew the Python test suite is *very* far 
from complete, but list() is an *extremely* crucial function,
isn't it? 

My impression is that Jython claims to be an implementation of 
Python in Java. Everybody understands the existance of bugs, but 
if functions are missing I'm not sure there is sufficient quality
control, leave alone a useful test suite. My bold guess is that 
it should be very easy to check automatically for each module in 
the std.lib. at least if the same classes and methods do exist 
in CPython and JPython.

Honestly, I don't think it makes much sense to maintain two
code bases without some degree of automatic testing...

When seeing such bugs, my immediate reaction (like that of most
others) is to think that not many people can be using this serious-
ly.

Regards,

Dinu

PS: BTW, how about this one:

[localhost:~] dinu% jython
Jython 2.1 on java1.3.1 (JIT: null)
Type "copyright", "credits" or "license" for more information.
>>> import os
>>> os.chdir('..')
Traceback (innermost last):
  File "<console>", line 1, in ?
  File ".../Jython-2.1/Lib/javaos.py", line 56, in chdir
OSError: [Errno 0] chdir not supported in Java: ..


From rich at worldsinfinite.com  Tue Feb 12 11:00:31 2002
From: rich at worldsinfinite.com (Rich Harkins)
Date: Tue, 12 Feb 2002 11:00:31 -0500
Subject: re.sub and empty groups
In-Reply-To: <3C693B12.1060504@chambersharrap.co.uk>
References: <3C693B12.1060504@chambersharrap.co.uk>
Message-ID: <200202121600.g1CG0VQ02195@localhost.localdomain>

On Tuesday 12 February 2002 10:56 am, Patrick Gaherty wrote:
> I'm trying something along the following lines:
>
> line = re.sub(r'(a)(b)?(c)',r'\3\2\1',line)
>
> and Python complains about empty groups when (b)? doesn't match.
>
> What's the best way of getting around this?
>
> Thanks in advance.
>
> Patrick Gaherty

Try moving the question mark into the group itself, for instance:

>>> print re.sub(r'(a)(b?)(c)',r'\3\2\1','abc')
cba
>>> print re.sub(r'(a)(b?)(c)',r'\3\2\1','ac')
ca

If you need a group where (b?) is you can use the non-grouping group, i.e.: 
((?:some_re)?).

This of course assumes that you want to insert nothing for the second group 
when the second group doesn't match.  If you need to insert something 
more complicated I would guess it is time to move to a function for a 
replacement string...

Hope this helps!
Rich



From db3l at fitlinxx.com  Tue Feb 19 13:19:37 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Feb 2002 13:19:37 -0500
Subject: Operating system requirements
References: <3C71A556.3E7BF27E@xatt.netx> <3C71CD62.23BF11AC@erols.com> <b34d3835.0202190041.40b22cc4@posting.google.com>
Message-ID: <uu1sdz64m.fsf@ctwd0143.fitlinxx.com>

rony.steelandt at bucodi.com (Rony) writes:

> Jerry Gitomer <jgitomer at erols.com> wrote in message 
> > Windows 95 or newer.
> 
> Well this is a very restricted answer ;)

Not to the original question, which was running Python on a _Windows_ PC :-)

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


From miracle at paradise.net.nz  Mon Feb 25 20:20:56 2002
From: miracle at paradise.net.nz (Matthew Sherborne)
Date: Tue, 26 Feb 2002 14:20:56 +1300
Subject: advocacy help ( Python vs VB )
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>
Message-ID: <000e01c1be63$d7ee1a30$08020196@Jonah>

Google uses python to script their cluster(s).

http://www.google.com/jobs/openings.html

(Search for python on that page, I counted 10)

Also check out these URLs:

http://home.rmi.net/~lutz/advocacy.html

http://py-howto.sourceforge.net/advocacy/advocacy.html

GBU
Matthew Sherborne

----- Original Message -----
From: "Stefan Kuzminski" <SKuzminski at FairIsaac.com>
To: <python-list at python.org>; <activepython at listserv.ActiveState.com>;
<python-help at python.org>
Sent: Tuesday, February 26, 2002 1:25 PM
Subject: advocacy help ( Python vs VB )


> I need to convince my co workers and erstwhile managers to build a
> production system in Python.  They're coming from a VB world and are
fearful
> to detach from the Redmond mothership.  I need hard facts, not warm and
> fuzzies about syntax but names of big companies using Python and things VB
> just *cannot* do ( threads, regular expressions.. ? ).
>
> Thanks,
> Stefan
> _______________________________________________
> ActivePython mailing list
> ActivePython at listserv.ActiveState.com
> To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
>




From jkraska at san.rr.com  Sun Feb 17 01:50:14 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 17 Feb 2002 06:50:14 GMT
Subject: Why Python is like BASIC ...  java "vs" python
References: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph> <mailman.1013826279.13649.python-list@python.org> <i4iu6ug73jr2lmpblj2e70e1tci043st0l@4ax.com>
Message-ID: <u3lu6u4e10n7s7se8k2bitcajp3j710loq@4ax.com>

>>Java is not at all similar to C++.  Different.
>
>This is also just silly (sic).

You misspelled "ridiculously ignorant".

C//



From andersjm at dancontrol.dk  Wed Feb  6 07:57:50 2002
From: andersjm at dancontrol.dk (Anders J. Munch)
Date: Wed, 6 Feb 2002 13:57:50 +0100
Subject: distutils, C extensions, Borland compiler, wxPython
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk> <mailman.1012939329.5851.python-list@python.org>
Message-ID: <3c61286b$0$89086$edfadb0f@dspool01.news.tele.dk>

"Gerhard H?ring" <gh_pythonlist at gmx.de> wrote in message
news:mailman.1012939329.5851.python-list at python.org...
> Le 05/02/02 ? 15:07, Anders J. Munch ?crivit:
> > Hi,
> >
> > I have built Python from source on Win2K using cygwin except with a
> > Borland compiler (BCB4) instead of gcc.  It took a fair amount of
> > hacks to get the configure script and makefiles to work (s/.o/.obj/ is
> > just the beginning) but it was worth it.
>
> I' currently doing something similar with mingw32 instead of Cygwin :-)
>
> The real question is if you can use dynamically loaded extension modules
> at all if you have all of the Python core in a python.exe.

And I thought the Python core was always in python.exe.  Strange .. in
my build the python core is in the .exe, and I haven't done anything
special to achieve that, just "make install".  Perhaps a difference
between the makefile in the root of the source tree and the PC
subdirectory stuff?

>
> I think I have heard something to the effect, that on Windows, DLLs
> can't link back to .EXEs, only to other DLLs.
>
> Which would mean that you can either build a python{major}{minor}.dll or
> you compile every single extension module (i. e. all of wxPython, for
> example) into one big python.exe.

Yup!  One build, one binary, one compiler, one consistent set of
compiler switches for everything => no hassle.  That's my philosophy.
Well ok _some_ hassle, but no nasty surprises when the alignment gets
mixed up or the runtimes in the .dll and the .exe don't cooperate.

- Anders




From nospam at bigfoot.com  Tue Feb 19 17:17:49 2002
From: nospam at bigfoot.com (Gillou)
Date: Tue, 19 Feb 2002 23:17:49 +0100
Subject: Alternative for \n when using telnetlib
References: <1103_1014153824@Johan>
Message-ID: <a4uigt$jnu$1@norfair.nerim.net>

Did you try the small example

http://www.python.org/doc/current/lib/telnet-example.html

Try to pass the username in the cmd line too :

$ telnet -a -l myusername host

HTH

"Johan" <aequitas at planet.nl> a ?crit dans le message news:
1103_1014153824 at Johan...
> When connecting to my adsl modem (alcatel speed touch pro) via telnetlib i
can read the date write my username. But then i when i give the \n to
proceed to the password it doesn't work. In interactive mode it has the
> same problem. It just writes a new line.
>
> Is there a alternate for the \n command?
>




From amuys at shortech.com.au  Tue Feb 12 19:54:43 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 12 Feb 2002 16:54:43 -0800
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com>
Message-ID: <7934d084.0202121654.1275c67e@posting.google.com>

joost_jacob at hotmail.com (J.Jacob) wrote in message news:<13285ea2.0202080524.e2e7948 at posting.google.com>...
> With a module like this included in the Python distribution you could
> make
> simple text based user interfaces that work on Windows, linux, *nix
> and Mac
> that dont flicker, that is, if it is well implemented.  It looks so
> simple but
> it really is not.  You would need to know how to do things on all
> platforms.
> I tried and gave up, besides i do not have access to much OS's to test
> things
> with.  Doing this for linux and unix is simple with the curses module,
> but
> Windows? Mac?  The pdcurses and WConio modules i found for Windows
> need
> compilation etc., couldnt test/install them properly (no MS compiler
> here).
> Cygwin Python has partially implemented curses but then you need to
> have Cygwin installed.

AFAIK cygwin's curses implementation is complete, at least it's
sufficient to permit lynx/mutt/links/etc to compile and run.  They
require far more x-platform functionality then you do :).  You also
don't require cygwin installed, just the associated dll's, in this
case the cygwin, cygncurses, and probably cygpanel/cygmenu dll's
(IIRC).  ie. <22k, which is hardly an excessive demand :).

The most likely restriction, and the reason I suspect it hasn't been
included in the Python windows distribution by default, is that cygwin
is GPL'd, so if you want to distribute a turn-key proprietary
application you will need to license an GPL exemption from RedHat
(http://www.redhat.com/software/tools/cygwin/buycontract.html).

OTOH, as the ncurses library itself is under the MIT license, if you
don't want to pay RedHat you could,
a) port it yourself.
b) let the user link the python wrapper against cygwin themselves
before using your application.

both of which are permitted under the GPL.

Still many UI projects have little or no exclusitivity or sale value,
and are developed strictly for their use value.  In such cases, just
pick your favourite OSS license and you can link freely against
cygwin.

Andrae Muys


From gbreed at cix.compulink.co.uk  Thu Feb 21 08:05:48 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Thu, 21 Feb 2002 13:05:48 +0000 (UTC)
Subject: Adding static typing to Python
References: <mailman.1014225213.11001.python-list@python.org>
Message-ID: <a52rbc$p1i$1@thorium.cix.co.uk>

Justin Sheehy wrote:

> Those terms are used fairly often in programming language
> discussions.  Many languages use the static/dynamic distinction to
> describe their own typing system.

Thanks for the examples.  Yes, static/dynamic typing does seem to exist in 
the real world :-)

> > The only definition I have to hand contradicts it:
> > <http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=strong+typing>
> 
> So, if we're just going to pick definitions off the web, here's one
> that appeared rather quickly on google:

Hang on, I wasn't "picking definitions off the web".  That's the only 
technical dictionary I had to hand.  I wouldn't expect anybody casually 
mentioning "strong typing" to have a better one.  I happened to find 
another in preparing this message

<http://www.harcourt.com/dictionary/def/9/9/2/7/9927800.html>

that equates strong with static typing.  It looks like "strong typing" 
doesn't have a universally agreed definition, so I suppose the best thing 
is to avoid the term.

Out of general interest, does anybody know of a dictionary I *can* rely on 
for these kind of definitions?

> A common interpretation/explanation of this is that dynamic typing
> assigns types to values, while static typing assigns types to
> variables.  This is in line with my statement about Python's strong
> yet dynamic typing.

That makes sense as regards static/dynamic, thanks.  So the argument for 
adding optional static typing is for names to have types associated with 
them, which is consistent with types being declared for function 
arguments.

If the type was declared only for the initial assignment, for example

def cons(element, stuff: tuple):
  stuff = list(stuff)
  stuff.append(element)
  return tuple(stuff)

would that still be static typing?

> > Taking "static typing" literally would probably mean that an object 
> > couldn't change it's type after creation.  That's almost always the 
> > case in Python.  If you mean identifiers have to stay bound to the 
> > same type, it'd mean
> >
> > x = 1
> > x *= 5L
> >
> > would be illegal, because x begins as an integer and then becomes a 
> > long.  
> 
> This shows that you don't yet understand Python's assignment rules.

How does it show this??????  How ever badly I understand the terminology, 
I thought I had a good knowledge of Pythonic assignment.  Certainly 
nothing in the explanation below is new to me.

> In the above, no object is changing type.  The following is what 
> happens:
> 
>  - An object of type integer is created.
>  - The name "x" is bound to that object.
>  - An object of type long is created.
>  - The name "x" is bound to that object.
> 
> The binding of names to objects is completely independent from the
> values of the objects.  Objects have types in dynamically-typed
> languages, but names may not.

Yes, and the example was of an identifier being bound to a different type. 
 I never said any objects were changing type.  Here's an example of that:

Python 2.2 (#1, Dec 31 2001, 15:21:18)
[GCC 2.95.3-5 (cygwin special)] on cygwin
Type "help", "copyright", "credits" or "license" for more information.
>>> class Test1(object): pass
...
>>> class Test2(object): pass
...
>>> test = Test1()
>>> test.a=1
>>> type(test)
<class '__main__.Test1'>
>>> test.__class__ = Test2
>>> type(test)
<class '__main__.Test2'>
>>> test.a
1


                 Graham

     (http://www.microtonal.co.uk/)


From threeseas at earthlink.net  Sun Feb  3 19:00:59 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:00:59 GMT
Subject: Attention Phil Hunt
References: <mailman.1012327189.31750.python-list@python.org>
Message-ID: <585.799T42T11732924threeseas@earthlink.net>

On 29-Jan-02 12:58:27 Laura Creighton <lac at strakt.com> wrote:
>Every minute you spend arguing here is a minute that the release of
>Herbivore is delayed.

protecting once psuedo position can be very important.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From zomba at dtmf.com.ar  Wed Feb  6 00:05:50 2002
From: zomba at dtmf.com.ar (Zomba)
Date: Wed, 6 Feb 2002 02:05:50 -0300
Subject: Ioctl! Need some help to make set
References: <3CD41CC4.5752F30C@wag.caltech.edu> <j47kmjiyuq.fsf@informatik.hu-berlin.de> <X7XA8.4743$CN3.179616@news2.tin.it> <j43cx6ilpq.fsf@informatik.hu-berlin.de> <wDeB8.6725$zW3.177926@news1.tin.it>
Message-ID: <002001c1aecb$f2917e60$020a0a0a@cotelviso.com.ar>

I made a
s=socket(AF_INET, SOCK_DGRAM)
res= fcntl.ioctl(s.fileno(), 0xc020690d, 'le1' + '\0' *29)

the 0xc020690d call, work pretty well, i can take from res (with
struct.unpack) the ip addr of my device, but the problem is:

fcntl.ioctl(s.fileno(), 0xc020690c, res)
and it said: "Errno 22 - Invalid Argument"

(I try this, because i make my own 'res' with other information, but it had
the same error)

Im working on OpenBSD 3.0, the 0xc020690d is OSIOCGIFADDR and 0xc020690c is
SIOCSIFADDR

Ideas??!

Thanks in advance ppls
Zomba






From johnroth at ameritech.net  Sat Feb 23 09:14:27 2002
From: johnroth at ameritech.net (John Roth)
Date: Sat, 23 Feb 2002 09:14:27 -0500
Subject: Adding static typing to Python
References: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu> <a55qob$ebu$1@thorium.cix.co.uk> <slrna7dhbm.5et.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <u7f8mhek950ec6@news.supernews.com>

A bit of a linguistic nit below:

"Quinn Dunkan" <quinn at regurgitate.ugcs.caltech.edu> wrote in message
news:slrna7dhbm.5et.quinn at regurgitate.ugcs.caltech.edu...
> On Fri, 22 Feb 2002 16:14:03 +0000 (UTC), gbreed at cix.compulink.co.uk
> <gbreed at cix.compulink.co.uk> wrote:
> >Quinn Dunkan wrote:
> >

> >> I'm not even sure why the smalltalk style 'object followed by
method'
> >> notation
> >> caught on, especially since 'function followed by args' has always
been
> >> firmly entrenched.
> >
> >It happens to match the usual English word order: "the class's
method"
> >rather than "the method of the class".  If the field had been
dominated by
> >French speakers, perhaps it would have been the other way round.
>
> Yeah, I've heard postfix function application described as
"subject-verb" and
> prefix application as "verb-subject", which indicates that people are
thinking
> linguisticly.  But I still think English has at least a strong
orientation
> around "the area of the circle" or "rotate the circle" or "the sorted
list" all
> of which are verb-subject.  Another hypothesis is that OO-heads wanted
to
> emphasize that their programs are conceptually data oriented, rather
than
> function oriented.  Could be the same reason they invented all that
funny
> "message" and "responds to" terminology.

Your first and third examples do not include a verb, so subject-verb or
verb-subject aren't really relevant. The actual linguistic issue is
whether noun
clauses are written modifier-noun or noun-modifier.

Writing a method call as
object.attribute.attribute.attribute(parameters)
has the distinct advantage that it is much easier to emit code: that's
the
order in which the program has to follow the object chain. Doing it in
the other order requires the compiler or interpreter to maintain a stack
before emitting code or executing the instruction.

I've never understood "message" as terminology either. I've
always made the totally unsupported assumption that way back
in the beginning, someone thought of each object instance as being
a separate thread, so that message actually meant something other
than 'function call with a misleading name.'

John Roth






From gh_pythonlist at gmx.de  Tue Feb 12 04:52:09 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 12 Feb 2002 10:52:09 +0100
Subject: Shell command
In-Reply-To: <20020212044512.A9351@node0.opengeometry.ca>
References: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net> <20020212044512.A9351@node0.opengeometry.ca>
Message-ID: <20020212095208.GA1925@lilith.hqd-internal>

Le 12/02/02 ? 04:45, William Park ?crivit:
> On Tue, Feb 12, 2002 at 10:35:21AM +0100, DelPiccolo Ivano wrote:
> > How can I do to launch a command in my python script ?
> > example : I want to launch the command : rm test.txt
> 
> os.system('rm test.txt')

Yep. But instead of most of that "simple" shell commands you can use
Python functions in the os and shutil modules, for example:

    os.unlink("test.txt")       # Remove the file

This is a little bit faster then executing a seperate shell process, and
looking into these modules is quite informative :-)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 11.7 ?C      Wind: 7.9 m/s



From ajeru at vknn.org  Tue Feb 19 19:58:10 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 16:58:10 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <slrna74ssf.21s.QnickQm@localhost.localdomain>
Message-ID: <24c39b2c.0202191658.3fd74528@posting.google.com>

QnickQm at alum.mit.edu (Nick Mathewson) wrote in message news:<slrna74ssf.21s.QnickQm at localhost.localdomain>...
> In article <24c39b2c.0202181531.187fad4c at posting.google.com>, 
>         Alexander Jerusalem wrote:
> > Has anyone heard of plans to add static type checking to Python?
> 
> Someone else has mentioned an old proposal of Guido's for adding
> optional type declarations in order to help the compiler catch errors.
> 
> Making a real static type system is harder than you'd think.
> Consider this function:
> 
> def fn(lst,item):
>    for i in lst:
>      if i == item:
>         return repr(i)
> 
> What is the type of fn?  You could call it:
> 
>      (list[int], int) -> str
> 
> But then you'd lose the polymorphism of the function.  To a first
> approximation, you might say:
> 
>      (list[A], A) -> str
> 
> But keep in mind that not any A will work...
> 
>      (list[A], A) -> str  s.t. A has __eq__(A) and __repr__() -> str
> 
> And that x==y is sometimes okay even if x.__class__ != y.__class__...
> 
>      (list[A], B) -> str  s.t. A has __eq__(B) and __repr__() -> str
> 
> But there's nothing that says you can't use a tuple or a string or a UserList 
> instead of a list...
> 
> 
>      (sequence[A], B) -> str  s.t. A has __eq__(B) and __repr__() -> str
> 
> And you aren't using _all_ of the sequence protocol: just iteration...
> 
>      (A, B) -> str   s.t. A has __iter()__ -> C 
>                      s.t. C has next() -> D (or raises StopIteration)
>                      s.t. D has __eq__(B) and repr() -> str
> 
> IMNSHO, some kind type inference is the only tractable answer.
> 
> And-what-if-repr-doesn't-return-a-string?-ly yrs,

You argue that there are cases where static typing gets in your way,
where it makes programs less flexible. That's true and that's the
reason why I my suggestion was to add it as an optional feature. It's
also clear that it's not easy to do and that you need generics if you
have types otherwise you get what Java has: ClassCastException.

I like multiple type based dispatching (multi methods) (not
inheritance based dispatching) and I don't see how this is possible
without declaring types although someone has suggested that CLOS does
it without static typing. I'm not sure how this is possible or if this
means that no type declarations are necessary.

Alexander


From threeseas at earthlink.net  Sun Feb 10 08:00:30 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 10 Feb 2002 13:00:30 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>	<3c62a989.3448437947@news>	<a3ucsk$12h8$1@nntp6.u.washington.edu>	<a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com>
Message-ID: <5659.806T1220T5243918threeseas@earthlink.net>

On 09-Feb-02 16:55:04 Jeff Grimmett <grimmtoothtoo at yahoo.com> wrote:
>Cliff Wells <logiplexsoftware at earthlink.net> wrote in message
>news:<mailman.1013203955.3594.python-list at python.org>...

>> Arguing with irrational oddballs on usenet is today's equivalent of the
>> Rubik's Cube.

>Only thing is, there is an actual solution to Rubik's Cube.

>I remember Timbo from when I used to hang out in c.s.amiga.*, years
>ago, so consider that if he's still at it, there's a pretty good
>chance that nothing anyone here can say will put an end to the
>nonsense. Killfile him and move on.

>I guess his Amiga broke, so he's looking to Python instead of ARexx to
>implement his ... whatever it is.

>Timbo, it won't work. You need to use Perl. Perl, or C#.

There are header files to these post, which show what system I'm using.
Certainly you know this.

You make posts to the newsgroup three days last august, then this one.
That's your complete history in this newsgroup. That suggest the above is
a troll and an act of intentional defamation of character, libel, slander,
or whatever you want to call a false witness. Hmmm, you haven't changed
yet, or is that what you are (see below)?


The Amiga was the first publicly available system to have all three primary
user interfaces (Command line interface, GUI (workbench) and Automation
Port Interface (API - screw you if you do not like my redefinition of this
term - But it is a correct redefinition that includes the USER.)

In the products the computer industry presents the consumers:
It always seems that there is something lacking, always an incomplete user
tools. Though the Amiga came really close, what was and is lacking is the
intelligence in control of the Amiga IP and as such the system was taken
down and made into an incomplete and problematic system for the user. The
spoiled pirating user base didn't help and the pirates grew up to insure
Amiga will always continue with such ignorance.

The smartest thing to happen so far with the Amiga platform is UAE and
AROS, but even these freely available products can have the the rug pulled
out from under them by the ignorant IP holders. And this will happen
should either of these products become relatively popular. Such IP control
and manipulation has been used before to stunt the Amiga Sales and it will
happen again.

For that, The Amiga is constantly a "dying system", as it has been for
most of it's existance. Amiga corps only sees what ideas it can extract
and try and stamp their IP ownership on, from the so called "user base"
or "Kommunity".

Python is available for many platforms, including the Amiga. Where I have
found that development is alot easier. Maybe on Linux it'd been easier
than windows but Amiga corps lead me to eventually have a windows box (an
example of lack of intelligence that effects the consumer, but what to
expect by PC born owners of Amiga IP?).

Whats the windows box give me that I don't have on the Amiga. Better web
browsing and some applications like allowing me to better deal with
accessing my digital camera and editing pics. Better web page editing
tools (for free). DVD support.

For python, maybe someday I'll figure out how to get tkinter to be
findable by python applications that use it, when I click on their icons,
and in a manner that anyone who installs it on a windows box won't have to
figure it out too. Untill then tkinter is a generall unaccessable GUI for
python and available for fewer platforms than what Python itself is
available. That's really to bad, given python could use a standard GUI
interface package available where ever python is.

What's the history of systems and their user interfaces?

MS DOS began with a Command Line Interface, then GUI but never did it
evolve it's Command Line Interface to have what is considered standard
functionality like command line history and the deception of resizing the
window is a great hinderance to use. The third UI in Windows is various
half ass attempts that are not really for the consumer to even know about.

The Mac began it's user interfaces with the GUI, never really having a
Command Line Interface - Though Now it might have one in OSX. Its third
user interface is limited or constrained to applescript.

Today the Mac might be the best supplier of a system with all three
user interfaces.

But then there is Linux and the Hurd. Also with nothing much more than a
Command Line interface and GUIs, but the potential of adding in the third
user interface on the GNU applications is tremendious. With a skelton code
frame, such conversion can happen very quickly across the platform.

So, there really is not yet an existing system having all three primary
User Interfaces openly available. Yet every one know there is value to be
had in such.

There can only be one reason for this failure. Nobody wants to give the
consumer such control to allow the consumer to do things for themselves.

And that's just the scientifically validated usefulness of the three
primary user interfaces.

Certainly such resistance to even such basics suggest what to expect of
anything that is built upon all three, such as the VIC and the power it
gives the users to do things for themselves, including autocoding.

Should I be supprised a troll from Amigadom shows up?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jason at tishler.net  Fri Feb  1 13:29:50 2002
From: jason at tishler.net (Jason Tishler)
Date: Fri, 1 Feb 2002 13:29:50 -0500
Subject: locale on cygwin
In-Reply-To: <3C5A607F.6060107@halfcooked.com>
References: <Xns91A1A678DEEE3andy47halfcookedcom@203.109.250.24> <20020125140805.GC1252@dothill.com> <3C5A607F.6060107@halfcooked.com>
Message-ID: <20020201182950.GA2196@dothill.com>

Andy,

On Fri, Feb 01, 2002 at 08:31:43PM +1100, Andy Todd wrote:
> Jason Tishler wrote:
> >On Fri, Jan 25, 2002 at 05:22:02AM +0000, Andy Todd wrote:
> >>Jason Tishler probably knows the answer to this off the top of his head,
> >>but ... 
> >
> >Not really, being locale challenged...
> >
> > [snip]
>
> Thanks for the input Jason,

You are welcome.

> but the plot thickens ;-)

Like I said above I am locale challenged, so I will bow out now...

Jason



From DeepBleu at DeepBleu.org  Mon Feb  4 09:27:23 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Mon, 4 Feb 2002 08:27:23 -0600
Subject: Problem while uploading a binary file using cgi module
References: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU>
Message-ID: <u5t6k65m7869d7@corp.supernews.com>

Wang
Please send the code indented next time.  You know, you may know it inside
out.  But poor readers like me do not know your intentions and indenting
makes it easier to read.

>
> elif form.has_key("filename"):
> item = form["filename"]
What are you trying to access by using item=form["filename"]?
try the following:
elif form.has_key("filename"):
    if len(form["filename"].value) > 0:    #Process if file is not an empty
file.
                                                          #byte value is
form["filename"].value
        fp=open('1.jpg','wb')    #you can also open this file in a defined
directory.
                                           # make sure to use os.sep such as
"/" if necessary
        # please note:  access Byte value of files with
form["filename"].value;
        #access file name itself using form["filename"].filename
        fp.write(form["filename"].value)
        fp.flush    #good to use under certain conditions.  check your
favourite python library doc
        fp.close()
DeepBleu




From gh_pythonlist at gmx.de  Sun Feb  3 20:46:12 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 4 Feb 2002 02:46:12 +0100
Subject: curses module
In-Reply-To: <13285ea2.0202031557.244bb70d@posting.google.com>
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <13285ea2.0202030404.5dfe9b66@posting.google.com> <a3j9du$181gf1$1@ID-7529.news.dfncis.de> <13285ea2.0202031557.244bb70d@posting.google.com>
Message-ID: <20020204014608.GA1180@lilith.hqd-internal>

Le 03/02/02 ? 15:57, J.Jacob ?crivit:
> [Karl M. Syring]>
> > This is utter nonsense. Pdcurses simply works on Windows.

True, but this doesn't say anything about building the Python curses
module againsat it.

In fact, I've tried and I can tell you that this isn't plug-and-play.
After seeing lots of undefined symbols, I gave up. It simply wasn't
worth the effort to me.

But you seem to be pretty sure that it's pretty easy, so ...

> Maybe -you- can get it to work?

... I'd also be interested if you get this to work.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 7.3 ?C      Wind: 3.3 m/s



From jjl at pobox.com  Tue Feb  5 14:09:47 2002
From: jjl at pobox.com (John J. Lee)
Date: Tue, 5 Feb 2002 19:09:47 +0000
Subject: REQ : encoding windows cp1252 => iso latin 1
In-Reply-To: <a3p68d$1a56$1@norfair.nerim.net>
References: <a3p68d$1a56$1@norfair.nerim.net>
Message-ID: <Pine.SOL.4.30.0202051904310.26540-100000@mimosa.csv.warwick.ac.uk>

On Tue, 5 Feb 2002, Gillou wrote:
[...]
> My XML parser (pyexpat) does not accept cp1252 character, and I'm looking
> for a function that can translate extra cp1252 characters to the closest ISO
> latin 1 encoding.

I realise this is overkill, but:

There is a simple Python wrapper of recode which I have a setup.py for and
which compiles on win32 with mingw... but though it seems quite stable on
unix, it crashes on windows, and I haven't got debugging working yet on my
machine.

So, if you feel like fixing it, you're welcome to it.


John



From dalke at dalkescientific.com  Wed Feb 13 04:11:57 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Wed, 13 Feb 2002 02:11:57 -0700
Subject: photo id ?!?!
References: <000c01c1b416$b85c9540$445d4540@Dell2> <mailman.1013563639.20235.python-list@python.org>
Message-ID: <a4daoi$np9$1@slb4.atl.mindspring.net>

David Ascher:
>As I mentioned to Andrew in private email, we'll make sure to address
>his concern at future Python conferences (I can't recall if OSCON has a
>similar requirement).

I was at the O'Reilly Bioinformatics conference the week before the
Python conference.  I can say for certain that I didn't need to present
ID, but I didn't go through the main registration because I was a speaker.
I had emailed in my photo for their web site and exchanged emails with
a couple of the conference organizers.  During registration we talked
about those emails, which gave us another way to verify identity much
more than a piece of plastic or paper could.  I was also at the
biohackathon beforehand (where I knew almost everyone), which provided
yet another way to verify identity.

                    Andrew
                    dalke at dalkescientific.com





From jjl at pobox.com  Sat Feb  2 16:49:44 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 21:49:44 +0000
Subject: [repost] win32 debugging-- python21_d.dll not found (fwd)
Message-ID: <Pine.SOL.4.30.0202022147160.597-100000@mimosa.csv.warwick.ac.uk>

[Sorry for repost: I left a rather long gap before replying to Mark's
suggestion, so it probably looked like the problem had been solved, and
I'm sure someone out there must have a clue about this.]

On Fri, 25 Jan 2002, Mark Hammond wrote:

> John J. Lee wrote:
> > I compiled python21_d.dll and python21_d.exe with MSVC, and stuck the
> > _d.dll in C:\winnt\system32 along with python.dll, and the _d.exe in
> > C:\Python21, alongside python.exe.  When I try using python_d.exe, with an
> > extension compiled with mingw32 against libpython21_d.a, I get a message
> > box complaining that python21_d.dll couldn't be found, yet
> > C:\winnt\system32 is listed in the search path given in that very message
> > box.  Are DLLs not found by filename?  Or what?
>
> Check that all dependent modules can be found.  Use the free "depends"
> tool to determine what other DLLs have been referenced by python21_d.dll
> are available.

Well, the only direct dependencies of python21_d according to depends.exe
are (as I'm sure you know!) standard windows dlls: kernel32, user32,
advapi32, shell32 and msvcrtd.

However:

1. I just realised that the error only occurs on importing the extension
   module, not on starting python_d.exe itself, which runs quite happily
   in interactive mode until you import _recode.

2. If, instead of looking at the python21_d DLL, I look at _recode_d.dll
   in depends.exe, I get a dependency on PYTHON21_D with a big question
   mark icon to the left of it.

I was under the impression that Python on Windows resided essentially in
the DLL.  If I have that correct, then since python_d.exe works fine in
interactive mode presumably the problem is with the _recode module rather
than my system setup.  Not exactly a helpful error from Windows...

So, why does depends.exe complain about _recode's dependency on
python21_d, when it seems quite happy with python21_d itself, and when,
what's more, they both sit in the very same directory (C:\winnt\system32)?
I guess if I knew that I'd know why I get the error on importing _recode
with python_d.exe.  I'm guessing this is related to the fact that the
extension was compiled with mingw, but exactly how, I'm not sure.

The extension module in question (_recode) does import ok on windows when
compiled against the non-debugging python, though it causes a memory
access error on actually using it (after my little test routine finishes
without error, around when the interpreter exits, though I don't know
exactly when this happens, obviously, since I haven't been able to debug
it yet).  Since this is somebody else's extension, written by someone more
competent with C than me, and given also the fact that it works fine on
unix, I suspect some cockup by yours truly is causing this, but what it
might be I don't know.

I've tried both a version (of the debugging DLL and .exe) I compiled
myself with MSVC, and using the standard distribution of debugging libs.
I'm using the standard python.org windows Python 2.1.1 distribution.  I'm
restricted to mingw for the extension itself because I haven't got MSVC to
compile it yet.

The error text is as follows (typed in by hand since it seems not to be
possible to cut-and-paste the damn thing):

The dynamic link library python21_d could not be found in the specified path
c:\Python21\DLLs;;C:\WINNT\System32;C:\WINNT\system;C\WINNT;d:\ProgramFiles\mingw\bin;

...etc., and from Python:

$ //c/Python21/python_d.exe
>>>
Adding parser accelerators ...
Done.
Python 2.1.1 (#20, Jul 20 2001, 01:19:59) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import _recode
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: DLL load failed: The specified module could not be found.
[5430 refs]
>>>

Thanks for any help!


John



From laymon at crd.ge.com  Mon Feb  4 15:08:43 2002
From: laymon at crd.ge.com (Marc Laymon)
Date: 4 Feb 2002 12:08:43 -0800
Subject: directly executing a python script in Unix
Message-ID: <60018f02.0202041208.67d3100e@posting.google.com>

I am just learning python.  I have written a few simple scripts which
run fine when invoked using "python myscript.py".  The Python manual
says you can execute a Python script directly In Unix by making the
first line "#!/usr/bin/python", then giving the command
"./myscript.py".  However, when I try this, I get an error "Command
not found.".  I get the same error under both Linux and Solaris. 
Using the same format for perl works fine.  Why doesn't the
"#!/usr/bin/python" line work for Python ?  (perl and python are both
in /usr/bin on my Linux system.)

Thanks.

Marc Laymon


From nelson at crynwr.com  Wed Feb 20 10:55:44 2002
From: nelson at crynwr.com (Russell Nelson)
Date: Wed, 20 Feb 2002 10:55:44 -0500 (EST)
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <oq8z9o2mx6.fsf@carouge.sram.qc.ca>
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
	<oq8z9o2mx6.fsf@carouge.sram.qc.ca>
Message-ID: <15475.50237.602572.554832@desk.crynwr.com>

Fran?ois Pinard writes:
 > [Russell Nelson]
 > 
 > > There are two Python ideoms that bother me:
 > 
 > >     while 1:
 > >         line = file.readline()
 > >         if not line: break
 > 
 > Hello, Russell.  Glad to see you here as well! :-)

Hello, Fran?ois.  Yes, I discovered Python about a year ago, after
pining (for the fjords) for a language without silly syntactic sugar
around its blocks.  Software Practice & Experience had an article over
20 years ago that pointed out that a small set of indentation rules
made block syntax tokens unnecessary.  I thought that was a great idea
... and then much later I discovered Python using those rules.

 > For the above, I like writing:
 > 
 >     line = file.readline()
 >     while line:
 >         ...
 >         line = file.readline()

I agree that this is useful.  A "while line = file.readline()"
creates the expectation that all lines are read at that point.  Then
again, "for line in file:" creates a stronger expectation!

 > >     while line = file.readline():
 > 
 > I would rather not see Python accepting this, first because it makes the
 > code less explicit, but also because it would more easily allow the error
 > of miswriting `==' as `=' unnoticed by the Python code reader.

Yeah.  Perhaps if there was a built-in function which returned the
result of the previous test, but that has its own dangers of being
used in the wrong place.

 > > I just think it would be easier to read programs if they could be
 > > written like this.
 > 
 > Programs may get slightly easier to write, I doubt it would really make
 > them more legible, however.  My feeling is that the current situation is
 > better than alternatives, all considered.

Well, I was just struck by the consistency of allowing "variable =
boolean" in "if" and "while".  Sort of like this in assembly language:

	add ax,bx
	jc  got_overflow
	mov result, ax

You do your test, but you still have the results of the computation
available to you.  Seems like a Good Thing(tm), but I'm happy to be
talked out of it.

-- 
-russ nelson              http://russnelson.com | Crypto without a threat
Crynwr sells support for free software  | PGPok | model is like cookies
521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | 



From peter at engcorp.com  Sun Feb 24 14:32:59 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 14:32:59 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com> <slrna784si.m30.akuchlin@dust.mems-exchange.org> <de3fc1ef.0202201835.7f49f4be@posting.google.com>
Message-ID: <3C793FEB.CD45FF95@engcorp.com>

William Tanksley Google wrote:
> 
> "A.M. Kuchling" <akuchlin at dust.mems-exchange.org> wrote:
> >    William Tanksley Google wrote:
> > > My friend was griping about one battery that Basic comes with but
> > > Python doesn't (at least not that I've seen): graphics.  Yes, I know
> 
> > turtle.py is cute, and requires Tkinter:
> 
> Good suggestion.  Where is it?

Mere seconds with Google could give you the answer.  Try "download"
and "turtle.py" for example.


From martin at v.loewis.de  Tue Feb 26 16:02:20 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 22:02:20 +0100
Subject: socket in python 2.2 vs 2.1
References: <mailman.1014746388.20320.python-list@python.org>
Message-ID: <m3wux0ufc3.fsf@mira.informatik.hu-berlin.de>

"Jason Orendorff" <jason at jorendorff.com> writes:

> (Martin:  255.255.255.255 is the limited broadcast address.)

I see. I thought this was socket.INADDR_BROADCAST, but this doesn't
actually work, either (and apparently never did). I'll investigate why
Python 2.2 rejects to this connect call.

Regards,
Martin


From robin at jessikat.fsnet.co.uk  Fri Feb 15 19:12:21 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 16 Feb 2002 00:12:21 +0000
Subject: win32 DeleteObject ??
References: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk>
 <uit8yuwhk.fsf@ctwd0143.fitlinxx.com>
Message-ID: <a+fHoXAlPab8EwQo@jessikat.fsnet.co.uk>

In article <uit8yuwhk.fsf at ctwd0143.fitlinxx.com>, David Bolen
<db3l at fitlinxx.com> writes
......
>Those are all methods on a pyCDC object which you get from calls like
>CreateDC().
>
>Note that this is all in the win32ui.pyd module, which has its own
>documentation and is located in the Pythonwin subdirectory rather than
>the win32 subdirectory in an installation.
.....
yes I know they're in there, but to use them I need to include a 650Kb
.pyd. At present I'm using anygui and it has quite a small foot print.
-- 
Robin Becker


From marklists at mceahern.com  Sun Feb 10 17:27:17 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 14:27:17 -0800
Subject: A newbie question about class
In-Reply-To: <3c66e1b6_4@corp.newsgroups.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEADKCAA.marklists@mceahern.com>

[Joshua Muskovitz]
> And yet, unless you do use things like "!= None", bugs in packages like
> FixedPoint would propogate.  It's a double edged sword.  Being able to use
> "is not None" is a good way to be able to work around a bug like this (if
> you don't have access to the source, for example), but I would hesitate to
> say that it is "better".

Hmm, here's why I think it's better.  (And I hope I'm not merely repeating
myself--or worse, contradicting myself!--if so, I apologize.)

If I use "is [not] None", I don't have to worry about it.

And to constrain my point, this again assumes that it's my code that's doing
the comparison.  I view the talk of better to be entirely contingent on what
the actual particular situation is, whether there's third party sources,
etc.

Is failing to compare to None a bug?  Maybe, maybe not.  If it's up to me,
I'd compare to None by identity.  Wouldn't you?

Of course, you may not know you're comparing to None:

	f = None
	doFoo(f)

	def doFoo(value):
		if value == g:
			...

In other words, what if None is on the left-hand side?  I guess all I can
say is that I consider that a different question.

I hope I've properly constrained and refined the sense in which I think it's
clearly better to compare to None by identity.

Cheers,

// mark




From * at spam.ruud.org  Thu Feb  7 21:26:12 2002
From: * at spam.ruud.org (ruud de rooij)
Date: 07 Feb 2002 21:26:12 -0500
Subject: python also allowed in google contest
Message-ID: <877kpo7lq3.fsf@bunty.ruud.org>

according to [1], it seems that python will also be allowed in the
google programming contest [2]:

| Google provides source code for the Programming Contest in C++. You
| may also choose to write your code in Java or Python, in which case
| you are responsible for implementing any necessary interface
| code. Your submission must include a Makefile and README, and must
| compile on Linux 2.2 or 2.4 using g++ (for C++ code) or standard Sun
| tools (for Java code) or Python version 2.2.

although it seems that the contest webpage [2] has not been updated
yet to reflect this.

        - ruud

[1] http://groups.google.com/groups?hl=en&group=google.public.programming-contest&selm=257fd06c.0202071532.14b1997b%40posting.google.com
[2] http://www.google.com/programming-contest/
-- 
ruud de rooij | *@spam.ruud.org | http://ruud.org


From fperez528 at yahoo.com  Tue Feb 12 07:44:18 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 12 Feb 2002 12:44:18 +0000
Subject: IPython 0.2.5
References: <a4ainp$gts$1@peabody.colorado.edu> <fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com>
Message-ID: <a4bre5$8ok$1@peabody.colorado.edu>

Terry Reedy wrote:

> You wrote further on your web page: "Hopefully Windows users will
> quickly migrate to XP (which is actually quite good in many respects)
> and we'll bury Win95 and children were they should have been buried
> years ago."
> 
> Don't hold your breath.  XP is a greater resource hog than ME is a
> greather resource hog than W98 is a greater resource hog than W95, so
> many (most?) users can't upgrade, even if they would want to pony up
> the upgrade fee and run the risk (in some cases, certainty) of
> disabling a hardware system and set of software than currently work.
> (XP does NOT run everything that runs under Wxx.)

Well, my feeling is that things aren't all that bad, given you have *memory*. 
I recently upgraded my oldie desktop to XP, an aging PIII-450 with a slowish 
hard disk, but with 256Mb of ram. And it's just fine. I turned off most of 
the eye candy and it runs quite well. It feels barely slower than it did 
under 98, and the gains in stability are priceless.

And it had lots of old hardware: a parallel port CD burner, an old Epson 
Photo 700 printer (parallel), an old winmodem without driver disks, etc. I 
had to get various updates on the web and move the printer to the  usb port 
(with a parallel/usb converter), but now *everything* works 100%.

I'm sure others may have had other experiences, but the key factor seems to 
be memory. Have >=256Mb of ram, a processor >= 400 Mhz and you should be 
fine. And most hardware and software makers from what I saw have free patches 
for XP on their websites.

At the end of the day,  I'm quite happy with the results. At least it doesn't 
crash every 5 minutes!!!

> That said, I may download and see what happens on my Win98 system.

Let me know how it goes. I didn't mean to sound hostile to Win9x users, it's 
just that I'm sick of dealing with the instability of that thing. And now 
that I've migrated to XP, I have no way of testing under 98. But with some 
luck things should work, even the unix/windows-dependent code is minimal, so 
I hope it will be fine with 98. If you find a problem and can fix it, send me 
the patch and I'll include it in a new release.

Cheers,

F.


From pj at sgi.com  Mon Feb 11 19:18:00 2002
From: pj at sgi.com (Paul Jackson)
Date: 12 Feb 2002 00:18:00 GMT
Subject: Reading Powerpoint on Linux [was: Slides on the web]
References: <mailman.1013135593.4897.python-list@python.org>
Message-ID: <a49mvo$auq6q$1@fido.engr.sgi.com>

Could someone care to recommend a tool for reading Powerpoint
slides on Linux (short of running a Windows emulator ...)?

Thanks.

P.S. -- If it weren't Guido, with his reputation aptly
	described elsewhere here last week as:

	> Yes.  And nine times out of ten, if you think you disagree
	> with Guido's intuition, you'll change your mind later

	I'd likely protest the usage of Powerpoint here.

	But it was Guido, and I look forward to viewing his
	presentations.

Paul "resistance may be futile, but so what?" Jackson

-- 
-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj at sgi.com> 1.650.933.1373


From wookiz at hotmail.com  Thu Feb 14 05:24:50 2002
From: wookiz at hotmail.com (wooks)
Date: 14 Feb 2002 02:24:50 -0800
Subject: Windows Message Hooks
Message-ID: <8852d33c.0202140224.13840b45@posting.google.com>

Can anyone direct me to a resource where I can find info Python
support for the above.

Thanks.


From ngps at madcap.netmemetic.com  Wed Feb 20 09:56:26 2002
From: ngps at madcap.netmemetic.com (Ng Pheng Siong)
Date: 20 Feb 2002 14:56:26 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <3C7368A0.5050702@bath.ac.uk> <KzNc8.17085$Zp2.1759804@atlpnn01.usenetserver.com> <a50aun$32h$1@ariadne.rz.tu-clausthal.de>
Message-ID: <a50deq$mke$1@violet.singnet.com.sg>

According to Lutz Schroeer  <Lutz.Schroeer at tu-clausthal.de>:
> Seems I missed that thread. After thinking it over I came to the conclusion
> that using cookies is the most practical way. It's a pity, I hoped there
> would have been a different solution, but passing an ID through all my URL
> doesn't really attract me ...or maybe... mmh, I really don't know. I think I
> try the URL-ID method and if I get completely messed up I use the cookies.

In summary, these are the three ways to store session-ids.

1. Use cookies.
2. Use encoded URLs.
3. Use a hidden field.

Take a look at the current snapshot of M2Crypto; the AuthCookie thingy may
help.

Cheers.


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



From jkraska at san.rr.com  Thu Feb 14 12:30:48 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 14 Feb 2002 17:30:48 GMT
Subject: watching the number of references to an object
References: <1013700953.2880.0.nnrp-01.c1c3e191@news.demon.co.uk>
Message-ID: <jhtn6u8lkbpviucvml02m8inj1khefbjkk@4ax.com>

On Thu, 14 Feb 2002 15:50:04 -0000, "Paul Brian" <paul1brian at yahoo.com> wrote:

>Dear all,
>
>Does anyone know how to monitor the references pointing to an instance of an
>object?

In addition to the other answers, please note that the number of
referrers will often be 1 higher than what you expect: the function
for getting the references itself temporarily grabs a reference,
IIRC.

C//



From ajeru at vknn.org  Tue Feb 19 22:01:19 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 19:01:19 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <slrna758ng.d1u.quinn@yak.ugcs.caltech.edu>
Message-ID: <24c39b2c.0202191901.5f6643f6@posting.google.com>

> It's only of the first argument, but that's what mono-dispatch means.
> Static typing having anything to do with multi-dispatch is incorrect,
> though.  The best-known object system with multi-dispatch is CLOS, which
> is certainly dynamic.

I just looked up what CLOS does. I don't know if you call that static
typing or not. In any case, CLOS does have type declarations in the
source code which is what I want and what enables the features I'd
like to see in Python. For example in CLOS you can have the following
method declarations:

 (defmethod Area ((C Circle)) (* pi (Square (Radius C)))
           (defmethod Area ((R Rectangle)) (* (Width R) (Height R)))
           (defmethod Area ((Sq Square)) (Square (Width Sq)))
           OR (defmethod Height ((Sq Square)) (Width Sq))
           
Alexander


From mhammond at skippinet.com.au  Mon Feb  4 05:43:28 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 10:43:28 GMT
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com> <a3hnft$cgu$1@newshost.accu.uu.nl> <3C5CC325.7B3FC988@engcorp.com> <flhp5u4jalahtcm1att17i5vuj2evqoott@4ax.com> <a3j97s$d98$1@newshost.accu.uu.nl>
Message-ID: <3C5E6620.8070405@skippinet.com.au>

Martijn Faassen wrote:


> By the way, to my great and enduring happiness I managed to turn it off
> under XFree86. Even the light doesn't go on when I press it! I've actually 
> remapped it to be an extra control key, but perhaps escape would be better
> for vi use. But I don't really dare getting used to it as I'd screw up
> as soon as I'd switch to anyone else's keyboard..

I discovered that in Win2k, under "Accessibility Options", there is a 
"toggle keys" option.  Enabling this plays a beep whenever caps or num 
lock is pressed.  This seems a good solution.

Mark.



From tismer at tismer.com  Fri Feb 22 14:42:01 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 22 Feb 2002 20:42:01 +0100
Subject: Request for Stackless Sponsorship
Message-ID: <3C769F09.20705@tismer.com>

Dear Stackless Python Community,

As you might have noticed, Stackless has changed quite
much:

- Continuations are gone.
- Platform independence is gone.
- Non-completeness is gone.
- Python-version-dependance is gone.
- Incomprehensibilits of code is gone.
- per-Python-Version maintenance is gone.

Stackless is now becoming a rock-solid implementation
of parallelism. It will boost many applications to a
performance that was not yet expected to be exposed
by an interpreted language.

Besides all these good news, there is a draw-back.
I have put many weeks of fulltime-hacking into
this project, and I cannot afford to continue.
But I want to continue, since it all looks so very
doable, so absolutely right to do. My heart is
beating for this project, again and more than ever.

Please, I need some corporate sponsorship to continue
my work. I assure you, it will work out. What I'm working
on will be the basis for many applications. I have been
through all the prototypes, they are no longer needed.
This will become production code from the first day.

This is an absolutely urgent request:
To keep this project carrying on, I need to make a living.
I am hereby seriously asking for companies which
would consider to support me by some amount of money.
Every sponsor will get his own page on stackless.com,
where they can show their use of stackless to any extent
if they like. I cannot continue unless I continue
full-time for a couple of weeks. I don't want to stop now,
since I'm at the very best solution ever, and it is infinite
fun for me. Too bad that fun doesn't feed.

Please HELP!
If you want Stackless to succeed right now with an
extent of years, please try to support my work
with some amount of corporate sponsorship. You will
get the best that I can offer: Stackless running on
your platform, with your individual modifications and
extensions, at incredible speed.

Thanks in advance - chris

p.s.: Sorry for abusing this list for this request.
I wouldn't do so if I wouldn't have to.

p.p.s.: Guido, forgive me.

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From threeseas at earthlink.net  Tue Feb  5 23:54:40 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:40 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <Xns91AB378D110ABahmebahhotmailcom@198.99.146.10>
Message-ID: <1417.802T2216T45354threeseas@earthlink.net>

On 04-Feb-02 08:26:21 ameoba <ahmebah at hotmail.com> wrote:
>"Timothy Rue" <threeseas at earthlink.net> wrote in
>news:922.799T2387T11875593threeseas at earthlink.net:

>> Its damn clear that there is a need to remove the sort of people posing
>> to be professional who are in essence just good liars, abstraction
>> manipulators, from the industry.

>WTF are you talking about?  The only thing that I've managed to come up
>with about your project is that it -heavily- abstracts the process of
>computing to the point where joe-6pack should be able to make a computro do
>anything.

>If you feel there is something wrong with being an 'abstraction
>manipulator' perhaps you shouldn't be talking to computer programmers.
>Computing is nothing but the building and manipulation of abstractions.  If
>it weren't for abstractions, then we'd still be swapping vaccuum tubes to
>program a computer.



>Don't even get me started on the hipocrasy of -you- using 'abstraction
>manipulator' as an insult.


When you bit flip an ameoba, does it look the same?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From googlegroups at sixtyten.org  Wed Feb  6 11:25:17 2002
From: googlegroups at sixtyten.org (Allan Crooks)
Date: 6 Feb 2002 08:25:17 -0800
Subject: Win32evtlogutil.ReportEvent() - more info please?
References: <p0815u82mna6i5dk1v0bhb3smvvakssqes@4ax.com> <3C50E89C.4090304@skippinet.com.au> <5l525uo5smpoj5lhiuu2htk1rcuammbhe2@4ax.com> <3C5136C6.103@skippinet.com.au>
Message-ID: <304d20df.0202060825.6f1fe7b5@posting.google.com>

Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C5136C6.103 at skippinet.com.au>...

> So win32evtlog.pyd is such a DLL.  pythonservice.exe is also such a 
> file.  win32evtlog.pyd has simply lots of "%1" messages - ie, no text at 
> all.   See 
> http://www.pythonpros.com/cgi-bin/cvsweb.cgi/PyWin32/win32/src/win32evtlog_messages.mc. 
>   pythonservice.exe has a few service related messages too - see 
> http://www.pythonpros.com/cgi-bin/cvsweb.cgi/PyWin32/win32/src/PythonServiceMessages.mc

Could you specify an eventId that will actually work for win32evtlog?
I've can't seem to find one that's valid at all... I've tried using
other DLL's, which work, it's just win32evtlog doesn't seem to have
any messages I can use (either that, or I'm doing something seriously
wrong)...

Thanks,
Allan.


From paul at boddie.net  Wed Feb 20 11:22:38 2002
From: paul at boddie.net (Paul Boddie)
Date: 20 Feb 2002 08:22:38 -0800
Subject: XML-RPC over HTTPS/SSL
References: <8rKc8.4$5vh.135400448@news.telia.no>
Message-ID: <23891c90.0202200822.43655a9b@posting.google.com>

"Thomas Weholt" <thomas at gatsoft.no> wrote in message news:<8rKc8.4$5vh.135400448 at news.telia.no>...
> 
> I keep seeing stuff about SSL/HTTPS-support in standard python, but ... ??
> Do you compile the interpreter with OpenSSL etc ?? My clients and servers
> will most probably run 60% on Linux and 40% windows, and re-compilation
> isn't always feasable for windows-users.

When building Python from source, it is possible to link to OpenSSL
and thus support secure sockets ("https" URLs) and so on - that's how
urllib supports such URL schemes, as far as I can tell. Whilst
M2Crypto provides replacement modules for urllib, I found that the
replacement functions and methods would cut received data at 4096
bytes, and since I was also using Windows without a C compiler to
hand, I needed to find something which plugged in to the Python
distribution I have (which is ActivePython 2.1). Fortunately, Robin
Dunn has provided some drop-in replacement libraries:

  http://alldunn.com/python/

Hats off to Robin, I say! :-)

Paul


From scarblac at pino.selwerd.nl  Tue Feb 26 12:05:01 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 26 Feb 2002 17:05:01 GMT
Subject: gnuplot woes, a second story
References: <mailman.1014696821.3069.python-list@python.org>
Message-ID: <slrna7nfrq.ouu.scarblac@pino.selwerd.nl>

chajadan <python at chajadan.net> wrote in comp.lang.python:
> When I try to run gnuplot from my Python2.2 installation, I get an error 
> (listed below)
> 
> I'm on windows 98, I installed NumPy, I unzipped gp371w32.zip into my main 
> python directory, and I tried changing the command to the gnuplot program 
> FROM gnuplot_command = 'pgnuplot.exe' TO gnuplot_command = 
> 'c:\Python20\gp371w32\pgnuplot.exe'

First guess: if this is really what you wrote, you need to double the
backslashes, use \\ instead of \ (since they're the escape sign in Python
strings).

-- 
Remco Gerlich


From lac at strakt.com  Fri Feb 15 11:56:12 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 15 Feb 2002 17:56:12 +0100
Subject: Tkinter Resize 
In-Reply-To: Message from jmsun@bioeng.ucsd.edu (Jeffrey) 
   of "14 Feb 2002 13:29:52 PST." <3daa1aa8.0202141329.63fdc057@posting.google.com> 
References: <3daa1aa8.0202141329.63fdc057@posting.google.com> 
Message-ID: <200202151656.g1FGuCGo006379@ratthing-b246.strakt.com>

You want to bind to the <Configure> event.
http://www.pythonware.com/library/tkinter/introduction/events-and-bindings.htm#AEN676

Laura Creighton



From jeff at ccvcorp.com  Thu Feb 14 17:19:49 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 14:19:49 -0800
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org>
Message-ID: <3C6C3805.37BD87FB@ccvcorp.com>


Gustavo Cordova wrote:

> >
> > I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> >
> > When I do a readline the I get back a string.
> >
> > How do I convert this string to a dictionary?
> >
> > Thanks.
> >
>
> You eval() it?
>
> >>> d = { "one":1, "two":2, "three":3 }
> >>> print d
> {'three': 3, 'two': 2, 'one': 1}
> >>> S = str(d)
> >>> print S
> {'three': 3, 'two': 2, 'one': 1}
> >>> dd = eval(S)
> >>> print dd
> {'one': 1, 'three': 3, 'two': 2}
> >>> d is dd
> 0
> >>>

This works, but is generally a Bad Idea(tm).  The problem
with eval/exec on arbitrary strings (and any string you read
in from a file is arbitrary) is that it's hard to be
positive that what you're eval/exec-ing is what you expect.
And if it's *not*, then many, many very bad things can
happen.  (Imagine someone "accidentally" replacing that
textfile with one that contains the line "import
os;os.system('rm -s /')" -- suddenly your entire filesystem
is blank....)

If you want to store the contents of a dictionary on a
permanent disk file, you should look at the Pickle and
Shelve modules.  If you're doing this a lot, or with lots of
data, then try using one of the generic db modules (anydbm,
bsddb, etc).

Jeff Shannon
Technician/Programmer
Credit International




From mikeb at mitre.org  Thu Feb  7 18:49:13 2002
From: mikeb at mitre.org (Mike Brenner)
Date: Thu, 07 Feb 2002 18:49:13 -0500
Subject: Python COM for Microsoft Project (the final solution)
Message-ID: <3C631279.C5A41BAA@mitre.org>

Last month the people on this list helped me out (see the history of this thread). 

Thanks. Here is the code that worked to have a Microsoft Project file do a SAVE AS tab-delimited text.

After following the advice from this thread, the next hurdle was to realize that tab-delimited text, which is expressed as 'format="MSProject.TXT"' in Visual Basic, is expressed as 'format=3' in COM.

My next challenge is to figure out why it is not automatically choosing the 'Map="Better Task Information". Instead, it is popping up a window and making the user select it with the mouse. That is the only remaining barrier to do a stream 100% automated SAVE AS COM commands.


def saveAs_tabDelimited(schedule):
    from win32com.client import Dispatch    
    import os
    format="MSProject.TXT";format=3;oo=0
    mp = Dispatch("MSProject.Application");mp.Visible = 0
    mp.Application.MapEdit(Name="Better Task Information",DataCategory=oo,
                           TableName="Task_Data",
                           FieldName="Start", ExternalFieldName="Start_Date",
                           Create=0, OverwriteExisting=1,
                           ExportFilter="All Tasks", ImportMethod=oo, HeaderRow=1,
                           AssignmentData=0, TextDelimiter="\t", TextFileOrigin=0,
                           UseHtmlTemplate=0, IncludeImage=0)
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Unique ID", ExternalFieldName="Unique_ID")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Name", ExternalFieldName="Task_Name")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Finish", ExternalFieldName="Finish_Date")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Actual Start", ExternalFieldName="Actual_Start")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Actual Finish", ExternalFieldName="Actual_Finish")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="Predecessors", ExternalFieldName="Predecessors")
    mp.Application.MapEdit(Name="Better Task Information", DataCategory=oo,
                           FieldName="% Complete", ExternalFieldName="Percent_Complete")
    print; print "Files to save in tab-delimited text format"
    for sched in schedule:
        filename=sched[0]
        text_filename=sched[1]
        try: os.remove(text_filename)
        except: pass
        print "    "+filename
        mp.FileOpen(filename)
        mp.Application.FileSaveAs(Name=text_filename, Format=format, Map="Better Task Information")
        mp.Application.FileClose()
    mp.Application.FileQuit()
    print "MP application is closed"
    del mp




From jwbaxter at spamcop.net  Mon Feb  4 01:10:09 2002
From: jwbaxter at spamcop.net (John W. Baxter)
Date: Sun, 03 Feb 2002 22:10:09 -0800
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>
Message-ID: <030220022210095545%jwbaxter@spamcop.net>

In article <_JU68.2091$3E5.208453 at newsread2.prod.itd.earthlink.net>,
Stede Troisi <stede at computer.org> wrote:

> I was thinking of doing an E-book (because I don't like the high price of
> regular books) on an introduction to programming and artificial intelligence
> in Python.
> 
> I was wondering if there would be interest in such a e-book and if it would
> sell?  I would love to here everyone's opinion.

It would have to be quite compelling (based on reviews) before I would
go to the trouble of obtaining the ability to read ebooks on some
machine, and then the trouble of reading at half speed (which is what
seems to be my onscreen/onpaper reading speed ratio).

I doubt that the ebook's time will have come before I'm gone (age 62).

  --John


From jeff at ccvcorp.com  Fri Feb 15 20:07:23 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 15 Feb 2002 17:07:23 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <a4k2si$1skr$1@norfair.nerim.net> <926k4a.3a.ln@127.0.0.1>
Message-ID: <3C6DB0CB.CC52E6B1@ccvcorp.com>


Carl Banks wrote:

> Gillou wrote:
> >>
> >> * NO LINE NUMBERS! Enough said.
> > ...
> >>   IF condition GOSUB line-number (I never did understand the necessity
> >>   for that kind of statement until well after I'd stopped using BASIC
> >>   entirely and moved to more "modern" languages like... Pascal).
> >
> > Only old men like you and me remember this
>
> Hi.
>
> I remember using calculated GOSUB, something like this, in GW-BASIC:
> 220 ON LEVEL GOSUB 1000,2000,3000,4000,5000,...
> It was for a crude level-based video game.  I'm 25 now.

I (regrettably) currently routinely use a (proprietary) dialect of BASIC (for
our proprietary database system), in which GOSUB remains the most advanced
control structure.  (And in which, all variables are global.  >shudder< )

Needless to say, I try to spend as much of my time working on the client side
of things, in Python...  :)

Jeff Shannon
Technician/Programmer
Credit International




From joshm at taconic.net  Thu Feb 28 21:32:12 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Thu, 28 Feb 2002 21:32:12 -0500
Subject: Want to monitor process..
References: <87550ef1.0202272315.15779bbd@posting.google.com> <f2lf8.385$Az6.1854@news.hananet.net> <97ae44ee.0202280901.3153bdfc@posting.google.com>
Message-ID: <3c7ee5d0_4@corp.newsgroups.com>

> I'm chuckling to myself. This is exactly what I'm doing at the moment
> too and it sucks. I'm so glad you asked this question, Haeyoung, and
> described it so well :)

I described a solution about six weeks ago (with code!).  Here's the link
(or search on google yourself for SingleInstance and Muskovitz):

http://groups.google.com/groups?q=SingleInstance+Muskovitz+group:comp.lang.p
ython

Basically, your app creates a lockfile, and periodically touches it.  An
external app can check the timestamp on the file, and if it is too old, then
your process has hung up and could be killed/restarted.  Alternately, you
can stick your app directly in cron.  When a new instance starts up, it can
check the lock and if things are ok, then it can just kill itself.

The code works on NT and Solaris.  I haven't tried other platforms.
--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From eppstein at ics.uci.edu  Wed Feb  6 22:39:20 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 06 Feb 2002 19:39:20 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu> <eppstein-80FFB2.17393106022002@news.service.uci.edu> <a3sq4m$1an2ll$1@ID-11957.news.dfncis.de>
Message-ID: <eppstein-FBDBD3.19392006022002@news.service.uci.edu>

In article <a3sq4m$1an2ll$1 at ID-11957.news.dfncis.de>,
 "Emile van Sebille" <emile at fenx.com> wrote:

> You can squeeze out another (ugly) 15-20% performance gain
> by making almost everything local and taking out the len check:
> 
> def flatten(x, isseq=operator.isSequenceType,
>     concat=operator.concat, reduce=reduce, map=map):
>     if isseq(x) and x[0] != x:
>         return reduce(concat, map(flatten, x), [])
>     else:
>         return [x]
> 

No, the len check is important, otherwise you get a crash when the input 
contains [] (that's also why I included the third arg to reduce).

Anyway, the reason I said that this is inefficient is that there are ways 
of making flatten take time linear in the length of the input, while the 
code I wrote is quadratic in the worst case, and its worst case (an 
already flattened list) is not especially unlikely. Using an algorithm 
with better asymptotic efficiency can make an improvement of much more 
than a few percent, for large inputs. For instance, when I test the code 
below, for n=10^5, my functional flatten takes 425 seconds while the mixed 
functional-imperative flat2 takes only 1:

def flatten(x):
        if operator.isSequenceType(x) and (len(x) != 1 or x[0] != x):
                return reduce(operator.concat, map(flatten, x), [])
        else:
                return [x]

def flat2(x):
        L = []
        def recurse(x):
                if operator.isSequenceType(x) and (len(x)!=1 or x[0]!=x):
                        map(recurse, x)
                else:
                        L.append(x)
        recurse(x)
        return L

def testflat(n):
        L = [0]*n

        start = time.clock()
        F = flatten(L)
        print "flatten time for", len(F), "items:", time.clock() - start
        
        start = time.clock()
        F = flat2(L)
        print "flat2 time for", len(F), "items:", time.clock() - start
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From aahz at panix.com  Thu Feb 28 08:57:05 2002
From: aahz at panix.com (Aahz Maruch)
Date: 28 Feb 2002 05:57:05 -0800
Subject: PEP 276 (was Re: Status of PEP's?)
References: <mailman.1014854027.3045.python-list@python.org>
Message-ID: <a5lcvh$4tt$1@panix2.panix.com>

[posted & e-mailed]

In article <mailman.1014854027.3045.python-list at python.org>,
 <James_Althoff at i2.com> wrote:
>
>Rather than restart the entire debate on PEP 276, please send me
>suggestions for specific changes to the PEP.

My primary complaint about the PEP is that the following idiom is
awkward, ugly, and prone to misunderstanding:

    if i in 5:
        ...

Please add this to your list of objections.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From simon at ping-pong.budig  Tue Feb  5 10:53:46 2002
From: simon at ping-pong.budig (Simon Budig)
Date: 5 Feb 2002 16:53:46 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net>
Message-ID: <3c60000a$1@si-nic.hrz.uni-siegen.de>

In article <uheowosxs.fsf at python.net>, Michael Hudson wrote:
> Simon Budig <simon at ping-pong.budig> writes:
>> Michael Hudson wrote:
>> > Simon Budig <simon at ping-pong.budig> writes:
>> [...]
>> >> BTW: The stupid fixed width limit of SF messed the patch up.
>> >> Maybe creating an attachment is more useful...
>> > 
>> > Well, anyone can see roughly what it does.  If it gets accepted (or
>> > more likely, noone complains for a couple of weeks), I'll check it in
>> > from your patch.
>> 
>> Great! If you do this, please make sure, that a similiar patch goes
>> into python 2.2.
> 
> Well, it can't go into 2.2.  If you mean 2.2.1, that's a little
> greyer; there might be people out there who rely on the current
> behaviour.

Hmm. Maybe it is possible to export this variable as an optional
parameter to readline.set_completer with the default on the
current behaviour? I think it is really annoying, that readline
in the current python-module implementation insists on completing
filenames when the completer returns no matches.

Bye,
	Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From andda715 at student.liu.se  Wed Feb 20 11:23:22 2002
From: andda715 at student.liu.se (Anders Dahlberg)
Date: Wed, 20 Feb 2002 17:23:22 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <a50hd2$6vt$1@panix2.panix.com>
Message-ID: <a50im3$bf1$1@news.island.liu.se>

"Aahz Maruch" <aahz at panix.com> skrev i meddelandet
news:a50hd2$6vt$1 at panix2.panix.com...
> In article <m2wux85hud.fsf at desk.crynwr.com>,
> Russell Nelson  <nelson at crynwr.com> wrote:
> >Henrik Motakef <henrik.motakef at web.de> writes:
> >> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> >>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> >>>>
> >>>> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> >>>> czignat?  Classifero en bragna!
> >>> Woas da Deifi wos des hoasn mog.
> >> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> >
> >Even better: we could switch languages within threads, to *really*
> >keep the monolinguists confused.
>
> B'vakasha

Ja, varf?r inte?





From fperez528 at yahoo.com  Sat Feb 16 14:26:52 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sat, 16 Feb 2002 19:26:52 +0000
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net>
Message-ID: <a4n4gu$eel$1@peabody.colorado.edu>

I.J. wrote:

> Is there a module for python that grants ussage of decimal numbers with more
> than 14 decimal places, including all scientific functions to do with those
> numbers?

No. Python uses C doubles underneath, and those give you only 14 places. 
There are, however, python bindings for the Gnu multiple precision libraries 
(google...). You'd have however to implement scientific libraries to use the 
arbitrary precision function calls instead. Not an easy task.

What you're asking for isn't particularly easy to find outside of systems 
like Mathematica which implement their own arithmetic subsystem, and can 
therefore do *any* calculation in arbitrary precision. Matlab, IDL, Octave, 
etc., all are limited to doubles. Designing a generic, arbitrary precision 
scientific library is a massive undertaking.

Cheers,

f.


From starx at pacbell.net  Sat Feb  9 22:23:03 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 10 Feb 2002 03:23:03 GMT
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4> <slrna6bnf4.9tv.quinn@hork.ugcs.caltech.edu>
Message-ID: <Xns91B0C533368E9RASXnewsDFE1@130.133.1.4>

Quinn Dunkan || Sat 09 Feb 2002 06:41:41p:

> On 9 Feb 2002 18:21:31 GMT, Philip Swartzleonard <starx at pacbell.net>
> wrote: 
>> [Origional]
 
> I recommend leaving the 'type' attributes out.
> 
> class Message: pass
> class Fighto(Message): pass
> class Spello(Message): pass
> 
> class Receiver: # i before e except after c!
>     def message(self, msg):
>         msgname = 'rec_' + msg.__class__.__name__
>         if hasattr(self, msgno):
>             # recommend leaving out the callable(), why let people
>             define # non callable rec_ attrs?
>             getattr(self, msgname)(message)

Oh, so that's how you do that. I wrote this only using what knowledge i had 
available on hand... that's what i wanted, but didn't know how at the time, 
i think :)

Oh, and i laredy gave up on the 'callable' thingie, and stuck something 
like this in the comments:

Bla bla bla... Do not name any attributes of subs of this class anything 
beginning with "rec_" (except functions that take one 'message' paramater)
or you may be burned. Thank you.

 
> To the other poster who had the idea of using dicts to alias messages,
> you can do the same with:
> 
> class F(Receiver):
>     def rec_fighto(self, message):
>         print "FIGHTO!!"
>     rec_hacko = rec_fighto

You have gained a point in understanding python. Only a thousand more!
That's a really nice way of doing something like that, i think :)

>>Also, is there any real reason to have the message classes inherit when
>>the only thing that seems to identify them is an attribute? 
> 
> No, unless you wanted messages to have some messages of their own,
> which I assume you did since otherwise you could have made them
> strings.  If you do want all Messages to respond to a few common
> messages (small 'm'), like 'sender' or 'time_sent', then inheritance is
> natural. 

Well, i definatly wanted to attach data to them... and just about 
everything has a sender, i think. I haden't thought of that and inheritance 
at the same time i guess :)



-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From daniels at mindspring.com  Wed Feb 20 21:32:04 2002
From: daniels at mindspring.com (Alan Daniels)
Date: 20 Feb 2002 18:32:04 -0800
Subject: HTMLifyer for Python code?
References: <mailman.1014238715.7462.python-list@python.org>
Message-ID: <5d283efe.0202201832.334d7609@posting.google.com>

"Bjorn Pettersen" <BPettersen at NAREX.com> wrote in message news:<mailman.1014238715.7462.python-list at python.org>...

> I just browsed through the Vaults, and did a quick Google search, but I
> couldn't find a tool for converting Python source code to
> syntax-highlighted HTML. Surely someone must have written something like
> this?

VIM (www.vim.org) will syntax-highlight Python code, and has a
"convert to HTML" option under its "Syntax" menu. It helps if a
VI-style editor is your cup of tea, I suppose, but if all you
want is the HTML conversion, it's a quick download that should
get the job done pretty quickly.


From op73418 at mail.telepac.pt  Wed Feb 27 14:02:08 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Wed, 27 Feb 2002 19:02:08 +0000
Subject: Status of PEP's?
Message-ID: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com>

Anyone knows how is the "acceptance status" of the following PEP's, that
is, is there any idea if they will be accepted for the next Python
realease:

PEP 276: Iterator for ints.
(I really love this one - as simple and "obvious" as it gets)

PEP 279: Enhanced generators

Thanks,
Gon?alo Rodrigues


From ykingma at accessforall.nl  Sun Feb 17 06:23:35 2002
From: ykingma at accessforall.nl (Ype Kingma)
Date: Sun, 17 Feb 2002 12:23:35 +0100
Subject: Java vs Python ... ???
References: <mailman.1013789007.14325.python-list@python.org>
Message-ID: <3C6F92B0.D34787F3@accessforall.nl>

Steve,

"Orr, Steve" wrote:
> 
> What about Jython? Do you get some of the benefits of Java along with
> easier, more productive Python programming?

You get allmost all of the benifits of Java and allmost all of the benifits
of Python. Jython is not Java vs. Python, it's Java and Python.

> 
> For instance, with the Oracle Application Server (not the database) you
> supposedly can have all your Java apps load balanced "for free" without
> having to write load balancing code. So I was wondering the other day, since
> Jython compiles to bytecode, will it run under applications servers (such as
> Oracle) and enjoy the same Java features those servers provide?
> 

To get that working you need to:
- write a java method that calls jython code (see www.jython.org),
- have that method run in a load balanced way with your server.

> I want to have my cake and eat it too!

What are you waiting for?

Have fun,
Ype


From sheila at thinkspot.net  Sat Feb 23 21:12:14 2002
From: sheila at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 18:12:14 -0800
Subject: [Tutor] Working with Email question  [new 'email' module in Python 2.2]
In-Reply-To: <LNBBLJKPBEHFEDALKOLCEENLNOAA.tim.one@comcast.net>
References: <6AD3F866286@kserver.org> <LNBBLJKPBEHFEDALKOLCEENLNOAA.tim.one@comcast.net>
Message-ID: <707E65554AE@kserver.org>

On Sat, 23 Feb 2002 20:17:04 -0500, Tim Peters <tim.one at comcast.net>
wrote about RE: [Tutor] Working with Email question  [new 'email' module
in Python 2.2]:

> Sheila, if you belive there's a bug, you should open a bug report. 
...<snip>...
> A bug report doesn't need a lot of argument unless it proves to be
> controversial; but since you haven't yet gotten this to the author's
> attention, you can't yet know whether Barry will love or hate the complaint.
> It doesn't really matter what Oleg or I may think:  we're not the experts
> for this package.

OK, bug report opened.
http://sourceforge.net/tracker/index.php?func=detail&aid=521937&group_id=5470&atid=105470

Sorry for not handling this in the proper manner. I guess I was looking
for corroboration or comments or...heck...I dunno.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From gs at styrax.com  Thu Feb 28 14:00:04 2002
From: gs at styrax.com (Garry Steedman)
Date: Thu, 28 Feb 2002 19:00:04 +0000
Subject: Python DB that supports numbers
In-Reply-To: <20020228125553.A2852587@vislab.epa.gov>
Message-ID: <3C7E7E34.24356.6EB01AA@localhost>

randall,

gadfly: http://www.chordate.com/gadfly.html

hth,

garry

On 28 Feb 2002, at 12:55, Randall Hopper wrote:

From:           	Randall Hopper <aa8vb at yahoo.com>
To:             	python-list at python.org
Subject:        	Python DB that supports numbers
Date sent:      	Thu, 28 Feb 2002 12:55:53 -0500

> I'm looking for a very simple Python DB which supports:
> 
>      1) ints and floats as keys and values, 
>      2) performs DB access in the client, and 
>      3) stores the DB on disk.  
> 
> Just a simple DB for quick projects without the setup headache of
> server-based database systems.
> 
> For example, on the following table I just want to: "select * where
> column_1 is 1".  I'm using raw binary files now, but I'd prefer
> something more self-describing like an SQL table.
> 
>      1       10       1.814  177.1   13.61   73.2
>      1       5        1.624  182.5   -9999   -9999
>      1       2        1.259  185.9   13.24   73.9
>      2       10       2.672  181.4   13.59   72.5
>      2       5        2.423  183.2   -9999   -9999
>      2       2        1.902  190.7   13.27   73
>      3       10       3.208  181.9   13.46   72.1
>      3       5        2.787  186.6   -9999   -9999
>      3       2        2.17   194.7   13.16   72.3
> 
> Does anything like this exist in Python?
> 
> Thanks,
> 
> Randall
> 
> -- 
> Randall Hopper
> aa8vb at yahoo.com
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list


+-------------------------------------------+
  Garry Steedman      mailto:gs at styrax.com  
  Styrax Associates  http://www.styrax.com/ 

  "The Good Man has no shape."            
+-------------------------------------------+



From mwh at python.net  Thu Feb 21 06:43:24 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 11:43:24 GMT
Subject: Reloading modules
References: <a51tt1$msq$1@news1.ucsd.edu>
Message-ID: <u664rm510.fsf@python.net>

Jeff Davis <jdavis at empires.org> writes:

> Is there a way to say that all calls to import should check for a newer 
> version, if a newer one exists then completely drop the old one and import 
> the new one?

An import hook could do this.  Wouldn't be trivial to write, but not
too hard, either.

> I am aware of the reload() function, but it has enough caveats that it 
> always leaves lingering questions.

Um, if reload() has problems, how would the above be much better?  I
suppose it would get round the having-to-reload-submodules problem.

> I usually end up just stopping and restarting the interpreter, which
> slows development/debugging. I would just like to be sure that any
> time after I import a module, I am using the newest version
> available at the time of import.  

> I am also unsure as to whether
> modules imported within a reloaded module are also reloaded.

Not if you call reload() on it, no.  The above-alluded to import hook
would solve this problem.

Cheers,
M.

-- 
  ARTHUR:  Don't ask me how it works or I'll start to whimper.
                   -- The Hitch-Hikers Guide to the Galaxy, Episode 11


From stephen.arndt at bethsteel.com  Thu Feb  7 14:37:26 2002
From: stephen.arndt at bethsteel.com (Stephen Arndt)
Date: 7 Feb 2002 11:37:26 -0800
Subject: IIS ASP Python 2.2 script works first time only
Message-ID: <f36760cb.0202071137.61bc8027@posting.google.com>

Have a simple Python script hello.asp to create a web page which
displays current date and time.

<%@LANGUAGE="Python"%>
<% python script code here %>

After upgrading from Python 2.1 to 2.2 with win32all-146.exe my
hello.asp script runs first time OK. Then just a blank page is
displayed in IE browser window next Refresh or activation.  I am
running Python 2.2 on WindowsNT 4.0 server and IIS 4.0.

Any ideas why the hello.asp script would generate HTML code only the
first time a new hello.asp file is copied to the server?


From martin.franklin at westgeo.com  Thu Feb 14 09:55:45 2002
From: martin.franklin at westgeo.com (Martin Franklin)
Date: Thu, 14 Feb 2002 14:55:45 +0000
Subject: Layout management in tkinter.
References: <3C6B9001.E5568D2C@lucent.com> <a4g7lr$oqb$1@norfair.nerim.net>
Message-ID: <a4gj46$1g4g$1@mail1.wg.waii.com>

Eric Brunel wrote:

> Hi Lukasz,
> 
> Lukasz Pawelczyk wrote:
>> Hello,
>> 
>> I try to change the geometry manager pack into grid in the code
>> and I have problem.
>> Why it does not work?
>> 
>> prz1=Button(main,text="Otworz",cursor="question_arrow",command=otworz)
>> prz1.grid(row=0)
>> 
>> prz2=Button(main,text="Zachowaj",cursor="dot",command=zachowaj)
>> prz2.grid(row=0,column=1)
>> 
>> prz3=Button(main,text="Zamknij",cursor="pirate",command=wyjscie)
>> prz3.grid(row=0,column=2)
>> 
>> plik=ScrolledText(main,background="grey")
>> plik.grid(row=1,column=0,columnspan=3)                - there the code
>> stop
> 
> You didn't mention where you got the ScrolledText widget, but apparently
> you did a "from ScrolledText import ScrolledText", isn't it? If it is, I
> do have the same problem: apparently, the creation or gridding of the
> ScrolledText tries to pack something in the "main" widget, which results
> in program hanging.
> 
> I would advise not to use the ScrolledText widget from the ScrolledText
> module since it is obviously not so well designed... Either use the Text
> and Scrollbar widgets from Tkinter (you'll have to bind them to each other
> explicitely, but it's quite easy), or use the ScrolledText widget
> from PMW (Python MegaWidgets) that you'll find at:
> http://pmw.sourceforge.net/
> You'll also find in PMW a great deal of other very useful widgets.
> 
> HTH
>  - eric -
> 

I have noticed Tkinter hanging when I (by mistake) pack and grid widgets 
into the same parent! You don't seem to be doing this however the 
ScrolledText widget uses a frame widget 'under the covers and this uses 
pack to 'pack' the scrollbar.  the grid() call to the scrolledtext widget 
is passed to this frame widget therefore without knowing it you are trying 
to pack and grid widgets into the same parent!.....


This is what I came up with :-

from Tkinter import *
from ScrolledText import ScrolledText


main=Tk()

f=Frame(main)
f.pack(fill='both', expand='yes')

prz1=Button(f,text="Otw?rz",cursor="question_arrow",command=None)
prz1.pack(side='left')

prz2=Button(f,text="Zachowaj",cursor="dot",command=None)
prz2.pack(side='left', expand='yes')

prz3=Button(f,text="Zamknij",cursor="pirate",command=None)
prz3.pack(side='right')

plik=ScrolledText(main,background="grey")
plik.pack(fill='both', expand='yes')           

main.mainloop()


HTH
Martin
















From jason at jorendorff.com  Tue Feb 19 22:14:13 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 19 Feb 2002 21:14:13 -0600
Subject: add-a-gram
In-Reply-To: <e9e6dd1f.0202191634.21da06b9@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIENODDAA.jason@jorendorff.com>

Jason wrote:
> Here is a puzzle that was floating around comp.lang.lisp, and I was
> wondering what a Python solution would be.

I solved this in Python in a few minutes, then converted it to C++
for efficiency and found that it ran slower in C++ (because I used
map<> and string.  Python's stuff is more efficient.)

Then I wrote my own hashtable and my own string type and managed
to get the C++ down to about 8x faster than the Python.
It solves their WORD.LST in <1 second on my PC.

I solved the other puzzles too...  They still didn't hire me.  :)

## Jason Orendorff    http://www.jorendorff.com/



From stojek at part-gmbh.de  Thu Feb 14 10:12:12 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Thu, 14 Feb 2002 15:12:12 GMT
Subject: help!: optimization of search problem (example code)
Message-ID: <3c6bcf70.16509609@news.easynews.net>

Hi,

after some days of playing around I am not getting
any further and have to ask for some more help.

Here is my problem. I have two large lists ("la" and "lb") of tuples.
Each tuple represents a point in space and has four entries:
x, y, z, pointID

Now I have to find out for each point of list "la" which point of
"lb" is closest. The result should be for all points of "la":

Point 123 of "lb" is the closest one to Point 5932 of "la"
where 123 and 5932 is the pointID.

So here is what I have so far.:

----------------------------------------------------------------
from random import *
from math import *
import bisect
import time


# timer for benchmarking
def startclock():
    d=time.clock()
    return d

def stopwatch(start):
    return time.clock()-start

#generating two arbtitrary lists
seed(0)
a={}
b={}

for i in range(20000):
    a[i]=(100*random(),100*random(),100*random())
    b[i]=(100*random(),100*random(),100*random())

la=[]
lb=[]

for i in range (20000):
    dummy=(a[i][0],a[i][1],a[i][2],i)
    la.append(dummy)
    dummy=(b[i][0],b[i][1],b[i][2],i)
    lb.append(dummy)

lb.sort() #source list

start=startclock()
maxdist=5.0 # I know that there are no two points with a distance > 5

for i in range(500): #should be range(len(la))
#slicing a part out of lb that is closer than maxdist for the first
#coordinate
    lp=(la[i][0]-maxdist,la[i][1],la[i][2],la[i][3])
    rp=(la[i][0]+maxdist,la[i][1],la[i][2],la[i][3])
    left=bisect.bisect_right(lb,lp)
    right=bisect.bisect_left(lb,rp)
    if right==len(lb): right-=1
    lsearch=lb[left:right] # only compare la[i] to this list

    mindist=10000
    minhit=-1 #pointID of closest point

    for j in lsearch:
# i can exclude points with coordinate 2 or 3 further away than
#maxdist
        if abs(j[1]-la[i][1])< maxdist and  
           abs(j[2]-la[i][2]) < maxdist:
#calculate distance:

d=sqrt(pow((j[0]-la[i][0]),2)+pow((j[1]-la[i][1]),2)+pow((j[2]-la[i][2]),2))
            if d < mindist:
                mindist=d
                minhit=j

    if i <= 10 :
   	print la[i],minhit,mindist

bench=stopwatch(start)
print bench, " sec"
--------------------------------------------------------------------------------------------------------

This takes 5.3 sec on my machine for 500 points. That is much too
long. Any idea how I could significanty improve this code?

Tanks,
Marcus


From op73418 at mail.telepac.pt  Mon Feb 25 14:29:03 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 19:29:03 +0000
Subject: On inner classes
Message-ID: <iv3l7ukajb2mnmsi19u050iapodl4itqd4@4ax.com>

Suppose we have something like

class <name>:
    class <name>:
        <whatever>
    <whatever>

that is, a class defined inside another. How can the inner class acess
the methods/attributes of the outer one?

Thanks for the help,
Gon?alo Rodrigues


From jason at jorendorff.com  Mon Feb  4 20:00:17 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 19:00:17 -0600
Subject: lambda expressions
In-Reply-To: <3c5e5962.591375@news.t-online.de>
Message-ID: <HFEKILOLEFEFMKAECNDLIEFGDAAA.jason@jorendorff.com>

Gerson Kurz wrote:
> Jason Orendorff wrote:
> > Gerson Kurz wrote:
> > > So, I see few limitations in the current implementation of lambda for
> > > the DEDICATED lambda-user.
> >
> > Lack of tail recursion could be a problem.
>
> I'm not really sure - I left university six years ago -, but isn't
>
> WHILE = lambda x,y: BOOL(x()) and (FALSE(y()) or WHILE(x,y))
>
> tail recursive?

Yes.

I should have said "Lack of tail call optimization could be a problem."

For spectators:  the reason tail recursion is important at all is,
any tail call can be compiled into (essentially) a goto.  This allows
functions like WHILE() to loop indefinitely without eating up memory.

In Python, tail calls are not optimized.  So in Python this uses
O(N) memory, where N = the number of times the loop executes.
Eventually you get "RuntimeError: maximum recursion depth exceeded".

A much more complete explanation of tail recursion:
http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-11.html#%_sec_1.2.1

## Jason Orendorff    http://www.jorendorff.com/



From chris.gonnerman at newcenturycomputers.net  Tue Feb  5 19:02:20 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 5 Feb 2002 18:02:20 -0600
Subject: OT again... is there an IMAP expert in the house?
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org> <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org> <mailman.1012224450.11377.python-list@python.org> <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu> <mailman.1012892049.28494.python-list@python.org> <Pine.OSF.4.44.0202042332410.184841-100000@goedel2.math.washington.edu> <mailman.1012915928.22449.python-list@python.org> <Pine.OSF.4.44.0202051154340.470451-100000@goedel3.math.washington.edu>
Message-ID: <005f01c1aea1$a26e7480$0101010a@local>

----- Original Message -----
From: "Eduardo Chappa" <chappa at math.washington.edu>


> *** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:
>
> :) > :) Tried that, didn't work.  I was using Netscape 4.75 as one client,
> :) > :) Outlook Express as the other, and concurrent deletion (even with
> :) > :) expunges) just didn't work.  Outlook Express never seemed to
> :) > :) notice Netscape doing deletions, and I couldn't get Outlook
> :) > :) Express to properly expunge without closing it.
> :) >
> :) > Just to be sure you did that. What are the first characters of the
> :) > first line of the folder that you want to share?
> :)
> :) Gah.  I didn't keep the files around, so I can't tell you for sure.  I
> :) did look at them with vi after conversion and they definitely weren't
> :) Unix mailboxes.
>
> Did you try two different sessions in Pine?. I have donde that with Pine
> several times so I know it works. At least if you succeed with one client
> you could discard the server and format as the reason why you are failing.

Brace yourself... I've never used Pine in my life.  You have a valid point,
though.





From flognat at flognat.myip.org  Mon Feb 11 09:00:38 2002
From: flognat at flognat.myip.org (Andrew Markebo)
Date: Mon, 11 Feb 2002 14:00:38 GMT
Subject: Emacs python mode and pdb.. shouldn't they cooperate??
Message-ID: <m3y9i0rud3.fsf@localhost.localdomain>

Hey emacs lovers!

How do you debug in your emacs?? Or you don't need to?? Once upon a
time, long, long ago I ported the xemacs gud-back/frontend, whatever
to emacs, but it is uncosy to use.. 

Isn't there support in the python-mode?? Have seen a couple of
variables for pdb, but haven't managed to get it working.. 

also.. 

I am trying to debug my script with python 2.2.. from IDLE i get a 
        valuerrror: invalid literal for float(): 399.0 
when I try to set the breakpoint on line 399... 

        /Andy

-- 
 The eye of the beholder rests on the beauty!


From ballabio at mac.com  Sat Feb  2 10:25:27 2002
From: ballabio at mac.com (Luigi Ballabio)
Date: Sat, 2 Feb 2002 16:25:27 +0100
Subject: How can I write this C code in Python?
In-Reply-To: <e959e7b.0202020710.417801c8@posting.google.com>
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <a05100300b881b459666b@[10.123.114.6]>

At 7:10 AM -0800 2/2/02, Fred wrote:
>Hello, I am just starting out with Python. I have a very basic book
>that is not helping me much. I have done some simple C programming in
>the past and I think if I could see the below C code done in Python it
>would get me over the beginners hump. I tried to do it in Python but
>got stuck right at the start, Python did not like it when I tried to
>say: for i = 1

Fred,
	I think you can find more info and tutorials at www.python.org.
Anyway, the for loops you are trying to write ca nbe expressed as:

for i in range(1,13):             # 1 is included, 13 is not
     for j in range(1,13):
         print "%4d" % (i*j),      # the comma at the end inhibits line breaks
     print                         # "print" alone prints a newline

Happy coding,
		Luigi


>//Simple C code to create a times table.
>
>#include <stdio.h>
>int main (void)
>
>{
>
>int i;
>int j;
>
>for (i = 1; i < 13; i++)
>	{
>	for (j = 1; j < 13; j++)
>	  printf("%4d", i*j);
>	        printf("\n");
>    }
>return 0;
>}
>--
>http://mail.python.org/mailman/listinfo/python-list


-- 



From bokr at oz.net  Fri Feb  1 16:47:18 2002
From: bokr at oz.net (Bengt Richter)
Date: 1 Feb 2002 21:47:18 GMT
Subject: Python like the old-time batch files.
References: <mailman.1012516669.13779.python-list@python.org>
Message-ID: <a3f2d6$qos$0@216.39.172.122>

On Thu, 31 Jan 2002 16:34:29 -0600, "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote:

>"MDK" <mdk at mdk.com> wrote in message
>news:<a3bhcf$176195$1 at ID-98166.news.dfncis.de>...
>> I was thinking back to years ago when I wrote batch files and I was able
>to
>> make these real nice double-line borders, change the color of things, and,
>> one of my favorite things, move the cursor up, down, left, or right in
>order
>> to write text or put lines anywhere.
>>
>> In Windows when I double click on a Python.py file it runs in a DOS-like
>> box.  Is it possible in Python to make those neat double-line borders,
>> change colors, and move the cursor to the exact position you want?
>>
>> Thanks.
>
>The borders I'm not sure about, but my WConio module can handle the rest.
>
>    http://newcenturycomputers.net/projects/wconio.html
>
>If anyone knows about drawing lines and boxes in a Windows Console and
>is willing to illuminate me, I'll add support to WConio for that also.
>
>Good luck.
>
Try this in your 'DOS' window:
 >>> for i in xrange(16):
 ...     for j in xrange(16):
 ...         print chr(i<<4|j),
 ...     print
 ...

I think you'll see all the typical ascii stuff,
including box/line characters.

BTW, I've had the thought that language processors should
treat all those characters as white space. That would allow
interesting readability effects in source code (at least
looking at it with appropriate font ;-).

Treating characters with double verticals as '#' (or C++ '//')
might also be interesting. You could have this as a translation
pragma for pre-processors, perhaps.

Regards,
Bengt Richter



From fperez528 at yahoo.com  Wed Feb 27 09:10:16 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Wed, 27 Feb 2002 14:10:16 +0000
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com> <29e28c51.0202270014.61dbd1d0@posting.google.com>
Message-ID: <a5ji37$ceb$2@peabody.colorado.edu>

Cimarron Taylor wrote:

> I have used the following tricks before with Java code, some of them
> should be useable with python.

[snipped]

I sincerely hope this was a joke (and as such, a really good one :). I can't 
seriously imagine you actually *doing* those things. 1 is fine, maybe 2.

Anyway, thanks for the good laugh!

f.


From terabaap at yumpee.org  Tue Feb 26 03:44:34 2002
From: terabaap at yumpee.org (Manoj Plakal)
Date: Tue, 26 Feb 2002 02:44:34 -0600
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au> <3C7ACEB4.B590E892@mediaone.net> <3C7AD777.7050401@skippinet.com.au> <3C7B46B5.3060001@yumpee.org>
Message-ID: <3C7B4AF2.3060105@yumpee.org>

Manoj Plakal wrote:
> Mark Hammond wrote:
>> Mark Barclay wrote:
>>
>>> Is there a way of coding a COM server that can be running before it 
>>> is accessed
>>> by a COM client? I'm not sure if this makes sense, but I'd like to 
>>> have a program
>>> running, that can be connected to, perhaps multiple times, without it 
>>> having to
>>> create multiple instances of teh server, just have multiple clients 
>>> using it.
>>
>> IIRC, it gets a little messy - you need to implement a LocalServer 
>> object, and add it to the running object table (ROT).  Searching old 
>> posts at groups.google.com is probably your best bet.
> 
>          Take a look at the CVS tree of qasid.sourceforge.net.
>          There's a small example of a wxPython GUI app that acts
>          as a COM server. The COMHostServer function in qasidserver.py
>          does the job of creating an instance of the server and
>          registering it in the ROT.


	   Whoops, I was referring to an old version of the code.
            You should look at file qasid_win32com.py, class
            QasidCOMHostThread, method QasidCOMHostServer (cross fingers).
            The sample GUI app is in testapp.py

            For the curious, Qasid is an idea I had for encapsulating
            platform-specific mechanisms for local intra-machine "desktop"
            RPC with a portable API. This is an API can be used by
            desktop apps to talk to each other for sending simple messages.
            Currently, you have to use platform-specific stuff
            like COM, or use XML-RPC/CORBA and implement a registration
            and lookup service, or do ad-hoc things like
            command-line parameters (Netscape/Opera/XMMS etc).

            The idea is to have a simple API wrapping
            the existing Win32 COM, AppleEvents, KDE DCOP and
            Gnome CORBA/ORBit. So far, there's the Win32 COM
            instantiation with a very thin API (start, register
            method, stop). There's a short blurb on qasid.sf.net
            about why CORBA and XML-RPC may not be a good idea
            for intra-machine (between desktop apps) communication,
            though they work great on the intra/internet. Haven't
            had too much time to work on this since the initial
            COM effort, but I do plan to get started on the
            Apple Events backend sometime soon.

            Manoj



From deathtospam43423 at altavista.com  Fri Feb  8 19:23:16 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 9 Feb 2002 00:23:16 GMT
Subject: Help: I want a chat script!
Message-ID: <Xns91B06F8651AusYourStandingInIt@130.133.1.4>

Ello. I'm a newbie programmer using the Python language. I've recently 
completed a couple of the Python beginner's tutorials out there on the web 
(Josh Cagliati's, and Richard Baldwin's).

I have this ambitious idea for a project. I want to make my own browser 
based chat room. 

It seems simple. I basically wrote a little Python subroutine which takes 
input via the Python IDE, and writes a little html document using IO 
commands. Voila! Un chat program! Albeit an extremely simple one at first. 
Now if I could just train the little bugger to take input from a web 
browser I'd be in business. I figured the easiest way to do this was to 
write an html form that writes the input to a text or dat file, and my prog 
can extract the data from that and print it back to web page

But ... I can't find any easy way to do that, to my suprise. I tried 
looking through all the html tutorials for information about forms. The 
nearest I could come to, specific to my requirements, was tutorials on how 
to set up your own messageboard. Apparantly, I have to get a web server 
that supports CGI (whatever that is), use a telnet client(whatever that is) 
to connect to it. I also have to learn PERL (What? A whole new programming 
language?), and ideally the server should be UNIX (A whole new operating 
system?).

I can't believe I have to learn 4 things at once just to get to get input 
from a browser! Surely there must be an easier way to do it?

If I could,  I would much rather be able to have the script on my PC, use 
my personal web server to host it. Then I'd invite a net buddy or two into 
it to help me perfect the script. And then I might think about actually 
putting the room on a proper server, if I don't lose interest in the 
project by then.


-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From asim_beg at hotmail.com  Wed Feb 20 00:28:30 2002
From: asim_beg at hotmail.com (Asim Beg)
Date: Tue, 19 Feb 2002 23:28:30 -0600
Subject: MySQLdb build on Solaris 7
Message-ID: <F203anSs21Ud9mTP2vZ000016c2@hotmail.com>

Jeff,

Here's an excerpt from setup.py

elif sys.platform == "sunos5": # Solaris 2.8 + gcc
    include_dirs = ['/usr/local/mysql/include']
    library_dirs = ['/usr/local/mysql/lib']
    libraries = [mysqlclient, "z"]
    runtime_library_dirs = ['/usr/local/lib:/usr/openwin/lib:/usr/dt/lib']
    extra_compile_args = ["-fPIC"]

and MySQL is installed at /usr/local/mysql. I can send the whole file if 
needed.


Asim


>From: Jeff Bauer <jbauer at rubic.com>
>To: python-list at python.org
>CC: asim_beg at hotmail.com
>Subject: Re: MySQLdb build on Solaris 7
>Date: Tue, 19 Feb 2002 22:15:40 -0600
>
>Asim writes:
> > My ISP has Python 2.1 installed and I can run Python
> > CGI programs fine. I now need to connect to MySQL. I
> > need to install MySQLdb and am having difficulty
> > getting the MySQLdb source to build. I have tried
> > both MySQL-python-0.9.0 and MySQL-python-0.9.1.
> >
> > The machine is a solaris 7 with Python 2.1 on it. When
> > I try to build it starts building and then goes into
> > an infinite loop with strange libmysqlclient.a error
> > messages..
>
>
>Asim,
>
>I have Andy Dustman's MySQL-python-0.9.1 interface
>running with Python 2.1 on Solaris.
>
>You probably need to edit setup.py to reflect the
>locations of the libraries and include files, especially
>if you have MySQL in non-standard directories.
>
>Look for the line of code that begins with ...
>
>   elif sys.platform == "sunos5": # Solaris 2.8
>
>... and edit the directories to reflect your
>local configuration.
>
> > Also since I cannot install MySQLdb in the standard
> > places on Solaris, how should I go about after the
> > build step.
>
>After installation, I've got the following installed:
>
>   /usr/local/lib/python2.1/site-packages/
>
>   CompatMysqldb.py        _mysql.so
>   CompatMysqldb.pyc       _mysql_exceptions.py
>   MySQLdb/                _mysql_exceptions.pyc
>
>... and ...
>
>   /usr/local/lib/python2.1/site-packages/MySQLdb/
>
>   __init__.py       constants/        cursors.pyc  times.pyc
>   __init__.pyc      converters.py     sets.py
>   connections.py    converters.pyc    sets.pyc
>   connections.pyc   cursors.py        times.py
>
>If you have a similar file/path structure somewhere in your
>importable PYTHONPATH, you should be okay.
>
>Jeff Bauer
>Rubicon Research


_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx




From NIDHI_M at infy.com  Sun Feb  3 01:07:49 2002
From: NIDHI_M at infy.com (NIDHI_M)
Date: Sun, 3 Feb 2002 11:37:49 +0530
Subject: Help needed on screen painters/ Forms manager for Python on Linux
Message-ID: <755FA95DB839D211856B0008C7287D930B9C323C@kecmsg02.ad.infosys.com>

> Hi
> 
> We are re-engineering an application in C with front end in JAM to use
> python ncurses library. This application will later run on Linux. Is
> there any screen painter which will help us re write the JAM forms
> into python ncurses.
> 
[NIDHI_M]  

1. There is something called snack. Will this be useful for character
based front -ends?
2. Does the tinter module come with red hat linux?

> Thanks
> Nidhi



From jkraska at san.rr.com  Sat Feb  2 23:30:11 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 04:30:11 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx> <mailman.1012707428.2108.python-list@python.org>
Message-ID: <ajfp5uork5menmspbkf1a198811bnaa0r2@4ax.com>

>This is hard for me because I can't see what the '}' goes with.
>Combine this with leaving out the braces on one-line compound
>statements, and I'll have real problems.

Well, the argument in favor "but you can: the trailing brace
matches the statement at its indentation level" makes some sense,
it's just that you and I probably use matching tools. If my
matching tool "took me from the trailing brace to the beginning
of the first matching statement just prior to the initiating brace"
I might be able to deal with it. But no.

C//



From park997 at aol.comnotospam  Tue Feb 12 12:23:38 2002
From: park997 at aol.comnotospam (Park997)
Date: 12 Feb 2002 17:23:38 GMT
Subject: Tk button help
Message-ID: <20020212122338.02073.00001446@mb-ch.aol.com>

Hi,

I have a Tk button with a command (function) bound to it. The first
time the program is run, the command is executed without the button
being pushed. After this pushing the button does nothing.

Here is the relevant (I think) code. I would appreciate help in
understanding and fixing this:

	def run(self,G):
		
		from tkSimpleDialog import askfloat
	
		self.N = askfloat('N', 'Leaf N (50 - 250 mmol m-2) ')




if __name__ == '__main__':
	
	test = assimilation()
	
	import Gen_Graph
	
	from Tkinter import *

	top = Tk()
		
	G = Gen_Graph.grapher(top)

	Button(top, text='Enter Leaf N', command = test.run(G)).pack(side=RIGHT, fill
= BOTH)
	
	top.mainloop()  

Thanks.

Wendell Cropper
	


From missive at frontiernet.net  Fri Feb 22 20:03:30 2002
From: missive at frontiernet.net (Lee Harr)
Date: Sat, 23 Feb 2002 01:03:30 -0000
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com> <nkUc8.19043$kX6.133584@NewsReader> <655abd4a.0202220114.21447213@posting.google.com>
Message-ID: <slrna7dqet.32.missive@whave.frontiernet.net>

>> >         self.webpage = urllib.urlopen(self.url).read()
>> 
>> Yes, this can cause Python to hang (thread passim).
>> 
>> It *seems* that using httplib() instead *might* avoid the hang.
>> 
>> Andrew ward
> 
> it seems i might be having the same sort of problem...
> 
> i am writing a special (http) server using SimpleHTTPServer and when i
> try to read from do_POST, it hangs...
> 

Which version of python are you using?

I was trying to implement a program which was interactive, and
used urllib in the background to transfer data to and from a
webserver.

With python2.1, urllib locked the system up completely until
it received a response.

Looking around on the web, I found there had been a bug report
about this just before 2.2, so I upgraded. It is much better, but
still not quite there for my application. Possibly on a faster
machine it would not be an issue.

Something to try, anyhow.



From sjmachin at lexicon.net  Sun Feb 17 19:39:44 2002
From: sjmachin at lexicon.net (John Machin)
Date: 17 Feb 2002 16:39:44 -0800
Subject: Embedding python with lcc-win32
References: <335764c1.0202161031.425dd81a@posting.google.com> <335764c1.0202170339.8db63a0@posting.google.com>
Message-ID: <c76ff6fc.0202171639.70564dd@posting.google.com>

ae98096 at aero.iitm.ernet.in (decay) wrote in message news:<335764c1.0202170339.8db63a0 at posting.google.com>...
> I got my simple python-embedded-in-C program (read below) to compile
> with lcc-win32 by doing the following:
[snip]

You may well be the first person to try using lcc-win32 with Python.
Apart from MS VC++ (which is used to build Python for Windows), you
may like to consider the command-line C/C++ compiler from Borland, or
MinGW (gcc for Windows) -- both (money-)free; there is considerable
experience with these in the Python community.

To get your "why is this getting an <unknown>" etc question answered,
you would have to (a) provide your code (test.c) and the actual error
output (b) hope that the problem was generic rather than
compiler-specific.


From tim.one at comcast.net  Thu Feb 28 18:54:46 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 28 Feb 2002 18:54:46 -0500
Subject: Status of PEP's?
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A1C5@natasha.auslabs.avaya.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEFPOAAA.tim.one@comcast.net>

[Delaney, Timothy]
> ...
> Are the following two equivalent though?

Not necessarily, but the possible difference isn't where you think:

> if 1 in obj:
>     pass
>
> if 1 in iter(obj):
>     pass
>
> i.e. is iter(obj) called implicitly in the first case?

Maybe.  First type(obj) is checked to see whether it implements __contains__
(at the C level, whether obj->ob_type->tp_as_sequence->sq_contains is
reachable and non-NULL).  If it is, iter() gets involved if and only if the
__contains__ method decides to use iter() (and, for example, dicts certainly
do not).  If type(obj) doesn't implement __contains__, then "in" tickles
iter(obj) repeatedly.




From phr-n2002a at nightsong.com  Tue Feb  5 05:29:33 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 02:29:33 -0800
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <mailman.1012902609.27332.python-list@python.org>
Message-ID: <7x7kpscjci.fsf@ruckus.brouhaha.com>

Gerhard =?iso-8859-15?Q?H=E4ring?= <gh_pythonlist at gmx.de> writes:
> Btw. it looks like random.randint/randrange really want ints and won't
> accept longs. A workaround could be to use a construction like:
> 
>     print random.randrange(0, sys.maxint) * 2L

That's always zero in the lowest bit.


From mcherm at destiny.com  Tue Feb 26 13:23:34 2002
From: mcherm at destiny.com (mcherm)
Date: Tue, 26 Feb 2002 13:23:34 -0500
Subject: functional programming with map()
Message-ID: <3C7BD2A6.6040700@destiny.com>

>> List comprehensions are declarative (no loops) and they avoid side effects
>> (no loop variables or temporaries)
> 
> No side effects?  Try this:
> 
> L = [x for x in range(10)]
> print x
> 

Yeah, but that's a bug.

<pulls hat down over head and ducks behind the desk>

-- Michael Chermside






From mcfletch at geocities.com  Fri Feb 15 17:27:47 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Fri, 15 Feb 2002 17:27:47 -0500
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com>
Message-ID: <3C6D8B63.5070909@geocities.com>

wxImages can be loaded from strings.  See the wxPython Wiki entry on 
working with images (particularly the sections on PIL and Numeric):

http://wiki.wxpython.org/index.cgi/WorkingWithImages

You can probably use PIL to get GIF/JPEG/Whatever formatted data into 
the RAW format for use by wxImage

HTH,
Mike

Paul Moore wrote:
> On 14 Feb 2002 23:34:12 +0100, Chris Liechti <cliechti at gmx.net> wrote:
[comments on loading various file types from zip files]
...
>>but then wxWindows Bitmap.LoadFile wants real file...
>>but: """wxBitmapFromXPMData(listOfStrings)  Create a bitmap from a 
>>Python list of strings whose contents are XPM data""" so you would be 
>>limited to XPM images.
>>
> 
> Looks like it's wxPython that has the problem...
> 
> Annoyingly, wxWindows has a wxImage constructor, wxImage(wxInputStream)
> to load from a stream, but wxPython *doesn't* seem to wrap this
> constructor. OK, so there would need to be a conversion from a Python
> file object to a wxInputStream (or something).
> 
> But some way of getting a wxImage from in-memory data (of arbitrary
> format) is really all I'm after - and there ought to be a way to get it.
> I've copied this to the wxPython list. Is there some way I've missed -
> or could a way be added?
> 
> Thanks,
> Paul Moore
> 
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From spenrose at well.com  Mon Feb 18 22:21:20 2002
From: spenrose at well.com (Sam Penrose)
Date: 18 Feb 2002 19:21:20 -0800
Subject: Python Gotcha with Octal Numbers
Message-ID: <1014088880.1837.17.camel@localhost.localdomain>

Charles Richmond  <richmond at ev1.net> wrote:
    Ed Fredkin is semi-famous. He was mentioned prominantly in David
    Levy's book _Hackers_. He also invented some data structures that
    are used today, notably the "trie". 
    
He means Steven Levy. 

http://www1.fatbrain.com/asp/bookinfo/bookinfo.asp?theisbn=0141000511

I finally got around to reading this a couple months ago, and I wish I
hadn't waited that long. It is a wonderful history of the culture of
computing in a few important places, notably MIT in the 60's and 70's.
If you are enough of a geek to frequent c.l.py., you really should
consider this book mandatory reading. A lot of programmer behavior that
seems randomly bizarre can be traced back to management decisions made
by government contractors lo these many years ago. Among other things,
it explains where Richard Stallman came from.

and-the-answer-is-not-"Mars"-ly yours,
s




From gh_pythonlist at gmx.de  Sun Feb  3 21:16:51 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 4 Feb 2002 03:16:51 +0100
Subject: Autocoding evolves from........
In-Reply-To: <972.799T919T11694325threeseas@earthlink.net>
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net>
Message-ID: <20020204021650.GA1333@lilith.hqd-internal>

Le 04/02/02 ? 00:00, Timothy Rue ?crivit:
> On 30-Jan-02 06:52:48 phil hunt <philh at comuno.freeserve.co.uk> wrote:
> >On 30 Jan 2002 00:48:40 GMT, Peter Seebach <seebs at plethora.net> wrote:
> 
> >>Anyway, it looks to me like Python is the wrong language for what you want
> >>to do.  C# is almost certainly the best choice for this; it's designed for
> >>the kinds of tasks you're looking at,
> 
> >Is Tim trying to lock people in to the Microsoft monopoly, then?
> 
> Why do you make up things by taking things out of context?
> 
> Are you naturally dishonest or do you work at it?
> 
> Autocoding will be a nice thing to have, it'll get rid of all the arrogant
> jerks.

Sure. Just like the 4GL buzz from the 1980ies revolutionized the
software industry. Not.

Other than that, I learnt something because of the thread you started,
namely that my mailer supports message scoring.

,----[ .muttrc_scoring ]
| score '~s .*' 5
| score '~s autocod' =0
`----

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 7.7 ?C      Wind: 3.4 m/s



From jkraska at san.rr.com  Mon Feb 25 23:26:00 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 04:26:00 GMT
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <jg3m7ugd2c3svhi4k9s2qp88jm8f2ib99s@4ax.com>

>1) add a local declaration, similar in syntax to Python's existing global
>   declaration,...

Python's current "global" directive notionally "imports" the global variable
into the local namespace and short-circuits the automatic creation of a local.
Yours is the opposite: instead of using an outer variable, force the creation
of a local. Personally, I like this.

The list part of me wants:

	x = 3
	let ( x ):
		x = 5

I might call it something other than "let" to make it Pythonic, though.

>2) Add a compiler option similar to Perl's "use strict",

This has been brought up before. Personally, I favor the idea of a directive
of some kind, on per module basis, that affects that module for that module's
data. One of the reasons you're not seeing too many people clamour for this
kind of thing, however, is that misspellings are (so it is claimed, and so my
own personal experience suggests) actually a relatively scarce cause of Python
problems.

You can get some of this kind of behavior _now_ by the way. Just implement
a class which doesn't allow setattr to work on an instance for any variable
which doesn't likewise exist in it class.

*Poof, Guido uses his time machine*

>   Example 2:
>     local x(int)
>     x = "banana"    # implementation-dependent behavior

I disagree here. If you've gone to the trouble of declaring something,
to get full use, you need a defined error; further, it would be helfpul
if this occured prior to execution.

There's some good cause for resistance to this. One of those things which
keeps Python Pythonic is it's concise expressivity. Declarations foil that.

C//



From tim at frontier.net.nospam  Thu Feb 28 18:52:24 2002
From: tim at frontier.net.nospam (Tim H)
Date: Thu, 28 Feb 2002 16:52:24 -0700
Subject: Can use python to play MP3 file?
References: <a5llkf$8a75f$1@ID-12869.news.dfncis.de> <3C7E6328.6050500@bigfoot.de>
Message-ID: <3c7ec28f$0$27085$4c41069e@reader1.ash.ops.us.uu.net>

"Gerhard H?ring" <gerhard at bigfoot.de> wrote in message
news:3C7E6328.6050500 at bigfoot.de...
> Ricky wrote:
> > I hear that PHP can online play MP3 file. Is there any method that can
make
> > me play MP3 by using Python CGI?
> >
> > Thanks for any reply!
>
> Do you want to play the MP3 file on the server, like a jukebox. Or do
> you want to stream MP3 files back to the client computer?
>
> (Of course you can do both with Python)
>
> Gerhard
>

Do you have examples, links, hints to doing each with python?

Thanks,
 Tim




From philh at comuno.freeserve.co.uk  Wed Feb 27 12:28:22 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 27 Feb 2002 17:28:22 +0000
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
Message-ID: <slrna7q5pm.edg.philh@comuno.freeserve.co.uk>

On Wed, 27 Feb 2002 15:05:01 +0000, Chris <nospam@[127.0.0.1]> wrote:
>In the olden days I played around with BASIC
>just for fun, writing trivial games and making patterns move on the 
>screen.
>
>Can I do that sort of thing with Python?

Yes

>I've at last got a new machine, running XP Home.
>Will this be OK?
>Or would it be wiser to use my old 200MMX?

Python is happy with either.

>Before I start wrestling with Python - and with thinking - and with 
>making my brain hurt - is Python the best language for this sort of 
>play?

Yes

>Would Java be better?

No.

Java will rot your brain.

>Will I be able to make a ball bounce around inside a box etc.?

I don't see why not.

-- 
<"><"><"> Philip Hunt <philh at comuno.freeserve.co.uk> <"><"><">
"I would guess that he really believes whatever is politically 
advantageous for him to believe." 
                        -- Alison Brooks, referring to Michael
                              Portillo, on soc.history.what-if


From mal at lemburg.com  Tue Feb 19 14:13:24 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 19 Feb 2002 20:13:24 +0100
Subject: ANN: eGenix.com mxCGIPython Version 0.4.0
References: <3C717A94.2C9ED2C5@lemburg.com>
Message-ID: <3C72A3D4.96BE28E1@lemburg.com>

"M.-A. Lemburg" wrote:
> 
> ________________________________________________________________________
> 
> ANNOUNCING:
>                  eGenix.com mxCGIPython Version 0.4.0
> 
>             An easy-to-install one-file Python Interpreter
> 
> ________________________________________________________________________
> 

Update:

Oleg Broytmann kindly sent me six files for Linux, Solaris and 
FreeBSD, Python 2.1.2 and 2.2:

	http://www.egenix.com/files/python/mxCGIPython.html

(The setup seems to work with Python 2.2 as well.)

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From sab at NOSPAM.freeuk.com  Fri Feb  8 10:58:39 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Fri, 8 Feb 2002 15:58:39 -0000
Subject: Tkinter menu checkbutton options??
Message-ID: <a40sgo$n5f$1@news8.svr.pol.co.uk>

Has anybody got any idea what option the Tkinter menu checkbutton can take.
Basically i have a menu using a checkbutton (i.e. like a little tick next to
the menu option) and want to start the program with the checkbutton on. and
also can i check the status of it later??

Thanks

G. Willoughby




From alexw at fast.no.nospam  Thu Feb 14 08:03:01 2002
From: alexw at fast.no.nospam (Alexander Wilkens)
Date: Thu, 14 Feb 2002 14:03:01 +0100
Subject: Problem freezing Win32 services
References: <a4g3s3$k0s$1@news.ost.eltele.no> <a4g5sl$f89$1@ID-59885.news.dfncis.de>
Message-ID: <a4gci5$pbh$1@news.ost.eltele.no>

"Thomas Heller" <thomas.heller at ion-tof.com> wrote in message
news:a4g5sl$f89$1 at ID-59885.news.dfncis.de...
> > I'm attempting to freeze a Windows NT service (in Win2000) written in
Python
> > 2.2 using win32all 142 and the freeze util bundled with Python in the
tools
> > directory, and I've encountered some problems. Freezing console apps
works
> > fine.
> <shameless plug>
> You could (should ;-) try out the newest py2exe release, which also can
> create service exe-files: http://starship.python.net/crew/theller/py2exe/

I tried py2exe and it seems to work better at least :-) I mananged to create
a service executable that can be started either from the Windows Services
GUI or by 'net start myservice'. However if I try to just run the service
executable I get:

Starting service - this may take several seconds - please wait...
Could not start the service - error 997

The real problem, however, starts when I move the executable and associated
pyds and dlls to another computer. I then get the following error in the
event viewer:

"The description for Event ID ( 3 ) in Source ( myservice ) cannot be found.
The local computer may not have the necessary registry information or
message DLL files to display messages from a remote computer. The following
information is part of the event:  ; e; N."

I suspect I may be missing some imports. py2exe gave me some warnings,
however I've tried to find out which it is with no luck. Is there any way to
get more descriptive errors (tracebacks?)? Running the service with
the -debug option outputs "Error 0xC0000003 - The instance's SvcRun() method
failed" and  " e: N" . Any suggestions?

--
Alex




From amuys at shortech.com.au  Wed Feb 27 19:00:43 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 27 Feb 2002 16:00:43 -0800
Subject: Oddity question
References: <mailman.1014745204.15257.python-list@python.org> <7934d084.0202262258.53fbe36@posting.google.com>
Message-ID: <7934d084.0202271600.499b2b74@posting.google.com>

amuys at shortech.com.au (Andrae Muys) wrote in message news:<7934d084.0202262258.53fbe36 at posting.google.com>...
> <aside>
<SNIP>
> </aside>

Cliff has suggested in a very polite private email that this aside was
a cheap shot.  On reflection I agree with him, so wish to offer a
public appology for it.  While it may have been appropriate in a FTF
discussion, such a comment divorced of all non-verbal cues is
inappropriate, in bad grace and showed poor judgement on my part.

Sorry Cliff.  I will do better next time.

Andrae Muys


From alanmk at hotmail.com  Tue Feb 19 06:28:59 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 19 Feb 2002 03:28:59 -0800
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <f72dac0d.0202190328.6c0c2c1b@posting.google.com>

Jim,

The 5-6 second delay you are experiencing is very likely caused by the
initialisation overhead of whatever JVM you are using.

You would see the same overhead if you avoided Jython altogether, and
wrote your "Hello World" CGI in pure Java (It's probably worth doing
this, just to prove that Jython is not the culprit here).

If you're going to do CGI with Java and/or Jython, you're going to
have to live with this JVM initialisation overhead: it's unavoidable,
unless you compile your Java/Jython into native code using a Java to
native code compiler (which is really going too far to solve such a
simple problem).

Is there any way that you can install a servlet engine on your
machine? If so, that's the way to go. This is the whole purpose for
Java servlets: to avoid the very JVM initialisation overhead you are
currently experiencing.

Java is just not a CGI language, and hence any language derived from
it, e.g. Jython, Jacl, will be equally unusable for CGI scripting.

Regards,

Alan.

"jimh" <jim at jim.jim> wrote in message news:<a4rjq4$4n8$1 at web1.cup.hp.com>...
> We have a bunch of Python (cgi) scripts producing html behind an Apache web
> server.  We also have a set of Java libraries that I want to access from
> these Python scripts, but I haven't been able to figure out how to do this.
> I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
> a one-line 'print "hi"' script, and this is, of course, not acceptable for a
> web server.


From quinn at retch.ugcs.caltech.edu  Tue Feb  5 01:28:30 2002
From: quinn at retch.ugcs.caltech.edu (Quinn Dunkan)
Date: 5 Feb 2002 06:28:30 GMT
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net> <IPV68.8359$Xk4.443398@news1.east.cox.net>
Message-ID: <slrna5uusd.qqe.quinn@retch.ugcs.caltech.edu>

On Sat, 02 Feb 2002 18:01:12 GMT, Andrew Replogle <replogle2 at cox.net> wrote:
>
>"Andrew Replogle" <replogle2 at cox.net> wrote in message
>news:LwK68.7172$Xk4.381742 at news1.east.cox.net...
>>    Anyone know of any books out there for python system administration or
>> should I just stick to perl for these tasks?
>>
>>
>Well, for example, I'd like to write a python script to tail a ftp log file
>& when a connection from a customer is made and the file transmits
>succesfully to email them a confirmation. I am usually very efficient if I
>have just a small example or concept of how to do this and just wanted to
>know if anyone had any suggestions on such a book. Thats all. :)

Well, dunno about any books, but here's a quick-n-dirty something for the
above:

tail -f ftp.log | grep 'connect from customer' | mail_line.py >errlog

mail_line.py:

#!/usr/local/bin/python
import sys, re
def main():
    s = sys.stdin.readline()
    customer = re.search(r'extract (customer)', s).group(1)
    customer_email = lookup_in_some_database(customer)
    p = os.popen('/usr/lib/sendmail %s' % customer_email, 'w')
    p.write(congratulatory_message)
    p.close()

while 1:
    try:
        main()
    except:
        import traceback
        traceback.print_exc()


You could probably do it all in shell with some 'sed' and 'read' hackery, but
anything more complicated than a pipeline in shell is what I use python for :)

Umm, stdio buffering is probably going to screw up the above.  Dumb substdio.


Sysadmin stuff in python is much like in C, since python simply wraps a lot of
libc calls, e.g. getpwnam(), os.system(), etc.  Of course you don't have to
worry about memory management and all the other C details.  Check out
the docs for the os and posix modules.


From tim-dated-1014947118.88a476 at catseye.net  Thu Feb 21 20:45:17 2002
From: tim-dated-1014947118.88a476 at catseye.net (Tim Legant)
Date: 21 Feb 2002 19:45:17 -0600
Subject: strings and sort()
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A1A8@natasha.auslabs.avaya.com>
References: <B43D149A9AB2D411971300B0D03D7E8BF0A1A8@natasha.auslabs.avaya.com>
Message-ID: <86664qe1ci.fsf@skitty.catseye.net>

"Delaney, Timothy" <tdelaney at avaya.com> writes:

> > From: caljason76 at yahoo.com [mailto:caljason76 at yahoo.com]
> >
> > I like Python, but the biggest problem I have with the language right
> > now is what low code density it has at times.
> 
> Huh? Low code density? Huh?

Maybe compared to Perl, where if your cat sits on the keyboard, the
resulting code is as likely to run as anything /you/ write.


Tim



From mwh at python.net  Mon Feb 25 04:45:55 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 25 Feb 2002 09:45:55 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com>
Message-ID: <u3czpzyas.fsf@python.net>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> David Eppstein <eppstein at ics.uci.edu> writes:
> > > : But what is the functional equvalent of:
> > > 
> > > : for x in items:
> > > :     x.f()
> > > 
> > > 
> > > 
> > > Here's one way to do it:
> > > 
> > > ###
> > > map(lambda x: x.f(), items)
> > > ###
> > 
> > I'd prefer
> > [x.f() for x in items]
> > 
> > It's not functional syntax, but so what?
> 
> Both of those build up a new list of the results, instead of
> discarding the values.  If the f function takes an integer and
> computes a 20-megabyte structure, you've got a problem.  This really
> calls for a generator comprehension as discussed a few weeks back.
> 
>    reduce(lambda x,y:(y.f(),0)[1], items, 0)
> 
> almost works, but invokes f an extra time at the end.
> 
> Someone see how to fix it?

for x in items:
    x.f()

?

Cheers,
M.

-- 
  Premature optimization is the root of all evil.
       -- Donald E. Knuth, Structured Programming with goto Statements


From max at alcyone.com  Wed Feb 27 20:54:37 2002
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 27 Feb 2002 17:54:37 -0800
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
Message-ID: <3C7D8DDD.B22E6BF1@alcyone.com>

Huaiyu Zhu wrote:

> Is there a way to do this
> 
> for k, v in dict.items():
>     dict[k] = f(v)
> 
> without involving a Python for loop?  I'd imagine that a function
> implemented in C could speed things up quite a bit.

In 2.2 you can use map on a dictionary; it is equivalent to mapping
against its keys.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From owen at xastrox.xwashingtonx.edu.invalid  Fri Feb 15 11:35:29 2002
From: owen at xastrox.xwashingtonx.edu.invalid (Russell E. Owen)
Date: Fri, 15 Feb 2002 08:35:29 -0800
Subject: Layout management in tkinter.
References: <3C6B9001.E5568D2C@lucent.com> <a4g7lr$oqb$1@norfair.nerim.net> <a4gj46$1g4g$1@mail1.wg.waii.com>
Message-ID: <owen-D5A9FE.08352915022002@nntp2.u.washington.edu>

In article <a4gj46$1g4g$1 at mail1.wg.waii.com>,
 Martin Franklin <martin.franklin at westgeo.com>, in response to a 
sequence that started with article <3C6B9001.E5568D2C at lucent.com>,
 Lukasz Pawelczyk <pawelczyk at lucent.com> wrote:

>I have noticed Tkinter hanging when I (by mistake) pack and grid widgets 
>into the same parent! You don't seem to be doing this however the 
>ScrolledText widget uses a frame widget 'under the covers and this uses 
>pack to 'pack' the scrollbar.  the grid() call to the scrolledtext widget 
>is passed to this frame widget therefore without knowing it you are trying 
>to pack and grid widgets into the same parent!.....
>
>
>This is what I came up with :-
>
>from Tkinter import *
>from ScrolledText import ScrolledText
>
>
>main=Tk()
>
>f=Frame(main)
>f.pack(fill='both', expand='yes')
>
>...
>plik=ScrolledText(main,background="grey")
>plik.pack(fill='both', expand='yes')           
>
>main.mainloop()

A follow-up comment: if this is the problem, you can still grid the 
widgets by making a separate frame for the ScrolledText widget, pack the 
ScrolledText into that frame and grid the frame. Something like:

scframe = Frame(main)
plik=ScrolledText(scframe...)
plik.pack(expand="yes", fill="both")
scframe.grid(...)

Also, ScrolledText is broken. It should return a frame containing the 
widgets, precisely so this kind of problem doesn't happen.

-- Russell
-- 
Return      owen
address     astro
garbled     washington
in header   edu


From tim.one at home.com  Thu Feb  7 22:21:06 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 7 Feb 2002 22:21:06 -0500
Subject: FixedPoint
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBGEFJKBAA.mark@mceahern.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEGHNKAA.tim.one@home.com>

[Mark McEahern]
> In the FixedPoint module found here:
>
>
<ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoi
nt.py.Z>
>
> there seems to be a bug comparing to None like this:
>
>   if f == None:
> ...
>     TypeError: can't convert to FixedPoint: None

Magnitude (as opposed to identity -- "is") comparision of a FixedPoint to a
non-numeric object makes no sense that I'm comfortable imposing, so I'm
happy to let it raise an exception.  Specific apps may require more, and
through the magic of subclassing you can supply whatever senseless <wink>
comparison behavior you like without touching the base code.

In any case, this isn't a maintained module.  Whatever time I may have been
able to devote to it I set aside in case Aahz hits snags implementing IBM's
Standard Decimal Arithmetic proposal:

    http://www2.hursley.ibm.com/decimal/

FixedPoint gets the job done for most who've tried it, but it's at odds with
IBM's proposed model, and I like IBM's well enough that I'd rather help it
become a standard.




From duncan at NOSPAMrcp.co.uk  Thu Feb 21 07:06:21 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Thu, 21 Feb 2002 12:06:21 +0000 (UTC)
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de>
Message-ID: <Xns91BC7AB03699duncanrcpcouk@127.0.0.1>

gerson.kurz at t-online.de (Gerson Kurz) wrote in 
news:3c74c3d6.20466328 at news.isar.de:

> I found an easier solution:
> 
> sort = lambda x:(x.sort(),x)[1]
> 
If you are heading that way:
sort = lambda x:x.sort() or x

-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From starx at pacbell.net  Mon Feb 18 06:41:55 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 18 Feb 2002 11:41:55 GMT
Subject: PyAppBar/wxAppBar opened at SourceForge
References: <adec2554.0202172056.62b0f735@posting.google.com>
Message-ID: <Xns91B92597C587BRASXnewsDFE1@130.133.1.4>

Keith Farmer || Sun 17 Feb 2002 08:56:39p:

> https://sourceforge.net/projects/pyappbar/
> 
> I decided to toss it out upon the unsuspecting masses.

It might have been a good idea to also toss out this description from 
the top of your sourceforge page.

     This project aims to create Application Toolbars (a class of
     dockable window under Win32) in Python. Examples of AppBars
     include the Win32 taskbar, Norton System Doctor's docking window,
     and Mirabilis's ICQ client. An proof-of- concept has been
     created.

> There are currently two main issues which I'm stuck on:
> * Resizing while docked doesn't work. I think there's a conflict with
> other messages kidnapping the mouse.
> * Moving from a docked to an undocked state -- it works, kinda. If you
> drag the mouse too quickly, it'll lose its grip. If you release the
> mouse button, and grab the (now undocked) window again, it's fine.
> 
> I'd certainly appreciate some help in finishing this up, if anybody's
> interested.

It may be beneficial to to try and document what you already have first. 
I couldn't find a single doc string, comment block, or file header to 
tell me what to do with this... maybe i'm looking in the wrong places 
but i couldn't even find definate information on something both tiny and 
important: which OS(s) it is designed for. I can only guess that it 
works on windows due to the DLLs available for download.

It is a nice idea though, if it does what i think it does =).

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From andy47 at halfcooked.com  Sun Feb 17 20:34:05 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Mon, 18 Feb 2002 01:34:05 +0000 (UTC)
Subject: DCOracle syntax question
References: <Wvmb8.659$jj4.576986@typhoon.ne.ipsvc.net>
Message-ID: <Xns91B97FBD3220andy47halfcookedcom@203.109.250.24>

"Jim St.Cyr" <jstcyr at mediaone.net> wrote in news:Wvmb8.659$jj4.576986
@typhoon.ne.ipsvc.net:

> I have a query of the form
> 
> select * from table where key = 'key' and date = (select max(to_char
(date))
> from table where key = 'key')
> 
> Through SQL*Plus this executes fine.  When executing through DCOracle I 
get
> an empty result.  I've tried both tag = 'query' and cursor.execute
(query).
> I think that either the subquery or the inline quotes is throwing things 
off
> but am at a loss.
> 
> Any pointers would be appreciated.
> 
> Thanks.
> 
> Jim
> 
> 
> 

Jim,

You are being confused by data type conversions. The problem (probably) 
lies in your sub-query;

AND date = (select max(to_char(date)) from table where key = 'key')

If the column 'date' is actually a DATE datatype column then comparing it 
to a to_char'ed version of 'date' will not necessarily equate. What you are 
seeing is that SQL*Plus plays nice and manages to match the values in 
'date' successfully but whatever other tool you are using is much stricter 
and doesn't.

Do you *need* to have that 'to_char' in your sub-query? I can't see any 
need for it. If you re-write your query as;

SELECT * 
FROM   table 
WHERE  key = 'key' 
AND    date = (select max(date)
               from   table 
               where key = 'key')

everything should be fine. Of course if the column 'date' is not a DATE 
datatype column then my advice is invalid ...

Regards,
Andy
-- 
Smiggins Hole 2010 - Unleash the Mongrel


From cfelling at iae.nl  Thu Feb 28 11:16:10 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 17:16:10 +0100
Subject: sorting
References: <3c7e3d12_7@news.teranews.com>
Message-ID: <a5ll4a$4ml$1@animus.fel.iae.nl>

Jeff McComas <jmackd1 at yahoo.com> wrote:
> As a newbie i find the python scripting language very interesting and useful.
> I am trying to write a script that will search a .txt file and have the script
> tell me how many of the same numbers are in the .txt file. I know you can do
> it with the f.count command but my problem is how can you do an input that

What is this f.count method you're talking about?

> makes the numbers strings? Or puts double quotes and a comma around a number?

I'm not sure I understand what you mean, but le'me try.

> This is what a raw input from a .txt file looks like:

> 34.6        25        78        89.2       45.7
> 23          92        47        20         85
> 56          78        23        49         92

Okee let's read this whole file in

   >>> text = open("some-file.txt").read()

And now let's split it up discarding all whitespace and new lines

   >>> number_strings = text.split()

We now have a list of strings, each string containing the decimal
representation of a number.  For fun let's turn those strings into
real numbers --below we use `list comprehension', it's believed to
make code readable once you understand what it does.  The square
brackets indicate we are creating a sequence, and that sequence
consists of things `x' turned into reals `real(x)'.  The things that
are thus turned into reals are all the strings `for x in' we've put in
the sequence `number_strings' above--

   >>> numbers = [ real(x) for x in number_strings ]

Now let's count them using a dictionary (count) to keep track of the
number of times we saw a specific number.  If you would like to treat
`2.0' as being different from `2', then we'll have to use the string
representations from the file (number_strings), otherwise we'd better
use the computer representations of those numbers (numbers)

   >>> count = {}
   >>> for x in numbers:  # or: for x in number_string:
   ...     try:
   ...          count[x] += 1   # increment the count for this number
   ...     except KeyError:     # or if that number is not yet in the dict
   ...          count[x] = 1
   ...

> out of that how can i pull out the numbers that are the same, say 78 = 2

So let's look this up --the % operator on strings let you fill bits
and pieces on marked places (%r and %d) in that string--

   >>> print "the number %r occured %d time(s)" % (78, count[78]))

or to get them all

   >>> for x in count:
   ...    print "the number %r occured %d time(s)" % (78, count[78]))
   ...
-- 
groetjes, carel


From barry at zope.com  Tue Feb 26 11:28:19 2002
From: barry at zope.com (Barry A. Warsaw)
Date: Tue, 26 Feb 2002 11:28:19 -0500
Subject: Scheduled downtime for all mail at python.org & zope.org
Message-ID: <15483.47011.951930.194502@anthem.wooz.org>

We plan on upgrading the mail server mail.python.org on Wednesday
27-Feb-2002, at 11am -0500 (EST, US/Eastern).  This will affect all
mail and mailing lists on the python.org and zope.org domains.

Downtime should be minimal; if all goes well, just 10-15 minutes.  You
shouldn't even notice it.  If you still cannot get email through by
noon our time, check the www.python.org home page for status, or email
Barry (barry at zope.com), Martijn (mj at zope.com), and Greg
(gward at mems-exchange.org).

-Postmaster @ {python,zope}.org



From bldrake1 at yahoo.com  Tue Feb  5 06:04:37 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 5 Feb 2002 03:04:37 -0800
Subject: HappyDoc: Can't get output
Message-ID: <fb072181.0202050304.35626d25@posting.google.com>

When I try to run HappyDoc, I get the following:

commandline:
c:\python21\happydoclib> python happydocwin.py <anyfile.py>
output:
--------------------
HappyDoc version 2.0
Docstring converters from C:\Python21\happydoclib\docstring
Docsets list from C:\Python21\happydoclib\docset
Formatters from C:\Python21\happydoclib\formatter
Initializing documentation set HappyDoc Generated Documentation...
Getting info for test_happydoc.py
Beginning to write...
Traceback (most recent call last):
  File "C:\Python21\happydoclib\happydocwin.py", line 73, in ?
    HappyDoc().run()
  File "C:\PYTHON21\happydoclib\CommandLineApp.py", line 313, in run
    exit_code = apply(self.main, tuple(self.remainingOpts))
  File "C:\PYTHON21\happydoclib\appclass.py", line 430, in main
    parsed_modules.write()
  File "C:\PYTHON21\happydoclib\docset\docset_MultipleFile.py", line
110, in write
    self._root_node = self.openOutput( self._root_name, self._title,
'' )
  File "C:\PYTHON21\happydoclib\happydocset.py", line 807, in
openOutput
    f = self._formatter.openOutput(name, title, subtitle)
  File "C:\PYTHON21\happydoclib\formatter\formatter_HTMLFile.py", line
275, in openOutput
    title1,
  File "C:\PYTHON21\happydoclib\formatter\fileformatterbase.py", line
101, in openOutput
    happydoclib.path.rmkdir(happydoclib.path.dirname(name))
  File "C:\PYTHON21\happydoclib\path.py", line 83, in rmkdir
    rmkdir(parent)
  File "C:\PYTHON21\happydoclib\path.py", line 83, in rmkdir
    rmkdir(parent)
  File "C:\PYTHON21\happydoclib\path.py", line 83, in rmkdir
    rmkdir(parent)
  File "C:\PYTHON21\happydoclib\path.py", line 85, in rmkdir
    os.mkdir(path)
OSError: [Errno 2] No such file or directory: '\\C:'
----------------------
Anyone know what's going on?  I say the same question by someone else
on the SourceForge HappyDoc forum.  Any help would be appreciated.

System: win98, ActivePython 2.1, HappyDoc 2.0

Barry Drake
bldrake1 at yahoo.com


From kdahlhaus at yahoo.com  Wed Feb  6 08:15:07 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 6 Feb 2002 05:15:07 -0800
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <3c603127.18127765@news.t-online.de>
Message-ID: <283adf56.0202060515.1672a2b0@posting.google.com>

> Something like that, only bigger, and with sourcecode:
> 
> http://www.p-nand-q.com/cgidocs.htm

That cgidocs app you have is great!


From Mike_ba at web.de  Tue Feb 12 06:31:49 2002
From: Mike_ba at web.de (Michael Bauer)
Date: Tue, 12 Feb 2002 12:31:49 +0100
Subject: Exiting threads silently
Message-ID: <a4au9e$u1v$01$1@news.t-online.com>

Hi,

i have a small problem when exiting threads. According to the docs, a 
silent exit should be possible without an exception.

from <http://www.python.org/doc/2.2/lib/module-thread.html>:

"exit() 
Raise the SystemExit exception. When not caught, this will cause the thread 
to exit silently."

Here is the relevant code
(This is part of a proxy module)

----------------------------------------
    def error(self, code, body):
        """Answer Client with an error page

        XXX Really not very nice page...

        """
        import BaseHTTPServer
        response = BaseHTTPServer.BaseHTTPRequestHandler.responses[code][0]
        try:
            self.wfile.write("HTTP/1.0 %s %s\r\n" % (code, response))
            self.wfile.write("Server: iOwl.net ProxyServer\r\n")
            self.wfile.write("Content-type: text/html\r\n")
            self.wfile.write("\r\n")
            self.wfile.write('<html><head>\n<title>%d %s</title>\n</head>\n'
                    '<body>\n%s\n</body>\n</html>' % (code, response, body))
            self.wfile.flush()
            self.wfile.close()
            self.rfile.close()
        except IOError:
            # probably user aborted browser so we dont have a socket left   
            # to write to
            pass

        # Exit thread
        thread.exit()
----------------------------------------

What happens is the error page gets displayed correctly in the browser, but 
therad.exit() dumps following exception to the console:

----------------------------------------
Exception happened during processing of request from ('127.0.0.1', 40400)
Traceback (most recent call last):
  File "/usr/lib/python2.2/SocketServer.py", line 458, in 
process_request_thread
    self.finish_request(request, client_address)
  File "/usr/lib/python2.2/SocketServer.py", line 253, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/usr/lib/python2.2/SocketServer.py", line 514, in __init__
    self.handle()
  File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 170, in handle
    server, client = self.Connect(host, port)
  File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 346, in Connect
    self.error(200, 'Error connecting to "%s" (%s)' % (host, err))
  File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 504, in error
    thread.exit()
SystemExit
----------------------------------------

Anybody has an idea how i can exit silently?

thanks, Mike

(Running on linux from scratch, kernel 2.4.17 with Python 2.2)


From sjmachin at lexicon.net  Mon Feb 11 03:27:27 2002
From: sjmachin at lexicon.net (John Machin)
Date: 11 Feb 2002 00:27:27 -0800
Subject: No explanation for weird behavior in re module!
References: <a470re$1cl30l$1@ID-78052.news.dfncis.de> <c76ff6fc.0202101814.5d18209d@posting.google.com> <87pu3c7osj.fsf@uol.com.br>
Message-ID: <c76ff6fc.0202110027.378461bd@posting.google.com>

synthespian <synthespian at uol.com.br> wrote in message news:<87pu3c7osj.fsf at uol.com.br>...
> sjmachin at lexicon.net (John Machin) writes:
> 
> > synthespian <synthespian at uol.com.br> wrote in message news:<a470re$1cl30l$1 at ID-78052.news.dfncis.de>...
> > > Hi-
> > > 
> > > 	I'm really intrigued by this behavior:
> > > 
> > > >>> import re
> > > >>> p = re.compile('^(der|die|das(\s\w+))')
> > > >>> m = p.match('die T?r, T?ren')
> > > >>> n = p.match('das Auto, Autos')
> > > >>> m.group(0)
>  'die'
> > > >>> m.group(1)
>  'die'
> > > >>> m.group(2)
>  [nothing!!!!]
> > > >>> n.group(0)
>  'das Auto'
> > > >>> n.group(1)
>  'das Auto'
> > > >>> n.group(2)
> > > 'Auto'
>  (snip)
> > see that this matches just 'die' also.
> > 
> > Anticipating the next raft of problems: (1) You will need to use the
> > re.UNICODE flag when you call re.compile(), otherwise \w will not
> > recognise the Unicode alphabetics (this *is* documented) (2) You may
> > need to give it an input whose Python type is 'unicode' -- being able
> > to see the umlaut on your screen is not sufficient evidence of this
> > :-) (3) You should get into the habit of using the raw string notation
> > with your regexes whether it is necessary or not, else you will be
> > bitten in the future.
> > 
> > Anyhow, the following works for me:
> > 
> > Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> > >>> p = re.compile(r'^((der|die|das)(\s\w+))',re.UNICODE)
> > >>> p.match('das Auto').groups()
>  ('das Auto', 'das', ' Auto')
> > >>> z = u'die T\u00FCr, T\u00FCren'
> > >>> p.match(z).groups()
> > (u'die T\xfcr', u'die', u' T\xfcr')
> 
> Hi-
> 	Thanks for helping. Unfortunately, the solution did not yet solve the problem at hand.
> 	You see, I must read lines forma a plain text file, which has the form "das Auto, Autos".

It can't be very plain, if you expect to be able to represent German
characters with umlauts and the "Eszett". Do you actually know what
character set you have in this "plain text file"? Unicode UTF-16?
Unicode UTF-8? "Latin 1" ? If not, I suggest that you create a small
file with the umlauts etc and use xd to dump it out and post what are
the hex representations of u-umlaut and eszett and so forth.

> 	The original program has to ask if "Auto" has "der, die or das" as the definite article.
> 	The problem is that I can't make Python read anything with non-ASCII character set.

The problem is not the reading, it is knowing what character set you
have, and using Python accordingly.

> 	The output you have suggested, "T\xfcr", is, for all practical purposes, unreadable.

That was *not* "suggested output". It was printing out the "repr" of
the data to show that correctly-formulated Python code, derived from a
brief perusal of the documentation, was in fact working on Unicode
data. 0x00FC is the Unicode 16-bit representation of 'Latin small
letter u with diaeresis'. In discussing these types of problems, it is
best to focus on the internal representation of the data, not what you
see on the screen. WYSIWYG? Well, what you see as identical on your
screen or printer can have two or more possible internal
representations.

> 	As I understood it from the other posts, the "\w+" on the regex will depend on my locales. 

(1) You have only one locale at a time. (2) Was the input file created
in your locale? (2) RTFM & re-read Tim's post: if you want to work in
Unicode, you will need to (a) supply Unicode data and (b) to use \w
properly,  supply the re.UNICODE flag to re.compile()

> 
> 	I have, in /etc/enviroment. the following set-up (native language: Portuguese):
> 
> 	LANG=C
> 	LANGUAGE=pt_BR
> 	LC_ALL=pt_BR
> 	LC_CTYPE=latin1
> 	LESSCHARSET=latin1
> 	NLSPATH=/var/catman
> 	MM_CHARSET=ISO-8859-1
> 
> 	Seems like non-ASCII is a real bother in Python...I'd expected this to have been better looked
> after in Python-2.0...
> 
We don't have enough information yet to know whether it's a
"non-ASCII" problem.
Let's assume for the moment that Unicode was a red herring, caused by
your "Other than the fact that 'T?r' has the '?' unicode charcater".
Let's also assume that your script and your data are both in
ISO-8859-1 aka Latin-1, which can hack both Portguese and German. Then
your only remaining problem should be, as both Tim and I posted, the
problem with precedence of | in a regex. Try fixing that first, and
see if you still have a problem ... if so, it is likely to be with the
'?' not being recognised as alphabetic --- then you will need to talk
to the locale experts.  Oh and try this: print '?'.isalpha()
Whether the result is 1, 0, or an exception should tell you something.

HTH,
John


From jason at jorendorff.com  Sun Feb 17 13:02:26 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 17 Feb 2002 12:02:26 -0600
Subject: Editable strings anyone?
In-Reply-To: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEDIDDAA.jason@jorendorff.com>

> I'm wondering how difficult it would be to add an editable (mutable)
> string class to Python 2.2 implemented in C under the new type/class
> unification scheme.

It would be a pain.  I think you'd find that you can't really
reuse much (if any) of the code from stringobject.c, since you're
changing the underlying representation of the data in memory.

I don't think you could subclass str to do this at all; you'd
want to create a new type.

On the other hand, creating a new type isn't so tough.  If str
and list didn't have 3 dozen methods each, I'd say it's a breeze.  :)


> [...] Does this sound like a reasonable design?  Does it fit in
> reasonably with the CPython implementation?  Would it require a
> lot of hacking to existing string functions like the re module?

No, this sounds good, and everything should work okay.  I think
you would have to sneakily close the gap whenever the re module
asks you for your buffer pointer, because (I think) re needs a
contiguous buffer.  Not sure about that.

> Python surprisingly lacks a counterpart to StringBuffer.
> Python programmers tend to say either a=b+c+d+e+f (potentially
> taking quadratic time in the total number of characters) or some
> kludge like a=''.join(b,c,d,e,f).

Lately I've been using str.join.  My first impression was that it's
a hack, but I've changed my mind.  It's surprisingly flexible.
I can tinker with the list - even sort() or reverse() it - and then
at the end, I get whatever separator I choose.  Nice.

## Jason Orendorff    http://www.jorendorff.com/



From matthias_j at lycos.com  Wed Feb 13 02:20:07 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 12 Feb 2002 23:20:07 -0800
Subject: Just for the gurus -- Remote admin password changing with privileged user ??
References: <d7d5ebdf.0202121332.4f7d3723@posting.google.com>
Message-ID: <d7d5ebdf.0202122320.362660ab@posting.google.com>

Thanks to all of you.


From aries at bitstream.net  Fri Feb 22 02:15:29 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Fri, 22 Feb 2002 01:15:29 -0600
Subject: running a program via exec*
References: <a54294$l0e@r02n01.cac.psu.edu>
Message-ID: <3c75ee73$0$79558$6536d21b@news.bitstream.net>

Rajarshi Guha <rxg218 at psu.edu> wrote in message
news:a54294$l0e at r02n01.cac.psu.edu...
> Hi,
>   this is somewaht related to my earlier post. When I use execl(progname)
> it seems to work. However it always looks for the progrma in the current
> directory. Is there any way to specify the path from the environment
> without having to type out the path to the program itself?
>
> TIA,
>
> --
> -------------------------------------------------------------------
> Rajarshi Guha                  | email: rajarshi at presidency.com
> 417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
> Dept. Of Chemistry             | ICQ  : 123242928
> Pennsylvania State University  | AIM  : LoverOfPanda
> -------------------------------------------------------------------
> GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A
> Public Key     : http://pgp.mit.edu/
> -------------------------------------------------------------------
> All syllogisms have three parts, therefore this is not a syllogism.

yep
Yes, of course there is a way,

If I understand your question correctly you are going
to run a script FROM in some directory lets say the directory
is named

on unix
/mypycode/mypydir

on windows
c:\mypycode\mypydir


Now in the script file you want the script to execute
a program which is in some other directory on your
system lets say its named myprogdir.  And lets say
the actual path to that location is

on unix
/stuff/junk/myprogdir

on windows
c:\stuff\junk\myprogdir

and the program name you want to execute is some
simple program that shows the date lets say its
named showdate and its located in the myprogdir directory

the program below

---
#put she bang here for unix posix systems

import os
import string    #for the str() built in method
#---import sys

#operating system hard drive variables
myname = os.name
print "the os is: ", myname
if os.name == 'posix':
    myrootdir = "/"
elif os.name == 'nt':
    myrootdir = "c:\\"
else: print "This machine OS is something other than nt or posix\n"
print "the root dir is: ",myrootdir

#will be used to show me what current directory I am in
wheredir = os.getcwd()

#list the files in this directory
mydirfiles = os.listdir(os.curdir)

print "my current dir is: ", wheredir
print "the file listing is: \n", mydirfiles
print "##########\n"

# change to the other programs directory location
thedir = os.path.join(str(myrootdir),'stuff', 'junk', 'myprogdir')
os.chdir(thedir)

print "CHANGED- NOW my current dir is: ", thedir

#list the files in this directory
showfiles = os.listdir(os.curdir)

print "CHANGED- NOW the file listing is: \n", showfiles
print "##########\n"

#--now launch your program file named showdate
os.system("showdate")





From starx at pacbell.net  Sat Feb  2 01:21:12 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 2 Feb 2002 06:21:12 GMT
Subject: Making a list-inhereted object return standard repr?
References: <mailman.1012624684.17642.python-list@python.org>
Message-ID: <Xns91A8E36591E4DRASXnewsDFE1@130.133.1.4>

Jason Orendorff || Fri 01 Feb 2002 08:32:57p:

> Philip Swartzleonard wrote:
>> Ok, so i decided to make my container class inherit from list, [...]
>> I want to change __repr__ and __str__ to do something more like the
>> first case again [...] 
> 
> class Foo(list):
>     def __repr__(self):
>         basic = object.__repr__(self)
>         values = list.__repr__(self)
>         return basic[:-1] + ' containing ' + values + '>'
> 
> ## Jason Orendorff    http://www.jorendorff.com/

Thanks! That's copy-and-paste'ly what i need. I knew it was something more 
or less obvious i was missing %D.

(although i may have to leave out the contains, as it gives me things like     
<__main__.Cell object at 0x015F0210 containing [<__main__.Player object at  
0x015E7870 containing []>]> ... so far every object is a container :)

(Incidentally, my other problem was that i was doing something like 'if 
new_container == self: complain', which suddenly turned into a comparasion 
between two empty lists instead of two object ids... IS fixed it fine :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From Bill.Scherer at VerizonWireless.com  Thu Feb 28 07:18:42 2002
From: Bill.Scherer at VerizonWireless.com (Scherer, Bill)
Date: Thu, 28 Feb 2002 07:18:42 -0500 (EST)
Subject: Rotor in Jython?
In-Reply-To: <7x8z9e45b8.fsf@ruckus.brouhaha.com>
Message-ID: <Pine.LNX.4.33.0202280716080.11077-100000@bagend>

On 27 Feb 2002, Paul Rubin wrote:

> "Scherer, Bill" <Bill.Scherer at VerizonWireless.com> writes:
> > I've a need to perform rotor decryption in Jython.  A quick 
> > search didn't reveal anything to help.  I've begun coding it in 
> > Python from rotormodule.c. My C knowledge is not strong, so there 
> > are spots that are surely not right.
> 
> I hope you're not the rotor module for computer security.  That cipher
> is nowhere near as safe to use as a modern one like AES.

Understood. Let's just say it's a "locks are for honest people" 
situation.  If I could get rotor decryption in Jython in the near 
term, it would be useful.  Otherwise I need to make changes 
elsewhere that I'd rather not have to do at this time...

> 
> If you just have some old rotor-encrypted files you have to decrypt,
> I'd say get to a CPython environment and use the existing rotor module
> to decrypt them and be done with it.
> 
> Don't use the rotor module for new traffic that needs good security.
> 

Thanks for your concern.

-- 
Bill.Scherer at Verizon Wireless
RHCE 807101044903581




From ballabio at mac.com  Thu Feb 21 09:05:46 2002
From: ballabio at mac.com (Luigi Ballabio)
Date: Thu, 21 Feb 2002 14:05:46 +0000
Subject: Stupid Lambda Tricks [LONG]
In-Reply-To: <3c74c211.20012718@news.isar.de>
Message-ID: <5.1.0.14.0.20020221135925.00a76650@mail.mac.com>

Hi,

At 09:51 AM 2/21/02 +0000, Gerson Kurz wrote:
>* Using IF statements
>
>4.16. Is there an equivalent of C's "?:" ternary operator?

<snip>

>This method is of course trivially defined as lambda:
>
>IF = lambda a,b,c:(a and [b] or [c])[0]

I wouldn't advise using anything of the sort of your IF thing, because it 
doesn't short-circuit. Try this:

def a():
     return 1
def b():
     print "lengthy b() calculation, possibly with side effects"
def c():
     print "lengthy c() calculation, possibly with side effects"

# Use the ternary trick:
 >>> (a() and [b()] or [c()])[0]
lengthy b() calculation, possibly with side effects

# Use you IF thing instead:
 >>> IF(a(),b(),c())
lengthy b() calculation, possibly with side effects
lengthy c() calculation, possibly with side effects

Bye,
         Luigi




From lac at strakt.com  Thu Feb 21 07:48:57 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 21 Feb 2002 13:48:57 +0100
Subject: Adding static typing to Python 
In-Reply-To: Message from Alexander Jerusalem <ajeru@gmx.net> 
   of "Thu, 21 Feb 2002 03:05:40 +0100." <5.1.0.14.0.20020221025222.02aad830@pop.gmx.net> 
References: <Message from ajeru@vknn.org (Alexander Jerusalem) of "20 Feb 2002 13:00:29 PST." <24c39b2c.0202201300.3dfe7dbc@posting.google.com> <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com> <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com> <24c39b2c.0202192354.765da777@posting.google.com> <mailman.1014210154.9297.python-list@python.org> <24c39b2c.0202201300.3dfe7dbc@posting.google.com>  <5.1.0.14.0.20020221025222.02aad830@pop.gmx.net> 
Message-ID: <200202211248.g1LCmvtB001743@ratthing-b246.strakt.com>

> Alexander Jerusalem:
> But there's another feature of static typing that I find extremely useful 
> and that's type based dispatching, ideally multi methods. And that's not 
> covered by dynamic protocoll detections as suggested in the paper as far as 
> I understand it.

Yes.  I'd like to discuss this more with more interested parties.
I don't think that type-based dispatching has to rely on _static_
types, just strong types.  But then mayhap I am using 'type based
dispatching' more loosely than is normally the case.

Laura



From ahimsa at inext.co.za  Thu Feb 14 00:37:19 2002
From: ahimsa at inext.co.za (ahimsa consulting)
Date: Thu, 14 Feb 2002 07:37:19 +0200
Subject: py2exe & McMillan's installer (was Re: Is making Python executables a reality?)
References: <E16bC9r-0000zp-00@mail.python.org>
Message-ID: <008001c1b519$d9300fa0$0100000a@ahimsa>

Hi Pythonites

Could someone kindly post the URL for downloading the py2exe program and
also the "Gordon McMillan's installer" that David referenced (see below).

Thanks folks

Andrew


>    1. Re: Is making Python executables a reality? (David Bolen)
> Subject: Re: Is making Python executables a reality?
> From: David Bolen <db3l at fitlinxx.com>
> Date: 13 Feb 2002 19:25:34 -0500
> Organization: Fitlinxx, Inc. - Stamford, CT
> To: python-list at python.org
>
>
> We use Gordan McMillan's installer (but others use py2exe) to package
> the application and required files, and then use InnoSetup to build
> the Windows installer.  Works great.
>
> Perhaps if you posted some more specifics about what didn't work for
> you with py2exe someone might be able to suggest a solution.
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/





From James_Althoff at i2.com  Thu Feb  7 19:11:00 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 7 Feb 2002 16:11:00 -0800
Subject: 2001 Turing Award
Message-ID: <OF6F5515F7.58E9AACF-ON88256B59.0083423F@i2.com>

[Keith Keller]
> Damn--I thought the award was for the person who best impersonated
> a Turing machine.  :)

[Mark McEahern]
>Do you have any nominees?

Seriously though, if you are interested in the history of object-oriented
programming, Simula (Dahl and Nygaard -- 2001 Turing Award winners) is a
fascinating place to start.

For example, check out:

SIMULA begin
  Graham M. Birtwistle, Ole-Johan Dahl, Bjoern Myhrhaug, and Kristen
Nygaard.
  Petrocelli/Charter, New York (1975).
  ISBN 0-88405-340-7.

http://www.isima.fr/asu/asubook.html

Of course, if you really want to dive into the primordial-soup mother lode
(of OO), go straight to Ivan Sutherland's Sketchpad thesis (MIT 1963).

http://arc.cs.odu.edu:8080/dp9/getrecord/oai_rfc1807/0018.mit.theses/1963-10

Alan Kay -- the creator of Smalltalk -- has called Sketchpad "the most
astonishing piece of work ever done by a single person in the field of
computing" (paraphrasing).

Happy reading,

Jim




From Christopher.Saunter at durham.ac.uk  Thu Feb 14 12:35:03 2002
From: Christopher.Saunter at durham.ac.uk (Christopher Saunter)
Date: Thu, 14 Feb 2002 17:35:03 +0000 (UTC)
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <X1va8.207034$KM2.8517807@atlpnn01.usenetserver.com>
Message-ID: <a4gsg7$m4i$1@sirius.dur.ac.uk>

Steve Holden (sholden at holdenweb.com) wrote:
: Chris: Welcome to c.l.py!
Thanks, and thanks to everyone else who has so far replied.  I'll have to
remember the 'holliday' bug with August ;-)

: It might be possible to introduce it as a redundant alternative, but
: generally speaking there's resistance to having more than one way to
: represent something (confuses learners), and this wouldn't anyway have saved
: you from the glitch you experienced.

Yeah.  It'd be nice if there could be a plan to switch to a more obvious
0o over several versions (as is being down with the integer divide) but
it's kind of nice to know that even python has some quirks....
Unfortunatly I started my computing with an 8 bit Amstrad, so I never
experience the old unix boxes...


: Serves you right for being anal about alignment when only the interpreter
: and you were going to read the code <0.9 wink>.

;-)  I guess it's kind of a 'nervous programming tick' or something...

Also, on the subject of number bases, is there an easy (read library)  way
to take a string containing a number represented as hex and convert it to
an int or float?  I've looked through the libraries, and I'm sure it's
there somewhere, just can't find it.....

Regards,
	Chris Saunter


From greg at cosc.canterbury.ac.nz  Mon Feb 25 21:23:39 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 26 Feb 2002 15:23:39 +1300 (NZDT)
Subject: Time to rename Stackless?
In-Reply-To: <3C7A151F.1090709@tismer.com>
Message-ID: <200202260223.PAA23591@s454.cosc.canterbury.ac.nz>

Christian Tismer <tismer at tismer.com>:

> It is still stackless in the sense that order of execution is not
> restricted by the C stack.

Even so, I still think it's a poor name, because it
focuses on an implementation detail rather than
telling you what it can do for you.

> We have now tasklets, proclets, and the stack slices are
> stacklets. :)

Hmmm... maybe it should be called Pythonlet...

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+



From asim_beg at hotmail.com  Wed Feb 20 16:11:21 2002
From: asim_beg at hotmail.com (Asim Beg)
Date: Wed, 20 Feb 2002 15:11:21 -0600
Subject: MySQLdb build on Solaris 7
Message-ID: <F126tRc0JFaTijlqzUr00002741@hotmail.com>

Problem solved.

import MySQLdb works like a charm...

I followed the instructions from this thread..

http://sourceforge.net/forum/message.php?msg_id=297985

Jeff, thanks for all your help so far..;-). If you look at the above you 
will see that the build/installation scripts can be more robust.


On to Python + Mysql adventures.....

Asim


>From: Jeff Bauer <jbauer at rubic.com>
>To: python-list at python.org, asim_beg at hotmail.com
>Subject: Re: MySQLdb build on Solaris 7
>Date: Wed, 20 Feb 2002 14:33:14 -0600
>
>Asim Beg wrote:
> > Frankly, I am surprised (shocked...) to find that an
> > important package such as MySQLdb for python would
> > not compile on Solaris...
>
>Let's get some perspective here.  MySQLdb python has compiled
>and is running on thousands of Solaris machines ... just not
>yours <0.5 wink> -- yet.
>
>Further questions:
>
>1. What is in your LD_LIBRARY path?
>
>2. Which version of the MySQL server is running
>on your Solaris machine?
>
>3. Did you install MySQL database server on the
>machine or did the admin?  If the latter, did they
>compile it on the machine or install a binary
>distribution (which might result in a version
>mismatch)?
>
>And one suggestion:  If all else fails you might
>consider compiling a local MySQL server from the
>sources -- producing the libs relative to your home
>directory -- and then building the MySQLdb python
>interface against those libs --- then testing your
>setup your local MySQL server.
>
>Jeff Bauer
>Rubicon Research




_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx




From gcordova at hebmex.com  Wed Feb 20 11:07:30 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 10:07:30 -0600
Subject: HTTP state management without cookies?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC2@SRM201002>

> Perhaps you could use some mod_rewrite magic to transform:
> 
> http://foo.bar/_SID_=2374820394/myscript.cgi/
> 
> into:
> 
> http://foo.bar/myscript.cgi?_SID_=2374820394
> 
> so the same trick will work for CGI?
> 
> - Geoff
> 

This will hit the QUERY_STRING parameter.

Another less used parameter is PATH_INFO, the rewrite could
be into something like this:

http://foo.bar/_SID_=2374820394/myscript.cgi/

into something like:

http://foo.bar/myscript.cgi/2374820394

Apache takes the "/2374820394" and puts it into PATH_INFO,
where you can take it from.

And that leaves you with QUERY_STRING untouched.

-gus



From jkraska at san.rr.com  Mon Feb 11 00:49:25 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 11 Feb 2002 05:49:25 GMT
Subject: Fear of autocoding! >Re: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net>
Message-ID: <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com>

>Very good example of someone who is afraid of autocoding.

You never did answer my question, by the way. If one has
become delusional, how does one become aware of it? What
metric does one apply to determine if one's convictions
are out of line with what is actually real?

C//



From phr-n2002a at nightsong.com  Mon Feb  4 02:23:27 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 03 Feb 2002 23:23:27 -0800
Subject: pickle security
References: <mailman.1012591743.10841.python-list@python.org> <7xit9g51lz.fsf@ruckus.brouhaha.com> <j4g04ihc7k.fsf@informatik.hu-berlin.de> <7xsn8im3uh.fsf@ruckus.brouhaha.com> <m3k7ttkaa6.fsf@mira.informatik.hu-berlin.de>
Message-ID: <7xd6zlvhg0.fsf@ruckus.brouhaha.com>

martin at v.loewis.de (Martin v. Loewis) writes:
> > The code to deal with quote strings calls eval, and without actually
> > examining the eval implementation (big, complicated) there's no way to
> > confirm that it can't somehow run arbitrary code found in the string.
> 
> Then, please do examine the code. Not doing so and still claiming
> there is a problem means to spread FUD.

There is a lot of code and the very size of the code means I might
miss something.  However, it's not FUD to say that a complicated piece
of code is harder to check for exploits than a simple one.

> > Microsoft does that and that's why we keep hearing about new
> > exploits being discovered in Microsoft products.  The code has to be
> > checked thoroughly, which is easier to do when the code is simple.
> 
> In this specific case, it is relatively easy to check that the eval is
> safe. It is not the case that the risk of an exploit has been
> ignored. Instead, it is the case that the risk has been analysed and
> found to be non-existing. If you chose to ignore this analysis, fine -
> but please don't tell people that there is a security risk.

I spent a few minutes checking the eval when I first reported the
pickle security bug, and the few minutes was enough to see that to do
a really thorough check, I'd have to spend hours, and I still might
miss something, or eval could change in a some future Python version
when no one is thinking about how it interacts with hostile pickles
(that's how pickles got to be insecure in the first place).  Since the
correct fix is to just not use eval, I don't want to spend hours
checking eval (and re-checking it with each new Python release) unless
someone is paying me.  Are you offering?

> In any case, there is a patch on SF that removes the eval, see
> 
> http://sourceforge.net/tracker/?func=detail&aid=505705&group_id=5470&atid=305470
> 
> The rationale for doing so is different, though: eval interns strings,
> which is undesirable for unpickling.

The patch is desirable for security reasons too, so I'm glad it's
there.  I'll look at the pickle code again next time I have an
application where I want to unpickle hostile data.  For now, I'm doing
ok with alternative serialization methods.


From richard.papworth at bbsrc.ac.uk  Mon Feb 18 13:20:49 2002
From: richard.papworth at bbsrc.ac.uk (Richard Papworth)
Date: 18 Feb 2002 10:20:49 -0800
Subject: PyGres fetchone problem
Message-ID: <695dd93e.0202181020.3c1cb4c5@posting.google.com>

Playing around with the PyGres module I came across the following
problem when using fetchone to iterate over a cursor, an idiom that
I've seen used in many examples:

Python 1.5.2 (#1, May  1 2001, 10:59:55) [C] on osf1V4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import pgdb
>>> connection = pgdb.connect(':aDatabase:::')
>>> cursor = connection.cursor()
>>> cursor.execute('select distinct database_id from database_table')
>>> cursor.rowcount
3
>>> while 1:   
...     aRow = cursor.fetchone()
...     if aRow == None: break
...     print aRow
... 
[5]
[5]
[5]
>>> cursor.execute('select distinct database_id from database_table')
>>> cursor.fetchall()
[[5], [6], [7]]
>>> 

The cursor.fetchone() loop should produce
[5]
[6]
[7]

However, although the cursor seems to be advancing through the result
set, only the first row of the result set is ever retrieved.

Am I doing something wrong here?

Cheers

Richard


From phd at phd.pp.ru  Fri Feb 15 09:25:27 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 15 Feb 2002 17:25:27 +0300
Subject: certification
In-Reply-To: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>; from thoa0025@mail.usyd.edu.au on Fri, Feb 15, 2002 at 01:50:54PM +0000
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>
Message-ID: <20020215172527.A20767@phd.pp.ru>

On Fri, Feb 15, 2002 at 01:50:54PM +0000, Trung Hoang wrote:
> i was wondering if i can get certified with python from an internationally
> recognised company?

   http://www.brainbench.com
   Python 1.5 - $50.

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



From dkuhlman at rexx.com  Wed Feb 13 11:37:51 2002
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: 13 Feb 2002 10:37:51 -0600
Subject: How to generate XML
References: <3c21e330@news.opennet.it> <j44rmko5s6.fsf@informatik.hu-berlin.de> <a0212d$57k$1@news1.xs4all.nl>
Message-ID: <1013618891.574031@rexx.com>

Irmen de Jong <usenet at nospam-irmen.cjb.net> wrote:
> 
> 
> "Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
> news:j44rmko5s6.fsf at informatik.hu-berlin.de...
> 
>> I'm amazed that people always start proposing difficult-to-use
>> libraries when this question comes up. I'd prefer to use
>>
>> print "<foo/>"

For the special case where you have a tree structure whose nodes
are instances of Python classes, implement an "export" method in
each class something like the following:

    def export(self, outfile, level):
        showIndent(outfile, level)
        outfile.write('<special>\n')
        for element in self.subelements:
            element.export(outfile, level + 1)
        showIndent(outfile, level)
        outfile.write('</special>\n')


where showIndent is, perhaps, as follows:

    def showIndent(outfile, level):
        for idx in range(level):
            outfile.write('    ')


Then call export(outfile, 0) on the root node of the tree.

I have had this special case/need quite often.  Following the above
conventions makes it quite ease to generate an XML representation
of my data structures.

  - Dave


-- 
Dave Kuhlman
dkuhlman at rexx.com
http://www.rexx.com/~dkuhlman


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


From kp87 at lycos.com  Thu Feb 14 01:54:46 2002
From: kp87 at lycos.com (kevin parks)
Date: 13 Feb 2002 22:54:46 -0800
Subject: filtering random values (sieves?)
References: <5e8bd451.0202130827.2df6f50b@posting.google.com> <a4e5na$1efggq$1@ID-11957.news.dfncis.de>
Message-ID: <5e8bd451.0202132254.5e90217b@posting.google.com>

Uhmm... yeah. stupid me. As amazing as it seems, it never occured to
me to just generate the values i wanted in the first place.

But what i was hoping to do was generate two somewhat large lists
algorhytmically and use one list to filter the other, which would be
sort of  like getting the interesection, like so:

def intersect(*args):
	intersection = []
	for x in args[0]:     			
		for foo in args[1:]:			
			if x not in foo: break	
        	else: 						
            		intersection.append(x)		
    	return intersection


Not sure that is the best way, but the idea of just choosing from the
items in the first place is, uhmm <embarrasing mode on> obvious
</embarrasing mode off>

It's hard  to find a rock big enough to crawl under. Boy sometimes the
answer really is right under your nose aint it?

best,

kevin


From fgeiger at datec.at  Tue Feb  5 02:16:27 2002
From: fgeiger at datec.at (F. GEIGER)
Date: Tue, 5 Feb 2002 08:16:27 +0100
Subject: CPython vs. Jython for Web applications?
References: <c775c8db.0202041126.5177e606@posting.google.com>
Message-ID: <a3o0um$229$1@rex.ip-plus.net>

> I'm considering writing a fairly sizable Web app in Python and would

Did you consider webware.sourceforge.net as a framework?

> motivation for considering Jython is "native" access to Java's more
> mature XML libraries for XSLT (e.g., Saxon) and SVG (e.g., Apache's

I use 4XSLT. It is not that fast but I experienced no stability problems (my
XML projects are pretty small, though). And as Brian pointed out it should
become faster in the next release.

Cheers
Franz

"John E. Barham" <jbarham at octopuswireless.com> schrieb im Newsbeitrag
news:c775c8db.0202041126.5177e606 at posting.google.com...
> Good day,
>
> I'm considering writing a fairly sizable Web app in Python and would
> appreciate advice on whether to use CPython or Jython.  My main
> motivation for considering Jython is "native" access to Java's more
> mature XML libraries for XSLT (e.g., Saxon) and SVG (e.g., Apache's
> Batik).
>
> The constraints are:
>  - data to be stored in a relational database, preferably open-source
>  - app must run on both Win32 and some free Unix, preferably FreeBSD
>  - interface will be generated using XSLT
>  - app must run in Apache using some form of embedding (i.e.,
> out-of-process overhead of classic CGI is unacceptable)
>
> I'm more interested in ease of development than optimal performance
> but performance still matters...
>
> Some questions:
>   Has anyone written a sizable Web app in Jython?
>   Are there real-world benchmarks comparing, say, Jython & Saxon vs.
> CPython & Pyana/Xalan?
>   How do Jython servlets compare to using something like mod_python in
> terms of speed and ease of use and scalability?
>   How painful is it to do Jython database access vs. the standard
> Python DBI?
>
> FWIW, I have a lot of experience with C++, CPython and XSLT but just
> an undergrad course's worth in Java.
>
> I appreciate that Jython's performance is dependent on the JVM, so
> advice would be appreciated there as well, especially regarding the
> performance of JVMs for the free Unices.
>
> Thanks in advance.
>
>   John




From mlh at vier.idi.ntnu.no  Wed Feb 13 13:24:57 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 13 Feb 2002 18:24:57 +0000 (UTC)
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com>
Message-ID: <slrna6lbrp.guj.mlh@vier.idi.ntnu.no>

In article <b16e4ef7.0202131021.2b105a66 at posting.google.com>, Stephen
Ferg wrote:
>I'm working on something that I call EasyGui.  It is still very much
>in the roughing-out stage, and I wouldn't announce it except for the
>fact that I've seen a couple of questions posted, to which EasyGui
>might be the answer.
>
>Here is a bit of a write-up...
[...]
>I'm still a newbie Pythonista myself, so I welcome suggestions for
>improvement -- especially if they come in the form of working code
>snippets!

Perhaps take a look at the Anygui project? One of its chief aims is to
be dead simple.

>-- Steve Ferg (Steve at Ferg.org)

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From mac at ceti.pl  Tue Feb 12 10:30:08 2002
From: mac at ceti.pl (Maciej Starzyk)
Date: Tue, 12 Feb 2002 15:30:08 +0000 (UTC)
Subject: Help with Installer.
References: <Xns91B3864EE6013maccetipl@149.156.4.11> <Xns91B36755D3F49gmcmhypernetcom@199.171.54.214>
Message-ID: <Xns91B3A738AD5C1maccetipl@149.156.4.11>

Gordon McMillan <gmcm at hypernet.com> wrote in
news:Xns91B36755D3F49gmcmhypernetcom at 199.171.54.214: 

> The only way I know of (and boy, I hope I've
> missed something!) is to install the executable
> on the user's PATH. It should be able to find
> itself, at the expense of some hunting and pecking.
> 
> -- Gordon
> http://www.mcmillan-inc.com/
> 

Thanks for your reply.
Unfortunately, I cannot put it in PATH. 
It is executed as a CGI script, and I have no 
write access to any directory in the apache's PATH.

Maciek


From mats at laplaza.org  Thu Feb  7 17:40:07 2002
From: mats at laplaza.org (Mats Wichmann)
Date: Thu, 07 Feb 2002 22:40:07 GMT
Subject: SourceForge Python foundry ????
Message-ID: <3c6301ac.188163695@news.laplaza.org>

What's up with the Python foundry on SourceForge? It's not listed in
any of their lists of foundries, although if you plug in "python" in
the standard path they use for a foundry it comes up fine...

-- mats


From sjmachin at lexicon.net  Thu Feb 14 17:41:13 2002
From: sjmachin at lexicon.net (John Machin)
Date: 14 Feb 2002 14:41:13 -0800
Subject: searching a list of tuples
References: <mailman.1013636242.28850.python-list@python.org>
Message-ID: <c76ff6fc.0202141441.c6ade8f@posting.google.com>

"Jason Orendorff" <jason at jorendorff.com> wrote in message news:<mailman.1013636242.28850.python-list at python.org>...
> Marcus Stojek writes:
> > I have a (huge) list of tuples :
> > list=(float 1,float 2,float 3,int).
> > 
> > The list has been sorted with list.sort()
> > Now I have to slice the list in a way
> > that:
> > 
> >   a <= float 1 <= b
> 
> Well, on my computer I can make a float that
> represents "infinity".  So this works:
> 
> import bisect
> 
> INF = 1e600
> left = bisect.bisect_right(mylist, (a, -INF, -INF, -INF))
> right = bisect.bisect_left(mylist, (b, INF, INF, INF))
> myslice = mylist[left:right]
> 

Couple of problems here:

(1) As you said, "on my computer". This trick does not work on all
platforms for floats, and cannot be extended to types (like integer)
that don't have a representation of "infinity". Pax Knuth, but cute
algorithms that depend on sentinels that "can't appear in the data"
are a PITA.

(2) The OP has misunderstood [see his post requesting help with
optimisation] and is calling bisect* with an actual data value as the
second arg; result: lossage if that data value actually appears in the
list. The OP wanted
a <= alist[i] <= b.

>>> alist = [10, 20, 30, 30, 30, 40]
>>> import bisect
>>> def slicer(srtdlist, lowval, highval):
...    lowinx = bisect.bisect_right(srtdlist, lowval)
...    highinx = bisect.bisect_left(srtdlist, highval)
...    return srtdlist[lowinx:highinx]
...
>>> slicer(alist, 0, 10)
[]
>>> slicer(alist, 0-1, 10+1)
[10]
>>> slicer(alist, 20-1, 30+1)
[20, 30, 30, 30]
>>> slicer(alist, 20, 30+1)
[30, 30, 30]


From db3l at fitlinxx.com  Fri Feb  1 20:38:58 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Feb 2002 20:38:58 -0500
Subject: Defending the rodent (was Re: Defending the Python lanuage...)
References: <mailman.1012606383.28987.python-list@python.org>
Message-ID: <uit9gslvx.fsf@ctwd0143.fitlinxx.com>

James_Althoff at i2.com writes:

> Of course, if Douglas Englebart had had his way, this would all be moot.
> Englebart -- who invented the computer mouse -- didn't like moving his hand
> off the keyboard either.  So for his famous NLS system he devised a
> one-handed "chording" keypad on which he typed (all characters) using his
> left hand while manipulating the mouse with his right.  Decades later he
> still maintained that this is *the* way to do it.  :-)
> 
> (QWERTY who?)

Drat, and after I spent all those years learning to type really fast
on a keyboard whose layout was designed to slow me down :-)

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


From jbauer at rubic.com  Tue Feb 19 23:15:40 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 19 Feb 2002 22:15:40 -0600
Subject: MySQLdb build on Solaris 7
Message-ID: <3C7322EC.61C07B4D@rubic.com>

Asim writes:
> My ISP has Python 2.1 installed and I can run Python 
> CGI programs fine. I now need to connect to MySQL. I 
> need to install MySQLdb and am having difficulty 
> getting the MySQLdb source to build. I have tried
> both MySQL-python-0.9.0 and MySQL-python-0.9.1.
>
> The machine is a solaris 7 with Python 2.1 on it. When 
> I try to build it starts building and then goes into
> an infinite loop with strange libmysqlclient.a error
> messages..


Asim,

I have Andy Dustman's MySQL-python-0.9.1 interface
running with Python 2.1 on Solaris.

You probably need to edit setup.py to reflect the
locations of the libraries and include files, especially
if you have MySQL in non-standard directories.

Look for the line of code that begins with ...

  elif sys.platform == "sunos5": # Solaris 2.8 

... and edit the directories to reflect your
local configuration.

> Also since I cannot install MySQLdb in the standard 
> places on Solaris, how should I go about after the
> build step.

After installation, I've got the following installed:

  /usr/local/lib/python2.1/site-packages/

  CompatMysqldb.py        _mysql.so
  CompatMysqldb.pyc       _mysql_exceptions.py
  MySQLdb/                _mysql_exceptions.pyc

... and ...

  /usr/local/lib/python2.1/site-packages/MySQLdb/

  __init__.py       constants/        cursors.pyc  times.pyc
  __init__.pyc      converters.py     sets.py
  connections.py    converters.pyc    sets.pyc
  connections.pyc   cursors.py        times.py

If you have a similar file/path structure somewhere in your
importable PYTHONPATH, you should be okay.

Jeff Bauer
Rubicon Research



From marknenadov at winningteam.com  Mon Feb 25 14:02:41 2002
From: marknenadov at winningteam.com (Mark J Nenadov)
Date: Mon, 25 Feb 2002 13:02:41 -0600
Subject: XML RPC Woes
Message-ID: <157242002212519241790@winningteam.com>

Brian,

After some investigation, it appears that the source
of that core dumping problem lies with mcrypt
rather than having to do with XMLRPC.

~Mark

_____________________________________________
Free email with personality! Over 200 domains!
http://www.MyOwnEmail.com




From gcordova at hebmex.com  Fri Feb 22 10:18:39 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 09:18:39 -0600
Subject: Indexes and Slices.
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AD7@SRM201002>

Regarding the current (2.2) behaviour of indexes
and slices (you know, array-like behaviour), here's
what I found out today.



## Start up pythonwin.

PythonWin 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2001 Mark Hammond (mhammond at skippinet.com.au)
 - see 'Help/About PythonWin' for further copyright information.

# Let's first create a simple, gentle class.

>>> class SliceTest:
... 	def __getitem__(self, *args):
... 		print "getitem: %s" % repr(args)
... 	def __setitem__(self, *args):
... 		print "setitem: %s" % repr(args)
... 	def __delitem__(self, *args):
... 		print "delitem: %s" % repr(args)
... 
>>> 

## And now, a simple, gentle object.

>>> st = SliceTest()

## Let's test the waters first.

>>> i = st[10]
getitem: (10,)

## OK, this was expected. Just returned the key
## in a one element tuple.

>>> i = st[10:20]
getitem: (slice(10, 20, None),)

## Neat, the slice notation becomes a slice
## object, which has a third element... why?
## Should I specify a third part to the slice?

>>> i = st[10:20:4]
getitem: (slice(10, 20, 4),)

## Yes, seems like we can say "from 10 to 20, by 4"

>>> # MOST INTERESTING!!

## Indeed.

>>> st[10] = 12
setitem: (10, 12)

## No surprises here.

>>> st[10:23] = "hola"
setitem: (slice(10, 23, None), 'hola')

## Hmmm... slice object again. Lets look further.

>>> st[10:23:3] = "hola"
setitem: (slice(10, 23, 3), 'hola')

## Yep, same behaviour. Neat.

>>> st[12,2] = "hola"
setitem: ((12, 2), 'hola')

## THIS was news.
## Commas in indexes convert the index into a tuple.
## So I can use more complex behaviours, like multi-
## dimensional indexing.

>>> # Hmmm... commas in key argument seem to convert to tuple automagically.

## Yep.

>>> st[12,'a',4] = "Some wierd place in space"
setitem: ((12, 'a', 4), 'Some wierd place in space')

## Nice generalization.
## Very useful, I suppose.

>>> # cool.

## Yes, very.

>>> 

Now, all this is VERY interesting, specially the last
part about automagically converting indexes with commas
in them to tuples.

This will make possible creating true matrix-like behaviours,
instead of hacking it with arrays-of-arrays.

right?
--
Gustavo C?rdova Avila
> gcordova at sismex.com
? 8351-3861 | 8351-3862



From gerson.kurz at t-online.de  Mon Feb 25 09:23:02 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 14:23:02 GMT
Subject: MSWord and Python
References: <3C7A4EF8.4010404@thomas-guettler.de>
Message-ID: <3c7a478a.4742062@news.t-online.de>

On Mon, 25 Feb 2002 15:49:28 +0100, Thomas Guettler
<st-newsgroups at thomas-guettler.de> wrote:

>1. Where can I find a introduction on the net?
http://p-nand-q.com/win32com_and_office.htm

>2. Where can I find the documentation of methods that MSWord
>  exposes by the use of COM?

If you have the Jan 2002 MSDN, go to

ms-help://MS.MSDNQTR.2002JAN.1033/off2000/html/woobjapplication.htm

If not, go to http://msdn.microsoft.com and work your way to

Office Solutions Development
-Microsoft Office
--Microsoft Office 2000
---Microsoft Office 2000 Language Reference
----Microsoft Word 2000 Reference
-----Microsoft Word Visual Basic Reference

>3. I want take a word template. Replace text in it, and save it again.
>  Has someone code examples fot something like this?
Hint: A word template is just an additional parameter for
word.Documents.Add() (I had to find that out myself, too). Be warned,
though, the Word API *sucks* (TM). I'll put up a more complete example
of how to generate word-from-html on my site in a few days, an
incomplete and buggy version is at http://p-nand-q.com/pydocgen.htm




From jdavis at empires.org  Sun Feb 24 15:44:22 2002
From: jdavis at empires.org (Jeff Davis)
Date: Sun, 24 Feb 2002 12:44:22 -0800
Subject: Problems with POST
References: <8f8ffe67.0202230614.7ea56c9c@posting.google.com>
Message-ID: <a5bjib$l1c$1@news1.ucsd.edu>

Can you run the second script as a cgi and load it with a web browser to 
determine which script is not behaving correctly? I have a suspicion that 
the first script is somehow not posting quite correctly.

Regards,
        Jeff

Sunit Joshi wrote:

> Hello
> Somehow I can't seem to get any data with POST although the same
> script (with modifications in the first one) works for GET . Here's
> the scripts: First one is to POST and next one to recieve the POSTED
> data.
> 
> #!C:\Python21\Python.exe -u
> #To POST and output the data - postTest.py
> 
> from httplib import HTTP
> from urllib import quote
> 
> quoteData = "Sunit Joshi"
> postData = 'txtName=' + quote(quoteData)
> 
> print 'Will POST ', len(postData), 'bytes:', postData
> 
> # begin HTTP request
> req = HTTP('sunitw2k:6080')
> req.putrequest("POST" ,'/PyCgi/formTest.py')
> req.putheader("Accept", "text/html")
> req.putheader("User-Agent", "postTest.py")
> 
> # Set Content-length to length of PostData
> req.putheader("Content-length", str(len(postData)))
> req.endheaders()
> 
> # Send POST data after ending headers
> req.send(postData)
> 
> replycode, message, headers = req.getreply()
> print "HTTP RESPONSE: ", replycode, message
> print 'HEADERS: ', str(headers)
> 
> # Get file-like object from HTTP response
> # and print received HTML to screen.
> fd = req.getfile()
> txtlines = fd.read()
> fd.close()
> print '\nReceived following HTML:\n'
> print txtlines
> 
> #************************
> #!c:\Python21\Python.exe -u
> # To send the data posted - formTest.py
> import sys
> import cgi
> 
> sys.stderr = sys.stdout
> 
> form = cgi.FieldStorage()
> print "Content-type: text/html"
> print
> if not form:
> print "<h2>Field 'txtName' not filled in.</h2>"
> print str(form)
> else:
> print '<h3>You entered <font color="#FF6600">%s</font></h3>' %
> form.getvalue('txtName','none')



From sjmachin at lexicon.net  Wed Feb 20 18:19:57 2002
From: sjmachin at lexicon.net (John Machin)
Date: 20 Feb 2002 15:19:57 -0800
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk> <mailman.1014230314.2890.python-list@python.org>
Message-ID: <c76ff6fc.0202201519.509ec73e@posting.google.com>

Gustavo Niemeyer <niemeyer at conectiva.com> wrote in message news:<mailman.1014230314.2890.python-list at python.org>...
> [...]
> > I need to create Large (>300 bit) python Long's from C. I don't need (and
> > in fact, don't want) to use the features of the (apparently obselete?)
> > mpz module, or gmpy, or whatever. I simply want to say:
> > 
> > val = PyLong_FromOctets(char *v, int vlen);
> 
> You're probably looking for PyLong_FromString().

It is possible that the OP wants, but is not looking very hard for, this:

[from Python 2.2 source, in include/longobject.h]
"""
/* _PyLong_FromByteArray:  View the n unsigned bytes as a binary integer in
   base 256, and return a Python long with the same numeric value.
   If n is 0, the integer is 0.  Else:
   If little_endian is 1/true, bytes[n-1] is the MSB and bytes[0] the LSB;
   else (little_endian is 0/false) bytes[0] is the MSB and bytes[n-1] the
   LSB.
   If is_signed is 0/false, view the bytes as a non-negative integer.
   If is_signed is 1/true, view the bytes as a 2's-complement integer,
   non-negative if bit 0x80 of the MSB is clear, negative if set.
   Error returns:
   + Return NULL with the appropriate exception set if there's not
     enough memory to create the Python long.
*/
extern DL_IMPORT(PyObject *) _PyLong_FromByteArray(
        const unsigned char* bytes, size_t n,
        int little_endian, int is_signed);
"""


From terabaap at yumpee.org  Tue Feb 26 03:26:29 2002
From: terabaap at yumpee.org (Manoj Plakal)
Date: Tue, 26 Feb 2002 02:26:29 -0600
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au> <3C7ACEB4.B590E892@mediaone.net> <3C7AD777.7050401@skippinet.com.au>
Message-ID: <3C7B46B5.3060001@yumpee.org>

Mark Hammond wrote:
> Mark Barclay wrote:
>> Is there a way of coding a COM server that can be running before it is 
>> accessed
>> by a COM client? I'm not sure if this makes sense, but I'd like to 
>> have a program
>> running, that can be connected to, perhaps multiple times, without it 
>> having to
>> create multiple instances of teh server, just have multiple clients 
>> using it.
> 
> 
> IIRC, it gets a little messy - you need to implement a LocalServer 
> object, and add it to the running object table (ROT).  Searching old 
> posts at groups.google.com is probably your best bet.


          Take a look at the CVS tree of qasid.sourceforge.net.
          There's a small example of a wxPython GUI app that acts
          as a COM server. The COMHostServer function in qasidserver.py
          does the job of creating an instance of the server and
          registering it in the ROT.

          Manoj




From oren-py-l at hishome.net  Tue Feb  5 10:29:35 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Tue, 5 Feb 2002 10:29:35 -0500
Subject: Inherit Class Dynamicly
In-Reply-To: <a3ok79$bvr$1@pegasus.csx.cam.ac.uk>
References: <a3ok79$bvr$1@pegasus.csx.cam.ac.uk>
Message-ID: <20020205152935.GA24765@hishome.net>

On Tue, Feb 05, 2002 at 12:49:56PM +0000, Xiao-Qin Xia wrote:
> Hi, everybody,
> 
> from Tkinter import *
> class NewWidget:
>         ...
>         def __init__(self, widget_name = "Button", **dict):
>                 # What is needed here to make the new born object(self) 
>                 # behavior like both the class specified by widget_name
>                 # and NewWidget?
> 

def NewWidget(widget_name, **dict):

    class DynamicallyGenerated(FixedParentClass, widget_classes[widget_name]):
       ...

    return DynamicallyGenerated(**dict)


A function returning a dynamically generated class is quite similar to a
C++ template.  It's probably a good idea to memoize such dynamically 
generated classes - creating a new class for each widget is rather
expensive.

	Oren




From skip at pobox.com  Thu Feb  7 23:26:20 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 7 Feb 2002 22:26:20 -0600
Subject: 2001 Turing Award
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBMEHNKBAA.marklists@mceahern.com>
References: <2jnu3a.4sq.ln@wombat.san-francisco.ca.us>
        <NCBBLFCOHHDIKCAFGCFBMEHNKBAA.marklists@mceahern.com>
Message-ID: <15459.21356.832793.817605@12-248-41-177.client.attbi.com>

    >> Damn--I thought the award was for the person who best impersonated a
    >> Turing machine.  :)

    Mark> Do you have any nominees?

I operate about as slow as a true Turing machine at times.  Perhaps I'll
nominating myself if I can unjam this damn paper tape that runs around my
waist...

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From gimbo at ftech.net  Tue Feb 26 17:20:51 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 22:20:51 +0000
Subject: what I would like to see in python to make a better "glue" language (Newbie)
In-Reply-To: <a5g3g1.3vsg8p3.1@kserver.org>
References: <899f842.0202261325.6ff3dd54@posting.google.com> <a5g3g1.3vsg8p3.1@kserver.org>
Message-ID: <20020226222051.GA2250@andy.tynant.ftech.net>

On Tue, Feb 26, 2002 at 01:44:33PM -0800, Sheila King wrote:

> I have recently had to do the same thing (parsing an nslookup
> command) in order to write a script. I would REALLY appreciate a DNS
> module as part of the standard library. The socket library just
> didn't quite do everything I needed.

Have you seen Andy Dustman's adns module?  I've used this with some
success in the past.

http://dustman.net/andy/python/adns-python

BTW, I agree with most of the OP's points, except that I'm hungrier
for SNMP than he is.  ;-)

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From tim at vegeta.ath.cx  Wed Feb 13 06:52:18 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Wed, 13 Feb 2002 11:52:18 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org>
Message-ID: <slrna6kkpj.is0.tim@vegeta.ath.cx>

Gustavo Cordova <gcordova at hebmex.com> graced us by uttering:
> And also "agua" which seems female, but it's a male subject: "el agua".

IIRC, "agua" _is_ feminine, but takes the masculine article when
singular. Mi diccionario dice que "agua" es feminina, tambi?n.

OTOH, la palabra "mar" es masculina _y_ feminina. Por ejemplo:

    "el agua en la mar"     # Correct
    "las aguas del mar"     # Correct

HTH
Tim Hammerquist
-- 
It's, uh, pseudo code.  Yeah, that's the ticket...
[...]
And "unicode" is pseudo code for $encoding.  :-)
    -- Larry Wall in <199808071717.KAA12628 at wall.org>


From roy at panix.com  Sun Feb 17 13:06:02 2002
From: roy at panix.com (Roy Smith)
Date: Sun, 17 Feb 2002 13:06:02 -0500
Subject: Why Python is like BASIC ...  java "vs" python
References: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph> <mailman.1013826279.13649.python-list@python.org> <i4iu6ug73jr2lmpblj2e70e1tci043st0l@4ax.com> <YBRb8.80754$3J1.2551057@news2.tin.it>
Message-ID: <roy-FCF055.13060217022002@reader2.panix.com>

Alex Martelli <aleax at aleax.it> wrote:
> It seems to me that this is a very good example of why "the distinction
> between interpreted and compiled *languages*", far from being "important",
> is utter, total, and absolute bunk.

The picture can be further complicated by the fact that some things that 
you *think* are compiled are really being interpreted.  The PowerPC-based 
Macintoshes have a M68k compatability mode.  As far as I understand it, 
it's basicly an M68k virtual machine.  So, you take a C program, compile it 
into M68k machine code, and run it on a PPC.  Is it compiled, or is it just 
some virtual machine running bytecode?  Kind of hard to say for sure.

The invention of things like dynamic libraries blurs the line between 
"dynamic interpreted languages" and "static, compile-link-and-execute 
languages" just a little bit more.

Vaxes used to do the same thing with pdp-11 binaries, but that was really 
more of a hardware compatability mode than a virtual machine in the sense 
we're talking about here.  Of course, the vax was a microcoded machine, so 
in a very real sense, you *were* running a software interpreter.

I think it's much more important, from the user perspective, to think about 
whether a language is interactive or not.  With Python (or BASIC, or lisp, 
or shell, or ...) if you start it up and type something at it, it types 
something back at you and waits for you to type something else.  Perl, for 
example, doesn't work that way (at least not with the implementations I've 
seen).  With Perl, if I want to "ask the computer", I can certainly start 
up a perl interpreter and type stuff at it, but I need to hit control-D to 
make it run it, then I'm back in the shell.

With Python, I can fire up he language, type a bunch of stuff at it, and 
play around until I get what I need (or figure out what I wasn't sure of).  
I do this many times a day when coding; it's easier than looking something 
up!  The fact that my python has readline support in it just makes it that 
much easier, but the fundamental good thing is that when it evaluates what 
I just typed at it, it leaves me in the system with everything I've done so 
far still there.  In this respect, it's very much like Basic (or lisp).


From syring at email.com  Wed Feb 20 01:45:34 2002
From: syring at email.com (Karl M. Syring)
Date: Wed, 20 Feb 2002 07:45:34 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1>
Message-ID: <a4vgnh$3eoao$1@ID-7529.news.dfncis.de>

"Carl Banks" <idot at vt.edu> schrieb>
<snip>
> Flight simulation: The earth's radius is about 20 million feet (about
> 7 million meters for my SI using friends), and the simulator must
> resolve, at a minimum, distances of about 0.1 feet.  That's a
> difference of 9 decimal places.  We then need several more places for
> numerical robustness.
<snip>

You surely are joking, Mr. Banks?
This is exactly the kind of misuse of floating point that was referred to in
a previous posting. Every computational geometry software would use a
rational data type here, although you might get away with simple point data
binning in this case.

Karl M. Syring




From l0819m0v0smfm001 at sneakemail.com  Sun Feb  3 00:55:38 2002
From: l0819m0v0smfm001 at sneakemail.com (Joshua Macy)
Date: Sun, 03 Feb 2002 05:55:38 GMT
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com>
Message-ID: <3C5C973E.6080607@sneakemail.com>

Paul Rubin wrote:

> 
> I just tried that and it throws an error:
> 
>     $ python
>     Python 1.5.2 (#1, Mar  3 2001, 01:35:43)  [GCC 2.96 20000731 (Red Hat Linux 7.1 2 on linux-i386
>     Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>     >>> from __future__ import division
>     Traceback (innermost last):
>       File "<stdin>", line 1, in ?
>     ImportError: No module named __future__
>     >>> 
> 
> That's with the version of Python currently being shipped with Red Hat
> Linux, so any scripts I distribute with that future import will crash
> for a lot of people.
> 



try:
     from __future__ import division
except ImportError:
     pass

Of course, the semantics of division will then vary depending upon 
whether the type is integer...but that's the problem that future 
division is trying to solve, so I don't see any way around that.

Joshua





From jeff at ccvcorp.com  Fri Feb 15 15:51:31 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 15 Feb 2002 12:51:31 -0800
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
Message-ID: <3C6D74D2.1CBD6AC6@ccvcorp.com>


Artur Skura wrote:

> In other words: I have to put text entries between text labels.
> Now can you suggest an easy, lazy way of doing this (with any
> toolkit) so things don't pile up and everything looks nice?

Using wxPython, you can set the text of a wxStaticText control to be
whatever you want:

    MyLabel = wxStaticText(parent, wxNewId(), "Starting Text")

    ....

    MyLabel.SetLabel("New, expanded text!")

wxPython also has sizers, which will automatically resize windows and
controls, so that you can change your text, have the sizer recalculate
the layout, and refresh your dialog.  (It can take a bit of fiddling to
get sizers to work the way you want, but once you get the hang of them,
they're *very* helpful!)  If you want to insert specific text into a
'template' sentence, then that's easy to do using Python's string
interpolation:

   MyLabel.SetLabel("Now serving %s!" % mydata)

And of course, you can dynamically add and remove controls from dialogs
and sizers, so you can easily change your dialog on the fly.

Jeff Shannon
Technician/Programmer
Credit International




From mwh at python.net  Tue Feb 26 08:20:27 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 26 Feb 2002 13:20:27 GMT
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com>
Message-ID: <usn7oxtpd.fsf@python.net>

amrit040 at yahoo.com (Amrit) writes:

> Hi 
> 
> I'm trying to write a C function callable from python. I'm getting
> some strange results :
> 
> Here's a typical run:
> """
> Python 2.0 (#2, Jun 18 2001, 15:29:44) 
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>> import lt
> >>> lt.__file__
> 'lt.so'
> >>> dir(lt)
> ['__doc__', '__file__', '__name__', 'fmm']
> >>> lt.fmm(range(15),-41)
> (0.0, -1.0)
> >>> lt.fmm(range(15),-41)
> TypeError: function requires at least one argument
> """
> 
> On calling the same function again with the same argument gives me a
> type error. This alternates. I get a typeerror and then a correct
> output and so on. Any ideas what could be wrong. My C code is pasted
> below

This sort of thing usually means an exception happened and you didn't
return NULL.

>   for (i = 0; i < n; i++)
>     {
>       obj[i] = PySequence_GetItem (vec, i+1);

This looks suspicious.

Cheers,
M.

-- 
  You have run into the classic Dmachine problem: your machine has
  become occupied by a malevolent spirit.  Replacing hardware or
  software will not fix this - you need an exorcist. 
                                       -- Tim Bradshaw, comp.lang.lisp


From syring at email.com  Sun Feb  3 05:16:52 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 11:16:52 +0100
Subject: Defending the Python lanuage...
References: <mailman.1012432610.4390.python-list@python.org> <83vgdfxe1j.fsf@panacea.canonical.org>
Message-ID: <a3j2mi$18b85b$1@ID-7529.news.dfncis.de>

"Kragen Sitaker" <kragen at pobox.com> schrieb
> "Tim Peters" <tim.one at home.com> writes:
> > Open source *is* like that, to the extent that free support is
> > relatively easy to get for as long as people like it enough to work on
it,
> > and even after that support for hire remains a possibility forever.
Would
> > that it were so for my copy of, e.g., Macsyma.
>
> One version of Macsyma is now open source:
> http://www.ma.utexas.edu/maxima.html

It is at http://maxima.sourceforge.net/, because it's original maintainer
died.

>
> > Software bascially sucks, and you've got to come to terms with that.
The
> > reality is that no software you write is going to have a useful lifetime
> > spanning even a decade without continual rewriting.
>
> I sure wish some parts of TeX, Emacs, Xlib, and CMUCL had been
> rewritten sometime in the last decade.  But they haven't, and I still
> manage to use them somehow...

There is a rewite of TeX called NTS (http://nts.tug.org/) written in, uhm,
Java. With the native compilers galore, it may get a breeze of new life.

Karl M. Syring




From idot at vt.edu  Sun Feb 17 16:54:23 2002
From: idot at vt.edu (Carl Banks)
Date: Sun, 17 Feb 2002 16:54:23 -0500
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com>
Message-ID: <fq8p4a.ll.ln@127.0.0.1>

Tim Roberts wrote:
> And the original poster should understand that this is true because such
> systems are of little practical use.  There are very, very few physical
> processes where floating point is useful AND precision of more than 6
> significant digits is really required.  Large integers can be useful (and
> Python supports them), but a requirement for high precision floats is
> usually a sign that the requirer does not understand his problem space.


Incredibly untrue.

Many physical and nonphysical systems require high precision, usually
because they involve calculating small differences between large
numbers.  Some examples off the top of my head:

Fractals: deep zooming.

Flight simulation: The earth's radius is about 20 million feet (about
7 million meters for my SI using friends), and the simulator must
resolve, at a minimum, distances of about 0.1 feet.  That's a
difference of 9 decimal places.  We then need several more places for
numerical robustness.

Optimization: Many optimization algorithms, especially in linear
programming, have numerical problems with single precision.

Differential equation solving: Certain differential equations have an
undesirable property called stiffness.  This happens a lot in fluid
mechanics.  High precision and other tricks are used to solve such
problems.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From fredrik at pythonware.com  Sun Feb 10 07:52:53 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 10 Feb 2002 12:52:53 GMT
Subject: Python + windows + problems with make.py
References: <mailman.1013342055.30975.python-list@python.org>
Message-ID: <F2u98.17443$n4.2988116@newsc.telia.net>

"gombosk at vas.birosag.hu" wrote:
> 3. I've got the next:
>
> reading exceptions from
> Traceback (most recent call last):
>   File "./Make.py", line 133, in ?
>     main()
>   File "./Make.py", line 73, in main
>     print "reading exceptions from", exceptions.__file__
> AttributeError: 'exceptions' module has no attribute '__file__'
>
> Yes, exceptions module has no attribue '__file__'
>
> What can I do

remove the print statement?

blame it all on me and barry?

:::

in recent versions of Python, exceptions is a built-in module
(on all platforms).

if you really need to know where the standard library is, check
the __file__ attribute of some other module (e.g. os or string),
use sys.prefix, or scan the sys.path list.

hope this helps!

</F>

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




From caljason76 at yahoo.com  Thu Feb 21 15:35:25 2002
From: caljason76 at yahoo.com (Jason)
Date: 21 Feb 2002 12:35:25 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <7xg03vsidi.fsf@ruckus.brouhaha.com>
Message-ID: <e9e6dd1f.0202211235.4e7be9aa@posting.google.com>

Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7xg03vsidi.fsf at ruckus.brouhaha.com>...

>   b = list(a)
>   b.sort()
> 

perfect. thanks

> In Python, that's considered immoral--it puts you in danger of
> forgetting that the sort operation modifies the list that you use it
> on.  For example, if you say
> 
>    a = b.sort()
> 
> expecting to get a sorted copy of b without clobbering b, you'd get
> a rude surprise.

I still think that the convenience of having sort() return the list
far outweighs the cost.  People would only get bit once, maybe twice,
before they learned the behavior of sort.

I like Python, but the biggest problem I have with the language right
now is what low code density it has at times.

-j


From duncan at NOSPAMrcp.co.uk  Tue Feb 12 04:07:04 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Tue, 12 Feb 2002 09:07:04 +0000 (UTC)
Subject: just a moron trying to get python working on win98 and personal web  server
References: <3C685DDE.AA8AFD10@kc.rr.com>
Message-ID: <Xns91B35B101A743duncanrcpcouk@127.0.0.1>

james <jdaily at kc.rr.com> wrote in news:3C685DDE.AA8AFD10 at kc.rr.com:

> is there an apache alternative.
> 
There is quite a reasonable build of Apache at www.opensa.org which runs on 
Win95, 98, NT and 2000.

-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From oren-py-l at hishome.net  Sat Feb  2 02:43:34 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Sat, 2 Feb 2002 09:43:34 +0200
Subject: Enhanced Generators - two-way generators
In-Reply-To: <004101c1ab5b$aa2e01c0$2df6a4d8@othello>; from othello@javanet.com on Fri, Feb 01, 2002 at 03:04:31PM -0500
References: <a3at9p$2he$1@bob.news.rcn.net> <20020201144050.GA28252@hishome.net> <004101c1ab5b$aa2e01c0$2df6a4d8@othello>
Message-ID: <20020202094334.A6843@hishome.net>

On Fri, Feb 01, 2002 at 03:04:31PM -0500, Raymond Hettinger wrote:
> > I'm in favor of one-way consumer functions to mirror generator functions,
> > but not two-way generators.
> 
> Why deny yourself the ability to pass data along with control flow.  I don't
> really change anything, control flow stays the same.  The proposal just
> gives a way to pass some data along the way.  If you HAVE TO send some
> data in, the existing alternative using global variables truly sucks
> (error prone, hard to read, unclear in its intention).  If you don't HAVE
> TO, then you've lost nothing as the argument to .next() and yield assignment
> are optional.

Try to think about how to describe this object in a few short sentences
and the problems become more apparent.  
  
Here's a description for the usage of iterators and generators:

   "An iterator has a next() method returning the next item; raises 
    StopIteration if no more data is available."

   "Calling a generator function returns a generator object implementing
    the iterator protocol."

Here's a description for the usage of sinks and consumers:

   "A sink has a submit() method to which one argument is passed; this 
    object is sent to the sink's destination for processing.".

   "Calling a consumer function returns a consumer object implementing
    the sink protocol."

Both of these descriptions are complete, self-contained descriptions of a
stream of objects.  It is easy to convert one to the other, connect streams
together, perform arbitrary transformations on streams, etc.

And a description for the usage of your extended iterators and generators:

   "An extended iterator has a next() method that may or may not take an
    argument. The return value of the next() method may or may not contain 
    relevant data."
 
   "Calling a generator function returns an object implementing the extended
    iterator protocol. This object may either generate data, consume data 
    or implement some two-way dialogue."

Yes, there is such a thing as too much flexibility.  This protocol is not
self-contained and requires additional agreements just to get the data
across.  

This year is the 30th anniversary of Unix pipes, invented by Doug McIlroy.  
I think this is enough experience to build on.  In practice, pipe 
programming is used for pipelines of processing stages, not for dialogue. 
Dialogues are used for terminals with human operators, but are very rare as
a building block inside a program.  They are mostly used when a tool 
designed for human interaction is used in an automated manner, not as a
first choice.

	Oren




From eppstein at ics.uci.edu  Wed Feb  6 12:44:03 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 06 Feb 2002 09:44:03 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de>
Message-ID: <eppstein-A86D20.09440306022002@news.service.uci.edu>

In article <a3rbad$19u08k$1 at ID-98166.news.dfncis.de>, "MDK" <mdk at mdk.com> 
wrote:

> For example:
> 
> ("cat",5,['dog',[3,3,1]],"zoo")
> 
> Would become:
> 
> ('c','a','t',5,'d','o','g',3,3,1,'z','o','o')

This is standard in LISP as "flatten". Here's an implementation using 
simple generators.

from __future__ import generators

def flatten(x):
        def flatgen(x):
                try:
                        xg = iter(x)
                        for y in xg:
                                if y == x:
                                        # prevent infinite recursion
                                        # when x is a character
                                        yield x
                                        return
                                yg = flatgen(y)
                                for z in yg:
                                        yield z
                except TypeError:
                        yield x
        
        return tuple(flatgen(x))

print flatten(("cat",5,['dog',[3,3,1]],"zoo"))
('c', 'a', 't', 5, 'd', 'o', 'g', 3, 3, 1, 'z', 'o', 'o')
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From aahz at panix.com  Thu Feb 28 08:58:46 2002
From: aahz at panix.com (Aahz Maruch)
Date: 28 Feb 2002 05:58:46 -0800
Subject: Killing threads
References: <1014894766.777152@seux119>
Message-ID: <a5ld2m$554$1@panix2.panix.com>

In article <1014894766.777152 at seux119>, VUG <plugin at airtel.net> wrote:
>
>Hi all. I'm doing an application in C who needs to execute simultaneous
>Pythons scripts. Under certain conditions I need to kill one threads. ?How
>can I do it? 

You can't.  You'll need to figure out a mechanism to poll for requests
to kill the thread.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From anton at vredegoor.doge.nl  Thu Feb  7 17:49:33 2002
From: anton at vredegoor.doge.nl (Anton Vredegoor)
Date: Thu, 07 Feb 2002 23:49:33 +0100
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org> <a3uukp$5dq$1@news.hccnet.nl>
Message-ID: <a3v0a2$6uh$1@news.hccnet.nl>

On Thu, 07 Feb 2002 23:17:31 +0100, anton at vredegoor.doge.nl (Anton
Vredegoor) wrote:

>Now what is it? Its a permutation engine for multisets. Read more
>about it at:

Sorry, this should be:

>
>http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.py
>
>or (html version):
>
>http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.py






From phr-n2002a at nightsong.com  Wed Feb  6 08:14:58 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 06 Feb 2002 05:14:58 -0800
Subject: Web Hosting - one report
References: <283adf56.0202060503.6556b483@posting.google.com>
Message-ID: <7x7kpqoip9.fsf@ruckus.brouhaha.com>

kdahlhaus at yahoo.com (Kevin Dahlhausen) writes:
> I thought I had found a place at 'Hostdepartment.com'  $5/month for
> 100Mb and a full shell. They have 1.6 installed.  You can compile your
> own python, but many modules do not build.  It turns out that they
> have disabled dynamic loading of modules from user accounts.  So if
> you have a swigged extension, python can load any shadow files, but
> the .so is never found.  That's a shame, as I was hoping to actually
> post an *answer* to this reoccuring question.  Personally, I need to
> be able to use extension modules - at least PIL, GD charts, etc.

If you can compile your own python, can't you statically link your
extension into it?


From eppstein at ics.uci.edu  Sun Feb 10 21:47:50 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Sun, 10 Feb 2002 18:47:50 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <eppstein-A87A6C.14430710022002@news.service.uci.edu> <a46uqm$gc7$1@peabody.colorado.edu>
Message-ID: <eppstein-D7569F.18475010022002@news.service.uci.edu>

In article <a46uqm$gc7$1 at peabody.colorado.edu>,
 Fernando P?rez <fperez528 at yahoo.com> wrote:

> Indeed, it's just the Spanish form of the 'ist' English suffix (pianist, 
> violinist, etc.) It simply means 'who performs an activity'.
> 
> Its use is more common in Spanish than in English, though (pianista, 
> violinista, but also futbolista, tenista, equilibrista, ...). Interestingly, 
> even though in Spanish the -a ending typically denotes a feminine noun, the 
> -ista ending applies to both male and female practitioners.

Another English-language example, seen in today's Los Angeles Times 
Magazine: fashionista.  Google reports about 4500 English examples...
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From emile at fenx.com  Fri Feb  1 13:17:12 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 10:17:12 -0800
Subject: Conventions for iterators
References: <mailman.1012555082.6918.python-list@python.org> <U2A68.17631$3J1.540244@news2.tin.it>
Message-ID: <a3emcv$17igt2$1@ID-11957.news.dfncis.de>

"Alex Martelli" <aleax at aleax.it> wrote in message
news:U2A68.17631$3J1.540244 at news2.tin.it...
> _timbot_ in a typo -- dies albo signanda lapillo...!-)].

Translation please.  You know we can only read dutch here... ;-)


--

Emile van Sebille
emile at fenx.com

---------



From jeff at ccvcorp.com  Thu Feb 28 15:40:20 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 12:40:20 -0800
Subject: wxPython blocks in thread?
References: <rk63czmu79v.fsf@pegase.inria.fr>
Message-ID: <3C7E95B4.60E828BD@ccvcorp.com>


Arnaud de La Fortelle wrote:

> I am a beginner with wxPython and tried to embed a GUI within a
> thread.  So I came with a simple implementation, but it does not work.
> I do not understand why the following application blocks. What the
> matter?

I seem to remember some discussion on the wxPython-users mailing list,
recently, that indicated that wxPython considers the "main thread" to be
the one that first imports wxPython.wx.  You're running your app in a
thread that has never imported wxPython -- the Python interpreter will
accept it, but the underlying wxWindows framework may be having problems.
(I may be misremembering the details on this, though -- you can check the
recent archives of the mailing list from www.wxpython.org.)

Can you invert your thread usage?  Instead of having some other program
fire off a thread that does the GUI, have the GUI part be the main thread
and then *it* fires off various worker threads.

If not, then you might be able to simply wait until you're *in* your
thread, to import wxPython and declare your wxApp, like so:

class GUIThread(threading.Thread):
    def Run(self):
        from wxPython.wx import *
        class MyApp(wxApp):
            def OnInit(self):
                ....
        self.app = MyApp(0)
        self.app.MainLoop()

Jeff Shannon
Technician/Programmer
Credit International



From sjmachin at lexicon.net  Sun Feb 10 21:14:20 2002
From: sjmachin at lexicon.net (John Machin)
Date: 10 Feb 2002 18:14:20 -0800
Subject: No explanation for weird behavior in re module!
References: <a470re$1cl30l$1@ID-78052.news.dfncis.de>
Message-ID: <c76ff6fc.0202101814.5d18209d@posting.google.com>

synthespian <synthespian at uol.com.br> wrote in message news:<a470re$1cl30l$1 at ID-78052.news.dfncis.de>...
> Hi-
> 
> 	I'm really intrigued by this behavior:
> 
> >>> import re
> >>> p = re.compile('^(der|die|das(\s\w+))')
> >>> m = p.match('die T?r, T?ren')
> >>> n = p.match('das Auto, Autos')
> >>> m.group(0)
>  'die'
> >>> m.group(1)
>  'die'
> >>> m.group(2)
>  [nothing!!!!]
> >>> n.group(0)
>  'das Auto'
> >>> n.group(1)
>  'das Auto'
> >>> n.group(2)
> 'Auto'
> 
> 	I'm using Python2.0 on a Debian potato system. 
> 	Why didn't m.group(2) produce 'T?r' as the output???
> 	Python2.0 is supposed to have Unicode support buil-in the re module right?
> 	Other than the fact that 'T?r' has the '?' unicode charcater, I fail to see any difference!

Your pattern is asking to match one of the following
(a) 'der'
(b) 'die'
(c) 'das' followed by a space then one or more alphabetic characters.
In other words, nothing to do with Unicode, and all to do with
operator precedence. You need to toss some parentheses in there.
Forget the grammar and try your pattern on 'die Auto, Autos'; you'll
see that this matches just 'die' also.

Anticipating the next raft of problems: (1) You will need to use the
re.UNICODE flag when you call re.compile(), otherwise \w will not
recognise the Unicode alphabetics (this *is* documented) (2) You may
need to give it an input whose Python type is 'unicode' -- being able
to see the umlaut on your screen is not sufficient evidence of this
:-) (3) You should get into the habit of using the raw string notation
with your regexes whether it is necessary or not, else you will be
bitten in the future.

Anyhow, the following works for me:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>>> p = re.compile(r'^((der|die|das)(\s\w+))',re.UNICODE)
>>> p.match('das Auto').groups()
('das Auto', 'das', ' Auto')
>>> z = u'die T\u00FCr, T\u00FCren'
>>> p.match(z).groups()
(u'die T\xfcr', u'die', u' T\xfcr')


From hgg9140 at seanet.com  Sun Feb  3 11:00:30 2002
From: hgg9140 at seanet.com (Harry George)
Date: 03 Feb 2002 08:00:30 -0800
Subject: Xmi parser
References: <3c5bb698$0$4609$626a54ce@news.free.fr> <m31yg49lyn.fsf@mira.informatik.hu-berlin.de> <3c5c49b3$0$4617$626a54ce@news.free.fr>
Message-ID: <m3r8o2r1wh.fsf@wilma.localdomain>

"Bruno Li?nard" <lienard.bruno at free.fr> writes:

> Yes I know, but given the complexity of the DTD and the number of tags and
> attributes, I think it is quite a lot of work, so if anybody had already
> done it... I need a parser that can read a XMI file and create the
> corresponding objects.  I have tried with xml_objectify, but it doesn't
> work. I didn't spend a lot of time to understand why..
> 
> Regards,
> 
> Bruno
> 
> 

You can check out pycase at:
   http://www.seanet.com/~hgg9140/comp/index.html

In the package is a copy of the UMLXMI DTD, and a program which
converts that to python classes which know how to write themselves out
to xml.  The program (xmidtd2py.py) could be extended to read xml as
well.  

This is very alpha -- it was good enough for the uses I have, but no
claims of completeness or cor4rectness, etc.

> 
> 
> "Martin v. Loewis" <martin at v.loewis.de> a ?crit dans le message news:
> m31yg49lyn.fsf at mira.informatik.hu-berlin.de...
> > "Bruno Li?nard" <lienard.bruno at free.fr> writes:
> >
> > > I am looking for a xmi parser. Does anybody here have already
> > > written one ?
> >
> > Given that XMI is an XML application, any XML parser should be able to
> > parse XMI. I recommend xml.parsers.expat (or the SAX and DOM wrappers
> > around it).
> >
> > Regards,
> > Martin
> >
> 
> 

-- 
Harry George
hgg9140 at seanet.com


From pinard at iro.umontreal.ca  Sun Feb 17 19:24:39 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 17 Feb 2002 19:24:39 -0500
Subject: high precision mathematics
In-Reply-To: <fq8p4a.ll.ln@127.0.0.1>
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu>
	<gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1>
Message-ID: <oqbsenwsag.fsf@titan.progiciels-bpi.ca>

[Carl Banks]

> Tim Roberts wrote:

> > [...] a requirement for high precision floats is usually a sign that
> > the requirer does not understand his problem space.

> Incredibly untrue.

Science people often need more precision.  DOUBLE existed in early FORTRAN.
C has `float' and `double' as well.  It would be rather surprising if all
involved people were that dumb! :-)

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



From threeseas at earthlink.net  Fri Feb 15 20:45:47 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:47 GMT
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org>
Message-ID: <4296.811T2599T12485989threeseas@earthlink.net>

On 11-Feb-02 19:37:51 Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
>----- Original Message -----
>From: "Jeff Shannon" <jeff at ccvcorp.com>
>>
>> Timothy Rue wrote:
>>
>> > But then group voting is like a conspiracy huh?
>>
>> Only for nonstandard definitions of 'conspiracy'.
>>
>> Conspiracy implies secrecy, and imposing the will of the few upon the
>> many. Voting is nonsecret (even if individual votes are secret, the
>> process is not) and implies determining the will of the many (at the
>> possible expense of the will of the few).

<rollseyes>

There are plenty more users and consumers of computers and computer
products than there is developers of such products. And it's clear from
the case against the richest man in the world by the most powerful
government in the world where the computer industry stands in matters of
ethics regarding the consumers/user. There is plenty being withheld from
the consumers, including freedom to create what they need and to do so
without abuses of the over complexity generated by the computer industry,
who continuely asks the consumer "where do you want to go today?" because
those in the computer industry simply are to disconnected from the
consumer/user base to know and lacking in ethics plenty enough to simply
give the consumer/user the tools to go where they want..

>Yes.  I apologized (to the list, not to Rue) because the suggestion is,
>well, non-Pythonic.  Tacky.  And I was/am sorry.  Jeff has it right here,
>though, that it was certainly no conspiracy; and it is the USENET way to
>publicly handle such things.

Oh really? It seems to me that the way to handle something you do not like
is to show as much outspoken disrespect for it as you can while holding
onto the illusion that you are justified in bad mouthing something. So
much for the mothers saying of "if you don't have something nice to say,
don't say anything".....

>> Then again, we shouldn't be surprised at you using nonstandard
>>definitions of words.

Non-standard is this week defined to be the same as it was last week, a
matter of convience...... How many ways can you label a "terminal"?
And even better yet, what is the computer industry calling a "terminal"
this week?

What of the double speak computer industry dictionary? The one that has
"we" in it for talking to the in a manner of them preceiving it as
inclusive of them but for those using it it really means not inclusive f
the public. And other such double meaning ........ standard
definitions.....?????

>Indeed.  He uses a nonstandard definition of "language" and applies
>"physics" to programming.  He can't program, you know; ten years on this
>project and nothing to show for it.  If I had tried to develop something
>like this for ten years and had nothing to show for it but a website,
>I'd give up.

According to the computer industry the only standard definition is the one
they are using this week but it will be different next week because that
is the only way to keep the consumer/users "stupid" and submissive.

There is a physics to abstraction manipulation, including such abstraction
manipulation as the double speak dictionary and programming abstractions
that boils down to bit manipulation of physical switches.

I have no doubt that those who have made a living off of such
manipulations of bits, abstraction and consumers certainly don't want to
be exposed for the wrongful manipulators that are.

And I suppose that explains you, too.


Clearly there is a priority here where the coding is not on top of the
list but rather ...... well what do you suppose Neo ment in that last
line of the Movie, The Matrix?

10 years is alot shorter than the 300 or so years it took the hindu-arabic
decimal system to propogate to the majority of the civilized population!!!

Besides there is enough to show for the time.

You don't understand the problems because you refuse to look at yourself.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From drew_csillag at yahoo.com  Tue Feb 19 09:07:43 2002
From: drew_csillag at yahoo.com (Drew Csillag)
Date: Tue, 19 Feb 2002 09:07:43 -0500
Subject: Adding static typing to Python
In-Reply-To: <20020219095832.GE2611@lilith.hqd-internal>; from gh_pythonlist@gmx.de on Tue, Feb 19, 2002 at 10:58:33AM +0100
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com> <mailman.1014085471.28317.python-list@python.org> <slrna73h0o.18i.grante@tuxtop.visi.com> <20020219095832.GE2611@lilith.hqd-internal>
Message-ID: <20020219090743.A1157@localhost.localdomain>

On Tue, Feb 19, 2002 at 10:58:33AM +0100, Gerhard H?ring wrote:
> Le 19/02/02 ? 03:18, Grant Edwards ?crivit:
> > In article <mailman.1014085471.28317.python-list at python.org>, Chris Gonnerman wrote:
> > 
> > > Ultimately, adding static typing to Python results in a different
> > > language; it's not Python anymore.
> > > 
> > > If you don't like Python because the typing isn't strong, use Ada,
> > > or C++, or Pascal... 
> > 
> > You might also want to take a look at Modula-3.  Anybody used
> > to Python will recognize many of the language features.
> 
> Modula-3 also looks great, as did Sather to mee. But both seem seem to
> have no compilers that are actively maintained.
> 
> So I'm still leaning towards Ada 95 and C++ for statically typed
> languages.

If you want really statically typed, you probably want to take a look
at Eiffel.

Drew



From Bill.Scherer at VerizonWireless.com  Wed Feb  6 07:16:47 2002
From: Bill.Scherer at VerizonWireless.com (Scherer, Bill)
Date: Wed, 6 Feb 2002 07:16:47 -0500 (EST)
Subject: Is anyone capable of explaining ?
In-Reply-To: <3C6069DA.18895.92952C@localhost>
Message-ID: <Pine.LNX.4.33.0202060712170.12958-100000@bagend.vzwcorp.com>

On Tue, 5 Feb 2002, A wrote:

> 
> Can anybody please answer the following question?
> 
> Question: How does web browser know what cookies should be 
> sent to server as a response?

The browser determines which cookies to include with a request by 
comparing the domain of the server it is about to contact with 
the domain of the cookies.   I didn't see any domain in the 
cookies you are setting; there should be one.

The template http cookie header:

Set-Cookie: name=value; expires=date; path=pathname; domain=domainname

See:
http://www.webreference.com/js/column8/http.html


-- 
Bill.Scherer at Verizon Wireless
RHCE 807101044903581





From aahz at panix.com  Wed Feb 27 00:32:18 2002
From: aahz at panix.com (Aahz Maruch)
Date: 26 Feb 2002 21:32:18 -0800
Subject: How to have multi-threaded server log to one file ?
References: <97ae44ee.0202261855.15c97708@posting.google.com>
Message-ID: <a5hr12$c24$1@panix2.panix.com>

In article <97ae44ee.0202261855.15c97708 at posting.google.com>,
Stephen <shriek at gmx.co.uk> wrote:
>
>When the server is run, each request from a client is handled by
>MyHandler() in a separate thread. If multiple requests are being run
>at the same time, how should I write my code so that they cal all
>write to the same log file without conflict ?  Should I use a global
>variable for the logfile, as below ~

Create a new thread for writing to the logfile and use Queue.Queue to
pass data to the writer thread.  See
http://starship.python.net/crew/aahz/
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From threeseas at earthlink.net  Fri Feb 15 20:45:44 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:44 GMT
Subject: Fear of autocoding!
References: <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <a49s16$5ao$1@newshost.accu.uu.nl>
Message-ID: <1195.811T1341T12074615threeseas@earthlink.net>

On 11-Feb-02 20:44:06 Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>phil hunt <philh at comuno.freeserve.co.uk> wrote:
>> Either that or we're part of the Evil Programmers Conspiracy to make
>> programming hard.

>I'm *not* a member of the PSU. The Python Secret Underground does not
>exist. I have no idea why people would think it does. Probably because
>they're all incompetent. Yeah..

>There is no conspiracy. Even if such a conspiracy existed, I could tell you
>that there is no reason to

Right!!!!

one person conspiracy is an oxymoron.

Just because you have many that have the same alterior motive doesn't mean
the alterior motive is a conspiracy.

Autocoding is something a given group of programmers (psuedo programmers
- re-inventors - code assemblers) would likely preceive as something that
can replace them.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From anthony_barker at hotmail.com  Tue Feb 26 16:25:15 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 26 Feb 2002 13:25:15 -0800
Subject: what I would like to see in python to make a better "glue" language (Newbie)
Message-ID: <899f842.0202261325.6ff3dd54@posting.google.com>

I had heard python referred to as a good glue language and decided to
try it out. So far, I think it is excellent, but as a newbie there
were a couple of things I found that can be improved. Note - this is
primarily from a System Admin or company internal development
perspective.

Very Important:
1. LDAP module should be included in the base distro. Active Directory
is already available via com (nice)

Important:
2. DNS really should be included in the base library, I emailed
Anthony Baxter and he replied, saying it was almost done. It is a
hassle to parse the dig (nslookup) output as I ended up doing, as it
is not portable.

3. Telnet Library is great - including an SSH module would be a nice
addition.

Less important:
4. Include an SNMP implementation, there are several around, but a
pure python one as a library would be nice

5. RRDTOOL - a nice wrapper would be nice ( there may be one already)

Already good (as far as I can tell)
XML Support
DBI Support
Windows COM support
TCPIP Services: SMTP.. etc
Mime/zip etc

One more idea for Administrators/Glue Implementers; lots of code
examples is key. As there is no CPAN to download sample scripts, the
library should have more examples of use. Admins don't have much time
to figure libraries out. The cookbook is a nice idea (from
ActiveState) and I picked up the /efbot's book but really I think that
some of the samples should be included in the library. I came to
Python from using some IBM tools. IBM includes examples for every
module/class and every method- which is nice as you can often copy and
paste.

Thanks,

Anthony


From jlh at cox.net  Fri Feb 15 19:17:37 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Sat, 16 Feb 2002 00:17:37 GMT
Subject: Paper "PDF Editing and Processing" at IPC 10?
References: <gherman-822C04.14540615022002@news.t-online.com>
Message-ID: <Byhb8.10471$SJ3.374979@news1.east.cox.net>

I was at the presentation.  It was a system that parsed pdf files in to
pythonic objects and structures and allowed for editing and saving existing
pdf docs.  It was written completely in python.  The down side is that the
presenter had no code to show or share and seemed quite taken aback when
requests were made.

He said in a "few months" after the code had been "cleaned up."  So, it
doesn't appear as though they will be releasing it.   Too bad, it was quite
interesting.

-Jeff

"Dinu Gherman" <gherman at europemail.com> wrote in message
news:gherman-822C04.14540615022002 at news.t-online.com...
> Hi!
>
> At IPC 10 there was the following presentation/paper:
>
>   Jyh-Yuan Deng and Shang-Kang Deng
>   Using Python as PDF Editing and Processing Framework
>
> that I'm interested in. I read the paper, but unfortu-
> nately the thing seems to have no name, the author
> doesn't reply to emails and the website and/or company
> of one of the authors, www.FlashMix.com, seems not to
> exist.
>
> Hence, I wonder, is this vaporware or has somebody seen
> it in action or even used it?
>
> Regards,
>
> Dinu
>
> .....................................................
> "If the Nuremberg laws were applied today, then every
> Post-War American president would have to be hanged."
>                                        (Noam Chomsky)




From tim.one at home.com  Sun Feb 10 18:05:01 2002
From: tim.one at home.com (Tim Peters)
Date: Sun, 10 Feb 2002 18:05:01 -0500
Subject: Bug or feature? 'abc'.split('') rejects empty separator
In-Reply-To: <20020210145042.A25774@glacier.arctrix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEAINLAA.tim.one@home.com>

[Bengt Richter]
>  >>> 'abc'.split('')
>  Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
>  ValueError: empty separator
>
> Wouldn't it make sense to return list('abc') ?

[Neil Schemenauer]
> It would also make sense to return ['a', 'b', 'c'].

Well, that's what list('abc') does return, so you're in violent agreement.
However, this still stands:

> Since it's not obvious what you want Python raises an error.

Indeed, my first thought was "OK, if it has to return *something*, then
since we're asking it to split on nothing, it shouldn't split at all:

    ['abc']

is what it should return."  That's also compatible with

    ''.join[['abc']] == 'abc'

People who think '' should split "everywhere" instead of "nowhere" should
really be arguing for 'abc'.split('') to return

    ['', 'a', 'b', 'c', '']

instead, since in any sense that '' could be said to "match", it matches at
4 slice positions in 'abc', not 2.




From plugin at airtel.net  Thu Feb 28 06:11:48 2002
From: plugin at airtel.net (VUG)
Date: Thu, 28 Feb 2002 12:11:48 +0100
Subject: Killing threads
Message-ID: <1014894766.777152@seux119>

Hi all. I'm doing an application in C who needs to execute simultaneous
Pythons scripts. Under certain conditions I need to kill one threads. ?How
can I do it? For example, I have the following module:

------------------------- module.py
def func1():
    code......

def func2():
    code......

def func3():
    code......
-------------------------
Then, from de application in C I load the previous module and run one of the
functions as a thread, like this:

// Load the module
Py_Initialize();
PyRun_SimpleString("import thread");
PyRun_SimpleString("from module import *");
// Launch some threads
PyRun_SimpleString("thread.start_new_thead(func1,args)");
....
PyRun_SimpleString("thread.start_new_thead(func2,args)");


So.... How can I from the C code KILL one of the threads I have started?. I
know that I can use the exit() function from the thread but I need the user
who create these functions doesn't have to put this kind of code inside
their functions.

Thanks in advance





From nhodgson at bigpond.net.au  Sun Feb 17 17:09:12 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Sun, 17 Feb 2002 22:09:12 GMT
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com>
Message-ID: <cSVb8.34653$Zu6.153988@news-server.bigpond.net.au>

Paul Rubin:

> I'm wondering how difficult it would be to add an editable (mutable)
> string class to Python 2.2 implemented in C under the new type/class
> unification scheme.  Right now, writing a text editor in Python seems
> difficult to do in any natural way.
> ...
> The usual method of representing the editable string as
> two contiguous regions of a chunk of memory separated by a gap, so you
> can add or delete characters by growing or shrinking the gap, and
> moving the gap as necessary, would be fine.

   I have considered creating a Python split buffer extension based upon the
classes I wrote for Scintilla, a GUI text editing component. On top of the
basic split buffer class, a text editor also often wants features for
tracking various things about the text such as line start positions,
styling, and undo. Performance is often more dependent upon these than on
the split buffer.
http://www.scintilla.org/

   There is a newer, better set of basic editing classes in SinkWorld, the
next version of Scintilla at:
http://scintilla.sourceforge.net/SinkWorld.html
   The major unresolved questions in providing a Python extension is what to
expose and how. It would be good to provide a basic split buffer, plus
allowing options to add undo, line position maintenance, styling and further
features. More contentious is whether to also provide callbacks from the
classes upon events such as insert text, as applications often use decoupled
code to mutate and display the text and so providing an observer interface
is a natural solution. There are then reentrance and performance issues,
especially with invoking Python code upon the lowest level actions.

>  1) There are functions to insert and delete single values into the array
>     but not multi-element slices.
> ...
> But it turns out that both operations are supported!  I discovered
> this by looking in the arraymodule.c source to see how the buffer
> interface was implemented, and then experimenting.  The docs for the
> array module gave no clue at all that these operations were supported.

   Damn! If I had known this then I would have developed at least an
experimental split buffer in Python.

   Neil




From phr-n2002a at nightsong.com  Mon Feb  4 21:06:56 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 18:06:56 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com> <7xheow3hl2.fsf@ruckus.brouhaha.com> <3C5F3A07.45126DAF@ccvcorp.com>
Message-ID: <7xg04gisvz.fsf@ruckus.brouhaha.com>

Jeff Shannon <jeff at ccvcorp.com> writes:
> I suspect that this is the crux of the whole issue.  My personal
> feeling is that the only reason anybody *cares* at all about 1.5.2
> is because of Red Hat, and thus, indirectly, the large number of
> ISP's using it because it's what came with RH.  Once RH starts
> shipping with a reasonable Python version (which *ought* to be with
> 8.0, whenever *that* comes out, but hopefully *soon*), then 1.5.2
> will begin fading as people upgrade their RH installations.  It
> probably won't completely disappear by the time that Python 3.0 is
> released, but it should be a negligible concern by then. 

OK, that takes care of 1.5.2.  I gather that 2.0.* never became very
widespread.  So I wonder now, whether more people are using 2.1.* than
1.5.2, and whether there will still be a significant amount of 2.1
installations when 3.0 is released.  If it sounds like there won't be,
then I'll probably just put

   if 3/2==1: <print upgrade-to-new-script message and exit>

at the top of my scripts and then code for 2.1.  When 2.1 gets more
displaced by 2.2 and later, I can switch over to import future division.


From eric.brunel at pragmadev.com  Wed Feb  6 04:42:05 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Wed, 6 Feb 2002 09:42:05 +0000
Subject: Saving Tkinter images?
References: <f88b21fc.0202051400.50a694ad@posting.google.com>
Message-ID: <a3qq2a$11m$1@norfair.nerim.net>

Hi Dustin,

> I'm interested
> if there is some way to save an image (eg. to a jpeg or png) that has
> been drawn to a Tkinter canvas.  I'm guessing there isn't but I
> thought I'd at least ask.

The only way I know to do that programmatically is to dump the canvas image 
as a postscript file using the "postscript" method on canvases, then use an 
external utility to convert the postscript to jpeg or png.
If you want to do that, there's an almost universal image conversion 
utility named "convert" that's part of ImageMagick 
(http://www.imagemagick.org), that runs almost everywhere.

HTH
 - eric -


From ivo at nospamAmaze.nl  Tue Feb  5 06:39:19 2002
From: ivo at nospamAmaze.nl (ivo at nospamAmaze.nl)
Date: 5 Feb 2002 11:39:19 GMT
Subject: Fun with lambda and map
References: <3c5eaf65.22609703@news.t-online.de> <a3mhnr$r2o$1@thorium.cix.co.uk>
Message-ID: <a3og97$6l0$2@news1.xs4all.nl>

gbreed at cix.compulink.co.uk wrote:
: Gerson Kurz wrote:

:> If you're in a masochistic mood, you can write this in one line:
:> 
:> fac = lambda n:map(lambda x:(map(lambda a:a and 1,[x[0]()])[0] and
:> x[1]()) or x[2](x[3]),[[lambda:not n,lambda:n==n,lambda
:> n:fac(n-1)*n,n]])[0]

: If you're allowing booleans and subtraction, how about

: fac = lambda x: not x or x*fac(x-1)

Well, I'm not allowing python operations, nor python numbers (the numbers
in the lambda system are presented as lambda's themselve, so a pythonic
'not' won't work on them. However, that could be fixed)

	Ivo

-- 
Drs. I.R. van der Wijk                              -=-
Brouwersgracht 132                      Amaze Internet Services V.O.F.
1013 HA Amsterdam                                   -=-
Tel: +31-20-4688336                          Linux/Web/Zope/SQL
Fax: +31-20-4688337                           Network Solutions
Web:     http://www.amaze.nl/                    Consultancy
Email:   ivo at nospamAmaze.nl                         -=-



From gh_pythonlist at gmx.de  Tue Feb 19 04:44:09 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 10:44:09 +0100
Subject: Subclassing a read-only class
In-Reply-To: <qS9c8.83559$3J1.2671693@news2.tin.it>
References: <mailman.1014046530.13105.python-list@python.org> <qS9c8.83559$3J1.2671693@news2.tin.it>
Message-ID: <20020219094407.GC2611@lilith.hqd-internal>

Le 18/02/02 ? 16:21, Alex Martelli ?crivit:
> Gerhard H?ring wrote:
> 
> > Hello,
> > 
> > For compatibility testing, I'd like to put an additional method into
> > window object I get from the curses module. Unfortunately, if I try to
> > do so, I get:
> > 
> >     TypeError: object has read-only attributes
> > 
> > I'm pretty sure there is a solution, probably with a proxy object. But
> > currently I just don't get it.
> 
> You can generally substitute automatic-delegation in lieu of
> inheritance (as long as nobody is so uncouth as to typetest,
> e.g. with isinstance -- in which case, you're in trouble).
> 
> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52295 gives
> a simple example ('inheriting' [via automatic delegation] from a
> file object back in Python 2.0 / 2.1 when you couldn't actually
> just subclass file as you can today) -- I hope you find the idea
> easy to adapt to your specific needs.

Thanks, that was what I was after. I just didn't remember where to find
an example of this pattern.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.0 ?C      Wind: 6.1 m/s



From brueckd at tbye.com  Thu Feb  7 13:38:43 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 7 Feb 2002 10:38:43 -0800 (PST)
Subject: encapsulation, typing & other questions
In-Reply-To: <3C62C5C5.7FEF0B10@daylight.com>
Message-ID: <Pine.LNX.4.33.0202071028550.5155-100000@ns0.tbye.com>

On Thu, 7 Feb 2002, Ragu Bharadwaj wrote:

> Hi Guys,
> 	I'm considering Python for enterprise level development and being new
> to Python but having used ObjC & Java a lot, have the following
> questions. Any answers would be appreciated.

Hi Ragu!

> 1. Are there plans to include type support for method/function calls in
> Python? Seems like it would lead to much faster error detection &
> correction.

I don't know of any plans to do this, and the benefits are probably less
than you'd imagine. This issue tends to come up as a theoretical problem
much more often (if ever) than it does in practice.

> 2. How easy is it to use design patterns in Python? A lack of typing
> suggests it wouldn't be too easy, but I'd like to hear back from folks
> that have tried this.

Any particular patterns? Obviously patterns revolving around how to deal
with type systems won't be of much use, but lots of other patterns are
very applicable.

> 3. Are there plans to include method & variable encapsulation in the
> object models in python? The answer I've heard is that since it's
> possible to overcome encapsulation by long or dirty ways, it isn't worth
> it. But my comeback is that if one sticks to the rules while coding,
> this won't happen and the resulting gains are worth it.

Python does have method and variable encapsulation, and it eliminates the
accidental use of "protected" members. It is not, however, a babysitter
for people who cannot follow coding standards.

> 4. How advanced is db access in Python? Does it use connection pooling
> and other optimizations like the JDBC drivers?

There are extension modules for the most popular databases (Oracle, MySQL,
Postgres, etc). Also, you *can* use the JDBC drivers themselves - just use
Jython. It works very, very well.

-Dave




From pythonlist at zumnet.com.br  Thu Feb 14 13:50:39 2002
From: pythonlist at zumnet.com.br (Rafael Mentz Aquino)
Date: Thu, 14 Feb 2002 16:50:39 -0200
Subject: Curses & Colors
Message-ID: <3C6C06FF.DAFFBB43@zumnet.com.br>

Hi, there...

I'm creating some simple interfaces with curses with python under
FreeBSD,
but I'm having some problems to change colors. I tried to do something
in
/etc/termcap, but when I run curses.can_change_color(), I always obtain
0.
My env term is cons25, but I don't know if python uses the same
terminal.
I'm running python 2.1.1. Can anybody help me...? ;-))

Thanks!

Rafael Mentz Aquino
Porto Alegre - RS - Brazil




From marklists at mceahern.com  Sat Feb  9 21:03:40 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sat, 9 Feb 2002 18:03:40 -0800
Subject: A newbie question about class
In-Reply-To: <a50631f6.0202091635.1dc42d4b@posting.google.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEOEKBAA.marklists@mceahern.com>

[newgene]
> Hi, group,
>      I have a newbie question about building a class:
>
> eg.
>
> class A:
>    def __init__(self,x=None):
>        self.x=x
>        if self.x != None:
>             self.y=x**2
>        else:
>             self.y=x**2

1.  When comparing to None, it's better to use one of the following:

	if x is not None:...
	if x is None:...

2.  Consider using property, which is new with Python 2.2 and requires a
new-style class (thus we subclass from object):

class A(object):

	def __init__(self, x=None):
		self._x = x

	def getX(self):
		return self._x

	def setX(self, x):
		self._x = x

	x = property(getX, setX)

a = A(1)
print a.x

Cheers,

// mark




From eppstein at ics.uci.edu  Fri Feb 22 03:18:59 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Fri, 22 Feb 2002 00:18:59 -0800
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1> <7xd6yyfdep.fsf@ruckus.brouhaha.com>
Message-ID: <eppstein-6DBBB2.00185922022002@news.service.uci.edu>

In article <7xd6yyfdep.fsf at ruckus.brouhaha.com>,
 Paul Rubin <phr-n2002a at nightsong.com> wrote:

> Am I missing something?  Python has the usual bit operations that work
> on ints.

I found the automatic integer-long conversions messed up some C idioms for 
bit twiddling:

# wrong way to count number of nonzero bits in a <=32-bit integer
def countBits(x):
        i = 0
        while x != 0:
                x &= x - 1
                i += 1
        return i

# better
def countBits(x):
        i = 0
        if (x < 0):   # prevent infinite loop from int->long autoconversion
                x ^= (1<<31)
                i = 1
        while x != 0:
                x &= x - 1
                i += 1
        return i
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From mwh at python.net  Fri Feb 22 05:40:21 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 22 Feb 2002 10:40:21 GMT
Subject: giant stdlib tuples (was: Re: Help with code)
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com> <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu> <uadu3m5dz.fsf@python.net> <slrna7agtq.1vk.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <uadu1lrum.fsf@python.net>

quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan) writes:

> On Thu, 21 Feb 2002 11:35:37 GMT, Michael Hudson <mwh at python.net> wrote:
> >quinn at schilling.ugcs.caltech.edu (Quinn Dunkan) writes:
> >> It's always seemed to me that things like time, pwd, grp, and stat
> >> are pretty non-pythonic.  You can't even really say it's because of
> >> the thin wrapper thing because writing and using a StructType or
> >> something wouldn't be much harder than using tuples.
> >> 
> >> Of course, I've learned to carry around my own little wrappers, but
> >> I've always wondered why they were that way in the first place.
> >
> >Used Python 2.2 yet?
> 
> Aha!  I hadn't seen this one advertised anywhere, and they disguise
> themselves as tuples.

Well, they have to, or lots of code would break...

> pwd and grp still don't return structs, but it's a good start.
> 
> Deserves more advertisement, I think:

Perhaps.  Don't try and pickle the results until 2.2.1 gets released...

Anyone feel like contributing code to similarly enhance pwd and grp?

Cheers,
M.

-- 
  "The future" has arrived but they forgot to update the docs.
                                        -- R. David Murray, 9 May 2000


From loewis at informatik.hu-berlin.de  Mon Feb 18 11:46:12 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 18 Feb 2002 17:46:12 +0100
Subject: super(obj)?
References: <mailman.1014045511.10659.python-list@python.org>
Message-ID: <j4sn7yybzf.fsf@informatik.hu-berlin.de>

Gustavo Niemeyer <niemeyer at conectiva.com> writes:

> AFAICS, this seems to be the most useful version of super:
> 
> super(obj.__class__, obj)

It would usually be wrong. Consider

class Base(object):
  def doit(self):
    pass

class Derived1(Base):
  def doit(self):
    super(Derived1, self).doit()

class Derived2(Derived1):
  def doit(self):
    super(Derived2, self).doit()

obj = Derived2()
obj.doit()

This could does the expected: Derived2.doit call2 Derived1.doit calls
Base.doit. If you change Derived1 to

class Derived1(Base):
  def doit(self):
    super(self.__class__, self).doit()

then self.__class__ would be Derived2, so the Derived1 would call
itself recursively, instead of performing a super call.

HTH,
Martin


From bldrake1 at yahoo.com  Fri Feb  8 19:46:42 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 8 Feb 2002 16:46:42 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <st888.7971$oI4.33917985@newsserver.ip.pt> <a3retd$19cgik$1@ID-11957.news.dfncis.de> <d0380df9.0202080516.7a76905d@posting.google.com>
Message-ID: <fb072181.0202081646.24900468@posting.google.com>

Hugo,
I tried to duplicate your results.  Here is the output in Idle:

Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import array
>>> myArray = array.array('f', [.5, .5, .5, .5])
>>> myArray
array('f', [0.5, 0.5, 0.5, 0.5])
>>> n = 10
>>> for i in range(len(myArray)):
	    myArray[i] += (1.0/n)

>>> myArray
array('f', [0.60000002384185791, 0.60000002384185791,
0.60000002384185791, 0.60000002384185791])
>>> for i in range(len(myArray)):
	    if (myArray[i] > 0.0) and (myArray[i] < 1.0):
		    exit(0)
		
Traceback (most recent call last):
  File "<pyshell#11>", line 3, in ?
    exit(0)
>>> len(myArray)
4
>>> def TestFunc(min, max):
	    for i in range(len(myArray)):
		    if (myArray[i] > min) and (myArray[i] < max):
			    print "I quit!"

			
>>> TestFunc(0.0,1.0)
I quit!
I quit!
I quit!
I quit!
>>> import sys
>>> def TestFunc(min, max):
	    for i in range(len(myArray)):
		    if (myArray[i] > min) and (myArray[i] < max):
			    print "I quit!"
			    sys.exit(0)

			
>>> TestFunc(0.0,1.0)
I quit!
Traceback (most recent call last):
  File "<pyshell#32>", line 1, in ?
    TestFunc(.0,1.)
  File "<pyshell#31>", line 5, in TestFunc
    sys.exit(0)
SystemExit: 0
>>>

As you can see I tried with exit(0) and without.  In both cases the
loop terminates as you wanted.

It might help if we knew what computer system your using and the
version of Python.  Have you tried using the debugger to inspect
memory; printing out intermmediate values, etc.?  The dubugger is
integrated with Idle and it's quite good.

Barry

hugomartires at hotmail.com (Hugo Martires) wrote in message news:<d0380df9.0202080516.7a76905d at posting.google.com>...
> "Emile van Sebille" <emile at fenx.com> wrote in message news:<a3retd$19cgik$1 at ID-11957.news.dfncis.de>...
> > "Hugo Martires" <hmartires at cultalg.pt> wrote in message
> > news:st888.7971$oI4.33917985 at newsserver.ip.pt...
> > > but i don?t need all the numbers, i only need 0.6
> > 
> > See 4.98 in the FAQ at http://python.org/doc/FAQ.html
> 
> I have read it but still confused.
> Se my point:
> --> n=0
> 
> myArray = array.array('f', [0.5, 0.5, 0.5, 0.5]
> 
> myArray[i] = myArray[i] + (1.0/n)
> 
> myArray[0] = 0.600000023 ("something like this")
> 
> for i in range ( len(myArray) ):
> 	if (myArray[i] > 0.0) and (myArray[i]<1.0):
> 		exit(0)
> 
> The problem is that he it never exit, unless i do like this:
> for i in range ( len(myArray) ):
> 	if (myArray[i] > 0.0001) and (myArray[i]<0.9999):
> 		exit(0)
> 
> that is not the better way :(
> 
> ------------------------------------------------------------------
> I have tried with:
>     myArray=array.array('f')
>     myArray=array.array('d')
> and even with a list:  myArray=[]
> 
> I know that the problem is in the internal storage of the numbers, but
> i canot understand how to solve that.
> I just want to make a simple calculation and an if statement.
> 
> Tanks for help


From anton at vredegoor.doge.nl  Thu Feb  7 17:17:31 2002
From: anton at vredegoor.doge.nl (Anton Vredegoor)
Date: Thu, 07 Feb 2002 23:17:31 +0100
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org>
Message-ID: <a3uukp$5dq$1@news.hccnet.nl>

On Thu, 7 Feb 2002 08:01:24 -0600, "Chris Gonnerman"
<chris.gonnerman at newcenturycomputers.net> wrote:

>Ah, perhaps you are right.  I was in a particularly bad mood at that point.

Enough of this. I have decided to add some code to this project. The
code shares some characteristics with the autocoding project in that I
myself have trouble understanding what I have wrought, but it has the
advantage of a more limited objective. It is in need of some
simplyfing and better documentation. I already succceeded in producing
a smaller module that does mostly the same but it has less
functionality. What I want is to condense this module into one or two
lines of code as accomplished in another related thread (you will know
which one if you've read it). After that it should be expanded again
for readability.

Now what is it? Its a permutation engine for multisets. Read more
about it at:

http://home.hccnet.nl/a.vredegoor/multiperm.py

or (html version):

http://home.hccnet.nl/a.vredegoor/multiperm.py

The relevance for the autocoding project is that if there are nine
possible things one can do, and if 

- these actions are taken sequentially, -instead of in parrallel
(admittedly a dangerous assumption)-
- it is known beforehand how many times each action is taken

then this module can produce all possible sequences of actions that
can be taken and can assign a unique number to each one of them.

After this, needing to find the input and output of the different
processes will be the remaining problem, as well as finding the right
permutation index number.

I'm desperately in need of simplyfing this code, so if any programmers
reading the autocoding module are now in need of some code that is
more accessible I would appreciate constructive as well as
reductionistic comments about it.

Anton.




From robin at jessikat.fsnet.co.uk  Sat Feb  2 09:21:04 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 2 Feb 2002 14:21:04 +0000
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
 <313g3a.d22.ln@wombat.san-francisco.ca.us>
 <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk> <37d27e024b.tim@worthy.demon.co.uk>
Message-ID: <PLPSljAQX$W8Ewpx@jessikat.fsnet.co.uk>

In article <37d27e024b.tim at worthy.demon.co.uk>, Tim Howarth
<tim at worthy.demon.co.uk> writes
>In message <6vTfsIAio7W8EwNp at jessikat.fsnet.co.uk>
>          Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>
>> well if I could get the win32com extensions to change passwords on
>> different machines for me I would really like that
>
>Why win32com rather than win32net (where it's easy)?
>
well I assumed we have to do it through DCOM, but if it's easy with
win32net what's the easy way? I'm not a winnt/2k/9x person and don't
really understand the system internals at all.
-- 
Robin Becker


From ruediger.maehl_nospam at web.de  Fri Feb  1 02:07:42 2002
From: ruediger.maehl_nospam at web.de (Ruediger Maehl)
Date: Fri, 1 Feb 2002 08:07:42 +0100
Subject: [py2exe] How can I specify additional module search paths
References: <newscache$wbmsqg$4ed$1@www-neu.dzsh.de> <a3bqhs$179dje$1@ID-59885.news.dfncis.de>
Message-ID: <newscache$htduqg$dgm$1@www-neu.dzsh.de>

"Thomas Heller" <thomas.heller at ion-tof.com> schrieb im Newsbeitrag
news:a3bqhs$179dje$1 at ID-59885.news.dfncis.de...
>
> > [I am changing sys.path during program execution]
>
> the solution is simple: don't do that. Don't modify sys.path in your
program.
>
> There are several reasons:
> - It make it impossible for py2exe to find the modules you need (although
>   you *could* modify sys.path inside your setup-script *shudder*)

One of my solutions as preliminary workaround. The other was to
add the additional path to $PYTHONPATH for py2exe's execution time.

> - Even if py2exe would find your modules, the frozen executable would not,
>   at least not in the way you would expect. The frozen executable runs
>   with an empty sys.path [], but finds the modules via an installed import
>   hook, which is searched first.
>   Modifying the path now would instruct the executable to additionally
search
>   for modules or packages in the *file system*.

Yes, that does not work.

> Why do you want to modify the sys.path in your program?

Because I have a library directory for each application package, that
contains only application relevant modules.

Now that I found py2exe will work for me (upto now, at least),
I must perhaps consider a different approach.
Btw, any problems with the reportlab package?

Where do _you_ place application specific modules?
Or do you include all these directories in $PYTHONPATH?

Many regards,

R?diger




From jimd at vega.starshine.org  Sun Feb 17 13:43:45 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 17 Feb 2002 18:43:45 GMT
Subject: __class__ for 2.2 types
References: <2ec1bc1c.0202071240.44185531@posting.google.com> <a3upaq$1arn1e$1@ID-59885.news.dfncis.de>
Message-ID: <a4otl1$44d$1@news.idiom.com>

In article <a3upaq$1arn1e$1 at ID-59885.news.dfncis.de>, Thomas Heller wrote:

>"Larry" <wrbt at email.com> wrote in message news:2ec1bc1c.0202071240.44185531 at posting.google.com...
>> hi, rookie here, I was hoping someone could explain to me why this happens:

>> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>> Type "help", "copyright", "credits" or "license" for more information.
>> >>> [].__class__
>> <type 'list'>
>> >>> int.__class__
>> <type 'type'>

>> why doesn't the int one return <type 'int'>?

> int is the *type* of an integer, not an integer itself.

> See:
>>>> list.__class__
> <type 'type'>
>>>> 2.__class__
>  File "<string>", line 1
>    2.__class__
              ^
>SyntaxError: invalid syntax
>>>> 2 . __class__
><type 'int'>

>Thomas

 That's an interesting "gotchya"  I understand why 2.__class__ is a
 syntax error (ambiguous with a float: 2.0 etc).  But I'm surprised 
 that 2 . __class__ works; the spaces seem unnatural here. 

 (2).__class__ makes sense because it uses parentheses in a natural 
 way (to disambiguate precedence and/or parsing in an expression). 






From rjones at ekit-inc.com  Tue Feb 26 18:06:15 2002
From: rjones at ekit-inc.com (Richard Jones)
Date: Wed, 27 Feb 2002 10:06:15 +1100
Subject: Need advice for obfuscating Python code
In-Reply-To: <Pine.LNX.4.33.0202261511020.13167-100000@ns0.tbye.com>
References: <Pine.LNX.4.33.0202261511020.13167-100000@ns0.tbye.com>
Message-ID: <200202262306.XAA09222@crown.off.ekorp.com>

On Wed, 27 Feb 2002 10:28, brueckd at tbye.com wrote:
>
> [snip solution which encodes the module bytecode in some manner]
>
> That should work. Of course, since the code that decrypts and loads your
> code will be visible, you probably want to use a long key (or another form
> of encryption) and install the customer importer from a C program that
> runs an embedded interpreter (installs the importer and then runs the rest
> of your program).

Or you could bundle the encoded bytecode in a C module, which makes the 
solution a lot neater.

None of this is a substitute for a good license and some manner of 
identifying the code when it is stolen. One suggestion I read from the timbot 
involves engineering in specific, obscure-case bugs.

Don't rely on smoke and mirrors to protect your IP. This applies to any 
programming language out there. Witness the number of warez sites with 
cracked programs to offer.


    Richard



From salmonia at cf.ac.uk  Wed Feb 13 07:59:46 2002
From: salmonia at cf.ac.uk (Alan james Salmoni)
Date: Wed, 13 Feb 2002 12:59:46 GMT
Subject: controlling web browser in windows
Message-ID: <1103_1013605186@netnews.cf.ac.uk>

Hi,

My problem is basically this:

I am doing some psychological research into search engine usability and need to have control of a web browser. I am using Python (cos it's simple - I am a psychologist, not 
a programmer!) and looked at the webbrowser module but need to control the back button (so that when a subject completes a task, the back button event is automatically 
called) and have access to the history list.

Does anyone have any suggestions as to how this could be done. I can use IE, Netscape, or Opera (only on Windows though).

Alan Salmoni
HCI Group
Cardiff University.





From quinn at cruzeiro.ugcs.caltech.edu  Fri Feb 15 22:23:33 2002
From: quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan)
Date: 16 Feb 2002 03:23:33 GMT
Subject: Fear of autocoding!
References: <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <1476.811T2040T12515414threeseas@earthlink.net>
Message-ID: <slrna6rk5j.2pn.quinn@cruzeiro.ugcs.caltech.edu>

On Sat, 16 Feb 2002 01:45:49 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>Which you could have done by not doing anything, in comparison to what you
>have done.

Do people laughing at you make it harder for you to write your thing?

Awww... sorry.  Other people will entertain themselves that way, though, if you
give them the opportunity.  Obviously you're entertained by it too, or else you
wouldn't come back and post when you could be writing up your gizmo.

So everyone wins, and we all keep each other occupied.  Usenet is such a
friendly place.

If it takes 14 years to write 1 of 9, then Armageddon for programmers is at
2114.  By then we'll have already exploded when the unix 32bit epoch rolled
over.

Start your clocks:
http://timecube.com/index15.html


From aries at bitstream.net  Wed Feb 20 10:23:58 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Wed, 20 Feb 2002 09:23:58 -0600
Subject: Guido wins FSF Award
References: <mailman.1013906967.6705.python-list@python.org>
Message-ID: <3c73bdf1$0$79553$6536d21b@news.bitstream.net>

Tim Peters <tim.one at home.com> wrote in message
news:mailman.1013906967.6705.python-list at python.org...
> [Paul Rubin]
> > Congratulations to Guido for winning the fourth Free Software Foundation
> > Award, for inventing and implementing Python as Free Software:
> >
> >   http://www.fsf.org/press/2002-02-16-FSF-Award.html
>
> Hooray for Guido!  Shows how much the world cares about secure pickles
> <wink>.
snip

what are revealing the Nine Pythonic Actions underlying all that we do.
can you explain what that means ??????  thanks
Joe F.





From kirill_simonov at mail.ru  Sun Feb 24 01:53:59 2002
From: kirill_simonov at mail.ru (Kirill Simonov)
Date: Sun, 24 Feb 2002 08:53:59 +0200
Subject: Tkinter + Entry + cp1251 cyrillic
In-Reply-To: <26428104.0202231657.639cec52@posting.google.com>; from redskin@tut.by on Sat, Feb 23, 2002 at 04:57:19PM -0800
References: <26428104.0202231657.639cec52@posting.google.com>
Message-ID: <20020224085359.A966@abulafia.local>

On Sat, Feb 23, 2002 at 04:57:19PM -0800, Andrey Sidorenko wrote:
> I use Active Python 2.2 on Windows 2000 and have trouble - "Entry"
> class can't display cyrillic letters (cp1251 encoding) correctly in
> entry field. Can anybody help me with this weird problem?

Try to replace "Python22\Lib\lib-tk\FixTk.py" with the following:

import sys, os

if not os.environ.has_key('TCL_LIBRARY'):
    tcl_library = os.path.join(sys.prefix, "tcl", "tclX.Y")
    os.environ['TCL_LIBRARY'] = tcl_library


Kirill.




From cbbrowne at acm.org  Sat Feb 23 00:42:41 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Sat, 23 Feb 2002 00:42:41 -0500
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com> <mailman.1014434138.12990.python-list@python.org>
Message-ID: <m3d6ywg3e6.fsf@chvatal.cbbrowne.com>

The world rejoiced as Gustavo Niemeyer <niemeyer at conectiva.com> wrote:
>> > - Platform independence is gone.
>> 
>> I hope not...  I would hope to hear that "platform indepedence has
>> _ARRIVED_"

> That's the human nature. Tismer writes half a dozen mails, some with
> unbelivable results, and some asking for help. One of the first
> answers he gets is complaining about something.

It wasn't intended as a "complaint;" I was somewhat surprised, and was
making a fairly much value-free comment (that could be taken more than
one way :-(), somewhat hoping that the answer was "oops; typo" perhaps
accompanied by some suitably sarcastic come-backs.

The expansion on the matter has been educational.
-- 
(reverse (concatenate 'string "moc.adanac@" "enworbbc"))
http://www.ntlug.org/~cbbrowne/spiritual.html
Rules of the Evil Overlord #216. "If my Legions of Terror are defeated
in a battle, I will  quietly withdraw and regroup instead of launching
a haphazard mission to assassinate the hero."
<http://www.eviloverlord.com/>


From phr-n2002a at nightsong.com  Tue Feb 26 02:19:20 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 23:19:20 -0800
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com> <5uGe8.4184$LN5.796@nwrddc01.gnilink.net>
Message-ID: <7xwux04snb.fsf@ruckus.brouhaha.com>

"logistix" <logstx at bellatlantic.net> writes:
> The type advice is really impossible in a dynamic language and it's not
> going to help out a native compiler anyways.  A better example (than Example
> 2) of code biting you in the ass is:

Of course it's possible in a dynamic language, and it helps compilers.
Lisp compilers have worked that way since the 60's or even earlier.

> def foo(x):
>     local y(int)
>     y = x
> 
> This will only be able to successfully statically analyze if every call made
> to foo is passing in a statically typed value.  So for any of the "advice"
> to work, all vars need to be typed. 

Obviously that example can't be checked statically by the compiler.
It can be checked dynamically at runtime (debug mode) or the compiler
can generate code that assumes the advice is correct and which crashes
if the advice is wrong (optimized mode).

> This is exactly the way you have to validate info in "real
> languages" like C++ (and I imagine java). RTTI is required if your
> class heirarchies get too crazy for static checking (such as a toy
> lisp interpreter I wrote, all lisp objects inherited from a base
> "object" class).

If you've written a toy Lisp implementation, you might also like to
look at how a real Lisp compiler works.  They use declarations for
optimized code generation, while remaining dynamic.  

> This would remove all dynamic properties from python, which is
> probably undesirable.  If you do need/want to validate the type at
> runtime, you might as well use the builtin type() function.

Why would I want to use the builtin type() function everywhere I want to
validate the type, if Python can do the check for me automatically based
on the type advice?


From emile at fenx.com  Tue Feb 26 11:53:33 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 08:53:33 -0800
Subject: Time to rename Stackless?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org> <915a998f.0202260438.246be39c@posting.google.com> <mailman.1014738044.20018.python-list@python.org>
Message-ID: <a5gevj$77e6d$1@ID-11957.news.dfncis.de>

"Christian Tismer"
>
> Not bad!
> Should I start a contest?
>

<wonder>Do we get to vote with stackless?</wondering>

<if voting>No</if>
<else proposed_name=stackless--revisited</fi>

<why_you_ask>

For the anticipated pleasure of one day seeing 'stackless' as part of
the standard release.

<now_stop_asking>

I hope you've found funding to continue.  Do we send contributions to
Kaunstr.?

--

Emile van Sebille
emile at fenx.com

---------



From bit_bucket5 at hotmail.com  Mon Feb 18 11:38:13 2002
From: bit_bucket5 at hotmail.com (Chris)
Date: 18 Feb 2002 08:38:13 -0800
Subject: List unchanged in function--why?
Message-ID: <fa7108b0.0202180838.6557992a@posting.google.com>

This must have something to do with references etc, but I'm not sure
what's going on.

def changeList(theList):
  theList = ['a']

aList = ['x', 'y']
print aList
changeList(aList)
print aList


Running the above results in

['x', 'y']
['x', 'y']

Can someone explain?  I was expecting the second line printed to be
['a']

Thanks,
Chris


From rasmussn at lanl.gov  Tue Feb 19 18:29:19 2002
From: rasmussn at lanl.gov (rasmussn at lanl.gov)
Date: Tue, 19 Feb 2002 16:29:19 -0700
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <7F6415ED-2590-11D6-9E6A-0030657EBC3E@lanl.gov>

On Tuesday, February 19, 2002, at 09:57 AM, Craig Maloney wrote:
>
> I'm doing some scientific computing and have
> amassed a moderate set of classes.  Due to the
> heavy computational demands, I don't think I
> can justify rewriting everything in an interperated
> language; nice as that would be now that I'm
> playing around with what I've created.

> I would, of course, like to have some sort of scripted
> environment (e.g. Python) in which to "play"
> with my objects.

> I'm making essential use of templates and polymorphism.
>
> I have no experience with Boost.Python, but from what I
> understand, it would not involve writing wrapper definitions
> (ala SWIG or IDL), right?  Are there other options?

You should consider SILOON which was created for scripting
scientific code.  It works particularly well with C++
and works by parsing user source directly.  No wrapper definitions
or IDL is required.  See http://www.acl.lanl.gov/siloon.

A new version is coming out in a couple of months which generates
Python extensions for Fortran 95 (again by parsing user source
code to discover Fortran procedures to export to Python).

> Or should I just wait for .NET/mono to come out? ;)

Of course.

Craig




From akakakhel at attbi.com  Sun Feb 24 21:53:02 2002
From: akakakhel at attbi.com (Ali K)
Date: Mon, 25 Feb 2002 02:53:02 GMT
Subject: images
Message-ID: <iGhe8.20623$vP.68360@rwcrnsc51.ops.asp.att.net>

I would like to know how to show a picture in Python so that it shows up on
my window.

Please Reply.

Thank You.

Ali K.




From phr-n2002a at nightsong.com  Thu Feb 21 01:12:01 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 22:12:01 -0800
Subject: Recursively traverse linked list -- help!
References: <mailman.1014240283.12938.python-list@python.org> <7xadu3wwr5.fsf@ruckus.brouhaha.com> <mR%c8.57438$X64.20569489@news1.rdc2.pa.home.com>
Message-ID: <7xzo23weha.fsf@ruckus.brouhaha.com>

"Terry Reedy" <tjreedy at home.com> writes:
> > Oh yes, you're correct.  However, the tail call should still be
> > optimized, compiled as an indirect jump to f.
> 
> I can accept 'maybe could' (see below) but 'should'?  What sense of
> shouldness are you using?  

The normative sense of what one can expect of a compiler that implements
tail recursion optimization.  That's separate from the question of
whether compilers should be expected to implement TRO.  For Scheme,
it's required and you can expect it.  For Python, it's not required
and you shouldn't expect it.  OK?


From loewis at informatik.hu-berlin.de  Sun Feb  3 09:30:47 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Feb 2002 15:30:47 +0100
Subject: The object named 'type', in 2.2
References: <mailman.1012717865.20281.python-list@python.org>
Message-ID: <j4bsf6hc2w.fsf@informatik.hu-berlin.de>

"Jason Orendorff" <jason at jorendorff.com> writes:

> I don't know where you can find more information, sorry.
> It might be that recursively looking at __class__es is not what
> you want after all; maybe you should walk x.__class__.__bases__
> instead.

There is, of course, definitive information on all that: The Python
source code proper.

Regards,
Martin


From mdk at mdk.com  Fri Feb  1 10:39:05 2002
From: mdk at mdk.com (MDK)
Date: Fri, 1 Feb 2002 10:39:05 -0500
Subject: Help with time
Message-ID: <a3ecqt$16bk9d$1@ID-98166.news.dfncis.de>

I am passing to a function how long in milliseconds I want to wait for data;
a timout limit.

For example:

data = getdata(300)

I just can not figure out how to loop until time expires.  Here is what I
have and does not work.

import time

def getdata(ms):
    starttime = time.time()
    while len(data) < 20:
        now = time.time()
        print '%.3f seconds %.3f' % (now - starttime, timeout / 1000)
        if now - starttime > timeout/1000:
            break
    return data

What am I doing wrong?

Thanks.




From adalke at mindspring.com  Tue Feb 12 16:30:26 2002
From: adalke at mindspring.com (Andrew Dalke)
Date: Tue, 12 Feb 2002 14:30:26 -0700
Subject: photo id ?!?!
Message-ID: <04a501c1b40c$7d17bf00$0201a8c0@josiah.dalkescientific.com>

Dear Guido, Foretec, and the Python community,

Last week I was almost unable to attend the Python conference.  Not
because I was sick, or unable to travel, but because I refused to
present photo id during registration.  [1]

I pointed out that half the conference probably knows who I am and
could vouch for me.  That was not allowed.  (On reconsideration,
that's more like 1/4th to 1/5th, but my point is that I could
establish a trust network.)

I offered to show my business card, and that was not allowed.  Photo
id or I would have to leave.

I said then that I would not attend with that requirement (especially
since it was sprung on me at the last moment), and asked for a refund.
I was told I would not get a refund without showing a photo id.
Catch-22.

Frustrated I stood there not knowing what to do.  I pointed out they
should recognize me from the previous times I've been to the
conference and that I recognized a few of them, at least by face.
At this point, Fred Drake appeared and said "Hi Andrew" and welcomed
me to the conference.  (Thanks Fred!) The Foretec people finally
capitulated and allowed me to attend without showing photo id, after
saying they did somewhat recognize me from last year.

Attending the Python conference for five years, sending in bug reports,
improvements, documentation, contributing to comp.lang.python,
helping people learn the language, writing articles about Python,
promoting Python in the computational life sciences, founding
several open-source Python based projects including Biopython, presenting
papers, and posters at previous conferences, and organizing a BoF at
this one ... has some benefits.

I was told at this point that the reason for the photo id
requirement was 1) to prevent fraud, and 2) in this day and age
we cannot be too careful.

I understand the fraud part.  I know of two relevant forms of
fraud - illegal use of a credit card and identity theft.  If I
had known beforehand I would have sent a check a month early so
that it would clear before I attended, or I would have brought
cash with me to pay at the conference itself.  As to the identity
theft, I had with me the printout from my registration, which
contained the confirmation code.  That should be a unique and
relatively well hidden information.

Upon review of the conference registration materials, I see that
Foretec charges $50 for substitutions made to the reservation.
I suppose I could defraud them of that charge if I weren't me, but
were that the case I would still rather pay the extra $50 than
present photo id.  I still would complain about the unjustified
claim of theft implicit in that charge.  I also note that people
who registered on-site, in person, were still required to show id,
so this can't be the sole reason for this policy.

Therefore I assert that there are other ways to handle the fraud
issue than to require photo id.

As to the "in this day and age" claim.  First off, I find it
outrageous to believe that showing photo id for the conference has
any bearing on the state of terrorism in the world.  After all,
everyone on those planes showed photo id.  Israel has mandantory
id yet also has many acts of political violence.  The UK had
mandantory id but that didn't eliminate with violent protests related
to Ireland, while it did enforce discriminitory practices.  As
far as I can research, nationwide mandantory identification papers
does not reduce crime nor does it prevent acts of violence.

But in a deeper, more philosophical sense, I will not show photo
id for a conference because it represents an undue and completely
inappropriate government influence on my activities.  I have
two forms of photo id - a driver's license and a passport.  The
first gives me license to drive and the second one gives me the
right to enter and leave national boundaries.  Both are revocable
at will by the government.  Neither has any relevance to the Python
community.

Why should revocation of those two forms of id give just cause
to prevent my from attending the Python conference?  For that matter,
it means that having my wallet stolen has an all too serious and
drastic effect on my life.  (Had it been stolen in DC I would have
been unable to show ID; and I don't know how I would have gotten home!
For fear of that, I left my driver's license at my friends' house
while I played tourist in DC after the conference.)

In addition, stretching more into my personal history, I come from
a family of non-denominational Protestant missionaries, who have
worked in Baptista-era Cuba, Velasco and Lara-era Ecuador, and
Duvalier-era Haiti.  These were all autocratic governments.  One
of the things I took to heart from family stories was that freedom
and liberty includes the right to travel and conduct business without
having to get government permission for everything.

I am proud that is it possible even now in the US for people to live
and prosper without needing any sort of photo id and suffer only
relatively few difficulties.  I could have taken a bus, train,
hitchhiked, biked, chartered a plane, or even canoed to get to DC,
without needing ID along the way.

(Let me caution that I am not claiming the US is a police state or
other sorts of rash claims.  I am stating one of several reasons for
why I have such strongly held beliefs on this matter.  I happen to
also believe this matter can be resolved such that Foretec and I are
both happy with the result.)

I have attended many conferences and meetings over the last 16 years,
ranging in size from 20 to 6,000.  I've helped organize several
conferences - the largest at about 1/2 the size of Python10.  The only
time I needed to show id was when the meeting was on a military base,
and the base (not the conference organizers) needed to identify non-US
citizens.  In that case, we had plenty of warning beforehand that IDs
were needed.  Otherwise, I know from experience that IDs are not needed
for a conference like this. [2]

When I complained about this to Michel Sanner, he kidded me that I
was acting like a real American - complaining about government
control and living in the Southwest where life is more free.  I take
that as a compliment.  I chose my lifestyle carefully and I want to
keep it that way.  I will not voluntarily allow government intrusion
where it has no business, and I will continue to deny meaningless
access to government issued documents [3].

I would like to attend my 6th Python conference next year, but
cannot do so without compromising my strongly held beliefs unless
Foretec changes their policies.  What I ask as a result of this
complaint is one of two possibilities:

  - Foretec provide a mechanism to let people attend without showing
photo id.  I outlined two ways to do this: I can pay in cash at the
desk, or I can send in a check beforehand and present a confirmation
number.

  - The Python organizers choose someone other than Foretec to
handle the conference; someone who does not have this requirement.
To guarantee that that is possible, I will offer now to help organize
next year's conference here in Santa Fe.  At the very least, the
downtown hotels are within blocks of a lot of restaurants, shops,
and galleries, which would be a nice change from the last four
conferences where we all stayed in the hotel the whole time.

                    Andrew Dalke
                    dalke at dalkescientific.com


[1] "photo id" - I own two pieces of photo id; my driver's license
and my passport.  Both are government issued and revocable on
demand by the government.  In the above I deliberately assume
"government issued photo id" because there were no guidelines for
what constituted a valid photo id.  I had my camera with me and
somehow doubt that taking a picture of myself, making a printout,
and signing my name would have counted.

[2] I do know that some people will gatecrash.  The numbers from
conferences I've attended suggests that that number is very small,
and additional hassle needed to screen everyone is not worth it.

[3] it helps that I don't smoke and never bought alcohol at a
restaurant until I was 29, but I do understand the need to verify
age.  It also helps that I pay in cash or debit/credit card, so don't
need to show id at the stores for a check.





From nospam at bigfoot.com  Mon Feb 11 14:03:32 2002
From: nospam at bigfoot.com (Gillou)
Date: Mon, 11 Feb 2002 20:03:32 +0100
Subject: Zope cookbook?
References: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au>
Message-ID: <a4944q$1ie7$1@norfair.nerim.net>

Yes,

Even if there is more and better docs (Web an books), the Zope learning
curve is still long.
Lots of essential stuff to know are still in HowTo's in the Zope site.

First learn the principles of acquisition. The rest will follow (nearly)
naturally.
--Gilles




From tismer at tismer.com  Tue Feb 26 19:27:40 2002
From: tismer at tismer.com (Christian Tismer)
Date: Wed, 27 Feb 2002 01:27:40 +0100
Subject: Time to rename Stackless?
References: <m3pu2sancq.fsf@chvatal.cbbrowne.com> <a5fqbi$p5q$1@thorium.cix.co.uk> <pan.2002.02.26.15.05.03.465651.1294@uchicago.edu>
Message-ID: <3C7C27FC.4010503@tismer.com>

Ben Wolfson wrote:

 > On Tue, 26 Feb 2002 05:08:34 -0600, gbreed wrote:
 >
 >
 >>>There have got to be some puns heading in the direction of "seriously
 >>>stacked."  Hmm...
 >>>
 >>Wonder Python!
 >>
 >
 > Builds strong threads and coroutines twelve ways?


If you spell all 42 ways, I'll consider.

BTw I'm not sure whether the subject is renaming
Stackless Python. We (Of course Jean-Claude Wippler and me)
are currently stackless-ing Tcl and Perl as well.

A new naming challenge!

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
       where do you want to jump today?   http://www.stackless.com/






From jkraska at san.rr.com  Fri Feb  8 12:09:51 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 17:09:51 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com> <slrna67k0u.for.philh@comuno.freeserve.co.uk>
Message-ID: <br186u8qg3bu2hpsf0dcgs1ldih7gir4rg@4ax.com>

On Fri, 8 Feb 2002 13:18:22 +0000, philh at comuno.freeserve.co.uk (phil hunt) wrote:

>On Fri, 08 Feb 2002 01:42:51 GMT, Courageous <jkraska at san.rr.com> wrote:
>>
>>>It's just that being forced -- "for my own good" -- not to use certain
>>>features because someone has decreed them intrinsically bad, doesn't
>>>appeal to me at all.
>>
>>Sure. You realize that a hygienic macro doesn't prevent you from
>>modifying things by consequence, right? All it does is require for
>>you to mean it; this caters to Python's phenonomenon of least
>>surprise. Imagine the following
>>
>>>>> i = 3
>>>>> j = 4
>>>>> mymacro i
>>>>> print j
>>5
>>
>>That would vex some folks.
>
>It would vex me in someone else's code unless there was a good reason for
>it and it was well documented.
>
>> But that's exactly what a non hygienic
>>macro can do.
>
>And it's exactly the behaviour I want a macro to be able to do.

>defmacro incj:
>   j = j + 1

>In a non-hygenic macro system it is easy to see what underlying Python
>code gets produced. In your system it isn't -- 

Perhaps only because it hasn't been explained well enough. The only
way the above macro definition could work without issuing an exception
is if j already exists in the global namespace. Using current python
forms, it would simply fail. That's because as a hygienic macro, it
has nothing but it's own namespace (as defined at the caller) to
manipulate, and without offering the explicit lexical closure, there's
nothing in it. This is the error you would get:

>>> j = j + 1
Traceback (most recent call last):
  File "<pyshell#1>", line 1, in ?
    j = j + 1
NameError: name 'j' is not defined
>>> 

With the following modification, it would work as expected:

defmacro incj (j):
	j = j + 1

C//



From threeseas at earthlink.net  Sat Feb  9 11:44:13 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 09 Feb 2002 16:44:13 GMT
Subject: Repost of original >: autocoding project proposal
References: <5512.785T1955T13695089threeseas@earthlink.net>
Message-ID: <6206.805T2889T7505727threeseas@earthlink.net>

Reposting as a matter of 20/20 hindsight in reference to what all has
evolved from this proposal. (has therre really been any productive
direction achieved....)

On 20-Jan-02 22:01:49 Timothy Rue <threeseas at earthlink.net> wrote:
>I'm posting this to this python group because a sample of what is
>mentioned below is done in python. And wouldn't it cool to have python
>used in an autocoding development environment?


>This posting is to introduce or suggest an open source software project
>to produce a core tool that will allow an autocoding environment to be
>developed, allowed to evolve in the open source software community and
>spirit of. I believe we can overcome many of the problems that proprietary
>autocoding systems inherently have (not to mention that existing
>autocoding systems appear to be field/domain specific limited and that
>it may be possible to beat this too.)

>First:
>Ghostscript PDF viewers for the following PDF links (if you don't already
>have a PDF viewer) http://www.cs.wisc.edu/%7Eghost/

>The following information represents what is probably the best of what is
>available thru the Web on the subject of autocoding. (via google)

>HIRTS DARP Working Group on Autocoding, 18th, 19th April 2000
>http://www.cs.york.ac.uk/hise/darp/pdf/autocoding.pdf
>(Brings up various issues which will help you focus in on what
>"autocoding" is and what some of the issues to solve, are.)

>The follow up to the above is:
>May 8th thru 10th, 2001 "DARP HIRTS Workshop" paper by Jakob Engblom:
>http://www.artes.uu.se/mobility/reports/hirts_report1.0.pdf
>(See pages 5-6 section 3.2.5 The Use of Tools in Aerospace)

>In summary, Though autocoding is being used to some extent, it is a
>future hope, since in general it has a bug density which is an order of
>a magnititude lower than manual code.  Point being is that this is
>leading edge stuff, an opportunity for OSS to shine.

>The above paper mentioned the SCADE autocoding product:
>http://www.esterel-technologies.com/scade/
>(See code generator part of Product overview, Benefits, Toolset Features)

>Autocoding as it applies to the medical industry:
>http://www.ahima.org/journal/coding/coding.0110.1.htm
>(Since it was mentioned in a paper above, know it's a product of a
>different nature.)

>To help show why I believe OSS efforts can shine when it comes to such a
>project as Autocoding:

>QinetiQ - Analysis of the Impact of Open Source Software
>http://www.govtalk.gov.uk/interoperability/egif_document.asp?docnum=430

>and From the Conference on the Public Domain, Nov. 9-11. 2001 at Duke Law
>School "Coase's Penguin, or, Linux and the Nature of the Firm" by Yochai
>Benkler: http://www.law.duke.edu/pd/papers/Coase%27s_Penguin.pdf

>It is also worth mentioning, to my undersandng, that the GNU efforts are
>becomming more modular in nature and there is also the Hurd that is
>modular or componet based from the ground up. This is a consistant and
>fitting direction in accord with an open autocoding development and use
>environment.

>OK, so although this project is not AeroSpace "Mission Critical", it does
>not hurt to make the quality target of the project to be of such high
>standards. Actually, what I believe is that the core (as mentioned at the
>beginning of this post) can be made to be of high quality itself, where
>the rest, the coding knowledge base or what ever you want to call the
>pre-autocode dictionary, will be of whatever quality it is created and
>improved to be (as is the way and spirit of the OSS community.)

>Where to start?

>Automating what was done manually requires the identification of, and
>ability to apply, the manual action set we use, but have the computer do
>it. A USPTO Published comment introducing these identified actions and
>suggestions of how they may be applied, including autocoding, is here:
>http://www.mindspring.com/~timrue/KNMVIC.html

>The bottom part of my home page regarding the
>"Virtial Interaction Configuration":
>http://www.mindspring.com/~timrue/

>An older paper on the Virtual Interaction Configuration:
>http://www.mindspring.com/~timrue/KC.html

>Why don't I do this shell and/or library myself?

>I don't consider myself a programmer having the experience and knowledge
>of better ways of coding somethings and as such I'm sure this core of
>functionality can be coded better and faster than what I could do, but
>there is what I have done, and that includes some python programming that
>can be used and run to show/communicate some of the concepts of the
>Virtual Interaction Configuration. And I can provide insight as to how it
>may be used for such things as autocoding.  Until recently, this past
>week, I wasn't aware "autocoding" was an actual goal and practice of
>anyone, though I have used the term for sever years now, and apparently,
>given the above HIRTS DARP papers, my perspectives and thoughts on the
>matter have been along the lines of what's been going on. Though I do
>believe The VIC as a core can solve some of the problems facing commercial
>autocoding packages (but this is something to get into later, when I can
>communicate but showing).

>Besides the VIC core, there is the pre-automation code base(s) to create
>for whatever programming language(s) people want to use. And that's
>something clearly beyond my ability to directly do, at least in the
>beginning. Besides, there are many other things the VIC can be used for
>besides autocoding. Consider it a tool for general automation...

>At any rate I do believe Autocoding is a worthwhile goal that the GNU
>community can shine at. And I think some of you, in consideration of the
>HIRTS DARP link contents and the USPTO comment, will too.

>If interested in helping, email me at mailto:threeseas at earthlink.net
>If interested in using such a tool, then say so here, let others know.



---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jaco at puxedo.org  Wed Feb 13 02:48:31 2002
From: jaco at puxedo.org (Jaco Greeff)
Date: Wed, 13 Feb 2002 09:48:31 +0200
Subject: Extracting bzip2 & gzip files
References: <mailman.1013534839.29210.python-list@python.org> <m3d6zaw66r.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3C6A1A4F.9060500@puxedo.org>

Martin v. Loewis wrote:

> It works fine for me, just the way you posted it 


This is strange. The file definately decompresses via the command-line. 
I might have a funny somewhere else - the file I'm trying to decompress 
is done straight after an HTTP download - maybe there is an open file or 
two lying around.

I know that the file is downloaded successfull, otherwise it would give 
the same error via the command-line. What I'll do is to invoke the 
function on a "static"/not donaloaded file which I know exists on the 
local drive.

Thanks for trying.

Greetings,
Jaco




From christophertavares at earthlink.net  Sun Feb  3 04:59:59 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Sun, 03 Feb 2002 09:59:59 GMT
Subject: "Private" attributes - a possible idea (maybe PEP-worthy).
References: <mailman.1012679645.17201.python-list@python.org>
Message-ID: <zS778.4551$Hb6.359000@newsread1.prod.itd.earthlink.net>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.1012679645.17201.python-list at python.org...
> [Chris Tavares]
> > How about changing the name mangling rule so that the current
> > number of base classes is added to the secret string?
>
> I'm not sure what "current number of base classes" means.  In
>
> class X(Y):
>     __pvt = 1
>
> what's the current number of base classes?  Keep in mind that
name-mangling
> is done at compile-time, and has only the static text of the module being
> compiled to examine, so the only feasible answer I can come up with is
"1".
> But that wouldn't help (if Y derived from another class named X which
> derived from Z, the count would be 1 on both incarnations of X).
>

Yeah, I see your point. I've been stuck with static languages too long. It
could probably be made to work somehow, but it's not worth the dev effort.

Another idea would be to generate a guid for each class, and use THAT in the
name mangling instead of the counter. That would guarantee uniqueness. Be
kinda hard to manually unmangle, though. Oh, and you'd need a guid generator
on non-windows platforms. But that's a really useful thing to have anyway.
:-)

-Chris





From leader730 at hotmail.com  Tue Feb 12 21:21:25 2002
From: leader730 at hotmail.com (Joseph Youssef)
Date: Tue, 12 Feb 2002 21:21:25 -0500
Subject: newbie asking for info about Python
Message-ID: <xkka8.2932$VX6.245554@weber.videotron.net>

Hello, I know nothing about Python but before I start learning I would like to know what Python is usually "used" for, I mean I don't think it's made to write some kinda application or something, so what exaclty is it used for?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020212/e59db7a3/attachment.html>

From loewis at informatik.hu-berlin.de  Mon Feb 25 05:56:53 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 11:56:53 +0100
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com> <m3eljb9tcv.fsf@mira.informatik.hu-berlin.de> <mailman.1014602200.11854.python-list@python.org> <j4lmdhn9to.fsf@informatik.hu-berlin.de>
Message-ID: <j4wux1lthm.fsf@informatik.hu-berlin.de>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:

> Do you know whether I can reproduce this on a non-Russian Windows
> installation, eg. on a German XP installation?

I could answer this question myself. If you change the system
locale/codepage to "Russian" (for non-Unicode applications), and also
change the keyboard layout, then Python 2.2 will still display Latin
characters in response to keypresses. If Tcl finds its coding
directory, it will display the proper Cyrillic characters.

BTW, if you only change the keyboard layout (and leave the system code
page as cp1252), Tcl will display Latin characters no matter whether
it finds its encodings directory or not.

Regards,
Martin


From jim at jim.jim  Fri Feb  8 16:46:56 2002
From: jim at jim.jim (jimh)
Date: Fri, 8 Feb 2002 13:46:56 -0800
Subject: TRYING to like python, but...
Message-ID: <a41h0m$ear$1@web1.cup.hp.com>

I untarred, configured and built python 2.1.2.  It all seems to be working
fine.

I wanted to try out the xml.dom stuff, so I looked over the Python Library
Reference sections on xml.dom and tried one of the examples.  This is what
happened:

   xml.sax._exceptions.SAXReaderNotAvailable: No parsers found

Well, THAT'S helpful.

Nothing in Setup about xml.dom or xml.sax, so I assume that Setup is ok.
Besides, I WAS able to import and start using xml.dom.

So I get online and start searching around for this error message.  Find
some vague references to needing "expat" (whatever that is).  Ok, I remember
vaguely something about expat in Setup, so I go back to Setup and see that
TO USE A STANDARD PYTHON LIBRARY, I need to DOWNLOAD ADDITIONAL code (expat)
from SOME PLACE ELSE.

So I download it, compile it, update EXPAT_DIR in Setup to point to it (just
like the comments in Setup say).

So I ./configure again, and Make again, and, guess what??? IT STILL DOESN'T
WORK!!!  (xmlparse.h: no such file).

so PLEASE PLEASE PLEASE tell me that I don't really have to go through ALL
OF THIS CONFUSING MESS just to get a STANDARD python library to work.
(xml.dom IS in the "Python Libary Reference", but there is NO mention of all
these additional steps required to actually USE it).

To be honest, if python is really this difficult to use, I am going to try
something different.

Someone please help me out here.

Jim Haffey







From shalehperry at attbi.com  Thu Feb 28 01:17:11 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 27 Feb 2002 22:17:11 -0800 (PST)
Subject: Regular expression problem
In-Reply-To: <YXff8.148576$dh.38567308@typhoon.nyroc.rr.com>
Message-ID: <XFMail.20020227221711.shalehperry@attbi.com>

On 28-Feb-2002 Asheesh Laroia wrote:
> I've been trying to use sgmllib, actually, to delete all the other tags.
> 
> It just doesn't handle the <@ [...] > condition well.  It refuses to
> parse it, treating it as text.
> 

The reason is this:

starttagopen = re.compile('<[>a-zA-Z]')
tagfind = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9]*')

near the top of sgmllib.py.

Changing them in your code will allow the parser to understand the tag. 
However there is another problem which requires more work.  When a tag is found
the parser tries to run 'start_' + tag.  start_ at Trap() is not a valid python
name.  You could redefine the function which calls the handlers so that it
looks for perhaps start_atTrap().  This would allow you to use the SGMLParser
for all of your parsing needs, but may also be overkill for the problem.



From rony.steelandt at bucodi.com  Mon Feb 18 09:04:03 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 18 Feb 2002 06:04:03 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com>
Message-ID: <b34d3835.0202180604.7a7605ab@posting.google.com>

Well i did try it with the same error !

Anyone has an idea ??

rony.steelandt at bucodi.com (Rony) wrote in message news:<b34d3835.0202180007.78d987fd at posting.google.com>...
> This is interresting ... never sawn this one
> I'll try it immediatly
> 
> Thanks
> 
> 
> Michael Kelly <mkelly2002NOSPAM at earthlink.net> wrote in message news:<rfrv6ucpn8osftj7e2mbj34e1o46271v3q at 4ax.com>...
> > On 16 Feb 2002 08:57:15 -0800, rony.steelandt at bucodi.com (Rony) wrote:
> > Something I found in Python using COM,
> > (after Mark pointed it out to me that is)
> > is that if you have "out" parameters other than
> > the retval you should use them in multiple assignment.
> > I wrote a wrapper for Delphi TIniFile that got a similar
> > error only in Python.  It turned out I was passing a
> > param to procedures with "out" parameters like:
> > 
> > myIniFile.ReadSection('section', sectionlist)
> > 
> > when I should have done:
> > 
> > sectionlist = myIniFile.ReadSection('section')
> > 
> > In VB or VC++ I would have to pass the 2 params
> > in the parens as expected.
> > 
> > IOW, since Python has multiple assignment and
> > VB and VC++ do not, any out param should be
> > assigned to rather than passed as a param.
> > 
> > Anyway the error was something to do with
> > a type error so I'm pretty sure you're getting
> > burnt by the same thing.


From jason at jorendorff.com  Wed Feb  6 13:24:21 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 12:24:21 -0600
Subject: Fun with lambda and map
In-Reply-To: <a3og4o$6l0$1@news1.xs4all.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLEEMKDAAA.jason@jorendorff.com>

ivo wrote:
> (btw, I got a solution by email that is pure lambda, but I
> first have to figure out how it works)

I meant to send it to the list.

fac = lambda n: n(lambda c: lambda q:
                  q(succ(c(true)))(mul(c(true), c(false)))
                  )(lambda q: succ(zero))(false)

Be careful; it's O(n!).  <wink>

my-code-runs-real-fast-it's-that-darn-to_int-function-ly y'rs,
## Jason Orendorff    http://www.jorendorff.com/



From phr-n2002a at nightsong.com  Wed Feb 27 16:32:45 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 13:32:45 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com> <7xk7sznsoa.fsf@ruckus.brouhaha.com> <3C7D3B2C.80A97BA2@ccvcorp.com>
Message-ID: <7x664i8vb6.fsf@ruckus.brouhaha.com>

Jeff Shannon <jeff at ccvcorp.com> writes:
> > Also, there's no way to make sure that the "del temp" at the end of
> > a block actually gets run, unless you put try...finally around it,
> > which is a big mess.
> 
> I may be in way out of my depth here, but I don't think that this is
> true.  If you explicitly delete an object, then that name should be
> guaranteed to be unbound.  What's *not* guaranteed, is whether or
> not an unreferenced object's __del__() method gets called -- the
> object may still exist in memory, unreachable, somewhere.  However,
> that name is *not* going to reference it properly.
> 
> At least, that's my understanding of it.

What I mean is if you say

  if x:
    temp = frob()
    mumble(temp)
    foo()
    bar()
    del temp

and the foo() or one of the other calls raises an exception that's caught
by some outer block in the same function, then temp hasn't been deleted,
and in fact the code could even depend on it not being deleted.  So del
doesn't give you the same effect as local scope.


From gimbo at ftech.net  Thu Feb 28 11:41:44 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Thu, 28 Feb 2002 16:41:44 +0000
Subject: Status of PEP's?
In-Reply-To: <a5lk0a$8enfq$1@ID-11957.news.dfncis.de>
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl> <a5lk0a$8enfq$1@ID-11957.news.dfncis.de>
Message-ID: <20020228164144.GA6305@andy.tynant.ftech.net>

On Thu, Feb 28, 2002 at 07:51:35AM -0800, Emile van Sebille wrote:
> 
> > for i in 6: print i
> > for i in 6,7: print i
> > for i in 6,7,8: print i
> >
> > You wouldn't think the first line does something different, would you?
> 
> Is that really any different from what we have now:
> 
> for i in "Spam": print i
> for i in "Spam","and": print i
> for i in "Spam","and", "Eggs": print i

Yes, it really is.  "Spam" is a sequence, whereas 6 isn't.

> I find the best argument for adding an iter method to ints is that it
> allows me to say:
> 
>     for ii in iter(len(mylist)):
> 
> which I find _much_ less confusing than saying:
> 
>     for ii in range(len(mylist)).
> 
> The first reads as:
> 
>     'for ii iterating over the len of mylist'
> 
> whereas the second:
> 
>     'for ii in the list of numbers starting with 0 to 1 less than the
> len of mylist'

I find the second much more explicit and much less ambiguous, and
therefore far preferable.

To me, "the len of mylist" is the number of items in the list - a
single value, not some other list.  If I don't happen to know what
range() does because I'm a newbie, it's easy to find out and get a
clear, unambiguous, unsurprising answer.

> If this PEP is added, the coincidental effect of ints invoking their
> iterators in appropriate contexts is, in my mind, no different than what
> we do today with strings.
> 
> However, the clarity and intent conveyed by allowing this, particularly
> for newbies, more than offsets the gotcha's, which, once learned, is the
> same set of gotcha's in place for strings.

I disagree - strings are sequences, integers aren't.

-Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From cmaloney at physics.ucsb.edu  Thu Feb 28 12:17:54 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: Thu, 28 Feb 2002 09:17:54 -0800
Subject: Scripting C++ -- 2 -- a more concrete example.
Message-ID: <3C7E6642.2030202@physics.ucsb.edu>

Thanks to all for the great leads to wrapper generators for 
multi-language projects, but I'm still confused about my initial 
question, so I'll try to be a bit more concrete.

Suppose that I have a class Crystal, and a class Atom.  I'd like to be 
able to do the following:

-----------------
Crystal c;
...
Atom a1;
Atom a2;
...

c.addAtom(a1);
c.addAtom(a2);
...
c.computeEnergy();
---------------------

Atom is a simple enough data structure that default copy works.

Now here comes the concrete question.  If I want to use this class 
library from a scripting language (e.g. Python) what are the pros and 
cons of using wrapper code (e.g. Boost.Python or SILOON from lanl) vs CORBA.

To me the main differences are:

1) Interface generation.
If I use CORBA, I would have to write IDL interfaces on top of the 
library, since there is no reverse mapping from C++ as there is in java. 
  It seems contrary to the spirit to write the IDL *after* I've written 
the class library -- indicating a lack of regard to issues of 
distributedness in designing the class library. (see my comment at the end)

If I use wrapper generators, it seems like there is a chance (e.g. using 
SILOON/PDT or BOOST/PDT) that it would be possible to have the wrapper 
code automatically generated.

This is a major boon as I would not have to convert 
std::valarray<double> to sequence<double> or whatever monkeying I would 
have to do.  I can imaging this getting pretty hairy to the point of 
unworkable.

2) Marshalling and Objects-by-value.
When the "Crystal" computes its energy it should *NOT* have to marshall 
calls to get at the atom data structures.  This is an operation that 
will be repeated about 2.85 bijillllion times.  In CORBA, this is no 
problem -- I would either have Crystal::addAtom take a valuetype (atoms 
are easy enough to serialize), or I would use some "co-location" trick 
so that the ORB knows that the atom *really* lives in the same address 
space as the crystal.  Preferrably the former.

If using a wrapper generator (e.g. SILOON) I don't see a way around the 
marshalling of the call to get at the atom data structure.

Of course, this would be taken care of by a revamp of the design of the 
class library... maybe making the crystal an atom factory that produces 
new atoms in its own address space.  But I would like to avoid thinking 
about multi-language and distribution issues when designing the core of 
the library.  Is this dream just simply self-contradictory?

Thanks all,
Craig



From magnus at thinkware.se  Thu Feb 21 07:16:48 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 13:16:48 +0100
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <258fd9b8.0202201701.5f8e4e42@posting.google.com> <7xpu2zsjii.fsf@ruckus.brouhaha.com>
Message-ID: <3C74E530.2030604@thinkware.se>

Paul Rubin wrote:

> But there's no reason sort couldn't sort in place and return the list.
> d.keys() already makes a temporary list with the keys in it after all.
> The temporary list can be sorted in place without disturbing the
> dictionary.


But the dictionary is just a special case. Sort sorts any list.

You should ask Guido, but I think it's made like this to reduce
confusion. It DOES an in-place sort. If

 >>> a = [1,5,3]
 >>> b = a.sort()
 >>> print b
[1, 3, 5]

worked, it would not be obvious that both a and b now refered to the
same, now sorted list. Like this:

>>> a = [1,5,3]

 >>> b = a

>>> b = b.sort()
>>> print b
[1, 3, 5]

Most people would probably think that it was equivalent with:

 >>> a = [1,5,3]
 >>> b = a[:]
 >>> b = b.sort()
 >>> print b
[1, 3, 5]

But if it was implemented like that, all sorts would use much more
memory.



From oren-py-l at hishome.net  Fri Feb  1 09:40:50 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Fri, 1 Feb 2002 09:40:50 -0500
Subject: Enhanced Generators - the difference between sources and iterators
In-Reply-To: <a3at9p$2he$1@bob.news.rcn.net>
References: <a3at9p$2he$1@bob.news.rcn.net>
Message-ID: <20020201144050.GA28252@hishome.net>

On Thu, Jan 31, 2002 at 02:53:15AM -0500, Raymond Hettinger wrote:
> At the suggestion of  Kragen Sitaker, the arguments for xmap(), xzip(), and
> indexed() are now called collections rather than sequences.

I prefer the term source. 

iterable: 
   anything for which iter() is valid.

iterator: 
    an iterator is a temporary object that may be iterated once
    x is an iterator if iter(x) is x

source: 
    A source is a persistent object that may be iterated multiple times.
    Python containers are sources.  Returns a fresh, independent iterator 
    for each iter() call

generator: 
    An iterator created by calling a generator function.  

generator function:
    A magic function returning a generator.  A generator function is NOT
    a source because it does not have an __iter__ method.  However,
    a generator function with no arguments is a convenient way to implement
    an source's __iter__ method.

My implementation of xmap, xfilter and xzip returns a source, not an
iterator.  It can be iterated multiple times, just like the output of map,
filter or zip.

The logical opposite of a source is a destination.  Destinations should
be able to produce a temporary object mirroring an iterator.  I call such 
an object a sink (PLUG: see http://tothink.com/python/dataflow)

> Addressed Alex Martelli's question about why .submit() was used instead of
> overloading .next()

Why invent a new protocol when on already exists in the Python library? the
HTMLparser object has the method .feed() to implement pushing data to a
sink.


I'm not sure what you expect two-way generators to be used for.  If the two
directions of communications are used as the upstream and downstream
directions of a filter in stream they should be independent, not using the 
same generator object.  Are they meant for bidirectional communication 
between two entities?  Dialogues are rather rare as a form of interconnection 
between parts of the same program.  Dialogues are common in communication 
but I don't believe that was your intention.

I'm in favor of one-way consumer functions to mirror generator functions,
but not two-way generators.  

	Oren



From brueckd at tbye.com  Thu Feb 28 11:01:04 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 28 Feb 2002 08:01:04 -0800 (PST)
Subject: Rotor in Jython?
In-Reply-To: <7x8z9e45b8.fsf@ruckus.brouhaha.com>
Message-ID: <Pine.LNX.4.33.0202280748000.16013-100000@ns0.tbye.com>

On 27 Feb 2002, Paul Rubin wrote:

> "Scherer, Bill" <Bill.Scherer at VerizonWireless.com> writes:
> > I've a need to perform rotor decryption in Jython.  A quick
> > search didn't reveal anything to help.  I've begun coding it in
> > Python from rotormodule.c. My C knowledge is not strong, so there
> > are spots that are surely not right.
>
> I hope you're not the rotor module for computer security.  That cipher
> is nowhere near as safe to use as a modern one like AES.

Actually, that's not necessarily true. AES and friends are
cryptographically much stronger than rotors for key lengths of a given
size, but with rotors the computational cost of encryption/decryption is
unaffected by key length, so if you have a good source of nearly random
bits to use in making long keys then you can use rotors for encryption
that is, for all intents and purposes, unbreakable.

Put another way: sorting by key length, AES beats rotors hands down.
Sorting by computational cost, rotors always win because you can just make
the keys longer (without having to reimplement your crypto).

-Dave




From shenwang at ameritech.net  Sun Feb 24 15:08:23 2002
From: shenwang at ameritech.net (Shen Wang)
Date: Sun, 24 Feb 2002 20:08:23 GMT
Subject: Is there a python parser generator that is similar to JavaCCfrom webgain?
References: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com> <m38z9jpz4a.fsf@mira.informatik.hu-berlin.de> <B89D9263.3864%pereira@cis.upenn.edu> <m3it8n9ti8.fsf@mira.informatik.hu-berlin.de> <B89E5855.38AA%pereira@cis.upenn.edu>
Message-ID: <XKbe8.9545$pC1.1637045@newssrv26.news.prodigy.com>

Thanks, I will take a look.


"Fernando Pereira" <pereira at cis.upenn.edu> wrote in message
news:B89E5855.38AA%pereira at cis.upenn.edu...
> On 2/24/02 3:25 AM, in article
m3it8n9ti8.fsf at mira.informatik.hu-berlin.de,
> "Martin v. Loewis" <martin at v.loewis.de> wrote:
>
> > Fernando Pereira <pereira at cis.upenn.edu> writes:
> >
> >> JavaCC takes grammar specs (lexical patterns and rules) and produces a
> >> parser (a Java class) using a LL(k)-type of rule compilation (that is,
> >> deterministic recursive descent with lookahead).
> >
> > Ok, I guess the closest thing for Python like this is YAPPS:
> >
> > http://theory.stanford.edu/~amitp/Yapps/
>
> Close, but more limited. JavaCC allows k lookahead for k > 1, Yapps just k
=
> 1.
>
> -- F
>




From phr-n2002a at nightsong.com  Wed Feb 27 23:14:46 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 20:14:46 -0800
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net>
Message-ID: <7x4rk244zt.fsf@ruckus.brouhaha.com>

netzapper at magicstar.net (A. Jones) writes:
> It's my second day Pythoning and I've written this code to encypher
> text (or files, ostensibly) into two pieces, neither of which can be
> used to determine the message.
> 
> Any problems pointed out, clunky lines spotted, or security problems
> would be most appreciated.  Also, if you've seen something like this,
> if you'd tell me, please.

What the heck?  There's no encryption function in there anywhere.

If you want convenient low-to-medium security encryption, use the
rotor module already supplied Python.

If you want higher security, use one of the various crypto modules
available through Vaults of Parnassus or elsewhere.


From aleax at aleax.it  Fri Feb  1 22:23:12 2002
From: aleax at aleax.it (Alex Martelli)
Date: Sat, 02 Feb 2002 03:23:12 GMT
Subject: Error opening file ! Help, please
References: <Xns91A9CA356801ilen712l@193.70.192.103>
Message-ID: <AYI68.19256$3J1.608629@news2.tin.it>

gimar wrote:

> Sometimes when I try to create an output file executing:
> 
> f=open('\paul','w')
> 
> I receive the following error message:
> "TypeError: object of type 'list' is not callable".
> This has no sense for me. Can you explain me what's happening ?

Looks like you're rebinding name 'open' to some list object.  If
you're sure you have no "open = ..." around, likeliest mistake on
your part is probably a "from X import *" for some value of X.


Alex



From starx at pacbell.net  Sun Feb 10 06:47:05 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 10 Feb 2002 11:47:05 GMT
Subject: Bootstraping 'frozen' python... anyone done this before?
Message-ID: <Xns91B1267CA69F8RASXnewsDFE1@130.133.1.4>

I've been wondering how to distribute an application program in python 
for a while now -- most of the disutils and installing and information 
seems to assume you're working with libraries. I was also wondering how 
to make a standalone version (for people who don't have all the right 
versions on hand and don't want to get them, or people who just don't 
have python), and leave the code in the open, and how to make it 
reasonably sized. I think i've got a solution that works for me now (and 
realizing it was sort of a 'oh, duh' experience). Just make a standalone 
out of a simple script with all of the libs you need and run the real 
program from there.

Another benefit of this is that it seperates all of the supporing crap 
(which totals at 11 megs, though i should try and get rid of all the TK 
stuff and it will be around 7), which only ever changes when i upgrade 
my development libraries, and the small, continully changing code that 
actually defines my program, which is now at about 370k including a few 
tile-textures.

Here's a copy of the bootstrapping loader-script i came up with, even 
with minimal runtime configuration abilities. (It's so minimal because i 
didn't want to import anything at all to do it, and i thought it should 
be exceedingly simplistic at the time :)

This works great with the Macmillian Installer module-thing, but you 
could probably swing it with py2exe too.

Ok, no more posting w/o more sleep for now =)

# Project Includes
import wxPython.wx
import PyCrust.crust
import OpenGL
import OpenGLContext.wxcontext
import Image
import random
import crements
# Whatever you need to have on hand in the finished program.
# (Could be other ways if you don't want to import them immediatly,
# but i load them all before doing anything anyway)

try:
    cfg_file = open("execframe.cfg")
    bootscript_name = cfg_file.readline()
    bootscript_name = bootscript_name.strip()
    cfg_file.close()
except IOError:
    print "*** No execframe.cfg, using default boot-script name ***"
    bootscript_name = "sproject.py"

print '*** Loading boot-script: ', bootscript_name, '***'
try:
    execfile( bootscript_name, {'__builtins__':__builtins__,
                                '__name__':'__main__'} )
except IOError:
    print '*** File error, boot-script not found ***'

print '*** Press Enter to Exit ***'
raw_input()

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From rdsteph at earthlink.net  Sun Feb 10 20:33:39 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Mon, 11 Feb 2002 01:33:39 GMT
Subject: Are there any Python books or sites with good exercises?
References: <g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r@4ax.com> <Xns91B2604B33F77andy47halfcookedcom@203.109.250.24>
Message-ID: <3C672052.1D284C30@earthlink.net>

Andy Todd wrote:
> 
> Stefan Franke <franke at ableton.com> wrote in
> news:g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r at 4ax.com:
> 
> > (WRT your personal definition of 'good', of course). I'm going to
> > prepare a Python introductory course for experienced programmers
> > and I'm looking for handy exercises which hopefully provide them
You might also try out the tutorials and book reviews at
http://www.awaretek.com/plf.html
The beazley advanced tutorial is especially good



> > numerous Aha! effects.
> >
> > Thanks,
> > Stefan
> >
> >
> >
> 
> http://diveintopython.org/
> 
> Regards,
> Andy
> --
> Contents free posts a speciality


From richard.papworth at bbsrc.ac.uk  Thu Feb 21 10:59:05 2002
From: richard.papworth at bbsrc.ac.uk (Richard Papworth)
Date: 21 Feb 2002 07:59:05 -0800
Subject: mxDateTime in Python2.2 on Tru64
Message-ID: <695dd93e.0202210758.3b64f79a@posting.google.com>

Hi

After upgrading from Python 1.52 I'm having trouble getting mx
Extensions to work with Python 2.2.

Compilation went OK (although config.h was missing from the Python
includes directory). However, when I try to load up the DateTime
module I run into problems:

Python 2.2 (#2, Feb 20 2002, 11:44:40) 
[GCC 2.95 19990728 (release)] on osf1V4
Type "help", "copyright", "credits" or "license" for more information.
>>> from mx import DateTime
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/nfs3/users/postgres/usr/local/lib/python2.2/site-packages/mx/DateTime/__init__.py",
line 8, in ?
    from DateTime import *
  File "/nfs3/users/postgres/usr/local/lib/python2.2/site-packages/mx/DateTime/DateTime.py",
line 441, in ?
    MaxDateTime = DateTime(sys.maxint / 366 - 1,12,31)
OverflowError: signed integer is greater than maximum
>>> 

The other mx extensions seem to be OK. Does anyone know what's going
wrong?

Thanks in advance for any help.

Richard


From sholden at holdenweb.com  Mon Feb 11 09:09:18 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 09:09:18 -0500
Subject: Question on if, while, for
References: <e959e7b.0202110545.3dff102@posting.google.com>
Message-ID: <8gQ98.150037$KM2.6737812@atlpnn01.usenetserver.com>

"Fred" <c at tech.usd253.org> wrote in message
news:e959e7b.0202110545.3dff102 at posting.google.com...
> Hello,
>
> I have written a simple program that I want to simplify. I used if and
> elif. But I am sure this can be simplified with for or while. Or what
> if I wrote a function to do the testing?
>
> Please DO NOT re-write this for me I want to figure it out for myself.
> I would just like to know what is the BEST way to do it:
>
> 1) For loop (nested??)
> 2) While loop
> 3) Function
> 4) Leave it alone it's fine??
>
> I am a rookie here learning Python with my son, very impressed so far!
> Fred
> -----------------------------------------------------------------------
>
> num = input("Enter the numerator: ")
> den = input("Enter the denominator: ")
>
> if num > den: #Test for improper fraction.
>     i = num/den #This will only return the whole part, it discards
> the remainder.
>     j = num%den #This will return only the remainder.
>
>     if j != 0: #If j is equal to 0 you will have a whole number, no
> remainder,
> #skips down to else.
>
> #If j is not equal to 0, begin test to find lowest terms of the
> fraction.
>         if j%9 == 0 and den%9 ==0:
>             a = j/9
>             b = den/9
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
> #Prints the value of i, the whole part, reduces the fracton to lowest
> terms and prints it.
>
>         elif j%8 == 0 and den%8 ==0:
>             a = j/8
>             b = den/8
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%7 == 0 and den%7 ==0:
>             a = j/7
>             b = den/7
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%6 == 0 and den%6 ==0:
>             a = j/6
>             b = den/6
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%5 == 0 and den%5 ==0:
>             a = j/5
>             b = den/5
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%4 == 0 and den%4 ==0:
>             a = j/4
>             b = den/4
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%3 == 0 and den%3 ==0:
>             a = j/3
>             b = den/3
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%2 == 0 and den%2 ==0:
>             a = j/2
>             b = den/2
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>         elif j%1 == 0 and den%1 ==0:
>             a = j/1
>             b = den/1
>             print "You entered an improper fraction, simplified it is:
> %d %d/%d" % (i,a,b)
>
>     else:
>         print "You entered an improper fraction, simplified it is: %d"
> % (i)
> #You have a whole number, which is stored in i, this prints it.
>         #If you typed in 4/2 you would be here.
>
>
> #This section is for a proper fraction.
> elif num%9 == 0 and den%9 ==0:
>     a = num/9
>     b = den/9
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%8 == 0 and den%8 ==0:
>     a = num/8
>     b = den/8
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%7 == 0 and den%7 ==0:
>     a = num/7
>     b = den/7
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%6 == 0 and den%6 ==0:
>     a = num/6
>     b = den/6
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%5 == 0 and den%5 ==0:
>     a = num/5
>     b = den/5
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%4 == 0 and den%4 ==0:
>     a = num/4
>     b = den/4
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%3 == 0 and den%3 ==0:
>     a = num/3
>     b = den/3
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%2 == 0 and den%2 ==0:
>     a = num/2
>     b = den/2
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),
>
> elif num%1 == 0 and den%1 ==0:
>     a = num/1
>     b = den/1
>     print "The fraction you entered is %d/%d in lowest terms." %
> (a,b),

Sorry about my mail program wrpaaing your code. Just some basic hints, then
it's up to you.

1. You could make the whole program repeat by using a "while 1:" loop which
you break out of when the user hits RETURN without making any inputs. This
loop would effectively go around the logic you already have.

2. Whenever you see the same code repeated several times with minor
variations you should think about building a loop that repeats the same
code, varying things as necessary.

3. There are arithmetical algorithms for finding the "LCD" (lowest common
denominator) of two numbers. If you worked out this value on the numerator
and denominator entered by the user, you would easily find whether the
fraction was in its lowest terms.

4. As you build up experience in programming, you will often find better
ways to do things you have already done in inefficient ways. Don't worry in
the least about this, it's a normal part of growth as a programmer.

5. You are to be congratulated on building a working program of any kind. If
more parents took a similar interest in their kids' education it wouldn't
hurt. You just don't have to mind when the kids start writing code five
times as fast as you!

comp.lang.python is a pretty friendly group overall. Welcome, enjoy.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From josh.winters at webstream.net  Mon Feb 18 18:09:13 2002
From: josh.winters at webstream.net (josh.winters at webstream.net)
Date: Mon, 18 Feb 2002 18:09:13 -0500
Subject: We would like to possibly get some info on your company
Message-ID: <E16cwuH-0001oN-00@mail.python.org>

Hello, 

We would like to possibly get some info on your company in an effort to explore the ways that we might be able to work together. We may be able to save you money and offer you the benefits of our reseller program.

We have been developing and hosting web sites since 1997. We offer design, programming, hosting and webcasting and videoconferencing. We support Linux, NT and AS400. 

Please forward this to the proper party, or respond to the address below. You can also visit our web site at http://webstream.net for more information on our services. 

If by e-mail: 
josh.winters at webstream.net 

If by mail: 
WebStream Internet Solutions 
Outsourcing/Purchasing 
2200 W.Commercial Blvd. Suite 204 
Ft. Lauderdale, FL 33309 USA 

Thank you very much. 

Sincerely, 

Josh Winters 
josh.winters at webstream.net 
http://webstream.net 
Design * Programming * Hosting * WebCasting * Since 1997




From m.faassen at vet.uu.nl  Fri Feb  8 17:32:47 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Feb 2002 22:32:47 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu> <a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org>
Message-ID: <a41jmf$cd2$1@newshost.accu.uu.nl>

Cliff Wells <logiplexsoftware at earthlink.net> wrote:
> On 8 Feb 2002 20:23:50 GMT
> Martijn Faassen wrote:

>> Besides, even if I were arguing with the fellow I might've been
>> doing it for amusement. But this type of person doesn't seem quite
>> what to make of me most of the time, so they just don't respond..

> Arguing with irrational oddballs on usenet is today's equivalent of the
> Rubik's Cube.

I never managed to solve those either. Is it still equivalent for those who
do mananage to solve the cube?

Regards,

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


From m.faassen at vet.uu.nl  Fri Feb  8 15:26:21 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Feb 2002 20:26:21 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <mailman.1013159835.28909.python-list@python.org>
Message-ID: <a41c9d$bf4$2@newshost.accu.uu.nl>

Oleg Broytmann <phd at phd.pp.ru> wrote:
> It is amazing how fast a dozen of lamers, few trolls and a free software
> advocate can destroy the good old friendly spirit of c.l.py.

Am I the only one who noticed you including this 'free software advocate'?
I remember a few amusing threads with you in it. :)

Come on, folks, the Rue threads have been amusing! I mean, the guy thinks
there's a conspiracy theory, such fertile territory for PSU postings!

Regards,

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


From phr-n2002a at nightsong.com  Sun Feb 17 08:01:47 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 05:01:47 -0800
Subject: Editable strings anyone?
Message-ID: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com>

I'm wondering how difficult it would be to add an editable (mutable)
string class to Python 2.2 implemented in C under the new type/class
unification scheme.  Right now, writing a text editor in Python seems
difficult to do in any natural way.  I'm messing with Amit Patel's very
cool Proxy3 web proxy (which does HTML editing on the fly) and find
myself really wanting something like this.

I guess the editable string class it would support the current string
operations and list operations.  The underlying implementation should
support fast editing operations (insertions and deletions of slices)
even for very large (multi-megabyte) strings, at least in the usual
case where consecutive editing operations happen in the same region of
the string.  The usual method of representing the editable string as
two contiguous regions of a chunk of memory separated by a gap, so you
can add or delete characters by growing or shrinking the gap, and
moving the gap as necessary, would be fine.

Example:

    e = editable_string('potato')   => potato
    e.insert(0,'baked ')            => baked potato
    x,y = re.search('ak',e).span()  => 1,3
    e[x:y] = 'oil'                  => boiled potato

Does this sound like a reasonable design?  Does it fit in reasonably
with the CPython implementation?  Would it require a lot of hacking to
existing string functions like the re module?

Java has a StringBuffer class which Java programmers use idiomatically
whenever they want to build a big string from smaller ones.  Python
surprisingly lacks a counterpart to StringBuffer.  Python programmers
tend to say either a=b+c+d+e+f (potentially taking quadratic time in the
total number of characters) or some kludge like a=''.join(b,c,d,e,f).
Editable strings would be even more general than StringBuffer.


From gerson.kurz at t-online.de  Mon Feb  4 08:22:55 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 13:22:55 GMT
Subject: Problem while uploading a binary file using cgi module
References: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU>
Message-ID: <3c5e8a67.13140484@news.t-online.de>

On Sun, 3 Feb 2002 21:49:59 -0500, "J. Wang" <jw9 at cse.buffalo.edu>
wrote:

>
>My problem is that it works fine with a text file, but failed everytime
>with a binary file such as jpeg file, even a very small (5k) binary file.

Rather than writing

#!e:/Python22/python.exe

use

#!e:/Python22/python.exe -u

IIRC there is an entry in the FAQ for this, although I cannot find it
right now.


From ahimsa at inext.co.za  Tue Feb 12 00:52:48 2002
From: ahimsa at inext.co.za (ahimsa consulting)
Date: Tue, 12 Feb 2002 07:52:48 +0200
Subject: Possible dumb query
References: <E16aVdi-0005rV-00@mail.python.org>
Message-ID: <000b01c1b389$8509fbc0$0100000a@ahimsa>

Chris & Wayne

Thanks for your rapid response. I'll check it out tonight as you suggest.
The directory doesn't bug me, I was just curious about its role.

Cheers guys!!

Andrew

> Message: 15
> From: "ahimsa consulting" <ahimsa at inext.co.za>
> To: <python-list at python.org>
> Subject: Possible dumb query
> Date: Tue, 12 Feb 2002 07:24:44 +0200
>
> Greeting Pythonites
>
> This might qualify as one of those "really dumb questions", but:
>
> Under Linux RH 7.2 when doing a search for which python ("$ which python")
I
> have discovered that I have one entry for python as /usr/bin/python and
one
> entry for python2.2 as /usr/bin/python2.2. My question is this: if I were
to
> delete the directory /usr/bin/python what impact - if any - will it have
on
> any of the libraries/ etc. that I would require to use python2.2
>
> Once again, my apologies if this is a dumb query, but gotta learn somehow.
>
> Much obliged
>
> Andrew
>





From fruhstuck at rutherfurd.net  Thu Feb 14 23:41:33 2002
From: fruhstuck at rutherfurd.net (Oliver Rutherfurd)
Date: Thu, 14 Feb 2002 23:41:33 -0500
Subject: '/' not allowed in cookie name?
In-Reply-To: <a4hffl$8m$1@nntp9.atl.mindspring.net>
Message-ID: <016201c1b5db$0cb5c970$0200a8c0@diego>

> -----Original Message-----
> From: python-list-admin at python.org 
> [mailto:python-list-admin at python.org] On Behalf Of Andrew Dalke
> Sent: Thursday, February 14, 2002 5:58 PM
> To: python-list at python.org
> Subject: Re: '/' not allowed in cookie name?
> 
> So token cannot contain "/", which means "1/2" is not a valid 
> cookie name.
> 
> As to if some clients and servers let you use illegal cookie 
> names ... that's a different thing.

Thanks for your help Andrew.  I figured someone would have detected it
before, if Cookie.py was wrong, but better to ask than assume.

Regards,
-Ollie

Ollie Rutherfurd
oliver at rutherfurd.net




From zvonkoka at yahoo.com  Sat Feb 23 21:50:58 2002
From: zvonkoka at yahoo.com (zvonko)
Date: Sun, 24 Feb 2002 03:50:58 +0100
Subject: cannot start pythonwin
References: <a58k0o$31ur$1@as201.hinet.hr> <3C7835EB.4040809@skippinet.com.au>
Message-ID: <a59kcd$4fsn$1@as201.hinet.hr>

Thank you!

i reinstalled python using admin install and pythonwin works perfectly now.



"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C7835EB.4040809 at skippinet.com.au...
> zvonko wrote:
> > i've just installed python 2.2 and the pythonwin extensions.
> > however, when i try to start pythonwin, i get the following error:
> >
> >
> > The application cannot locate win32ui.pyd (1157)
> >
> > but the file is in the same directory where the pythonwin is.
>
> This is probably because you installed Python itself for the current
> user, rather than for the entire machine.  In this scenario, win32all
> usually fails to install right at the end.
>
> The workaround is to move python22.dll from your Python directory into
> your SYSTEM32 directory, then reinstall win32all.
>
> Mark.
>




From just at xs4all.nl  Sun Feb 24 06:53:24 2002
From: just at xs4all.nl (Just van Rossum)
Date: Sun, 24 Feb 2002 12:53:24 +0100
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones> <a58885$5acoc$1@ID-12869.news.dfncis.de> <zVPd8.4096$jC3.163217@news1.east.cox.net> <a5acho$5j86i$1@ID-12869.news.dfncis.de> <slrna7hkcg.i6a.quinn@baht.ugcs.caltech.edu>
Message-ID: <just-74F5E1.12532424022002@news1.xs4all.nl>

In article <slrna7hkcg.i6a.quinn at baht.ugcs.caltech.edu>,
 quinn at baht.ugcs.caltech.edu (Quinn Dunkan) wrote:

> On Sun, 24 Feb 2002 17:41:17 +0800, Ricky <store_li at sina.com> wrote:
> >I see now. Thank you! But I am still confuse, why I write???? method can't
> >use filename as a parameter.
> 
> Many files don't have names.
> 
> E.g. sockets, pipes, StringIOs, etc.

However for methods like ConfigParser's write() it is quite common to 
accept a filename __or__ a file object.

Just


From akakakhel at attbi.com  Fri Feb  1 23:01:27 2002
From: akakakhel at attbi.com (Ali K)
Date: Sat, 02 Feb 2002 04:01:27 GMT
Subject: converting between data types
Message-ID: <rwJ68.2444$Ri2.7632@rwcrnsc54>

I would like to know how to convert a string into a number.

Please Reply.
Thank You.


Ali K.




From aa8vb at yahoo.com  Thu Feb 28 13:41:01 2002
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 28 Feb 2002 13:41:01 -0500
Subject: auto LD_LIBRARY_PATH loading?
In-Reply-To: <3C7E7092.5060101@nmt.edu>; from jeff@nmt.edu on Thu, Feb 28, 2002 at 11:01:54AM -0700
References: <3C7E7092.5060101@nmt.edu>
Message-ID: <20020228134101.B2853090@vislab.epa.gov>

Jeff Shipman - SysProg:
 |I need to be able to have DCOracle run without me explicity
 |setting the LD_LIBRARY_PATH before starting python. Is there
 |a way I can modify python to automatically add certain
 |paths to LD_LIBRARY_PATH?

Usually it's an argument to the linker.  Check your ld docs for details.  
Solaris: -R<path> ; IRIX: -rpath <path> etc.  How you pass this to the
Python build nowadays I'm not sure -- maybe 'env LDFLAGS="-R..." configure'

Randall

-- 
Randall Hopper
aa8vb at yahoo.com



From andy47 at halfcooked.com  Tue Feb  5 21:55:40 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Wed, 6 Feb 2002 02:55:40 +0000 (UTC)
Subject: cx_Oracle on cygwin
Message-ID: <Xns91AD8D95B2B69andy47halfcookedcom@203.109.250.24>

Has anyone successfully compiled cx_Oracle (2.4) under cygwin again 
Oracle 8.1.7 running on Windows 2000? 

I've grabbed the source package from Anthony's web site and am running 
into a few 'opportunities'. To save me from removing what little hair 
I've got left I wondered if anyone else has successfully done this before 
me?

My problem is (I believe) that I need to tweak the setup.py but I have 
absolutely no knowledge of make and its falling over trying to find the 
right libraries.

Out of the box 'python setup.py build' doesn't work because it assumes I 
am on Unix and looks for the Unix specific Oracle client library clntsh.

To get round this hiccup I've changed the setup.py so that it uses the 
same arguments for cygwin as it does for Windows. To do this I changed 
line 13 from;

if sys.platform == "win32":

to;

if sys.platform == "win32" or sys.plaform == "cygwin":

Now everything is fine until we reach this stage;

gcc -shared -Wl,--enable-auto-image-base build/temp.cygwin-1.3.9-i686-
2.2/cx_Oracle.o -L/Tools/Lib -L/usr/lib/python2.2/config -loci -
lpython2.2 -o build/lib.cygwin-1.3.9-i686-2.2/cx_Oracle.dll -s
/usr/lib/gcc-lib/i686-pc-cygwin/2.95.3-5/../../../../i686-pc-
cygwin/bin/ld: cannot find -loci
collect2: ld returned 1 exit status
error: command 'gcc' failed with exit status 1

Reading 'man' for gcc this *seems* to be a problem because the linker is 
looking for a file called liboci.a which, naturally, doesn't exist under 
windows. I think I should be linking to oci.dll, but I don't know enough 
to be sure.

Any suggestions, pointers or re-education would be most welcome.

Andy
-- 
Contents free posts a speciality


From zak at mysql.com  Tue Feb 26 15:41:17 2002
From: zak at mysql.com (Zak Greant)
Date: 26 Feb 2002 13:41:17 -0700
Subject: eWeek Poll: Which database is most critical to your organization?
Message-ID: <1014756079.692.671.camel@linux>

Good Day All,

eWeek has posted a poll that asks which database server is most critical
to your organization.

The poll tracks the following databases:

 - IBM DB2
 - Informix
 - Microsoft SQL Server
 - MySQL
 - Ingres
 - Oracle
 - PostgreSQL
 - Sybase Adaptive Server Enterprise
 - Sybase SQL Anywhere
 - any object database

It is interesting to note that Oracle has shown a rather large jump
between yesterday and today - about 7% IIRC - a sign that the sample
size of the poll should become larger. :)

If you would care to vote, visit:
http://www.eweek.com/article/0%2c3658%2cs=708&a=23115%2c00.asp


Note: This message is also being delivered to the Perl DBI, PHP 
general, PostgreSQL Advocacy, MySQL general and Ruby general 
mailing lists. However, to prevent possible cross-posting, 
I am delivering each message individually. :)


Ciao!

--zak

-- 
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /    Zak Greant <zak at mysql.com>   
 / /|_/ / // /\ \/ /_/ / /__   MySQL AB,  Advocate
/_/  /_/\_, /___/\___\_\___/   Calgary, Canada 
       <___/   www.mysql.com   403.244.7213




From nanotech at europa.com  Mon Feb 25 16:09:59 2002
From: nanotech at europa.com (czrpb)
Date: Mon, 25 Feb 2002 21:09:59 -0000
Subject: Compiling for HPUX and AIX
In-Reply-To: <x61yf9nx1r.fsf@notus.mii.dynalabs.de>
Message-ID: <a5e977+r51h@eGroups.com>

Michael:

That seemed to work, thanks!

Could I bother you to explain why and more importantly why Python
itself can not figure this out?

thanks!! -- Quentin

--- In python-list at y..., Michael Piotrowski <mxp at d...> wrote:
> "czrpb" <nanotech at e...> writes:
> 
> > I am having the following issues compiling:
> >
> > HPUX 11.0
> > =========
> 
> [...]
> 
> I just tried building Python 2.2 on HP-UX and had no problems; here's
> my build script:
> 
> ---------------------------------------------------------------------
> #! /usr/bin/sh
> 
> unset LANG
> export CCOPTS='+z'
> 
> make distclean
> 
> CC=cc ./configure --without-gcc --without-cxx \
>  --prefix=/opt/python && make
> ---------------------------------------------------------------------
> 
> HTH
> 
> -- 
> Michael Piotrowski, M.A.                                  <mxp at d...>
> -- 
> http://mail.python.org/mailman/listinfo/python-list




From tim at vegeta.ath.cx  Mon Feb  4 05:05:40 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Mon, 04 Feb 2002 10:05:40 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx> <tdnzo2r9sen.fsf@shell01.TheWorld.com>
Message-ID: <slrna5sn0n.srn.tim@vegeta.ath.cx>

Chris Jones <clj at theWorld.com> graced us by uttering:
> Tim Hammerquist <tim at vegeta.ath.cx> writes:
> 
> [...]
> 
>   My eyes just can't parse:
>   
>   if ( some_condition() ) {
>     {
>       for (i = first_i();i <= 10; i++)
>         {
>           initialize_some_stuff();
>           while ( some_other_condition() )
>             {
>               do_some_more_stuff()
>             }
>         }
>     }
>   
> I can, and I notice you've got unbalanced braces (look at your first two
> lines). 

Spot on! Yup, hard to break those K&R habits, even when trying to
demonstrate the alternatives.

I guess that just goes to show it's truly more a matter of what you're
used to than what's "better", whatever that is. <wink>

Thx.
Tim Hammerquist
-- 
Usenet is essentially a HUGE group of people passing notes in class.
    -- R. Kadel


From tatebll at aol.com  Tue Feb 12 07:02:41 2002
From: tatebll at aol.com (Bill Tate)
Date: 12 Feb 2002 04:02:41 -0800
Subject: PYTHONPATH on the Windows Registry?
References: <3c673b3e$1_2@nopics.sjc>
Message-ID: <cb4ba455.0202120402.5ea47551@posting.google.com>

"Ixokai" <usenet at ixokai.net> wrote in message news:<3c673b3e$1_2 at nopics.sjc>...
> Question-- if Python on Windows (w/o win32all) is running on a machine,
> without any PYTHONPATH/etc set in the registry, but with an environment
> variable set, will it set its 'current' python path into the registry? Or
> set anything at all? I want a zero-impact run of Python on the system. :)
> 
> Thanks.
> 
> --S
If you create/set an environment variable via the control panel
(system -> environment variables) the variable will be set in the
registry until removed (normally environment variables are stored
somewhere like HKEY_LOCAL_MACHINE\SYSTEM\ControlSet01\Session
Manager\xxxxxx - it varies by what version of windows you're running).
 If you set environment variable via a *.cmd or *.bat file, then those
settings exist for the length of the user's session and AFAIK no
temporary registry settings are made.


From jonathan at onegoodidea.com  Mon Feb  4 03:53:15 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Mon, 04 Feb 2002 08:53:15 +0000
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com> <7xbsf7dqj9.fsf@ruckus.brouhaha.com>
Message-ID: <B883FC7B.4309%jonathan@onegoodidea.com>

On 3/2/2002 6:32, in article 7xbsf7dqj9.fsf at ruckus.brouhaha.com, "Paul
Rubin" <phr-n2002a at nightsong.com> wrote:

> So much for compatibility then.

I think I missed the beginning of this thread so I'm confused. Exactly which
new backwards incompatible feature of the recent Python versions is it that
is breaking people's old scripts?

So far I've seen complaints of 'x in somedict' which only affects writing
new code for the new Python interpreter, future division which is not
enabled yet by default, and the import __future__ mechanism, which again is
only meant to be called by new code on the new interpreter.

If you need to run code on 1.5.2 then don't use any of the new features;
this is a given. This goes just as much for library evolution as core
language changes.

Which *old* code of someone's has actually been broken so that it will not
run correctly on the newer interpreters?

Jonathan



From osuchw at ecn.ab.ca  Fri Feb 15 17:07:34 2002
From: osuchw at ecn.ab.ca (waldekO)
Date: 15 Feb 2002 14:07:34 -0800
Subject: VB code -> Python
References: <3C6CDDC8.4ED3B329@stop.spam.volvo.com>
Message-ID: <c5b3dbee.0202151407.38f1dce9@posting.google.com>

Mikael Lex?n <mikael.lexen at stop.spam.volvo.com> wrote in message 

> So my questin is how do I translate the GetObject(...) thing in to
> Python code. 

To get info about user groups here you have two options here (thanks
to Mark Hammond of course)

1. You can use Windows API calls from win32.  Look at win32netdemo.py
located in win32\demos directory
2. Or newer ADSI based interface.  Examples are in
\win32comext\adsi\test.py

>>from win32com.adsi import *
>>me = ADsGetObject('WinNT://CAMELLOTNT/osuchwa',IID_IADsUser)
>>me.get_FullName()
u'Waldemar Osuch'

waldekO


From jwbaxter at spamcop.net  Wed Feb 13 10:51:36 2002
From: jwbaxter at spamcop.net (John W. Baxter)
Date: Wed, 13 Feb 2002 07:51:36 -0800
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org>
Message-ID: <130220020751360934%jwbaxter@spamcop.net>

In article <mailman.1013611523.15053.python-list at python.org>, Simon
Brunning <SBrunning at trisystems.co.uk> wrote:

> Probably to late to do anything about it now - think of all the code which
> would break!

Did the elves at AT&T pick up the "leading 0 means octal" from
something earlier, or did they invent this stupidity themselves?

Either way, we're stuck with it.

   --John (could the zero-less Romans have been right after all?)


From sjoshi at ingr.com  Wed Feb 13 15:45:32 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 13 Feb 2002 12:45:32 -0800
Subject: Help! IIS and Python
Message-ID: <8f8ffe67.0202131245.375fe55d@posting.google.com>

Hello
I'm trying to use IIS5.0 with Python 2.1 and whenever I try to import
modules I get an error. The test.asp and error are below.
thanks
Sunit
sjoshi at ingr.com

<%@ Language=Python %>
<%
import string
server = Request.ServerVariables("SERVER_NAME")
server = string.upper(server)
clientIP = Request.ServerVariables("REMOTE_ADDR")
%>
<HTML><HEAD><TITLE>Python Test Page</TITLE></HEAD>
<BODY>
<p>
<h3><font color="#990033">You are coming from <b><font
color="#669900"><%=clientIP%></font></b>
to <%=server%></font></h3>

<hr>
PythonAspTest
</BODY>
</HTML>
Error Type:
Python ActiveX Scripting Engine (0x80020009)
Traceback (innermost last): File "<Script Block >", line 3, in ?
server = string.upper(server) File "C:\Python21\lib\string.py", line
60, in upper return s.upper() File
"C:\Python21\win32com\client\dynamic.py", line 438, in __getattr__
raise AttributeError, "%s.%s" % (self._username_, attr)
AttributeError: <unknown>.upper


From grante at visi.com  Wed Feb 13 16:16:50 2002
From: grante at visi.com (Grant Edwards)
Date: Wed, 13 Feb 2002 21:16:50 GMT
Subject: How to capture output of os.spawnv()?
Message-ID: <6JAa8.34711$Wf1.6786156@ruti.visi.com>

I'd like to capture stdout/stderr from a program run with
os.spawnv().  

I've searched through c.l.p archives, and the usual answer
seems to be "use popen2 et al".  I don't want to use popen
since it runs the program as a shell command and I don't want
the shell to mess with the strings I'm passing as arguments to
the programs.

I can write something using the dup2(), frok(), and execv(),
but I thought maybe something already existed somewhere...

-- 
Grant Edwards                   grante             Yow!  I'm working under
                                  at               the direct orders of WAYNE
                               visi.com            NEWTON to deport consenting
                                                   adults!


From mhammond at skippinet.com.au  Thu Feb 14 18:31:03 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 14 Feb 2002 23:31:03 GMT
Subject: Windows Message Hooks
References: <8852d33c.0202140224.13840b45@posting.google.com>
Message-ID: <3C6C4920.4000301@skippinet.com.au>

wooks wrote:
> Can anyone direct me to a resource where I can find info Python
> support for the above.
> 
> Thanks.

I don't think you will find direct support.  Adding such support to the 
win32gui module should not be difficult.

Mark.





From kp87 at lycos.com  Fri Feb 15 11:13:03 2002
From: kp87 at lycos.com (kevin parks)
Date: 15 Feb 2002 08:13:03 -0800
Subject: Using Python to generate MIDI files (Mac)
Message-ID: <5e8bd451.0202150813.20720fc5@posting.google.com>

Anyone have a working module to generate MIDI files? with Python? I
have seen this asked before, but not really answered. I know that
there is a module to read MIDI files, and a few shakey attempts to
send MIDI data in real-time out a port, but I specifically want to
generate a midi file that i can then further mangle in other programs.

I would like to be able to read in midi files, parse them and muck 'em
up and
write them out again, but that all really depends on being able to
write the MIDI file in the first place...

Just curious. I would have thought that someone might have done this
buy now. I am surpised there is not much call for MIDI modules in
Python. MIDI (though in general total crap for some musical
application (like anything not equal-tempered, which pretty much kills
anything i am doing) (ironic isn't it? MIDI was made for music!, see
skinny and other protocols for what should have been) MIDI is ok for
some music applications and for controlling (music, lighting, etc.)

blah...blah...blah...

Anyone.


make
a
midi file
w/ python...

ever...

-kp--


From nas at python.ca  Wed Feb 20 20:00:18 2002
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 20 Feb 2002 17:00:18 -0800
Subject: How do you conver output of "popen('ls')" to a list?
In-Reply-To: <3C7391F6.CB6CAE01@macrozon.net>; from husam@macrozon.net on Wed, Feb 20, 2002 at 01:09:26PM +0100
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com> <mailman.1014244477.1738.python-list@python.org> <3C7391F6.CB6CAE01@macrozon.net>
Message-ID: <20020220170018.A2667@glacier.arctrix.com>

husam wrote:
> os.listdir('.') gets you all the files in the '.' dircetory. This want
> work if i want to process some selected files. In contrast to this,
> "os.popen() allows you to pass arguments like "ls *txt" to retrieve only
> *.txt files.

Sorry, that's still not a good argument for two forks and god knows how
any system calls.  Try this:

    import glob
    txt_files = glob.glob("*txt")

Cheers,

  Neil



From alex at laplaza.org  Wed Feb 20 12:13:25 2002
From: alex at laplaza.org (Alex Chavez)
Date: Wed, 20 Feb 2002 10:13:25 -0700
Subject: Python 2.1.1 on Solaris 7 broken
Message-ID: <26C20D46-2625-11D6-B442-0030656BCC12@laplaza.org>

I just upgraded Python from 1.5.1 to 2.1.1 on a Solaris 7 Sparc machine 
and I get the following error when I try running a scrip.

root at elvis cron $ python -i run_queue
Traceback (most recent call last):
   File "run_queue", line 26, in ?
     from Mailman import OutgoingQueue
   File "/usr/local/mailman/Mailman/OutgoingQueue.py", line 41, in ?
     import time
ImportError: ld.so.1: python: fatal: libgcc_s.so.1: open failed: No such 
file or directory
 >>> ZZ^Z
[2]+  Stopped                 python -i run_queue


I'm not sure if the gcc version had anything to do with it but here is 
that info.

root at elvis cron $ gcc -v
Reading specs from /usr/local/lib/gcc-lib/sparc-sun-
solaris2.7/2.8.1/specs
gcc version 2.8.1
root at elvis cron $



Help would be greatly appreciated.

alex

......:::  Alex Chavez  :  System Admin :::......
La Plaza Telecommunity : www.laplaza.org
:   Taos.  New Mexico   :   505.758.1836    :
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::




From sdm7g at virginia.edu  Fri Feb  1 10:54:10 2002
From: sdm7g at virginia.edu (Steven Majewski)
Date: Fri, 1 Feb 2002 10:54:10 -0500 (EST)
Subject: What is a "bus error"?
In-Reply-To: <ku3d0l2ubj.fsf@lasipalatsi.fi>
Message-ID: <Pine.OSX.4.40.0202011045320.764-100000@d-128-54-224.bootp.virginia.edu>


On 1 Feb 2002, Erno Kuusela wrote:

> In article
> <Pine.LNX.4.43.0202011202560.4970-100000 at zeon.well.ox.ac.uk>, Kenny
> <kccw at well.ox.ac.uk> writes:
>
> | 	I recently installed Python 2.2-5 from Fink (Unix Software for
> | MacOSX) and tried running a few sample scripts/programs.
> | 	However, some of these programs now give a "bus error" message -
> | what does this mean?
>
> it means python got a SIGBUS signal from the kernel, ie it crashed.
> SIGBUS commonly results from attempting to access unaligned memory.

... or chasing a nil pointer or useing something that's not a pointer
as a pointer.

> so it seems that there is a bug in the fink packaging of python, or
> python itself. you could try asking the fink people and/or compiling
> python yourself and seeing if that works. (if it doesn't, file a
> bug in the python bug tracker.)

You really shouldn't be getting a bus error from Python or any of
the modules packaged with it. Did you do a 'from test import testall'
in Python or 'make test' from the shell ?

Some other modules not packaged with python might have bugs that
would cause a bus-error (I know pyobjc still has at least one)
or it  might be misconfigured.

If all else fails, just get a copy of the 2.2 sources from
python.org or sourceforge and build from scratch. After unpacking
the tarfile, cd to the Python-2.2 directory and do:

./configure
make
make install
limit stack 4096
make test

-- Steve Majewski




From Mitch.Chapman at bioreason.com  Wed Feb  6 14:50:10 2002
From: Mitch.Chapman at bioreason.com (Mitch Chapman)
Date: Wed, 06 Feb 2002 12:50:10 -0700
Subject: ZODB.FileStorage -- copy?  save as?
References: <3C606CCE.ACA1472D@bioreason.com> <b0f083db.0202051909.2a8f0327@posting.google.com>
Message-ID: <3C6188F2.72793984@bioreason.com>

Jeremy Hylton wrote:
> 
> Mitch Chapman <Mitch.Chapman at bioreason.com> wrote in message news:<3C606CCE.ACA1472D at bioreason.com>...
> > I'd like to be able to do something like "save as" with
> > ZODB.FileStorage.FileStorage or a derivative thereof.  I'm having a
> > hard time even identifying alternatives, mostly because I don't yet
> > understand the FileStorage implementation.
> 
> I hope you don't need to understand the implementation in order to use
> it.  I'll grant that there's a lot of folklore about how to manage a
> FileStorage that isn't documented anywhere.

That was my hope, too.  The problem is, I'm dealing with very large 
object graphs, and am trying to find a way to implement "Save As..." 
without tying up the application for several minutes at a time.  
So I was looking for a way to do a (fast) file copy rather than 
a (slow) copy-and-save of a large Python object graph.

> [...] It's more 
> likely that you'd want to copy objects at the ZODB level; load an
> object graph out of one storage and create copies of all the objects
> in a new storage.

Yes, that's basically it:  I want to load an object graph out of 
one storage, make changes to the graph, commit the changes to
a new storage, and henceforth use the new storage.  Ideally I'd
like to be able to do this *without* having to copy all of the
objects in the graph or -- worse yet -- having to create new copies
of an associated host of "wrapper" objects.

(Never mind what those wrapper objects suggest about the
application architecture :)

Just to be sure I understand, are you suggesting something like
this?
------------------------------------------------------------------------
oldGraph = oldConnection.root()["graph"]
...
newGraph = Graph()
newConnection.root()["graph"] = newGraph

# Create copies of all objects from oldGraph, in newGraph:
newGraph.assign(oldGraph)

# Don't commit any pending changes to the old store.
oldConnection.close()

# *Do* commit all changes to the new store.
get_transaction().commit()
------------------------------------------------------------------------

> Hope this was helpful.

Yes, it was.  Obviously, I still have questions (feel free to 
answer :), but you've already helped a lot.  Thanks.

-- 
Mitch


From infinitystwin.SPAM at IS.BAD.yahoo.com  Thu Feb  7 23:33:32 2002
From: infinitystwin.SPAM at IS.BAD.yahoo.com (Greg Krohn)
Date: Thu, 7 Feb 2002 22:33:32 -0600
Subject: ActivePython and wxPython
References: <RiH88.37389$Jq.1973082@news2.calgary.shaw.ca>
Message-ID: <a3vl00014vc@enews3.newsguy.com>

"Andrew" <anon at anon.com> wrote in message
news:RiH88.37389$Jq.1973082 at news2.calgary.shaw.ca...
> Hello,
>
> Sorry if this question is inappropriate for this newsgroup, but I was
> wondering if there are known problems using wxPython with ActivePython?  I
> try to run the first example (from www.wxpython.org) in the wxPython
> tutorial but get a crash when I attempt to create a wxFrame object.  I
> stepped into the code and it's failing on an 'apply' call.
>
> Thanks,
> Andrew

Yeah, that happens. If you try it from the command line you should be fine
(i.e. 'c:\>python myscript.py'). I _think_ this has something to do with
Pythonwin's GUI getting in a fight with wxPython's GUI. Again, I _think_
they both use MFC.

greg

is _think_ a magic method?




From phd at phd.pp.ru  Thu Feb 21 04:30:58 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 21 Feb 2002 12:30:58 +0300
Subject: current default string encoding
In-Reply-To: <3c749db4.11089435@news.cis.dfn.de>; from luc.saffre@mail.ee on Thu, Feb 21, 2002 at 07:15:03AM +0000
References: <3c749db4.11089435@news.cis.dfn.de>
Message-ID: <20020221123058.A15829@phd.pp.ru>

On Thu, Feb 21, 2002 at 07:15:03AM +0000, Luc Saffre wrote:
> How or where is this current default string encoding defined? Can i

   http://www.python.org/doc/FAQ.html#4.102

> change it during run-time?

   No.

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



From james.anderson at setf.de  Wed Feb 13 09:30:13 2002
From: james.anderson at setf.de (james anderson)
Date: Wed, 13 Feb 2002 15:30:13 +0100
Subject: python to lisp translation
Message-ID: <3C6A7825.714DF612@setf.de>

?
is there a python to lisp translator out there any where?

thanks.
...


From andy47 at halfcooked.com  Mon Feb 25 18:22:19 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Mon, 25 Feb 2002 23:22:19 +0000 (UTC)
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <Xns91C1695EC1D48andy47halfcookedcom@203.109.252.31>

mtaylor at lowebroadway.com (Moray Taylor) wrote in 
news:d906b2c1.0202250818.57dcfcc7 at posting.google.com:

> Hi,
> 
> I am desperatly seeking a good python editor, so far JEdit is the best
> I have found, but since it is written in Java, I find it painfully
> slow, and PyBrowse, although good, a little flaky.
> 
> Can anyone reccommend a good python editor for UNIX/linux/IRIX. I
> don't mind paying for it, but $300 for SlickEdit is a little steep.
> 
> Before all you Emacs/vi fanatics start ranting at me, I have tried,
> and hate, them both.
> 
> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.
> 
> Somebody please help, before I have to buy a faster computer for
> JEdit.
> 
> Thanks in advance
> 
> Moray

I can't believe anyone could hate vim, its the best editor in the world ;-)

But if you must, you must. The standard answer to this FAQ is at; 

http://www.python.org/editors

Regards,
Andy
-- 
Contents free posts a speciality


From ivano.delpiccolo at degroof.be  Thu Feb 14 03:18:08 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Thu, 14 Feb 2002 09:18:08 +0100
Subject: Win DLL
Message-ID: <3c6b72a7$0$12224$4d4efb8e@news.be.uu.net>

I want to use home made DLL.
How can I use it in python ?





From m.gillies at ucl.ac.uk  Wed Feb 27 10:35:16 2002
From: m.gillies at ucl.ac.uk (Marco Gillies)
Date: 27 Feb 2002 07:35:16 -0800
Subject: Problems compiling python Debug under windows
Message-ID: <757c9986.0202270735.446b8a0f@posting.google.com>

I'm having a bit of a nightmare trying to write a python extension in
C++ under windows. I traced the original problem to the fact that I
should be using the debug version of python with a debug extension. So
I'm trying to compile python as debug. Which is where the problems
start.

I've got as far as compiling the pythoncore module (the python dll and
executables). However, I also need to compile the external modules,
_sre, _socket, parser etc. I've compiled these and the .pyd files are
there and I've tried to import them into python the give an error:

Import Error:dynamic modules does not define init function

I've tried to run DUMPBIN on the .pyd file to see what names it
exports and it doesn't seem to export anything. I think this is
because USE_DL_EXPORT isn't defined. I tried defining this but then
the modules won't compile. They give various external symbol
unresolved errors of the form:

socketmodule.obj : error LNK2001: unresolved external symbol
_PyExc_TypeError
socketmodule.obj : error LNK2001: unresolved external symbol
_PyTuple_Type

etc. etc.

My suspition is that, as USE_DL_EXPORT defines both DL_EXPORT and
DL_IMPORT to be __declspec(dllexport) some symbols in header files
that are being imported by the module I'm compiling are defined as
DL_IMPORT and should be import symbols but are in fact being defined
as export symbols by the DL_IMPORT macro.

I guess I'm doing something wrong, probably I'm defining USE_DL_EXPORT
in the wrong place (I'm doing it in the Visual C++ properties dialog
which defines the symbol on the command line while compiling the
module). I'm using Visual C++ .NET.

Does anyone have any ideas 

thanks 

Marco


From jeff at ccvcorp.com  Mon Feb 18 14:43:30 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 18 Feb 2002 11:43:30 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com> <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com>
Message-ID: <3C715962.A42F1E48@ccvcorp.com>


Michael Kelly wrote:

> On 18 Feb 2002 06:04:03 -0800, rony.steelandt at bucodi.com (Rony) wrote:
>
> >Well i did try it with the same error !
> >
> >Anyone has an idea ??
>
> Out of curiosity, did they give you a type library
> with the control?

This is important.  As I understand it, without a type library, pythoncom
has no way of knowing which params are [out] params.  Since Python doesn't
have reference semantics like those of C++ or Visual Basic, it cannot use
[out] params in the "normal" way.  So if you're using an interface that
makes use of [out] params, you *need* the type library in order to be able
to use it from Python.  If you're not sure whether you've got one, try
Makepy, as suggested...

Jeff Shannon
Technician/Programmer
Credit International




From jeff at ccvcorp.com  Mon Feb 25 15:57:08 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 25 Feb 2002 12:57:08 -0800
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu> <Vg%c8.4179$XT1.116344@news1.east.cox.net> <a5cnvo$2mdu$1@news.idiom.com>
Message-ID: <3C7AA524.C217A237@ccvcorp.com>


Jim Dennis wrote:

> >> Why is eval dangerous?
>
>
>
>  Regulars on this newsgroup display a knee jerk reaction to eval().
>  This could be construed as a healthy wariness in some cases.

Another reason for the 'knee jerk reaction' is that eval() tends to lead to
unclear code.  There's almost always some other way to do what you want, that is
much more explicit, and therefore easier to read and understand.  In particular,
other ways are less likely to have hard-to-trace side effects.  To my mind,
eval() is magic -- sometimes it's good to have, but you should always be wary
about using it.  Less magic is better.  (I won't go quite so far as to call
eval() black magic, but in some contexts it's definately pretty gray...  ;) )

Jeff Shannon
Technician/Programmer
Credit International




From alanmk at hotmail.com  Sun Feb  3 11:07:41 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 3 Feb 2002 08:07:41 -0800
Subject: Idle wait for outstanding threads?
References: <f72dac0d.0201300518.35860565@posting.google.com> <4v7g5uccpi2s9inb88uiq1cfdvlq6k5csi@4ax.com> <f72dac0d.0201310334.1a442342@posting.google.com> <a3bg0s$9n1$1@serv1.iunet.it>
Message-ID: <f72dac0d.0202030807.6d39773d@posting.google.com>

"Alex Martelli" <aleax at aleax.it> wrote:

> I see several other have already recommended Python standard module Queue,
> and I heartily second the recommendation.  Many useful multithreaded
> architectures can be put together most simply with Queue as the only
> synchronization mechanism.

Indeed, I have decided to go with a Queue.Queue solution, involving a
pre-created pool of worker threads. It makes things so much easier.
The many advantages include

1. No thread creation overhead. (See below).
2. Simpler code (which is of course much more likely to operate
*correctly*)
3. Prevents the possibility of inadvertently creating too many
threads, i.e. more threads than the OS will allow at once.

> But for your problem as you outline it elsewhere, timeouts don't
> seem strictly necessary.  You could have 4 dedicated working threads
> each waiting on a separate Queue instance; when a work-request
> arrives on the working thread's Queue, the thread does the work,
> posts the result-description to a single results-Queue (common to
> all working threads), and loops back to waiting.  The main thread,
> when a request arrives, slices it up into work-requests, posts each
> to the appropriate working thread's Queue, then goes pluck the N
> expected result-descriptions from the single results-Queue, collates
> overall results, and responds to the request.

I still think the timeout is required. The actual work being done is
database queries, which can fail for a number of reasons, or possibly
take an excessively long period of time. I need to catch those
situations and deal with them explicitly, rather than have requests
hang, while they are waiting for query results that are never going to
arrive, or take an inordinately long time.

> You probably don't want to incur the overhead of starting and
> terminating N working threads per request; the above-sketched
> "thread-pool" approach is generally more effective.  

After doing some quick timings of thread creation and
destruction/collection/joining/(?) I decided that the overhead wasn't
worth it. On my 1.2 Ghz Athlon, creation and destruction of a thread
took 400 ?-secs. Not huge, but when multiplied by 5 or 10 threads, it
starts to look significant.

> Couldn't some architecture patterned on this example satisfy
> your needs?

Indeed it can, as described above.

Thanks for the assistance.

Regards,

Alan.


From c at tech.usd253.org  Sat Feb  2 10:10:32 2002
From: c at tech.usd253.org (Fred)
Date: 2 Feb 2002 07:10:32 -0800
Subject: How can I write this C code in Python?
Message-ID: <e959e7b.0202020710.417801c8@posting.google.com>

Hello, I am just starting out with Python. I have a very basic book
that is not helping me much. I have done some simple C programming in
the past and I think if I could see the below C code done in Python it
would get me over the beginners hump. I tried to do it in Python but
got stuck right at the start, Python did not like it when I tried to
say: for i = 1

Can anybody help here? Thanks!

Fred  c at tech.usd253.org

//Simple C code to create a times table.

#include <stdio.h>
int main (void)

{

int i;
int j;

for (i = 1; i < 13; i++)
	{
	for (j = 1; j < 13; j++)
	  printf("%4d", i*j);
	        printf("\n");
   }
return 0;
}


From bokr at oz.net  Wed Feb 27 19:32:17 2002
From: bokr at oz.net (Bengt Richter)
Date: 28 Feb 2002 00:32:17 GMT
Subject: Source syntax escapes, new raw string representation. Was: Re: PEP 263 comments
References: <mailman.1014614501.21492.python-list@python.org>
Message-ID: <a5jtqh$14l$0@216.39.172.122>

On Mon, 25 Feb 2002 06:20:20 +0100, "Martin v. Loewis" <martin at v.loewis.de> wrote:

>To make some progress on PEP 263, I suggest that some of the open issues
>are resolved as follows:
>
>- Comment syntax: I suggest to use the form
>  -*- coding: <coding name> -*-
>  Emacs already recognizes this syntax, as does patch #508973 
>  on IDLEfork. The other proposed syntaxes should be removed from the
>  PEP.
>
Is this to be a *special purpose* (coding declaration) comment-context-limited
escape mechanism, or an *open ended* one? If special, I think Emacs can just
as well adapt to Python as vice versa. ISTM ad hoc special escapes in comments
are often the beginning of a convention for alternative out-of-band info, and
should be considered in that light. Cf. HTML ;-/

Perhaps it is time to (re?)consider a standard Python mechanism for embedding OOB
info at arbitrary places in Python source, analogous to XML's <![CDATA[ ... ]]> and
<? ... ?> (BTW, xml CDATA has a HUGE**N wart (IMO) in that it is not nestable.
They should have taken a clue from mime delimiting IMO).

I prefer orthogonal general purpose mechanisms to ad hoc syntactic escape warts,
so I would prefer <? ... ?> as a vehicle for defining encoding. <?py blah blah?>
could mean eval('blah blah') in some defined python environment and replacing
the <? ...?> with any string returned ('' if None), and then continue processing,
starting with the replaced text. This could be used for wholesale preprocessing
or simple mode flag control side effects, as in <?py set_special_flag(1)?>
(assuming that would get eval'd in a useful context).

In conjunction with a pythonic CDATA [1], it would permit wrapping a whole source
file (or starting at the second line):

#!/usr/bin/python
<?py recode(q'--unique delimiter[1]--'
... rest of source ...
--unique delimiter[1]--)?>

('recode' here has no special meaning. It would depend on the configured execution
context for <?py ...?>)

Thus the source could be arbitrarily reprocessed before being normally interpreted.
If you want to set a special effect flag, you could just say <?py set_special(1)?>.
If you want today's date embedded in place, write <?py now()?>, assuming that
function was appropriately defined in the execution context for <?py ...?>.
BTW, if we had an alternate assignment operator (perhaps ':=') that would make
flag=1 an expression when written as flag := 1, then we could write
<?py flag:=1? instead of <?py eval(compile('flag=1','','exec'))?>

There are side effect and execution environment issues to think about, but I think
the general mechanism could be very powerful, and ways could be created to configure
its operation via site.py etc.

[1] I am proposing a special q'...' string to be very like an r'...' string except that
the string in quotes following the q is interpreted as a mime-style arbitrary delimiter
string, and the whole value of the q'delim'...delim string representation is exactly
just the characters between the delimiters. E.g., you could

    assert q'<-=delim=->'content here<-=delim=-> == 'content here' # this would be true

without getting an error.

Note the lack of quotes around the final delimiter string, since it itself is
the final delimiter. This can also be used to solve the final unescaped
backslash problem for quoting windows paths:

    q'|'c:\foo\bar\|

Also note nestability, assuming you guarantee unique delimiter strings:

    q'<::outside::>'q'|'c:\foo\bar\|<::outside::>

A null q delimiter could be defined to imply delimiting by the end of the file or
other representation container. I.e., q''<-- content up to EOF -->

Escapes are recognized according to raw string rules inside the quotes of the
q delimiter string, so it has a final backslash problem, but that shouldn't be
too hard to live with.

q'delim' ... delim should be able contain *anything*, including <? and """ and #... and
r'whatever', etc, but <? ?> should perhaps not take precedence over ordinary string
and comment contexts. You could argue both ways. Also, a different target than py in the
xxx slot of <?xxx ...?> should act per xml PI specs, probably.

A q string could theoretically allow putting unescaped arbitrary binary data in
a source file, though many editors would have problems dealing with it. Even so,
there might be some use for that.

Thinking about it, <? ..?> and q'delim'...delim might be worth separate PEPs
irrespective of PEP 263 relevance? Opinions?

(I decided to retitle this post, since that is the actual focus here).

Regards,
Bengt Richter



From grante at visi.com  Mon Feb 18 22:18:31 2002
From: grante at visi.com (Grant Edwards)
Date: Tue, 19 Feb 2002 03:18:31 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com> <mailman.1014085471.28317.python-list@python.org>
Message-ID: <slrna73h0o.18i.grante@tuxtop.visi.com>

In article <mailman.1014085471.28317.python-list at python.org>, Chris Gonnerman wrote:

> Ultimately, adding static typing to Python results in a different
> language; it's not Python anymore.
> 
> If you don't like Python because the typing isn't strong, use Ada,
> or C++, or Pascal... 

You might also want to take a look at Modula-3.  Anybody used
to Python will recognize many of the language features.

-- 
Grant Edwards                   grante             Yow!  I'm having an
                                  at               emotional outburst!!
                               visi.com            


From niemeyer at conectiva.com  Wed Feb 20 13:51:33 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Wed, 20 Feb 2002 15:51:33 -0300
Subject: Large integers from C
In-Reply-To: <7xadu49jct.fsf@ruckus.brouhaha.com>
References: <20020220.163039.1504569917.30753@ic.ac.uk> <7xadu49jct.fsf@ruckus.brouhaha.com>
Message-ID: <20020220155132.A4275@ibook.distro.conectiva>

> Convert to hexadecimal and do intval = int(hexval, 16).
> Really, no joke.

You probably meant long(str, base):

>>> int('fffffffffffffffffffffffffff', 16)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: int() literal too large: fffffffffffffffffffffffffff
>>> long('fffffffffffffffffffffffffff', 16)
324518553658426726783156020576255L
>>> 

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From msoulier at nortelnetworks.com_.nospam  Mon Feb 25 17:37:41 2002
From: msoulier at nortelnetworks.com_.nospam (Michael P. Soulier)
Date: 25 Feb 2002 22:37:41 GMT
Subject: raw_input on password
Message-ID: <slrna7lf63.4pb.msoulier@pmerd071.ca.nortel.com>

    Hey people. 

    My apologies if this has been asked before. I didn't see it...

    Is there a way to get raw_input() to mask the response, so it can be used
to request a password safely?

    Thanks,

    Mike

-- 
Michael P. Soulier, QX41, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to Unix


From mhuening at zedat.fu-berlin.de  Sat Feb 16 13:46:40 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 16 Feb 2002 18:46:40 GMT
Subject: win32com and MS Excel
Message-ID: <Xns91B7C929EB6A6mhueningzedatfuberli@130.133.1.4>

Hello,

The following works just fine:

---
from win32com.client import Dispatch
ll = [['aaa', 333, 'ccc']] * 1000
xlApp = Dispatch("Excel.Application")
xlApp.Visible = 1
xlApp.Workbooks.Add()
t = xlApp.Workbooks(1).Sheets(1)
t.Range(t.Cells(1,1), t.Cells(len(ll),3)).Value = ll
---

But when I increase the size of the list (of lists), 
my script gets veeerryy slow. With 10000 it is still acceptable, 
but with 50000 elements you'll have to wait some minutes for
Excel to show the data. And with 100000, I get a fatal error. 
Where is the problem? Is it in win32com or Excel?
And would it be possible to speed things up? How?
(Win98, Excel 2000, ActivePython 2.1)

Thanks, Matthias


From paul at boddie.net  Thu Feb  7 08:01:54 2002
From: paul at boddie.net (Paul Boddie)
Date: 7 Feb 2002 05:01:54 -0800
Subject: High-level languages, large projects, GNOME and the .NET controversy
Message-ID: <23891c90.0202070501.78df2371@posting.google.com>

Slightly off-topic, but the recent controversy stirred up by Miguel de
Icaza and Richard Stallman on the future of GNOME and its relation to
.NET technologies produced an interesting response:

  http://www.newsforge.com/article.pl?sid=02/02/06/1536255&mode=nocomment

Here's the relevant section:

    [Miguel]

    I have written and maintained many lines of code as part of
    my GNOME work.  Ximian has developed Evolution which consists of
    roughly 750,000 lines of code. 

        Large software projects expose a set of problems that can be
    ignored for smaller projects.  Programs that have long life times
    have different dynamics when it comes to memory management than
    smaller programs.

        There is a point in your life when you realize that you have
    written enough destructors, and have spent enough time tracking
    down a memory leak, and you have spend enough time tracking down
    memory corruption, and you have spent enough time using low-level
    insecure functions, and you have implemented way too many linked
    lists [1]

        [1] indeed, GNOME uses Glib which is a massive step up from
    the Unixy libc APIs.

It occurs to me that there are still many people who choose not to use
languages of a higher level than C/C++ for development, and it would
be interesting to know what their reasons are. Of course, writing a
graphical toolkit *purely* in Python is somewhat ambitious, and there
are clearly performance issues to be considered, but I still can't
understand why people would want to write *entire* applications (of
the nature discussed above) in C/C++ these days.

Can anyone enlighten me? Perhaps I should have cross-posted to
comp.lang.c, or whatever. :-)

Paul


From mmartine at tc.fluke.com  Fri Feb  1 13:25:28 2002
From: mmartine at tc.fluke.com (Mat Martineau)
Date: Fri, 1 Feb 2002 10:25:28 -0800
Subject: Windows, RS232, and Python
In-Reply-To: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org>
References: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org>
Message-ID: <Pine.LNX.4.44.0202011011490.18577-100000@firebolt.tc.fluke.com>

Using the serial port in Windows isn't well documented - I only found
out how to do it by finding an old comp.lang.python posting.  A little
later on, USPP came along:

http://balder.prohosting.com/ibarona/en/python/uspp/uspp_en.html

It's intended as a cross-platform serial library, but you can just
look at the Windows code to see how to use win32file for serial port
access.

It sounds like you're hoping to implement something bigger, but
hopefully this information will at least help you out with the serial
port part of your project.


Mat


On Fri, 1 Feb 2002, Spencer Ernest Doidge wrote:

> I want to have control of every aspect of RS232 transmitting and
> receiving in Win9x/NT with Python commands.
>
> Question 1:
> What such functionality is already built into Python and where is it
> documented?
>
> Whatever functionality is lacking, I will implement in Win32
> API. My final product will present a complete GUI to the end user, but
> behind-the-scenes folks will use extended Python from  the command line.
>
> Question 2:
> Between two approaches, which would work better:
>
> ...
>
> To anyone who takes the time to read and answer this post, I extend my
> thanks in advance.
>
> Spencer Doidge
>
>



From cliechti at gmx.net  Tue Feb  5 15:47:29 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 5 Feb 2002 21:47:29 +0100
Subject: how can i check if a variable is set?
References: <mailman.1012938670.4213.python-list@python.org>
Message-ID: <Xns91ACDE2F5F13cliechtigmxnet@62.2.16.82>

"maximilianscherr" <MaximilianScherr at T-Online.de> wrote in 
news:mailman.1012938670.4213.python-list at python.org:

> how cna i check if a variable is set?
> like:
> 
> if variable a doesn't exist then set it to 0
> if it exists do nothing...

try to use it and handle the name NameError exception if it doesn't 
exists. or try has_attr() on class objects. or use x.__dict__.get
(varnamestr,default) [vars().get('a',0) for globals].

what best depends on your particular problem.

chris

-- 
Chris <cliechti at gmx.net>



From cliechti at gmx.net  Thu Feb 28 20:23:01 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 1 Mar 2002 02:23:01 +0100
Subject: Can python handle operating a modem?
References: <a5ltoj$m59$1@newsg2.svr.pol.co.uk>
Message-ID: <Xns91C418D5A8C4Ecliechtigmxnet@62.2.16.82>

"G. Willoughby" <sab at NOSPAM.freeuk.com> wrote in news:a5ltoj$m59$1
@newsg2.svr.pol.co.uk:
> Can python handle operating a modem? i.e. can i say, lift the hook 
> on the line to see if there is a dial tone? and test this? are 
> there any modules available for this?

two days ago i wrote a program that wait untils it sees a special 
window and then dials the phone to call me (the pc is in the basement 
doing some work and i don't want to wait there...).

i used my lib: http://pyserial.sf.net
and then i sent out "AT" commands.


-- 
Chris <cliechti at gmx.net>



From gh_pythonlist at gmx.de  Mon Feb  4 07:36:47 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 4 Feb 2002 13:36:47 +0100
Subject: [repost] win32 debugging-- python21_d.dll not found (fwd)
In-Reply-To: <Pine.SOL.4.30.0202022147160.597-100000@mimosa.csv.warwick.ac.uk>
References: <Pine.SOL.4.30.0202022147160.597-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <20020204123646.GA1187@lilith.hqd-internal>

Le 02/02/02 ? 21:49, John J. Lee ?crivit:
> [problem with mingw compiled Python extensions in debugging mode]

If all you want to do is debug a Python extension compiled with mingw32,
the following HOWTO from me might help you (watch out: *one* line):

http://groups.google.com/groups?q=python+howto+debugging+open+source&hl=en&selm=849vl9.6b1.ln%40gargamel.hqd-internal&rnum=1

Related to that, I've put a patch on Sourceforge to get rid of the
necessity of renaming the "_d" extension away when using mingw. The
statements from Tim Peters et al. about "_d" are also interesting. The
URL is (*one* line):

http://sourceforge.net/tracker/index.php?func=detail&aid=462754&group_id=5470&atid=305470

Sorry if neither resources help you.

Btw. I'm currently porting Python to a 100% pure mingw build process
using MSYS. I've already a running python.exe :-) The rest is getting
rather tricky, though :-/

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.9 ?C      Wind: 1.4 m/s



From philh at comuno.freeserve.co.uk  Thu Feb  7 18:43:42 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:43:42 +0000
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com>
Message-ID: <slrna6649e.d87.philh@comuno.freeserve.co.uk>

On Thu, 7 Feb 2002 09:51:53 -0500, Steve Holden <sholden at holdenweb.com> wrote:
>"Orr, Steve" <sorr at rightnow.com> wrote in message
>news:mailman.1012938609.3651.python-list at python.org...
>> Python Newbie here.
>>
>> I need to migrate a web application from PHP to Python and I have a
>> requirement to save state info WITHOUT using cookies where much of the
>state
>> data MUST not be visible to the end users. The PHP application allows
>users
>> to logon to one of several databases using any of a number of Oracle
>> user/passwords. The logon info needs to be saved for queries on other
>pages
>> in the app. With PHP it's easy to avoid cookies by saving this info using
>> its session handling functions. Is there an EASY way to do this in Python
>> without having to adopt the Mother of all frameworks?
>>
>Disclaimer: I have never used PHP.
>
>However, *if* PHP allows you to associate requests with session state, then
>*either*:
>
>a) it does use cookies, you just don't know about them, or
>b) it puts session references in the URL.
>
>There simply isn't any other available mechanism.

You could go by the client's IP address. Though IMO this is a bit iffy.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From maxm at mxm.dk  Tue Feb 12 08:49:40 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 12 Feb 2002 13:49:40 +0000
Subject: Why prefix with an underscore?
References: <a4b52q$1dq8gi$1@ID-98166.news.dfncis.de>
Message-ID: <3C691D74.1050100@mxm.dk>

MDK wrote:


> I am wondering why I see some people's code with function definitions
> prefixed with an underscore.
> 
> For example:
> 
> def _myfunct(foo):
> 
> Does this mean something to the code or is it for some other reason?
> 

It is a convention to tell other programmers that they don't want other 
programmers to call those functions.

A hidden method of an object, only meant to be used by the object itself.

regards Max M



From kjones at docstrategies.com  Thu Feb 28 14:42:17 2002
From: kjones at docstrategies.com (Keith Jones)
Date: Thu, 28 Feb 2002 14:42:17 -0500
Subject: Newbie scopes question
References: <vvcq7uso4lujr5i5caodd6j2dv781322hh@4ax.com>
Message-ID: <3c7e83cf$1@buckaroo.cs.rit.edu>

I do believe it will work for you if you change the line

 > print attribute

to

> print self.attribute

you need to tell python that attribute is a member variable. Otherwise, py
will think it's a local var, or failing that, a global var; and of course,
in your case. attribute doesn't exist in either of those scopes.

Hope that clears things up.

--

Keith Jones
http://www.codeexamples.org




From doug at mcnaught.org  Fri Feb 22 09:36:36 2002
From: doug at mcnaught.org (Doug McNaught)
Date: 22 Feb 2002 09:36:36 -0500
Subject: Python 2.1.2 regression crash on Linux/Sparc
References: <m3it8ql5zr.fsf@abbadon.mcnaught.org> <uwux5kcfk.fsf@python.net>
Message-ID: <m3adu1lh1n.fsf@abbadon.mcnaught.org>

Michael Hudson <mwh at python.net> writes:

> Doug McNaught <doug at mcnaught.org> writes:
> 
> > I have an UltraSparc running Linux (Red Hat 6.2 + erratea updates).  I
> > would love to get Zope running on it.  
> > 
> > I got Python 2.1.2, and did 'configure --with-threads'.  The make goes
> > fine.  When I run the regression test, I get:
> 
> Trying compiling mathmodule.c without optimization?

Worth a try--thanks for the suggestion!

-Doug


From jason at jorendorff.com  Wed Feb 20 21:47:31 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 20:47:31 -0600
Subject: print and unicode strings
In-Reply-To: <m38z9o6347.fsf@mira.informatik.hu-berlin.de>
Message-ID: <HFEKILOLEFEFMKAECNDLAEDDDEAA.jason@jorendorff.com>

Martin v. Loewis wrote:
> "Jason Orendorff" <jason at jorendorff.com> writes:
>
> > I think IDLE should do this by default.  It's fine for
> > Python to "refuse to guess" encodings, but in the case of Tk
> > it's UTF-8 on all platforms... right?
>
> Well, not exactly. Tkinter supports Unicode strings as primary data
> type for text, with byte strings (e.g. UTF-8) only being the second
> choice. So IDLE should set sys.stdout to a stream that accepts Unicode
> objects as-is.

Thank you for the hint.  I've submitted the patch (a one-liner):

http://sourceforge.net/tracker/index.php?func=detail&aid=520483&group_id=547
0&atid=305470

> Also, setting the stream for the console to cp437 is not ideal; it
> would be much better if Python would use the Unicode console API
> (i.e. WriteConsoleW) if it detects that sys.stdout is a console.  If
> you then use the Lucida Console font (instead of the raster font), you
> might be even able to display COMET in a console.

That sounds nice.  I'll try to implement this next.

## Jason Orendorff    http://www.jorendorff.com/




From SBrunning at trisystems.co.uk  Tue Feb 19 08:54:30 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 19 Feb 2002 13:54:30 -0000
Subject: ODBC on unix w/o mxODBC
Message-ID: <31575A892FF6D1118F5800600846864DCBCC9E@intrepid>

> From:	Paul Rubin [SMTP:phr-n2002a at nightsong.com]
> > You could always make a SWIG wrapper for iODBC - I started one to see
> > how hard it might be - but in the end it's probably cheaper (time is
> > money, so they say) to license mxODBC.
> 
> I hadn't heard of iODBC before, but just found the home page www.iodbc.org
> and it mentions that a Python bridge (whatever that is) already exists.
 
So far as I can tell, mxODBC *is* the Python bridge for iODBC.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From m.hadfield at niwa.co.nz  Sun Feb  3 16:07:49 2002
From: m.hadfield at niwa.co.nz (Mark Hadfield)
Date: Mon, 4 Feb 2002 10:07:49 +1300
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org>
Message-ID: <a3k9am$p0c$1@newsreader.mailgate.org>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.1012678565.11760.python-list at python.org...
> What the Python developers do get testy about is living with Guido's
> seemingly inconsistent demand that Python code use 4-space indents with no
> hard tabs, C code use 8-space hard-tab indents.  Oddly enough, the more
> intelligent of us agree with him, but nobody can explain why <wink>.

Who *are* the more intelligent of you?

---
Mark Hadfield
m.hadfield at niwa.co.nz  http://katipo.niwa.co.nz/~hadfield
National Institute for Water and Atmospheric Research





From gcordova at hebmex.com  Fri Feb 22 12:34:23 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 11:34:23 -0600
Subject: Newbie: Crazy, but Quick
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE1@SRM201002>

> 
> For that matter, you could also store the Python objects 
> directly in the SQL database.  My own opinion is that SQL
> is so easy that it's a no-brainer to use it whenever possible.
> A problem I've seen many times is people utilize some shortcut
> because they feel the program will never need anything more,
> only later to discover that the program has utility far
> beyond what they expected, but the shortcut they've taken 
> cripples it, and they end up rewriting (or worse, they don't).
> 

Has a python wrapper for SQLite been written yet?

SQLite, for those who haven't checked it out, is a
C library (or dll) which implements a SQL data store,
complete with foreign keys, indexes, etc etc.

I think it's modeled after the PostgreSQL engine,
bit I could be very mistaken. The author has done
a few comparisons with PostgreSQL, and says that
it's about 4 times faster.

Also, it stores the whole database in a single file,
which is mighty convenient.

>
> What if she decides to maintain historical data?  Then to 
> keep the files reasonably sized, she'll have to start 
> juggling files.
> 
> Software has a tendency to grow beyond the developer's 
> expectations, so I consider it unwise to take shortcuts
> when a better long-term  solution isn't much more difficult
> to implement.

Indeed.

> 
> Not only that, but storing the data in a SQL database 
> separates the data storage from the program, so in the
> future she could use that  data from some other program,
> not just the Python one.
>
> Then if she  decides she wants to access the data from,
> say, a web server, it's not a big problem. 
>

Yep.

There's the hassle of a db server. I KNOW, to us geeky
people setting up a MySQL, an mSQL (ewww), or a pgsql
engine isn't a hassle at all, and it pays back quite
nicely by separating data storage from the application.

But to a non-developer it may be more of a hassle than
to us.

-gus



From graz at mindless.com  Mon Feb 18 16:38:36 2002
From: graz at mindless.com (Graham Ashton)
Date: Mon, 18 Feb 2002 21:38:36 +0000
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <Trec8.7391$ZO2.359506@stones>

On Mon, 18 Feb 2002 19:13:08 +0000, jimh wrote:

> I have downloaded and tried out Jython, but it takes 5-6 seconds just to
> run a one-line 'print "hi"' script, and this is, of course, not
> acceptable for a web server.

Indeed. Did you try using a JIT and the jythonc byte code compiler? I'm no
expert by the way, having only tried jython-ing one of my python apps for
the first time yesterday.

It was very impressive seeing a swing GUI driving my Python libraries
though, to complement the GTK GUI that I'd already built.

-- 
Graham


From danb_83 at yahoo.com  Fri Feb 22 19:22:42 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 22 Feb 2002 16:22:42 -0800
Subject: re. help needed  :(
References: <a56838$qb8$1@newsg4.svr.pol.co.uk>
Message-ID: <ad052e5c.0202221622.7107832c@posting.google.com>

"G. Willoughby" <sab at NOSPAM.freeuk.com> wrote in message news:<a56838$qb8$1 at newsg4.svr.pol.co.uk>...
> Hi,
>     Im trying to replace the single word "lamb" and the word lamb followed
> by any punctuation in this string:
> 
> x="Mary had a little lamb called lamb, but didn't like the lamb's mood"
> 
> i want to replace all of the single words "lamb[punctuation]" but not the
> word "lamb's" to achieve this:
> 
> x="Mary had a little pig called pig, but didn't like the lamb's mood"
> 
> any help with be greatly appreiciated!

Here's one way:

x = x.replace("lamb's", "XspamX").replace("lamb", "pig").replace("XspamX", "lamb's")


From bobnotbob at byu.edu  Sat Feb 23 14:52:10 2002
From: bobnotbob at byu.edu (Bob Roberts)
Date: 23 Feb 2002 11:52:10 -0800
Subject: string substitutions
Message-ID: <c4e6b17d.0202231152.5f87765d@posting.google.com>

What would be a good way to replace every one or more spaces (" ") in
a string with just one space?  Or replace any number of newlines with
just one?


From mlh at vier.idi.ntnu.no  Wed Feb  6 00:17:34 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 6 Feb 2002 05:17:34 +0000 (UTC)
Subject: lists, uppercase?
References: <mailman.1012427749.9821.python-list@python.org>  <Vy_58.60751$8b.2485331@atlpnn01.usenetserver.com> <slrna5je5t.86s.mlh@vier.idi.ntnu.no>  <3C59CD86.A173232D@engcorp.com>
Message-ID: <slrna61f3e.h9.mlh@vier.idi.ntnu.no>

In article <3C59CD86.A173232D at engcorp.com>, Peter Hansen wrote:
>Magnus Lie Hetland wrote:
>>   if myString.isupper():
>>       # it's all uppercase
>
>Almost, but not quite:
>
>From 2.2.6.1 String Methods in the 2.2 documentation:
> isupper() :  Return true if all cased characters in the string are uppercase 
> and there is at least one cased character, false otherwise.
>
>In other words, it is possible for a string to contain characters which are
>not uppercase and yet isupper() returns true.  

And by "not uppercase" you mean "not cased", such as numbers and
special characters? I'd say the definition is pretty natural.

>So it all depends on the real requirement.  What does "totally
>uppercase" mean to the OP?

Sure.

>-Peter

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From pepotas2002 at yahoo.com  Mon Feb 18 06:10:11 2002
From: pepotas2002 at yahoo.com (pepotas2002 at yahoo.com)
Date: Mon, 18 Feb 2002 06:10:11 -0500
Subject: Interested in a home-based business?
Message-ID: <0QJE4S.4Y9R6U.pepotas2002@yahoo.com>

Have a PC?   Interested in a home-based business?

Become a successful entrepreneur and achieve the American Dream! Benefit from the Internet revolution, be a part of an explosive industry, work from home and produce incomes like this 
  

=> Amy L. works part-time around 4 kids and earns over $15,000 per month 

?Use our proven methodology and tools to achieve success?

We have all read about the fast growing, changing markets of today and all opportunities to generate extra income.  The difficulty is not in finding specific opportunities but in having the experience, support, and organization to properly begin and act on these opportunities.  The frustration is from knowing that now is the time to act, but not knowing what is the next step and how to get there. 
  
Want to learn more?? 
Phone Toll Free 1-888-240-7137
 
  
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 
This email was sent to you by the Global  Marketing Group, 1601 NW 97th Ave, SJO3016, Miami FL 33102.  011-506-267-7139.  This email cannot be considered spam as it was sent in compliance with all existing and proposed email legislation.  If you wish to be removed from this mailing, please send an email to cabeyitos at yahoo.com with remove in the subject line.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++




From oren-py-l at hishome.net  Fri Feb  8 12:04:48 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Fri, 8 Feb 2002 12:04:48 -0500
Subject: range(seq)?
In-Reply-To: <slrna67foh.dl7.mlh@vier.idi.ntnu.no>
References: <slrna67foh.dl7.mlh@vier.idi.ntnu.no>
Message-ID: <20020208170448.GA78376@hishome.net>

On Fri, Feb 08, 2002 at 12:05:37PM +0000, Magnus Lie Hetland wrote:
> I just had an idea... I'm sure there are others out there who don't
> particularly like the much-used (and necessary) expression
> range(len(seq))... One proposed solution is to make integers iterable,
> so one can do:
> 
>   for x in len(seq): ...
> 
> That is one way of doing it, of course (i'm not sure I'm all that fond
> of it, but that's another matter ;)
> 
> Then it occurred to me that perhaps it could be done the _other_ way:
> 
>   for x in range(seq): ...

I like it.  It makes much more sense for the range function to understand
container types than for integers to be iterable.  An integer is an atom, 
there is nothing 'in' it.  

> I.e. let range() take a sequence (or iterable) as an argument, and

The length of some iterables may not be known in advance without actually
stepping through them with .next(), but anything with a __len__ will do.

> Of course, we have the proposed indices(seq)... But by allowing
> range() to work on iterables, we wouldn't have to add another
> function. And I can't see any significant code breakages (but I'm sure
> there are some... :)

range() currently takes an int, long, float or a class with an __int__ 
method.  Constructing a case where range(seq) does not raise a TypeError 
but would behave differently under this proposal would require a class with 
both both __len__ and __int__ methods returning different values.  This 
seems rather unlikely.

> So, why is this a stupid idea? (I'm sure it is...)

It's a stupid idea because people hate it when someone shows them a much 
simpler way to do something and makes them feel stupid for not thinking of 
it themselves and coming up with more complicated solutions instead...

	Oren




From sholden at holdenweb.com  Wed Feb 13 08:55:16 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 08:55:16 -0500
Subject: controlling web browser in windows
References: <1103_1013605186@netnews.cf.ac.uk>
Message-ID: <Zeua8.206395$KM2.8473296@atlpnn01.usenetserver.com>

Alan:

There is a COM object in the Windows environment which exposes most of the
functionality of Internet Exploder. This would proably give you the most
satisfactory control of the experimental (?) environment, though you would
have to learn how to "drive" COM objects form Python. Hammond and Robinson's
"Python Programming on Win32" probably explains most of what you need to
know about that.

Since I am unable to easily find a reference to the object in question I'll
leave that to someone who uses it regularly!

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py

"Alan james Salmoni" <salmonia at cf.ac.uk> wrote in message
news:1103_1013605186 at netnews.cf.ac.uk...
> Hi,
>
> My problem is basically this:
>
> I am doing some psychological research into search engine usability and
need to have control of a web browser. I am using Python (cos it's simple -
I am a psychologist, not
> a programmer!) and looked at the webbrowser module but need to control the
back button (so that when a subject completes a task, the back button event
is automatically
> called) and have access to the history list.
>
> Does anyone have any suggestions as to how this could be done. I can use
IE, Netscape, or Opera (only on Windows though).
>
> Alan Salmoni
> HCI Group
> Cardiff University.
>
>
>




From Scirious at Scirious.com  Fri Feb 15 14:49:33 2002
From: Scirious at Scirious.com (Scirious)
Date: Fri, 15 Feb 2002 19:49:33 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <Xns91B6B55E421A2Scirious@140.99.99.132>

Ok! Once again thaks for your rsponse.
Scirious.


From lac at strakt.com  Wed Feb  6 12:23:55 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 06 Feb 2002 18:23:55 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary? 
In-Reply-To: Message from skywalkerpackage@hotmail.com (Johannes Eble) 
   of "Wed, 06 Feb 2002 09:46:46 GMT." <3c60fb80.199334@news.online.de> 
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org>  <3c60fb80.199334@news.online.de> 
Message-ID: <200202061723.g16HNuGo016535@ratthing-b246.strakt.com>

Johannes Eble writes:
> Hello Laura,
> 
> thanks again for your explanations.
> 
> Now I have another problem:
> 
> I want to use my own widgets in the same scheme. For example, let's
> say I have created the widget
> 
> LabelEntry
> 
> which just combines a label and an entry side by side in a row. The
> widget inherits from Frame:
> 
> class LabelEntry(Frame):
>     """Valid options: labelText, padx, pady"""
>     def __init__(self, master=None, labelText='', padx=4, pady=4):
>         Frame.__init__(self, master)
>         self.label = Label(self, text=labelText)
>         self.entry = Entry(self)
>         self.label.pack(side=TOP, anchor=W)
>         self.entry.pack(side=TOP, anchor=W, padx=padx, pady=pady)
> 
> This doesn't work together with an optionDict={'labelText':'hello',
> 'padx':'6'}
---------------------------------------------------^^^^^^^^

That is your problem.  Tkinter labels want to get their text
set with text='hello', not labelText='hello'

You have to make some design decisions here.  If you want to pass one big
dictionary, you will then have to unpack it, parse it, and generate
the correct keys for the correct widgets.  This is do-able; it is
what PMW does.  But I didn't like that approach.  I just use 2 dictionaries.

You also mention:
>But then, how do I handle default values, and how do I ensure that the
>objects are created with the right parameters? Maybe I should use a
>local dictionary.

[snip]

>Is there an elegant solution to this problem?

Well, I don't know if you would call this elegent but ...


from __future__ import nested_scopes
import Tkinter

class LabelledWidget(Tkinter.Frame):
    def __init__(self, parent, widgetName=Tkinter.Entry,
                 labelDict={}, widgetDict={}, **frameDict):
        
        Tkinter.Frame.__init__(self, parent, **frameDict)

        """
        let us say that you always want the label to be written on a
        cyan background in purple letters, unless you ask differently.
        The widget you just want on a green background.
        """
        
        ldefaults={'bg': 'cyan', 'fg' : 'purple'}
        wdefaults={'bg': 'green'}
        # overwrite any of the defaults with the dictionary you passed
        ldefaults.update(labelDict)
        wdefaults.update(widgetDict)
       
        self.label = Tkinter.Label(self, **ldefaults)
        self.label.grid(row=0, column=0, sticky='w')
        self.widget = widgetName(self, **wdefaults)
        self.widget.grid(row=0, column=1, sticky='e')

########
if __name__ == '__main__':
    root = Tkinter.Tk()
    widgetList = (
        [Tkinter.Entry,
         {'text':'An entryfield label:'},{}],
        [Tkinter.Button,
         {'text':'A button label:', 'bg':'peachpuff'},
         {'text':"I'm a button"}],
        [Tkinter.Label,
         {'text':'A label label:'}, {'text':"I'm a label", 'fg':'honeydew'}],
    )
    # for testing.
    def doit(widgetlist, part, method, **dict):
        for w in widgetlist:
            object = getattr(w, part, None)
            if object:
                command = getattr(object, method, None)
                if callable(command):
                    command(**dict)

    r = 0
    widgets=[]
    for name, ld, wd in widgetList:
        w = LabelledWidget(root, name, ld, wd)
        widgets.append(w)
        w.grid(row=r, col=0, sticky='ew')
        r += 1

    r = 0
    for w in widgets:
        w.grid(row=r, col=0)
        r += 1

    cmdList = (
        ['pink label', 'label', 'configure', {'fg':'pink'}],
        ['yellow widget', 'widget', 'configure', {'bg':'yellow'}],
        )

    c = 0
    for txt, part, method, dict in cmdList:
        button = Tkinter.Button(root, text=txt,
                                command=lambda w=widgets,
                                p=part, m=method, d=dict :
                                doit(w, p, m, **d))
        button.grid(row=r, col=c, sticky='w')
        c += 1
    button = Tkinter.Button(root, text='Exit', command=root.destroy)
    button.grid(row=r, col=c)
    
    root.mainloop()
--------------------------------------

Laura Creighton



From aahz at panix.com  Wed Feb  6 08:56:26 2002
From: aahz at panix.com (Aahz Maruch)
Date: 6 Feb 2002 05:56:26 -0800
Subject: Please comment  on Draft PEP for Enhanced Generators
References: <a38c1u$945$1@bob.news.rcn.net>
Message-ID: <a3rcma$1n9$1@panix2.panix.com>

In article <a38c1u$945$1 at bob.news.rcn.net>,
Raymond Hettinger <othello at javanet.com> wrote:
>
>I have written a draft PEP summarizing proposed enhancements to generators:
>http://users.javanet.com/~othello/download/genpep.htm

I don't like xmap() returning a list.  It should return an iterator in
all cases.  Or am I just making the mistak of reading your docstring
instead of your code?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From brueckd at tbye.com  Tue Feb 12 22:27:43 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 12 Feb 2002 19:27:43 -0800 (PST)
Subject: Why no documentation of basic data structure performance?
In-Reply-To: <eppstein-0310EF.13300812022002@news.service.uci.edu>
Message-ID: <Pine.LNX.4.33.0202121910540.21138-100000@ns0.tbye.com>

On Tue, 12 Feb 2002, David Eppstein wrote:

> Does anyone else notice a serious lack of timing information in the Python
> documentation?

Not meaning to sound nit-picky, but languages can't have timing
information, but implementations can. My point is that information
would not necessarily be true for other implementations, e.g. Jython.

Maybe the best place for that sort of info would be in a note on the
specific implementation that accompanies the CPython distribution.

> documentation?  I have the impression that one of the biggest negative
> points for Python wrt other languages is the perception that it's slow, but
> that a lot of that perception may come from programmers not having a good
> model of how much time each Python operation takes

Possibly, but most of that perception is held by people who haven't
actually used Python that much, so timing info wouldn't help dispel the
myth much.

While the info you're talking about can be valuable in some cases, in most
cases it just doesn't matter, for example, which list operation you use -
and putting that sort of info in the main documentation encourages us to
think about optimization when we shouldn't. For example, if I'm not paying
attention I catch myself doing this too much:

meth = obj.method
for item in seq:
  meth(item)

Why? Because it saves a lookup to obj.method. But unless I'm trying to
remove a specific bottleneck I'm solving a non-problem and cluttering my
code at the same time. Shame on me!

> (Not that Python is alone in this lack -- can anyone tell me whether the
> Java Vector class uses a constant-amortized-time resizing strategy when not
> given a capacityIncrement initialization parameter?)

Again you won't and shouldn't see this in the API documentation, but in a
separate "implementation notes" doc or something.

Somebody had a Python optimization page I found useful (was it Skip?) -
maybe you could report your findings to him/her and have the page updated.
IMO, that's where this sort of info belongs - on a web page or document
that we can find when we need it but just far enough out of the way that
we don't spend too much time looking at it.

-Dave




From aahz at panix.com  Thu Feb 28 15:45:30 2002
From: aahz at panix.com (Aahz Maruch)
Date: 28 Feb 2002 12:45:30 -0800
Subject: New PEP (281): Loop Counter Iteration with range and xrange
References: <slrna6fu1v.goj.mlh@vier.idi.ntnu.no>
Message-ID: <a5m4ta$rjs$1@panix2.panix.com>

In article <slrna6fu1v.goj.mlh at vier.idi.ntnu.no>,
Magnus Lie Hetland <mlh at vier.idi.ntnu.no> wrote:
>
>I've submitted a draft PEP describing the idea of letting range and
>xrange accept sequences (or possibly, for xrange, any iterable) and
>return/yield one index per element. It's closely related to PEP 212,
>but instead of proposing new functions/syntax, it proposes extending
>the capabilities of existing functions.

Have to say that I really don't like this suggestion as proposed.  IMO,
range() with a sequence should permit only one argument; at the very
least, if more than one argument is permitted, the second and third
arguments should be restricted to integers.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From max at alcyone.com  Thu Feb 21 15:22:00 2002
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 21 Feb 2002 12:22:00 -0800
Subject: What values are considered false?
References: <mailman.1014312336.21048.python-list@python.org> <a53i07$njv$1@slb2.atl.mindspring.net>
Message-ID: <3C7556E8.F2F041DB@alcyone.com>

Andrew Dalke wrote:

> Not quite.  You called the fuction 'foo', which doesn't
> define a return value so the value None is used instead.
> To compare the function to None, do
> 
>   foo is None
> 
> and you'll get 0.

Which really shouldn't be surprising, since a function that doesn't do
anything except return None is not the same thing as None.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From me at mikerobin.com  Fri Feb 22 00:20:58 2002
From: me at mikerobin.com (Michael Robin)
Date: 21 Feb 2002 21:20:58 -0800
Subject: Record Audio (Automation)
References: <140415ad.0202211224.40d8b34b@posting.google.com> <a53nl0$6ld@news.or.intel.com>
Message-ID: <52e5ab5f.0202212120.225912b8@posting.google.com>

"DanS" <danscott at chocolatefudge.com> wrote in message news:<a53nl0$6ld at news.or.intel.com>...
> The best thing here would be to write your own Visual Basic app and use the
> MS Multi Media control...
> 
> Write your program to take command line input.. and well, thats about it..

This shouldn't be necessary as you should be able to use the 
OLE automation interface to the MCI control from Python directly.
However, the MCI control has lots of baggage you don't need.
The mciSendString() API was specifically designed to do 
multimedia from scripting languages. (Originally Macromind Director's
MMP files.) If mciSendString isn't already wrapped, you should be able
to use CallDLL. 

example strings:
   "play foo.wav from start to 1434"
   "open baz.wav alias s"
   "record s"
   "close s"

See http://msdn.microsoft.com/library/default.asp?url=/library/en-us/multimed/mci_7vvt.asp
for details.

mike
 


> 
> Dan
> 
> 
> 
> 
> 
> "T H" <tth_1 at yahoo.com> wrote in message
> news:140415ad.0202211224.40d8b34b at posting.google.com...
> > I need a simple way to record audio with a scripting language in
> > Windows.  I
> > think this can be done by executing sndrec.exe or sndrec32.exe (the
> > sound recorder that comes with windows that can be executed by going
> > to Programs->Accessories->Entertainment->Sound Recorder), however I
> > can't seem to be able to make it record.  I know you can use this
> > program to PLAY a WAV file by executing
> > sndrec32.exe /play /close filename.wav
> >
> > However I was not able to make it record by executing
> > sndrec32.exe /record /close filename.wav
> >
> > The command above I tried is just a guess, i have no idea if there is
> > a /record parameter.  Any idea where I can find a list of command line
> > parameters for sndrec32.exe?
> >
> > If anyone knows any other way of automating the task of recording
> > audio using a scripting language (which executes a free audio
> > recording program), please let me know.  Any help is much appreciated.


From writeson at earthlink.net  Tue Feb  5 19:51:33 2002
From: writeson at earthlink.net (Doug Farrell)
Date: Wed, 06 Feb 2002 00:51:33 GMT
Subject: wxPython and Boa-Constructor
References: <88bc63c6.0202040509.74258657@posting.google.com> <3C5FA0D3.2000806@wanadoo.fr>
Message-ID: <p6%78.12294$Hb6.1017733@newsread1.prod.itd.earthlink.net>

Thanks for the heads up, I didn't know what you were talking about, but then
checked SourceForge.net and saw there is a new Boa-Constructor version
available. And yes, you're right, it works well.

Doug

"kondor" <kondor at wanadoo.fr> wrote in message
news:3C5FA0D3.2000806 at wanadoo.fr...
> I have python 2.2 and the last version of wxpython
(http://www.wxpython.org) and
> it work fine.
>
>
> Doug Farrell wrote:
>
> > Hi all,
> >
> > I have Python 2.2 and wxPython 2.3 installed on my machine. I've also
> > been curious about using Boa-Constructor to generate some code and
> > look at what it generates as a tutorial to me on what to do. I've
> > downloaded the latest Boa-Constructor, but it doesn't like the version
> > of Python/wxPython that I've got installed. Does anyone know what I
> > might have to do to get it running or when it might be upgraded again?
> >
> > Thanks
> > Doug Farrell
> >
>
>




From pawelczyk at lucent.com  Mon Feb 18 06:49:23 2002
From: pawelczyk at lucent.com (Lukasz Pawelczyk)
Date: Mon, 18 Feb 2002 12:49:23 +0100
Subject: Layout management in tkinter.
References: <3C6B9001.E5568D2C@lucent.com> <a4g7lr$oqb$1@norfair.nerim.net> <a4gj46$1g4g$1@mail1.wg.waii.com> <owen-D5A9FE.08352915022002@nntp2.u.washington.edu>
Message-ID: <3C70EA43.417EFD3E@lucent.com>

Hello,

Thanks it realy works. :)

I have another question.
How to make in grid method something like expand in pack?
sticky=W+E+N+S does not work.

Lukasz


From m.faassen at vet.uu.nl  Mon Feb 11 20:50:45 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Feb 2002 01:50:45 GMT
Subject: Coming of age...
References: <mailman.1013466258.26610.python-list@python.org>
Message-ID: <a49sdl$5ao$2@newshost.accu.uu.nl>

Paul Prescod <paul at prescod.net> wrote:
[snip slashdot reference]

Noticed the same thing. 

> This is one more datapoint over a few I've collected in the last few
> weeks that somehow Python has crossed some kind of line from being a
> "like to try it someday" language to an "everybody cool is using it
> these days" language. Despite waiting for this moment for several years
> I am somewhat, er, nervous about what this means for the future...

> Be careful what you wish for.

The benefits do in my opinion outweigh the disadvantages. While 
comp.lang.python is terribly busy these days, I don't think it's grown
as quickly as the use of Python itself, so it's still useful/interesting.

The advantages are that it's now much easier to use Python in projects,
sell Python based solutions, meet other Python users, hire people who
know Python, and so on. And there's just far more Python
based stuff out there that we can share.

Another interesting thing the other day is that on comp.object someone
was asking which language they should try to experiment with OO, and
there came lots of suggestions to use Ruby (Python was in there, but Ruby was
mentioned a *lot*). Seems that language has been crossing a barrier 
as well recently.

Regards,

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


From tismer at tismer.com  Fri Feb 22 10:14:46 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 22 Feb 2002 16:14:46 +0100
Subject: Stackless: 1,000,000 Switches in 0.6 seconds
Message-ID: <3C766066.3080400@tismer.com>

Sorry, I can't hold it back.

Not only that tasklets are smaller than expected, they
are also really fast.

I was stunned when I measured switching speed on an AMD Athlon XP 1500+
machine. A python script switched tasklets 1,000,000 times. The usual
Python call overhead was included. I measured this without switching and
got about 0.3 seconds. That means: Context switching by an automated,
internal scheduler without the interpreter overhead comes at the same
cost as a Python function call.

The website http://www.stackless.com/ has been updated again.
-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
       where do you want to jump today?   http://www.stackless.com/






From jason at jorendorff.com  Wed Feb 20 15:01:43 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 14:01:43 -0600
Subject: Warning about "for line in file:"
In-Reply-To: <just-8C3043.11330718022002@news1.xs4all.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com>

Just van Rossum wrote:
> It took me a while to understand what you meant, and the followups 
> didn't make it much clearer either... But now that I do, I'm just as 
> baffled as you are...
> 
> Here's what happens on my box:
> [...]
>
> This is highly unintuitive to me, indeed so much so that I would like to 
> call it a bug!

I agree!  This could be improved by having file.__iter__()
always return the same object.  But that doesn't solve the problem
in full.

One possibility is to change file.__iter__() to return
iter(file.readline, ''), and have users call file.xreadlines()
explicitly, and accept the semantic oddness, if they need the
extra speed.  :-\

Another possible hack would be to have xreadlines.next() call
file.seek() on the underlying file object, to make sure the two
are in sync.  That sounds like a pain to me.

## Jason Orendorff    http://www.jorendorff.com/



From syver-en+usenet at online.no  Sat Feb 16 19:54:20 2002
From: syver-en+usenet at online.no (Syver Enstad)
Date: Sun, 17 Feb 2002 00:54:20 GMT
Subject: win32gui
References: <a4monr$hmg$1@tyfon.itea.ntnu.no>
Message-ID: <u3d01kjy8.fsf@online.no>

"Jorg R?dsj?" <jorgelr at stud.ntnu.no> writes:
> Can anyone give us, or point us to some resources when it comes to
> using
> 
> win32gui?

The book python win32 programming by Hammond and Robinson has a
chapter on this. It's a book you need if you are going to use Python
on win32, plus it's a whole lot of fun to use/read.

-- 

Vennlig hilsen 

Syver Enstad


From marklists at mceahern.com  Mon Feb 11 21:02:33 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Mon, 11 Feb 2002 18:02:33 -0800
Subject: Unwieldy Python considered harmful.
In-Reply-To: <3c686bc9_4@corp.newsgroups.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEEPKCAA.marklists@mceahern.com>

[Joshua Muskovitz]
> I can't say it enough.  Refactor, refactor, refactor.
> 
> http://www.cnn.com/2002/US/02/11/killer.snake.ap/index.html

darwin award?

// m



From jkraska at san.rr.com  Thu Feb 28 13:11:16 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 28 Feb 2002 18:11:16 GMT
Subject: Status of PEP's?
References: <mailman.1014911568.7916.python-list@python.org>
Message-ID: <c2ts7ugv266j88crahvf8mlca4cmvenumm@4ax.com>

>  for i in len(mylist):
>
>Creates an iterator over the indices in mylist.

I just don't think so. It would be better to have something
which behaves slightly more like a classic for form. This
could almost be achieved by some sort of first class range
object thusly. In fact, one might even use slice notation:

for i in [0:len(mylist):1]:

... where the stride is optional of course.

C//



From mgash at trisystems.co.uk  Wed Feb 20 11:20:45 2002
From: mgash at trisystems.co.uk (Mark Gash)
Date: Wed, 20 Feb 2002 16:20:45 -0000
Subject: Adding GUI to an application
Message-ID: <31575A892FF6D1118F5800600846864DCC76AA@intrepid>

Hi,

I have written an application that will check the percentage free on the
local harddisks and compare against a predetermined level. 
The application is written purely for Win32 and uses the performance monitor
modules to extract the disk information.
The level is set by using a parameter within an ini file. 
If the level is exceeded, a mail is sent to the administrator. 
Once the comparison is made (and mail sent as required), the process then
goes to 'sleep' for 2 hours.
The process then wakes up again, and the comparison repeats.

The application is confirmed as working as required.

I wish to amend the application so that (i) it runs as before, (ii) there is
an icon in the taskbar, via which the application can be stopped.

As a newbie to programming, I am finding this task difficult, to say the
least. The problem that I am having, is that by using the
win32gui_taksbar.py from the win32/demos module, I can get the icon down in
the taskbar - but as the diskchecking application 'goes to sleep' rather
than completes/exits, the control is never returned to the win32gui_taskbar
to await the next interupt.

I feel that this change will have to be multi-threaded, but as I have said,
I am a newbie to programming and am 'drowning' whilst trying to complete
this.

Full code listing can be sent/attached if required.

Many Thanks for your help in advance.

Mark Gash
mgash at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From eric.brunel at pragmadev.com  Mon Feb  4 14:17:16 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 4 Feb 2002 19:17:16 +0000
Subject: Hiding the DOS console on a console application on Windows
References: <a2ru6g$2dgv$1@norfair.nerim.net> <a2u199$idd$0@216.39.172.122>
Message-ID: <a3mj0r$14o5$1@norfair.nerim.net>

Hi, and thanks for the answer,

> I'm sure you can build a straight windows gui app, and call AllocConsole()
> in the win32 API to pop up an associated console window if you want one.
> You don't have to build as a console app. Just do the AllocConsole() call
> when you want a console window.

We finally figured that out by ourselves, and did it. Unfornutaley, the 
AllocConsole functions does create a console window, and we still didn't 
want to see it.
We finally figured out a solution to make it disappear, and here it is, 
just for the record.
At first, the problem seemed simple: we only had to get the window handle 
of the console window, and pass it to ShowWindow as first argument with 
SW_HIDE as the second one. But we soon found out that there were no simple 
way to get the window handle for the console window...
So we finally used an o-so-dirty trick we found... in Micro$oft Visual C++ 
documentation! It is simply to change the title of the console window to a 
supposedly unique title via SetConsoleTitle, then get the handle of the 
window with that title via FindWindow...
Here is the actual C code we included in an external module:

// Don't make the console disappear if there was already one
if (AllocConsole())
        {
        // format a "unique" newWindowTitle
        wsprintf(newWindowTitle,"%d/%d",
                        GetTickCount(),
                        GetCurrentProcessId());
        // change current window title
        SetConsoleTitle(newWindowTitle);
        // ensure window title has been updated
        Sleep(40);
        // look for newWindowTitle
        hwndFound=FindWindow(NULL, newWindowTitle);
        // If found, hide it
        if ( hwndFound != NULL)
                ShowWindow( hwndFound, SW_HIDE);
        }

This is what I call an awful API! I'd never used such an ugly trick for 
years! ;-)
 - eric -


From no at spam.com  Wed Feb 27 11:41:14 2002
From: no at spam.com (Tokomar)
Date: Wed, 27 Feb 2002 16:41:14 GMT
Subject: Tkinter list boxes, selection emphasis, and the keyboard focus
References: <K0Qe8.48006$BA5.2207030344@newssvr14.news.prodigy.com> <gfRe8.6453$c15.3356020763@newssvr21.news.prodigy.com> <mailman.1014816224.23572.python-list@python.org>
Message-ID: <K_7f8.48513$Pm7.2510201023@newssvr14.news.prodigy.com>

Laura Creighton wrote:

> Make your listboxes like this:
> lb = Listbox(exportselection=0)
> 
> http://www.pythonware.com/library/tkinter/introduction/x5453-patterns.htm
> 
> I don't understand all of your problem so I don't know if this will fix
> all of it or only some of it.

Thanks; that solved the technical problem.

As pursuing the esthetic problem would invoke the name of the dread spirit 
from the grave, the Common User Access, I shan't be pursuing that when 
children are present.

-- 
Tokomar


From jvahn+USENET at short.circuit.com  Sun Feb 10 18:12:55 2002
From: jvahn+USENET at short.circuit.com (James Vahn)
Date: 10 Feb 2002 23:12:55 GMT
Subject: Using pydoc
Message-ID: <a46upn$u92$2@gonzo.circuit.com>

I've an existing (linux) script that begins "#!/usr/bin/python -i" and
contains (amongst other things) the lines:

        import parted
        import partition

Calls seem to be made like:

        drvobj = partition.Partition(drv)
        drvobj.create_partition_table()
        drvobj.commit_changes()

My question is:  How do I find documentation for those things?

    ~$ pydoc partition.Partition
        no Python documentation found for 'partition.Partition'


-- 


-- 


From sholden at holdenweb.com  Mon Feb 11 08:26:16 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 08:26:16 -0500
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org> <afe3b146.0202100036.f221c9a@posting.google.com> <ZND98.140690$KM2.6229113@atlpnn01.usenetserver.com> <uvgd4alnn.fsf@online.no>
Message-ID: <LDP98.149559$KM2.6717139@atlpnn01.usenetserver.com>

"Syver Enstad" <syver-en+usenet at online.no> wrote in message
news:uvgd4alnn.fsf at online.no...
> "Steve Holden" <sholden at holdenweb.com> writes:
> > >
> >
There-are-other-things-that-one-does-with-one-hand-or-the-other-but-we-don't
> >
> > -mention-ly-yours
>
> And then there are those that have to do it with both.
>
> --
>
> Two handed-ly-yours Syver Enstad

I should point out for the record that due to an angle-bracket-eating
nanovirus the remarks attributed to me were actually made by Brian Lenihan.
I hope regular readers know I would *never* mention anything we don't
mention.

Syver, I'm not quite sure whether you're bragging or complaining.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From phd at phd.pp.ru  Tue Feb 12 08:45:26 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Tue, 12 Feb 2002 16:45:26 +0300
Subject: Why prefix with an underscore?
In-Reply-To: <a4b52q$1dq8gi$1@ID-98166.news.dfncis.de>; from mdk@mdk.com on Tue, Feb 12, 2002 at 08:24:41AM -0500
References: <a4b52q$1dq8gi$1@ID-98166.news.dfncis.de>
Message-ID: <20020212164526.S4309@phd.pp.ru>

On Tue, Feb 12, 2002 at 08:24:41AM -0500, MDK wrote:
> I am wondering why I see some people's code with function definitions
> prefixed with an underscore.
> 
> def _myfunct(foo):
> Does this mean something to the code or is it for some other reason?

   from amodule import * does not import names with the underscore.

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



From bokr at oz.net  Wed Feb 20 03:34:10 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 08:34:10 GMT
Subject: add-a-gram
References: <e9e6dd1f.0202191634.21da06b9@posting.google.com> <7xpu30ew4o.fsf@ruckus.brouhaha.com>
Message-ID: <a4vn22$92v$0@216.39.172.122>

On 19 Feb 2002 18:15:51 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:

>caljason76 at yahoo.com (Jason) writes:
>> Here is a puzzle that was floating around comp.lang.lisp, and I was
>> wondering what a Python solution would be.  I am still learning Python
>> and am very curious how Python solutions differ from other languages,
>> like Perl or C.  Are there some set of maxims to the Python
>> programmer, and are there Python idioms for problem solutions?
>> 
>> Appended to the message is the problem, you can find the problems and
>> word list http://www.itasoftware.com/careers/programmers.php
>
>I did both of those problems (add-a-grams and nine 9's) in Python but
Me too, but my first translation of the mystery fcn blew the stack and
I jumped to a wrong conclusion *<8^( Still haven't gotten back to it.

>I don't feel it's appropriate to publish the solutions right now.  If
>you want to see my solutions I'll send them to you by email.
If your answer for nine 9's was less than 200 using Python, you have
some more fun coming ;-)

Regards,
Bengt Richter



From calves at coelce.com.br  Wed Feb 27 05:33:14 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Wed, 27 Feb 2002 07:33:14 -0300
Subject: Equations with Regular Expressions
Message-ID: <29A97D00F387D411AC7900902770E148041B7272@lcoeexc01.coelce.net>

Hi everybody,
Could someone of you give me and example of how can I translate
'2x^3-2x^2+15' into '2*(x**3)-2*(x**2)+15' using regular expressions.

Carlos Alberto
COELCE/DPRON-Departamento de Projetos e Obras Norte
Fone: 677- 2228
e-mail: calves at coelce.com.br
\|||/
(o o)
--ooo0-(_)-0ooo--

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020227/c278a948/attachment.html>

From gcordova at hebmex.com  Fri Feb 15 14:50:41 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 13:50:41 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A8C@SRM201002>

> 
> Take a latin alphabet. Take a lot of exceptions. Throw in some weird
> letters. Take some English words. Change a lot of spelling. 
> Mix in some more spooky letters.
> Make rules. Make exceptions to these rules.  Make
> even more exceptions. Shuffle. The resulting language will be ....
>

PERL?????

>
> German :)
> 

Ah. Oh.

>
> (I have very little knowledge of German and some of English :)
> 
> Baczek
>

:-D

-gustavo



From mwh at python.net  Tue Feb  5 09:13:34 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 14:13:34 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de>
Message-ID: <upu3kovyv.fsf@python.net>

Simon Budig <simon at ping-pong.budig> writes:

> In article <u4rkwql8p.fsf at python.net>, Michael Hudson wrote:
> [discussion of me about being lazy creating a sf account...  :-)   ]
> > That's what I did when I submitted it:
> > 
> > https://sourceforge.net/tracker/index.php?func=detail&aid=513235&group_id=5470&atid=305470
> > 
> >:) you might want to type your email address into the "monitor" field.
> 
> Ok, did that   :-)
> 
> Btw - I already had submitted a Bug-Report via a friend:
> https://sourceforge.net/tracker/index.php?func=detail&aid=511655&group_id=5470&atid=105470
> 
> You might want to link them as duplicate/related.

I made a comment to that effect.

> BTW: The stupid fixed width limit of SF messed the patch up.
> Maybe creating an attachment is more useful...

Well, anyone can see roughly what it does.  If it gets accepted (or
more likely, noone complains for a couple of weeks), I'll check it in
from your patch.

Cheers,
M.

-- 
  MAN:  How can I tell that the past isn't a fiction designed to
        account for the discrepancy between my immediate physical
        sensations and my state of mind?
                   -- The Hitch-Hikers Guide to the Galaxy, Episode 12


From dalke at dalkescientific.com  Wed Feb 13 04:03:52 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Wed, 13 Feb 2002 02:03:52 -0700
Subject: photo id ?!?!
References: <mailman.1013553742.986.python-list@python.org>
Message-ID: <a4da1h$ute$1@slb0.atl.mindspring.net>

Brian Quinlan:
> they wouldn't even consider letting me in wearing only a balaclava!

*shudder* DC was entirely too cold to think about that.

But that's not a requirement of the conference.  It's a requirement
of the venue.  The two places you listed wouldn't work out for a
conference:
  Cedar trails -- http://www.bright.net/~ctnr/Pages/rules.html has
     a fascist abuse against people with body piercing below the neck
  Twin Falls -- most of us don't live in a place that makes NSW
      Australia an easy visit

But the real problem is that neither have the facilities to run a
computer conference with 250 or so people.  Eg, for Twin Falls
    "Currently we have an external toilet and shower (with
     Hot water!),"
doesn't make me feel comfortable that they can string in a network
connection, have the facilities to run four different tracks, and
be able to feed everyone.

>If I understand correctly, your
>objection is that you are dependant on the government for your photo id
>and would prefer not to be?

There are two objections:
  1. your summary
  2. I don't believe the reasons told for requiring ID in the first place

and a complaint:
  3. there is no provision for people like me who prefer to not present
ID, or for people who lost their ID.

>Presumably you could get photo id from some non-governmental source? I
>took a look at my various pieces of photo id and found:
>1. An (old) student card
>2. A PADI diving certification card
>3. A video rental card
>4. An (old) bank card with my picture on it
>5. A season's (once a week) skiing pass
>
>Plus all the government stuff.

Umm, sure, but like I said, I could also print out a photo of me,
sign the name "Guido van Rossum" to it, write the words "Official
Dalke Scientific Software Internal Identification Document" on the
top and present that.  (I own my company - I can do these sorts
of things.)

I think I do have 14 year old student ID I keep around in a box
somewhere.  I don't have a diving certification, I don't rent
videos (your video store takes your picture?! Why!?)  My bank
card has no picture on it, and the one time I tried snowboarding
I fell on my butt so many times my tailbone started hurting, so
I swore off that activity until my next reincarnation.

I am not kidding when I say that I only have two forms of photo id.

>Presumably the security requirements of the Python conference are not so
>onerous that none of these pieces of non-governmental id would have been
>acceptable.

I don't believe there are any security requirements of the conference
that necessitate any sort of photo ID, when other credentials (like my
confirmation number) should be sufficient.

My business card (w/o photo) wasn't enough.  Being able to have
previously authorized people already at the conference verify who I
am wasn't enough.

                    Andrew Dalke
                    dalke at dalkescientific.com





From mdk at mdk.com  Wed Feb 13 16:55:54 2002
From: mdk at mdk.com (MDK)
Date: Wed, 13 Feb 2002 16:55:54 -0500
Subject: Is making Python executables a reality?
Message-ID: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>

I have tried several tools to make a Python exe, i.e. py2exe which just
gives me errors.  It looks as though changes were made to how it is supposed
to work but now it does not work for me.

The programs that are supposed to come bundled with Python for this task do
not seem to exist on my PC after the Python install.

What percentage of people make exe files?  Do most people have end-users
download the Python environment?

What is everybody doing in this regard?

Thanks.




From jason at jorendorff.com  Wed Feb  6 16:57:04 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 15:57:04 -0600
Subject: Is there anyway to obtain client's IP address using cgi module?
In-Reply-To: <Pine.SOL.4.30.0202061627250.17425-100000@hadar.cse.Buffalo.EDU>
Message-ID: <HFEKILOLEFEFMKAECNDLAENFDAAA.jason@jorendorff.com>

J. Wang wrote:
> I am trying to create a python CGI script. [...] I'd like to log the
> user's IP address [...]

import os
os.environ['REMOTE_ADDR']

More: http://hoohoo.ncsa.uiuc.edu/cgi/env.html

## Jason Orendorff    http://www.jorendorff.com/



From michael at stroeder.com  Fri Feb  8 11:29:55 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 08 Feb 2002 17:29:55 +0100
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com> <slrna6649e.d87.philh@comuno.freeserve.co.uk> <NkQ88.105036$KM2.4008630@atlpnn01.usenetserver.com>
Message-ID: <3C63FD03.1AC94EDE@stroeder.com>

Steve Holden wrote:
> 
> phil hunt" <philh at comuno.freeserve.co.uk> wrote:
> > You could go by the client's IP address. Though IMO this is a bit iffy.
> >
> Extremely iffy when you realise that many X terminal users may run their web
> clients on the same CPU, not to mention Lynx users on timesharing terminals.

Even more iffy if you think of web proxies. Or web proxy farms with
many siblings doing load-balancing which causes every HTTP request
to come from another web proxy with different IP address.

Ciao, Michael.


From skip at pobox.com  Tue Feb 12 22:45:39 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 21:45:39 -0600
Subject: searching backwards in a string
In-Reply-To: <7x8z9y46l1.fsf@ruckus.brouhaha.com>
References: <mailman.1013549132.16176.python-list@python.org>
        <7x8z9y46l1.fsf@ruckus.brouhaha.com>
Message-ID: <15465.57699.381206.105954@12-248-41-177.client.attbi.com>

    Paul> I propose adding a "direction flag" to the search function:

    Paul>    match = pattern.search(text, 50000, -1, direction=-1)

    Paul> here 50000 and -1 are the starting and ending locations.

If start > end isn't direction == -1 implied?

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From danb_83 at yahoo.com  Wed Feb 27 21:49:46 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 27 Feb 2002 18:49:46 -0800
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com> <29e28c51.0202270014.61dbd1d0@posting.google.com> <a5ji37$ceb$2@peabody.colorado.edu>
Message-ID: <ad052e5c.0202271849.164dd36c@posting.google.com>

Fernando P?rez <fperez528 at yahoo.com> wrote in message news:<a5ji37$ceb$2 at peabody.colorado.edu>...
> Cimarron Taylor wrote:
> 
> > I have used the following tricks before with Java code, some of them
> > should be useable with python.
> 
> [snipped]
> 
> I sincerely hope this was a joke (and as such, a really good one :). I can't 
> seriously imagine you actually *doing* those things. 1 is fine, maybe 2.

Have you ever read "How to Write Unmaintainable Code"?

http://www.mindprod.com/unmain.html


From syring at email.com  Sun Feb  3 07:11:43 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 13:11:43 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <13285ea2.0202030404.5dfe9b66@posting.google.com>
Message-ID: <a3j9du$181gf1$1@ID-7529.news.dfncis.de>

"J.Jacob" <joost_jacob at hotmail.com> schrieb^
> > > Why in the std distribution of Python 2.2 there isn't the curses
> > > module?
> [Gerhard H?ring]
> > If you're using Windows, that explains it ;-) There is indeed no curses
> > module included in the Python binary distribution for Windows. IIRC the
> > reason is that there is no ncurses library readily available on win32 to
> > build the curses support against.
>
> There have been many more questions about curses for Windows.  There
> is even no alternative in the Python distribution, you can download
> 3rd party packages like WConio or console but you have to compile them
> yourself and you are out of luck if you do not have that MSVC++
> compiler the Python team choose if you want to communicate with their
> libraries.  This is a bad situation and it comes up again and again
> when you [have to] use Windows.  For a programming class in school
> [never did see a school using linux except university] it would be
> good if Python came with something like curses for Windows, the
> flickering menus look bad and give Python a bad reputation.  And no,
> you do not want to start a programming class with explaing TKinter and
> callbacks hehe :-)

This is utter nonsense. Pdcurses simply works on Windows.

Karl M. Syring



From danb_83 at yahoo.com  Tue Feb 26 22:50:56 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 26 Feb 2002 19:50:56 -0800
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <ad052e5c.0202261950.72991328@posting.google.com>

hankc at nospam.com (HankC) wrote in message news:<3c7c1b5b.34558051 at news.la.sbcglobal.net>...
> Greetings:
> 
> I'm 2 days new to Python and hope I can get some pointers here...
> 
> I have a sorted, delimited text file such as:
> 
> Arty,1000
> Bobby,2000
> Charlie,3000
> 
> Knowing a name, I want to return the entire line from the file.  My
> approach/problems (any comments appreciated):

If your data file is small enough to fit in memory, you can use

dict = {}
datafile = open(..., "r")
for line in datafile.readlines():
   name = line.split(",")[0]
   dict[name] = line[:-1] # strip trailing \n
datafile.close()

Now, for example, dict["Bobby"] == "Bobby,2000"


From michael at rcp.co.uk  Fri Feb  1 04:08:31 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Fri, 1 Feb 2002 09:08:31 +0000 (UTC)
Subject: An improvement to Queue
References: <Xns91A793581DD79michaelrcpcouk@194.238.50.13> <a3blop$a29$1@serv1.iunet.it> <Xns91A7A14C232EBmichaelrcpcouk@194.238.50.13> <a3ccif$rls$1@panix2.panix.com>
Message-ID: <Xns91A85CFFB7ED2michaelrcpcouk@194.238.50.13>

aahz at panix.com (Aahz Maruch) wrote in news:a3ccif$rls$1 at panix2.panix.com:

> What I still don't get is why you're waiting on an event rather than on
> data.  It seems pointless.  Just have the worker thread grab data and
> process it.
 
The worker thread has other things on its mind: when it does a job, it 
wants the whole job fully set up and ready to go.


From dima at xenon.spb.ru  Tue Feb  5 19:53:23 2002
From: dima at xenon.spb.ru (Dmitry Rozmanov)
Date: Wed, 6 Feb 2002 03:53:23 +0300
Subject: graphics module, ICC profiles, sRGB colorspace
Message-ID: <a3puq4$9oi$1@news.wplus.spb.ru>

Hi.

Is there any special module or pakage for Python to deal with ICC profiles
and colorspaces?

As far as I know PIL knows nothing about color management. Am I right?

Thank you.

    ---Dmitry.

P.S. It would be nice if you send a replay to e-mail as well. ;).




From hughett at mercur.uphs.upenn.edu  Wed Feb 13 11:27:08 2002
From: hughett at mercur.uphs.upenn.edu (Paul Hughett)
Date: 13 Feb 2002 16:27:08 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net>
Message-ID: <a4e44s$tli$1@netnews.upenn.edu>

John W. Baxter <jwbaxter at spamcop.net> wrote:
: In article <mailman.1013611523.15053.python-list at python.org>, Simon

: Did the elves at AT&T pick up the "leading 0 means octal" from
: something earlier, or did they invent this stupidity themselves?

: Either way, we're stuck with it.

It could be worse.  In one language I worked in, "10" was an octal integer,
"10." was a decimal integer, and "10.0" was floating point.  Nice language
otherwise, though.

Paul Hughett


From logiplexsoftware at earthlink.net  Mon Feb  4 12:09:52 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 4 Feb 2002 09:09:52 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <a3kb3p$se44@news1.gtech.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3d46d$tfn2@news1.gtech.com>
	<Wls68.14178$om6.357057@news1.tin.it>
	<mailman.1012590723.7301.python-list@python.org>
	<a3kb3p$se44@news1.gtech.com>
Message-ID: <20020204090952.2c9d7f9b.logiplexsoftware@earthlink.net>

On Mon, 4 Feb 2002 08:42:58 +1100
Peter Milliken wrote:

> G'Day Cliff,

Peter =)

> > I much prefer the "continuous ongoing code review" idea better than a
> > formalized procedure (if I understand what you mean by it correctly).
> > Working in pairs provides this as a side-effect rather than having a
> formal
> > procedure for code-review and seems less time-consuming overall.
> >
> 
> Not sure what you mean by "continuous ongoing code review". I used to be
a
> "team leader", I have "looked after" groups of programmers between 1985
and
> 2000, bitter experience taught me to "review early and often" i.e.
diivide
> the programmer's work in small units and as each unit is completed review
> it - never, ever, ever let the stuff build up to the point where after an
> unfavourble review you are faced with the decision to either miss
schedule
> because it should be re-written or forge ahead regardless and hope that
it
> will come good through some miracle of extra effort on some ones part
:-).
> There are so many what I will class as "poor" programmers in the
industry,
> that this is a necessary technique when you have to "manage" other
> programmers. So I assume that "continuous ongoing code review" is similar
to
> my idea of "early and often"? :-)

I think so.  My personal experience in working closely with another
programmer was that we tended to write code appropriate for the other
person to read and often discussed approaches to coding problems.  After
discussions, prototype code was reviewed (informally) to see how our ideas
worked out, if there were shortcomings and the like.  We never had a
"formal" review, but since we were comfortable working with each other and
had much mutual respect, we were able to criticize and accept criticism
freely.

> >
> > Actually, when I say "art", I am using it in the sense of "skill", as
in
> > "martial art" or "art of design" versus the broader sense that includes
> > unrestrained personal interpretation.  In this sense "art" means taking
> > basic, established techniques and internalizing them to the point where
> > they are second-nature. At that point, personal expression becomes an
> > extension of those techniques - not a deviation from them.  My
experience
> > has been that programmers at this level feel a sense of pride and
personal
> > accomplishment and _appreciate_ code review (otherwise, who will ever
see
> > and appreciate their accomplishment?).  I suspect this to be a major
> > driving force behind open-source software - it's the equivalent of a
> public
> > demonstration of one's skills.
> >
> 
> I think you are correct about the open source motivation etc, but in any
> typical project/company I have worked for, it is *relatively* rare to
find
> people who welcome a review of their code. The people you find in Open
> Source stuff are the people who view programming as a "joy" (hobby)
rather
> than a "job". These people, in my experience, tend to be better
programmers
> (probably because they really do "care" and have pride in their work?
:-)).
> But the majority of people I have worked with over the years forget about
> programming the minute they step out the corporate doors, it is nothing
more
> than a job to them - very sad :-). These people tend to be the people who
> view their work as "Art" rather than increasing their skills as an "art"
:-)
> They do no external reading, they are not interested in seeing how other
> people have solved similar problems etc etc.

People who don't enjoy their work inevitably suck at it.  This is
undoubtedly a factor in why it is not uncommon to find that a self-taught
programmer is often quite superior to someone who actually obtained a
degree in CS.  It seems that quite often people are drawn to a field
because that field is in vogue and looks to be financially rewarding, so
they get a degree in that field, but to them it's simply a job (of course
this does not apply to the majority of people pursuing degrees, but it is
not uncommon), whereas a self-taught individual usually becomes proficient
because they are deeply interested in the subject.

Of course, the ideal programmer is the self-taught individual who later
goes to college to raise their level of skill.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From tonygreen at techie.com  Tue Feb  5 08:31:03 2002
From: tonygreen at techie.com (Tony Green)
Date: 5 Feb 2002 05:31:03 -0800
Subject: Jython Embedding - how do I compile and exec with params
Message-ID: <cefe12a8.0202050531.33412d38@posting.google.com>

I am writing a scripting extension mechanism to a big server project
we've built over the last year and a half.

What I want to acheive is
- system authors upload scripts (via a web front end)
- the system checks and compiles the script
- stashes the compiled script away against a named event.

When the even occurs the compiled script is called by the system with
a list of parameters (current user id, various numbers describing the
system state).

Now I can upload and compile the script just fine using :
PyCode somePyCode = __builtin__.compile ( script, "<>", "exec");
...and deal with any errors ...and stash the script away against a
named event...etc.

And it is easy enough to call this code with
aPythonInterpreterInstance.exec ( somePyCode )

but that doesn't pass my parameters through.

It seems to me there's three obvious ways to do this:

(1) specify that each script has a named target (eg. 'def
onEvent(eventid , userid):') and use one of the PyCode.call() methods
- but they aren't sensibly documented in the javadocs so it could take
a while to figure out what's really going on.

(2) specify that each script contains a class def and implements a
__call__ method - but I can't see how to get the PyClass object out of
the PyCode object returned by __builtin__.compile

(3) put the variables into the local namespace and then execute the
script (without a named method or class) - but this means that I would
have to clean down the local namespace each time (- and anyway what is
it local to?).

(Did I mention execution speed is important to this app and I need to
service multiple threads?)

Thanks in advance,
Tony


From jason at jorendorff.com  Mon Feb 18 23:48:13 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 22:48:13 -0600
Subject: Adding static typing to Python
In-Reply-To: <a4s51k$8pd1@news1.gtech.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEJODDAA.jason@jorendorff.com>

Peter Milliken:
> When I compile a Ada program cleanly I know that 95% of the bugs 
> are out of it - before I even start testing! :-)

Brian Quinlan:
> I would say that 99% of my bugs are due to design mistakes [...]
> or semantically incorrect algorithms[.]

Forgive me if I am a bit dubious about these numbers.

I have added a sys.excepthook in the Lib/site.py file
on the PC where I do most of my work.  It logs uncaught
exceptions (except in interactive mode).  In a week or two,
I should have some real numbers to share.

Obviously this is a crude and subjective measurement,
but it's better than (ahem) nothing.

## Jason Orendorff    http://www.jorendorff.com/



From rsteiner at visi.com  Wed Feb 13 13:08:50 2002
From: rsteiner at visi.com (Richard Steiner)
Date: Wed, 13 Feb 2002 12:08:50 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org>
 <130220020751360934%jwbaxter@spamcop.net>
 <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net>
Message-ID: <yuqa8oHpvCEW092yn@visi.com>

Here in alt.folklore.computers,
claird at starbase.neosoft.com (Cameron Laird) spake unto us, saying:

>In article <130220020751360934%jwbaxter at spamcop.net>,
>John W. Baxter <jwbaxter at spamcop.net> wrote:
>			.
>>Did the elves at AT&T pick up the "leading 0 means octal" from
>>something earlier, or did they invent this stupidity themselves?
>			.
>OK, folks, how far back *can* we trace this?  My
>mind associates it with DEC systems, going back
>to the '60s, but I couldn't find any confirmation
>of that in a quick search.

I've seen the leading-zero-means-octal convention used rather heavily
in documentation dated 1967-1968 in a UNIVAC 1106/1108 context.

It seems to have been a fairly well-established convention at that
point in time, at least in the UNIVAC world.

-- 
 -Rich Steiner >>>---> http://www.visi.com/~rsteiner >>>---> Eden Prairie, MN
   OS/2 + BeOS + Linux + Solaris + Win95 + WinNT4 + FreeBSD + DOS + PC/GEOS
            + Fusion + vMac + Executor = PC Hobbyist Heaven! :-)
           Press CTRL-ALT-INS-DEL-END-HOME-SHIFT-PAUSE to continue.


From jason at jorendorff.com  Fri Feb  8 21:34:33 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 8 Feb 2002 20:34:33 -0600
Subject: Perl programmers?
In-Reply-To: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEFJDBAA.jason@jorendorff.com>

> I know some of y'all still keep roots to the Perl community, so I was
> wondering if there's any kind of definitive nickname for Perl
> programmers, the way we call ourselves Pythonistas?

I believe it's "Perl hacker", as in "Just Another Perl Hacker".

## Jason Orendorff    http://www.jorendorff.com/



From pajer at alum.mit.edu  Sun Feb 24 21:32:48 2002
From: pajer at alum.mit.edu (Gary Pajer)
Date: Sun, 24 Feb 2002 21:32:48 -0500
Subject: installing and finding PMW
Message-ID: <3C79A250.B24F6CC5@alum.mit.edu>

Uncle!

I'm learning python. I'm using IDLE.   I just installed (or tried to)
pmw on my Win98 machine.

I can't import pmw.   I get a msg that the module can't be found.

I've installed pwm in c:\program
files\gary\python22\lib\site-packages\pmw
sys.path seems to include this directory.  I've added this directory to
PATH and PYTHONPATH.
I notice that there is no file called pmw.py --- how does import work in
this case?

Where should I put PMW????   How should the environment variables be
set???

Thanks,
Gary



From marklists at mceahern.com  Sun Feb 10 23:08:46 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 20:08:46 -0800
Subject: FixedPoint
In-Reply-To: <HFEKILOLEFEFMKAECNDLEEKBDBAA.jason@jorendorff.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBCEBEKCAA.marklists@mceahern.com>

[Jason Orendorff]
> Yep.  If you know what's in _norm() you can do better than I
> can.  :)  Also, you might pull the first return statement out
> of the try.

_norm basically pulls the number and the precision out of the FixedPoint
instances.  So that would give us this:

    def __cmp__(self, other):
        try:
            xn, yn, p = _norm(self, other)
        except TypeError:
            xn, yx = self, float(other)
        return cmp(xn, yn)

The only problem is:  It doesn't work with other as None because this fails:

	float(None)

which leads us back to testing for None:

	def __cmp__(self, other):
		if other is None:
			return 1
            try:
                  xn, yn, p = _norm(self, other)
            except TypeError:
                  xn, yx = self, float(other)
            return cmp(xn, yn)

I'm inclined at this point to think I'd only do this if I couldn't change
the code that's doing the foo == None.

// mark




From phr-n2002a at nightsong.com  Thu Feb 21 21:39:26 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 18:39:26 -0800
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1>
Message-ID: <7xd6yyfdep.fsf@ruckus.brouhaha.com>

Jim Richardson <warlock at eskimo.com> writes:
> I have an application that does a calculation on a single byte of data,
> after breaking it into an upper and lower nibble of four bits. How can I
> do this in python? To clarify.
> Take a single byte, break off the upper 4 bits, and do some stuff, then
> do similar stuff with the lower nibble, and then munge them back
> together to create a byte again. This is for a checksum to inteface with
> a bit of embedded stuff. I know how I can do this in C, but as that
> would be the sole bit of copiled code in the app, I'd prefer to avoid
> it if possible. 


Am I missing something?  Python has the usual bit operations that work
on ints.

a = do_some_stuff(byte >> 4)        # upper nibble
b = do_similar_stuff(byte & 0x0f)   # lower nibble
new_byte = (a << 4) + b     # assume a and b are nibble results


From tim at worthy.demon.co.uk  Mon Feb 11 08:17:28 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Mon, 11 Feb 2002 13:17:28 GMT
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com> <69c619074b.tim@worthy.demon.co.uk> 
 <7xlme0fcku.fsf@ruckus.brouhaha.com>
Message-ID: <b82222074b.tim@worthy.demon.co.uk>

In message <7xlme0fcku.fsf at ruckus.brouhaha.com>
          Paul Rubin <phr-n2002a at nightsong.com> wrote:

> Tim Howarth <tim at worthy.demon.co.uk> writes:

> > 0xFFFFFFFFL
> > to become
> > -1
> 
> That's the result I get in Python 2.2.  I don't have 2.1 on this system.
> 1.5.2 raises an exception when it tries to convert 0xFFFFFFFFL to hex.

Stupid me, yes so it does - I think I might have tried using 8FFFFFFF
for some reason !

> > besides I grew up believeing eval==evil.
> 
> I guess you could use binascii and struct to convert the hex string to
> a binary string and then pack the binary string into an integer.
> Ouch.

Exactly, it's a lot of bother for a simple request!

I'll probably stick with my bitshift version.
 
-- 
___
 |im    ---- ARM Powered ----


From trentm at ActiveState.com  Fri Feb  8 17:11:26 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Fri, 8 Feb 2002 14:11:26 -0800
Subject: TRYING to like python, but...
In-Reply-To: <a41h0m$ear$1@web1.cup.hp.com>; from jim@jim.jim on Fri, Feb 08, 2002 at 01:46:56PM -0800
References: <a41h0m$ear$1@web1.cup.hp.com>
Message-ID: <20020208141126.C4813@ActiveState.com>

> so PLEASE PLEASE PLEASE tell me that I don't really have to go through ALL
> OF THIS CONFUSING MESS just to get a STANDARD python library to work.
> (xml.dom IS in the "Python Libary Reference", but there is NO mention of all
> these additional steps required to actually USE it).
> 

You *could* use ActivePython. It comes with Python expat interface setup for
you.
    http://www.activestate.com/Products/Download/Get.plex?id=ActivePython
Note that this is a shameless plug -- I work for ActiveState.



Trent

-- 
Trent Mick
TrentM at ActiveState.com



From ullrich at math.okstate.edu  Thu Feb  7 11:25:27 2002
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Thu, 07 Feb 2002 16:25:27 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>
Message-ID: <3c62a989.3448437947@news>

On 6 Feb 2002 02:27:13 -0800, sandskyfly at hotmail.com (Sandy Norton)
wrote:

An amusing aspect of all this: I just browse clp occaisionally,
since I'm an abstraction manipulator, not a programmer. So I
overheard comments about this guy before seeing any examples
of his output. Before seeing the things he wrote I was gonna
say "you guys got it easy - you should see sci.math". But
the funny thing is the guy reminds me _exactly_ of the
sci.math guy I had in mind: If you can't see what a genius
he is you're incompetent, you're all gonna be fired when
the Truth comes out, etc...

>"Whereof one cannot speak, thereof one must be silent."

Sorry, I wasn't listening.

David C. Ullrich


From mtaylor at lowebroadway.com  Thu Feb 28 06:34:42 2002
From: mtaylor at lowebroadway.com (Moray Taylor)
Date: 28 Feb 2002 03:34:42 -0800
Subject: PyQT and shaped windows in X
Message-ID: <d906b2c1.0202280334.35f7942f@posting.google.com>

Hi,

Can anyone point me in the direction of an example of shaped windows
using QT with Python, is the XLib module required? Is it even
possible?

BTW I have tried asking in the PyQT/PyKDE mailing list, but it wasn't
considered to be a PyQT question, go figure.

Thanks v much in advance

Moray


From miracle at paradise.net.nz  Mon Feb 25 14:54:05 2002
From: miracle at paradise.net.nz (Matthew Sherborne)
Date: Tue, 26 Feb 2002 08:54:05 +1300
Subject: Is it possible in Python too?
References: <3C7A6BA1.26999.19EEC27@localhost>
Message-ID: <001e01c1be36$2f740a90$08020196@Jonah>

Yes but I don't know how.

Tell me how to do it in another language, and I'll tell you how in python ;)

GBU
Matthew
----- Original Message -----
From: "A" <printers at sendme.cz>
To: <python-list at python.org>; <tutor at python.org>;
<activepython at listserv.ActiveState.com>; <python-help at python.org>
Sent: Tuesday, February 26, 2002 4:51 AM
Subject: Is it possible in Python too?


>
> Hi,
> Is possible to set up a proxy name (address) in MS Explorer
> browser from a running  Python program?
> Thank you for your help.
> Ladislav
>
>
>
> _______________________________________________
> ActivePython mailing list
> ActivePython at listserv.ActiveState.com
> To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
>




From Mitch.Chapman at bioreason.com  Tue Feb 26 16:11:47 2002
From: Mitch.Chapman at bioreason.com (Mitch Chapman)
Date: Tue, 26 Feb 2002 14:11:47 -0700
Subject: python stripping LD_LIBRARY_PATH?
References: <3C7BDD3A.9050905@nmt.edu>
Message-ID: <3C7BFA13.EDEFC99F@bioreason.com>

Jeff Shipman - SysProg wrote:
> 
> I never got a response to this and I really don't
> understand why it's happening. I would appreciate
> any help anyone can give me. If there's another place
> I should be posting this, please let me know.
> 
> -------- Original Message --------
> 
> I've got a wrapper so I can run a SUID python script [...]

> The problem I'm having is that the LD_LIBRARY_PATH
> is not showing up in os.environ. 
> [...]
> Why is python stripping out this environment variable?
> 
> I'm using:
> 
> ActivePython 2.1, build 211 (ActiveState)
> based on Python 2.1 (#1, Jun 15 2001, 02:52:41)
> [GCC 2.95.2 20000220 (Debian GNU/Linux)] on linux2


Hi Jeff,

Python is not removing LD_LIBRARY_PATH.  The runtime loader 
ld.so is doing so.  I think this is for security reasons, 
i.e. to prevent someone from slipping a bogus shared library
implementation into the runtime environment of a privileged Unix
executable.  Or something like that :)

See the ld.so(8) man page.  Here's an online copy turned up 
by Google:
	http://www.cwi.nl/~aeb/linux/man2html/man8/ld.so.8.html

The relevant excerpts:
	The shared libraries needed by the program are searched 
	for in various places:
	...
	o Using the environment variable LD_LIBRARY_PATH. Except 
	  if the executable is a setuid/setgid binary, in which 
	  case it is ignored. 

-- 
Mitch Chapman
Mitch.Chapman at bioreason.com


From borcis at geneva-link.ch  Mon Feb  4 09:45:19 2002
From: borcis at geneva-link.ch (Boris^2)
Date: Mon, 04 Feb 2002 15:45:19 +0100
Subject: Python-formed XML ? Re: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org>
Message-ID: <3C5E9E7F.EE0DEF8B@geneva-link.ch>

Gerard Braad wrote:
> 
> Hello,
> 
> For a long time now, I had the idea of XML-Formed Python.

What about a two-way transform allowing a nested indentation
expression of XML as well ? For a long time now, I had the
idea that much of XML logic can be captured with python-style
nesting, eg :

<FOO bar=parm1 baz=parm2> some data </FOO>

gives maybe

tag foo(bar=parm1,baz=parm2) :
	some data


Boris


From msoulier at nortelnetworks.com_.nospam  Wed Feb 27 16:39:02 2002
From: msoulier at nortelnetworks.com_.nospam (Michael P. Soulier)
Date: 27 Feb 2002 21:39:02 GMT
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com> <%rbf8.71116$X64.25961788@news1.rdc2.pa.home.com>
Message-ID: <slrna7qkg6.4pb.msoulier@pmerd071.ca.nortel.com>

On Wed, 27 Feb 2002 20:37:15 GMT, Terry Reedy <tjreedy at home.com> wrote:
> 
>>>> def prcnt(counter = [0]):
> ...   print "The current count is %d" % counter[0]
> ...   counter[0] = counter[0] + 1
> ...
>>>> prcnt()
> The current count is 0
>>>> prcnt()
> The current count is 1

    But that doesn't work with multiple instances of prcnt. They all share the
same reference to the default list you created. 

    Mike

-- 
Michael P. Soulier, QX41, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks, SDE Pegasus
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to Unix


From millioner at wallst.ru  Sat Feb 23 12:34:54 2002
From: millioner at wallst.ru (Vladimir)
Date: Sat, 23 Feb 2002 22:34:54 +0500
Subject: =?ISO-8859-1?Q?=F0=E5=E0=EB=FC=ED=FB=E9 =E7=E0=F0=E0=E1=EE=F2=EE=EA,
 =E2?=
	=?ISO-8859-1?Q?=FB=F1=EE=EA=E8=E9 =E4=EE=F5=EE=E4?=
Message-ID: <a58kl0$bm8$587@news.raid.ru>

????????? ??????? ??????? ? ????? ??????????????????? ??????-??????? ??? ????????? ?????. WMGetter - ??? ??? ???????? ???? ? ????????? ? ??????. ???? ?? ?????? ???? ?? ????? ????, ?? ??????? ????? ?????? ????????????? ? ??????? ????? ?????? ? ?????????? ???????. ?? ???????? ???? ???? ? ????? !  http://www.wmgetter.com/vmikheev/







From phr-n2002a at nightsong.com  Mon Feb 18 20:40:14 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 18 Feb 2002 17:40:14 -0800
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com> <ad052e5c.0202181707.69c22978@posting.google.com>
Message-ID: <7xsn7yqmf5.fsf@ruckus.brouhaha.com>

danb_83 at yahoo.com (Dan Bishop) writes:
> > I have downloaded and tried out Jython, but it takes 5-6 seconds
> > just to run a one-line 'print "hi"' script, and this is, of
> > course, not acceptable for a web server.
> 
> Because you expect that running a ten-line script will take a minute?

Huh?  The 5-6 seconds is probably startup overhead so the ten-line
script would take the same 5-6 seconds, which is already unacceptable
for a web server.  Even 1 second is enough to make the server feel
unresponsive.


From db3l at fitlinxx.com  Wed Feb 20 20:25:18 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Feb 2002 20:25:18 -0500
Subject: why does popen2 silently ignore IOError?
References: <mailman.1014229773.31846.python-list@python.org> <a50vu1$29da$1@nntp6.u.washington.edu>
Message-ID: <uy9hn4oe9.fsf@ctwd0143.fitlinxx.com>

"Donn Cave" <donn at u.washington.edu> writes:

> popen is actually a bit better than popen2 here, though, inasmuch as when
> you close the file object you get a special return value with the exit
> status.  To get this value with popen2, you need to use the Popen3 class
> and call its wait() member function.

Or on Windows, which doesn't have the Popen3/4 classes, you get the
exit status on close for any of the os.popen# calls (in Python 2.0 or
greater since before that only os.popen was in the standard library
for Windows).  For the popen2-popen4 calls, the exit code is retrieved
when you close the final file from the set returned by the initial
popen# call.

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


From ajs at ix.netcom.com  Sun Feb 10 15:09:26 2002
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Sun, 10 Feb 2002 15:09:26 -0500
Subject: Perhaps I am just dumb
Message-ID: <000701c1b26e$ef4e0920$a9e1fea9@carol>

Steve writes -

>Well, if I might be allowed to blow my own trumpet, "Python >Web
Programming" ...

A bit out of context, but I would like to take the opportunity
to commend Steve' efforts.  He rightly assumes that it is the
Web technology some of us need/want to understand better (get hands on
with), and that we choose to do so via Python (either because we are already
comfortable with it, willing to learn it, or find Python particularly well
suited for this type of exploration).

Being a Big Shot now, I was able to skip over most of the Python
introduction, and get right to the heart of the Web technology
presentations, from which I am learning much and having good fun.

This and the new O'Reilly book "Python & XML" by Jones and Drake are a great
one, two punch.

I expect to be dangerous after having absorbed them.

Or at least a lot less mystified.

There may come a point when I follow considerably more than my current 26%*
of the discussion on the python-list.

Art

* statistics corrected for the Rue autocoding thread





From MaximilianScherr at T-Online.de  Sun Feb 24 04:18:28 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Sun, 24 Feb 2002 09:18:28 -0000
Subject: thread question
In-Reply-To: <a4kj1a$p57$1@panix2.panix.com>
Message-ID: <a5ab54+p15f@eGroups.com>

can i acces attributes, or functions of a thread(like if i started 
the class with threading.Thread(None, someclass)) like i can access 
classes? and if yes how? if no, how can i do it in another way?

--- In python-list at y..., aahz at p... (Aahz Maruch) wrote:
> In article <mailman.1013817865.2645.python-list at p...>,
> maximilianscherr <MaximilianScherr at T...> wrote:
> >
> >if a thread (object) is for example a class, would it be possible 
to 
> >call the class objects of this thread from another thread?
> >or how could something similar be done?
> 
> Absolutely.  See http://starship.python.net/crew/aahz/
> for some examples.
> -- 
>                       --- Aahz  <*>  (Copyright 2002 by aahz at p...)
> 
> Hugs and backrubs -- I break Rule 6                 
http://www.rahul.net/aahz/
> Androgynous poly kinky vanilla queer het Pythonista
> 
> "usenet imitates usenet"  --Darkhawk
> -- 
> http://mail.python.org/mailman/listinfo/python-list




From cliechti at gmx.net  Sat Feb 23 16:41:42 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 23 Feb 2002 22:41:42 +0100
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com>
Message-ID: <Xns91BEE76FABF7Ccliechtigmxnet@62.2.16.82>

bobnotbob at byu.edu (Bob Roberts) wrote in 
news:c4e6b17d.0202231152.5f87765d at posting.google.com:

> What would be a good way to replace every one or more spaces (" ") 
> in a string with just one space?  Or replace any number of newlines 
> with just one?

>>> ' '.join("h  e    l l   o".split())


-- 
Chris <cliechti at gmx.net>



From risnes at alfanett.no  Wed Feb 27 13:03:38 2002
From: risnes at alfanett.no (André Larsen Risnes)
Date: Wed, 27 Feb 2002 18:03:38 GMT
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
	<3c7c5eb3.159302234@news.isar.de>
Message-ID: <20020227191758.0939a577.risnes@alfanett.no>

On Wed, 27 Feb 2002 04:32:53 GMT
gerson.kurz at t-online.de (Gerson Kurz) wrote:

[...]
> Well, "Python Obfuscation" is my middle name ;) For starters, there is
> 
> - the "How to write unmaintainable code" FAQ
> 
> http://mindprod.com/unmain.html
> 

Lol! This is good, i especielly liked the "17. Bedazzling Names" item.
Evil >:-)

-- 
Andr? Risnes




From jkraska at san.rr.com  Thu Feb 14 12:28:40 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 14 Feb 2002 17:28:40 GMT
Subject: Recommendations for cross-platform GUI Bindings ?
References: <xBMa8.217$B92.52511@news.xtra.co.nz>
Message-ID: <ictn6uomd63papjtut186c67lickq0cjj9@4ax.com>

>	I'm writing an Open Source software application that allows children to
>create their own video games, and would like to hear anyone's
>recommendations for GUI bindings. The key things I'm looking for are:

For low level graphics, pygame is the best of the lot. It's on
top of SDL.

For windowing, Tk, WxWindows, and QT are the king of the crop,
although I'm not sure how well the QT bindings are being maintained.

C//



From syring at email.com  Tue Feb  5 12:18:55 2002
From: syring at email.com (Karl M. Syring)
Date: Tue, 5 Feb 2002 18:18:55 +0100
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com> <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu> <M6M78.23304$l93.4247257@newsb.telia.net> <7xu1swe3bv.fsf@ruckus.brouhaha.com> <hYT78.14612$tg4.168818@vixen.cso.uiuc.edu>
Message-ID: <a3p45v$19iupe$1@ID-7529.news.dfncis.de>

"John" <nobody at nobody.com> schrieb im Newsbeitrag
news:hYT78.14612$tg4.168818 at vixen.cso.uiuc.edu...
>
> Is there any available code that does the below?

And above, there is
http://www.nmr.mgh.harvard.edu/Neural_Systems_Group/gary/python.html

Karl M. Syring

>
> > but the OP asked for a generator where you could just specify what
> > skewness you want.  The random module doesn't give a direct way to do
> > that.  You have to pick a distribution, then crunch out some integrals
> > and solve some equations to get the distribution parameters for the
> > skewness you want.
> >
> > Further description:
> > http://www.everything2.net/index.pl?node=skewness
>
>



From aahz at panix.com  Sun Feb 17 21:06:45 2002
From: aahz at panix.com (Aahz Maruch)
Date: 17 Feb 2002 18:06:45 -0800
Subject: basic books/guides on multithreading programming?
References: <mailman.1013813246.12449.python-list@python.org> <7xlmdrmymb.fsf@ruckus.brouhaha.com>
Message-ID: <a4pnjl$7n1$1@panix2.panix.com>

In article <7xlmdrmymb.fsf at ruckus.brouhaha.com>,
Paul Rubin  <phr-n2002a at nightsong.com> wrote:
>
>There's a pretty good book called (I think) Java Threads.  I looked at it
>ages ago so I don't remember too much specific from it.  Python's threading
>module is sort of a subset of Java threads.

Not really.  Modeled on Java threads, yes, but Java requires threads on
all implementations where Python simply requires OS-level threads (and
punts if they're not available).  There are also some semantic
differences due to the Global Interpreter Lock.  Best to approach Python
threads as their own thing -- I've seen several people go astray because
they figured Python (technically, CPython) threads were the same thing
as Java.

That said, a book on Java threads can be a good resource for someone
wanting to learn about using Python threads, but you'd better have my
slides handy.  See http://starship.python.net/crew/aahz/
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From paul1brian at yahoo.com  Thu Feb 21 05:50:20 2002
From: paul1brian at yahoo.com (Paul Brian)
Date: Thu, 21 Feb 2002 10:50:20 -0000
Subject: Python UK Conference - April 4 & 5 2002
Message-ID: <1014287724.17116.0.nnrp-13.c1c3e191@news.demon.co.uk>

Dear all,

As you may have seen in other postings, the UK is hosting a Python
conference in April.

The two day conference is packed full of talks on cutting edge applications
of Python, with time to meet, socialise and even discuss code, life and love
with fellow Pythonistas. You even get access to the full ACCU conference,
attended by luminaries such as Bjarne Stroustrup.

The conference is kindly being hosted by the ACCU (The Association of C & C
+ + Users) at the Heritage Motor Centre, Warwick, (near Stratford Upon Avon,
and Oxford) and it runs for the perfect length of two days, on Thursday 4th
of April and Friday 5th April 2002.

You can register online at www.accuconference.co.uk , and further details
are available there too.  Please note that Python attendees should pay the
daily rate for the conference (?130 /day for ACCU members, ?155 else) for
the one or two days they choose to attend.

I do hope as many of you who can will be able to attend, even in these ROI
focused times.

Yours

Paul Brian
"I still have not got a signature for my newsreader"




From tripps81 at yahoo.com  Fri Feb 15 17:57:09 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Sat, 16 Feb 2002 05:57:09 +0700
Subject: basic books/guides on multithreading programming?
Message-ID: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>

which ones do you recommend? examples in any language is fine, 
though i prefer python, perl, or ruby. what i would like is 
guidance on what is thread safety, how to achieve it, what are 
the issues in multithreading programming, the global interpreter 
lock, and other basics.

t




From alans at killergame.com  Sun Feb 17 22:33:01 2002
From: alans at killergame.com (alan scales)
Date: 17 Feb 2002 19:33:01 -0800
Subject: wxPython:  EVT_CLOSE question
References: <c0a56015.0202151702.1f29d19e@posting.google.com> <slrna6tjb6.e10.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <c0a56015.0202171933.7d828ae1@posting.google.com>

Vadim Zeitlin <zeitlin at seth.lpthe.jussieu.fr> wrote in message news:<slrna6tjb6.e10.zeitlin at seth.lpthe.jussieu.fr>...
> On 15 Feb 2002 17:02:45 -0800, alan scales <alans at killergame.com> wrote:
> > i have some controls (in particular wxStyledTextCtrl) - they are NOT
> > frames or dialogs.  I want to be able to trap when they are closed (or
> > destroyed)(e.g. to save data). I was hoping to use EVT_CLOSE and
> > provide a handler.  however it does not seem to work.
> > 
> >   1. is it supposed to work? (it is only documented for
> > frames/dialogs)
> 
>  No, it isn't supposed to work for the simple controls. It only works for he
> top level windows.
> 
> >   2. if not, what can be done to detect when a control is closed?
> 
>  The best would be to do it in EVT_CLOSE of the frame containing the control.
> You also can use EVT_WINDOW_DESTROY but this could be too late to do anything
> meaningful with the control.
> 
>  Regards,
> VZ

the frame containing the control isn't being closed (the controls in
question are sort of being "paged in and out" - maybe this is
pathological application). however, the EVT_WINDOW_DESTROY might be
made to work.


thanks, 
as


From oren-py-l at hishome.net  Fri Feb  1 04:42:14 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Fri, 1 Feb 2002 11:42:14 +0200
Subject: Please comment on PEP XXX: Enhanced Generators
In-Reply-To: <000b01c1aa3a$1207a300$4dd8accf@othello>; from othello@javanet.com on Thu, Jan 31, 2002 at 04:31:14AM -0500
References: <a3at9p$2he$1@bob.news.rcn.net> <20020131092440.GA19083@hishome.net> <000b01c1aa3a$1207a300$4dd8accf@othello>
Message-ID: <20020201114214.A2549@hishome.net>

> 1.  Four new built-in functions:  xmap, xfilter, xzip, and indexed

I have already implemented these functions (except indexed) two months ago 
in my dataflow library (http://www.tothink.com/python/dataflow).  

My implementation has a very significant difference: just like the xrange 
function, the resulting object is a restartable source, not a one-time 
iterator.  Naturally, the source is restartable only of all arguments to 
the functions are restartable sources themselves.

> 2.  Generator comprehensions:  [yield int(x) for x in floatgen()]

The only problem I see with this syntax is that the square brackets are
misleading - the resulting object is not a list.  How about using parens
instead?  They are not strictly required, but just like for tuples, they
help disambiguate the syntax.

> 3.  Feeding data into a generator using an optional argument to .next()

Python currently has generators that produce data.  Below are three other
significant dataflow configurations:

1. consumers
2. transformations - receive data from an upstream source, perform some 
transformation on it and send it to their downstream consumer.
3. dialogues - like transformations, dialogues they have an input and an 
output, but they are both connected to the same logical entity, not an 
upstream and a downstream.

Consumers functions are relatively simple. Like generators, they do not 
require full-fledged multithreading, just the suspension of a single 
function context.  Like generator functions, consumer functions would allow 
the state machines of the producer process and the consumer pto run 
independently.

Dialogues can be implemented without threads, too, but in that case they
would require the producer process and the consumer process to run in 
lockstep, producing and consuming an item of data for each step.  Letting
them produce and consumer data at different rates would require either 
the use of null values, full threading support or decoupling the rates 
with a queue.

Transformations can be implemented easily using a generator function that
takes an upstream iterator as an argument.  Transformations may be chained
together, if necessary.  The input and output rates of a tranformation are 
decoupled without any need for threads of queuing: the transformation may 
call the upstream iterator's next more or less than once per yield statement.

Transformations could also be implemented as a consumer function that
takes a downstream sink object as an argument, but the upstream view looks
more natural and can be implemented today using generator functions.

My opinion:

We know enough about consumers to implement them now, if a good syntax is 
found and they are deemed important enough by the BDFL.  Transformations 
are easy with the current generators.  Dialogues are a form of dataflow 
more often found in communication systems than as a method of data flow 
between two parts of the same program.  Communication systems have many 
other issues to deal with than just the two-way exchange of data so two-way 
generators are probably not the right way to support them.

   dataflowingly yours,  

	Oren


Appendix: Emulating consumer functions with generator functions: 

  def consumer(args):
      <initialize>
      more = [None]
      yield more
      while more:
          <use data in more[0]>
          yield more
      <cleanup>

driving a consumer function:

    for mailbox in consumer(args):
        if <more data>:
            mailbox[0] = <data>
        else:
            del mailbox[:]

the long version:

    itr = consumerfunction(args)
    mailbox = itr.next()
    while <more data>
      mailbox[0] = <data>
      itr.next()
    mailbox[:] = []
    try:
      itr.next()
      raise 'Shouldn't be here'
    except StopIteration: pass




From magnus at thinkware.se  Thu Feb 21 07:04:23 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 13:04:23 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee> <3C74DD65.3060000@thinkware.se>
Message-ID: <3C74E247.3090506@thinkware.se>

Sorry for my multiple replies. Netscape is playing
dirty tricks with me... (This time I'll just press
send once, and wait to see if it appears in the news
group before I try again, whatever error message it
gives me.)

Magnus Lyck? wrote the same stuff three times...:
[snip]



From paul at boddie.net  Thu Feb 28 11:22:05 2002
From: paul at boddie.net (Paul Boddie)
Date: 28 Feb 2002 08:22:05 -0800
Subject: Thread support in NetBSD
References: <a5i40k$1k59$1@news.uit.no> <m3lmdf1dv8.fsf@mira.informatik.hu-berlin.de> <a5iarm$1nvp$1@news.uit.no>
Message-ID: <23891c90.0202280822.575c6ccc@posting.google.com>

"Anders ?erg" <andersaa at stud.cs.uit.no> wrote in message news:<a5iarm$1nvp$1 at news.uit.no>...
> 
> > If the pthreads module is suited for your thread library, you can try
> > setting CPPFLAGS and LDFLAGS before invoking configure. If that fails,
> > you need to adjust configure.in.
> 
> What should i set these flags to? Sorry but i don't know much about these
> things..

If you're running bash or something similar (with a $ prompt), try:

  CPPFLAGS=/path/to/headers LDFLAGS=/path/to/libraries ./configure ...

I think this should work. You could always have a look at config.log
to see where the configure program goes wrong.

Paul


From tim.one at comcast.net  Thu Feb 21 15:41:47 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 21 Feb 2002 15:41:47 -0500
Subject: newbie floating point question
In-Reply-To: <pan.2002.02.21.12.47.20.378591.4031@R191.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEFDNOAA.tim.one@comcast.net>

[Alex]
> Why does this happen ?
>
> >>> .2
> 0.20000000000000001
>
> Is this the correct behavior for the Python 2.2 interpreter ?

It's both correct and inexcusable, maddening and highly desirable.  The main
difference between Python and other languages you may have used here is that
Python isn't lying to you (well, it is, but not as badly as most others):

    http://www.python.org/doc/current/tut/node14.html




From bc90021 at bc90021redundantdot.net  Mon Feb 18 17:24:54 2002
From: bc90021 at bc90021redundantdot.net (bc90021)
Date: Mon, 18 Feb 2002 22:24:54 GMT
Subject: Newbie question...
References: <3C716B27.3090408@bc90021redundantdot.net> <mailman.1014067413.6584.python-list@python.org>
Message-ID: <3C717F1B.8080102@bc90021redundantdot.net>

X-No-Archive: Yes

> ... but of course hacking the RPMs can be fun. When I was using SuSE, I
> always changed the .SPEC file and recompiled the source RPMs. Then
> changed all the other Python related .SPEC files, if necessary and
> rebuilt everything Python related. Most of the time, this is pretty
> straightforward and a simple replace of "1.5" with "2.2" in the .SPEC
> file is enough.
> 
> I don't want to discourage you :-)
> 
> Gerhard


Thank you for all your help.  You won't discourage me - I've been 
playing around with RedHat since 5.0, and half the fun of RedHat is 
getting it to work *properly*.  ;)

I've never tried hacking the RPMs, as I've always just tried to solve 
the dependencies, but that sounds like it could be fun!  :)

bc90021



From ncstate1997 at yahoo.com  Tue Feb  5 23:25:40 2002
From: ncstate1997 at yahoo.com (Ed)
Date: Wed, 06 Feb 2002 04:25:40 GMT
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1012940770.12866.python-list@python.org>
Message-ID: <8f288.61264$A51.25251725@typhoon.southeast.rr.com>

Me neither.  I lost my non-dominant left arm over a year ago.

Let me know if I can help on that autocoder!

Eddie


"Jimmy Retzlaff" <jimmy at retzlaff.com> wrote in message
news:mailman.1012940770.12866.python-list at python.org...
> But about the only hand skill relevant to Python is typing, which I
> (and probably most of us) do both-handed...

Not me. I just had surgery on my dominant right hand last Thursday so
I'm temporarily non-dexterous. :(

Since I haven't been able to write much Python since, c.l.py has been a
welcome diversion!

Maybe ol' Tim Rue can come up with an autocoder concept that only uses
keys on the left half of the keyboard. I'd volunteer to help implement
it, but I type too slowly with this bloody stump. ;)

Jimmy





From arnoldjsph at netscape.net  Sat Feb  2 19:24:18 2002
From: arnoldjsph at netscape.net (Joe Arnold)
Date: Sat, 02 Feb 2002 19:24:18 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <20020202134649.468$Gg@news.newsreader.com>
Message-ID: <3C5C8332.3080101@netscape.net>

Hernan M. Foffani wrote:


> For example, if your CEO states: "We are an Oracle partner, so
> everything in PL/SQL" how much is left to say?


"Adios"



From mwh at python.net  Fri Feb  1 05:46:12 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 1 Feb 2002 10:46:12 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de>
Message-ID: <ur8o5xytl.fsf@python.net>

Simon Budig <Simon.Budig at unix-ag.org> writes:

> Hi all.
> 
> Something is broken with the completion of readline:

The more I learn about readline, the less I like about it.  Not that
I'd want to live without it, of course.

> It seems that readline has a fallback to filename completion when
> no matches are available. Even if I use my own completion function:
> 
> >>> def nullcompleter (text, state):
> ...    print "\nBuh!"

<aside>
That's probably a bad idea; you have no idea what state the terminal
will be in when this function gets called...
</aside>

> ...    return None
> ... 
> >>> rlcompleter.readline.set_completer(nullcompleter)
> >>> foo<TAB>
> Buh!
> <TAB>
> Buh!
> 
> foo.gif     foo.txt     foo2.gif    foobar.jpg  foot        
> >>> foo
> 
> there is this filename fallback.
> 
> Is this a known Problem? 

Well, I knew about it.  I don't regard it as a huge problem.

> Is there an evil hack to avoid this?

Dunno.  Doubt it, somehow.  You'll need to look at the documentation
for readline, though.

Cheers,
M.

-- 
40. There are two ways to write error-free programs; only the third
    one works.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From andy at reportlab.com  Tue Feb 26 04:54:54 2002
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 26 Feb 2002 09:54:54 -0000
Subject: advocacy help ( Python vs VB )
In-Reply-To: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>
Message-ID: <LKENLBBMDHMKBECHIAIACEKNCEAA.andy@reportlab.com>

> I need to convince my co workers and erstwhile managers to build a
> production system in Python.  They're coming from a VB world and
> are fearful
> to detach from the Redmond mothership.  I need hard facts, not warm and
> fuzzies about syntax but names of big companies using Python and things VB
> just *cannot* do ( threads, regular expressions.. ? ).
>
> Thanks,
> Stefan

If you can find it, please look at the first chapter of "Python Programming
on Win32", which was written to be "shown to your boss" in this situation.
And look at the mid section which shows how you can embed Python engines in
VB GUIs.  Your best bet may be to show the two languages working in
cooperation and not in competition.

- Andy Robinson








From usenet at thinkspot.net  Mon Feb  4 04:01:43 2002
From: usenet at thinkspot.net (Sheila King)
Date: Mon, 04 Feb 2002 01:01:43 -0800
Subject: Possible bug in Email  and Multifile Modules ?
Message-ID: <a3kmhn.3vvfim9.1@kserver.org>

Some friends and I are working on a script that filters incoming email. We
ran across some difficulties today with a certain test message that a
friend is using to test the script.

The original script is using mimetools and multifile to parse MIME
multipart messages (some parts of it were written when we only had access
to 1.5.2 on our web host).

The pertinent parts of the script are as follows (note--I didn't write this
code...written by a "python newbie" friend of mine):

============(begin code)==================
import sys, os
import mimetools, multifile

header = mimetools.Message(sys.stdin,0)
msgtype = header.gettype()

if msgtype[:10] == "multipart/":
	file = multifile.MultiFile(sys.stdin,0)
	file.push(header.getparam("boundary"))
	while file.next():
		submsg = mimetools.Message(file)
		nm = submsg.getparam("name")
		if nm <> None:
			nm = nm.lower()

			if nm in bannedfiles:
				if append_status == "on":
					msg_bannedfiles = msg_bannedfiles + msg_break +
"(Reason: " + quote + nm + quote + " file not permitted.)"
				print msg_bannedfiles
				sys.exit(exitcode)

			if bannedextensions:
				tempvar = nm.split(".")
				lastindex = len(tempvar) - 1
				ext = tempvar[lastindex]
				if ext in bannedextensions:
					if append_status == "on":
						msg_bannedextensions = msg_bannedextensions +
msg_break + "(Reason: " + quote + ext + quote + " files not permitted.)"
					print msg_bannedextensions
					sys.exit(exitcode)

  		fnm = submsg.getparam("filename")
		if fnm <> None:
			fnm = fnm.lower()
			if fnm in bannedfiles:
				if append_status == "on":
					msg_bannedfiles = msg_bannedfiles + msg_break +
"(Reason: " + quote + nm + quote + " file not permitted.)"
				print msg_bannedfiles
				sys.exit(exitcode)

			tempvar = fnm.split(".")
			lastindex = len(tempvar) - 1
			ext = tempvar[lastindex]
			if ext in bannedextensions:
				if append_status == "on":
					msg_bannedextensions = msg_bannedextensions + msg_break
+ "(Reason: " + quote + ext + quote + " files not permitted.)"
				print msg_bannedextensions
				sys.exit(exitcode)
	file.pop()
============(end code)==================

Someone was testing the code out today with a message containing a known
virus and reported the following error results:

> btw, I think I found one of the bugs in Python2.2. The script bombs 
> out when I feed it an email that contains the Sircam virus. The error is;
> multifile.Error: sudden EOF in MultiFile.readline()
> When I switch back to Python1.5.2, the problem doesn't exist. 

On the chance the I might fare better with the new email module, I got a
copy of the email message from him and tried the following in an
interactive session, with results as shown:

=============(interpreter session)====================
>>> msg = email.message_from_file(open('virus.txt'))
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "E:\PYTHON\PYTHON22\lib\email\__init__.py", line 35, in
message_from_file
    return _Parser(_class).parse(fp)
  File "E:\PYTHON\PYTHON22\lib\email\Parser.py", line 40, in parse
    self._parsebody(root, fp)
  File "E:\PYTHON\PYTHON22\lib\email\Parser.py", line 116, in _parsebody
    raise Errors.BoundaryError(
BoundaryError: Couldn't find terminating boundary:
====_ABC1234567890DEF_====
=============(end interpreter session)================

Now, it turns out that this message is not well formed. Instead of the
correct closing boundary at the end of the message, it has a boundary which
indicates that another part should follow.

I can see no way to read this message in using the email module. This makes
the email module something that I would be unable to use to scan incoming
mail.

I reported my findings to this other tester:

>  You will see that the final boundary is missing the two closing
>  hyphens. Therefore, it looks like there should be another message
>  part, but there isn't one.
> 
> Is it possible that Norton removed the last part or something? Are you
> 100% sure that this is the intact message as you received it? If so,
> then I guess it isn't a correctly formed message.
> 
> Still, this seems to me to be a problem with the modules in Python. I
> will report it in the newsgroup and see what they say.
> 
> I was able to import the message you sent into Agent and Agent did
> show it as having two attachments, the second one being named
> stuff.MP3.pif.

He responded:

>  No parts were removed, it's 100% intact. And you're right, it's not
>  well-formed, but that's the whole idea behind it. It tricks Outlook
>  into executing the attachment.
> But not being well-formed shouldn't break anything...
> 


Curious to hear what others, more experienced than I, think about this. I
can provide a copy of the virus message and the complete code for the
current script upon request.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_





From robin at jessikat.fsnet.co.uk  Thu Feb 14 04:02:29 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 09:02:29 +0000
Subject: py2exe & McMillan's installer (was Re: Is making Python executables a reality?)
References: <E16bC9r-0000zp-00@mail.python.org>
 <mailman.1013665162.16163.python-list@python.org>
Message-ID: <vTOazvAl03a8Ews2@jessikat.fsnet.co.uk>

In article <mailman.1013665162.16163.python-list at python.org>, ahimsa
consulting <ahimsa at inext.co.za> writes
>Hi Pythonites
>
>Could someone kindly post the URL for downloading the py2exe program and
>also the "Gordon McMillan's installer" that David referenced (see below).
>
>Thanks folks
>
>Andrew
....
even better try http://www.vex.net/parnassus/ where you will find all
things Pythonian in a searchable database.
-- 
Robin Becker


From martin at v.loewis.de  Sat Feb 16 16:44:23 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 22:44:23 +0100
Subject: using lists in C-extensions?
References: <3c6e711a.9871609@news.easynews.net>
Message-ID: <m33d01je4o.fsf@mira.informatik.hu-berlin.de>

stojek at part-gmbh.de (Marcus Stojek) writes:

> There is this strange "O" respectively "O&" which might
> be suitable, but I don't know how.

The appropriate format is inded O (or perhaps O!, passing PyList_Type),
passing the address of a PyObject* variable. Then, the PyObject*
will point to a PyListObject.

For this pointer, you can then use functions like PyList_Size,
PyList_GetItem, etc.

> So could anyone give an example or tell me where to look?

Just ask yourself what C functions of the standard library expect
lists. file.writelines comes to mind (although in 2.2, it uses the
METH_O optimization, which may complicate the understanding).

> EXAMPLE:
> 
> list=[]
> for i in range(100000):
>     list.append(i)
> 
> listnew=map((lambda x:x+1),list)  # this should be coded in C

This should look like

PyObject *
mymap(PyObject *self, PyObject *args)
{
   PyObject *l, *result, *item;
   int size, i;
   if (!PyArg_ParseTupe(args, "O!", &PyList_Type, &l))
     return NULL;
   size = PyList_Size(l);
   result = PyList_New(size);
   if (!result) return NULL;
   for (int i=0; i < size; i++) {
     item = PyList_GET_ITEM(l, i);
     if (!PyInt_Check(item)) {
        PyErr_SetString(PyExc_TypeError, "list contains non-integers");
        Py_DECREF(result);
        return NULL;
     }
     item = PyInt_FromLong(PyInt_AsLong(item) + 1);
     PyList_SET_ITEM(result, i, item);
   }
   return result;
}

Please refer to

http://www.python.org/doc/current/api/listObjects.html

for the exact meaning of the list API, in particular with regard to
reference counting.

If you want to support not only list, you need to use the abstract
sequence operations, notice that those have difference reference
counting semantics.

Regards,
Martin


From phr-n2002a at nightsong.com  Tue Feb 19 23:45:10 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 20:45:10 -0800
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org>
Message-ID: <7xd6z07odl.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> I solved the other puzzles too...  They still didn't hire me.  :)

It's kind of interesting.  They wrote me back saying my solutions were
correct and invited me to send a resume.  I replied basically that I
wasn't really looking for a job in Boston, but had fun with the
puzzles.

They're spending quite a lot of $$ running those banner ads and I
doubt if they have that many openings, so they must be looking for
"perfect" candidates.  I'm sure they get quite a few responses.


From cliechti at gmx.net  Thu Feb 14 17:34:12 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 14 Feb 2002 23:34:12 +0100
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com>
Message-ID: <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82>

Paul Moore <gustav at morpheus.demon.co.uk> wrote in
news:efao6us1tt3ut2hp6aor19p7qfqad3k4t1 at 4ax.com: 
...
> Hmm. Looks more messy than I'd hoped. Needs a bootstrap module, and
> then extra hacking to handle open/read. And I guess the fact that 
> I'm doing this manually means that there's not much chance of 
> something like a wxPython toolbar "just working" with icons stored
> in the archive. 

well you could pack the stuff in a zipfile and retreive the data with 
the zipfile module. you get the data as string but you can wrap that 
again in a cStringIO object to get a file-like object.

but then wxWindows Bitmap.LoadFile wants real file...
but: """wxBitmapFromXPMData(listOfStrings)  Create a bitmap from a 
Python list of strings whose contents are XPM data""" so you would be 
limited to XPM images.

py2exe uses a zip as storage and adds a loader in front of it a la 
self-extract zip. those files can be opened with e.g with WinZip maybe 
you could add files this way too (never tried myself)?

as described above getting data from a file-like object to an 
extestion in c isn't that painless. of course you could allways create 
a temp file (os.tmpfile()), copy the stuff from the zip to that file 
and use it (Bitmap.LoadFile(temp)) etc.


> I guess it's not worth the hassle.

the following could be worth a try:
  -make a py2exe dist
  -add data files

the running programm does:
  -extract data to a temp file for extensions that need a file
   (i.e. for Bitmap.LoadFile())
  -store thedata in a cStringIO object for pure python stuff
   (config files, readme.txt for an help dialog etc.)

> The reason I liked the Java approach, BTW, is that a complete
> application comes in a single file, which you can just double click 
> on and it works. Nothing to unpack, put it anywhere you like, run 
> off a floppy/CD/zip disk, no dependencies (other than needing Java
> installed). But it's just a convenience thing - not enough to endure
> writing all my code in Java :-)

yes its easy when you have a jar but i don't understand why "java" 
just doesn't cut off ".class" so that you could launch those files by 
doubleklick too....

> Nice to keep an eye on what the competition can do, though...

<sarcasm>
yeah, they finaly made it to include it a select like interface for 
e.g. sockets "to write high performace/scalable servers" and regex'es 
are now included in the 1.4 release thats out these days, wow. again 
ten years and they *might* catch up to python...
</sarcasm> ;-)

chris

-- 
Chris <cliechti at gmx.net>



From dongkui at huawei.com  Mon Feb 25 20:14:42 2002
From: dongkui at huawei.com (dongkui)
Date: Tue, 26 Feb 2002 09:14:42 +0800
Subject: ask help about raise, thank u
Message-ID: <000c01c1be62$f84aab00$490a0b0a@huawei.com>

pls give me one example about raise which have three para, and the third para is traceback obj

like 
raise "a", "b", c #c as traceback object

thank u
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020226/ccf73d23/attachment.html>

From emile at fenx.com  Wed Feb 27 01:49:28 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 22:49:28 -0800
Subject: ask help about raise, thank u
References: <mailman.1014686626.31452.python-list@python.org> <c76ff6fc.0202260144.70e52642@posting.google.com> <a5g41c$70hbr$1@ID-11957.news.dfncis.de> <c76ff6fc.0202261244.2abd9a02@posting.google.com>
Message-ID: <a5hvru$7ichb$1@ID-11957.news.dfncis.de>

"John Machin"
> "Emile van Sebille"
> > Isn't this what raise by itself does?
> >
>
> Quite possibly. I profess ignorance -- see the disclaimer in my post.
> I posted merely in case the cognoscenti ignored an unfortunate whose
> post turned up (in Google Groups at least) encoded in base64.
>
> Could you please enlighten me on the following:
>
> If a bare 0-arg raise is (mostly) equivalent to:
>    x, y, z = sys.exc_info()
>    raise x, y, z
> (apart from the circular reference problem) then what is the 3-arg
> raise useful for?

Well, now you're raising the same question suspiciously absent but
implied in my original reply to the OP.  ;-))  I imagine it could be
used in a "smart" error handler, of course, but it's not clear to me.

>
> The reference manual says:
> """If a third object is present, and it is not None, it should be a
> traceback object (see section 3.2), and it is substituted instead of
> the current location as the place where the exception occurred. This
> is useful to re-raise an exception transparently in an except
> clause."""
>
> There is a very interesting 39-message thread back in April 1998 that
> actually documents the conception & birth of 0-arg raise --
> Google("raise traceback"). Seems like 3-arg raise is now not very
> useful, and the manual could perhaps point out that 0-arg raise now
> handles the "re-raise transparently" functionality.
>
> I note that of the 4 examples of 3-arg raise in urllib.py, one seems
> to have the circular reference problem.

Thanks for looking at those.  Now that you have, you know more than I
about their potential use. ;-)

>
> While seeking enlightenment: What mechanism did you use to search all
> of Lib/*.py for 3-arg raises?

I wrote a q&d in python, then poked through the results.  I didn't save
it, but here's one that does a better job recognizing lines:

import os, glob, tokenize, StringIO

def outsink(*args): pass

pyfiles = glob.glob(os.path.join(os.path.dirname(os.__file__),
'url*.py'))
for file in pyfiles:
    pgmlines = open(file).readlines()
    buf = []
    for line in pgmlines:
        buf.append(line)
        line = ''.join( [x.strip() for x in buf ]).strip()
        try:
            tokenize.tokenize(StringIO.StringIO(line).readline, outsink)
            if line.startswith('raise') and line.count(',')>1:
                print '\n\n%s:%s' % (file, line)
            buf = []
        except: pass


See you,

Emile van Sebille
emile at fenx.com






From andiaye at chello.nl  Mon Feb 11 14:38:05 2002
From: andiaye at chello.nl (Arona Ndiaye)
Date: Mon, 11 Feb 2002 20:38:05 +0100
Subject: Is pippy alive?
References: <Pine.LNX.4.33L2.0202111041410.8657-100000@cheetah.jpl.nasa.gov>
Message-ID: <005201c1b333$a2f5b440$eb368418@a2000.nl>

Does the Pippy team still need help with the GUI ? The LispMe Gui model
seems ok ? Has anyone contacted Fred Bayer ? Want me to do it ? I could also
totally duplicate the GUI, or create one...I'm a PalmOS coder and I also
worked at Palm at some point....
----- Original Message -----
From: <Michael.E.Ressler at jpl.nasa.gov>
To: "Jeffery D. Collins" <jcollins at boulder.net>
Cc: <python-list at python.org>
Sent: Monday, February 11, 2002 7:54 PM
Subject: Re: Is pippy alive?


> On Sun, 10 Feb 2002, Jeffery D. Collins wrote:
> > With regard to floating point objects, others have also asked for this
> > feature to be reenabled.  I'm now looking into the code to determine
> > the extent of the necessary modifications.
>
> I'm no programmer, but if you need an enthusiastic beta tester, I hereby
> volunteer. I am also on the documentation teams of a couple of open-source
> projects. If you should need assistance with the docs, perhaps I can lend
> a hand there.
>
> Concerning floating point, I assume you are aware of the "mathlib" package
> for the Palm? It comes with many other opensource Palm apps, and is built
> into the Visors if I recall correctly. Having Pippy use mathlib for most
> of the higher order math functions (sin, cos, log, etc.) should save quite
> a bit on space; of course, you will still have to provide the core
> floating point add, mult, etc. since the Palm doesn't have an FPU. I
> played around with Quartus Forth for a while, so I know the trouble
> floating point can cause on a Palm. It's also because of (the drudgery of)
> Quartus Forth that I got interested in Pippy :-)
>
> Anyway, let me know if I can contribute in some non-programming capacity.
> I'll gladly repay my "whine" with a little bit of work. I actually became
> aware of Python through Pippy, and its a neat language that I've already
> begun to use in a number of work projects. Being able to do the same
> things (at least the simpler ones) on my Palm would be way too cool :-)
>
> Mike
>
> --
> Dr. Michael Ressler
> Research Scientist
> Astrophysics Research Element
> Jet Propulsion Laboratory
> ressler at cheetah.jpl.nasa.gov
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list





From sdm7g at Virginia.EDU  Mon Feb  4 16:10:13 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Mon, 4 Feb 2002 16:10:13 -0500 (EST)
Subject: [OSX] milter won't build
In-Reply-To: <644f6688.0202041212.1f3cfabf@posting.google.com>
Message-ID: <Pine.OSX.4.43.0202041601140.415-100000@d-128-61-180.bootp.virginia.edu>


On 4 Feb 2002, rdack wrote:

> now i get:
> cc -bundle -flat_namespace -undefined suppress
> build/temp.darwin-5.2-Power Macintosh-2.1/miltermodule.o -lmilter -lsm
> -o build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
> /usr/bin/ld: can't locate file for: -lsm
>
> what file is it looking for? why can't it find it? what is option
> '-lsm'? could i just remove it? what file would it be in?

"-lsm"  means it wants to link to  libsm.[ a, so, ... ].
Without a "-L" switch to add other library search directories, the
default is to look in /usr/lib.

I have no idea what "libsm" is -- that's not part of Darwin/Mac OS X.
Some common libs, like libm for example, are merged into the system
library, so you can just leave off "-lm" and it'll link to the
math libs without it -- so it's worth trying if you don't know
what it is or where it comes from.

-- Steve Majewski





From owen at xastrox.xwashingtonx.edu.invalid  Fri Feb 15 13:23:46 2002
From: owen at xastrox.xwashingtonx.edu.invalid (Russell E. Owen)
Date: Fri, 15 Feb 2002 10:23:46 -0800
Subject: Warning about "for line in file:"
Message-ID: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu>

I tried taking advantage of
  for line in file:
(a recent addition to the language) and ran into some odd behavior.

I read some lines using one for loop, to get to the start of the data of 
interest. E.g.:
for line in file:
  if startingcondition(line):
    break

I then some more lines from the same file using a similar for loop, but 
found that the lines were not contiguous! Instead there was a huge gap 
between the two sets of reads, as if the first for loop read in a lot of 
extra data that it never spit out.

I am guessing this is some form of optimization, but it is very 
surprising behavior. I expected "for line in file:" to act as an 
iterator -- read some now, read some later, whenever you read you get 
the next line.

Anyway, beware of treating file too much as an iterator.

There are plenty of workarounds. In my case the fileinput module 
actually was a better match for my needs (interleaving data from several 
sets of files). The readline or xreadline file methods work fine, of 
course. One could presumably use the iterator returned by 
file.__iter__(), but I didn't try that.

-- Russell
-- 
Return      owen
address     astro
garbled     washington
in header   edu


From philh at comuno.freeserve.co.uk  Sun Feb 10 20:40:33 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 11 Feb 2002 01:40:33 +0000
Subject: Is Parrot dead?
References: <B88C9617.532F%Roman.Fischer@teleweb.at> <3C66CDEE.EDA48DC2@ix.netcom.com>
Message-ID: <slrna6e88h.sme.philh@comuno.freeserve.co.uk>

On Sun, 10 Feb 2002 15:45:50 -0400, Antaeus Feldspar <feldspar at ix.netcom.com> wrote:
>"Ing. Roman Fischer" wrote:
>> 
>> For a long time I do not find informations about Parrot (the programming
>> language). O Reilly still estimates July 2001 for the "Programming in Parrot
>> in a Nutshell"-book and even Amazon (which is not very shy to offer books
>> long before they are available) does not have it in its offer. I also did
>> not find any real complete definitions of this language.
>> 
>> So is this language dead?
>> 
>> Roman
>
>I think I can clear up this confusion.  ^_^  It stems from a huge,
>elaborate April Fool's joke in April of 2001, in which Guido van Rossum
>and O'Reilly books and many other open-source leaders participated.  The
>full story can be found here:
>http://www.oreilly.com/news/parrotstory_0401.html .  
>
>However, that is not the end of the story.  There wasn't ever a real
>language named Parrot that combined the features of Perl and Python. 
>However...  there is now a virtual machine in development that should be
>able to run both Perl and Python code, as well as other interpreted
>languages.  Since the lead developer of this project is the very man who
>orchestrated the Parrot-the-programming-language hoax, Simon Cozens, the
>obvious name for this virtual machine was Parrot.

There is also a GUI-speciofication language I wrote called Parrot. (See
my website for details).

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From aahz at panix.com  Sat Feb  2 09:45:41 2002
From: aahz at panix.com (Aahz Maruch)
Date: 2 Feb 2002 06:45:41 -0800
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012618084.26466.python-list@python.org>
Message-ID: <a3gu2l$om7$1@panix2.panix.com>

In article <mailman.1012618084.26466.python-list at python.org>,
Jason Orendorff <jason at jorendorff.com> wrote:
>Aahz wrote:
>> Jason Orendorff wrote:
>>> Alex Martelli wrote:
>>>> 
>>>> Refusing to guess in presence of ambiguity is one of Python's
>>>> tenets, though.
>>>
>>> It's the language that doesn't guess.  The language designer is
>>> notorious for picking one thing, seemingly arbitrarily, from a number
>>> of reasonable alternatives.
>>
>> Mind expanding that?  (I could guess, but I don't want to flame you
>> until I know for sure what you're saying. ;-)
>
>My point was that when the BDFL has to choose between a number of
>reasonable alternatives, he often picks one without seriously trying to
>justify the choice.  (Hence "BDFL pronouncements" in many PEPs: 201,
>202, 223, 255...)  In PEP 255:  "No argument on either side is totally
>convincing, so I have consulted my language designer's intuition."

Yes.  And nine times out of ten, if you think you disagree with Guido's
intuition, you'll change your mind later.  As Alex (I think it was Alex)
pointed out a few months ago, Guido's ability to do good design work
vastly outstrips his ability to *explain* his thought processes.

Therefore, as long as you emphasize "*seemingly* arbitrary", you won't
get flamed by me.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From phr-n2002a at nightsong.com  Wed Feb 20 11:31:45 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 08:31:45 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com> <7xu1scew8m.fsf@ruckus.brouhaha.com> <slrna76g9l.rag.yhatakka@kenny.evolvit.fi> <7xeljg7h7b.fsf@ruckus.brouhaha.com> <RtNc8.17064$Zp2.1755158@atlpnn01.usenetserver.com> <NWPc8.55790$X64.20050659@news1.rdc2.pa.home.com>
Message-ID: <7xit8s868e.fsf@ruckus.brouhaha.com>

"Terry Reedy" <tjreedy at home.com> writes:
> def f(val):
>   global f
>   f = val and f1 or f2
>   return f(val)

This is not tail recursion.  The "global f" declaration shadows
the "f" that names the function.


From martin at v.loewis.de  Sat Feb 16 16:29:05 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 22:29:05 +0100
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <86n0yaw8u4.fsf@apo.hmotakef.int> <3C6E95FA.F95C914B@earthlink.net> <dlyb8.3101$tu6.303907@newsread2.prod.itd.earthlink.net>
Message-ID: <m37kpdjeu6.fsf@mira.informatik.hu-berlin.de>

"Chris Tavares" <christophertavares at earthlink.net> writes:

> Whitespace in XML is a little odd. It depends on the DTD/Schema wether it's
> significant or not. The normal solution is for the XML parser to have a
> "preserveWhitespace" setting, which will guarantee that you get everything
> (newlines and all). However, I took a spin through the minidom docs and
> couldn't find anything. Hmm... guess that's why it's a "mini" DOM. Have you
> tried using either SAX or 4DOM instead?

That is not the issue here. Preserving white space is only an option
for text content, and even if it is activated, line-break
normalization will still occur.

Regards,
Martin


From simc at xtra.co.nz  Tue Feb 12 04:33:48 2002
From: simc at xtra.co.nz (Simon Crosby)
Date: Tue, 12 Feb 2002 22:33:48 +1300
Subject: Networking Problems
Message-ID: <3C68E17C.3030009@xtra.co.nz>

Hi

I have recently tried programming a text network game using sockets in 
python as detailed in "Programming Python" 2nd ed. I developed this on 
linux and used telnet as a client. Then I tried connecting to the game 
server with windows telnet. This showed the text very differently, 
ignoring newlines and when prompting for input takes any character as a 
newline. It does not matter if the server is run on linux or windows.
Attached is my networking code.

eg :

  
   +---------------------------------------------------+
                                                             |                Are
 na Dungeon                      |
                                          | Our Motto: No quests, just fights!
              |
                     +---------------------------------------------------+
  
  
                Made by: Simon Crosby, simc at xtra.co.nz
  
  
                                                      Press:
>                                                            N - start a New chara
> cter
>     O - Open a existing character
>                                  Q - Quit Arena Dungeon
>  
>          
>                  >What? Invalid choice
>                                       >What? Invalid choice
>                                                            >What? Invalid choice

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: arena_net.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20020212/83dbc5a4/attachment.ksh>

From joshm at taconic.net  Wed Feb 20 21:14:54 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Wed, 20 Feb 2002 21:14:54 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com>
Message-ID: <3c7455a8_4@corp.newsgroups.com>

> In my experimentation 0, None, "", [], and () all seem to be
> considered false.  Just to confound me none of them are == to each
> other.  Coming from a Scheme world this is confusing.  Are there any
> other values that are false?  Why aren't () and "" the same thing?
> Don't both describe an immutable sequence of values (basically, why
> aren't strings either lists or tuples)?

Both are valid questions.

Philosophically, I would say that () and "" are different because tuples can
contain objects of any type, while strings can only contain characters.

As for why 0, None, "", [], and () are not == each other, it is because each
has a type associated with it.  Wouldn't you agree that objects of different
types are generally considered different?

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')





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


From anthony at ekit-inc.com  Wed Feb 20 20:16:05 2002
From: anthony at ekit-inc.com (Anthony Baxter)
Date: Thu, 21 Feb 2002 12:16:05 +1100
Subject: Idiots guide to code objects, frame objects and settrace? 
In-Reply-To: Message from martin@v.loewis.de (Martin v. Loewis) 
   of "20 Feb 2002 07:58:05 BST." <m3d6z063nm.fsf@mira.informatik.hu-berlin.de> 
Message-ID: <200202210116.g1L1G5511008@burswood.off.ekorp.com>


>>> Martin v. Loewis wrote
> "Tim Rowe" <tim at eltig.freeserve.co.uk> writes:
> 
> > Is there an idiots guide to code objects, frame objects and settrace
> > anywhere?
> 
> Anybody using those would look like a very smart idiot; so no.  Or,
> perhaps, the Python source code would be an acceptable guide to such
> people (it certainly is to me).

I find also that modules such as 'Lib/traceback.py' are very useful.

Anthony

-- 
Anthony Baxter     <anthony at interlink.com.au>   
It's never to late to have a happy childhood.




From peter.milliken at gtech.com  Mon Feb 18 19:07:19 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Tue, 19 Feb 2002 11:07:19 +1100
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <a4s51k$8pd1@news1.gtech.com>

There is no such effort to my knowledge. There is a tool called PyChecker
that attempts to fill the gap, I have no experience with it though.

The loose typing of Python is perceived as one of its big features :-).
There are regular discussions on the usefulness of Python for "larger"
projects. Many claim that they use it without any problems - I suspect that
it is from a background of ignorance and lack of metrics i.e. how much time
is lost in their projects to misnamed variables etc? The only way in Python
to find these is through (exhaustive) testing - such rigorous testing
*requires* you to develop an exhaustive test suite that is capable of
running periodically i.e. nightly etc to catch such errors. Whether the
people who support Python for larger projects have the discipline within
their project team to maintain such an effort is unclear - there has been no
mention of it at all. So I doubt that the proponents of Python in these
instances have any idea of the true cost of this language "feature" - they
appear to be wound up in how quickly they can produce a working prototype
(or whatever) so I suspect their day to day work involves very small,
manageable projects where the lost efficiency due to loose typing isn't
really that noticable for them i.e. they "live with it" :-).

Personally, I have used Python for several small, personal projects and the
odd "script" that searches text files - even here, the only perceived
advantage is the library support, I often wonder whether I wouldn't be
better off using my real language of choice (Ada) because of it's strong
typing - it too has interfaces to the standard GUI tools etc. My next
personal project will not be in Python, it is a "cute" language and allows
some nice facilities i.e. creating a script at run-time and then running it
can be very powerful in the correct circumstances - so I will probably use
an Ada-Python API for these situations and get the best of both worlds :-).
When I compile a Ada program cleanly I know that 95% of the bugs are out of
it - before I even start testing! :-) Of course, it takes me longer to write
up front but overall development effort (in my subjective opinion - since I
have no metrics :-)) is that it is much lower.

Good luck,
Peter

"Alexander Jerusalem" <ajeru at vknn.org> wrote in message
news:24c39b2c.0202181531.187fad4c at posting.google.com...
> Has anyone heard of plans to add static type checking to Python? I
> like Python very much but I don't use it for bigger projects because I
> fear the testing effort. For example I just refactored a rather large
> Java codebase. And when I change a method name for example, all I have
> to do is run everything thru the compiler and it will show me all
> places in other classes where this method is called, so I can go there
> and chenge the method call as well.
>
> In Python I'd have to test the whole software suite including each and
> every branch in each and every if statement to be sure that I didn't
> forget something. To build such a test suite would take almost as long
> as it takes to build the whole software and I'd have to maintain the
> test suite as well. That's just unafordable.
>
> On the other hand, Python is the best language that currently exists
> and I would like to use it on bigger projects not just for prototyping
> as some suggest.
>
> Regards,
> Alexander Jerusalem




From brian at sweetapp.com  Sat Feb 23 12:38:58 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sat, 23 Feb 2002 09:38:58 -0800
Subject: Python XML-RPC server as CGI
In-Reply-To: <a58hff$5ep4i$1@ID-76296.news.dfncis.de>
Message-ID: <003101c1bc90$f9768090$445d4540@Dell2>

Victor wrote:
>   Hello:
> 
>   I'm using Python 2.1 I want to run an xmlrpc server as a python CGI
> script, executed via Apache: it is possible with the Python 2.1's
standard
> xmlrpc library?

Python 2.1 didn't include an XML-RPC library, so I'm presuming that you
are referring to the PythonWare module xmlrpclib?

I've placed an enhanced version of the Python 2.2 XML-RPC server library
on my website:
http://www.sweetapp.com/xmlrpc/


The documentation for the 2.2 version is the most up to date:
http://www.python.org/doc/current/lib/module-SimpleXMLRPCServer.html

Here is a simple CGI usage example:

from SimpleXMLRPCServer import CGIXMLRPCRequestHandler

server = CGIXMLRPCRequestHandler()
server.register_function(pow)
server.handle_request()

Cheers,
Brian




From jeff at ccvcorp.com  Fri Feb 22 13:14:36 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 22 Feb 2002 10:14:36 -0800
Subject: Reloading modules
References: <mailman.1014332376.19219.python-list@python.org> <Z2rd8.77777$n73.1505518@atlpnn01.usenetserver.com>
Message-ID: <3C768A8C.975B2E3F@ccvcorp.com>

Steve Holden wrote:

>
> if file[-1] in "co":
>     file = file[:-1]

Given the source of 'file' in this case, this is probably not
significant, but there's a possible gotcha here if empty strings are a
possibility.

>>> empty = ''
>>> empty[-1]
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
IndexError: string index out of range
>>> empty[-1:]
''
>>>

Jeff Shannon
Technician/Programmer
Credit International




From fperez528 at yahoo.com  Sat Feb  9 12:24:38 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sat, 09 Feb 2002 17:24:38 +0000
Subject: Perhaps I am just dumb - On Perl, Python and context...
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com>
Message-ID: <a44eni$2hj$1@peabody.colorado.edu>

A little discussion on the issue of context which has been floating around my 
brain for a while.

wooks wrote:

>> One of the basic philosophies of Python design is that it is wrong for
>> the language to guess when presented with something that's ambiguous.
> 
> Which seems to be the opposite of the Perl philosophy - we will try to
> do the right thing.

Up front let me say I come from a background (in programming) of Perl (most 
recently, lots of other stuff before that) and having switched to Python, I 
refuse to go back. But I have given this particular issue some thought, 
because it is highly defended by the Perl community and in the long run I 
ended up understanding why *I* can't really live with it.

Perl is a language heavily inspired by one feature of human communication: 
the usage of context to resolve the inherent ambiguity in language. L. Wall 
is a brilliant man with a background in linguistics, and he has obviously 
thouhgt this through and through, coming to the conclusion that using this 
particularity of human languages in computer languages would be a good idea. 
At first I thouhgt the same, loved it and enjoyed some of the benefits it 
provides: conciseness, the immense expressive power of implicit knowledge, 
flexibility. You can write three lines of Perl which, when resolved in their 
current context, do amazing things.

The problem is, there is a fundamental difference between the way that 
'context' is created in human communication and in computer languages. For 
humans, context is something that occurs in *time*: you live through a 
conversation, your mind is aware of that context, and therefore you can try 
to resolve the ambiguities easily because you are immersed in the context. 
This is precisely what makes an out-of-context quotation so difficult to 
judge. In Perl, the computer doesn't have a problem because it *builds* that 
context, via a complex set of rules, when it parses the code.

But now think of the human working in perl: when you are writing the code, 
intensely concentrated on the problem, the context is clear to you. You use 
all the power of implicitness, the decisions are -at the moment- obvious to  
you and the result is beautiful, concise and powerful. Now it's 3 months 
later, you haven't  thought about this problem at all and you need to figure 
out what the code does. Good luck. At first sight, the code will be ancient 
egyptian: a magic incantation devoid of any apparent meaning. Until you spend 
the time to rebuild the context slowly, carefully, you won't be able to 
understand what the hell that code is supposed to do.

So in perl there are two options: either you maintain always the context 
around your code as human readable information, in the form of detailed 
comments, or every time you go back to your code you'll have to rebuild that 
context by re-understanding the problem. That's why most of my old perl code 
is either full of super-verbose comments or totally useless now.

Python is based on a fundamentally different choice: make each statement 
explicit and clear enough that the chance for ambiguity is as small as 
possible. This does mean producing slightly more verbose code, but in the end 
I find it's worth it. First of all, because the conciseness of perl is an 
illusion: if you want your code to be of any use in the long run, you better 
comment the hell out of it. Not so in python, I have found that my python 
code is very sparsely commented, yet extremely readable even months later. So 
much for the conciseness debate. And the gains in clarity, reusability and 
overall efficiency are priceless to me.

So in the end, this abstract discussion points to a very pragmatic 
conclusion: I prefer python because I can be more productive *in the long 
run* than I ever was in Perl. Maybe in Perl I can whip a 3 liner that does 
something that will take a bit longer to do in python. But life is more than 
3-liners, and it's in the big, complex problems that extend in time where the 
differences really matter.

So there you have, my rather philosophical thoughts about the issue you 
brought up. I know you asked about 'down to earth' information, and I hope 
this doesn't discourage you :) Believe me, this newsgroup is *very* useful 
for super-practical matters of syntax and semantics. It's just that many of 
its participants are interested enough in languages that these 
meta-discussions also generate quite a bit of postings.

Welcome!

Fernando.


From James_Althoff at i2.com  Thu Feb 28 18:17:17 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 28 Feb 2002 15:17:17 -0800
Subject: PEP 276 (was Re: Status of PEP's?)
Message-ID: <OF85647636.477BD4AC-ON88256B6E.007DD753@i2.com>

[Carel Fellinger]
> you being such a heavy weight that I fear
> Guido might be lured into thinking this to be a good idea
> that I felt compelled to react in public.

Well, thanks (I think <wink>), but I can assure you that I hold no special
brownie points and have zero ability to "lure" Guido (or anyone on the
Python team) into anything.  So, no need to lose any sleep over that one.
(On the other hand, if anyone knows *how* to "lure" any of them, be sure to
let me know <wink>).


[Samuele Pedroni offers this perspective sans endorsement <wink>]
> in mathematics it is not uncommon to
> identify the natural numbers with the set
> of their predecessors
>
> 0 = {}
> 1 = {0} = {{}}
> 2 = {0, 1}  {{},{{}}}
> etc

[Carel Fellinger]
> Not only newbie unfriedly, you need a firm grasp of modern set
> based number theory to appreciate it, it seems.

As fond as we all surely are of the neat mathematical underpinnings
suggested here and before, I believe that most programmers -- regardless of
skill level (and pedigree in the field of mathematics) -- would be able to
understand -- and certainly easily remember the second time -- that
iterating on n means taking each of 0,1,2, ..., n-1 in turn; this without
having to think about -- much less have "a firm grasp of" -- number theory
(modern, set-based, or otherwise <wink>).  At least if one has enough
background to understand "iteration" to begin with.

Let's try that again.

Q: What does it mean to iterate a number n?

A: Take each of 0,1,2, ..., n-1 in turn.

Q: When would I want to do that?

A: When you need the indices of an indexed-collection, for example.

(No set theory needed <wink>).

I certainly acknowledge the feelings of those who don't like the suggested
idiom.  At the same time, claims of its alleged *great difficulty* seem
overblown.


[Carel Fellinger]
> So please explain why you think it natural to loop over a number.

The "rational" section of PEP 276 talks about iterating through the indices
of an indexed collection.  These, as we know, are related to the length (an
integer) of the indexed collection.


[Carel Fellinger]
> here in your PEP you make it look like some minority
> of nitwits objected

I don't remember writing anything about nitwits.  But that does remind me
of the joke about "when failing to find a needed wit, they substituted two
half-wits, instead".  Seriously though, I don't recall and didn't intend
any insult to those who don't like PEP 276's suggestion of an iterator for
integers.  Please feel free to point out the actual text that you find
offensive.


[Aahz Maruch]
> My primary complaint about the PEP is that the following idiom is
> awkward, ugly, and prone to misunderstanding:
>
>   if i in 5:

I will note this.

However given that current Python seems able to do type-checking on strings
used with "in" as in, for example:

>>> 1 in 'spam'
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: 'in <string>' requires character as left operand
>>>

it's not obvious (to me, anyway) why the same (throwing a TypeError, that
is) could not be done for ints assuming one believes that "i in 5" is so
bad.   On the other hand, as noted by Bjorn Pettersen, "if i in
len(mylist):" seems like a pretty straightforward and handy thing to write.
I would use it were it available.  And I don't think it would confuse any
of the developers I work with. YMMV


[Remco Gerlich]
> Compare:
>
> for i in 6: print i
> for i in 6,7: print i
> for i in 6,7,8: print i
>
> You wouldn't think the first line does something different, would you?

[Emile van Sebille]
> Is that really any different from what we have now:
>
> for i in "Spam": print i
> for i in "Spam","and": print i
> for i in "Spam","and", "Eggs": print i

[Andy Gimblett]
> Yes, it really is.  "Spam" is a sequence, whereas 6 isn't.

I agree with Emile on this one.

>>> for s in 'spam': print s
...
s
p
a
m
>>> for s in 'spam','eggs': print s
...
spam
eggs
>>>

Two different results.  One needs to understand how and why.  Sometimes one
wants to treat a string as a unit, other times as a sequence.  It's useful
and convenient to be able to do both.


[Courageous]
> for i in [0:len(mylist):1]:

[Bjorn Pettersen]
> And exactly this notation has been rejected in PEP 204 --
> basically because it's butt-ugly in a Perl'ish sort of way
> Besides, this goes under the section of solving the "interval"
> problem which PEP 276 *explicitly* says it is not doing (and which
> I personally don't find very compelling, especially if PEP 276 is
> adopted).

I agree with Bjorn.  A lot of alternatives for full-blown intervals were
proposed during previous discussions of PEP 276.  Many of them require new
syntax.  Much of such new syntax, though logical and nice in theory, ends
up looking pretty cluttered when used in common examples.  PEP 276 attempts
to reduce clutter for a common case with minimal change to Python (no new
syntax at all).  In any event, proponents of alternatives should take the
time to write a PEP.


[David Eppstein]
> Maybe it's as natural as Perl's ability to concatenate numbers?

I don't appreciate your sarcasm.  Maybe others do.


Jim




From shriek at gmx.co.uk  Tue Feb 26 21:55:47 2002
From: shriek at gmx.co.uk (Stephen)
Date: 26 Feb 2002 18:55:47 -0800
Subject: How to have multi-threaded server log to one file ?
Message-ID: <97ae44ee.0202261855.15c97708@posting.google.com>

Consider a very simplistic server using the SocketServer module ~

import SocketServer

class MyServer(SocketServer.ThreadingTCPServer):
    allow_reuse_address = 1
    
class MyHandler(SocketServer.StreamRequestHandler):
    def handle(self):
        # Do something interesting.
        # Log any errors <------------------------- HOW ? 

def start(host, port):
    server = MyServer((host, port), MyHandler)
    server.serve_forever() # serve forever
    

When the server is run, each request from a client is handled by
MyHandler() in a separate thread. If multiple requests are being run
at the same time, how should I write my code so that they cal all
write to the same log file without conflict ?  Should I use a global
variable for the logfile, as below ~

logfile = open("errors.log", "rw")

class MyServer(SocketServer.ThreadingTCPServer):
    allow_reuse_address = 1
    
class MyHandler(SocketServer.StreamRequestHandler):
    def handle(self):
        # Do something interesting.
        global logfile
        logfile.write("fubar")


Or is there a more elegant solution that would take care of
concurrency ?
Is there a good document where I can learn how to do this in Python ? 

Thank you,

Stephen


From shalehperry at attbi.com  Fri Feb 22 15:30:18 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Fri, 22 Feb 2002 12:30:18 -0800 (PST)
Subject: subclassing float (or not)
In-Reply-To: <a567mp$4vkk5$1@ID-59885.news.dfncis.de>
Message-ID: <XFMail.20020222123018.shalehperry@attbi.com>

On 22-Feb-2002 Thomas Heller wrote:
> I'm currently porting my own rewrite of Konrad Hinsen's PhysicalValue
> module to Python 2.2.
> Basically a PhysicalValue instance is a float conbined with a unit.
> Should I subclass float or should I not? What would I gain by doing
> this (except maybe some memory savings when I use __slots__)?
> 

Generally when you subclass you are saying "I am a special type of foo" where
foo is your parent.  I think in this instance it would make more sense to make
a new class which contains a float and a unit.  Give it methods so it acts how
you expect.



From donn at u.washington.edu  Tue Feb 26 20:20:05 2002
From: donn at u.washington.edu (Donn Cave)
Date: 27 Feb 2002 01:20:05 GMT
Subject: Problems with select
References: <pan.2002.02.26.04.17.29.538545.11917@onesite.org> <slrna7nfkh.2bd.rmunn@rmunn.dyndns.org> <pan.2002.02.27.01.03.24.922015.11917@onesite.org>
Message-ID: <a5hc85$o1s$1@nntp6.u.washington.edu>

Quoth Alain Tesio <alain at onesite.org>:

| Hi, thanks for the explanation, but select still returns files where
| there is nothing to read, after the end of the stream.

Yes, that will happen.  select doesn't care whether a read will return
"something" or "nothing", as long as it will return.  Once you have
gotten an empty string back (at "end of file"), you should remove the
file from the list of files.

And there's another problem that you won't see in your example, but may
be a problem in real life.  To see it , change your command to
"echo one; echo two; sleep 2; echo three"

You would expect "one" and "two" to show up together, but "two" will
actually show up with "three".  Because your readline() reads all the
pipe system buffer into a local process buffer, where select doesn't
see it.  You may be able to solve this by unbuffering the file objects,
or you might find that it simplifies matters to forget about the file
objects and use the file descriptors you already get with fileno(),
with os.read().

	Donn Cave, donn at u.washington.edu


From graz at mindless.com  Sat Feb 23 05:09:24 2002
From: graz at mindless.com (Graham Ashton)
Date: Sat, 23 Feb 2002 10:09:24 +0000
Subject: XML RPC Woes
References: <20020222.142716.1960709859.1525@d57-47-79.home.cgocable.net>
Message-ID: <FPJd8.417$PE2.43414@stones>

On Fri, 22 Feb 2002 19:27:16 +0000, Mark Nenadov wrote:

> However, all of a sudden I have not been
> able to get it to work at all. The server just keeps on returning "500
> Internal Error" now.

I usually find that's caused by a bug in my code that knackers things up.
Why don't you post an example?

-- 
Graham


From mhammond at skippinet.com.au  Fri Feb 15 08:02:11 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 15 Feb 2002 13:02:11 GMT
Subject: Windows Message Hooks
References: <8852d33c.0202140224.13840b45@posting.google.com> <3C6C4920.4000301@skippinet.com.au> <8852d33c.0202150309.187b311f@posting.google.com>
Message-ID: <3C6D073E.9030708@skippinet.com.au>

wooks wrote:
> Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6C4920.4000301 at skippinet.com.au>...
> 
>>wooks wrote:
>>
>>>Can anyone direct me to a resource where I can find info Python
>>>support for the above.
>>>
>>>Thanks.
>>>
>>I don't think you will find direct support.  Adding such support to the 
>>win32gui module should not be difficult.

> . I came across this book.
> 
> http://www.oreilly.com/catalog/subhookvb/desc.html
> 
> I am wondering whether it would be possible to apply any techniques I
> learn from it to Python via the win32gui module that you mentioned. I
> presume the book tells you what API calls to make and you just make
> them from Python rather than VB. Yes?

Probably - but without seeing the book I really can't be sure.

Mark.





From emile at fenx.com  Wed Feb 27 15:08:36 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 12:08:36 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com> <7xk7sznsoa.fsf@ruckus.brouhaha.com> <3C7D3B2C.80A97BA2@ccvcorp.com>
Message-ID: <a5jem3$7toi7$1@ID-11957.news.dfncis.de>

"Jeff Shannon" <jeff at ccvcorp.com> wrote in message
news:3C7D3B2C.80A97BA2 at ccvcorp.com...
>
>
> Paul Rubin wrote:
>
> > Also, there's no way to make sure that the "del temp" at the end of
> > a block actually gets run, unless you put try...finally around it,
> > which is a big mess.
>
> I may be in way out of my depth here, but I don't think that this is
true.  If
> you explicitly delete an object, then that name should be guaranteed
to be
> unbound.

Unless, of course, the name shadowed something.  Consider:

>>> list = range(3)
>>> del list
>>> list
<built-in function list>

--

Emile van Sebille
emile at fenx.com

---------



From eddie at holyrood.ed.ac.uk  Fri Feb  1 10:03:32 2002
From: eddie at holyrood.ed.ac.uk (Eddie Corns)
Date: Fri, 1 Feb 2002 15:03:32 +0000 (UTC)
Subject: Why it does NOT work on Linux ?
References: <mailman.1012571348.11942.python-list@python.org>
Message-ID: <a3eao4$gmo$1@scotsman.ed.ac.uk>

"A" <printers at sendme.cz> writes:

>I have the following part of program that finds ItemID numbers.
>Here, for example, are two
>146759 and 146700 .
>This program works well under windows but on Linux it does not 
>find any number. Can you please help?
>Thanks.
>Ladislav

>####################
>import re
>Text="""<tr BGCOLOR="#FFFFFF">
>                      <td valign="top" align="left"><a 
>href="lead.asp?ItemID=146759">[CN] Oak, Foiled &amp; Antique 
>Furniture</a></td>
>                      <td valign="top" align="center">18/12/2001</td>
>                    </tr><tr BGCOLOR="#FFFFFF">
>                    <td valign="top" align="left"><a 
>href="lead.asp?ItemID=146700">[CN] Oak, Foiled &amp; Antique 
>Furniture</a></td>
>                      <td valign="top" align="center">18/12/2001</td>
>                    </tr>"""

>IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
>Results=re.findall(IDs,Text)
>print Results

>##############
>t

Using the text exactly as above it wouldn't work because the patterns you're
searching for have newlines in the middle of them.  Changing either the
compiled pattern or the text works fine for Python 2.1 on Linux.

You probably don't want to hear people saying that this is a really hairy way
of parsing HTML anyway or talking about htmllib etc. but this is the sort of
reason why it's better to do it 'properly' :)

Eddie


From ajeru at vknn.org  Tue Feb 19 03:59:42 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 00:59:42 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <24c39b2c.0202190059.41874b92@posting.google.com>

The argument that exhaustive testing should be done anyway is
certainly true. I nevertheless tend to think that it's better to let
the compiler do as much as possible in terms of error checking and
then use testing to find the true runtime errors. It just takes much
less time than starting a test run every time just to find all my
typos. I don't know if there are test suites that let you run all the
branches of all your methods and show you all typos at a time. I
suspect that I'd have to start the test suite for each and every typo
but I'm not knowledgable enough when it comes to Python test suites to
be sure of that claim. I'm gonna look at what PyChecker can do for me.
Thanks for the tip.

There's another thing that bugs me with the missing static typing and
that is that it makes code less readable. I don't know what I can do
with a function parameter when I see it. For example, when I see a
function like def computeSalary(person)... I don't know at first sight
if person is the id of person, a person object and if it's an object
what its exact class in a Person class hierarchy is. I have to find
the place where the constructor is called to find that out. That's not
a big problem for small programs but if I have many classes that use
inheritance, it's really not easy to figure out what interface a
parameter actually supports. This is, however, a minor problem if
everyone on the team agrees to document parameter types in the doc
string. The language doesn't enforce it though.

And the final argument I have for static type checking is that it
enables method dispatching based on parameter types. In a statically
typed language you can create two methods that have the same name but
differ on the parameter types. The correct method will be called for
you depending on the type of the argument you pass in your call. That
makes for a quite flexible way of extending a program. You can just
add another method with the same name and another type to handle a
special case without touching the existing methods.

I think type checking could be added as an optional feature, like in
Dylan, without hurting the character of Python.

I'm currently using Python to write generators that produce Java code
from an XML representation. So the runtime system is type checked by
the Java compiler anyway. I tried to do the generator in Java first
but the text processing capabilities of Java are just terrible, so I
came to use Python and I like it for many reasons. It's just that I'm
using much more time on debugging than I used to in Java where I
didn't have to care too much about typos because the compiler
complained anyway...

I'm not, however, completely satisfied with static type checking as
well because I don't see why the compiler forces me to have a
parameter comply with a static interface that has 10 methods when the
method that uses the parameter acutally only uses one of those
methods. I want the compiler to check if the parameter type I'm
passing has what the method needs but nothing more.

Actually languages that support templates or generics or however
parameterized types are called in different languages, do just that.
They generate a method for the type that's actually used in a method
call if they see that the type has everything the method needs. I
liked C++ templates (and some other features in C++) but I don't use
C++ anymore because its lack of garbage collection makes it much less
productive than other languages.

Regards,
Alexander Jerusalem


From tanzer at swing.co.at  Mon Feb 25 02:50:06 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 25 Feb 2002 08:50:06 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: Your message of "Sun, 24 Feb 2002 22:19:01 PST."
             <bnlj7u8kteb8he8v9p403051ld9pd77cm2@4ax.com>
Message-ID: <E16fFte-0002H7-00@tswings.swing.cluster>

Tim Roberts <timr at probo.com> wrote:

> >Apparently German, as Perl, has many ways of saying the same thing.
>
> It's worth pointing out, just for those who might be thinking differently,
> that the only actual German in the previous posting is the word
> "ACHTUNG"...

Not quite. `Das`, `und`, `nicht` and several others are German words.
Still a tiny minority though.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From rich at worldsinfinite.com  Mon Feb 11 13:42:59 2002
From: rich at worldsinfinite.com (Rich Harkins)
Date: Mon, 11 Feb 2002 13:42:59 -0500
Subject: Global Variables in Modules
In-Reply-To: <20020211111025.63f4c956.jsc-news@bluesine.com>
References: <20020211111025.63f4c956.jsc-news@bluesine.com>
Message-ID: <200202111843.g1BIh0K03171@localhost.localdomain>

On Monday 11 February 2002 01:10 pm, Jeff Clement wrote:
> [stuff about importing a module and giving it globals snipped]

Although I would expect that this is generally considered bad form, you might 
look into execfile() then.  Execfile() loads and executes a file within a 
globals and locals dictionary that you optionally provide (if only Perl let 
me specify namespaces to an eval so easily).  

Or, if you're going to run the thing multiple times you could (this is a 
little cleaner):

# Do this once...
code=compile(open(filename).read(),filename,'exec')

# For each run do *ONE* of the following...
exec code					# Simple form
exec(code,globals,locals)			# Alternative form...

The downside, of course, is that expectations that your called script has 
must be provided by the calling script.  However, this can be *VERY* handy 
for complex pythonic configuration scripts, where you want to provide some 
set of objects to be configured and let a simple python script be your 
program's rc file.

Hope this helps...
Rich





From thomas at weholt.org  Thu Feb 28 15:14:46 2002
From: thomas at weholt.org (Thomas Weholt)
Date: Thu, 28 Feb 2002 20:14:46 GMT
Subject: Simple threading thing
References: <Rlsf8.136$5vh.170729984@news.telia.no> <j4vgchebcy.fsf@informatik.hu-berlin.de>
Message-ID: <Wcwf8.1866$eJ6.34464@news2.ulv.nextra.no>

I need a thread, no forking. Should run in Linux and Win32. I need a thread
that functions in the background, running beside the main server, with a
reference to the main server so it can maintain states etc. in it. The
purpose of the process is to mainly remove expired sessions ( remember this
is a webserver ), log certain values and monitor the server, and work as
sort of a janitor.

Does that make it any clearer?

Thomas

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4vgchebcy.fsf at informatik.hu-berlin.de...
> "Thomas Weholt" <thomas at gatsoft.no> writes:
>
> > I got a webserver based on BaseHTTPServer using threads. When it starts
I
> > want to create a seperate process, a thread, that goes into the
background
> > and checks the server on given intervals, clearing expired sessions etc.
I
> > cannot get the created thread to pass control back to the server and let
it
> > continue. It just sleeps the specified interval and then does it thing
then
> > sleeps again. The server never gets into action at all.
> >
> > Any clues?
>
> What operating system? What do you mean by "create a separate process"?
>
> Regards,
> Martin




From duncan at NOSPAMrcp.co.uk  Mon Feb 18 04:33:56 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Mon, 18 Feb 2002 09:33:56 +0000 (UTC)
Subject: lambda to Sort Dictionary.items() by Value
References: <a4p47i$13ed$1@news.idiom.com>
Message-ID: <Xns91B95FAEB3D9Eduncanrcpcouk@127.0.0.1>

jimd at vega.starshine.org (Jim Dennis) wrote in
news:a4p47i$13ed$1 at news.idiom.com: 

> i = h.items()
> i.sort(lambda (k1,v1),(k2,v2): cmp(v2,v1) ) # sort by value, not by key
> i=string.join(map(lambda (x,y):x, i),"")  # extract letters from each
> return (h,i)
> 
Apart from the extremely nasty way i is reused, I would have said the 
biggest improvement here would be to get rid of the lambda calls 
altogether:

    i = [(val, key) for (key, val) in h.items()]
    i.sort() # by val
    i.reverse() # Get largest first.
    result = string.join([key for (val, key) in i], "")
    return h, result

Avoiding a comparison function for sort is generally a good idea if you can 
arrange it. You could also use Python 2.2's iteritems method, but I suspect 
it won't make much difference here. Another option for the first line (in 
Python 2.2) is:
    i = [(h[key], key) for key in h]


-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From pedronis at bluewin.ch  Thu Feb 28 00:05:14 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Thu, 28 Feb 2002 06:05:14 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu>
Message-ID: <3c7dbd8f_4@news.bluewin.ch>

> Samuele Pedroni wrote:
>
> > Maybe this is a useful insight:
> >
> > in mathematetics it is not uncommon to
> > identify the natural numbers with the set
> > of their predecessors
> >
> > 0 = {}
> > 1 = {0} = {{}}
> > 2 = {0, 1} {{},{{}}}
> > etc
>
> Ok, I won't argue this notation exists in some areas of mathematics, but
it's
> definitely _not_ a common one for the 'general public'. I've been doing
> mathematical physics work for years, so at least I have a passing
familiarity
> with common notations and conventions in math.
>
> And it's definitely not an intuitive one.
...
> Python is a general purpose language, not one to be used only by number
> theorists. Putting at a level as basic as the integers a convention which
> allows things like 'for i in 3' to be valid iteration code really sounds
> bizarre and counter intuitive to me.
>
> But maybe it's just me.
>

Stop please, I have said that I support the PEP: no

this was just an insight about the only  POV I could find to make
sense out of the proposal, btw it seems that the supporters
have not done such an effort.

For me it is fair if someone says that if this is the kind of argument
needed to make sense out of the proposal, then the proposal
cannot fly. Or if the supporter want to use it or are stimulated
to find something better or start a crusade to teach model theory to
everyone.

Look at it as purely informative.

regards, Samuele Pedroni.









From starx at pacbell.net  Tue Feb 12 04:40:55 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 12 Feb 2002 09:40:55 GMT
Subject: Best python way to do 'object messaging'
References: <mailman.1013462838.13259.python-list@python.org>
Message-ID: <Xns91B3111ABE559RASXnewsDFE1@130.133.1.4>

Gustavo Cordova || Mon 11 Feb 2002 01:15:25p:

> Have to taken a look at some of the available MUDs? I believe
> some of them are scriptable as you say, you define an object
> class, which can be anything in the game; you define lotsa
> properties and methods which control the way it interacts
> with the environment.
> 
> The objects you create can be weapons, plants, scrolls,
> monsters, seemingly anything.
> 
> Good luck on your quest :-)

Well, text adventures are the closest thing to this, and I have played 
with tools for making them on and off for years, though i've never 
actually made anything worth a damn. TADS is one of the major languages 
for creating modern text adventures (yes, people still do), and it 
(along with an obscure macintosh language 'Prograph') was where I 
finally 'got' a lot of OO concepts, and i bring a lot of ideas about 
things like object containment and other structural stuff from here.

http://www.tads.org/ for info on this
http://www.firthworks.com/roger/parsifal/index.html for links to 
everything that has anything to do with the Interactive Fiction scene.

Hm, i just realized something wierd. I see quite a resembelence of 
differences between Tads and Inform (the other major IF language, seems 
to be object oriented line noise to me...), and Python and Perl...

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From robin at jessikat.fsnet.co.uk  Thu Feb 28 19:26:14 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 1 Mar 2002 00:26:14 +0000
Subject: win32+popen4
Message-ID: <$4LupWAmqsf8Ew0o@jessikat.fsnet.co.uk>

What's the correct way to give a command to popen4 under win32 when the
command/arguments may have spaces in it. EG I want to use

"c:\program files\dimbo\bongo.exe" "c:\program files\dumbo\my file.dat"

as the command.
-- 
Robin Becker


From logiplexsoftware at earthlink.net  Thu Feb  7 15:57:32 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 7 Feb 2002 12:57:32 -0800
Subject: Autocoding evolves from........
In-Reply-To: <200202071740.g17HeXGo019034@ratthing-b246.strakt.com>
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk>
	<972.799T919T11694325threeseas@earthlink.net>
	<slrna5rpmb.19d.philh@comuno.freeserve.co.uk>
	<3943.801T1063T14354065threeseas@earthlink.net>
	<3C616C5A.4D1BBE1C@ccvcorp.com>
	<2300.802T2206T14055804threeseas@earthlink.net>
	<3C62190A.A54E9A57@engcorp.com>
	<00b901c1af9e$9de1d080$0101010a@local>
	<200202071740.g17HeXGo019034@ratthing-b246.strakt.com>
Message-ID: <20020207125732.028c896b.logiplexsoftware@earthlink.net>

On Thu, 07 Feb 2002 18:40:33 +0100
Laura Creighton wrote:

> 
> Voting people out of a newsgroup, as a concept, disgusts me.
> 
> Laura Creighton


Okay, we have a dissenter in our midst: who wants to vote Laura off the
list?


(Hi Laura - I couldn't resist ;)


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From phr-n2002a at nightsong.com  Thu Feb 21 13:28:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 10:28:36 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014314855.32175.python-list@python.org>
Message-ID: <7x8z9m665n.fsf@ruckus.brouhaha.com>

Gustavo Cordova <gcordova at hebmex.com> writes:
> > Anyway, =/== is a dead horse--didn't we agree that using := for assigments
> > in expressions totally fixes it?  So there's no reason to keep coming
> > back to it.
> > 
> 
> Er, no, we didn't :-)
> 
> _YOU_ and _I_ agreed that it's more readable and less
> error prone than simply using "=" in an expression.
> That's it. Don't include me in your revolution.

Yes, by "we" I meant you and I.  So in your opinion, does := fix the =/==
problem or not?

> > If you want Python to catch more typos for you,
> > you're better off asking for Python to check for
> > variable declarations (I don't mean static types,
> > I just mean "local x") and flag undeclared variables
> > like you can with perl's "use strict" option.
> > That will prevent a lot more bugs than := is likely to cause.
> > 
> 
> Now this is just plain nasty.

It's not intended as nasty.  Misspelled variable names are a source of
bugs in Python programs, to a great extent than in languages where you
have to declare variables.

> > And it fills more space on the screen, ...
> 
> So, get a larger monitor, or use a smaller font, or make the editor
> window larger, or change your text mode (if using text mode), or ...

That reaches a limit and then you can't do it any more.


> > > But in the end, neither of our subjective opinions
> > > matter - Guido's does. That's probably why this
> > > language is so great.
> > 
> > Guido doesn't write our code; we do.
> 
> That's all nice and useful. I'm not defending Guido, but hey, let's
> keep this civil, ok?

I'm not attacking Guido.  I believe Guido set out to design a language
that could easily be used by a wide range of different programmers.
If instead he'd set out to design a language that mapped onto his own
particular thought structures (which is what Larry Wall set out to
do), he'd have probably gotten a much more idiosyncratic language,
like perl, but different.  Since nobody can predict perfectly how
other programmers are going to react to things, their reports of their
experiences are an important source of data.

> *If* it weren't for Guido, we/you wouldn't *have* Python to hack on.

What are you getting at?  I'm not attacking Guido.

> Second, Guido's a pretty smart person. So maybe
> he did give quite a bit of thought about the
> expression/asignment thing.

He gave thought to augmented assignments too, and decided not to put
them into Python 1.x.  Then it turned out that a lot of people wanted
them, so they went into 2.0.

> "Language" is the expression of a person's thoughts, and thoughts
> and ideas expressed in a Language tend to reflect the structure of
> that persons thoughts.
> 
> On the other hand, the inverse process also happens: "Language"
> tends to impose, with time, it's structure and semantics upon it's
> users mind. Like a kind of feedback. The Object modifying the
> Observer.

That theory was popular in linguistics (type "Sapir-Whorf hypothesis"
into Google) for a while, but was eventually discredited and is now
considered bogus, at least at the lowest fundamental level.

> One thing which is quite delightful about Python is it's
> high-quality, simple-to-use, and quite powerful, included
> library. Even if lots of modules have "amateur" origins, it's of
> very high quality, and usability.

Sure, but there's always room for improvement.

> So. If the Language (Python) changes, in a way that it doesn't
> impose a Neat, Orderly, Clean, programming style, then I really
> believe that the Language will --passively-- encourage sloppy code.

I don't see your idiomatic 6-line while loop as being neater or
more orderly as the straightforward 2-line version that uses :=
in the control expression.

> Thursday!!

One day to TGIF?


From jeff at ccvcorp.com  Thu Feb 21 17:16:02 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 14:16:02 -0800
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com> <3C74AAA0.D3A050EE@macrozon.net>
Message-ID: <3C7571A2.41D4E9E2@ccvcorp.com>


husam wrote:

> Jeff, thanks alot for the enlightening remarks. I will try them. By the way,
> "glob.glob('*pdb') does read the files properly, but when I pass each of the to
> the other program for processing it doesn't work. I only work with files red
> with "popen".  A side problem when using "popen('ls *pdb')" I get an error
> message saying that "ls" got to many arguments. I got the same error in the
> shell. So I worked around this problem by copying all the *pdb file to a
> separate directory.

I'm not sure exactly what you're getting from your popen('ls...'), but glob()
returns the filenames with no path info.  You may simply need to append the path
back onto the filename, which you can do with os.path.join(directory, file) or
perhaps os.path.abspath(file) -- join() lets you specify any directory, while
abspath() normalizes its filename based on the current directory (which can be set
with os.chdir() ).

Jeff Shannon
Technician/Programmer
Credit International




From m.faassen at vet.uu.nl  Thu Feb  7 08:50:00 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Feb 2002 13:50:00 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org>
Message-ID: <a3u0m8$il4$1@newshost.accu.uu.nl>

Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
> The time has come, my friends and colleagues, to bring the issue at hand
> to a vote:

> Should Timothy Rue go away?

> All in favor, post AYE.  All opposed, post NAY.

> (Yes, this is a waste of bandwidth, but in my opinion not as bad as Rue.)

Please, I think it's far worse than Rue. I don't want this group to start
turning into a group that 'votes' others off. Either ignore him or killfile
him or just talk to the fellow. Having a vote is only a weird popularity 
contest more about the people doing the voting than the person being voted
on in my opinion, and we could do without.

I think this group is already going down in friendliness rather rapidly; 
I may be misguided but I think we'd just have enjoyed Rue or ignored him
if he'd stopped by a few years ago. It wouldn't have led to so much flaming as
it does now.

Finally, my opinion is that I agree with Rue, and we're all poseurs and 
auto-coding is a major breakthrough in programming, and I'm waiting
eagerly for Rue to release his long awaited system. :)

Poseur-ly yours,

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


From peter at engcorp.com  Sat Feb  2 23:57:09 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 02 Feb 2002 23:57:09 -0500
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <b34d3835.0201290851.f1bf021@posting.google.com> <a3hnft$cgu$1@newshost.accu.uu.nl>
Message-ID: <3C5CC325.7B3FC988@engcorp.com>

Martijn Faassen wrote:
> 
> ... And of course
> in comp.lang.python it's very frequent some thread on some boring issue
> that's been doing over and over again will grow some interesting
> outcroppings on how morality evolved or how there could be life on
> other planets and why the capslock ended up in the bizarre place it is
> today (why, anyway?).

Come on, that's *obvious*.  Look at your keyboard: if you took
away the Caps Lock key, there'd be a big empty space right there!
Where else would they have put it?  Sheesh...

-Peter-ly yr's ;-)


From pan-news at asheeshenterprises.com  Wed Feb 27 17:51:25 2002
From: pan-news at asheeshenterprises.com (Asheesh Laroia)
Date: Wed, 27 Feb 2002 22:51:25 GMT
Subject: Regular expression problem
Message-ID: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com>

I have some SGML input (PageMaker 6.5 tagged text), and I want to be able
to recognize (and delete) a tag.  That tag looks like:

	<@Trap Body text:>

It may also look like <@Trap Body text: useless-data>.

So, I tried the regular expression r"<@.?>".  That doesn't match the
above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
using, and why doesn't this work?

Thanks in advance!

-- Asheesh Laroia.

PS: An example of the tag "in the wild" is the following string:

<@Trap Body text=<FONT "Times"><CCOLOR
 "Black"><SIZE 11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE 
58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK 0><CLEADING -0.05
><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE 0><G+AFTER 0><GALIGNMENT "justify
"><GMETHOD "proportional"><G& "ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW 1><GKORPHAN 
1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0 25>>


From justin at iago.org  Tue Feb  5 11:06:14 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 05 Feb 2002 11:06:14 -0500
Subject: RFC -- HYGIENIC MACROS IN PYTHON
In-Reply-To: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> (Courageous's
 message of "Tue, 05 Feb 2002 05:36:47 GMT")
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <x7heovzzex.fsf@ra.iago.org>

Courageous <jkraska at san.rr.com> writes:

> I am currently experimenting with an implementation of hygienic macros
> in a custom Python parser as part of a larger experiment. 

Seems like a fun experiment, but I'm not sure where I think it ought
to end up.  Some languages (such as Scheme) have a clear need for
user-defined forms.  Python, both in structure and culture, does not
seem to be crying out for macros.

You are clearly aware of this; the reason I'm bringing it up is that
I'm not sure what the point of the experiment is.  If it is really
just an experiment on the nature of macros, declaration, whitespace,
etc., then I think it's great that you're doing it and am interested
in the results.

As far as a more practical purpose, though...  I don't see it.

-Justin

 




From phr-n2002a at nightsong.com  Fri Feb  1 16:34:18 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 01 Feb 2002 13:34:18 -0800
Subject: Enhanced Generators - the difference between sources and iterators
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org>
Message-ID: <7xelk451k5.fsf@ruckus.brouhaha.com>

Oren Tirosh <oren-py-l at hishome.net> writes:
> I'm not sure what you expect two-way generators to be used for. 

Same thing as coroutines?


From brian_l at yahoo.com  Tue Feb 12 23:34:01 2002
From: brian_l at yahoo.com (Brian Lenihan)
Date: 12 Feb 2002 20:34:01 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013555062.4397.python-list@python.org>
Message-ID: <afe3b146.0202122034.35dd1142@posting.google.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message news:<mailman.1013555062.4397.python-list at python.org>...
> > From: brian_l at yahoo.com [mailto:brian_l at yahoo.com]
> > "Steve Holden" <sholden at holdenweb.com> wrote in message 
> > news:<ZND98.140690$KM2.6229113 at atlpnn01.usenetserver.com>...
> > > "Brian Lenihan" <brian_l at yahoo.com> wrote in message
> > > news:afe3b146.0202100036.f221c9a at posting.google.com...
>  
> > > > I don't throw like a girl right-handed, so maybe I am actually
> > > > ambidextrous.
> > > >
> > > Not to mention gender-biased? :-)
> > 
> > Of course I'm gender biased.  I love girls.  I'm also sure there are
> > plenty of girls who can throw better than I can.  I left out the fact
> > that I'm also a switch-hitter because I thought it would only add to
> > the confusion.
> 
> For those of use who have no interest in (I presume) baseball, what the hell
> is a switch-hitter?
> 
> Tim Delaney

I was referring to baseball.  A switch hitter can bat from either side
of the plate. The term is also a euphemism for being bi-sexual.


From kmj9907 at cs.rit.edu  Thu Feb 14 16:54:54 2002
From: kmj9907 at cs.rit.edu (Keith Jones)
Date: Thu, 14 Feb 2002 16:54:54 -0500
Subject: Mutex a file?
References: <a4h98k$ao2f$1@ID-98166.news.dfncis.de>
Message-ID: <3c6c2e10$1@buckaroo.cs.rit.edu>

You're going to need to lock the file. Here's something google found me for
cross-platform file locking. Hopefully, this will help.

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/65203

--
-Keith Jones
http://www.codeexamples.org





From gmfawcett at operamail.com  Fri Feb 22 11:51:32 2002
From: gmfawcett at operamail.com (Graham Fawcett)
Date: 22 Feb 2002 08:51:32 -0800
Subject: Record Audio (Automation)
References: <140415ad.0202211224.40d8b34b@posting.google.com> <a53nl0$6ld@news.or.intel.com> <52e5ab5f.0202212120.225912b8@posting.google.com>
Message-ID: <3d799735.0202220851.719507ca@posting.google.com>

> > The best thing here would be to write your own Visual Basic app and use the
> > MS Multi Media control...
> > 
> > Write your program to take command line input.. and well, thats about it..
> 
> This shouldn't be necessary as you should be able to use the 
> OLE automation interface to the MCI control from Python directly.
> However, the MCI control has lots of baggage you don't need.
> The mciSendString() API was specifically designed to do 
> multimedia from scripting languages. (Originally Macromind Director's
> MMP files.) If mciSendString isn't already wrapped, you should be able
> to use CallDLL. 

Sam Rushing's DynWin module (
http://www.nightmare.com/~rushing/dynwin/ ) helps to make real clean
code for MCI calls (and other Win32 things). It's not necessary for
MCI calls but I like the cleanness of the code.

Here's a Recording object for you to try, using DynWin.

import os
from dynwin import windll
winmm = windll.module ('winmm')

def send(s):
	 winmm.mciSendString( windll.cstring(s), 0, 0, 0)

class Recording:
	def __init__(self, name, dirpath,
			samplesize = 16, # or 8
			samplerate = 44100, # or try 22050, 11025, ...
			channels = 1): # or 2 for stereo, of course!
		self.name = name
		self.samplesize = samplesize
		self.samplerate = samplerate
		self.channels = channels
		self.filename = '%s.wav' % (os.path.join(dirpath, self.name))
		send('open new type waveaudio alias  %s' % (self.name))
		send('set %s bitspersample %s' % (self.name, self.samplesize))
		send('set %s samplespersec %s' % (self.name, self.samplerate))
		send('set %s channels %s' % (self.name, self.channels))
	def start(self):
		send('record %s' % (self.name))
	def stop(self):
		send('stop %s' % (self.name))
	def save(self):		
		send('save %s %s' % (self.name, self.filename))
		send('close %s' % (self.name))

def test():
	'''record 5 seconds of audio, and save to c:\temp\test.wav'''
	import time
	r = Recording('test', 'c:\\temp')
	r.start()
	time.sleep(5)
	r.stop()
	r.save()

if __name__=='__main__': test()

-- Graham Fawcett
-- Centre for Flexible Learning
-- University of Windsor


From graz at mindless.com  Sat Feb 23 05:14:19 2002
From: graz at mindless.com (Graham Ashton)
Date: Sat, 23 Feb 2002 10:14:19 +0000
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de>
Message-ID: <fUJd8.418$PE2.43612@stones>

On Fri, 22 Feb 2002 16:36:22 +0000, Ricky wrote:

> I intend to use ConfigParser module to manage my INI file. I can use get
> to retrieve data from my ini file, and use set to change option value in
> Python, just in python, after I closed python, I found that all changes
> I made in python lost. There is nothing altered in my ini file!
> 
> How can I make my ini file to be changed?

As you haven't posted any code I can't be sure, but I suspect from what
you've said that you've not called the write() method on your
ConfigParser object.

Read the module's docs.

-- 
Graham


From deathtospam43423 at altavista.com  Sat Feb 23 03:07:10 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 23 Feb 2002 08:07:10 GMT
Subject: newbie: another problem with file reading
Message-ID: <Xns91BEBD577BCA4AusYourStandingInIt@130.133.1.4>

In this instance, _before_ = 8221 , _after_ = 8305.

What this function should do, is read only up to position 8221 in _chat_, 
print a .gif, then read the rest of the _chat_ file. Why isn't it doing 
that? 


def get_posts(num, before):
    index = map(string.rstrip, open(chat_index).readlines())
    after = int(str(index[-1])) 
    g = int(str(index[-(num+1)])) #for some reason, need to add +1
                                  #maybe cuz of the way chat_log and      
	    	    	    	    	    #chat_index are set up
    print int(before)
    chat = open(chat_log, 'r')
    chat.seek(g) #seek to where the post the user wants is

    print chat.read(before) #should mean read up to previous position
    print "<img src = pointer.gif></br>"
    print chat.read() #read rest of file.
    print after
 


-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From gcordova at hebmex.com  Mon Feb 11 14:47:05 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Mon, 11 Feb 2002 13:47:05 -0600
Subject: Deep override
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A26@SRM201002>

I've never used it, but I can give it a try:

> 
> I have a module I'm writing for work that has a class that inherits
> from websucker.Sucker  I'm trying to override websuckers' inheritance
> of webchecker.Checker.checkforhtml so it will just return 1(I want it
> to download some files, not html).
> I've tried different ways to name in and even tried to def it (Python
> didn't like all those dotted names in a def).  Does anyone know how to
> get this to work?
> Thanks in advance for any insights.
> 

In your own program:
====================

import webchecker

def AlwaysOne(*args,*kw): return 1

webchecker.Checker.checkforhtml = AlwaysOne

class SomethingElse(webchecker.Checker):
	...etc etc...

====================

The important thing I'd think is to redefine
"checkforhtml" before Checked get's referenced
by your own class.

Dunno.

What does everyone else think?

Alas, I don't have my Python on this silly
employer's development machine.

Good luck!

-gustavo



From brown.2053 at osu.edu  Mon Feb 25 21:15:13 2002
From: brown.2053 at osu.edu (Locke)
Date: Mon, 25 Feb 2002 21:15:13 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu> <mailman.1014687041.32731.python-list@python.org>
Message-ID: <a5eqvu$79a$1@news.cis.ohio-state.edu>

so you are saying it sends the data in groups? how do i know how many calls
to recv i need to do before the server is done sending everything it needs
to send? because if i call recv after the server is done sending everything,
the program will just deadlock waiting. any suggestions? anybody know
anything about pop3?

thanks





From thomas at gatsoft.no  Fri Feb  1 04:51:23 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Fri, 01 Feb 2002 09:51:23 GMT
Subject: HTTP redirector/load-balancer ??
Message-ID: <vyt68.137$ti7.170992640@news.telia.no>

As a follow-up to my earlier posting "Simple threaded web-server based on
BaseHTTPServer" I'd like to know more about how people think the problems
discussed in the mention news-thread could be solved. I've worked with
java-based Enhydra earlier, and its Director ( a load-balancer tool. ). If I
run several BaseHTTPServers, preferrably on different machines, and had a
redirector/load-balancer that took each request and sent it to different
servers each time, possibly having some sort of priority related to each
server so that if traffic got high, the best servers could process more
traffic while slower servers got fewer requests etc, would this be a
solution? Even if I just run several instances of the same server, on
different ports on the same machine and the redirector sent requests to the
server-instances in turn?

What I'm getting at is; what is the problem with threads ( I know the
concept of threads, how to use them, what they do etc. , but not the
technical implications of them as implemented in Python) ? Is it that they
don't scale well if run by the same process? Would running several instances
of a server solve this if we had some sort of redirector to send requests to
different servers each time a request was handled?

In closing; have anybody made a web-server load-balancing tool in Python I
could use to test this ?

Thanks for everybodys input in my earlier posting.

Best regards,
Thomas Weholt




From bokr at oz.net  Sun Feb 10 14:25:18 2002
From: bokr at oz.net (Bengt Richter)
Date: 10 Feb 2002 19:25:18 GMT
Subject: Bug or feature? 'abc'.split('') rejects empty separator
Message-ID: <a46heu$8f2$0@216.39.172.122>

 >>> 'abc'.startswith('')
 1
 >>> 'abc'.endswith('')
 1
 >>> 'abc'.find('')
 0
 >>> 'abc'.rfind('')
 3
 >>> 'abc'.find('',1)
 1

AND:

 >>> z=list('abc')
 >>> z
 ['a', 'b', 'c']
 >>> ''.join(z)
 'abc'

BUT:

 >>> 'abc'.split('')
 Traceback (most recent call last):
   File "<stdin>", line 1, in ?
 ValueError: empty separator

Wouldn't it make sense to return list('abc') ?

or, ugly, but consistent with *find results
(finding the separator front, back, and between all):

 >>> ';a;b;c;'.split(';')
 ['', 'a', 'b', 'c', '']

Regards,
Bengt Richter



From maxm at mxm.dk  Wed Feb 13 17:25:34 2002
From: maxm at mxm.dk (Max M)
Date: Wed, 13 Feb 2002 22:25:34 +0000
Subject: Is making Python executables a reality?
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>
Message-ID: <3C6AE7DE.2010004@mxm.dk>

MDK wrote:

> What percentage of people make exe files?  Do most people have end-users
> download the Python environment?

Well from a population of 1 I can say that there is a 100% succes rate 
when using Gordons installer.

;-)

regards Max M



From peter at engcorp.com  Tue Feb 12 23:49:55 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 12 Feb 2002 23:49:55 -0500
Subject: import images (wxPython)
References: <5kja8.21380$Cg5.1177018@news1.calgary.shaw.ca>
Message-ID: <3C69F073.DBE85FF4@engcorp.com>

Andrew wrote:
> 
> Hello,
> 
> I was looking at some of the demo code and notice that they do an 'import
> images' in some files to get some predefined images.  Now I'm writing my own
> script and would like to make use of those images as well, but I get the
> following error...  ImportError: No module named images.
> I'm still learning python and am not 100% sure how the import statement
> works.  The images file is in the demo folder of the wxPython folder.  I
> tried things like import wxPython.demo.images, but that doesn't work either.
> Can somebody give me a hand?

It needs to be accessible in a folder in your pythonpath.  But why
not just copy the file to your own folder and use it locally?  The
code in the demo is for demonstration purposes, not to use as a library...


From thomas.heller at ion-tof.com  Wed Feb 13 13:17:48 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 13 Feb 2002 19:17:48 +0100
Subject: Reading and writing C structs from Python
References: <3C696FC7.1090600@erols.com>
Message-ID: <a4eakn$1el6h9$1@ID-59885.news.dfncis.de>

"Edward C. Jones" <edcjones at erols.com> wrote in message news:3C696FC7.1090600 at erols.com...
> Suppose I have a C struct for example
>
> typedef struct {
> float x;
> int n[4];
> } test;
>
> and a pointer to an instance of the struct
>
> test* t = (test*) malloc(sizeof(test));
>
> I would like to wrap this in a python object "py_t" so that (in Python)
> "x = py_t.x" reads "t.x" and "py_t.x = 13.0" writes to "t.x". Ditto for
> "py_t.n[i]".

This is exactly the kind of module (system?) I'm currently working on.
Your example would look like this:

>>> from ctypes import Structure, Array
>>> IntArray4 = Array(format="i", length=4)
>>> class PY_T(Structure):
...     _names_ = "x", "n"            # declare the field names
...     _formats_ = "f", IntArray4    # declare the field types
...
>>> py_t = PY_T() # creates an instance, there are other ways to create one if you already have a pointer to a C structure
<PY_T object read-write ptr 0x007F8078, size 20 at 0x00767D68>
>>> print py_t.x
0.0
>>> print py_t.n
<i_ARRAY_4 object read-write base 0x00767D68, ptr 0x007F807C, size 16 at 0x007F3BC8>
>>> print py_t.n[0]
0
>>> print py_t.n[3]
0
>>> print py_t.n[4]
Traceback (most recent call last):
  File "<stdin>", line 27, in ?
IndexError: invalid index

Still this needs some time to be finished...

Thomas




From a.clarke11 at pop.ntlworld.com  Wed Feb 20 09:33:09 2002
From: a.clarke11 at pop.ntlworld.com (a.clarke11)
Date: Wed, 20 Feb 2002 14:33:09 +0000
Subject: Button changing its own grid options
Message-ID: <3C73B3A5.60DBCBA9@pop.ntlworld.com>

Hi
Does anyone know if it is possible to allow a Tkinter button, when
pressed, to change its own grid options?
I've tried various command callbacks, but I'm having problems
referencing the buttons own grid info.
The idea is to get a button to move around, in a grid of other buttons,
when pressed. Any ideas or even better, code, would be appreciated.
Thanks in advance
Tony



From mwh at python.net  Tue Feb 26 10:16:56 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 26 Feb 2002 15:16:56 GMT
Subject: Contributing patches (was Re: urllib and proxy)
References: <a5ecas$qiq$1@zeus.polsl.gliwice.pl> <20020226100136.GA18498@andy.tynant.ftech.net> <mailman.1014733123.3107.python-list@python.org>
Message-ID: <un0xwxob7.fsf@python.net>

Andy Gimblett <gimbo at ftech.net> writes:

> Earlier on today I posted to this list a modified urlopen() which
> supports proxy specification.  How do I go about suggesting this as a
> modification to the standard python library?
> 
> I've seen these:
>     http://www.python.org/download/cvs.html
>     http://www.python.org/patches/
> 
> and they seem to suggest that I can just use my existing SourceForge
> account and submit a patch to the patch manager, and I presume then
> someone examines it, decides if it's a good thing or not, and checks
> it in or rejects it?

Yes, more or less.  It may take a while to get noticed, I'm afraid.
There's also the quite likely step of being asked to modify your
patches in some way before they go in.  There can be quite some
to-and-fro-ing on the tracker.

> Is this how it works?

More or less.

> Just thought I'd ask here before blundering in.

Blunder away!

Cheers,
M.

-- 
  Famous remarks are very seldom quoted correctly.
                                                    -- Simeon Strunsky


From amrit040 at yahoo.com  Tue Feb 26 15:35:29 2002
From: amrit040 at yahoo.com (Amrit)
Date: 26 Feb 2002 12:35:29 -0800
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com> <usn7oxtpd.fsf@python.net>
Message-ID: <71c0a06e.0202261235.2165c300@posting.google.com>

Thanx that fixes the problem. So basically i was trying to read one
more element than what the input had. This explains the bug but what i
don't understand is why does this thing alternate ! Shouldn't it be
consistent.

Amrit
Michael Hudson <mwh at python.net> wrote in message news:<usn7oxtpd.fsf at python.net>...
> amrit040 at yahoo.com (Amrit) writes:
> 
> > Hi 
> > 
> > I'm trying to write a C function callable from python. I'm getting
> > some strange results :
> > 
> > Here's a typical run:
> > """
> > Python 2.0 (#2, Jun 18 2001, 15:29:44) 
> > [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> > Type "copyright", "credits" or "license" for more information.
> > >>> import lt
> > >>> lt.__file__
>  'lt.so'
> > >>> dir(lt)
>  ['__doc__', '__file__', '__name__', 'fmm']
> > >>> lt.fmm(range(15),-41)
>  (0.0, -1.0)
> > >>> lt.fmm(range(15),-41)
> > TypeError: function requires at least one argument
> > """
> > 
> > On calling the same function again with the same argument gives me a
> > type error. This alternates. I get a typeerror and then a correct
> > output and so on. Any ideas what could be wrong. My C code is pasted
> > below
> 
> This sort of thing usually means an exception happened and you didn't
> return NULL.
> 
> >   for (i = 0; i < n; i++)
> >     {
> >       obj[i] = PySequence_GetItem (vec, i+1);
> 
> This looks suspicious.
> 
> Cheers,
> M.


From wookiz at hotmail.com  Wed Feb  6 13:38:48 2002
From: wookiz at hotmail.com (wooks)
Date: 6 Feb 2002 10:38:48 -0800
Subject: What is the raison d'etre for Special/Magic Methods
Message-ID: <8852d33c.0202061038.3b53b455@posting.google.com>

I know there is a section on this at www.diveintopython.org but the
site has been down all afternoon.

I just started learning Python this week.


From jason at jorendorff.com  Mon Feb 25 20:02:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 19:02:28 -0600
Subject: Editors for Python
In-Reply-To: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEDFDFAA.jason@jorendorff.com>

Moray Taylor wrote:
> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.

IDLE has both of these.  It comes with Python.  Type "idle" at a
prompt; it's probably there.

  Do "File -> Open Module" and open "urllib".
  Then do "File -> Class Browser".

The class browser isn't very keyboard-friendly, though.
But you've got all the source code.  It's written in Python.

See also http://idlefork.sourceforge.net/ .

## Jason Orendorff    http://www.jorendorff.com/



From marklists at mceahern.com  Thu Feb  7 13:02:55 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 10:02:55 -0800
Subject: FixedPoint 
In-Reply-To: <200202071657.g17GvjGo018700@ratthing-b246.strakt.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBCEHBKBAA.marklists@mceahern.com>

> From: Laura Creighton [mailto:lac at strakt.com]:
> I would tell the PyPgSql folks to please change the test so that their
> code will work when passed an argument which redefines the __cmp__
> function.  Its just something they need to learn -- which Andrew Dalke
> taught me hmm, 2 weeks ago now.

Of course, consider the code below, which shows (in a perhaps overly
simplistic way) that I may not realize I'm comparing to None.  Should I
handle that?  I don't know what the Pythonic way says--maybe it says I
should just let the error propagate in this case and trap it.  I lean
towards thinking I oughta make my __eq__ and the like smarter.

// mark

#! /usr/bin/env python

class foo:

    def __init__(self, value):
        self.value = value

    def __eq__(self, other):
        return self.value == other.value

    def __ne__(self, other):
        return not (self == other)

f = foo(1)
g = foo(1)

print (f == g)
print (f != g)

f = None

print (f == g)
print (f != g)





From jjl at pobox.com  Sat Feb  2 17:14:19 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 2 Feb 2002 22:14:19 +0000
Subject: Problem building gdbm module (python-2.1.1)
In-Reply-To: <roy-AE0761.14585102022002@news1.panix.com>
References: <roy-AE0761.14585102022002@news1.panix.com>
Message-ID: <Pine.SOL.4.30.0202022212580.597-100000@mimosa.csv.warwick.ac.uk>

On Sat, 2 Feb 2002, Roy Smith wrote:
[...]
> If I change the configure script to look for gdbm.h instead of gdbm/ndbm.h,
> it finds the library, builds the module, and seems to work just fine.
>
> Is this a bug, or am I just doing something wierd?

It does exist:

http://www.opengroup.org/onlinepubs/007908799/xsh/ndbm.h.html

So I doubt it's a bug.  Perhaps your gdbm is too old.


John



From logiplexsoftware at earthlink.net  Mon Feb  4 17:45:12 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 4 Feb 2002 14:45:12 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <a3n03m$tfn7@news1.gtech.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3d46d$tfn2@news1.gtech.com>
	<Wls68.14178$om6.357057@news1.tin.it>
	<mailman.1012590723.7301.python-list@python.org>
	<a3kb3p$se44@news1.gtech.com>
	<3C5EE7C2.4DF4A885@ccvcorp.com>
	<a3n03m$tfn7@news1.gtech.com>
Message-ID: <20020204144512.268214f5.logiplexsoftware@earthlink.net>

On Tue, 5 Feb 2002 08:53:27 +1100
Peter Milliken wrote:

> > I believe that the "continuous ongoing code review" being mentioned is
in
> > reference to the Extreme Programming practice of pair programming --
all
> code is
> > written jointly by two people, sitting at the same computer.  Thus,
> everything
> > is, in effect, being reviewed by the other person as it's being
written.
> This
> > should (in principle ;) ) greatly reduce the need for formal group
> reviews...
> > though I expect they'd probably still be desirable regardless.
> >
> >
> Really? What an unusual idea - I didn't realise we had such a surplus of
> programmers to advocate something like this :-)

This is a little more "extreme" than I was referring to as well.  The idea
of two programmers sitting at the same computer doesn't really go over well
with me... I was referring having two programmers sit side-by-side at
separate computers, working on different parts of the same program versus
having one programmer work in isolation.  Indeed, I don't know many
programmers with the patience to sit and watch someone else type ;) 
However, if they are each writing code for the other to use, then you get
many of the same benefits as code review and faster development to boot.  I
have nothing against "official" code review per se, but I suspect it falls
in the realm of things that are great but rarely happen.  Additionally, I
think there are side-benefits in productivity and morale when working in
pairs that aren't to be had many other ways.

> 
> Actually, it is just another attempt at fixing the same old problem. I
have
> no problems with trying new techniques such as these but at the end of
the
> day, if good old fashioned reviews are done properly (and in a timely
> manner! :-)) then you would achieve the desired effect i.e. solid, robust
> code. If the individuals involved approach the problem with the correct
> frame of mind then I have never, ever seen inspections/reviews fail! It
is
> only when it is not tackled with the right sort of commitment that it can
> appear not to work.

Exactly.  I still venture that they won't ever be "tackled with the right
sort of commitment" because it's no fun having to read other people's code,
and lack of enthusiasm always seems to win over policy and "requirements"
in the end...


> I get somewhat frustrated sometimes when I look at yet more attempts to
> solve the same problem when I believe there are already workable
solutions
> (this will, I am sure be disputed :-) but is only a statement based upon
> personal experience :-)). I think it is most likely the lack of
> self-discipline (in programmers, not themselves) that caused the authors
to
> look for alternative methods - also, the fact that they needed something
new
> to write about otherwise they wouldn't be able to publish a book! :-) I
must
> admit that in a weak moment, I purchased a copy of Extreme Programming, I
> read it on the train one evening (not the whole book! :-)) and saw enough
to
> make me put it where it is now, on my shelf gathering dust :-).

I haven't read much beyond magazine articles regarding XP myself, but I
suspect they realize the futility of traditional approaches (not that the
approaches wouldn't work... people just won't use them which amounts to the
same thing) and tried to develop new approaches that mirror what successful
developers do already.  For example, the "make one to throw away" approach
is something that most of us do already (or should, I've had _very_ few
projects where the spec was so perfect and my understanding of the tools so
complete that I've gotten it perfect the first time), they just made it
into a "formal" design phase. 

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From martin at v.loewis.de  Mon Feb  4 01:22:14 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 04 Feb 2002 07:22:14 +0100
Subject: Python 2.2 Modules/Setup lies?
References: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net> <mailman.1012768207.25312.python-list@python.org> <slrna5rc6q.dtp.jon+usenet@snowy.squish.net>
Message-ID: <m3sn8hkbqh.fsf@mira.informatik.hu-berlin.de>

jon+usenet at unequivocal.co.uk (Jon Ribbens) writes:

> Sorry, I don't understand your answer. Do you mean that the README
> file is completely wrong? This would appear to be a bit of an
> oversight if so. Python 2.2 most certainly is using the Setup file to
> configure itself, because it changes things if I alter it. 

Where did you read that you must edit it? My copy of the README says

# Fortunately, the configuration and build process has been
# streamlined for most Unix installations, so all you have to do is
# type a few commands, optionally edit one file, and sit back.

This suggests that editing Modules/Setup is optional, and indeed, it
is.

# You may want to pass options to the configure script or edit the
# Modules/Setup file after running configure -- see the section below
# on configuration options and variables.

You *may* edit Modules/Setup; there is no need to do so.

> On the other hand, it does appear to be including modules that are
> excluded in Setup. Is there any *correct* documentation for all of
> this somewhere? How do I configure things like where socket.c looks
> for SSL libraries?

If you don't configure anything, setup.py will look for it in a few
standard places. If that fails, you should not edit setup.py, but edit
Modules/Setup instead - this is where you specify include directories
for modules if any are needed. If the SSL libraries are in a different
location on your system (and still provided by the system), please
report that as a bug so that setup.py will look in more locations.

Modules not listed in Modules/Setup are not excluded, they are just
not explicitly configured.

Regards,
Martin


From rjones at ekit-inc.com  Thu Feb 14 17:57:36 2002
From: rjones at ekit-inc.com (Richard Jones)
Date: Fri, 15 Feb 2002 09:57:36 +1100
Subject: Java vs Python ... ???
In-Reply-To: <Xns91B5F26DFF836cliechtigmxnet@62.2.16.82>
References: <Xns91B5CBEF0A103Scirious@140.99.99.131> <Xns91B5F26DFF836cliechtigmxnet@62.2.16.82>
Message-ID: <200202142257.WAA14088@crown.off.ekorp.com>

On Fri, 15 Feb 2002 09:46, Chris Liechti wrote:
> Scirious <Scirious at Scirious.com> wrote in
>
> news:Xns91B5CBEF0A103Scirious at 140.99.99.131:
> > I've been learning JAVA for a while, but now I've heard about
> > Python.
> > Any coments concerning these two languages?
>
> that's one of the questions that is good for a flamewar :-O

So don't fan the flames....

> anyway here some answers biased towards python:
>
> Python programms are shorter.
> The Debug Cycle is shorter.
> You can allways try out things interactively.
> Python is FUN programming.

Please, always just respond with the prior discussions:

    http://www.python.org/doc/Comparisons.html

We get enough noise on the list without re-discussing all this crap all over 
again.


> Here are some nice slides that illustrate some differences (PPT):
> http://64.78.49.204/pub/eckel/LovePython.zip
> (from http://www.mindview.net/Books/Python/ThinkingInPython.html)

Please submit these links to the python.org webmaster.



    Richard

ps. yes, the jython reference was perfectly valid, thanks for pointing that 
out, I'd forgotten :)



From xtian at toysinabag.com  Mon Feb 25 21:31:28 2002
From: xtian at toysinabag.com (xtian)
Date: 25 Feb 2002 18:31:28 -0800
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <d111eafa.0202251831.5b3dc4e@posting.google.com>

mtaylor at lowebroadway.com (Moray Taylor) wrote in message news:<d906b2c1.0202250818.57dcfcc7 at posting.google.com>...
> I am desperatly seeking a good python editor, so far JEdit is the best
> I have found, but since it is written in Java, I find it painfully
> slow, and PyBrowse, although good, a little flaky.
[snip]
> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.

I know this is a silly question, but no-one else on the thread has
asked it (as far as I can see):
If those are your requirements, what's wrong with IDLE?

(I'm guessing that those aren't your only requirements, really. They
certainly wouldn't be my only ones. What else do you need in your
editor? :)

xtian


From threeseas at earthlink.net  Sun Feb  3 19:00:49 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:00:49 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1726.795T908T14258threeseas@earthlink.net> <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <1173.799T632T11653818threeseas@earthlink.net>

On 30-Jan-02 23:23:31 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:

>This is a discuss group (gnu,misc.discuss anyway). It is not a place to seek
>coding help. If you don't want a debate don't post to a discussion forum. If
>comp.lang.python is like every other comp.lang group I know then it is
>also not a place to seek coding help - other than specific language problems
>(such as how do I convert a string into a number?).

>You just don't understand how programmers work. And until you do you
>won't find many willing to help you. In fact if someone did think your ideas
>were good they would probably just fork your code and do it themselves
>without your involvement - since you seem to be going out of your way to turn
> people off your project.


Don't like the idea of autocoding huh?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From quinn at regurgitate.ugcs.caltech.edu  Fri Feb 22 18:26:42 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Feb 2002 23:26:42 GMT
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com> <CMfd8.5037$D6.14223@news.iol.ie> <u7b81629bo9o4a@corp.supernews.com> <slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu> <mailman.1014399281.7163.python-list@python.org>
Message-ID: <slrna7dkti.5et.quinn@regurgitate.ugcs.caltech.edu>

On Fri, 22 Feb 2002 09:23:24 -0800, Cliff Wells
<logiplexsoftware at earthlink.net> wrote:
>For that matter, you could also store the Python objects directly in the
>SQL database.  My own opinion is that SQL is so easy that it's a no-brainer
>to use it whenever possible.  A problem I've seen many times is people
>utilize some shortcut because they feel the program will never need
>anything more, only later to discover that the program has utility far
>beyond what they expected, but the shortcut they've taken cripples it, and
>they end up rewriting (or worse, they don't).

I don't find SQL to be a no-brainer.  You have to download the thing (MySQL at
least is massive... the sunos binary tarball I have is >5MB).  Then you have to
install it (create tables, blah blah, learn MySQL's complicated permission
system, etc.).  Then you have to turn it on (which means either getting root to
edit the system rcs or setting up some hackery to restart it if the server
reboots).  Not to mention that you have to learn enough SQL to do all this.
Oh yes, and if there are no binaries for your platform, I hope you happen to
have a decent C++ compiler lying around.

And even if it is easy to introduce giant complicated support program, should
you?  If linking in a massive shared library (or *any* shared library) is as
easy as tacking on a '-lpiggy', is it a no-brainer to link it in whenever
possible?  I don't think so.  How easy it is for the developer to create
something is only part of the equation (that's what the python "creed" of
legibility and simplicity is about, yes?).


Of course, you're right that you can easily marshal python objects to SQL, I
assume by pickling to a string.  But then you lose the "it's not python
specific" advantage of SQL.  And you still have to scrunch your data into the
relational rows/columns model.  What if it's hierarchical and lumpy?


>What if she decides to maintain historical data?  Then to keep the files
>reasonably sized, she'll have to start juggling files.

I think the *dbm type stuff scales ok up to mediumish files at least (there are
a few > 50MB gdbms lying around, INN uses it for article index or something).
I have no idea if it's happy all the way up to a few gigs.  If I wanted to use
it for that much data I'd test it out.

>Software has a tendency to grow beyond the developer's expectations, so I
>consider it unwise to take shortcuts when a better long-term solution isn't
>much more difficult to implement.

That's one point of view, and a reasonable one.  Another point of view is that
you should only design for what you know you need, instead of increasing
generality and complexity to deal with some hypothetical future situation.
If you think your program has acquired a new requirement, you decide if
it's really a requirement or if it should be done by another program, and if
it is, you modify your program, which hopefully is easy to modify since you've
been trying to keep things simple.

In the shelve case, it would be pretty easy to write an SQL-using backend
that implements the shelve interface.  And maybe you'll like that interface
better than the SQL-ish cur.execute(this_and_that) anyway.

>Not only that, but storing the data in a SQL database separates the data
>storage from the program, so in the future she could use that data from
>some other program, not just the Python one.  Then if she decides she wants
>to access the data from, say, a web server, it's not a big problem. 

You can write CGI scripts in python :)


From joost_jacob at hotmail.com  Fri Feb  8 06:49:57 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 8 Feb 2002 03:49:57 -0800
Subject: 2001 Turing Award
References: <mailman.1013127134.29799.python-list@python.org>
Message-ID: <13285ea2.0202080349.6f8f9552@posting.google.com>

[James_Althoff]
> 
> http://arc.cs.odu.edu:8080/dp9/getrecord/oai_rfc1807/0018.mit.theses/1963-10
> 
> Happy reading,
> 

That link has a bib. record but no thesis, the thesis itself is not
available online?


From rich at worldsinfinite.com  Tue Feb 12 09:41:33 2002
From: rich at worldsinfinite.com (Rich Harkins)
Date: Tue, 12 Feb 2002 09:41:33 -0500
Subject: read and write binary data (newbie)
In-Reply-To: <3c69359f.22621453@news.easynews.net>
References: <3c69359f.22621453@news.easynews.net>
Message-ID: <200202121441.g1CEfXQ01782@localhost.localdomain>

On Tuesday 12 February 2002 10:45 am, Marcus Stojek wrote:
> Hi,
>
> I am using Python quite a while for all kind of things. Now i have to
> read a file (Windows) in binary format that contains strings (\0
> terminated), int and floats (4 byte). and I am stunned because
> I don't know how to.
>
> to read strings I have written a function:
[snip]
>
> reading ints and floats is easy:
> x= dat.read(4)
>
> but how can i convert x into a number?
> and how can I write a float or int into
> a binary file?
>
> Could anybody please tell me where to look?
>
> thanks
> marcus

Try the struct module.  


Rich



From scalet at yebu.de  Sat Feb 16 04:54:38 2002
From: scalet at yebu.de (Karl Scalet)
Date: 16 Feb 2002 01:54:38 -0800
Subject: Thread and CGI
References: <mailman.1013786724.3651.python-list@python.org>
Message-ID: <a925ae9a.0202160154.2d7881c5@posting.google.com>

[..]
> Remember:
> 	REOPEN STDIN, STDOUT, STDERR TO /dev/null !!!!
> 
> I know this is a python group, but I originally did this
> in perl, so I'm posting the bit I used:
> 
> Because, if not, the child won't "daemonize", and the
> parent process will hang until the child finishes, or
> the browser times-out (upon which the child will be
> summarily terminated).

if you don't need those three handles, just close them, but
be aware not doing it with sys.stdout.close() rather then
use os.close(1), etc.

Just mention this because that drove me crazy the last days.
Got this from some other thread in this group, thnx, but don't
have the reference.

cheers, Karl


From Denys.Duchier at ps.uni-sb.de  Wed Feb 20 07:28:37 2002
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: Wed, 20 Feb 2002 13:28:37 +0100
Subject: Mystery Guest from Number Theory
References: <a501qd$1c3k$1@news.idiom.com>
Message-ID: <isg03wl4lm.fsf@elk.ps.uni-sb.de>

http://mathworld.wolfram.com/CollatzProblem.html variously identifies
it as Collatz's problem, Hasse's algorithm, Kakutani's problem,
Syracuse algorithm, Syracuse problem, Thwaites conjecture, and Ulam's
problem.  The element of the resulting sequence are called Hailstone
numbers.

Cheers,

-- 
Dr. Denys Duchier			Denys.Duchier at ps.uni-sb.de
Forschungsbereich Programmiersysteme	(Programming Systems Lab)
Universitaet des Saarlandes, Geb. 45	http://www.ps.uni-sb.de/~duchier
Postfach 15 11 50			Phone: +49 681 302 5618
66041 Saarbruecken, Germany		Fax:   +49 681 302 5615


From cfelling at iae.nl  Thu Feb 28 09:44:40 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 15:44:40 +0100
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <slrna7rdna.5m1.QnickQm@localhost.localdomain>
Message-ID: <a5lfoo$38q$1@animus.fel.iae.nl>

Nick Mathewson <QnickQm at alum.mit.edu> wrote:
> In article <3c7c1a91.150875561 at menuhin.netfront.net>, A. Jones wrote:
...
>>             if type(blurb[z]) == "<type 'list'>": #Stops nesting
>> strings

> #It's bad style to rely in the string value of a type.  Intstead, you

Not only is it bad style, it does do what he thinks it does:)

   >>> type(1)
   <type 'int'>
   >>> type(1) == type("<type 'int'>")
   0
   >>> type("<type 'int'>")
   <type 'str'>
   >>> type(1)
   <type 'int'>
   >>> type(1) == "<type 'int'>"
   0

but

    >>> str(type(1)) == "<type 'int'>"
    1

Type() returns a type object, the <type 'int'> thingy is merely a
string representtation of that object.  Ofcourse they are not equal.
To make this test work, one has to force the type object to return
its string representation, either with str(), repr() or backticks.

And one of the reasons it's bad style, is that in a next version of
Python the string representation might change. (not likely though)
-- 
groetjes, carel


From mwh at python.net  Thu Feb 14 11:07:01 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 14 Feb 2002 16:07:01 GMT
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de> <u8za6pskh.fsf@python.net> <3c61abe7$1@si-nic.hrz.uni-siegen.de> <uu1sta7s2.fsf@python.net> <3c62820d$1@
Message-ID: <upu38f3jt.fsf@python.net>

Simon Budig <Simon.Budig at unix-ag.org> writes:

> When you put it into 2.3(\.[0-9)? I am perfectly happy. I have
> patched my local 2.1.1 installation and am happy with it...  :-)

This is done now.

Cheers,
M.

-- 
  SCSI is not magic. There are fundamental technical reasons why it
  is necessary to sacrifice a young goat to your SCSI chain now and
  then.                                                  -- John Woods


From SWidney at ci.las-vegas.nv.us  Mon Feb 25 12:33:15 2002
From: SWidney at ci.las-vegas.nv.us (Scott Widney)
Date: Mon, 25 Feb 2002 09:33:15 -0800
Subject: [Tutor] unrelated bug issue
Message-ID: <D4EB5574F4A7D2119C9E00A0C9EA408D059F59AE@SOVEREIGN>

> From: Tim Peters
> 
> I got a better one:  suppose you shoved a banana down your 
> throat, then a carrot in each nostril.  Would it then be 
> the banana's or the carrots' fault that you couldn't breathe
> anymore? That's right:  it would be the carrots', because
> you didn't have any trouble breathing before they wound 
> up in your nose.  Or maybe it was the banana's fault after 
> all, since if it weren't there, the carrots' presence 
> wouldn't matter.  It's a puzzler.  Just so long as it's
> clear that it's never ever your fault <wink>.

Tim, you're missing the point. What he said is that even after he removed
the carrots from his nostrils, he still couldn't breathe. So either the
carrots were not completely removed, or it's the banana's fault. He did the
right thing in looking for carrot residue before removing the banana. He
should be commended for his insight....



From martin.franklin at westgeo.com  Mon Feb 18 05:17:44 2002
From: martin.franklin at westgeo.com (Martin Franklin)
Date: Mon, 18 Feb 2002 10:17:44 +0000
Subject: List-Box
References: <a4qfje$26flu$1@ID-96133.news.dfncis.de>
Message-ID: <a4qkae$2ag$1@mail1.wg.waii.com>

Christian H?hne wrote:

> geolistbox['Delete'] = Tkinter.Listbox(subframegeo['Delete'])



try:

geolistbox['Delete'] = Tkinter.Listbox(subframegeo['Delete'],
        exportselection='no')


From kragen at pobox.com  Sun Feb  3 01:10:37 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 03 Feb 2002 01:10:37 -0500
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012597563.30612.python-list@python.org> <v5F68.110829$8b.4204925@atlpnn01.usenetserver.com>
Message-ID: <833d0jyu1u.fsf@panacea.canonical.org>

"Steve Holden" <sholden at holdenweb.com> writes:
> Without any introduction of static typing into the language, it's extremely
> unlikely that a JIT would yield the same benefits as it does for Java,
> although recent experiments with compiler techniques have shown interesting
> speedups on restricted cases.

This is completely unfounded and false.  The JIT techniques used for
Java were developed for Self, a language just as dynamic as Python,
and compiled Self code to within a factor of 2 of C --- the same kind
of speed Java JITs get today.  The Self experiments were carried out
during the late 1980s and early 1990s, and they were not "restricted
cases".

> The major problem is that with dynamic (latent) typing, the JIT code has to
> examine the type of its operands in order to determine which of several
> type-based operations to perform. Java can avoid this due to its strong
> typing mechanism. [inflammatory remark snipped]

Java can avoid this in some cases, because the types of some variables
(those of primitive, non-Object types) are known at compile-time, but
just like Python, the types of most objects are not known at
compile-time; most classes and most methods are not final.  It's
optimizing these dynamic calls that is hard.

It's true that Java objects all have declared types, but that doesn't
save you as much as you'd think; it just allows more efficient dynamic
method lookup in cases where the JIT can't do anything smarter --- but
it's still too slow.



From kfarmer at thuban.org  Mon Feb 18 15:30:44 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 18 Feb 2002 20:30:44 GMT
Subject: Python on Windows CE
References: <a4ev4g012jn@enews4.newsguy.com> <3C6D587B.5040203@wi.mit.edu>
Message-ID: <Uvdc8.31$RP7.31260710@newssvr14.news.prodigy.com>

A revival of Python CE has been going on for a while, but nobody's gotten
around to compiling 2.2 for the Clio.  Check Hammond's pages for links to
the current Python CE effort.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Brian Kelley" <bkelley at wi.mit.edu> wrote in message
news:3C6D587B.5040203 at wi.mit.edu...
> I was using Python CE for my Vadem Clio a couple of years ago.  You
...
> Unfortunately, 1.5.2 seems to be the end of the CE 2.0 distibution that
> Mark Hammond worked on a long time ago.





From grante at visi.com  Wed Feb 20 13:09:52 2002
From: grante at visi.com (Grant Edwards)
Date: Wed, 20 Feb 2002 18:09:52 GMT
Subject: Guido wins FSF Award
References: <mailman.1014220773.23411.python-list@python.org>
Message-ID: <QDRc8.44470$Wf1.8418718@ruti.visi.com>

In article <mailman.1014220773.23411.python-list at python.org>, Simon Brunning wrote:
>> From:	grante at visi.com [SMTP:grante at visi.com]
>> > what are revealing the Nine Pythonic Actions underlying all that we do.
>> > can you explain what that means ??????
>> 
>> Bhuddism allusion?
>  
> Autocoding allusion, I think.

Ah.  I gave up on that thread long ago.

-- 
Grant Edwards                   grante             Yow!  A can of ASPARAGUS,
                                  at               73 pigeons, some LIVE ammo,
                               visi.com            and a FROZEN DAQUIRI!!


From db3l at fitlinxx.com  Fri Feb  1 18:54:09 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Feb 2002 18:54:09 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it>
Message-ID: <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com>

Alex Martelli <aleax at aleax.it> writes:

> If you have to *think* about it, as "memorization" would suggest, then
> you'll probably be happier going with the rodent.  The power of vi, for
> me (and most other vi-enthusiasts I've known) is that the *fingers*
> learn the patterns -- sort of like in needlepoint; your conscious brain
> can focus on higher-abstraction stuff.  (...)

As an Emacs-die-hard, I'll also agree strongly with this.  The way I
like to put it is that I touch type Emacs editing commands nearly as
quickly as I touch type normal text.  Most of the time, I'm rarely
even conscious of the mechanics of achieving an edit - rather I know I
want to "move that line over there" or something, and my fingers just
perform the right operations.  And whether it's vi or Emacs, I find
not having to leave the home keys and reach for a mouse (with all the
associated eye hand coordination involved) is an enormous time saver
for most editing tasks.

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


From meowbot at meowing.net  Sat Feb  9 02:47:22 2002
From: meowbot at meowing.net (A Meowbot)
Date: Sat, 09 Feb 2002 07:47:22 GMT
Subject: IOError: [Errno 4] Interrupted system call
References: <1013225254.610747@cswreg.cos.agilent.com>
Message-ID: <MPG.16ce9dd3d2399887989680@news.verizon.net>

Greg Weeks wrote:
> In Python 1.5.2, I'm getting the message
> 
>     IOError: [Errno 4] Interrupted system call
> 
> in a context that doesn't make sense to me.  My program has two threads
> (which won't appear useful out of context):
> 
> Thread 1: Fork a subprocess, wait for the subprocess to exit, and set
> an Event.
> 
> Thread 2: Send a SIGTERM signal to the subprocess and wait for the Event to
> be set.

[...]

You may be getting tripped up by the SIGCHLD generated by the terminated 
subprocess.  How are you trying to collect that zombie in thread 1?


From chris.gonnerman at newcenturycomputers.net  Tue Feb  5 01:51:58 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 5 Feb 2002 00:51:58 -0600
Subject: OT again... is there an IMAP expert in the house?
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org> <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org> <mailman.1012224450.11377.python-list@python.org> <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu>
Message-ID: <00e201c1ae11$b16be5c0$0101010a@local>

----- Original Message -----
From: "Eduardo Chappa" <chappa at math.washington.edu>


> *** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:
>
> :) Urrr... it doesn't solve my problem. Maybe IMAP isn't the solution
> :) either.
> :)
> :) The customer(s) I am trying valiantly to support want to be able to
> :) share a mailbox *concurrently*. The Mercury and UW servers don't seem
> :) to work that way. I haven't studied the protocol so I don't know if it
> :) is even possible.
>
> This is not a problem about the protocol you are using to read your
> folder. IMAP is the way that you communicate to the server. The way that
> the mail is stored in the server is the relevant point here. For example,
> the default for the UW server is unix, and unix (format) does not allow
> concurrent access to a mailbox. If you use the UW server, and you use mbx
> format, then you can have concurrent access to your folders. Convert your
> folders to mbx format and you'll be done.

Tried that, didn't work.  I was using Netscape 4.75 as one client, Outlook
Express as the other, and concurrent deletion (even with expunges) just
didn't work.  Outlook Express never seemed to notice Netscape doing
deletions,
and I couldn't get Outlook Express to properly expunge without closing it.

Like I said later, it seems to be a client problem.





From yoonseo at altavista.net  Tue Feb 26 14:46:16 2002
From: yoonseo at altavista.net (SungPil Yoon)
Date: 26 Feb 2002 11:46:16 -0800
Subject: [Q] building a fortran extension module
Message-ID: <d737306c.0202261146.5dcf6e02@posting.google.com>

Hi, all

I am trying to build a fortran extension module using f2py.
I have done it successfully on Sun, and the message from 'file' command is

    % file rinexSupmodule.so
    ELF 32-bit MSB dynamic lib SPARC Version 1, dynamically linked, \
    not stripped

Now, I am having trouble to do the same thing on HP. After I build an
extension module and use 'file' command then I got

    % file addmodule.sl
    PA-RISC1.1 shared executable dynamically linked -not stripped

The message says that the file is not a library and if I try to import
in python, I get this error message:

    % python -v
    >>> import add
    shl_load addmodule.sl
    /usr/lib/dld.sl: Bad magic number for shared library: addmodule.sl
    /usr/lib/dld.sl: Exec format error
    addmodule.sl: Exec format error
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
    ImportError: Failed to load addmodule.sl
    >>>

Could anybody figure out what causes this problem? I tried python 2.0.1 
and python 2.2, and had the same problem.

Any help will be appreciated.

sungpil


From sjmachin at lexicon.net  Tue Feb 26 15:44:39 2002
From: sjmachin at lexicon.net (John Machin)
Date: 26 Feb 2002 12:44:39 -0800
Subject: ask help about raise, thank u
References: <mailman.1014686626.31452.python-list@python.org> <c76ff6fc.0202260144.70e52642@posting.google.com> <a5g41c$70hbr$1@ID-11957.news.dfncis.de>
Message-ID: <c76ff6fc.0202261244.2abd9a02@posting.google.com>

"Emile van Sebille" <emile at fenx.com> wrote in message news:<a5g41c$70hbr$1 at ID-11957.news.dfncis.de>...
> "John Machin" 
> >       # re-raise the exception, with traceback giving the *original*
> >       # source of the problem (somewhere inside the 'try' suite),
> >       # not the 'raise' statement
> >       raise exc, data, tb
> > 
> 
> Isn't this what raise by itself does?
> 
> >>> def test(d=None):
> ...     try:
> ...             1/d
> ...     except TypeError:
> ...             return 0
> ...     except:
> ...             raise
> ...
> >>> test()
>  0
> >>> test(0)
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 3, in test
> ZeroDivisionError: integer division or modulo by zero
> >>>

Quite possibly. I profess ignorance -- see the disclaimer in my post.
I posted merely in case the cognoscenti ignored an unfortunate whose
post turned up (in Google Groups at least) encoded in base64.

Could you please enlighten me on the following:

If a bare 0-arg raise is (mostly) equivalent to:
   x, y, z = sys.exc_info()
   raise x, y, z
(apart from the circular reference problem) then what is the 3-arg
raise useful for?

The reference manual says:
"""If a third object is present, and it is not None, it should be a
traceback object (see section 3.2), and it is substituted instead of
the current location as the place where the exception occurred. This
is useful to re-raise an exception transparently in an except
clause."""

There is a very interesting 39-message thread back in April 1998 that
actually documents the conception & birth of 0-arg raise --
Google("raise traceback"). Seems like 3-arg raise is now not very
useful, and the manual could perhaps point out that 0-arg raise now
handles the "re-raise transparently" functionality.

I note that of the 4 examples of 3-arg raise in urllib.py, one seems
to have the circular reference problem.

While seeking enlightenment: What mechanism did you use to search all
of Lib/*.py for 3-arg raises?

Regards,
John


From othello  Wed Feb 27 23:05:57 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 23:05:57 -0500
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
Message-ID: <a5kabc$cmf$1@bob.news.rcn.net>

How about:

dict( map( lambda k,v: (k,f(v)), d.items() ) )

or

map( d.__setitem__, d.keys, map(f, d.values) )


Raymond

"Huaiyu Zhu" <huaiyu at gauss.almadan.ibm.com> wrote in message
news:slrna7r0hi.bkl.huaiyu at gauss.almadan.ibm.com...
> Is there a way to do this
>
> for k, v in dict.items():
>     dict[k] = f(v)
>
> without involving a Python for loop?  I'd imagine that a function
> implemented in C could speed things up quite a bit.  Something like
> either of the following
>
> mapvalue(f, dict)
> dict.mapvalue(f)
>
> Huaiyu




From aahz at panix.com  Wed Feb 27 00:26:54 2002
From: aahz at panix.com (Aahz Maruch)
Date: 26 Feb 2002 21:26:54 -0800
Subject: thread question
References: <mailman.1014542390.12789.python-list@python.org>
Message-ID: <a5hqmu$b3i$1@panix2.panix.com>

In article <mailman.1014542390.12789.python-list at python.org>,
maximilianscherr <MaximilianScherr at T-Online.de> wrote:
>--- In python-list at y..., aahz at p... (Aahz Maruch) wrote:
>> In article <mailman.1013817865.2645.python-list at p...>,
>> maximilianscherr <MaximilianScherr at T...> wrote:
>>>
>>>if a thread (object) is for example a class, would it be possible to 
>>>call the class objects of this thread from another thread?
>>>or how could something similar be done?
>> 
>> Absolutely.  See http://starship.python.net/crew/aahz/
>> for some examples.
>
>can i acces attributes, or functions of a thread(like if i started 
>the class with threading.Thread(None, someclass)) like i can access 
>classes? and if yes how? if no, how can i do it in another way?

Once again, go to that URL.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From jmarshal at mathworks.com  Thu Feb  7 11:27:00 2002
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 7 Feb 2002 16:27:00 GMT
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de>
Message-ID: <a3u9sk$890$1@news.mathworks.com>

Another possibility:

from mymod import *

def dosomething(x):
    import mymod
    mymod.dosomething(1000)


MDK <mdk at mdk.com> wrote:
> I have this:

> # myprog.py

> from mymod import *

> def dosomething(x):
>     dosomething(1000)

> # End code

> I want the dosomething(1000) run from the function in mymod.  However,
> instead it is running dosomething(x) in myprog.py.

> Note: Changing the names of dosomething() in either module is not an option.
> (The names depicted here are not the real names in the program but have been
> changed for clarity.)

> I know that one way to fix this is to change

> from mymod import *

> to

> import mymod

> But I would like to know if, without changing the import line, there is a
> way to tell Python to use the dosomething() from mymod.

> I tried mymod.dosomething(1000) but it returns an error:
> AttributeError: 'builtin_function_or_method' object has no attribute
> 'dosomething'

> Thanks!




From ajeru at vknn.org  Tue Feb 19 19:42:43 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 16:42:43 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com> <3C72AB39.28618CC3@ccvcorp.com>
Message-ID: <24c39b2c.0202191642.1bfbfaad@posting.google.com>

Jeff Shannon <jeff at ccvcorp.com> wrote in message news:<3C72AB39.28618CC3 at ccvcorp.com>...
> Alexander Jerusalem wrote:
> 
> > And the final argument I have for static type checking is that it
> > enables method dispatching based on parameter types. In a statically
> > typed language you can create two methods that have the same name but
> > differ on the parameter types. The correct method will be called for
> > you depending on the type of the argument you pass in your call. That
> > makes for a quite flexible way of extending a program. You can just
> > add another method with the same name and another type to handle a
> > special case without touching the existing methods.
> 
> In Python, you'd simply try to use the object in one way.  If that doesn't
> work, catch the exception and try the next.  Then you can clearly see what
> happens to different types of objects when you're applying the "same"
> operation to them.  You don't need to use multiple methods to have
> different behaviors, and at the same time, you don't have to add more
> methods when you add a new type that you want to behave the same as an old
> type...

Yes, IF I want it to behave the same as the old. And that's really a
good feature of dynamically typed languages. But if you want the new
method do something else for a different type then you're stuck. You
have to change the one existing method an add a new case.

> 
> 
> > I'm not, however, completely satisfied with static type checking as
> > well because I don't see why the compiler forces me to have a
> > parameter comply with a static interface that has 10 methods when the
> > method that uses the parameter acutally only uses one of those
> > methods. I want the compiler to check if the parameter type I'm
> > passing has what the method needs but nothing more.
> 
> This is exactly what Python does, except that it does it at runtime
> instead of compile time; and really, it is *not* any slower to test this
> than it is to try a couple of compile cycles.

>From my experience I can say it is much slower but maybe one has to
get used to this style of programming.

>  Plus, as has been
> previously noted, PyChecker will scan for most such errors if it's used...

Yes that was a very valuable hint. I have already downloaded PyChecker
and it's very useful.

Alexander


From lac at strakt.com  Mon Feb 18 06:22:40 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 18 Feb 2002 12:22:40 +0100
Subject: List-Box 
In-Reply-To: Message from "Christian H hne" <Christian.Huehne@dlr.de> 
   of "Mon, 18 Feb 2002 09:57:16 +0100." <a4qfje$26flu$1@ID-96133.news.dfncis.de> 
References: <a4qfje$26flu$1@ID-96133.news.dfncis.de> 
Message-ID: <200202181122.g1IBMfGo011788@ratthing-b246.strakt.com>

You need to make your listbox with exportselection=0
http://www.pythonware.com/library/tkinter/introduction/x5453-patterns.htm

Laura Creighton



From jkraska at san.rr.com  Sun Feb 24 20:12:50 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 25 Feb 2002 01:12:50 GMT
Subject: Silly link question: python22_d.dll
References: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com> <m31yfaqzym.fsf@mira.informatik.hu-berlin.de>
Message-ID: <1d4j7u01o8v25qqkgnkjov4uupu8blfjr4@4ax.com>

>In PC/pyconfig.h, modify
>
>#ifdef _DEBUG
>#pragma comment(lib,"python23_d.lib")
>#else
>#pragma comment(lib,"python23.lib")
>#endif

Thank you for that. That was helpful.

C//



From hmartires at cultalg.pt  Mon Feb 25 10:32:33 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 15:32:33 -0000
Subject: Numeric arrays - initialising with binary data
References: <cf0ad9fb.0202250446.3df07a08@posting.google.com> <Pwqe8.596$xL1.7420993@newsserver.ip.pt>
Message-ID: <DMse8.612$Mk1.7309398@newsserver.ip.pt>

Sory...
I found it ;)
These are object methods.
Problem solved.


"Hugo Martires" <hmartires at cultalg.pt> wrote in message
news:Pwqe8.596$xL1.7420993 at newsserver.ip.pt...
> Where are this method's ??
> I import the array module, but i don't have them.
> I'm using python 1.5
>
> Tanks
>
> "Jesper Olsen" <jolsen at mailme.dk> wrote in message
> news:cf0ad9fb.0202250446.3df07a08 at posting.google.com...
> > Hi,
> >
> > The build in array type has the methods
> >
> > array.tostring()
> > array.fromstring()
> >
> > which make it easy to initialise arrays with binary data
> > read from a file.
> >
> > The Numeric array, however, apperantly only has the tostring method.
> >
> > How can I initialise a Numeric array (eg Float32), with binary data?
> >
> > I can off course do this by using the Numeric C-API, but I am sure
> > there must be an easier way.
> >
> > I am surprised the Numeric array type does not have the same methods
> > as the build in type, if the intention is in time to replace this
> > module...
> >
> > Cheers
> > Jesper
>
>




From aleax at aleax.it  Fri Feb  1 17:26:06 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 22:26:06 GMT
Subject: Code Reviews (was: Defending the Python lanuage...)
References: <mailman.1012596484.27614.python-list@python.org>
Message-ID: <2CE68.18845$3J1.585650@news2.tin.it>

Michael Chermside wrote:
        ...
> PS: For those XP'ers out there, I include pair programming as a limit
> case of code reviews.

Great take -- helps selling Agile Programming to point-haired bosses,
they HAVE heard that code reviews are good (don't call it XP or they'll
think you're talking about MS's latest product line, though).

[Great post, anyway...!]


Alex



From newgene at bigfoot.com  Sun Feb 10 12:03:11 2002
From: newgene at bigfoot.com (newgene)
Date: 10 Feb 2002 09:03:11 -0800
Subject: A newbie question about class
References: <mailman.1013306655.4125.python-list@python.org>
Message-ID: <a50631f6.0202100903.1455abc0@posting.google.com>

"Mark McEahern" <marklists at mceahern.com> wrote in message news:<mailman.1013306655.4125.python-list at python.org>...
> [newgene]
> > Hi, group,
> >      I have a newbie question about building a class:
> >
> > eg.
> >
> > class A:
> >    def __init__(self,x=None):
> >        self.x=x
> >        if self.x != None:
> >             self.y=x**2
> >        else:
> >             self.y=x**2
> 
> 1.  When comparing to None, it's better to use one of the following:
> 
> 	if x is not None:...
> 	if x is None:...

Could you explain a little more why "is not" is better than "!="? 

> 
> 2.  Consider using property, which is new with Python 2.2 and requires a
> new-style class (thus we subclass from object):
> 
> class A(object):
> 
> 	def __init__(self, x=None):
> 		self._x = x
> 
> 	def getX(self):
> 		return self._x
> 
> 	def setX(self, x):
> 		self._x = x
> 
> 	x = property(getX, setX)
> 
> a = A(1)
> print a.x
> 
> Cheers,
> 
> // mark

Thank you and other experts above. I got the solution for this problem now.


Newgene


From sdm7g at Virginia.EDU  Wed Feb 20 17:04:40 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Wed, 20 Feb 2002 17:04:40 -0500 (EST)
Subject: Why don't you use 'os.listdir(os.curdir)' instead?  [was: How do
 you conver output of "popen('ls')" to a list?]
In-Reply-To: <3C737245.BC5D7164@macrozon.net>
Message-ID: <Pine.OSX.4.43.0202201703380.381-100000@d-128-61-180.bootp.virginia.edu>


On Wed, 20 Feb 2002, husam wrote:

> thanks for reply.
>

You're very welcome!

-- Steve





From threeseas at earthlink.net  Mon Feb 11 00:40:51 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 11 Feb 2002 05:40:51 GMT
Subject: Fear of autocoding! >Re: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com>
Message-ID: <7643.807T2626T856092threeseas@earthlink.net>

Very good example of someone who is afraid of autocoding.

Come to think of it, it is a wonderment that autocoding is found in high
tech industries like aerospace and military but not so well known to be
used in lesser non-critical fields, such as general programming.

You'd think that such a complicated compound complexity task as autocoding
would be something needing to develeop and evolve from such lower and
non-critical field first to sort out all the bugs.

In fact the field of autocoding is rather young and immature and there
seems to be little available as to how it is done.

All of which seem to me to be rather interesting as the claim of
autocoding is one of higher quality, more stable and bug free code.

Of course you can even drop all of that and simple read all the words I've
written to this newsgroup and compare that to all the words others have
written in response and it's very clear what you will see is consistant
with both a fear factor on the part of responders and the lack of
autocoding development on lesser non-critical software development
directions.

Simply put, the frabicated excuse you have written up only show off your
ability to manipulate abstractions. But I do wonder, do you also
manipulate abstractions so well in your coding practices? Or do you simple
reinvent (meaning you are replaceable by autocoding development)?

I am not a troll, I'm very serious about what I present and have answered
most all the reasonable questions that have been presented me. Then I find
such links refering to software design as http://mukesh.ws/uml.html and
http://www.laputan.org/mud/mud.html and realize some of the things others
have presented me, that I thought might be reasonable, amounts to total
opiniated BS. They are intitled to their opinion I suppose, but don't
even they know it's opinion?

Lets' all face the fact that the truth of the matter is that you and
others are afraid of autocoding, that it will in fact remove the easy job
you have and perhaps expose that it is potentially better at coding then
you are. Just as fearful as the animation industry was when Newtek
introduce a low cost high quality video production switcher and animation
tool set. But what really happened is that those who were good at
animation only got better and gave us better, as more talent was in fact
introduced into the market and put to work.

You are fearful of change too, you like the way things are. And this shows
in what you write below, relying on "old" practices. But things change,
what was once a productive attitude has become an abusive excuse to
fabricate false witness against something you do not like.

It is very clear that there has been far more insinuation and attacks upon
me and my intention here than any anger I've expressed against it.

Your comments below only add to that unfairness. But it certainly shows
your skil and talent at abstraction manipulation.

Maybe there have been some not so negative responses I appear to have
skipped over, but the fact is that I've seen far to much negativity and
anti-user power from to many different sources to find anything less than
actual coding help to be questionable in intent.

I've had offers, but of varing conditions, none of which support in action
the open and free, as in GPL, advancement of the project. Though I've had
some help in optimizing the IQ code in Arexx.

I honestly and seriously do not believe that any programmer worth his salt
is so stupid that they do not understand the description of the nine
commands. For they are basic actions of which examples are found all over
the place in the field of programming. From understand the basic function
of having control of input and output direction, to the value and use of
variables and list of variables, to keeping track of where you are in
doing something, to looking up or accessing the meaning of something,
whether in a programmers manual page of a function to a word in a
dictionary and how applying constrains helps you to quicker locate what it
is you are looking for. And certainly there is enough example of
identifying system componets that plug and play is possible.

When so called programmers say they do not understand, and insist on such
understanding failure and feedback loop refusal, then I assume they are
insulting themselves. So why would I need to?

The fact is, you are afraid of losing control over your position. Where
what you should be doing is looking forward to how much more you could do
with having such a tool that is shared amoung all users. Unless of course
you are just a psuedo programmer having no ability to be creatively original.

Go ahead and apply some blind justice to the words I have written to this
newsgroup in comparison to the words of those who have responded. And
remember, I am only one person as opposed to the many who have responded.

On 10-Feb-02 20:18:13 Andrae Muys <amuys at shortech.com.au> wrote:
>rony.steelandt at bucodi.com (Rony) wrote in message
>news:<b34d3835.0202080025.1d69d198 at posting.google.com>...
>> YOU SHOULD BE ASHAME !!!!!!!!!!!!!!!!!!!!!
>Ashamed?  No, indeed he should be applauded for his stand for signal
>over noise!
>
>> Who on earth you think you are, to just post a message on usenet
>> saying don't talk to that guy
>>
>> 1. Nobody or nothing gives you that right.
>> 2. It's a attack to peoples intelegence here, i assume they very well
>> can decide by themself if they want to dicusse or not
>> 3. It's against the spirit of usenet
>> 4. You are not God to decide this
>> 5 ....

>(referencing <4egfvd$2pl2 at news-s01.ny.us.ibm.net>#1/1 from
>compl.lang.rexx in 1996)

>A quick grep through the deja archives indicates you have been on
>usenet long enough to know that what you're spouting here is absolute
>nonsense.  Everyone, by virtue of their involvement in a newsgroup
>community, has the right to defend that group's signal-noise
>ratio---you have been on usenet too long to claim ignorance of this.
>It is not an attack on anyone's intellegence, rather an appeal to the
>same, a request that they step back and reconsider the value of the
>(now rediculously extended) cascade they are contributing to.  That
>they pause, before automatically replying to the troll in our midst,
>and consider the effect of this threads continued existence on c.l.py
>itself.

>It appalls me that someone who has been involved in usenet for so
>long, would so willfully and deliberately, disregard the nature and
>culture of usenet communities in this way.  To appeal to the 'spirit
>of usenet' in defense of such a position maybe understandable, if
>mistaken, in a newbie making their first post, but for someone who
>appears to have been active here for over 5 years?

>The spontaneous, self-organising, emergent communities of people
>formed around this asyncronous text medium is the spirit of usenet you
>so glibly refer to.  The reaction of this community (by flame, by
>appeal, and eventually by ostracising) to defend itself against the
>parasitic troll, spammer, or pathological cross-poster, is a
>legitimate manifestation of this spirit.  An appeal by active
>participants of a newsgroup for silence, for signal over noise, is
>likely the purest expresion of the 'spirit of usenet' I can imagine.

>Andrae Muys


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From aahz at panix.com  Mon Feb  4 11:33:11 2002
From: aahz at panix.com (Aahz Maruch)
Date: 4 Feb 2002 08:33:11 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com> <7xzo2pqopt.fsf@ruckus.brouhaha.com> <WZx78.40291$KM2.1247277@atlpnn01.usenetserver.com>
Message-ID: <a3md47$h98$1@panix2.panix.com>

In article <WZx78.40291$KM2.1247277 at atlpnn01.usenetserver.com>,
Steve Holden <sholden at holdenweb.com> wrote:
>"Paul Rubin" <phr-n2002a at nightsong.com> wrote ...
>> 
>> The rand module also disappeared, completely gratuitously as far as I
>> can tell.  That's what broke the Redhat scripts.
>
>I had overlooked that.

The rand module was already deprecated in 1.5.2; it's in
Lib/lib-old/rand.py; it's not even in the docs.

ObShesh: Sheesh
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From thomas at gatsoft.no  Thu Feb 28 10:51:13 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Thu, 28 Feb 2002 15:51:13 GMT
Subject: Simple threading thing
Message-ID: <Rlsf8.136$5vh.170729984@news.telia.no>

I got a webserver based on BaseHTTPServer using threads. When it starts I
want to create a seperate process, a thread, that goes into the background
and checks the server on given intervals, clearing expired sessions etc. I
cannot get the created thread to pass control back to the server and let it
continue. It just sleeps the specified interval and then does it thing then
sleeps again. The server never gets into action at all.

Any clues?

Best regards,
Thomas Weholt




From winelight at spamcop.net  Mon Feb 11 12:58:18 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Mon, 11 Feb 2002 17:58:18 -0000
Subject: Newbie question: Why does read() method of urllib hang?
Message-ID: <%CT98.13104$HC4.90692@NewsReader>

I can't help feeling I'm doing something stupid. Most of the time:

    mystring=u.read()

works fine, but sometimes it just hangs. Is this simply because the remote
server is not responding? I would have thought that would cause
urllib.urlopen() to hang, not u.read().

Or have I made some fundamental mistake in error handling or failing to
clean up, or do I need threads, or...!

Any help greatly appreciated.

Andrew Ward





From trentm at ActiveState.com  Wed Feb 20 17:37:09 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Wed, 20 Feb 2002 14:37:09 -0800
Subject: HTMLifyer for Python code?
In-Reply-To: <Xns91BBEF384B311cliechtigmxnet@62.2.16.82>; from cliechti@gmx.net on Wed, Feb 20, 2002 at 11:27:41PM +0100
References: <mailman.1014238715.7462.python-list@python.org> <Xns91BBEF384B311cliechtigmxnet@62.2.16.82>
Message-ID: <20020220143709.A4573@ActiveState.com>

> > tool for converting Python source code to syntax-highlighted HTML. 
> 
> the SciTE editor which you can finde here:
> http://www.scintilla.org/
> 
> has a save as HTML option in the filemenu. (that editor component is 
> used in PythonWin and wxWindows)

Brian's SilverCity (which he just mentioned in this thread) uses Scintilla's
lexers so that above step has already been automated for you there.

Trent


-- 
Trent Mick
TrentM at ActiveState.com



From landshark at verticaladdiction.net  Thu Feb 14 01:24:36 2002
From: landshark at verticaladdiction.net (Dave Marotti)
Date: Thu, 14 Feb 2002 06:24:36 -0000
Subject: undefined symbols when embedding python
Message-ID: <slrna6mm3i.2aj.landshark@burrito.verticaladdiction.net>

Hello,

I'm using SIP to wrap a C++ class (very basic) and then embed the python
interpretor in another C++ program.  From the new C++ program I import the
wrapped c++ class and make some python calls.

Using SIP to wrap the C++ class worked fine.

Using the new pythong class in the python interpretor via the python executable
works fine.

However, when I try to run my C++ program which contains the embedded python
interpretor, any call which imports my wrapped C++ class causes undefined
symbols:

ImportError: ./libtestwrapc.so: undefined symbol: Py_InitModule4

I've linked my embedded C++ program to libpython2.0.a.

I've toyed with different linkings, but now I'm completely stumped.

Any help would be appreciated.  Thanks.

-dave


-- 
+-----------------------------+----------------------------------------------+
| Dave Marotti                | Looking for a Visio alternative for *nix?    |
| lndshark ! speakeasy net    | Kivio : http://thekompany.com/projects/kivio |
+-----------------------------+----------------------------------------------+
| I just don't trust anything | I suppose that in a few more hours I will    |
| that bleeds for 5 days and  | sober up. That's such a sad thought. I think |
| doesn't die. - Mr. Garrison | I'll have a few more drinks to prepare myself|
+-----------------------------+----------------------------------------------+



From jason at jorendorff.com  Sun Feb 24 00:27:53 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 23 Feb 2002 23:27:53 -0600
Subject: HELP! Obscure Python Reentrancy Question
In-Reply-To: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEMBDEAA.jason@jorendorff.com>

Courageous wrote:
> What now?

Are you using any actual threads?

Perhaps you could just build a non-threading version
of Python.  I've never done this on Windows, so I don't
know that it's possible.  But I imagine it would get rid
of your problems with the GIL; in such a system there's
nothing for the GIL to protect... (?)

## Jason Orendorff    http://www.jorendorff.com/



From Chris at ShowProgramming.com  Fri Feb 22 02:27:59 2002
From: Chris at ShowProgramming.com (Christopher Duncan)
Date: Fri, 22 Feb 2002 02:27:59 -0500
Subject: OT - Programmer war stories
Message-ID: <cmmd8.3512$2v1.204919@e3500-atl1.usenetserver.com>

Hi, folks.

I'm trying to reach as many seasoned programmers
as I can, hope you don't mind the off topic post.
As part of the research I'm doing for a follow up
to my recent book, "The Career Programmer:
Guerilla Tactics for an Imperfect World" (Apress),
I'm compiling a book of war stories from those of
you who have been out there dealing with all the
insanity that Corporate America has to offer.

In The Career Programmer, I talked about
overcoming the real world obstacles we run into
trying to get decent software out the door
(arbitrary deadlines, clueless management,
Marketing, fuzzy requirements, never any time for
testing, corporate politics - all that stuff they
never taught us in school). I've had tremendous
feedback to the book, and one of the things that
struck me was that as I told stories about the
ridiculous situations I've had to cope with in
my own career, everyone asked if I was looking
over their cubicle walls because they were
dealing with the same things in their shop. So,
if figured it would be fun to do a follow up
book relating the war stories that some of you
have been through. None of my books are tied to
a particular programming language, operating
system or platform, so it doesn't matter what
kind of coding you do. We all deal with the
same sort of craziness.

I'll paraphrase the stories, and of course I
won't use any real names (personal or company) -
we get enough grief from management as it is, no
point inviting any more. I'll be happy to include
your name in the Acknowledgements, but of course
will omit it if you prefer your story to be
anonymous. This book will be in the same, er,
conversational style (that's what my publishers
told me was the politically correct phrase for
being a wise guy) as The Career Programmer was.
I try not to take myself too seriously in this
business - keeps my antacid bills down. But then,
if you've read the last book you probably figured
that out about the time you encountered the
night watchman's attack Chihuahua...

Anyway, if you've got a good war story about
programming in Corporate America, I'd love to
hear it. I may change the format of how I gather
& organize the stories on my web site, so just go
to www.ShowProgramming.com/TheCareerProgrammer.asp
and look for the link on War Stories. That gives
me the flexibility to reorganize the database &
data input page whenever I want without leaving dead
links. And of course, you know I'll have to rewrite
the UI at least a couple of times. There's always
one more tweak.

Oh, by the way, since I get emails from programmers
all over the world, I get teased a little here &
there about my use of "Corporate America", since
programming is obviously an international community.
And from what I've heard, the business world is equally
insane in every country. However, America is the only
one I have experience with. And besides, it was easier
to write than Corporate <Insert Your Country Here>.
Hey, I got lazy, what can I say?

Anyway, if you've got a good story about how crazy this
business can get and wouldn't mind me sharing it with
others, please drop by and add it to the database.
Make sure you leave me an email address that will still
be good when the book comes out (these things tend to
take months to get out the door - sound familiar?).

And thanks very much for taking the time, I truly
appreciate it.

Happy Coding!

Christopher Duncan








From aleax at aleax.it  Mon Feb 18 07:24:53 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 12:24:53 GMT
Subject: There is more than one way to do it - and for no apparent reason.
References: <3C70ECB1.2090709@mxm.dk>
Message-ID: <po6c8.82758$3J1.2641405@news2.tin.it>

Max M wrote:
        ...
> Basically you can say that JScript uses the dictionary notation for
> getting attributes. As fas I can see this is smarter than the way we do

Maybe you can't see very far....?-)

adict = dict()
somerandomwordfromsomewhere = 'get'
adict[somerandomwordfromsomewhere] = 23

Now what should adict.get be?  The number 23, or a bound method that you
can save somewhere and later call with an argument or two?

If items and attributes were confused, you basically couldn't use
bound methods any more -- you could never be sure if by some
mischance somebody has set an item called get, or keys, or items,
or setdefault, or ... whatever other name you might like to use.


> I know it's just syntactic sugar, but it would be so sweet.
> Is there any reason why not to do it?

Hmmm, the risk to one's teeth, maybe?


Alex



From jolsen at mailme.dk  Thu Feb 28 06:07:51 2002
From: jolsen at mailme.dk (Jesper Olsen)
Date: 28 Feb 2002 03:07:51 -0800
Subject: Numeric Python Installation woes
References: <m3elj6ocvs.fsf@mycomp.tky.hut.fi>
Message-ID: <cf0ad9fb.0202280307.2951ec2c@posting.google.com>

I am not sure distutils were part of python 1.5,
and the most recent Numeric package (21) requires python 22 anyway...

Does distutils work for you otherwise, or is this only a Numeric problem?

I have recently installed Numeric 21 on both windows and Linux,
without any problems.

On linux, unless you are the super user, you should use the
--home option, to specify where you want to install the package.

Cheers
Jesper


Karthikesh Raju <karthik at james.hut.fi> wrote in message news:<m3elj6ocvs.fsf at mycomp.tky.hut.fi>...
> Hi All,
> 
> Tried installing Numeric Python from the sources as:
> 
> >> python setup.py install
> 
> error was:
>   File "setup.py", line 170
>     print >>rpm_script, rpm_install_text
>            ^
> SyntaxError: invalid syntax
> 
> When i tried 
> 
> >> python2 setup.py install 
> 
> got this:
> running install
> Traceback (most recent call last):
>   File "setup.py", line 187, in ?
>     ext_modules = ext_modules
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/core.py", line 138, in setup
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/dist.py", line 893, in run_commands
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/dist.py", line 912, in run_command  File "/usr/lib/python2.2/cmd.py", line 112, in ensure_finalized
>     
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/command/install.py", line 268, in finalize_options
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/sysconfig.py", line 408, in get_config_vars
>   File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/sysconfig.py", line 313, in _init_posix
> distutils.errors.DistutilsPlatformError: invalid Python installation: unable to open /usr/lib/python2.2/config/Makefile (No such file or directory)
> 
> 
> No clues, Am i missing something. 
> 
> i have both python 1.5 and python 2.2 and apparently 2.2 version does
> not respond to the upward arrow key for the previous command. Seems
> like i have completely played around with  something major.
> 
> Any help would be great,
> 
> Thankx in advance
> karthik


From jason at jorendorff.com  Wed Feb  6 12:59:22 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 11:59:22 -0600
Subject: RFC - Hygienic macros in Python
In-Reply-To: <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEMHDAAA.jason@jorendorff.com>

For the benefit of the discussion, here is another
example of what a macro could do:

  # I hope this syntax is readable; it's just an
  # example of how Courageous might choose to
  # have things look.
  defmacro synchronized(lock_expr, code_block):
      lock = ^lock_expr
      lock.acquire()
      try:
          ^code_block
      finally:
          lock.release()

This defines a new "synchronized" statement,
such as thread-happy people are always looking for
in Python:

  synchronized fnord.lock:
      fnord.foo()
      fnord.baz(bar)

Another possible example.  Suppose you have written
a library that provides expect-like functionality
and you wish to add that functionality to Python's
syntax (because it's very awkward to call it as a
library; you have to wrap all the code in
functions - very inconvenient).

The syntax you want is like this:

  scan myfile:
      case r'[Ll]ogin':
          send(username + '\n')
      case r'[Pp]assw(or)?d':
          send(password + '\n')
      case r'[%$>]':
          foo()
          bar()
          send(command + '\n')

Using a macro, you could add this syntax to the language.
(I don't know what the code for the macro itself would look
like.)

Ruby allows a small-but-useful subset of this functionality,
but the idea really comes from Lisp.  The "hygienic" part comes,
I think, from Scheme.

## Jason Orendorff    http://www.jorendorff.com/



From SBrunning at trisystems.co.uk  Tue Feb 19 04:22:42 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 19 Feb 2002 09:22:42 -0000
Subject: How do I view Pages on Palm
Message-ID: <31575A892FF6D1118F5800600846864DCBCC90@intrepid>

> From:	jphammock [SMTP:jphammock at email.msn.com]
> downloaded python but dont know how to use it to view web pages on my palm
> any help.
 
Hmmm. Perhaps you want Plucker? <http://www.plkr.org/>

At least *some* of this is written in Python...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From hamish_lawson at yahoo.co.uk  Thu Feb 14 11:26:39 2002
From: hamish_lawson at yahoo.co.uk (hamish_lawson)
Date: Thu, 14 Feb 2002 16:26:39 -0000
Subject: Is there a reverse operator to the '%' operator on string ?
In-Reply-To: <a4glqq$2oj$1@animus.fel.iae.nl>
Message-ID: <a4gofv+gtk5@eGroups.com>

Carel Fellinger wrote:

> Boris Boutillier wrote:

> I want to get the dictionary from the name and the Format something 
> like ...
> 'w%(width)dn%(nWord)d_%(fileName)s' op "w16n12_Toto" 
> which will give {'width':16,'nWord':12,'fileName': 'Toto'}

> What about regex's, like:
> 
> >>> import re
> >>> matcher = re.compile(r"^w(\d+)n(\d+)_(.*)$").match
> >>> matcher("w16n12_Toto").groups()
> ('16', '12', 'Toto')

Adding named groups to Carel's code gives you a dictionary of the 
matches:

>>> import re
>>> pattern = r"^w(?P<width>\d+)n(?P<nWord>\d+)_(?P<fileName>.*)$"
>>> matcher = re.compile(pattern).match
>>> matcher("w16n12_Toto").groupdict()
{'width': '16', 'nWord': '12', 'fileName': 'Toto'}


Hamish Lawson





From phr-n2002a at nightsong.com  Tue Feb 12 18:35:06 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 15:35:06 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org>
Message-ID: <7x8z9y46l1.fsf@ruckus.brouhaha.com>

"Mark McEahern" <marklists at mceahern.com> writes:
> That's probably true.  Would you be so kind as to post an example of the
> input (string, complicated regexp) and output (do you want it to tell you
> the starting position from left to right of the first match found backwards
> in the string?) you'd expect?

Input string: contents of some random html file

Pattern: match the beginning of an html open or close tag:

  pattern = re.compile("<(/?)(.*?)\b")

How do you search backwards in the html file, starting from location 50000?

I propose adding a "direction flag" to the search function:

   match = pattern.search(text, 50000, -1, direction=-1)

here 50000 and -1 are the starting and ending locations.


From nospam at [127.0.0.1]  Thu Feb 28 15:12:06 2002
From: nospam at [127.0.0.1] (Chris)
Date: Thu, 28 Feb 2002 20:12:06 +0000
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
 <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>
 <54abe19d.0202271315.4e699ed0@posting.google.com>
 <b7yVwiC1Cff8EwBv@[127.0.0.1]>
 <54abe19d.0202281032.77cceb54@posting.google.com>
Message-ID: <k+KKxkKW8of8Ew3u@[127.0.0.1]>

In article <54abe19d.0202281032.77cceb54 at posting.google.com>, Anders 
Schneiderman <schneida at seiu.org> writes
>Chris <nospam@[127.0.0.1]> wrote in message news:<b7yVwiC1Cff8EwBv@[127.0.0.1
>
>> Can you do ordinary 2D graphics stuff without VPython - simply using the
>> ordinary Python?
>
>Other folks have replied re: VPython, but one more toy you might check
>out is Alice, at http://www.alice.org/.  Alice is a tool that lets you
>create interactive 3D graphics, and it uses Python as its scripting
>language.  I've never got the chance to play w/ it, but I've heard
>people rave about how much fun it is.  Might be worth a look.

Thankyou.  I've downloaded the plug in to see what it can do.
-- 
Chris


From rony.steelandt at bucodi.com  Mon Feb 18 03:07:13 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 18 Feb 2002 00:07:13 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com>
Message-ID: <b34d3835.0202180007.78d987fd@posting.google.com>

This is interresting ... never sawn this one
I'll try it immediatly

Thanks


Michael Kelly <mkelly2002NOSPAM at earthlink.net> wrote in message news:<rfrv6ucpn8osftj7e2mbj34e1o46271v3q at 4ax.com>...
> On 16 Feb 2002 08:57:15 -0800, rony.steelandt at bucodi.com (Rony) wrote:
> Something I found in Python using COM,
> (after Mark pointed it out to me that is)
> is that if you have "out" parameters other than
> the retval you should use them in multiple assignment.
> I wrote a wrapper for Delphi TIniFile that got a similar
> error only in Python.  It turned out I was passing a
> param to procedures with "out" parameters like:
> 
> myIniFile.ReadSection('section', sectionlist)
> 
> when I should have done:
> 
> sectionlist = myIniFile.ReadSection('section')
> 
> In VB or VC++ I would have to pass the 2 params
> in the parens as expected.
> 
> IOW, since Python has multiple assignment and
> VB and VC++ do not, any out param should be
> assigned to rather than passed as a param.
> 
> Anyway the error was something to do with
> a type error so I'm pretty sure you're getting
> burnt by the same thing.


From gcordova at hebmex.com  Wed Feb 20 17:07:06 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 16:07:06 -0600
Subject: HTTP state management without cookies?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC5@SRM201002>

> 
> Paul Rubin wrote:
> > 
> > <base href=http://foo.bar/cgi-bin/_SID_=36987348732/>
> 
> Are you sure all browsers handle this properly?
> 
> Ciao, Michael.
> 

Only for relative links.

-gustavo



From fperez528 at yahoo.com  Wed Feb 27 16:40:07 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Wed, 27 Feb 2002 21:40:07 +0000
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch>
Message-ID: <a5kcf4$qei$1@peabody.colorado.edu>

Samuele Pedroni wrote:

> Maybe this is a useful insight:
> 
> in mathematetics it is not uncommon to
> identify the natural numbers with the set
> of their predecessors
> 
> 0 = {}
> 1 = {0} = {{}}
> 2 = {0, 1}??{{},{{}}}
> etc

Ok, I won't argue this notation exists in some areas of mathematics, but it's 
definitely _not_ a common one for the 'general public'. I've been doing 
mathematical physics work for years, so at least I have a passing familiarity 
with common notations and conventions in math.

And it's definitely not an intuitive one. It's ok to have particular 
notations which are a bit unusual when they simplify working in a  certain 
problem domain, and all fields of mathematics have them. In cases they may 
even run against established conventions and still be accepted because of 
their usefulness (think of the implied summation for repeated indices in all 
relativity calculations, and the 3/4 range implied in those for latin/greek 
indices). But that's in very specific domains, where you can define well 
under what conventions you are operating at the beginning of the book/paper, 
whatever.

Python is a general purpose language, not one to be used only by number 
theorists. Putting at a level as basic as the integers a convention which 
allows things like 'for i in 3' to be valid iteration code really sounds 
bizarre and counter intuitive to me.

But maybe it's just me.

Cheers,

f.


From mdk at mdk.com  Thu Feb 14 10:45:14 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 10:45:14 -0500
Subject: Need registry editing guidance
Message-ID: <a4gm2d$525q$1@ID-98166.news.dfncis.de>

Have Python 1.5 and win32all

Are there any examples for reading and writing registry entries?

Note1: I may be adding a very long string but I understand that there may be
a size limitation for a string value built into the registry.  I would add
it as a binary value but it looks as though win32api only allows you to
write string values.

Any thoughts or insights?

Note2: Downloading another registry manipulation module is not an option.

Thanks




From skip at pobox.com  Sat Feb  2 22:48:17 2002
From: skip at pobox.com (Skip Montanaro)
Date: Sat, 2 Feb 2002 21:48:17 -0600
Subject: print statement in lambda expression
In-Reply-To: <NW178.16270$bIE.6837@news1.bloor.is>
References: <NW178.16270$bIE.6837@news1.bloor.is>
Message-ID: <15452.45825.932477.18478@12-248-41-177.client.attbi.com>

    Michael> I just recently found that I couldn't put a print statement in
    Michael> a lambda expression as a binding in Tk. Could someone explain
    Michael> why?

Lambdas are restricted to a single expression in their body.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From philh at comuno.freeserve.co.uk  Thu Feb  7 08:36:59 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 13:36:59 +0000
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>
Message-ID: <slrna650nr.ahr.philh@comuno.freeserve.co.uk>

On Thu, 07 Feb 2002 01:04:58 -0500, Peter Hansen <peter at engcorp.com> wrote:
>Timothy Rue wrote:
>> 
>> On 06-Feb-02 12:48:10 Jeff Shannon <jeff at ccvcorp.com> wrote:
>> >The problem you're running into, is that you've failed to convince anyone
>> >that *you* understand the real problems involved, enough to be able to solve them.
>> >[...]
>> Since you have taken it upon yourself to be the spokes person for
>> programmers, 
>
>Actually, we elected him.  Everything he says goes for *all* of us here.
>Keep that in mind as you read the following; Jeff has our complete
>and unconditional confidence and support in this...
>
>> I'll have to assume that all of you are so lost that you
>> can't tell the difference between three deminsional reality and three
>> deminsional data arrays in the two deminsional world of programming.
>
>That's clearly true.  Our spokesman is the best of us, and yet even
>he is clearly (so _you_ say... I'm too stupid to tell) out to lunch.

One thing that puzzles me is why Tim wants help: as he says, we're
all too stupid to understand his grand design (as well as being 
arrogant). 

Since he is so much cleverer and far-seeing than everyone else, he'll
have to code it himself, as no-one else is clever enough to understand it.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From phr-n2002a at nightsong.com  Tue Feb 12 16:04:27 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 13:04:27 -0800
Subject: searching backwards in a string
References: <mailman.1013525800.15169.python-list@python.org>
Message-ID: <7x7kpi76p0.fsf@ruckus.brouhaha.com>

"Mark McEahern" <marklists at mceahern.com> writes:
> > I have a string and want to search backwards for a regexp, starting at
> > a given index, like if you search backwards in a text editor.  Is
> > there a way to do that with re?  The docs aren't clear about what's
> > supposed to happen if the ending position to the pat.search method is
> > less than the starting position, but I tried it and it doesn't seem
> > to do what I want.  Any advice appreciated--thanks.
> 
> Short answer:  Reverse the string first.

Thanks but I don't think that solution works if the pattern being
searched for is a complicated regexp.  Even if the pattern is a simple
string, it has to be reversed too.


From lac at strakt.com  Mon Feb 25 04:42:51 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 25 Feb 2002 10:42:51 +0100
Subject: Parse variable into variables >Re: Apology 
In-Reply-To: Message from "Timothy Rue" <threeseas@earthlink.net> 
   of "Mon, 25 Feb 2002 01:45:48 GMT." <2240.820T2407T12845765threeseas@earthlink.net> 
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com>  <2240.820T2407T12845765threeseas@earthlink.net> 
Message-ID: <200202250942.g1P9gptB011318@ratthing-b246.strakt.com>

> 
> I can read a line from a file but then I want to parse the line into a set
> of variables. How do I do the parse?
> 
> If I have to do this a baby step at a time then so be it.

Somebody recommended regular expressions.  They are quite difficult to
use.  Don't use them unless you really need to -- and most of the time
python programmers do not need to.

If by 'parse the lines into a set of variables' you mean 'split
the lines into whitespace separated words' then this will work for you.

lac at ratthing-b246:~/scratch$ python
Python 2.1.2 (#1, Jan 18 2002, 18:05:45)
[GCC 2.95.4  (Debian prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> line='a b c d e f g h i jj kk ll mm nnn ooo ppp qqq rrr ssss tttt'
>>> print line
a b c d e f g h i jj kk ll mm nnn ooo ppp qqq rrr ssss tttt
>>> words=line.split()
>>> print words
['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'jj', 'kk', 'll', 'mm', 'nnn', 'ooo', 'ppp', 'qqq', 'rrr', 'ssss', 'tttt']
>>> print words[0], words[6]
a g
>>>
-------------
Happy hacking,
Laura Creighton



From roy at panix.com  Sat Feb  2 10:29:38 2002
From: roy at panix.com (Roy Smith)
Date: Sat, 02 Feb 2002 10:29:38 -0500
Subject: How to distribute code with required modules?
Message-ID: <roy-CE500A.10293802022002@news1.panix.com>

I work in a shop that uses a lot of perl for internal tools.  I've been 
writing stuff (again, internal tools) in python, and trying to figure out a 
way to get the language to be more accepted.  Yesterday, I hit upon a 
stumbling block.

The way we distribute tools is we just check them into a tools CVS module.  
Everybody does "cvs update" on a regular basis, and gets any new tools that 
have been added to the repository.  I've written a lot of python classes 
for handling the types of data files we work with.  All of my tools depend 
on some combination of those classes.  I created a subdirectory of tools 
called pylib, and check all my class modules into there.  Then I set by 
PYTHONLIB environment variable to "$HOME/sandbox/src/tools/pylib".

So, here's the problem.  Somebody goes to run one of my tools, and it can't 
find my classes.  I explain to the guy that he just needs to set PYTHONPATH 
and all will be well.  I get a bit of pushback because none of his perl 
tools require you to set a PERLLIB" variable; he just has code which looks 
at $0, chops off the last component (leaving the directory from which the 
script was run), and then appends the name of the module he wants:

if ($0 =~ m/(.*\/)/) {
    $rundir = $1;
}
require "$rundir/module.pm";

He thinks my python scrips should do something similar (which would 
certainly be possible).  I'm trying to convince him that it would be very 
un-pythonic to do it that way.

Any thoughts?


From caljason76 at yahoo.com  Sun Feb 17 20:31:04 2002
From: caljason76 at yahoo.com (Jason)
Date: 17 Feb 2002 17:31:04 -0800
Subject: execute file interactively
Message-ID: <e9e6dd1f.0202171731.6cf2a5e2@posting.google.com>

When using the Python interpreter interactively, how can I execute a
file, so I can use the function definitions and data defined in it
interactively.


From starx at pacbell.net  Fri Feb  1 23:23:40 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 2 Feb 2002 04:23:40 GMT
Subject: Making a list-inhereted object return standard repr?
Message-ID: <Xns91A8CF7752ED0RASXnewsDFE1@130.133.1.4>

    	Ok, so i decided to make my container class inherit from list, 
considering that that's basically what it is. Now i've run into a problem 
where my self_parent exception is being raised for whatever reason, and the 
first thing i try is sticking a 'print self, assigend' sort of statement at 
the beginning of the function. The problem is, instead of getting something 
like '<__main__.mapcell instance at 0xf00fee>', i get '[]'. I want to 
change __repr__ and __str__ to do something more like the first case again 
(possible with an added 'containing [...]' element), but the only 
information i can find to build it out of is what's available from id() and 
type(), ... i can't find something that lets me mimic the first part of the 
message (id is useful, type isn't). 

    	Thanks.

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From guido at python.org  Thu Feb  7 21:34:02 2002
From: guido at python.org (Guido van Rossum)
Date: Thu, 07 Feb 2002 21:34:02 -0500
Subject: Slides on the web
Message-ID: <200202080234.g182Y2323550@pcp742651pcs.reston01.va.comcast.net>

I've put some new sets of powerpoint slides on the web, at

    http://www.python.org/doc/essays/ppt/

Scrolling to the end, you'll find:

OSCON 2001 Keynote 
      The slides I used for my keynote-in-absentia at the Open Source
      Convention 2001, in San Diego, CA.

Introduction to Python
      A 3 hour introductory tutorial that I gave at LinuxWorld in New
      York City, January 2002. It covers Python 2.2 (but mostly sticks
      to timeless features).

What's New in Python 2.2
      What the title says; a talk that I gave at LinuxWorld in New
      York City, January 2002, and earlier at a Washington, DC ZPUG
      meeting.

Python 10 DevDay Keynote
      The slides I used for my Developers Day keynote at the Python10
      Conference, February 2002 in Alexandria, VA.

Note: I've given up making HTML.  There are plenty of open source
tools that can read PowerPoint files now, and the HTML that PowerPoint
generates gets worse with each version it seems. :-(

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



From thomas.heller at ion-tof.com  Fri Feb  1 04:13:08 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Feb 2002 10:13:08 +0100
Subject: ANN: py2exe version 0.3.0 released
Message-ID: <mailman.1012673766.20189.clpa-moderators@python.org>

py2exe is a distutils extension to convert python scripts
into windows exe-files (plus maybe a few dlls), able to run
on computers without requiring a python installation.


New in release 0.3.0:

  The first release which can create windows NT services:
  A sample service with detailed instructions is provided.
  Builds version resources into the executables.
  Other small bug fixes as well.


Download from the usual location:
http://starship.python.net/crew/theller/py2exe/


Thomas











From stallard at capgemini.fr  Fri Feb 15 11:07:08 2002
From: stallard at capgemini.fr (Stephane Tallard)
Date: Fri, 15 Feb 2002 17:07:08 +0100
Subject: cygwin and python
References: <u6oe6n32m9pjfb@corp.supernews.com>
Message-ID: <3c6d2f5e@news.capgemini.se>

Hi all,

I use python under cygwin bash-rxvt . Trying to use omniorbPy,  I'm facing a
strange behaviour :

When executing omniorbPy example ( command line "python  echo_srv.py" )
script under ugly Command.com, everything is OK.

When executing the same command line under rxvt , nothing happens: the
script never ends. Moreover if I add print "Hello" at the beginning of the
script, "Hello" is never prompted ...

It seems that cygwin changes the behaviour of python interpretor ....

Anyone has an idea ? .

Stephane








From sholden at holdenweb.com  Tue Feb 12 22:12:44 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Feb 2002 22:12:44 -0500
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com>
Message-ID: <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7x8z9y46l1.fsf at ruckus.brouhaha.com...
> "Mark McEahern" <marklists at mceahern.com> writes:
> > That's probably true.  Would you be so kind as to post an example of the
> > input (string, complicated regexp) and output (do you want it to tell
you
> > the starting position from left to right of the first match found
backwards
> > in the string?) you'd expect?
>
> Input string: contents of some random html file
>
> Pattern: match the beginning of an html open or close tag:
>
>   pattern = re.compile("<(/?)(.*?)\b")
>
> How do you search backwards in the html file, starting from location
50000?
>
> I propose adding a "direction flag" to the search function:
>
>    match = pattern.search(text, 50000, -1, direction=-1)
>
> here 50000 and -1 are the starting and ending locations.

Paul, this thread's probably now old enough for you to tell us what the real
problem is! Why exactly do you need to search backwards from the 50,000th
character to find the beginning of an HTML tag?

trust-me-i'm-a-programmer-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From afp_randjohnson at yahoo.com  Wed Feb 27 04:37:39 2002
From: afp_randjohnson at yahoo.com (Mike Johnson)
Date: 27 Feb 2002 01:37:39 -0800
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de> <a5j53i$oms$1@wanadoo.fr>
Message-ID: <8z9f2rks.fsf@fido.trinitycapital.com>

"Riss Nicolas" <alien275 at libertysurf.fr> writes:

> I jave used it but when y tried to connect it makes an error:
>    File "c:\test.py", line 9, in ?
>     tn = telnetlib.Telnet("192.168.0.9",32)
>   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 190, in __init__
>     self.open(host, port)
>   File "C:\PROGRA~1\python22\lib\telnetlib.py", line 219, in open
>     raise socket.error, msg
> socket.error: (10061, 'Connection refused')
> 
> Programm:
> 
> import getpass
> import sys
> import telnetlib
> password = getpass.getpass()
> tn = telnetlib.Telnet("192.168.0.9",32)
> tn.read_until("Administrator Password : ")
> tn.write(password + "\n")
> tn.write("session\n")
> 
> 
> Where is the mistake ?

Wrong port. Try:

tn = telnetlib.Telnet("192.168.0.9", 23)


- Mike Johnson



From gh_pythonlist at gmx.de  Mon Feb 18 14:44:43 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 20:44:43 +0100
Subject: Python calling Java
In-Reply-To: <a4rjq4$4n8$1@web1.cup.hp.com>
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <20020218194442.GA8913@lilith.hqd-internal>

Le 18/02/02 ? 11:13, jimh ?crivit:
> We have a bunch of Python (cgi) scripts producing html behind an Apache web
> server.  We also have a set of Java libraries that I want to access from
> these Python scripts, but I haven't been able to figure out how to do this.
> I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
> a one-line 'print "hi"' script, and this is, of course, not acceptable for a
> web server.
> 
> Any ideas?

There IMO is no easy solution to this problem. It might be easiest to
rewrite all the CGI scripts as Jython (or Java) servlets.

There are a bunch of other possibilies I can imagine, but they require
more effort. Trying to get jython cgi scripts to work with FastCGI would
be one.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.0 ?C      Wind: 5.8 m/s



From tim.one at home.com  Wed Feb 13 23:29:43 2002
From: tim.one at home.com (Tim Peters)
Date: Wed, 13 Feb 2002 23:29:43 -0500
Subject: Using tabs instead of spaces in IDLE?
In-Reply-To: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOECENMAA.tim.one@home.com>

[Tom Bryan]
> Is there a way to use tabs exclusively for indentation in IDLE?

Yes, provided you're happy with indentwidth==8.

> Basically, I'd like tabwidth to equal identwidth while permitting
> developers to adjust the indentwidth.

You're out of luck there, if they want anything other than 8 for both.

>  I see that AutoIndent.py says
>
> # indentwidth is the number of characters per logical indent level.
> # tabwidth is the display width of a literal tab character.
> # CAUTION:  telling Tk to use anything other than its default
> # tab setting causes it to use an entirely different tabbing algorithm,
> # treating tab stops as fixed distances from the left margin.
> # Nobody expects this, so for now tabwidth should never be changed.
> usetabs = 1
> indentwidth = 4
> tabwidth = 8    # for IDLE use, must remain 8 until Tk is fixed
>
> But I don't quite understand that explanation.

Don't be so timid <wink>:  change the code and see what happens.  The
meaning of the comment will soon become painfully clear.  Have you ever used
a word processor (as opposed to a text editor)?  Tk switches to "word
processor" tabs when you change its default tab setting.

> ...
> It doesn't look like IDLE is friendly to that type of customization.

Not exactly:  the Tk text widget is the hangup here.  IDLE's autoindent code
is shared by PythonWin, which latter, being based on the Scintilla text
widget, doesn't suffer the same tab surprises.

> I know that emacs is.  I'm currently investigating the other editors
> that my team uses, such as CodeWright.

CodeWright won't be a problem.  But stick to IDLE's default (== the Emacs
python-mode default), and no editor anywhere will be a real problem.




From usenet at thinkspot.net  Fri Feb  8 16:14:48 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 08 Feb 2002 13:14:48 -0800
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <mailman.1013159835.28909.python-list@python.org>
Message-ID: <a40j09.3vvegs9.1@kserver.org>

On Fri, 8 Feb 2002 12:16:07 +0300, Oleg Broytmann <phd at phd.pp.ru> wrote in
comp.lang.python in article
<mailman.1013159835.28909.python-list at python.org>:

> It is amazing how fast a dozen of lamers, few trolls and a free software
> advocate can destroy the good old friendly spirit of c.l.py.

My thoughts exactly. :(

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From graz at mindless.com  Wed Feb 20 14:58:27 2002
From: graz at mindless.com (Graham Ashton)
Date: Wed, 20 Feb 2002 19:58:27 +0000
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <slrna738ab.akv.gerhard@lilith.my-fqdn.de>
Message-ID: <W9Tc8.37$PE2.5068@stones>

On Tue, 19 Feb 2002 00:50:19 +0000, Gerhard H?ring wrote:

> Ada 95 is also one of my favourite languages, but I've never used it for
> a real project (well, I once tried it use it for putting my gateway
> computer online/offline, ... [snip]

I did one of those in Python, primarily to play around with XML-RPC. It's
called landialler, on sourceforge.

-- 
Graham


From justin at iago.org  Tue Feb 26 21:32:02 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 26 Feb 2002 21:32:02 -0500
Subject: what I would like to see in python to make a better "glue"
 language (Newbie)
In-Reply-To: <a5g3g1.3vsg8p3.1@kserver.org> (Sheila King's message of "Tue,
 26 Feb 2002 13:44:33 -0800")
References: <899f842.0202261325.6ff3dd54@posting.google.com>
	<a5g3g1.3vsg8p3.1@kserver.org>
Message-ID: <x7vgcjbqot.fsf@ra.iago.org>

Sheila King <usenet at thinkspot.net> writes:

> I'm curious about the reference to Anthony Baxter and a possible DNS module
> for including in the standard library. More info? Comments?

The module being referred to started out as a module written by Guido,
and has been extended by Anthony and others.

It is only available right now via CVS.

(http://sourceforge.net/projects/pydns/)

I've used it.  It's very useful.  I would be happy if this made it
into the standard library at some point.

-Justin

 




From peter at engcorp.com  Wed Feb 13 23:06:06 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 13 Feb 2002 23:06:06 -0500
Subject: Opening random line from file with random.shuffle()
References: <87bses6g1p.fsf@uol.com.br>
Message-ID: <3C6B37AE.EB1E3DF7@engcorp.com>

synthespian wrote:
> 
>  I need to get a random line from a file.  I read in thins NG about a 
> "poorly" documented random.shuffle(), but I can't get it to work:

I just checked the documentation.  It seems well enough documented
to me, even indicating that it does not return anything but shuffles
the sequence in place.  Right above it is the choice() method,
which clearly behaves as you wish.  I find the comment about 
poor documentation quite curious in this case.


From emile at fenx.com  Tue Feb 26 11:42:56 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 08:42:56 -0800
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb 26)
References: <5A65FB556DC4B92B.47C0A0259510F4DA.27E89041417BA1D1@lp.airnews.net>
Message-ID: <a5ge8i$74oc1$1@ID-11957.news.dfncis.de>

"Jason Orendorff" 

>     Alex Martelli exposes with utter lucidity why Visual Basic,
>     and not Python, is the risky and speculative technology.
> http://groups.google.com/groups?th=5773a3f58c6b95aa
> 

Oops, wackyparsed that as:

    Alex Martelli exposes the utter ludicrity why Visual Basic,
    and not Python, is the risky and speculative technology.

;-)


Some good references to posts I missed.

Thanks Jason!

-- 

Emile van Sebille
emile at fenx.com

---------



From emile at fenx.com  Sun Feb 17 14:36:59 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 17 Feb 2002 11:36:59 -0800
Subject: nested scopes
References: <CfTb8.129031$th4.36971458@news02.optonline.net> <zsTb8.129039$th4.37009591@news02.optonline.net>
Message-ID: <a4p13e$1r4ju$1@ID-11957.news.dfncis.de>

"Nicola Kluge" <NKluge at optonline.net> wrote in message
news:zsTb8.129039$th4.37009591 at news02.optonline.net...
> Nevermind, I found that it doesn't work in IDLE, but it works in
cmdline
> Python.
>

from __future__ import nested_scopes  needs to be the first thing done,
and in idle you never get to be first...

It does work, of course, in idle under 2.2.

--

Emile van Sebille
emile at fenx.com

---------



From replogle2 at cox.net  Sat Feb  2 00:10:03 2002
From: replogle2 at cox.net (Andrew Replogle)
Date: Sat, 02 Feb 2002 05:10:03 GMT
Subject: Python System Administration
Message-ID: <LwK68.7172$Xk4.381742@news1.east.cox.net>

   Anyone know of any books out there for python system administration or
should I just stick to perl for these tasks?




From aahz at panix.com  Wed Feb 20 11:03:46 2002
From: aahz at panix.com (Aahz Maruch)
Date: 20 Feb 2002 08:03:46 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com>
Message-ID: <a50hd2$6vt$1@panix2.panix.com>

In article <m2wux85hud.fsf at desk.crynwr.com>,
Russell Nelson  <nelson at crynwr.com> wrote:
>Henrik Motakef <henrik.motakef at web.de> writes:
>> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
>>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
>>>> 
>>>> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
>>>> czignat?  Classifero en bragna!
>>> Woas da Deifi wos des hoasn mog.
>> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
>
>Even better: we could switch languages within threads, to *really*
>keep the monolinguists confused.

B'vakasha
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"Not everything in life has a clue in front of it...."  --JMS


From martin at v.loewis.de  Tue Feb 26 05:17:33 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 11:17:33 +0100
Subject: PEP 263 comments
In-Reply-To: <3C7B5E35.129E5501@lemburg.com>
References: <200202250520.g1P5KKD01484@mira.informatik.hu-berlin.de>
	<3C7B5E35.129E5501@lemburg.com>
Message-ID: <m31yf8fsxu.fsf@mira.informatik.hu-berlin.de>

"M.-A. Lemburg" <mal at lemburg.com> writes:

> Thanks for the comments. I've update the PEP at SourceForge...

Thanks! Is there anything that needs to be resolved before we can ask
for BDFL pronouncement? According to PEP 1, the PEP editor must be
informed that the PEP is ready for review; I suggest it is (after this
batch of changes).

> I've added a two phase approach to the PEP: first we only
> handle Unicode literals, then we do the whole file in a later
> step.

Sounds good.

> I did not add the IDLE changes to the PEP. Please upload them
> as feature request to SF.

Ok, I will.

Regards,
Martin



From jason at jorendorff.com  Mon Feb  4 20:43:31 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 19:43:31 -0600
Subject: heres a dumb question....
In-Reply-To: <a3mp6v$u04$1@wiscnews.wiscnet.net>
Message-ID: <HFEKILOLEFEFMKAECNDLCEGEDAAA.jason@jorendorff.com>

Jeremy Bush writes:
> [...] cant seem to figure out how to get keyboard input...
> ive read the whole readme that came with it, the tutorial,
> everything.

raw_input() gets a line of input.

Unlike C++, you ordinarily get a line at a time, and split it
to get the information you want.

For example, if you expect the user to type in numbers
separated by commas, you'd do:

line = raw_input('> ')   # see? you can provide a prompt
fields = line.split(',')
for s in fields:
    x = float(s)  # convert string to float
    print "you entered the number:", x

If you "import sys", then you have sys.stdin, which is a file object.
File objects have various methods for input.  Viz:
  http://www.python.org/doc/current/lib/bltin-file-objects.html

Enjoy Python!

## Jason Orendorff    http://www.jorendorff.com/




From usenet at thinkspot.net  Sat Feb 16 22:41:12 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 16 Feb 2002 19:41:12 -0800
Subject: How to get CNAME with sockets ?
References: <a4lpg1.3vvdg2v.1@kserver.org> <a4n5sl$9uj$0@216.39.172.122>
Message-ID: <a4mckp.3vvdvvf.1@kserver.org>

On 17 Feb 2002 02:52:05 GMT, bokr at oz.net (Bengt Richter) wrote in
comp.lang.python in article <a4n5sl$9uj$0 at 216.39.172.122>:

> I don't know what you are really doing, but that sounds like
> systematically dialing all phone numbers in various area codes
> until you get Arnold Schwarzenegger ;-)
> 
> Can't you get a copy of the 'phone directory and find him yourself?
> Or do you need to visit with all those others?
> 
> Or is the metaphor totally inapplicable? ;-)

I don't think the metaphor is *totally* inapplicable, but I can see why
you would draw that conclusion without looking at the code.

Here is a snippet from the original Perl module:

==============(begin snippet)==============
    BUILDLIST: 
    for ( "a".."z", "1..9999" ) { 
        my $query = $resolver->query ($_.'.'.$zone);
        my $rr; next BUILDLIST unless $query;
        foreach $rr ($query->answer) { 
            my $pushed = 0;
            if ($rr->type eq "A") { 
                push @list, $rr->address; 
                $pushed = 1;
            } elsif ($rr->type eq "CNAME") { 
                if ($rr->cname eq 'list.terminator') { 
                    pop @list if $pushed;
                    last BUILDLIST;
                } elsif ($rr->cname eq "skip") { 
                    pop @list if $pushed;
                    next BUILDLIST;
                }
            }
        }
============(end snippet)==================

It's more like, the hosts will be
a.zone
b.zone
c.zone
etc...
until there are no more hosts. The last host will have a CNAME that
resolves to list.terminator. I don't know WHY the original code adds
subdomains 1..9999 onto the list, since there are currently only TWO
razor servers that I can discover in the default zone, which is
razor.vipul.net

In my own Python adaptation, I did this:

===========(begin snippet)=================
subdomainlist = list(string.ascii_lowercase)

zone = 'razor.vipul.net'
host_IP_list =[]
response_times = {}

for subd in subdomainlist:
    host = '%s.%s' % (subd, zone)
    try:
        IPaddy = socket.gethostbyname(host)
        host_IP_list.append(IPaddy)
    except socket.error, e:
        pipe = os.popen('dnsquery -t CNAME %s' % host)
        data = pipe.read()
        if data.find('list.terminator') > -1:
            break
===========(end snippet)=================

Notice that I left the 1..9999 off.

By the way, that isn't the complete translation of the razor-discover
client. Only a snip. And it isn't portable, either. Designed to work on
my web host with tools they have installed.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From jonathan at onegoodidea.com  Fri Feb  1 14:36:18 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 01 Feb 2002 19:36:18 +0000
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu>
Message-ID: <B8809EB2.411E%jonathan@onegoodidea.com>

On 1/2/2002 18:09, in article a3ell7$29bm$1 at nntp6.u.washington.edu, "Donn
Cave" <donn at u.washington.edu> wrote:

> Can I ask what operating system that experiment was performed on?
[...]
> That's true, but "executable" in this context usually means the binary,
> i.e., python itself.  Years ago, some UNIX platforms introduced a hack
> that made setuid effective for script files, but today that is universally
> recognized to have been misguided.  That's why I'm curious about yours.

Hah! How curious. Of course, you're right. I had forgotten this. Though it
makes sense from the user's point of view - if being somewhat insecure
(personally I prefer it in this world where the difference between compiled
programs and scripts is blurred).

To answer your question, I ran my test on Mac OS X. It seems OS X still
retains the older semantics for script execution. Unsurprising since it is
based on an old BSD.

> Perl is subject to the same limitation, but has optional support for
> setuid.  How can that be?  I'm no Perl whiz, but I assume you install
> perl itself setuid root, and then it checks its input file for setuid
> and either sets its ID accordingly or reverts to the invoker's ID.
> This option is not the default and is rather discouraged as I recall it.

Hmmmm... to return to the original question, in the absence of perl playing
tricks with setuid, I would suggest a wrapper as being the best bet.

You'll need to write a small program along the lines of:

    #include <stdlib.h>

    int main( int argc, char* argv[] )
    {
        char* args[] = { "python", "/my/script.py", NULL };
        execv( "/usr/bin/python", args );
    }

Compile this, setuid it and then drop that into the cgi directory instead of
the Python script. That should do the trick I think. If you can't compile
stuff on the box or upload executables compiled elsewhere, then you're in
trouble.

Something to ask the providers is if they are willing to consider using
Apache's suexec to invoke your CGIs. I have used that in the past to allow
customers to run things as themselves.

Jonathan



From reanes at banta-im.com  Tue Feb  5 11:40:12 2002
From: reanes at banta-im.com (Robert Eanes)
Date: 5 Feb 2002 08:40:12 -0800
Subject: when does python free memory?
Message-ID: <10b3f29a.0202050840.7641923a@posting.google.com>

Hi-
Reading the docs and searching the list archives, it seemed that
python should release memory as soon as the last reference to an
object is deleted.  However, I can't seem to prove this in practice. 
Look at this example:

Python 1.5.2 (#1, May 11 2000, 11:19:54)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
>>> memsize = blah.readlines()
>>> print memsize
['23240 2360\012']
>>> z = ["LSDKFJLSKDFJSDF"]*10000000 #allocate a large list
>>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
>>> memsize = blah.readlines()
>>> print memsize
['23240 41424\012']
>>> del z #get rid of the large list
>>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
>>> memsize = blah.readlines()
>>> print memsize
['23240 41424\012']

The first number that comes out of the "print memsize" statement is
the pid, the second is the amount of memory used.  It doesn't seem to
be actually freeing the memory that was allocated to variable z.  Now
I played around with it a little more and if I reallocate another list
it does re-use that memory:

>>> x = ["LSDKFJLSKDFJSDF"]*10000000
>>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
>>> memsize = blah.readlines()
>>> print memsize
['23240 41424\012']

So apparently python knows that the memory formerly allocated to z is
available, but it doesn't tell the OS.  Is this the case, and if so,
is there any way around it?  I have an application that needs to run
80-150 python processes that can stay around for 15min to an hour, so
I'd like them not to hold onto the maximum amount of memory that they
use throughout the whole life of the process.

Thanks for any help,
-Robert


From lbergman at abi.tconline.net  Sat Feb  9 15:39:10 2002
From: lbergman at abi.tconline.net (Lewis Bergman)
Date: Sat, 9 Feb 2002 14:39:10 -0600
Subject: (mod_python) - add_handler gives index of dir
Message-ID: <200202092039.g19KdBq19300@lewis.abi.tconline.net>

I have posted this to the mod_python list but haven't gotten any ideas yet. I 
wonder if anyone here has worked with mod_python and the publisher?

As indicated, I have used an add_handler statement. This statement, as 
follows, attempts to dynamically change the resultant page viewed by the 
permissions of the user. No errors are reported in the apache log file (which 
is the case if I do something I shouldn't. On execution with a browser I get 
an auth popup like I should, am authed or rejected as appropriate. But 
instead of my expected page I get an index of the directory. Can anyone here 
point me in the correct direction please?

def authenhandler(req):
	"""Provide the correct page for the privilege of user."""

	pw = req.get_basic_auth_pw()
	user = req.connection.user
	if user == "admin" and pw == "admin":
		req.add_handler("PythonHandler", "setup::admin")
		return apache.OK
-------snip rest------------------------------------

The setup.admin is a class in the same directory which contains different 
functions to generate different pages. A part of it follows:

class setup:

	def _frame(frameLeft, frameRight):
		"""Sets up frames.
	
		frameleft and frameright will be replaced by the correct menu function.
		"""
		html = """\
			<html>
			<head>
				<title>ISPMgr</title>
				<link rel=stylesheet type="text/css" href="ispmgr.css" title="stylesheet">
			</head>
			<frameset cols="20%%,*">
				<frame src="%s" scrolling="no" name="menu">
				<frame src="%s" scrolling="no" name="main">
			</frameset>"""% (frameLeft, frameRight)

		return html

	def admin():
		"""Provide all options to admin users."""
		r = _frame('menu.admin', 'main.customer')
		return r
----rest snipped-----------------------------


-- 
Lewis Bergman
Texas Communications
4309 Maple St.
Abilene, TX 79602-8044
915-695-6962 ext 115



From jason at jorendorff.com  Sun Feb  3 12:04:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 11:04:57 -0600
Subject: lambda expressions
In-Reply-To: <3c5d4c21.38617375@news.t-online.de>
Message-ID: <HFEKILOLEFEFMKAECNDLAEBEDAAA.jason@jorendorff.com>

Gerson Kurz writes:
> So, I see few limitations in the current implementation of lambda for
> the DEDICATED lambda-user.

Lack of tail recursion could be a problem.

## Jason Orendorff    http://www.jorendorff.com/



From sholden at holdenweb.com  Wed Feb 13 07:13:36 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 07:13:36 -0500
Subject: DBI possibly bogus error message
References: <h37c5ugr0t6dn3gt37mf7em7oddrjrddf2@4ax.com>
Message-ID: <KLsa8.205875$KM2.8417487@atlpnn01.usenetserver.com>

"Lee" <lee at jamtoday.com> wrote ...
> I'm a brand new Python user.  I've got Active Python (and Active Perl)
> installed on my wintel sysem.
>
> I'm using the pyperl module to reach  perl's DBI and DBD-Oracle
> modules.
>
> I follow the example in dbi2.py (which has an error, in the connect
> statement  by the way,instead of DBI:Oracle:database=blahblah  I
> needed DBI:Oracle:blahblah).
>
> After some mucking about, I got an error message "Cant locate
> Python/Object.pm in @INC (@INC contains: ...<my path to Perl/lig and
> Perl/site/lib> )
> BEGIN failled--compiation aborted
>
> After that unpromissing beginning, lo and behold, I see the data I've
> selected from my oracle table neatly printed out as advertized.
>
> Anybody have a clue as to whats going on?
>
> Is the error truly bogus? If not, whats the problem and how do I fix
> it? If so, how do I surpress it?
>
Welcome to c.l.py.

Using pyperl to access Oracle is like taking a bus from New York to San
Francisco so you can fly to Los Angeles. There are perfectly good *Python*
modules for database access, the best-known for Oracle being cxOracle from
Computronix and DCOracle2 from Zope Corporation.

Save pyperl for when you *absolutely have* to use Perl-only functionality in
Python, and you'll limit your grief. In other words, "Don't do that!"

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From phr-n2002a at nightsong.com  Thu Feb 21 10:11:35 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 07:11:35 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014303815.10286.python-list@python.org>
Message-ID: <7xbseizx7c.fsf@ruckus.brouhaha.com>

Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> > You then use a server side rewrite rule to map the SID path component
> > into a query parameter.
> 
> I meant that WITHOUT using a rewrite rule, you would have to encode the
> session ID unto every URL as a query param.

Yes, I think we're in agreement.  mod_rewrite lets you do all kinds
of neat stuff.


From tripps81 at yahoo.com  Tue Feb 19 13:19:34 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Wed, 20 Feb 2002 01:19:34 +0700
Subject: Adding static typing to Python
In-Reply-To: <1ovc8.54286$X64.19378143@news1.rdc2.pa.home.com>
References: <mailman.1014098489.15429.python-list@python.org>
 <a4ss56$akj6@news1.gtech.com>
Message-ID: <5.1.0.14.2.20020220011822.0a66a790@202.95.131.154>

On 19/02/2002 23:51, Terry Reedy wrote:
>To me, this says that the 'strong typing' of C is a bit of a fraud!

but who says C is ever a strongly typed language? you can always 
defeat the compiler's check with typecasting.

t




From tim.golden at viacom-outdoor.co.uk  Thu Feb  7 04:35:01 2002
From: tim.golden at viacom-outdoor.co.uk (Tim Golden)
Date: Thu, 7 Feb 2002 09:35:01 -0000 
Subject: Autocoding evolves from........
Message-ID: <5F5E25C9F098D3119F140008C784457302D55EE6@TDI_NT9>

AYE

________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________



From phr-n2002a at nightsong.com  Wed Feb 20 18:37:18 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 15:37:18 -0800
Subject: Recursively traverse linked list -- help!
References: <mailman.1014240283.12938.python-list@python.org>
Message-ID: <7xadu3wwr5.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> > > def f(val):
> > >   global f
> > >   f = val and f1 or f2
> > >   return f(val)
> > 
> > This is not tail recursion.  The "global f" declaration shadows
> > the "f" that names the function.
> 
> That's not true.  The f that names the function is global anyway.
> It's the assignment that causes the problem.

Oh yes, you're correct.  However, the tail call should still be
optimized, compiled as an indirect jump to f.


From quickdry at users.sourceforge.net  Wed Feb 20 22:02:56 2002
From: quickdry at users.sourceforge.net (Quick Dry)
Date: Thu, 21 Feb 2002 14:02:56 +1100
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <3c746343$0$16639$afc38c87@news.optusnet.com.au>

"Kevin Dahlhausen" <kdahlhaus at yahoo.com> wrote in message
news:283adf56.0202201551.7026d548 at posting.google.com...
> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

Infinology (www.infinology.com) provide ASP, PHP, Perl and Python on Linux
servers.

Python isn't officially supported as far as I know, but it is there.

Pricing starts at around $16 per month IIRC

Steven




From eppstein at ics.uci.edu  Mon Feb 25 14:13:01 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Mon, 25 Feb 2002 11:13:01 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <slrna7l176.3al.quinn@vomit.ugcs.caltech.edu>
Message-ID: <eppstein-883A5E.11130125022002@news.service.uci.edu>

In article <slrna7l176.3al.quinn at vomit.ugcs.caltech.edu>,
 quinn at vomit.ugcs.caltech.edu (Quinn Dunkan) wrote:

> >I'd prefer
> >[x.f() for x in items]
> >
> >It's not functional syntax, but so what?
> 
> Sure it is.  It's an expression, works best when x.f() has no 
> side-effects, and provides nicer syntax for mapping and filtering, two 
> popular functional concepts.  It's also borrowed from haskell, which, 
> although functional, has a lot of syntax sugar that turns into function 
> application.

To me it's like the difference between using coordinates or avoiding them 
in linear algebra.  A true functional syntax, to me, would avoid referring 
to the individual items in the list and just combine the list with a 
function, so "map(f, items)" is purer to me than "[f(x) for x in items]", 
because it avoids using any x's.  Which in turn is purer than
    L = []
    for x in items:
        L.append(f(x))

But, as I said, for the x.f() case, I prefer the list-comprehension syntax, 
because simplicity and ease of understanding are more important than purity.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From sanjay at jarna.com  Mon Feb 18 18:05:40 2002
From: sanjay at jarna.com (Sanjay)
Date: 18 Feb 2002 15:05:40 -0800
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com> <3C6D08C1.2000307@skippinet.com.au> <9d4c7023.0202151038.1d3307f@posting.google.com> <3C6DAF39.5080808@skippinet.com.au>
Message-ID: <9d4c7023.0202181505.1646f9e6@posting.google.com>

Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6DAF39.5080808 at skippinet.com.au>...
> Sanjay wrote:
> > What I am trying to do is something that was somewhat discussed on a
> > previous thread, but I'm not sure the thought was fully completed. 
> > Specifically, I would like to write an exchange server event sink by
> > implementing two com interfaces, IExStoreAsyncEvents and
> > IExStoreSyncEvents.  The problem I'm having is that when the python
> > event listener is loaded as a com+ application, and I register this
> > application for events with exchange, I get the error TypeError ...
> > IID not supported.  From looking at the previous threads, it seems as
> > this problem is caused by not being able to handle arbitrary vtable
> > interfaces, such as the interface methods of IExStoreAsyncEvents and
> > IExStoreSyncEvents.  If, however,  the two com classes
> > IExStoreAsyncEvents and IExStoreSyncEvents were defined in a pyd, then
> > pythoncom would have knowledge of these interfaces, and properly call
> > my application.
> > 
> > I hope I'm explaining the problem clearly. I am a little confused as
> > to how to get around this with pythoncom.
> 
> You need t6o check out the win32com\servers\test_pycomtest.py.  This 
> shows how arbitary vtables must be loaded by your code before it is 
> recognized.  This in-turn implies that the interface is described in a 
> type library.
> 
> Mark.

Ok, I looked at test_pycomtest.py and changed my interface to match,
i.e. I set _public_methods_ to an empty list, _com_interfaces_ to
"[IExStoreSyncEvents]", I created a new _reg_clsid_ value, and
implemented the two methods in the interface.  I then registered the
class using register.UseCommandLine() with debugging turned on.  Then
I told Exchange to notify this object when new events happen.  In the
windows Event viewer, I now see the following error when events occur
in Exchange and it tries to call my python com object:

    Exception: C0000005
    Address: 0x1E12E6A5
    Call Stack: 
    python22!PyDict_Next + 0xC5
    ole32!StringFromCLSID + 0xB91
    ...
    ...

There doesn't seem to be anything strange in the Python Trace
Collector window.  I see a message saying "Object with win32trace
dispatcher created (object=None)", followed by 3 calls to
QueryInterface with unsupported IIDs (Unregistered, ISTdMarshalInfo,
and IExternalConnection) which are all interfaces I should not need
implemented here.

Any idea what the problem is?

Sanjay


From jkraska at san.rr.com  Sun Feb 24 15:10:48 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 24 Feb 2002 20:10:48 GMT
Subject: Silly link question: python22_d.dll
Message-ID: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com>

Okay,

I have this problem I've been getting around the hard way.
In VC6.0, when I have a program which talks to Python C/API
functions, and my program is compiled in Debug configuration,
it expects to find the python22_d.dll. I don't always want
or need this, however, and was wondering how's it's possible
to just accept python22.dll as part of my build?

I know how to build Python from source, and that's how I do
it now, but that's a pain sometimes.

Furthermore, why the heck doesn't the binary release include
python22_d.dll???

C//



From josh at nowhere.com  Mon Feb 25 11:42:37 2002
From: josh at nowhere.com (Joshua Marshall)
Date: 25 Feb 2002 16:42:37 GMT
Subject: The good "import" convention?
References: <mailman.1014613140.15550.python-list@python.org>
Message-ID: <a5dpht$i32$1@news.mathworks.com>

Jason Orendorff <jason at jorendorff.com> wrote:

> A single set of imports, at the top of each file, is better,
> because that's the convention.  "Prefer the standard to the
> offbeat."

Convention is useful in decision-making as a last resort.

Generally, where I import depends on a few things.  If only one or two functions
in my module use the functionality of the imported module, then I prefer to have
the imports inside my function definitions.

Having the import statements inside of function definitions also avoids some
circularity issues, if you have interdependent modules.  (Let's not turn that
into a design flamewar.  Sometimes it's useful to have circularly-dependent
modules.)

Otherwise, if the imported module is used heavily, I usually put the import
statement at the top of my module.


From mdk at mdk.com  Fri Feb  1 09:26:32 2002
From: mdk at mdk.com (MDK)
Date: Fri, 1 Feb 2002 09:26:32 -0500
Subject: Question regarding win32comport_demo.py
References: <a3bsm7$16jf7i$1@ID-98166.news.dfncis.de> <3C59CC8E.8AC2E485@engcorp.com>
Message-ID: <a3e8ip$17ckca$1@ID-98166.news.dfncis.de>

Actually, yes, that does help.

The reason behind this is I am trying to uderstand where I am getting the
data from as the example uses a continusly looping thread to get the data
which, in light of what you are saying, is not necessary.  Once the
CreateFile() is done then I can just access the memory buffer.

Thanks.

"Peter Hansen" <peter at engcorp.com> wrote in message
news:3C59CC8E.8AC2E485 at engcorp.com...
> MDK wrote:
> >
> > Does this program read from the Serial Port's Memory Buffer, the UART
> > Buffer, or something else?
>
> It does a ReadFile() in only one place, where it uses the handle which
> was returned from a previous call to CreateFile().  The way serial
> i/o works on most machines is that the UART generates a hardware
> interrupt when new data is received.  That interrupt triggers execution
> of an interrupt service routine (ISR), usually (and in the case of Win32)
> provided by the OS or in a driver.  The ISR adds the data into a
> memory buffer the size of which might be controllable via an API call
> (and again, in the case of Win32, I believe it is).  The file object
> which wraps the serial port is connected to this memory buffer so
> that reads come from there.
>
> Does that help?  You didn't describe the reason behind your question,
> which might be more helpful in soliciting a useful reply.  Why
> do you think you need to know this type of detail?
>
> -Peter




From robin at jessikat.fsnet.co.uk  Sat Feb  2 05:06:26 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 2 Feb 2002 10:06:26 +0000
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
 <313g3a.d22.ln@wombat.san-francisco.ca.us>
Message-ID: <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk>

In article <313g3a.d22.ln at wombat.san-francisco.ca.us>, Keith Keller
<kkeller at speakeasy.net> writes
>In article <LwK68.7172$Xk4.381742 at news1.east.cox.net>,
>       "Andrew Replogle" <replogle2 at cox.net> writes:
>>    Anyone know of any books out there for python system administration or
>> should I just stick to perl for these tasks?
>
>Way to start a holy war.  :)
>
>Did you have something specific in mind?
>
>--keith
>
well if I could get the win32com extensions to change passwords on
different machines for me I would really like that
-- 
Robin Becker


From sdm7g at mac.com  Thu Feb  7 14:09:58 2002
From: sdm7g at mac.com (Steven Majewski)
Date: Thu, 7 Feb 2002 14:09:58 -0500
Subject: How did TimBL's speech at Python 10 go yesterday?
In-Reply-To: <HGx88.95599$KM2.3475019@atlpnn01.usenetserver.com>
Message-ID: <4747DACE-1BFE-11D6-9BC0-003065A7CFBA@mac.com>

On Thursday, February 7, 2002, at 11:11  AM, Steve Holden wrote:

> His slides are available at
>
>     http://www.w3.org/2002/Talks/0206-python

Thanks, Steve -- I didn't write down the URL and I was having trouble 
finding this.


> I especially liked the AaronSwartz link on the last slide. :-)
>

I believe it was supposed to be:

<http://logicerror.com/webizingPython>

"logicerror.com" not "logerror.com"

(I found that while unsuccessfully googling for the URL you gave above!)

BTW: Do you happen to know the link given at the end of the Student Video 
?
( Or the name of the high school ? )
I think it was on opencontent.org but I couldn't find anything there.

-- Steve Majewski
(Who was standing next to you at the video, but didn't get a chance to say 
"Hi"
  before you left.)





From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 09:05:15 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 08:05:15 -0600
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4>
Message-ID: <002801c1b23b$fc793120$0101010a@local>

----- Original Message ----- 
From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>


> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
> news:mailman.1013316023.10782.python-list at python.org: 
> 
> > "PERL type CGI's" should be read "scripts".  You can't run Python,
> > Perl, Ruby,
> > Lua, etc. from this particular server.
> > 
> > Try Xitami.  It is quite compatible with Python.  http://xitami.com
> > 
> Thanx. I'm downloading it as we speak. 

I have used it a little; Steve Holden wrote about it in his book which
I just finished reading.
 
> So with this server, will I be able to just send info direct from my 
> browser form to a file? If I can do that, then I'm up and running.

I'm not sure what you are saying here.  You don't send info "to a 
file," you send it to a CGI script.  Your script must handle the data 
from there.

> [snip]
> > 
> > The form looks good, but where is script.cgi?
> > 
> In the same directory as the html form, cgi-bin.

Ah... what I meant was, that you should *post* the source to 
script.cgi.  Anyway, if your form is HTML it is bad form to have it in 
the cgi-bin folder.

GENERALLY (though it's not real safe to say this) CGI scripts can be 
stored in the cgi-bin folder, or in the htdocs/html/webroot folder; 
this depends on webserver settings.  The Xitami default allows CGI 
scripts only in the cgi-bin folder, but you can define that cgi-bin is 
actually the same as webroot.  Xitami recognizes executable files by 
extension on Windows, so your script should likely be called script.py 
(assuming it is Python).

Gah.  So.  Put script.py in the Xitami cgi-bin folder, and your form in 
the Xitami webpages folder, and if you are using the default 
configuration, you should be in good shape.  Start Xitami and then surf 
to http://localhost/yourform.html (whatever you called it) and when you 
click the Submit button, the form data is sent to 
http://localhost/cgi-bin/script.py (which should be in the action= 
attribute of the form tag, perhaps abbreviated as /cgi-bin/script.py).  
The script receives the data, does whatever, and then must send a new 
page to the web browser... this is why I want to see your script.






From john_coppola_r_s at yahoo.com  Wed Feb 13 14:16:13 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Wed, 13 Feb 2002 11:16:13 -0800 (PST)
Subject: __slots__
Message-ID: <20020213191613.48345.qmail@web11801.mail.yahoo.com>

Hello Python developers!  I have been reading about
Python 2.2's new features and I have to say there's
lots of good stuff going on. The feature I dedicate
this message to is slots.  This ubiquitous new feature
has the potential to boost performance of python quite
considerable and it is the first time python has ever
hinted at, for lack of a better word, "being static".

I used this word loosely.  What I mean by this is that
typically one could set attributes arbitrarily on
python objects, but now cometh slots, if __slots__ is
defined, one could set only attribute names defined in
slots if __slots__ exists.

In introductory essay on the new python object system,
"Unifying types and classes in 2_2", I found Guido's
answer for the existence of __slots__ unsatisfactory. 
As implied from the introductory essay, __slots__ was
used as a means to eliminate a double dictionary for
the dictionary subclass in the
example.  That is one application of slots, but I
believe there is a more powerful way to use this new
feature.

First let me discuss some notational issues I have
with __slots__.  It is possible to use any object that
supports iteration as the __slots__ object. I feel
that this is not a good idea.  Consider the following
example:

    class Foo(object):
        __slots__=['a','b','c']

    foo=Foo()

Bad, but you could do it.

    foo.__class__.__slots__.append('d')

Suppose someone wrote code to initialize some instance
like so:

    for attr in foo.__class__.__slots__:
        setattr(foo,attr,0)

Granted this is decidedly bad code but the problem
this creates is that attribute "d" does not exist in
foo or any instance of Foo.  So an attribute
error will be raised. Slot attributes are read only
thus del foo.__slots__ is not possible.  So neither
should __slots__.append be legal.  So the answer
is to enforce __slots__ is tuple, or for a single
attribute __slots__ is string, when generating code
objects.

Another issue I have with __slots__ is that there
exists better notation to do the same thing.  Why not
introduce new a keyword into the Python language so
that "static attribute sets" are created like so:

    class Foo(object):
        attribute a
        attribute b
        attribute c

Perhaps this syntax could be expanded further to
include strict typing:

    class Foo(object):
        attribute a is str
        attribute b is float
        attribute c is int

I feel that this notation is much more lucid and could
lead to even more interesting notation:

    class Foo(object):
	attribute a is str
	attribute b is float
	attribute c is property
	attribute g is property of float

I hope that it is understood from this context that
"is" compares the type(<attrname>) to trailing type in
the declaration.  The meaning of the modifier "of" in
the last line is a bit ambiguous at this moment. 
Perhaps it enforces the type passed into fset, and the
return value of fget.


We have for the first time static object attributes in
Python that prior to 2.2 never existed.  Is the python
interpreter  __slot__ aware?  Instead of the usual
variable name lookup into __dict__ (which is not
supported by instances who's class implements
__slots__), a slot attribute could be directly
referenced via pointer indirection.   Does the
interpreter substitute references for slot
attributes?  If my question is not clear let me
further elaborate:

    foo.a = "Some value"
    foo.b = 10.0
    foo.c = "spam"

is really   (in pseudo-code)

    Py_XDECREF(self -> some_slot_offset + 0)
    (foo -> some_slot_offset + 0) = <"Some value">
    Py_XDECREF(self -> some_slot_offset + 1)
    (foo -> some_slot_offset + 1) = <10.0>
    Py_XDECREF(self -> some_slot_offset + 2)
    (foo -> some_slot_offset + 2) = <???>
    (don't know how to deal with properties yet)

Does the interpreter substitute "references" instead
of a naming lookups in the case that instance foo
supports slots?  For that matter any namespace
could utilize this new approach, even the local
namespace of a new code block. Slots could be built
implicitly for a local namespace.  This could lead to
significant performance boost where typically, speed
is order O(N), slots comparatively could allow order
O(N) divided by 10.


Please send your comments.  If anything is ambiguous
(and i'm sure there's plenty), let me know and I will
try to clarify what I am talking about.  If anyone
feels that this could be a PEP candidate that would be
cool.

Sincerely,
John Coppola
john_coppola_r_s at yahoo.com



__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com



From usenet at thinkspot.net  Tue Feb 26 16:44:33 2002
From: usenet at thinkspot.net (Sheila King)
Date: Tue, 26 Feb 2002 13:44:33 -0800
Subject: what I would like to see in python to make a better "glue" language (Newbie)
References: <899f842.0202261325.6ff3dd54@posting.google.com>
Message-ID: <a5g3g1.3vsg8p3.1@kserver.org>

On 26 Feb 2002 13:25:15 -0800, anthony_barker at hotmail.com (Anthony_Barker)
wrote in comp.lang.python in article
<899f842.0202261325.6ff3dd54 at posting.google.com>:

> Important:
> 2. DNS really should be included in the base library, I emailed
> Anthony Baxter and he replied, saying it was almost done. It is a
> hassle to parse the dig (nslookup) output as I ended up doing, as it
> is not portable.

I have recently had to do the same thing (parsing an nslookup command) in
order to write a script. I would REALLY appreciate a DNS module as part of
the standard library. The socket library just didn't quite do everything I
needed.

I'm curious about the reference to Anthony Baxter and a possible DNS module
for including in the standard library. More info? Comments?

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From emile at fenx.com  Wed Feb  6 09:34:50 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 06:34:50 -0800
Subject: CSV revisited
References: <slrna62f14.pef.mlh@vier.idi.ntnu.no>
Message-ID: <a3rf88$1ag9qf$1@ID-11957.news.dfncis.de>

"Magnus Lie Hetland" <mlh at vier.idi.ntnu.no> wrote in message
news:slrna62f14.pef.mlh at vier.idi.ntnu.no...
> A little while ago I started a thread about adding CSV support to the
> standard library. There was (as expected) some controversy about this,
> but I think the idea had enough support that it might be turned into a
> formal proposal... However, I'm not sure wheter this is the sort of
> thing that should be formalised as a PEP... Any opinions in the
> matter? (Not pro/con the proposal itself... ;)
>

It seems to me that if a module is written that is used and properly
tested to read the varied CSV formats in use, that it could then be
submitted for inclusion in the standard library.  This shouldn't require
a PEP I think.

Of course, from personal experience and after the last thread, it also
seems to me that CVS is as much a standard as RS232 was.  ;-)


--

Emile van Sebille
emile at fenx.com

---------



From karthik at james.hut.fi  Wed Feb 27 16:04:07 2002
From: karthik at james.hut.fi (Karthikesh Raju)
Date: 27 Feb 2002 23:04:07 +0200
Subject: Numeric Python Installation woes
Message-ID: <m3elj6ocvs.fsf@mycomp.tky.hut.fi>

Hi All,

Tried installing Numeric Python from the sources as:

>> python setup.py install

error was:
  File "setup.py", line 170
    print >>rpm_script, rpm_install_text
           ^
SyntaxError: invalid syntax

When i tried 

>> python2 setup.py install 

got this:
running install
Traceback (most recent call last):
  File "setup.py", line 187, in ?
    ext_modules = ext_modules
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/core.py", line 138, in setup
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/dist.py", line 893, in run_commands
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/dist.py", line 912, in run_command  File "/usr/lib/python2.2/cmd.py", line 112, in ensure_finalized
    
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/command/install.py", line 268, in finalize_options
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/sysconfig.py", line 408, in get_config_vars
  File "/var/tmp/python2-2.2-root/usr/lib/python2.2/distutils/sysconfig.py", line 313, in _init_posix
distutils.errors.DistutilsPlatformError: invalid Python installation: unable to open /usr/lib/python2.2/config/Makefile (No such file or directory)


No clues, Am i missing something. 

i have both python 1.5 and python 2.2 and apparently 2.2 version does
not respond to the upward arrow key for the previous command. Seems
like i have completely played around with  something major.

Any help would be great,

Thankx in advance
karthik


From mcfletch at geocities.com  Tue Feb 12 03:21:07 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Tue, 12 Feb 2002 03:21:07 -0500
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <mailman.1013255897.29941.python-list@python.org> <VkV98.2452$yV3.182935@newsread2.prod.itd.earthlink.net>
Message-ID: <3C68D073.3080609@geocities.com>

Very nice module.  I just did a re-write of my watcher classes using 
dispatcher.  Why did I keep the watcher pattern?  Well, it makes 
creating pre-event processing possible (my one class needs to do some 
processing to determine conflicts, and until 2.2 object-based-classes 
(with properties) can live in ZODB, I'm stuck with that).

Would be nice if there was a simple download somewhere (had to re-format 
the code after Mozilla mangled it with cut-and-paste).  Should put an 
explicit license in there (as ActiveState's legal policy says that the 
license included with the contribution applies.

Also looked at the References.py module in anygui.  Didn't spend much 
time on it, as dispatcher is much closer to what I need.  Was one of 
these modules based on the other (your comments on the ASPN site seem to 
suggest there's some relationship)?

Good job all around. Thanks,
Mike


Patrick K. O'Brien wrote:
> You might find this useful:
> 
> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/87056
> 
> This module, dispatcher.py, provides global signal dispatching services
> suitable for a wide variety of purposes, similar to Model-View-Controller or
> Model-View-Presenter patterns. This particular implementation allows a
> looser coupling than most Observer patterns. It also does transparent
> cleanup through the use of weak references and weak reference callbacks.
> This version defaults to using weak references, but provides an option to
> not use weak references for those cases where weak references are
> problematic (lambdas and such).
> 
> --
> Patrick K. O'Brien
> Orbtech
...




From jhenry19 at ford.com  Wed Feb 13 12:22:08 2002
From: jhenry19 at ford.com (Jeff Henry)
Date: Wed, 13 Feb 2002 12:22:08 -0500
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com> <7x1yfq2gdf.fsf@ruckus.brouhaha.com> <c76ff6fc.0202130114.6c1ab5ae@posting.google.com>
Message-ID: <3C6AA0C0.CA9FFAA5@ford.com>

John Machin wrote:
> 
[snip]
> pending bag ... using a regex backwards to find the opening tag is so
> innovative that I'm totally gobsmacked.
> 
I love this group! Yesterday I learned "wackyparse", and today
"gobsmack"!

[snip]
> 
> Looks like the effbot gets to be gobsmacked too.
> 
Ah, but will he be wackyparsed? Inquiring minds want to know! :)


From jason at jorendorff.com  Wed Feb 27 17:10:12 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 27 Feb 2002 16:10:12 -0600
Subject: recursive problem
In-Reply-To: <292c8da4.0202271007.1b3af9ab@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAEMPDFAA.jason@jorendorff.com>

> I've been looking at the "mystery M function" question [...]
> What I'm doing here is storing in a Dictionary the result of m
> where I get one, so I can use it later. As I said it still blows.
> Any clues for how to improve my algorithm appriciated.

The technique you're using is called memoization, and it's a
smart thing to try.

My advice:  don't count on Python to get the answer.  Try it
for yourself with pencil and paper.  Remember to memoize the
results yourself.  But don't start with m(4, 4, 4).  Start
smaller.

## Jason Orendorff    http://www.jorendorff.com/



From jeff at ccvcorp.com  Fri Feb  8 15:10:15 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 12:10:15 -0800
Subject: What's the equivalent in Python for Perl's CPAN?
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com> <a3v903$kgk$1@panix2.panix.com> <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>
Message-ID: <3C6430A6.4DD9316E@ccvcorp.com>

Mike Carifio wrote:

> The closest thing I seem to be able to find is ActiveState's Package Manager
> (http://www.activestate.com/Products/ActivePython/more_information.plex).
> Is this what I should use or is something else preferred?
>
> So, what's the equivalent in Python for Perl's CPAN?

Well, if you'd tried searching Google's archives of this newsgroup, you'd have
doubtless found the answer, as it's been discussed many times.  (You'd probably
also have gotten better responses to a question phrased as "is there?" instead
of "what is?", especially if your message had actually *had* a body (even if it
was simply a repeat of the subject line), but...  ;) )

The answer is that there *is* no direct equivalent.  Yet.  There's a project
underway (Siphon... or was it Ciphon?) that has more-or-less the same
functionality as CPAN, but it is, IIRC, still in testing, and has a very limited
selection of modules uploaded.

Of course, the other problem with duplicating CPAN, is in finding sites to host
that network of mirrors....

Jeff Shannon
Technician/Programmer
Credit International





From marklists at mceahern.com  Tue Feb  5 15:48:49 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 5 Feb 2002 12:48:49 -0800
Subject: What is the command to determine an object's type?
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBGEADKBAA.marklists@mceahern.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBEEAEKBAA.marklists@mceahern.com>

> MDK wrote:
> How can I tell if an object is a string, a list, or a dictionary in a
> program using if and ifelse so that I can perform a differnent action
> depending on the object's type?

I posted code that will do the trick and then realized, I should probably
add this...

It goes without saying, of course, that you may want to investigate the
Strategy design pattern.  Eliminating that sort of conditional type checking
is Strategy's strong suit.

Cheers,

// mark




From jason at jorendorff.com  Sun Feb 17 14:51:50 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 17 Feb 2002 13:51:50 -0600
Subject: Redefining "\n"
In-Reply-To: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEEEDDAA.jason@jorendorff.com>

Dale Strickland-Clark wrote:
> Is there some way to redefine the "\n" escape sequence?
> 
> When preparing strings for the Windows clipboard, I need \x0d\x0a for
> a newline which is a bit cumbersome to insert all over the place.
> 
> It would be neat if I could make "\n" insert some other sequence.
> 
> I know I can do a translation before sending the string to the
> clipboard but this would be nicer.

I know what you're asking... but I can't help thinking that maybe
what you really want is:

def clip(s):
    s = s.replace('\n', '\r\n')
    import win32clipboard as cb
    cb.OpenClipboard()
    cb.EmptyClipboard()
    cb.SetClipboardData(cb.CF_TEXT, s)
    cb.CloseClipboard()

Calling into Win32 is the cumbersomeness-determining step.

## Jason Orendorff    http://www.jorendorff.com/



From brian at sweetapp.com  Sun Feb  3 14:24:42 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sun, 3 Feb 2002 11:24:42 -0800
Subject: SilverCity 0.1.0
In-Reply-To: <hB978.33300$Ni2.188457@news-server.bigpond.net.au>
Message-ID: <002101c1ace8$6e86c240$445d4540@Dell2>

Neil Hodgson wrote:
> Not sure which level you are interested in. 

Whatever level you are willing to provide :-) Really, I am trying to
make sure that my API does not make it difficult to retrieve useful
information.

> The _SilverCity interface
> looks straightforward. 

Yes, it's pretty simple. I'll probably write a test suite in the next
few days that will exercise/demonstrate all of its features. The only
feature that is not demonstrated, I believe, is that PropertySets act
like mapping types e.g. propset['tab.timmy.wingy.level'] = 3

I have a Scintilla features request though:
- expose the name of the lexer in the LexerModule class (it'd like to
include the lexer's name in __repr__ and exceptions) 

> There could be a script for generating language
> modules or outlines of language modules from the constants:
generate("CPP",
> "SCE_C_*"). 

Great idea! I don't think that the event handler names can be exactly
what I want (I like to segment identifiers by name i.e. SCE_C_COMMENTDOC
=> handle_c_comment_doc) but I think that saving me all the typing is
well worth it!

What did you think about the higher-level API? i.e. lexing to events,
the HTML generators, etc? 

> BTW, the verbatim lexical class that got the "XXX what is
> this?" comment is for C#'s multiline verbatim strings: 
> @"Just like ""Python"" uses triple quotes for long texts."

Cool, thanks. 

Thanks a lot for the feedback, Neil!

Cheers,
Brian




From fperez528 at yahoo.com  Tue Feb 19 16:32:17 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 19 Feb 2002 21:32:17 +0000
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu>
Message-ID: <a4v902$5pb$1@peabody.colorado.edu>

Rajarshi Guha wrote:

> Hi,
> is?it?possible?to?have?a?variable?contain?a?function?name,?and?use?the
> variable to call the function?
> An example:
> 
> def func:
> some?code
> 
> funcvar = 'func'
> 
> Now use funcvar (somehow!) to call func
> 
Just use

funcvar = func

(no quotes) and you're done. In python, 'everything is a name' (which points 
to an object). func is just a name pointing to a function object, and the 
assignment makes funcvar be a second name pointing to the same thing. Think 
of it as C pointers which do all referencing/dereferencing/memory management 
magically for you.

cheers,

f


From brown.2053 at osu.edu  Mon Feb 25 19:41:47 2002
From: brown.2053 at osu.edu (Locke)
Date: Mon, 25 Feb 2002 19:41:47 -0500
Subject: size of socket.recv buffer
Message-ID: <a5elgo$5bu$1@news.cis.ohio-state.edu>

The argument to recv is supposed to be the buffer, acording to the
documentation. But the doc doesnt say: buffer measured in what? Is it bytes?
Octets? (by the way, what is the diff between a byte and an octet?)

See, I am writing a pop3 client type thing, after requesting a message from
the server, my first s.recv tells me how long the message will be in octets.
So then I call s.recv again passing the number of octets as the argument to
s.recv. Shouldn't this download the entire message? What actually happens is
that it cuts off right in the middle of the message.

What's my problem? Is there an upper limit on what the buffer argument can
be to recv? Or does pop3 just refuse to send the whole message with one call
to recv? I don't think it is that, because what it sent me was like 600some
characters, which isnt a power of two or any other computer-ish number.

Thanks




From lac at strakt.com  Tue Feb  5 10:06:25 2002
From: lac at strakt.com (Laura Creighton)
Date: Tue, 05 Feb 2002 16:06:25 +0100
Subject: Inherit Class Dynamicly 
In-Reply-To: Message from Xiao-Qin Xia <xx758@cam.ac.uk> 
   of "Tue, 05 Feb 2002 12:49:56 GMT." <a3ok79$bvr$1@pegasus.csx.cam.ac.uk> 
References: <a3ok79$bvr$1@pegasus.csx.cam.ac.uk> 
Message-ID: <200202051506.g15F6QGo013953@ratthing-b246.strakt.com>

> Hi, everybody,
> 
> from Tkinter import *
> class NewWidget:
>         ...
>         def __init__(self, widget_name = "Button", **dict):
>                 # What is needed here to make the new born object(self) 
>                 # behavior like both the class specified by widget_name
>                 # and NewWidget?
> 
> Thanks from
> 
> Xiao-Qin
> 

I do it like this:

import Tkinter
class MyButton(Tkinter.Button): # single
    """
    This is a simple wrapper for the Tkinter Button, which just makes
    an alias, 'data' for 'text' in the button.  
    """
    def __init__(self, parent, **kw):
        """
        Function name: __init__
        Description:   Initialise MyButton.
        Arguments:     parent - the parent frame,
                       **kw   - a dictionary containing all the option=value
                       pairs as {opt1:val1, opt2:val2 ...}  If you put
                       things in here which are not valid options to a
                       Tkinter.Button you must remember to del them
                       from the dictionary before you call
                       Tkinter.Button.__init_
        """
        self.parent = parent
        if kw.has_key('data'):
            kw['text'] = kw['data']
            del kw['data']
     
        Tkinter.Button.__init__(self, parent, **kw)
	
        #This assumes you have 2.0 or better.  If you have 1.5.2
        #apply(Tkinter.Button.__init__, (self, parent), kw)
     
    # go add some more methods here as part of MyButton.  Here is a
    # good one for me to add since I just made an alias.

    def cget(self, property):
        """
        Function name: cget
        Description:   Add return values for the additional attribute
                       data in case somebody wants to find it this way.
        Arguments:     None
        Returns:       the data that is contained in the widget
        """
        if property == 'data':
            return(Tkinter.Button.cget(self, 'text'))
        else:
            return(Tkinter.Button.cget(self, property))

###########################################################
if __name__ == '__main__':
    root = Tkinter.Tk()
    w = MyButton(root, width= 30, fg='green', 
        text='This is a very green button')
    widgetList = (
        ['show', lambda x = w: x.grid(row=0, col=0, columnspan=5)],
        ['pink', lambda x = w: x.configure(fg='pink')],
        ['yellow', lambda x = w: x.configure(bg='yellow')],
        ['Exit', root.destroy],
        )
    column = 0
    for txt, cmd in widgetList:
        button = Tkinter.Button(root, text=txt, command=cmd)
        button.grid(row=1, col=column, sticky='w')
        column += 1
    root.mainloop()

----------------------------------------------------
 If you want multiple inheritance, that is fine too, i.e.

 class MyButton(MyWidgetMixin, Tkinter.Button): 

   but somewhere in the __init__ you need to do this
   MyWidgetMixin.__init__(self, whatever_you_want)

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

Laura Creighton



From azndizzy at asia.com  Tue Feb  5 21:21:25 2002
From: azndizzy at asia.com (David Pham)
Date: 5 Feb 2002 18:21:25 -0800
Subject: To all Higher Level....A True Beginner
Message-ID: <ba59f11e.0202051821.5c41221e@posting.google.com>

Hey! Everyone!, I like to know if any of you programmres recommend any
books, online tutourial for myself, or anyone else that is a beginner.
I am a beginner programmre. Chose Python cause heard it can do a lot
of things. But never have programme before. So I home you guys can put
out some help. The only language I have tried out is probably Qbasic
and VBdos thats all, but nothing came out.


From jkraska at san.rr.com  Wed Feb  6 19:39:35 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 00:39:35 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <mailman.1013007554.26921.python-list@python.org> <slrna62o1d.b33.philh@comuno.freeserve.co.uk>
Message-ID: <qjj36usrp60jbtjcvfh817p1br8sd7j0bn@4ax.com>

>Any programming system that takes this approach is one that I simply
>am not interested in learning to use.

I find it strange that you'd be programming in Python. Minimalism
is the One True of Python. Consider the paucity of forms. BDFLisms
abound. Not that I'm complaining (well, except for generators, ha).

C//



From bokr at oz.net  Thu Feb 28 18:46:47 2002
From: bokr at oz.net (Bengt Richter)
Date: 28 Feb 2002 23:46:47 GMT
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr> <a5iuq9$7701i$1@ID-11957.news.dfncis.de> <a5j53i$oms$1@wanadoo.fr> <8z9f2rks.fsf@fido.trinitycapital.com> <a5j7fb$fb$1@wanadoo.fr> <a5k352$etg$0@216.39.172.122> <a5lu00$29p$1@wanadoo.fr>
Message-ID: <a5mfh7$9fa$0@216.39.172.122>

On Thu, 28 Feb 2002 19:51:06 +0100, "Riss Nicolas" <alien275 at libertysurf.fr> wrote:

>Has somebody a python program where it is an telnet connection ?

You appear to be running windows. What happens if you get into your browser
and put

    telnet://192.168.0.9:32

in the location (i.e., what you are trying to connect to with your code)?
Do you get a telnet client screen with a login prompt?

Did you try \r\n in place of plain \n in your code?

Regards,
Bengt Richter



From deathtospam43423 at altavista.com  Sat Feb  9 20:26:29 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 10 Feb 2002 01:26:29 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4>
Message-ID: <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4>

Philip Swartzleonard <starx at pacbell.net> wrote in
news:Xns91AFAD9B27944RASXnewsDFE1 at 130.133.1.4: 

> Well, as far as i can remeber, CGI is just an interface, you can use it
> with Perl or Python or C or whatever. You probably will have to learn
> it, but it should just be a modlue and some extra commands IN python...
> and you'll have to configure whatever web server you're using to use
> CGI program, that's probaly the hard part.
> 
Done. I downloaded a personal web server with CGI capabilities.

[snip]

> Anyway, you'd only need unix sometimes if you used that kind of server,
> and you don't even want a server... if you can get your computer to run
> CGI's, and check out 11.2 (or so) of the library reference manual for
> the CGI commands, you should be most of the way there.
> 
> Sorry for the longness :)
> 
Thanx for the info. 
I have been looking through 11.2. It seems to mainly deal with how to 
process the information once you've already got it from the web browser. 
For instance this example from the manual ...

form = cgi.FieldStorage()
if not (form.has_key("name") and form.has_key("addr")):
    print "<H1>Error</H1>"
    print "Please fill in the name and addr fields."
    return
print "<p>name:", form["name"].value
print "<p>addr:", form["addr"].value

... the first line appears (correct me if I'm wrong) to take what ever data 
has been inputed to the cgi and store it in a variable called "form".

But I haven't managed to do that yet. :( Everytime I go to my servers cgi-
bin with my browser, and enter information on a form, I can't find the 
information anywhere.

Basically all I want to be able to do at this stage is take information 
from a form on a web browser, and place it in a file. What's the easiest 
way to do this? 

There are html tutorials on how to process forms ... but like I say, they 
all seem to want you to learn PERL and get a UNIX server. 
-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From martin at v.loewis.de  Mon Feb  4 01:53:37 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 04 Feb 2002 07:53:37 +0100
Subject: pickle security
References: <mailman.1012591743.10841.python-list@python.org> <7xit9g51lz.fsf@ruckus.brouhaha.com> <j4g04ihc7k.fsf@informatik.hu-berlin.de> <7xsn8im3uh.fsf@ruckus.brouhaha.com>
Message-ID: <m3k7ttkaa6.fsf@mira.informatik.hu-berlin.de>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> The code to deal with quote strings calls eval, and without actually
> examining the eval implementation (big, complicated) there's no way to
> confirm that it can't somehow run arbitrary code found in the string.

Then, please do examine the code. Not doing so and still claiming
there is a problem means to spread FUD.

> Microsoft does that and that's why we keep hearing about new
> exploits being discovered in Microsoft products.  The code has to be
> checked thoroughly, which is easier to do when the code is simple.

In this specific case, it is relatively easy to check that the eval is
safe. It is not the case that the risk of an exploit has been
ignored. Instead, it is the case that the risk has been analysed and
found to be non-existing. If you chose to ignore this analysis, fine -
but please don't tell people that there is a security risk.

In any case, there is a patch on SF that removes the eval, see

http://sourceforge.net/tracker/?func=detail&aid=505705&group_id=5470&atid=305470

The rationale for doing so is different, though: eval interns strings,
which is undesirable for unpickling.

Regards,
Martin


From jeff at ccvcorp.com  Thu Feb 28 14:52:42 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 11:52:42 -0800
Subject: Extremely Newbie wants to program and pleads to be pointed in the 
 right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
	 <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>
	 <54abe19d.0202271315.4e699ed0@posting.google.com> <b7yVwiC1Cff8EwBv@[127.0.0.1]>
Message-ID: <3C7E8A8A.6B16887C@ccvcorp.com>


Chris wrote:

> Can you do ordinary 2D graphics stuff without VPython - simply using the
> ordinary Python?

Sort of, depending on just what sort of graphics you want, and how
complicated you want to make them.

Core Python doesn't include any direct graphics support, because it's
intended to run on a wide variety of different computers, which handle
graphics in totally different ways (and in some cases, have *no* screen on
which to draw graphics!).  So no matter what, you'll need to use *some*
toolkit for your graphics.  The standard Python distribution *does* ship with
one such toolkit -- it's called Tkinter.  It (like many of the toolkits
available) is intended mostly for creating GUIs (graphical user interfaces --
think "windows programs" instead of "Console (DOS-box) programs"), but it
does include plain-old-drawing support.  (Look especially at the Canvas
widget).  Tkinter isn't the most powerful or flexible toolkit around, but
it's okay, and seems to fairly easy to pick up.

However, just because of the nature of cross-platform graphics and GUIs,
doing graphics is not necessarily an easy place to begin.  The Basic that
you're used to, made simple graphics easy by expecting to only ever be run on
that single platform; Python tries to be much more flexible and widely
applicable, but this does make this one particular aspect a bit harder to get
started in.  (But if you're trying to do anything *more* than toy examples,
Python is *much* easier!)  So don't necessarily rush straight for the
graphics toolkits -- spend a bit of time playing with text-only programs.
You can still do quite a bit, and have a lot of fun, without ever touching
graphics at all.

If you want to see some code examples, project suggestions, etc, you might
want to look at Useless Python (www.lowerstandard.com/python).  There's also
a Python Tutor mailing list
(http://www.python.org/psa/MailingLists.html#tutor), where there's lots of
people who will answer any questions you might have, at *any* level of
ability.

Enjoy!

Jeff Shannon
Technician/Programmer
Credit International




From ivan at kupol.ru  Wed Feb 20 00:08:52 2002
From: ivan at kupol.ru (Ivan Bykov)
Date: Wed, 20 Feb 2002 09:08:52 +0400
Subject: socket in python 2.2 vs 2.1
Message-ID: <514095158.20020220090852@kupol.ru>

Hello!

This code working in Python 2.1 and not working in Python 2.2.
Testing under Windows 98 and 2000.
This code want to send UDP packets for all.
===
PythonWin 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2001 Mark Hammond (mhammond at skippinet.com.au) - see 'Help/About PythonWin' for further copyright information.
>>> import socket
>>> tst=socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
>>> tst.connect(('255.255.255.255',5555))
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "<string>", line 1, in connect
gaierror: (7, 'getaddrinfo failed')
>>>
===
It`s bug?

-- 
Best regards,
 Ivan Bykov                         mailto:ivan at kupol.ru





From emile at fenx.com  Mon Feb 25 06:57:47 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 03:57:47 -0800
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com>
Message-ID: <a5d95n$6g35h$1@ID-11957.news.dfncis.de>

"Janez Jere" <janez.jere at void.si> wrote in message
news:qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367 at 4ax.com...
> Hi, I am experimenting with IPC, my current implementation is very
> slow, so I am asking for some hints, how to optimize it.
> Currently I know that cStringIO would cut execution time for 30%. but
> I don't want to drop unicode support
>


Define slow.  I ran some tests on various sizes, etc, and it seems OK.
At a certain size, os.write() appears to be blocked, and never returns.
That's slow... ;-)  Are you hitting something like that?  If not, look
into the profiler to see where your bottleneck is at.

--

Emile van Sebille
emile at fenx.com

---------



From cezary at bigfoot.com  Wed Feb 27 01:23:23 2002
From: cezary at bigfoot.com (Cezary Biernacki)
Date: Wed, 27 Feb 2002 07:23:23 +0100
Subject: Precompiled regular expressions slower?
References: <3c7bb531$0$3930$b45e6eb0@senator-bedfellow.mit.edu> <slrna7nm7u.11s.huaiyu@gauss.almadan.ibm.com>
Message-ID: <3C7C7B5B.5020009@bigfoot.com>

Huaiyu Zhu wrote:

> On Tue, 26 Feb 2002 11:17:53 -0500, Peter Bienstman <pbienst at mit.edu> wrote:
> 
> Yes.  Use re1.search(line) instead.  
> 
> The times of the following three tests are:
> 1.94322395325   # search(string, string)
> 2.62431204319   # search(re_obj, string)
> 0.667925000191  # re_obj.search(string)
> 
> Someone care to explain why the second ends up slower than the first?

Look at _compile in sre.py (the standard library):

def _compile(*key):
     # internal: compile pattern
     p = _cache.get(key)
     if p is not None:
         return p
     pattern, flags = key
     if type(pattern) not in sre_compile.STRING_TYPES:
         return pattern
     try:
         p = sre_compile.compile(pattern, flags)
     except error, v:
         raise error, v # invalid expression
     if len(_cache) >= _MAXCACHE:
         _cache.clear()
     _cache[key] = p
     return p

In first case pattern is found in "_cache" and directly returned.
In second case "if type(pattern) not in sre_compile.STRING_TYPES" must 
be also tested.


--
CB



From sjoshi at ingr.com  Tue Feb 19 09:56:43 2002
From: sjoshi at ingr.com (Sunit Joshi)
Date: 19 Feb 2002 06:56:43 -0800
Subject: Problem with 4Dom xslt processor
Message-ID: <8f8ffe67.0202190656.4fb11dda@posting.google.com>

Hello all
I'm trying to use the 4DOM's xslt processor I keep getting this error
(below) when trying to append a stylesheet. Any pointers are greatly
appreciated.

thanks
Sunit

>>> from xml.xslt.Processor import Processor
>>> proc = Processor()
>>> proc
<xml.xslt.Processor.Processor instance at 007E45D4>
>>> proc.appendStylesheetUri('ships.xsl')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "C:\Python21\_xmlplus\xslt\Processor.py", line 95, in
appendStylesheetUri

    sty = self._styReader.fromUri(styleSheetUri, baseUri)
  File "C:\Python21\_xmlplus\xslt\StylesheetReader.py", line 297, in
fromUri
    ownerDoc, stripElements)
  File "C:\Python21\_xmlplus\xslt\minisupport.py", line 58, in fromUri
    return self.fromStream(stream, baseUri, ownerDoc, stripElements)
  File "C:\Python21\_xmlplus\xslt\StylesheetReader.py", line 304, in
fromStream
    self.initParser()
  File "C:\Python21\_xmlplus\xslt\StylesheetReader.py", line 344, in
initParser
    if self._force8Bit:
AttributeError: StylesheetReader instance has no attribute
'_force8Bit'


From shalehperry at attbi.com  Wed Feb 13 19:02:16 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 13 Feb 2002 16:02:16 -0800 (PST)
Subject: Container Objects Enforcing Contents
In-Reply-To: <ff8171ee.0202131528.1a9aabfe@posting.google.com>
Message-ID: <XFMail.20020213160216.shalehperry@attbi.com>

On 13-Feb-2002 jim kraai wrote:
> Greetings,
> 
> I'm looking for suggestions on making container objects which must,
> may, or must not allow particular kinds of container objects to be
> added.
> 
> Any suggestions on a general approach for this?

Let's see .......

a constraints approach:

class Foo:
    def __init__(self, item):
        try:
            # look for a method of item that must exist
        except AttributeError:
            raise ConstraintException

This way you could add only items which support a method or have a certain
attribute.

You could also try to enforce types by ensuring the type of item matches types
you know about.  This often leads to a maintenance nightmare though.



From loewis at informatik.hu-berlin.de  Wed Feb 27 11:38:24 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Feb 2002 17:38:24 +0100
Subject: python stripping LD_LIBRARY_PATH?
References: <3C7BDD3A.9050905@nmt.edu> <3C7BFA13.EDEFC99F@bioreason.com> <3C7CFED4.9000507@nmt.edu>
Message-ID: <j4g03mkhhb.fsf@informatik.hu-berlin.de>

Jeff Shipman <shippy at nmt.edu> writes:

> What can be done about this? Is there a place in
> the python internals I can add the necessary information
> to allow me to start up? I really must run this wrapper
> suid to another user and it really must connect
> to Oracle.

Call setuid before calling exec in the wrapper.

Regards,
Martin


From python at chajadan.net  Tue Feb 19 03:16:09 2002
From: python at chajadan.net (chajadan)
Date: Tue, 19 Feb 2002 00:16:09 -0800
Subject: help - inserting values into database tables
In-Reply-To: <5.1.0.14.0.20020218215652.00a41ec0@63.118.78.184>
Message-ID: <5.1.0.14.0.20020219001529.00a75070@63.118.78.184>

It seems I have ISDN on the brain, because I changed the correct "DNS" to 
the incorrect "DSN" partway through and never went back ;)

--chajadan

At 09:57 PM 2/18/02 -0800, you wrote:
>Hi Joe =)
>
>Your problem shouldn't be too hard to fix.
>
>Using odbc, you can't open the database file directly like that, since 
>odbc supports so many different database standards, it needs to know which one.
>
>This requires a DNS. In order to set up a DNS for MS Access, just go to 
>your control panel, and you should have an option called "OBDC 
>Datasources" or similar. Click that, and it will show you your current 
>DSNs. There are several types of DSNs, but you will want to make this file 
>a System DSN, so click that tab. Select New, and choose the correct 
>database driver, in this case MS Access. Then you have to name the DSN, 
>anything you'd like (avoid spaces), chose a login/password or leave that 
>blank, and finally show it the database file this whole DSN should be 
>referring to.
>
>Then, you can odbc.odbc("myDsnName")
>
>=)
>
>Let me know if you have any other questions and I'll try to help. I work 
>on Win98, so some of what I described may look slightly different under 
>XP, but it should be there.
>
>Good luck!
>
>--chajadan
>aka charlie
>
>At 06:35 PM 2/18/02 -0600, you wrote:
>
>>Hi again,
>>Just wondering if you can help?
>>
>>I installed the python ver 2.0(from CD)for windowsX on a windows98
>>system.  Then I installed the windows32all stuff.
>>So are the dbi and the odbc modules should now be installed
>>on my system now, or do I need some other package ??????
>>
>>It seems I can import the dbi then the odbc with no errors...
>>But my simple script below won't run; has an error that I
>>don't understand ???????
>>
>>I am running from the command line, heres my environ stuff
>>******* Do I need to put in another path to Lib ?????
>>
>>COMSPEC=C:\WINDOWS\COMMAND.COM
>>windir=C:\WINDOWS
>>PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\JDK11~1.8\BIN;C:\PYTHON20;
>>
>>******** Heres the error I get upon running from prompt
>>
>>C:\codepy>python conntest.py
>>
>>Traceback (most recent call last):
>>   File "conntest.py", line 8, in ?
>>     conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
>>dbi.operation-error: [Microsoft][ODBC Driver Manager] Data source name not
>>found and no default driver specified in LOGIN
>>
>>The Manufacturers.mdb is the ms-access database of ver office XP
>>2002.  It is just a small database with a small table in it.
>>Very simple kind.  The file is in that codepy directory also.
>>So I don't understand ?????????????????
>>What is DSN, does it allow for a ms-access file??????????
>>
>>Well anyways heres my test database connection script, I am
>>just trying to connect and print some output from the database.
>>I think I am doing this correct but not sure ??????????
>>
>>The script:  conntest.py listing
>>#! /usr/bin/env python    #not needed for windowspython
>>
>>#test script to connect to a ms-access database
>>
>>import dbi
>>import odbc
>>
>>conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
>>rs = conn.cursor()
>>rs.execute("select * from ManufName")
>>print rs.description
>>
>>rs.close
>>conn.close
>>
>>Can you shine any light my way ???????????
>>
>>Thank you (working on a windows system today(UGH...)),
>>Joe F.
>>#####
>>
>>----- Original Message -----
>>From: chajadan <python at chajadan.net>
>>Newsgroups: comp.lang.python
>>Sent: Monday, February 18, 2002 2:34 PM
>>Subject: Re: help - inserting values into database tables
>>
>>
>> > I always use PythonWin, part of the win32all distribution. It's a
>> > color-coded ide, with a basic syntax check, and has one nifty advantage
>> > over any other input editor I've tried: when you call a function, it takes
>> > the document strings and actually gives you a little note about what kind
>> > of input it's expecting!
>> >
>> > So all in all it should be the same thing as it ever was, but now in the
>> > environment your particular task seems to ask for. The only "new" things
>> > will be the other possible windows-only features you might want to learn
>> > and use in your time on windows.
>> >
>> > Happy computing!
>> >
>> > --chajadan
>> >
>> > At 08:15 AM 2/18/02 -0600, mylinuxboxroot wrote:
>> > >Hey thanks,
>> > >This is great, and I have the project now that it will be developed on a
>> > >windows
>> > >server so I guess I don't need to worry about going from linux to the
>> > >MS-Access
>> > >database.  I will develop it right on the server using the win32all
>> > >distribution that
>>
>>snip
>
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From marklists at mceahern.com  Sat Feb 16 17:54:09 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sat, 16 Feb 2002 14:54:09 -0800
Subject: "implemented by subclass"
In-Reply-To: <slrna6t9de.nt2.philh@comuno.freeserve.co.uk>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEIJKDAA.marklists@mceahern.com>

[phil hunt]
> Consider the following python code:
> 
> class Log:
>    def write(self, text): 
>       """(implemented by subclass)"""  
> 
> Is this the most pythonic way to do it? If not, what would be
> better? How about rasing a ShouldBeImplementedBySublass exception?
> 
> How do other people do it?

If it MUST be implemented by subclass, I'd raise an exception here.

Otherwise, pass oughta do.

// m



From jason at jorendorff.com  Sat Feb  2 09:52:21 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 2 Feb 2002 08:52:21 -0600
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
In-Reply-To: <a3gu2l$om7$1@panix2.panix.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEOICPAA.jason@jorendorff.com>

Aahz wrote:
> [...] nine times out of ten, if you think you disagree with Guido's
> intuition, you'll change your mind later.  As Alex (I think it was Alex)
> pointed out a few months ago, Guido's ability to do good design work
> vastly outstrips his ability to *explain* his thought processes.
> 
> Therefore, as long as you emphasize "*seemingly* arbitrary", you won't
> get flamed by me.

I agree with all that.  Not to take any credit away from the BDFL,
but I also think that about 4 times out of 10, what's important is not
his intuition, but simply that one (any) of the alternatives be chosen
and endorsed, so we can get on with it.

## Jason Orendorff    http://www.jorendorff.com/



From logiplexsoftware at earthlink.net  Fri Feb  1 14:20:58 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 11:20:58 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <b34d3835.0202010038.6a907bff@posting.google.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<b34d3835.0202010038.6a907bff@posting.google.com>
Message-ID: <20020201112058.4e5700ae.logiplexsoftware@earthlink.net>

On 1 Feb 2002 00:38:10 -0800
Rony wrote:

> Cliff Wells <logiplexsoftware at earthlink.net> wrote in message
news:<mailman.1012498757.26825.python-list at python.org>...
> > However, your last statement that managers make _all_ the decisions was
an
> > unfortunate one.  It's better if the technicians make the decisions and
> > then managers _approve_ them.  This may be splitting hairs, but when it
> > comes to deciding on things like what language to develop in, who is
going
> > to be better informed: the people /using/ that language or someone who
is
> > more focused on the overall picture?  Tool choices are a detail and
best
> > left to those using the tools.  Managers should focus on personnel
> > coordination, feature specifications, overall design, etc - the big
> > picture.  It's when they start trying to specify which
> > editor/debugger/compiler will be used that friction occurs.  
> 
> I'm sorry but i definitifly don't agree with that, and here i'm
> speaking trough personnal experience !

[snip]

> the company the problem  was mine at the end. So i'm convinced that
> managment has to be involved in such decissions.

If you reread my statement, I never suggested that the manager not be
/involved/, rather that he should give adequate weight to the engineer's
opinion.  Clearly the manager still needs to approve such decisions, but if
it is to be denied it needs to be for the right reasons and with
consideration to the future, not just the present.

These arguments may apply somewhat less to you, since your presence on this
list implies that you are actively pursuing technical knowledge, but in
general managers tend to be far less knowledgable on these things.

Regards,

-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From sholden at holdenweb.com  Thu Feb  7 09:34:33 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 09:34:33 -0500
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <jAe88.87233$KM2.2811767@atlpnn01.usenetserver.com> <a3sklt$7vr$1@panix2.panix.com>
Message-ID: <Mfw88.94809$KM2.3437354@atlpnn01.usenetserver.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:a3sklt$7vr$1 at panix2.panix.com...
> In article <jAe88.87233$KM2.2811767 at atlpnn01.usenetserver.com>,
> Steve Holden <sholden at holdenweb.com> wrote:
> >"Spencer Doidge" <spencer at spencerdoidge.com> wrote ...
> >>
[he wants to pass arguments by reference]
>
> Um, that's not IMO/IME a particularly useful way to look at it.  I don't
> remember who first proposed it, but I find talking about "names" and
> "bindings" to be much clearer.  When talking about parameter passing, one
> would say that the function parameters create new names that get bound to
> the passed in objects.  Any time one uses assignment on the function
> parameter names (or any other name in Python, for that matter), the name
> gets rebound to a different object.
>
Maybe we should beat this one to death with a rock ;-)  I should point out
this is not for Aahz' benefit, since he knows it already.

Any time one uses assignment, one causes a name to be bound to an object.
Argument passing is actually binding a name (the argument name, present in
the function's local namesspace) to an existing object -- whatever was
passed as an argument, whether it was written as a complex expression (which
could create an object "on the fly") or simply as an existing name (which
would be a simple reference to an existing object).

Assignment to the argument name inside the function, as Aahz points out,
simply rebinds the argument name to a new value. Which means that such an
assignment will not change the object passed as an argument.

Assignment to *a component of the argument*, however, is a completely
different case (as I think someone else has already pointed out). So, if you
pass in to a function an argument arg, the assignment

    arg = something_new

will leave the argument object completely unaltered. However, the assignment

    arg[0] = something_new

will modify a list argument, and this modification will be visible outside
the function as well as inside it. You have mutated the object passed as an
argument, and so all names bound to that object will access the mutated
value. Similarly with arg.attr = something.new, arg.append(something), and
so on.

> If the object is mutable and mutated, all names that are bound to that
> object see the changes.  Mutation, by definition, *never* involves a
> rebinding of the object being mutated (that is, mutation never takes
> place in an assignment -- though of course the assignment expression
> might involve mutation as a side-effect).
>
You have to be careful to distinguish here between simple assignments (x =
expr) and augmented assignments (x += expr), since the latter are explicitly
allowed to mutate the object referenced by the name, but need not do so if
the implementation of the object concerned chooses not to. The former, I
agree, are simple rebindings.

> (Yes, you're talking about binding, but you don't go whole hog, which
> confuses the issue.)
>
I didn't get where I am today by not going the whole hog <wink>.

luckily-tim's-at-developers'-day-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From deblnonospammmy at raqia.com  Fri Feb  8 14:18:03 2002
From: deblnonospammmy at raqia.com (David Lees)
Date: Fri, 08 Feb 2002 19:18:03 GMT
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <3C642470.572F86B3@raqia.com>

Skip,

At least in the windows world, there is a free Powerpoint viewer on
Microsofts site.  Take a look at:
http://office.microsoft.com/downloads/default.aspx?Product=PowerPoint&Version=95|97|98|2000|2002&Type=Converter|Viewer

David Lees


Skip Montanaro wrote:
> 
> Guido posted a bunch of talk slides to the Python website in PowerPoint
> format.  I understand his reasons for not creating HTML versions (mostly
> because the HTML generated by PPT sucks).  Still, that leaves people without
> access to PPT sort of out in the cold.  What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT slides
> to other, more accessible, formats?  If we can get a reasonable (but fairly
> short) list of tools, I can probably see they get referenced on the relevant
> page of the Python website.
> 
> Thanks,
> 
> --
> Skip Montanaro (skip at pobox.com - http://www.mojam.com/)

-- 
debl


From eric.brunel at pragmadev.com  Thu Feb  7 12:35:21 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Thu, 7 Feb 2002 18:35:21 +0100
Subject: Getting correct file names under Windows
References: <a3ud3i$17g3$1@norfair.nerim.net>
Message-ID: <a3udm6$18b0$1@norfair.nerim.net>

Oops, I forgot: it's Python 2.1.1 on Win2K...
 - eric -




From jason at jorendorff.com  Thu Feb 14 16:13:05 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 15:13:05 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <a4h851$a8g7$1@ID-98166.news.dfncis.de>
Message-ID: <HFEKILOLEFEFMKAECNDLKEJODCAA.jason@jorendorff.com>

> I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> When I do a readline the I get back a string.
> How do I convert this string to a dictionary?

It's better to save data using pickle.
  http://www.python.org/doc/current/lib/module-pickle.html

import pickle
mydata = {'hello1': [1, 0, 0]}

# How to store data
f = open('mydata.dat', 'wb')
pickle.dump(mydata, f, 1)
f.close()

# How to load data back
f = open('mydata.dat', 'rb')
mydata = pickle.load(f)
f.close()

## Jason Orendorff    http://www.jorendorff.com/



From h.jehadalwan at student.kun.nl  Sun Feb 24 04:28:17 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Sun, 24 Feb 2002 10:28:17 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl> <a56o39$7dg$1@animus.fel.iae.nl> <3C7761D4.7050100@student.kun.nl> <a58k1m$8t1$1@animus.fel.iae.nl> <a58kjb$927$1@animus.fel.iae.nl>
Message-ID: <3C78B231.1010805@student.kun.nl>

Carel Fellinger wrote:

> Carel Fellinger <cfelling at iae.nl> wrote:
> 
> 
>>When you write to `i' you don't add arguments to the commandline used
>>to start `profit',  you're writing to the inputstream of `pdbs', quit a
>>
> 
> oeps, `pdbs' should read `profit'
> 
> 

hee, NEDERLANDER, leuk je te ontmoeten hier. Jammer dat er geen 
Nederlandse newsgroup is voor python.


groeten.





From jriveramerla at yahoo.com  Fri Feb  1 14:23:45 2002
From: jriveramerla at yahoo.com (Jose Rivera)
Date: 1 Feb 2002 11:23:45 -0800
Subject: Is thera a Metakit for Python 2.2  release ??
References: <11e94203.0201311433.519b556@posting.google.com> <L_v68.295688$pa1.74188709@news3.rdc1.on.home.com>
Message-ID: <11e94203.0202011123.44c67b2a@posting.google.com>

Hi..
Yes...I'm running it on Windows

It helps with programs that keep a log of transfers of files, reports
on files missing, monitoring the load of data to MSSQL Server
(DataWareHouse Project) keeping record of what has been loaded and it
works really good, I just wanted something simple, not to use a
relational database for this.

I tried "shelve" but the problem it is that we are talking of about
3000 files per day, and I want to keep information of two months...
categorized per city.

When I tried shelve I had an error... I tought was by the amount of
the information... so using metakit was a solution.

And if we take in account the speed that gives python to develop I
liked the results.


From gh_pythonlist at gmx.de  Mon Feb  4 00:57:07 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 4 Feb 2002 06:57:07 +0100
Subject: what about python ??
In-Reply-To: <832e3897.0202032107.25b3e70f@posting.google.com>
References: <832e3897.0202032107.25b3e70f@posting.google.com>
Message-ID: <20020204055706.GA3769@lilith.hqd-internal>

Le 03/02/02 ? 21:07, lennoncao ?crivit:
> pls,tell me what about python and where it can been used!

So you already read all of http://www.python.org/ The FAQ, in
particular, is at

    ==> http://www.python.org/doc/FAQ.html

Want to learn Python?

    ==> http://www.python.org/doc/Newbies.html

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.1 ?C      Wind: 4.7 m/s



From donn at drizzle.com  Wed Feb 13 18:00:16 2002
From: donn at drizzle.com (Donn Cave)
Date: Wed, 13 Feb 2002 23:00:16 -0000
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com>
Message-ID: <1013641215.358571@yasure>

Quoth grante at visi.com (Grant Edwards):
| I'd like to capture stdout/stderr from a program run with
| os.spawnv().  
|
| I've searched through c.l.p archives, and the usual answer
| seems to be "use popen2 et al".  I don't want to use popen
| since it runs the program as a shell command and I don't want
| the shell to mess with the strings I'm passing as arguments to
| the programs.

Check it out.  Not popen, but as they say, "popen2 et al."
It will exec the command on its own, no shell, if you give it
an argument list.  If you give it a string, it will use the
shell.

	Donn Cave, donn at drizzle.com


From magnus at thinkware.se  Wed Feb 20 19:55:45 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 16:55:45 -0800
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com>
Message-ID: <258fd9b8.0202201655.23bff72e@posting.google.com>

The problem is caused by the Helper in site.py:

import __builtin__

...

class Helper:
        def __repr__(self):
            import inspect
            if inspect.stack()[1][3] == '?':
                self()
                return ''
            return '<Helper instance>'
        def __call__(self, arg=None):
            import pydoc
            pydoc.help(arg)

__builtin__.help = Helper()

If I run this in cygwin, I get the same phenomena there.
Here's a snippet of output:

8<--
obals': <built-in function globals>, 'int': <built-in function int>, 'float': <b
uilt-in function float>, 'SyntaxError': <class exceptions.SyntaxError at 0xa0170
74>, 'pow': <built-in function pow>, 'vars': <built-in function vars>, 'help':
Welcome to Python 2.1!  This is the online help utility.

If this is your first time using Python, you should definitely check out
the tutorial on the Internet at http://www.python.org/doc/tut/.
8<--

"del __builtin__.help", or removing the above from site.py solves the 
problem, but removes the nice help facility... But changing 

__builtin__.help = Helper()

to

__builtin__.help = Helper

should make all systems go, except that I need to type "help()" to
reach the help menu. But maybe some fix in the Helper class could
sense the vars() problem. I'm too tired to figure that out now.


From tatebll at aol.com  Thu Feb 21 08:25:29 2002
From: tatebll at aol.com (Bill Tate)
Date: 21 Feb 2002 05:25:29 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com>
Message-ID: <cb4ba455.0202210525.6687556f@posting.google.com>

"Peter Milliken" <peter.milliken at gtech.com> wrote in message 
> Personally, I have used Python for several small, personal projects and the
> odd "script" that searches text files - even here, the only perceived
> advantage is the library support, I often wonder whether I wouldn't be
> better off using my real language of choice (Ada) because of it's strong
> typing - it too has interfaces to the standard GUI tools etc. My next
> personal project will not be in Python, it is a "cute" language and allows
> some nice facilities i.e. creating a script at run-time and then running it
> can be very powerful in the correct circumstances - so I will probably use
> an Ada-Python API for these situations and get the best of both worlds :-).
> When I compile a Ada program cleanly I know that 95% of the bugs are out of
> it - before I even start testing! :-) Of course, it takes me longer to write
> up front but overall development effort (in my subjective opinion - since I
> have no metrics :-)) is that it is much lower.

Peter,
Perhaps you should also use adjectives like "toy-like," "child-like,"
"foolish," "trivial," and "not a real language" while your at it.  You
offer an opinion without supporting metrics and effectively relegate a
language to the realm of triviality.  Should I presume, that you also
lump the python developer community within the realm of "caution:
children at play?"  Perhaps you could spare us the "cute" adjectives. 
Your remarks smack of intellectual superiority.

You're right to assert there are risks and costs associated with using
a dynamically typed language.  However, are you of the opinion that
this is a community that doesn't understand what those risks and costs
are?  Perhaps this is a community that understands, better than others
I suspect, how/when to make practical tradeoffs and how to apply the
necessary tradecraft to know when to use python and when not to?

Bill


From * at spam.ruud.org  Wed Feb 27 12:04:16 2002
From: * at spam.ruud.org (ruud de rooij)
Date: 27 Feb 2002 12:04:16 -0500
Subject: Python and Daemons
References: <02022215115601.01886@localhost.localdomain> <mailman.1014416559.28231.python-list@python.org> <8b3354da.0202270849.6b994141@posting.google.com>
Message-ID: <87wuwy26wf.fsf@bunty.ruud.org>

> > def daemonize():
> >     "cause this process to becaome a daemon"
> >     if os.fork() != 0:
> >         os._exit(0)
> >     os.setsid()
> >     if os.fork() != 0:      # 2nd fork is widely recommended
> >         os._exit(0)         # though not always needed
> >     debugmsg("daemonized")

> Can someone say why the 2nd fork is widely recommneded?

http://www.erlenstar.demon.co.uk/unix/faq_2.html#SEC16

3. fork() again so the parent, (the session group leader), can
   exit. This means that we, as a non-session group leader, can never
   regain a controlling terminal.

        - ruud
-- 
ruud de rooij | *@spam.ruud.org | http://ruud.org


From nelson at crynwr.com  Wed Feb 20 10:29:11 2002
From: nelson at crynwr.com (Russell Nelson)
Date: Wed, 20 Feb 2002 10:29:11 -0500 (EST)
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <31575A892FF6D1118F5800600846864DCBCCC1@intrepid>
References: <31575A892FF6D1118F5800600846864DCBCCC1@intrepid>
Message-ID: <15475.49175.229691.859111@desk.crynwr.com>

Simon Brunning writes:
 > for line in file:
 >     line.whatever()

Cool.  Good reason to upgrade to 2.2.

 > >     matched = re.match(r'egexp(.*)', line)
 > >     if matched:
 > >         print matched.group(1)
 >  
 > Assignments and expressions are different things in Python...

Indeed, and that's a good thing.  But I'm proposing an alternate
syntax for the if statement, specifically:
	"if" <variable> "=" <expression":.

-- 
-russ nelson              http://russnelson.com | Crypto without a threat
Crynwr sells support for free software  | PGPok | model is like cookies
521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | 



From tdelaney at avaya.com  Thu Feb 28 20:45:28 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 1 Mar 2002 12:45:28 +1100 
Subject: Status of PEP's?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1CA@natasha.auslabs.avaya.com>

> From: Samuele Pedroni [mailto:pedronis at bluewin.ch]
>
> PS: I have a Master in Mathematics.

Burn him!

Tim Delaney



From hmartires at cultalg.pt  Thu Feb 21 09:40:44 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Thu, 21 Feb 2002 14:40:44 -0000
Subject: MySQLdb
Message-ID: <XD7d8.203$zZ4.2649220@newsserver.ip.pt>

Where i can find an "how to" to learn working with MySQLdb module.
I have downloaded from sourceforge and install it, but there is no much
information on how to use it.
Any one knows where i can get it ?

Tanks in advance




From MaximilianScherr at T-Online.de  Sat Feb  2 17:47:11 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Sat, 02 Feb 2002 22:47:11 -0000
Subject: hex to normal chars?
Message-ID: <a3hq9f+2q3f@eGroups.com>

how can i transfer hex numbers to normal chars, like "41" to "A" etc.?




From gh_pythonlist at gmx.de  Tue Feb 26 19:15:39 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Wed, 27 Feb 2002 01:15:39 +0100
Subject: [ANN] JPE 2.0 The Java-python Extension
In-Reply-To: <3C7BFAB1.643BE736@arakne.com>
References: <3C7BFAB1.643BE736@arakne.com>
Message-ID: <20020227001539.GA1598@lilith.hqd-internal>

Le 26/02/02 ? 13:14, Frederic Giacometti ?crivit:
> This is the official announcement of the second version of JPE.
> [...]

Good to hear you've continued this project. I tried to build it on my
Debian system, which does have a shared Python library by default, but I
didn't get very far.

I wonder how this is supposed to work:

libplus/config/cfgsys.py, line 169:

self.ISSHARED = sysbuild.LDLIBRARY == sysbuild.SHLIBRARY

I couldn't find the string SHLIBRARY variable anywhere in jpe/, libplus/
or my Python sources except in this file, at this line.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.8 ?C      Wind: 5.0 m/s



From fredrik at pythonware.com  Wed Feb 13 11:19:24 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 16:19:24 GMT
Subject: UTC time conversion to internal time
References: <49ja8.17872$N31.902613@ozemail.com.au>
Message-ID: <gmwa8.25527$l93.4755349@newsb.telia.net>

Colin Brown wrote:

> I wish to convert UTC times to internal time representation (the inverse of
> time.gmtime). There does not seem to be any function to do this. Does anyone
> have a clean solution please.

in recent Python versions, you can use calendar.timegm:

>>> import calendar
>>> help(calendar.timegm)
Help on function timegm in module calendar:

timegm(tuple)
    Unrelated but handy function to calculate Unix timestamp from GMT.

:::

if it's not available in your Python version, you can use this one instead:

# time-example-4.py
# from O'Reilly's "Python Standard Library"

import time

def _d(y, m, d, days=(0,31,59,90,120,151,181,212,243,273,304,334,365)):
    # map a date to the number of days from a reference point
    return (((y - 1901)*1461)/4 + days[m-1] + d +
        ((m > 2 and not y % 4 and (y % 100 or not y % 400)) and 1))

def timegm(tm, epoch=_d(1970,1,1)):
    year, month, day, h, m, s = tm[:6]
    assert year >= 1970
    assert 1 <= month <= 12
    return (_d(year, month, day) - epoch)*86400 + h*3600 + m*60 + s


</F>

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




From gh_pythonlist at gmx.de  Sun Feb  3 14:13:49 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Sun, 3 Feb 2002 20:13:49 +0100
Subject: Xython - XML-Formed Python
In-Reply-To: <1012751401.4431.13.camel@acropolis>
References: <1012751401.4431.13.camel@acropolis>
Message-ID: <20020203191348.GA1476@lilith.hqd-internal>

Le 03/02/02 ? 16:49, Gerard Braad ?crivit:
> Hello,
> 
> For a long time now, I had the idea of XML-Formed Python. This way it
> would be possible to send Python code across a network connection.

IMO no need for XML there, marshal should be able to do that just as
fine.

> The only problem was, I didn't had much time to realise this. Way back
> in August I saw someone (mlh) had the same idea and in December the
> project page was included on the DDJ Python-URL. JEEJ!
> 
> I wasn't aware of this and hadn't worked on it, until now... Work on the
> initial draft has started. If you are willing to help, don't hesitate
> and contact us...
> 
>   //sourceforge.net/projects/xython/
>   //hetland.org/python/xython.php
>   //proj.survion.com/?page=xython
> 
> Any help would be appreciated. We hope this will become a standard for
> inclusion of Python in XML documents.

I also had this idea once. I thought of it when my boss told me the new
Java app should use an XML config file in order to be buzzword compliant
=:-) 

A very simple way to put Python code in XML would IMO be to base64
encode the Python source and then put it in a XML element. Or just use a
CDATA section. I can't think of a good reason to do this in reality,
though, except for the buzzword compliance.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.1 ?C      Wind: 1.0 m/s



From rhymes at myself.com  Fri Feb 15 14:51:16 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Fri, 15 Feb 2002 20:51:16 +0100
Subject: Highlights & Cursors
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com> <3daa1aa8.0202141551.1dae3651@posting.google.com>
Message-ID: <2apq6ucuh9n7s446rrcsdi2fu7eis0qaqu@4ax.com>

On 14 Feb 2002 15:51:27 -0800, jmsun at bioeng.ucsd.edu (Jeffrey) wrote:

> In terms of cursors, that's a different story.  Note that this is
>off the top of
>my head.  But I'll assume that you are tyring to change the cursor? 
>For the same widget, what you do is (this is off the top of my head)
>some_widget.bind("Any-Enter", change_cursor_enter ) where
>change_cursor is a function that changes the cursor to some other
>special cursor.  Then you have to bind the other way around too like
>some_widget.bind("Any-Leave", change_cursor_leave )....

My question is: how can i change the cursor in a frame or some others
widgets?

In the doc "an-introduction-to-tkinter" downloaded from pythonware.com
there is a chapter explaining the widget styling (ch. 6).
The last paragraph is about the cursor styles

Here is the paragraph:

"
The cursor option control which mouse cursor to use when the mouse is
moved over the
widget. If this option isn't set, the widget uses the same mouse
pointer as its parent.
Note that some widgets, including the Text and Entry widgets, set this
option by default.

"

how could i change this option?

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From bokr at oz.net  Sat Feb 16 18:02:55 2002
From: bokr at oz.net (Bengt Richter)
Date: 16 Feb 2002 23:02:55 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122> <xyuk7tdku0w.fsf@sex.ifi.uio.no> <m3y9hthzhu.fsf@mira.informatik.hu-berlin.de>
Message-ID: <a4moev$j05$0@216.39.172.122>

On 16 Feb 2002 22:45:49 +0100, martin at v.loewis.de (Martin v. Loewis) wrote:

>Jan-Fredrik Braseth <janfbr at sex.ifi.uio.no> writes:
>
>> > # I would change opt to c in the following line
>> >                     self.msg('IAC %d not recognized' % ord(opt))
>> 
>> But if there really is something wrong with that, shouldn't someone
>> send a bug report or something to the creators of Python?
>
>Please do: send to to sf.net/projects/python. A patch might help even
>more; please use a context or unified diff.
>
>
Oops, you were faster than I thought would happen. Sorry ;-/

If Jan-Fredrik finds the change improves things, the diff will look
something like:
_________________________________________________________________

--- \Python22\Lib\telnetlib.py.orig     Sun Oct 07 04:53:32 2001
+++ \Python22\Lib\telnetlib.py  Sat Feb 16 14:54:27 2002
@@ -421,7 +421,7 @@
                     else:
                         self.sock.send(IAC + DONT + opt)
                 else:
-                    self.msg('IAC %d not recognized' % ord(opt))
+                    self.msg('IAC %d not recognized' % ord(c))
         except EOFError: # raised by self.rawq_getchar()
             pass
         self.cookedq = self.cookedq + buf
_________________________________________________________________

Let's see what J-F finds.

Regards,
Bengt Richter



From phr-n2002a at nightsong.com  Thu Feb 21 06:22:02 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 03:22:02 -0800
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com> <Xns91BC6120D2B5duncanrcpcouk@127.0.0.1>
Message-ID: <7xbsejkrl1.fsf@ruckus.brouhaha.com>

Duncan Booth <duncan at NOSPAMrcp.co.uk> writes:
> The main differences between a coroutine and a generator are:
> 
> 1) Coroutines maintain their own stack. This allows a coroutine to yield 
> not just from the function you originally called, but also from any nested 
> function no matter how deep.
> Python's generators can only yield directly from the generator as they do 
> not attempt to maintain any kind of stack.

Is that right?  I thought the classic coroutine example is supposed to
also work with generators:

  def traverse(tree):
     if tree.left: traverse(tree.left)
     yield tree.value
     if tree(right) traverse(tree.right)

Am I missing something?


From gerson.kurz at t-online.de  Mon Feb  4 08:44:58 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 13:44:58 GMT
Subject: autocoding project proposal
References: <mailman.1011605371.27355.python-list@python.org>
Message-ID: <3c5e9001.14574312@news.t-online.de>

On Mon, 21 Jan 2002 01:28:24 -0800 (PST), "Sean 'Shaleh' Perry"
<shalehperry at attbi.com> wrote:

>By autocoding I assume you mean running a program which outputs python code
>which is then run.  

Say, wouldn't that be simlpy "a Quine"? (see
http://www.nyx.net/~gthompso/quine.htm)

If so, here is one in Python:

http://p-nand-q.com/python/quine6.txt


From newgene at bigfoot.com  Fri Feb 22 15:35:13 2002
From: newgene at bigfoot.com (newgene)
Date: 22 Feb 2002 12:35:13 -0800
Subject: The good "import" convention?
Message-ID: <a50631f6.0202221235.5858bdcd@posting.google.com>

Hi, group,
     When I write a python module, I got a question about the location
of "import" statement.

1. import all needed modules in the top of the code. Some modules
might be used in only one function.

import module1,module2,...
import ...
def f1():
   ....
def f2():
   ....

2. import needed modules in each functions. Some common modules like
"os", might be imported in multiple functions.

def f1():
   import module1
   ...
def f2():
   import module1,modules2
   ...

Which one is better or they are the same for the performance?


Thanks.

NewGene


From ivano.delpiccolo at degroof.be  Wed Feb 13 05:14:15 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Wed, 13 Feb 2002 11:14:15 +0100
Subject: popen3 Object
References: <3c69138a$0$12226$4d4efb8e@news.be.uu.net> <WIda8.5918$ZO2.295633@stones>
Message-ID: <3c6a3c71$0$7109$4d4efb8e@news.be.uu.net>

Thanks It's more than enought
Thank yu very much.
"Graham Ashton" <graz at mindless.com> wrote in message
news:WIda8.5918$ZO2.295633 at stones...
> On Tue, 12 Feb 2002 13:07:30 +0000, DelPiccolo Ivano wrote:
>
> > How can I print the stdout of the popen3
> >
> > exemple :
> >
> > fileobjet = os.popen3("ls -la", "true" ) print ??????????
>
> Using Python 2.1 on Linux:
>
> ratchet% python
> Python 2.1 (#8, Aug 22 2001, 16:46:14)
> [GCC 2.95.2 20000220 (Debian GNU/Linux)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>> import os
> >>> os.popen3("echo foo", "t")
> (<open file '(fdopen)', mode 'w' at 0x8134478>, <open file '(fdopen)',
> mode 'r' at 0x8134168>, <open file '(fdopen)', mode 'r' at 0x8132e00>)
>
> So we know that popen3 returns three open file objects, the first one
> open for writing and the second two for reading. If you check the docs
> you'll see that they are the child's stdin, stdout and stderr. So,
> returning to the interactive python session:
>
> >>> stdIn, stdOut, stdErr = os.popen3("echo foo")
> >>> while 1:
> ...     line = stdOut.readline()
> ...     if line == "":
> ...         break
> ...     else:
> ...         print line,
> ...
> foo
> >>>
>
> Does that help any?
>
> --
> Graham




From missive at frontiernet.net  Wed Feb  6 21:47:30 2002
From: missive at frontiernet.net (Lee Harr)
Date: Thu, 07 Feb 2002 02:47:30 -0000
Subject: Twisted Selector
References: <u5rsqqb9tamaad@corp.supernews.com> <LBe88.87264$KM2.2829250@atlpnn01.usenetserver.com>
Message-ID: <u63qm2cnomt801@corp.supernews.com>

>> I am looking at this Twisted tutorial:
>> http://twistedmatrix.com/page.epy/philosophy.html
>> and having trouble with the very first code.
>> 
> Moshe? Glyph?
> 

Is anyone else using this Twisted thing?
It looks cool to me, but I can't fix it :(



From rony.steelandt at bucodi.com  Fri Feb  8 03:25:55 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 8 Feb 2002 00:25:55 -0800
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>
Message-ID: <b34d3835.0202080025.1d69d198@posting.google.com>

YOU SHOULD BE ASHAME !!!!!!!!!!!!!!!!!!!!!

Who on earth you think you are, to just post a message on usenet
saying don't talk to that guy !!

1. Nobody or nothing gives you that right.
2. It's a attack to peoples intelegence here, i assume they very well
can decide by themself if they want to dicusse or not
3. It's against the spirit of usenet
4. You are not God to decide this
5 ....

I could go on for hours like that

Think about this

Rony


sandskyfly at hotmail.com (Sandy Norton) wrote in message news:<b03e80d.0202060227.368f6ea7 at posting.google.com>...
> "Whereof one cannot speak, thereof one must be silent."


From donn at drizzle.com  Sun Feb  3 14:51:59 2002
From: donn at drizzle.com (Donn Cave)
Date: Sun, 03 Feb 2002 19:51:59 -0000
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <mailman.1012745045.21128.python-list@python.org> <3c5d7e28_2@news.nwlink.com>
Message-ID: <1012765918.861234@yasure>

Quoth "Guy Gascoigne - Piggford" <guy at wyrdrune.com>:
| Actually I did look at them, one of them seemed somewhat limited for what I
| needed and the other uses an external expect library for functionality.
| What I want need has to execute under jython - so I can extend it in java
| but not C (at least easily).
|
| So that leaves me with my original syntax question.

Which I missed, but when I had thoughts like that once upon a time,
the best I could think of was to use the exception mechanism.  Or
abuse it, if you like.  Make pattern objects, feed them to the expect
parser in a try block, which would raise it as an exception to dispatch
to the appropriate except block.  Horrible.

But you must have support for pseudotty devices, which is a low level
functionality that is not so portable.  To my knowledge available only
on UNIX and similar platforms, and not on all such for Python.  Have
that in Java?  I don't know, but if not, you're sunk.

	Donn Cave, donn at drizzle.com


From martin_byrne2 at yahoo.co.uk  Fri Feb 22 03:33:56 2002
From: martin_byrne2 at yahoo.co.uk (B)
Date: 22 Feb 2002 00:33:56 -0800
Subject: Python ActiveX Scripting Engine error
References: <508cb8ae.0202210917.6d97e157@posting.google.com> <3C752D64.9090100@mxm.dk>
Message-ID: <508cb8ae.0202220033.1c3e7cd7@posting.google.com>

Max M <maxm at mxm.dk> wrote in message news:<3C752D64.9090100 at mxm.dk>...
> Hmm I sometimes experience those. I think it is when I have PythonWin 
> open at the same time an I try to run a script with errors. Then it sort 
> of "gets stuck" on the iis.
> 
> A restart of IIS solves it for me.
> 
I dont have PythonWin open and restarting IIS makes no difference, the
error just keeps happening randomingly it seems.
I did read something about redirecting before a html page is build can
cause this but if anyone has any other ideas I'd like to hear them.

Cheers


From phil.rittenhouse at dspfactory.com  Tue Feb 26 12:28:02 2002
From: phil.rittenhouse at dspfactory.com (Phil Rittenhouse)
Date: Tue, 26 Feb 2002 12:28:02 -0500
Subject: Converting C strings to Python problem
Message-ID: <BwPe8.21867$X2.226876@nnrp1.uunet.ca>

Hi,

I'm trying to pass a path string (eg. "c:\spam\test.exe") from C to python,
but python is interpreting the string as "c:\spam<tab>est.exe".

I've run into this problem with converting C strings to Python strings using
Py_BuildValue().   When converting a C string with backslashes to a
python string, the string (including backslashes) is copied exactly.  Sounds
reasonable, except that, according to the doc's, Python
uses two character escape sequences to represent things like tabs (\t).

Should the conversion process convert backslashes in C to double backslashes
in Python?   What is the recommended method for doing the C string to python
string conversion?  Does PyString_FromString() behave differently?

Thanks for your help,
Phil
------------
Phil Rittenhouse
phil.rittenhouse at dspfactory.com
Software Developer
Dspfactory Inc.









From cfelling at iae.nl  Thu Feb 28 18:03:38 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 1 Mar 2002 00:03:38 +0100
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>        <mailman.1014861466.23611.python-list@python.org>        <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com> <mailman.1014929989.16553.python-list@python.org>
Message-ID: <a5md0a$an8$1@animus.fel.iae.nl>

Skip Montanaro <skip at pobox.com> wrote:
...
> Don't assume for loop iteration is your only culprit... ;-) List creation
> overhead can be as big a problem.  I added another test function called iter

yep, see my other posting:)

> which relies on 2.2's new "for k in dict" capability to completely avoid
> list creation.  (I also changed your range() to an xrange() to avoid the
> extra list allocation, and called clock() instead of time() to measure CPU
> time instead of elapsed time.  This is on a Linux system) Here's what I get
> on my laptop:

Ah a new contestent, double fun, better algoritme better timing (linux
too), but the real iter still wins, although by a very slight margin:)

>     iter 6.69
>     iteritems 28.36
>     mapvalue 8.12
...
  def iteriteritems(f, d):
          for k, v in d.iteritems():
                  d[k] = f(v)
...
>         timeit(iter, f, d)
          timeit(iteriteritems, f, d)
          timeit(iteritems, f, d)
>         timeit(mapvalue, f, d)

iter 10.95
iteriteritems 10.43
iteritems 52.15
mapvalue 12.88
-- 
groetjes, carel


From scarblac at pino.selwerd.nl  Tue Feb 26 06:12:51 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 26 Feb 2002 11:12:51 GMT
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014624160.28614.python-list@python.org>
Message-ID: <slrna7mr7g.nl9.scarblac@pino.selwerd.nl>

Christian Tanzer <tanzer at swing.co.at> wrote in comp.lang.python:
> 
> Peter Hansen <peter at engcorp.com> wrote:
> 
> ( lots of `gefingerpoken` schnapped)
> 
> > Apparently German, as Perl, has many ways of saying the same thing.
> > And it's equally readable.  <grin>
> 
> To a german speaker, this actually looks like dutch. Must be Pythonic,
> then. <wink>

To a dutch speaker, it looks like the funny half-german half-english it was
meant to be. But no Dutch at all there :)

-- 
Remco Gerlich


From jeff at ccvcorp.com  Thu Feb 14 14:58:13 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 11:58:13 -0800
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com>
Message-ID: <3C6C16D4.2B035A20@ccvcorp.com>


Paul Moore wrote:

> I have no requirement to include a Python interpreter, or C extensions,
> in the archive. The code is all pure Python. But, I do want to be able
> to include support files (text files, image files for use in wxPython
> toolbars, graphics for display on screen, sound files, etc etc) which
> can be picked up from the same archive.

If you require this being in a single archive file only for transfer
purposes, you can use distutils (standard in 2.1+, not sure about 2.0).
Distutils will create a single zipfile/.tgz/rpm that can be distributed
easily.  To install the package, it's only necessary to open the archive and
type 'python setup.py install'.  (It may also be able to create a
single-file self-installing .exe on Windows.)  Note that once you install
the package, it is no longer a single archive file, but is fully extracted
to its individual component files, so this may or may not meet your needs.

Jeff Shannon
Technician/Programmer
Credit International




From roy at panix.com  Mon Feb 18 11:18:59 2002
From: roy at panix.com (Roy Smith)
Date: Mon, 18 Feb 2002 11:18:59 -0500
Subject: Performances of Pyhton programs
References: <Xns91B9AA94FE461bairef2ctiecpfr@138.195.8.3>
Message-ID: <roy-0B7BBB.11185918022002@reader2.panix.com>

bairef2 at cti.ecp.fr (Fran?ois Baire) wrote:
> Are there any benchmark or does anoyone know what are the performances of a 
> program written in python (in terms of CPU and RAM) compared with other 
> languages (especially C++ and compiled languages) ?
>    I've met many articles which explain that python programs are slowler 
> because it uses the interpreter, but none says how much slower it is ! I 
> would like to know this for a rather big OO program which uses standard 
> fonctions of the language.

There is little doubt that a program written in C/C++ will be faster than 
the same algorithm implemented in Python (or any of the common "scripting" 
languages such as Perl or TCL).  How much slower depends to a very large 
degree on exactly what you're doing.

I've got one program in Python which is probably 100x slower than a C 
version because it does a huge amount of low-level string manipulation 
(building up strings character by character) which is exactly the wrong 
thing to be doing in Python.  When I profile it, over 99% of the CPU time 
is spent in the string module.  But, that means it takes 3 seconds to run 
instead of a fraction of a second, and it saved me days of development time.

On the other hand, I've written Python database applications which 
essentially ran at the same speed as a C version would because all the hard 
work was really being done inside Oracle.

You might want to check out http://www.bagley.org/~doug/shootout/ for a few 
random datapoints.


From bob at kontact-no-spam-please.com  Mon Feb 18 21:16:37 2002
From: bob at kontact-no-spam-please.com (bob)
Date: Mon, 18 Feb 2002 18:16:37 -0800
Subject: Recursively traverse linked list -- help!
Message-ID: <3C71B585.90F644EB@kontact-no-spam-please.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020218/ba99b88d/attachment.html>

From sameer_ at email.com  Sun Feb 17 22:18:54 2002
From: sameer_ at email.com (sameer)
Date: 17 Feb 2002 19:18:54 -0800
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com>
Message-ID: <6bd9f01b.0202171918.a58611@posting.google.com>

sameer_ at email.com (sameer) wrote in message news:<6bd9f01b.0202160255.77244c9e at posting.google.com>...
> Here's the situation:
> I have a python package (or two) to be installed on a windows machine
> (for now) which has python installed.  I am using distutils to create
> a windows installer which installs my python packages just fine.  Now,
> I want to also create some shortcuts on the Desktop as well as the
> Start Menu.  As of now, distutils doesn't allow me to do either of
> those things.  What are my options here to do either or both of these
> things.  Can the distutils package be modified easily to provide this
> kind of support on windows?  Or is there another python package
> available that would allow me to create the desktop and Start Menu
> shortcuts?

it's a pity no one has replied.  I suppose there are no sane options for this.


From phr-n2002a at nightsong.com  Thu Feb 14 01:15:50 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 13 Feb 2002 22:15:50 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <mailman.1013571921.18465.python-list@python.org> <7xvgd2589b.fsf@ruckus.brouhaha.com> <mailman.1013574741.31892.python-list@python.org>
Message-ID: <7xaducmvvt.fsf@ruckus.brouhaha.com>

Skip Montanaro <skip at pobox.com> writes:
>     Paul> here 50000 and -1 are the starting and ending locations.
> 
>     Skip> If start > end isn't direction == -1 implied?
> 
>     Paul> Nope.  I tried it and it didn't search backwards.  
> 
> That wasn't quite what I was asking.  Let me ask it a slightly
> different way.  Presuming the behavior can be changed so that
> searching in reverse works, shouldn't the ordering of start and end
> be sufficient to indicate the direction of the search?

That might have been an ok way for the implementers to have originally
designed the re module, but they designed and documented a different
way instead.  Changing the behavior now of already-documented
arguments could break existing programs.  It's better to add a new arg
to preserve compatibility, and it's a cleaner design anyway to use a
separate arg.


From phr-n2002a at nightsong.com  Thu Feb 21 03:14:39 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 00:14:39 -0800
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>
Message-ID: <7xlmdndzf4.fsf@ruckus.brouhaha.com>

"Eduard Smirnov" <Ed.Smirnov at mail.ee> writes:
> If you have any experience in writing applications in system programming
> language (e.g. in C, C++) and re-writing it in scripting language (e.g.
> Python) or vice versa, please compare your work in both implementations,
> describe your work/application?s functionality, approximate number of lines
> and time of development of each implementation, skill?s level needed.
> Any more comments about are welcome.
> 
> Additional thanks for copy your answer to my e-mail.

I don't know of studies involving Python but Bernie Greenberg's paper
about Multics Emacs (written in interpreted Lisp instead of PL/I)
might be interesting to you.


From sdm7g at virginia.edu  Mon Feb 25 15:02:30 2002
From: sdm7g at virginia.edu (Steven Majewski)
Date: Mon, 25 Feb 2002 15:02:30 -0500 (EST)
Subject: functional programming with map()
In-Reply-To: <k4we8.3526$N7.997040@ruti.visi.com>
Message-ID: <Pine.OSX.4.40.0202251438110.509-100000@d-128-54-224.bootp.virginia.edu>


...
> >>I'd prefer
> >>[x.f() for x in items]
> >>
> >>It's not functional syntax, but so what?
> >
> > Sure it is.

On Mon, 25 Feb 2002, Grant Edwards wrote:

> Methinks there are two definitions of "functional" at work here:
>
>  1) The traditional CS "no side-effects" definition.
>  2) Something that looks like a function  call.
>
> ?

"Functional Programming" is a programming style that uses functions
as first class objects, favors declarative over procedural, and avoids
side effects.

"Functional Programming Languages" are languages that support that
style of programming.

"Pure Functional Programming Languages" are languages that support
that style of programming and try to prevent you from programming
in any other style.


It's not so much two definitions as it is a disagreement over whether
"functional" is  a binary or an analog adjective. It's the same issue as
with "object oriented": i.e. can Python be considered an Object Oriented
Language if it's missing one of the "required" features ( like
encapsulation, for example) ?

For both Functional and O-O programming, the style evolved first, then
came better language support, and last came the academic purifiers.


List comprehensions are declarative (no loops) and they avoid side effects
(no loop variables or temporaries)  -- plus they were originally used
in a pure functional language (Haskell) -- sounds like functional
syntax to me!


-- Steve Majewski






From gerson.kurz at t-online.de  Mon Feb  4 08:12:14 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 13:12:14 GMT
Subject: WANTED: IBM 1401 AUTOCODING WAREZ [was: Autocoding - the resistance  to expect from psuedo programmers]
References: <mailman.1012829589.12111.python-list@python.org>
Message-ID: <3c5e885b.12615968@news.t-online.de>

On Mon, 4 Feb 2002 08:32:34 -0500 (EST), Steven Majewski
<sdm7g at Virginia.EDU> wrote:

>HEY DUDEZ !
> I WANT TO GET INTO THIS AUTOCODING THING.
> ANYONE KNOW WHERE I CAN GET SOME IBM 1401 WAREZ ?
>-- MULTICS-MAN

DUDE, I THINK YOU GOT IT ALL WRONG: AUTOCODING MEANS CODING WHILE
DRIVING IN AN AUTOMOBILE, KIND OF DRIVE-BY-SHOOTING FOR GEEKS.





Gerson
http://p-nand-q.com


From brian at sweetapp.com  Wed Feb 20 17:12:55 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Wed, 20 Feb 2002 14:12:55 -0800
Subject: HTMLifyer for Python code?
In-Reply-To: <60FB8BB7F0EFC7409B75EEEC13E2019221513D@admin56.narex.com>
Message-ID: <00be01c1ba5b$bf9c3400$445d4540@Dell2>

You could try either:
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52298

Or (bias):
http://silvercity.sourceforge.net/

The former is a pure Python module that is quite is to use. The later is
more flexible and can syntax-color HTML and XML strings embedded in
Python code.

> -----Original Message-----
> From: python-list-admin at python.org
[mailto:python-list-admin at python.org]
> On Behalf Of Bjorn Pettersen
> Sent: Wednesday, February 20, 2002 12:56 PM
> To: python-list at python.org
> Subject: HTMLifyer for Python code?
> 
> I just browsed through the Vaults, and did a quick Google search, but
I
> couldn't find a tool for converting Python source code to
> syntax-highlighted HTML. Surely someone must have written something
like
> this?
> 
> -- bjorn
> 
> --
> http://mail.python.org/mailman/listinfo/python-list




From mcherm at destiny.com  Mon Feb 18 08:33:20 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 18 Feb 2002 08:33:20 -0500
Subject: Python License Issues
Message-ID: <3C7102A0.4020307@destiny.com>

>> Hello from rainy London
>> > We have an issue with the python license
> 
> [...]
> 
>> We can consider accepting liability 
>> on all third party code we use, for such gross misconduct situations but is 
>> it technically possible to do so given the PSF license? Ideally we would 
>> like to negotiate a single license agreement with these clients (our license 
>> agreement) which provides additional liability over and above to that which 
>> PSF provides. However given clause 4 it seems that the client *must* 
>> directly be bound by the PSF license. 
>> > Any comments? 
>> > Pavlos
> 
> [Silly idea of mine about dual liscensing]


OK, I had a weekend to think it over, and I now realize my original 
suggestion was really, REALLY silly. There's a FAR simpler solution than 
dual licensing. Clause 4 states:


4. PSF is making Python 2.1.2 available to Licensee on an "AS IS"
     basis.  PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
     MPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND
     DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR
     FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON
     <version> WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.

So if YOU wanted to offer such a guarantee, YOU could provide it 
separately. In fact, I'd recommend that you charge a separate fee for 
selling what is essentially a liability policy -- you provide a warranty 
that Python IS fit for merchantability or for some purpose, or isn't 
infringing. If one of these turns out to be false, you'll make up any 
losses, expenses, or liability for policy holders.

Of couse I should point out that this is ALL theoretical. There is no 
way that you could provide such a service unless you are a good-sized 
insurance firm, and there is no corporate lawyer in the world who is 
going to allow such a policy to be written (despite the fact that (in my 
opinion) it would never have to pay out anything).

All told, I'm glad that suing each other over merchantability is NOT 
part of the Python culture.

-- Michael Chermside







From sagwy at earthlink.net  Wed Feb  6 17:49:20 2002
From: sagwy at earthlink.net (mo)
Date: Wed, 6 Feb 2002 15:49:20 -0700
Subject: Quixotic Thread:  Author and Intent Revealed....
Message-ID: <0a6301c1af60$859fad00$bb4b0b3f@gatesway>

YKW* is the alter ego of the BFDL.  He and the other Paters of Python
are preparing this community for the  momentous release of p_VIC
scheduled for release 4/1/02.

PS:  I realize that this premature disclosure will result in an order
from the Directorate of the PSU launching a squadron of Norwegian
Blues to deliver a massive quantity of Norwegian Blue quano in my
direction.;)

*YKW = You Know Who

Mo






From deathtospam43423 at altavista.com  Sat Feb  9 21:52:40 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 10 Feb 2002 02:52:40 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org>
Message-ID: <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
news:mailman.1013306296.2115.python-list at python.org: 

> Let's go over this step-by-step.  First, what web server did you
> download? Most docs assume Apache.
> 
It's called Lil' HTTP server by Summit Communications Networks
http://www.summitcn.com/lilhttp/lildocs.html

> You have a form (stored in a .html file or produced by a cgi script,
> doesn't matter) which contains something like this:
> 
>     <form method='POST' action='http://url.to.the/cgi-bin/script'>
> 
> "script" must be a script file in the cgi-bin directory.  You probably
> can shortcut it like this:
> 
>     <form method='POST' action='/cgi-bin/script'>
> 
> and this is probably better.
> 
The actual text on my html form is 

    	<FORM method="post" Action="script.cgi">

I have tried changing the file the file extension to script.text, 
script.dat, and even as you seem to recommend above, just script. I press 
the SEND button on the form and a blank form comes up, with the filename 
"script.cgi".

> The 11.2 docs then describe how you decode the data sent by the form.
> 
This is the problem. I write data on the form, but no data turns up in the 
file(?) Isn't there supposed to be? 

> If you are still having problems, please post code (and tell us what
> webserver software you are using).
> 
Thanx. Here's the form (minus the head and title)...

<BODY BGCOLOR="#FFFFFF">

<P>

<FORM method="post" Action="script.cgi">

<INPUT NAME="name" Size=50 Type="text"> <B>Your Name</B><BR>
<INPUT NAME="email" Size=50 Type="text"> <B>Your E-Mail Address</B><BR>

<B>Would you like a reply? </B>
<INPUT TYPE="radio" NAME="reply" VALUE="yes"> <B>Yes</B> <INPUT 
TYPE="radio" NAME="reply" VALUE="no"> <B>No</B><P>

<!-- This is the spot to specify your email address. -->
<INPUT TYPE="hidden" NAME="submitaddress" VALUE="user at server.com">

<B>Write to me below:</B><P>
<TEXTAREA NAME="feedback" ROWS=10 COLS=50></TEXTAREA><P>

<center>
<INPUT TYPE=submit Value="SEND">
<INPUT TYPE=reset Value="CLEAR">
</center>
</FORM>

</body>

-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From cokefiend66 at hotmail.com  Sun Feb  3 16:26:18 2002
From: cokefiend66 at hotmail.com (Bob)
Date: 3 Feb 2002 13:26:18 -0800
Subject: Executing User Entered Code?
Message-ID: <8efc7149.0202031326.72806bda@posting.google.com>

Hi,
I'm writing a network simulator using Python and Tkinter.  I'd like to
be able to allow the user to enter the code for the routing algorithm
in Python and then execute it for each node during the simulation.

Can anyone point me in the right direction here?  I've looked through
most of the docs and haven't been able to find what I'm looking for.

Thanks,
-Bob


From quinn at hurl.ugcs.caltech.edu  Fri Feb  8 20:33:00 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 9 Feb 2002 01:33:00 GMT
Subject: range(seq)?
References: <slrna67foh.dl7.mlh@vier.idi.ntnu.no>
Message-ID: <slrna68v2c.qvm.quinn@hurl.ugcs.caltech.edu>

On Fri, 8 Feb 2002 12:05:37 +0000 (UTC), Magnus Lie Hetland
<mlh at vier.idi.ntnu.no> wrote:
>I just had an idea... I'm sure there are others out there who don't
>particularly like the much-used (and necessary) expression
>range(len(seq))... One proposed solution is to make integers iterable,
>so one can do:
>
>  for x in len(seq): ...
>
>That is one way of doing it, of course (i'm not sure I'm all that fond
>of it, but that's another matter ;)
>
>Then it occurred to me that perhaps it could be done the _other_ way:
>
>  for x in range(seq): ...
>
>I.e. let range() take a sequence (or iterable) as an argument, and
>generate range(len(seq)). (I guess in mathematical terms, if one sees
>a sequence as a function, that domain(seq) would be more appropriate,
>but what the hey ;)
>
>Of course, we have the proposed indices(seq)... But by allowing
>range() to work on iterables, we wouldn't have to add another
>function. And I can't see any significant code breakages (but I'm sure
>there are some... :)

I've always thought that sequences should have a keys() method.  Then we could
replace range(len(seq)) with seq.keys().  The 'indices' generator would then
be spelled [].iterkeys().

It would make it easier to write functions that work on dicts and lists, where
I always wind up writing something like:

if hasattr(a, 'keys'):
    ks = a.keys()
else:
    ks = range(len(ks))
for k in ks:
    ...


I'm sure this has been suggested and rebuffed, though.


From jimmy at retzlaff.com  Sun Feb 17 12:30:13 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Sun, 17 Feb 2002 09:30:13 -0800
Subject: win32com and MS Excel
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F55A3@MAYOR.retzlaff.com>

> One more question: what is going to happen when I build a standalone 
> (using py2exe)? Will the generated file be included and affect the
speed 
> on another machine too?

There are a couple of options. There is a py2exe option to include
everything makepy has ever generated on the build machine, see:

http://groups.google.com/groups?hl=en&selm=9e1a14%24ilo5%241%40ID-59885.
news.dfncis.de

The approach I've used is as follows:

- Find the file generated by makepy (it's in win32com\gen_py)
- Copy that to your application directory and give it a readable name
(e.g., Excel.py)
- Change these two lines:

from win32com.client import Dispatch
xlApp = Dispatch("Excel.Application")

to:

import Excel
xlApp = Excel.Application()


Jimmy



From jim at publishingresources.com  Tue Feb 26 16:12:21 2002
From: jim at publishingresources.com (Jim Abrams)
Date: Tue, 26 Feb 2002 21:12:21 -0000
Subject: Help! IIS and Python
References: <8f8ffe67.0202131245.375fe55d@posting.google.com> <ZtAa8.211294$KM2.8800436@atlpnn01.usenetserver.com>
Message-ID: <Xns91C1A4DAB542Djimpublishingresourc@209.249.90.101>

"Steve Holden" <sholden at holdenweb.com> wrote in
news:ZtAa8.211294$KM2.8800436 at atlpnn01.usenetserver.com: 

> The problem here appears to be that you are probably getting a COM
> object of some sort back as a result of your call to
> Request.ServerVariables(), and when you try to apply the upper()
> function from the string module it complains that you  aren't apssing
> the right kind of object as an argument. 
> 
> If you replace
> 
>     server = Request.ServerVariables("SERVER_NAME")
> 
> with
> 
>     server = str(Request.ServerVariables("SERVER_NAME"))
> 
> for example, you will get a Python string back and all should be
> copacetic. 
> 
> I personally I find it so tiresome to interact with the Scripting
> environment in this way I have now moved away form IIS completely, and
> use Apache or Xitami as a web server. I recently had to go back to
> coding VBScript for IIS, and it was painful.

If, like me, you don't have the option of moving away from IIS, you can 
save yourself alot of headache, time, and resources by writing a nice 
wrapper object, sticking Response inside and using your own. You can then 
add sorely lacking functionality, like str conversion, to it as well as any 
additional shortcuts.

I do it for most of the built ins (Response, Request, Server, Appication, 
Session) and it's been a lifesaver.

Also (in reply to somewhere in this thread) you can do a quick:
from win32com.axscript.client.framework import SafeOutput
_out = SafeOutput(Response)

then use 
print >> _out, Stuff

to get the handy print str conversion. Which Response.Write doesn't do.
Don't use sys.stdout = etc and regular print.

In addition to HTMLgen, which is almost a requirement for Python in ASP, 
check out the string iterpolation module, it also has come it quite handy 
for me. Itpl.py

-Jim



From jason at jvoegele.com  Fri Feb  8 10:13:07 2002
From: jason at jvoegele.com (Jason Voegele)
Date: 8 Feb 2002 07:13:07 -0800
Subject: High-level languages, large projects, GNOME and the .NET controversy
References: <23891c90.0202070501.78df2371@posting.google.com> <3C63A0A7.5060507@thomas-guettler.de>
Message-ID: <91acf731.0202080713.652fa392@posting.google.com>

Thomas Guettler <st-newsgroups at thomas-guettler.de> wrote in message news:<3C63A0A7.5060507 at thomas-guettler.de>...
> Would be nice to use perl libraries in python. But I think that can
> be done easier without CIL (Common Intermediate Language. AFAIK somthing 
> like bytecode in java or .pyc).

Also the upcoming Parrot virtual machine for Perl will make this
easily done (presuming that there is a port of Python to Parrot).

> Would be nice to have the intellisence (dropdown-list if you type 
> object.) of the Visual Studio. Is there already something like this?

This is really hard to do with dynamic languages because the editor
has no reliable way of knowing the type of a variable.

Jason Voegele


From jeff at ccvcorp.com  Tue Feb 12 13:27:16 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 12 Feb 2002 10:27:16 -0800
Subject: I wish.............
References: <mailman.1013443881.3294.python-list@python.org> <3C684D2C.6040307@skippinet.com.au>
Message-ID: <3C695E84.D0A0EE33@ccvcorp.com>


Mark Hammond wrote:

> Just to be clear, I believe that Pippy is a port rather than an
> implementation.  It is a port that has ripped the guts out of the
> implementation (for size purposes) but that is still a port :)
>
> The only implementations I am aware of are CPython, Jython, Vyper and
> Python for .NET.  I also understand only the first 2 of them are
> currently viable.

Hm, then... is Stackless a reimplementation or a port? (Or something else?)
I don't know, myself, how extensively it is rewritten....

(Either way, I'm not certain of its viability, though it sounds like it's
being worked on again...)

Jeff Shannon
Technician/Programmer
Credit International




From jcollins at boulder.net  Tue Feb 19 21:31:23 2002
From: jcollins at boulder.net (Jeffery D. Collins)
Date: Tue, 19 Feb 2002 19:31:23 -0700
Subject: Python and GIS
References: <3C72F52F.5050901@sympatico.ca> <20020219173633.6da14f22.logiplexsoftware@earthlink.net>
Message-ID: <006801c1b9be$4f2564a0$8a7ba8c0@Ron>

OK

Fatal Error: Class format error: Illegal CP reference of type 1
AT+WS46=252;+WS45=0;+IAPPL=2;D
You might also check http://openev.sourceforge.net.

--
Jeffery Collins (http://www.boulder.net/~jcollins)

----- Original Message -----
From: "Cliff Wells" <logiplexsoftware at earthlink.net>
To: "Ramkumar Kashyap" <rkashyap at sympatico.ca>
Cc: <python-list at python.org>
Sent: Tuesday, February 19, 2002 6:36 PM
Subject: Re: Python and GIS


> On Tue, 19 Feb 2002 20:00:31 -0500
> Ramkumar Kashyap wrote:
>
> > Hi,
> >
> > I was wondering if there are any modules for GIS, Mapping, Layering, or
> > hooks to Arcinfo, Mapinfo, etc.
>
> A quick search on Google turned up:
>
> http://www.geocities.com/brucedodson.rm/avpython.htm
>
> --
> Cliff Wells, Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308  (800) 735-0555 x308
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From jason at jorendorff.com  Sun Feb  3 16:43:01 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 15:43:01 -0600
Subject: Elegant solution needed: Data manipulation
In-Reply-To: <12e3779b.0202031211.617e8d7e@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEBMDAAA.jason@jorendorff.com>

Mark wrote:
> No, the only real problem with Steve's code was a need
> to have infix expressions.

It certainly doesn't *require* you to use infix.
Is infix notation bad?

Anyway, text-level transformation of code is error-prone.
But other than that the code looks just fine.  I've included
another stab at the problem below.

## Jason Orendorff    http://www.jorendorff.com/

------------------------------------
# formula.py - Code for applying one or more formulas to tabular data

# Grab the formulas from the "formula_code.txt" file.
# Compile them.
e = open("formula_code.txt", "r")
source_code = e.read()
e.close()
my_code = compile(source_code, "arexpr", "exec")

# Open the file that has all the data.
# Read the first line, which has the names on it.
f = open("formula_data.txt", "r")
names = f.readline().split()

# Make a global namespace and import a bunch of stuff into it.
my_globals = {}
exec "from math import *" in my_globals
exec "from operator import *" in my_globals

# Now perform the calculations.
results = []
for line in f.xreadlines():
    # Grab the data and put it into the locals dictionary.
    my_locals = {}
    data = line.split()
    for i in range(len(names)):
        name = names[i]
        value = eval(data[i], my_globals, my_locals)
        my_locals[name] = value

    # Execute all the formulas.
    exec my_code in my_globals, my_locals

    # Stick the results in the results list.
    results.append(my_locals)
f.close()

# Figure out what the column names are.
first_row = results[0]
if first_row.has_key('output_columns'):
    column_names = first_row['output_columns'].split()
else:
    column_names = first_row.keys()

# Lastly, print rows of output.
print str.join('\t\t', column_names)
for row in results:
    outrow = []
    for name in column_names:
        outrow.append(str(row[name]))
    print str.join('\t\t', outrow)
------------------------------------

formula_code.txt:
------------------------------------
# formula_code.txt - formulas for formula.py

output_columns = 'height weight shoesize X Y Z'

X = add(mul(height, weight), shoesize)
Y = log(height) + log(weight) + 0.44 * log(shoesize)
# ...and in case you want to avoid infix notation...
Z = add(add(log(height), log(weight)), mul(0.44, log(shoesize)))
------------------------------------

formula_data.txt:
------------------------------------
height  weight  shoesize
1   2   3
4   8   12
9   6   3
8   7   3
6   5   4
3   2   1
1   7   8
------------------------------------



From phr-n2002a at nightsong.com  Thu Feb  7 19:42:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 07 Feb 2002 16:42:51 -0800
Subject: High-level languages, large projects, GNOME and the .NET controversy
References: <23891c90.0202070501.78df2371@posting.google.com>
Message-ID: <7xy9i4iz1w.fsf@ruckus.brouhaha.com>

paul at boddie.net (Paul Boddie) writes:
> It occurs to me that there are still many people who choose not to use
> languages of a higher level than C/C++ for development, and it would
> be interesting to know what their reasons are. Of course, writing a
> graphical toolkit *purely* in Python is somewhat ambitious, and there
> are clearly performance issues to be considered, but I still can't
> understand why people would want to write *entire* applications (of
> the nature discussed above) in C/C++ these days.
> 
> Can anyone enlighten me? Perhaps I should have cross-posted to
> comp.lang.c, or whatever. :-)

I worked on one pretty big Java system and although the development
process was much easier than C++ (actually I'd say the coding effort
was comparable, but the debugging effort was greatly reduced), the
resulting Java programs were performance dogs.  That was ok for the
stuff I was doing--it was a high end application and when the programs
ran too slow, we just bought more expensive Sun hardware.  If the
programs had to run on millions of people's low-cost home PC's, I'm
not sure we could have gotten away with using Java.


From aleax at aleax.it  Mon Feb 18 03:45:52 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 08:45:52 GMT
Subject: Why Python is like BASIC ...  java "vs" python
References: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph> <i4iu6ug73jr2lmpblj2e70e1tci043st0l@4ax.com> <YBRb8.80754$3J1.2551057@news2.tin.it> <roy-FCF055.13060217022002@reader2.panix.com>
Message-ID: <4b3c8.68161$om6.1880515@news1.tin.it>

Roy Smith wrote:
        ...
> whether a language is interactive or not.  With Python (or BASIC, or lisp,
> or shell, or ...) if you start it up and type something at it, it types
> something back at you and waits for you to type something else.  Perl, for
> example, doesn't work that way (at least not with the implementations I've

Use:
perl -d -e42

i.e., run the perl *debugger* (-d) on an arbitrary perl 'expression' (-e42)
and you get a 'DB<1> ' prompt that does let you do stuff.  Not quite as
neat as Python's interactive interpreter, but workable, in a typically
Perlish way (i.e., devious enough that you get to feel clever just for
figuring it out, while Python's typically so straightforward that when you
DO "figure out" something you often just feel almost dumb for not
having figured it out sooner -- maybe that kind of 'tiny egoboost' that
comes with Perl's approach is part of its psychological appeal:-).


> far still there.  In this respect, it's very much like Basic (or lisp).

I think Lisp was indeed the first language to come with the read/eval
loop approach -- however, if I recall correctly, it was also easier by
far in Lisp to "save all of my work so far" during an interactive session
(and for sure that was the typical way to work in other languages
based on interactive sessions, such as APL or Smalltalk).


Alex



From syver-en+usenet at online.no  Sun Feb 10 19:49:30 2002
From: syver-en+usenet at online.no (Syver Enstad)
Date: Mon, 11 Feb 2002 00:49:30 GMT
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org> <afe3b146.0202100036.f221c9a@posting.google.com> <ZND98.140690$KM2.6229113@atlpnn01.usenetserver.com>
Message-ID: <uvgd4alnn.fsf@online.no>

"Steve Holden" <sholden at holdenweb.com> writes:
> >
> There-are-other-things-that-one-does-with-one-hand-or-the-other-but-we-don't
> 
> -mention-ly-yours

And then there are those that have to do it with both. 

-- 

Two handed-ly-yours Syver Enstad


From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 19:11:07 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 18:11:07 -0600
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
Message-ID: <007701c1ad10$728b5360$0101010a@local>

Remember... no responses to Rue.  We all need to go cold (on this) turkey.




From peter at engcorp.com  Wed Feb  6 23:16:00 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 06 Feb 2002 23:16:00 -0500
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <m3n0yqt5ib.fsf@chvatal.cbbrowne.com> <1462.801T1396T14014714threeseas@earthlink.net> <a3sbp7$ih2$1@newshost.accu.uu.nl>
Message-ID: <3C61FF80.8ADD0CAE@engcorp.com>

Martijn Faassen wrote:
> 
> So all ended well -- the PSU gets its income and the pretentious semi-competent
> programmers retain their jobs until perpetuity. Occasionally of course a
> person of exceptional insight breaks into the secret, but the PSU has
> made sure they're booed away before they can do any harm.

So what made the PSU fail in this case?  Not that I mean to suggest
the PSU is not entirely in control of the situation but I


From philh at comuno.freeserve.co.uk  Sat Feb 16 21:05:08 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sun, 17 Feb 2002 02:05:08 +0000
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <861yflwo4s.fsf@apo.hmotakef.int>
Message-ID: <slrna6u3uk.phv.philh@comuno.freeserve.co.uk>

On 16 Feb 2002 14:29:55 +0100, Henrik Motakef <henrik.motakef at web.de> wrote:
>Piet van Oostrum <piet at cs.uu.nl> writes:
>
>> BASIC a very good language for teaching????
>> Well, it depends on what you want to teach, I guess.
>
>Or, in the words of Edsger Dijkstra:
>
>"It is practically impossible to teach good programming style to
>students that have had prior exposure to BASIC: as potential
>programmers they are mentally mutilated beyond hope of
>regeneration."

Dijkstra's full of shit.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From sholden at holdenweb.com  Fri Feb 22 07:59:14 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 07:59:14 -0500
Subject: Pythonistas near Rancho Cordova/Sacramento CA
Message-ID: <5grd8.78385$n73.1511733@atlpnn01.usenetserver.com>

On my travels again, teaching next week. If there are any Pythonistas in the
area (especially those with a fondness fopr beer and a knowledge of local
microbreweries) please email me about an evening get-together.

Please note that due to the demise of my laptop all email arrangements must
be completed in the next 48 hours :-(

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py





From s.keim at laposte.net  Wed Feb 13 11:36:13 2002
From: s.keim at laposte.net (sebastien)
Date: 13 Feb 2002 08:36:13 -0800
Subject: Having a mare with Classes
References: <3C6A4BC4.4060409@unite.com.au>
Message-ID: <7da8d8f8.0202130836.427996ea@posting.google.com>

Graeme Matthew <graeme.matthew at unite.com.au> wrote in message news:<3C6A4BC4.4060409 at unite.com.au>...
> Please can someone help, im having an absolute mare, I have Wesley Chuns 
> book on core python programming and I cannot get casses to work, im just 
> using a simple test from the examples in the book.
> 
> OK I HAVE A CLASS PERSON
> 
> class Person:
> 
>      def printName(self):
>          print 'You printed it'
> 
> I HAVE A SCRIPT:
> 
> #!/usr/local/bin/Python2.2
> 
> 
> import sys
> 
> sys.path.append('/www/cgi-bin')
> 
> import Person
> person = Person()
> 
> person.printName()
> 
> I keep on getting this error:
> 
> #! /usr/local/bin/Python2.2
> 
> 
> TypeError: call or non function (type module)
> 
> Please if you can help it would be much appreciated.
> 
> Thanks graeme

In fact you have created a module (the Person.py file) which define
the class Person.

try this:
#!/usr/local/bin/Python2.2
 
 
import sys
 
sys.path.append('/www/cgi-bin')
 
import Person
person = Person.Person()  ## here we call the class Person of the
module Person
 
person.printName()


From logiplexsoftware at earthlink.net  Wed Feb 27 17:44:14 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 27 Feb 2002 14:44:14 -0800
Subject: Thank you Python!!!!!
In-Reply-To: <BC74F8F7E87DD511990C000103E2FEFA2F36@XCHG>
References: <BC74F8F7E87DD511990C000103E2FEFA2F36@XCHG>
Message-ID: <20020227144414.790fa32d.logiplexsoftware@earthlink.net>

On Wed, 27 Feb 2002 17:11:04 -0500
Python wrote:

> 
> A big thanks to Mark Hammond/Andy Robinson (for "Python programming on
> Win32), Guido van Rossum, and all those who help maintain Python.
> 
> I was able to program an script that walks through a directory and
appends
> many thousands of faxes together.  Lowest outside bid:  $10,000.   I did
it
> for the price of Mark's book, an activeX dll that cost $180 (PIL might
have
> worked, but the documentation was lacking - at least for my level of
> experience), and a few hours (ok, maybe a few + a few:).  My previous
> scripting/programming experience:  WinNT login scripts (i.e. almost
none).
> Ideas for future use of Python:  Many!!!

Yes, but your solution, while it may work and do exactly what you want, is
obviously inferior since it wasn't written in a "real" progamming language.

<wink>

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From marklists at mceahern.com  Fri Feb  1 18:48:27 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Fri, 1 Feb 2002 15:48:27 -0800
Subject: comparing DateTime instances
In-Reply-To: <XCF68.111054$8b.4223878@atlpnn01.usenetserver.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEGGKAAA.marklists@mceahern.com>

Steve Holden wrote:
> It is indeed telling you the DateTime values are equal. Did you think of
> trying the equality operator?

Yeah, I did, but it was failing because of rounding errors.  That's why I
felt I had to resort to cmp, since it allows an optional accuracy parameter.
I just got all flummoxed because my brain wanted it to return true for
equality.  This is despite the fact that I've defined __cmp__ for my own
classes in the past and been bitten by the same confusion.  I guess my brain
just doesn't work that way.

> [...]I suspect you are going to enjoy Python.

That's true, I enjoy it immensely.  Thanks for the helpful explanation.

Cheers,

// mark




From rxg218 at psu.edu  Wed Feb 20 14:51:41 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Wed, 20 Feb 2002 14:51:41 -0500
Subject: a question regarding conciseness
Message-ID: <a50umn$1dou@r02n01.cac.psu.edu>

Hi,
  I have a dictionary and would like to loop over the elements in sorted 
order.
The dictionary is of the form:

{'1':'option1', '2':'option2', '3':'option3'}

However when I use:

for i in d.keys().sort():
  do something

I get:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: loop over non-sequence

But I know that d.keys() returns a list to which the sort() function should 
be applicable.

When I do:

k = d.keys()
k.sort()
for i in k:
  do something

it works fine.
Why can't I use d.keys().sort() directly? And is there any other more 
concise (elegent?) way to loop over sorted dictionary keys?

TIA,

-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
Dept. Of Chemistry             | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
CChheecckk yyoouurr dduupplleexx sswwiittcchh..



From pballard at ozemail.com.au  Tue Feb 26 21:37:41 2002
From: pballard at ozemail.com.au (Peter Ballard)
Date: 26 Feb 2002 18:37:41 -0800
Subject: Can I import an arbitary file?
References: <9d5509fa.0202251811.11837b04@posting.google.com> <a5esub$oau$1@peabody.colorado.edu>
Message-ID: <9d5509fa.0202261837.1dc552d2@posting.google.com>

Fernando P?rez <fperez528 at yahoo.com> wrote in message news:<a5esub$oau$1 at peabody.colorado.edu>...
> Peter Ballard wrote:
> 
> > I've written a program (an SRAM generator, if you must know) which
> > has some configuration info (for different SRAM sizes etc.) in a separate
> > file. This configuration file is python code.
> > 
> > What I want to do is import this file. The problem is, this file could
> > be any name at all. What I want to do is parse my command line, and
> > find out the name of my configuration file, and then import the file of
> > that name. Is that possible?
> 
> look at __import__

Thank you. For posterity (and so everyone can have a good laugh
at my coding style), here's what I did...

Previously I used to copy the file (say, sram_8x8.py)
to configtmp.py and have the line "import configtmp" in my code.

Now, the following code appears to be completely equivalent:

##############

import imp

configfilename = "sram_8x8"   # or whatever

fileinfo = imp.find_module(configfilename)
configtmp = imp.load_module("foo", fileinfo[0], configfilename, fileinfo[2])

##############


Thanks also to the other person who replied, but by the time
I saw that I already had my code working :)

Regards,
Peter Ballard


From jason at tishler.net  Fri Feb 15 12:48:23 2002
From: jason at tishler.net (Jason Tishler)
Date: Fri, 15 Feb 2002 12:48:23 -0500
Subject: cygwin and python
In-Reply-To: <3c6d2f5e@news.capgemini.se>
References: <u6oe6n32m9pjfb@corp.supernews.com> <3c6d2f5e@news.capgemini.se>
Message-ID: <20020215174823.GI600@hp.com>

Stephane,

On Fri, Feb 15, 2002 at 05:07:08PM +0100, Stephane Tallard wrote:
> I use python under cygwin bash-rxvt . Trying to use omniorbPy,  I'm facing a
> strange behaviour :
> 
> When executing omniorbPy example ( command line "python  echo_srv.py" )
> script under ugly Command.com, everything is OK.
> 
> When executing the same command line under rxvt , nothing happens: the
> script never ends. Moreover if I add print "Hello" at the beginning of the
> script, "Hello" is never prompted ...
> 
> It seems that cygwin changes the behaviour of python interpretor ....
> 
> Anyone has an idea ? .

Are you using Win32 or Cygwin Python?  I presume the Win32 one, because
it has "display issues" under rxvt.  So do other Win32 apps that write
directly to the console.

When I need to use Win32 Python, I run it directly under bash.  As long
as the CYGWIN environment variable contains "notty", then output should
show up.

I would like to use rxvt for everything too.  Unfortunately, it does
not seem to be possible...

Jason



From mark at mceahern.com  Thu Feb  7 03:30:03 2002
From: mark at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 00:30:03 -0800
Subject: FixedPoint
Message-ID: <NCBBLFCOHHDIKCAFGCFBGEFJKBAA.mark@mceahern.com>

In the FixedPoint module found here:


ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoin
t.py.Z

there seems to be a bug comparing to None like this:

  if f == None:

Demo:

    $ python
    Python 2.2 (#1, Dec 31 2001, 15:21:18)
    [GCC 2.95.3-5 (cygwin special)] on cygwin
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import FixedPoint
    >>> f = FixedPoint.FixedPoint(1.0)
    >>> if f == None:
    ...     print "that's not true!"
    ...
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
      File "FixedPoint.py", line 272, in __cmp__
        xn, yn, p = _norm(self, other)
      File "FixedPoint.py", line 415, in _norm
        y = FixedPoint(y, x.p)
      File "FixedPoint.py", line 216, in __init__
        raise TypeError("can't convert to FixedPoint: " + `value`)
    TypeError: can't convert to FixedPoint: None

This doesn't happen if you use either of the following:

	if f is None:
		...

	if not f:
		...

I ran into this because PyPgSql (v2.0) uses the first method:

    File "/usr/lib/python2.2/site-packages/pyPgSQL/PgSQL.py", line 1434, in
  _quote

      if value == None:

I'm not sure what the best fix is here but one that works is to modify the
__cmp__ method for FixedPoint to anticipate comparison to None.  Here's the
current __cmp__:

    def __cmp__(self, other):
        xn, yn, p = _norm(self, other)
        return cmp(xn, yn)

The fix would look like this:

    def __cmp__(self, other):
        ## To avoid crashing on if x == None and the like:
        if other is None:
            return 1
        xn, yn, p = _norm(self, other)
        return cmp(xn, yn)

Or perhaps it'd be better to muck with __eq__ and __ne__?  Well, anyway,
that mod to __cmp__ works.

Cheers,

// mark




From l.h.allen at verizon.net  Sat Feb 23 21:40:29 2002
From: l.h.allen at verizon.net (Lloyd Hugh Allen)
Date: Sat, 23 Feb 2002 21:40:29 -0500
Subject: [Tutor] unrelated bug issue
References: <LNBBLJKPBEHFEDALKOLCEENLNOAA.tim.one@comcast.net>
Message-ID: <3C78529D.F4E7D754@mail.verizon.net>

Tim Peters wrote:
> Without a bug report, nothing will change.  I often open one for a poster
> when a posted behavior is obviously a bug, or when the poster seems unable
> to articulate their case clearly.  This one doesn't fit either case, so I'm
> leaving it to you:
> 
>     http://sf.net/tracker/?group_id=5470&atid=105470
> 
> Barry may or may not agree it's a bug, but-- honest --he doesn't bite,
> except on Thursdays <wink>.

So I tried Ruby for a bit, got frustrated by their ide and figured that
I'd try it later in a proper linux environment since the windows version
seems to be (I think, anyway) built on CygWin. When I uninstalled Ruby,
IDLE stopped working (no message--just didn't start. Command line python
worked fine). Turned out that this was because Ruby had put a few lines
in the autoexec.bat, one of which said "please use the Tcl library in
the Ruby directory", which no longer existed. Cleared out the
autoexec.bat, now Python works again. Is this a Python bug (ie, should
Python tell the user that the Tcl library doesn't exist or at least give
some sort of feedback), a Ruby bug (that the uninstaller didn't clean up
after itself), or both?



From jeff at ccvcorp.com  Thu Feb 21 14:17:19 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 11:17:19 -0800
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net>
Message-ID: <3C7547BF.463DD2AD@ccvcorp.com>

husam wrote:

> Since I'm a newbie to python, I thought it must
> be possible to automate this process.

Yes, this is a great use for Python.  :)


> I wrote this little script which
> does the job nicely:
>
> listit=os.popen('ls *pdb')

> pdbs=listit.readlines()
> listit.close()

This would probably be better done as

import glob
pdbs = glob.glob('*pdb')

(This will build your file list with significantly less overhead.)


> The problem is how can I save the output to a file?

If you use the popen2() function, instead of popen(), you'll get two file
objects back, one of which is stdin and the other is stdout.

passin, passout = os.popen2('/home/chemistry/profit ')

Now, you can call

passin.write( 'your commands here' )

For getting output, you can take a couple of approaches.  If you just need
to archive everything that is output by your program, a simple

outfile = open('myoutput.txt', 'w')
outfile.write( passout.read() )

should suffice.  You can also read a single line of output at a time ( with
passout.readline() ), but be aware that this call won't return until there's
either something to read, or the pipe closes (i.e., your piped program
terminates), in which latter case it will return an empty string ( '' ).

Jeff Shannon
Technician/Programmer
Credit International





From jgardn at alumni.washington.edu  Thu Feb 28 02:05:39 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Thu, 28 Feb 2002 16:05:39 +0900
Subject: [LONG] Custom debugger using popen.Popen3
Message-ID: <vukf8.382$Az6.1803@news.hananet.net>

I am working on Linux 2.4.14 with Python 2.1.2.

Okay, I got popen.Popen3 working fine for standard python scripts. Example:
test.py=================
print "Oh yeah!"
END CODE=========

debug.py================
from select import select
from fcntl import fcntl
from FCNTL import F_SETFL, O_NONBLOCK
from popen2 import Popen3
from pprint import pformat

# Start the other process
dbg = Popen3("python test.py", 1)

# Non-blocking, please!
fcntl(dbg.fromchild.fileno(), F_SETFL, O_NONBLOCK)
fcntl(dbg.tochild.fileno(),   F_SETFL, O_NONBLOCK)
fcntl(dbg.childerr.fileno(),  F_SETFL, O_NONBLOCK)

# Process input
while 1:
    (r, w, e) = select([dbg.fromchild, ], [], [dbg.childerr, ], 1)
    for i in r:
        fromdbg = dbg.fromchild.read()
        if fromdbg:
            print "Input: %s" % pformat(fromdbg)
    for i in e:
        fromdbg = dbg.childerr.read()
        if fromdbg:
            print "Error: %s" % pformat(fromdbg)
    retval = dbg.poll()
    if retval != -1:
        print "Finished with '%d'" % retval
        break
    else:
        print "Still alive..."
END CODE=========

And here's the finished product:
$ python debug.py
Input: 'Oh yeah!\n'
Finished with '0'
$

Pretty cool, huh? I am in the driver's seat with other processes with only 
this little bit of code. (Python RULEZ!)

Let's upgrade to the new version that starts the debugger.

Let's try it from Bash:
$ python /usr/local/lib/python2.1/pdb.py test.py
> /home/jgardn/projects/debug/<string>(0)?()
(Pdb) 

If I type "q<enter>", it will do this:
$ python /usr/local/lib/python2.1/pdb.py test.py
> /home/jgardn/projects/debug/<string>(0)?()
(Pdb) q
$

Humm, so let's emulate this with my script:
debug.py (v0.2)===========
from select import select
from fcntl import fcntl
from FCNTL import F_SETFL, O_NONBLOCK
from popen2 import Popen3
from pprint import pformat

# Start the other process
dbg = Popen3("python /usr/local/lib/python2.1/pdb.py test.py", 1)

# Non-blocking, please!
fcntl(dbg.fromchild.fileno(), F_SETFL, O_NONBLOCK)
fcntl(dbg.tochild.fileno(),   F_SETFL, O_NONBLOCK)
fcntl(dbg.childerr.fileno(),  F_SETFL, O_NONBLOCK)

# Process input
sent_q = 0 # Whether or not I have told it to stop
while 1:
    w = []
    if not sent_q:
        w.append(dbg.tochild)
    (r, w, e) = select([dbg.fromchild, ], w, [dbg.childerr, ], 1)
    for i in r:
        fromdbg = dbg.fromchild.read()
        if fromdbg:
            print "Input: %s" % pformat(fromdbg)
    for i in e:
        fromdbg = dbg.childerr.read()
        if fromdbg:
            print "Error: %s" % pformat(fromdbg)
    for i in w:
        dbg.tochild.write('q\n')
        print "Wrote 'q\\n'"
        sent_q = 1
    retval = dbg.poll()
    if retval != -1:
        print "Finished with '%d'" % retval
        break
    else:
        print "Still alive..."
END CODE=========

The excitement is building... what will happen? Will it work? Will it not? 
LET'S FIND OUT!

$ python debug.py
Wrote 'q\n'
Still alive...
Still alive...
Still alive... (in 1 second intervals.)

Sure enough, the two processes are alive.

My question is this: How come? Why? Why doesn't the debugger print anything 
out to stdout? Is it "readline" that is screwing this up? Why would 
readline screw it up? It doesn't make any sense... any help would be 
greatly appreciated.

If I can't get this figured out, I will probably have to write a server 
that is based off of bdb and have the client connect to it...

Jonathan



From tim at worthy.demon.co.uk  Sat Feb  2 09:52:44 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Sat, 02 Feb 2002 14:52:44 GMT
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net> <313g3a.d22.ln@wombat.san-francisco.ca.us> <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk> 
 <37d27e024b.tim@worthy.demon.co.uk> <PLPSljAQX$W8Ewpx@jessikat.fsnet.co.uk>
Message-ID: <ac5588024b.tim@worthy.demon.co.uk>

In message <PLPSljAQX$W8Ewpx at jessikat.fsnet.co.uk>
          Robin Becker <robin at jessikat.fsnet.co.uk> wrote:

> In article <37d27e024b.tim at worthy.demon.co.uk>, Tim Howarth
> <tim at worthy.demon.co.uk> writes
> >In message <6vTfsIAio7W8EwNp at jessikat.fsnet.co.uk>
> >          Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
> >
> >> well if I could get the win32com extensions to change passwords on
> >> different machines for me I would really like that
> >
> >Why win32com rather than win32net (where it's easy)?
> >
> well I assumed we have to do it through DCOM,

You need to use com to attack the clever new features in Active
Directory Win2k+ but plain NT account info win32net is simple.

Best is to download the samples for the win32all package

(not sure, can't remember URL, I have file as - PythonWin32_samples.zip)

or buy Mark Hammonds Python on Win32 book,

or use the help with win32allxxx (and win32api help)


> but if it's easy with win32net what's the easy way?

import win32net

userinfo=win32net.NetUserGetInfo('','tim',2)
userinfo['password']='secret'
win32net.NetUserSetInfo('','tim',2,userinfo)


'' is server name, in this case server it is run on (no need for name)
2 is data level - what to get/set


-- 
___
 |im    ---- ARM Powered ----


From cliechti at gmx.net  Thu Feb 14 17:46:43 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 14 Feb 2002 23:46:43 +0100
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <Xns91B5F26DFF836cliechtigmxnet@62.2.16.82>

Scirious <Scirious at Scirious.com> wrote in 
news:Xns91B5CBEF0A103Scirious at 140.99.99.131:
> I've been learning JAVA for a while, but now I've heard about 
> Python.
> Any coments concerning these two languages?

that's one of the questions that is good for a flamewar :-O
anyway here some answers biased towards python:

Python programms are shorter.
The Debug Cycle is shorter.
You can allways try out things interactively.
Python is FUN programming.

Here are some nice slides that illustrate some differences (PPT):
http://64.78.49.204/pub/eckel/LovePython.zip
(from http://www.mindview.net/Books/Python/ThinkingInPython.html)

Anyway you might have a look at www.jython.org. its python implemented 
in java. you can access all java classes from within python (and vice 
versa). you can e.g. programm a swing or awt gui with python, its 
realy cool.

-- 
Chris <cliechti at gmx.net>



From grante at visi.com  Sun Feb  3 23:23:23 2002
From: grante at visi.com (Grant Edwards)
Date: Mon, 04 Feb 2002 04:23:23 GMT
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net> <MPG.16c760cdaa8ee3639896d5@newnews.dircon.co.uk> <PSi78.6815$3E5.522096@newsread2.prod.itd.earthlink.net>
Message-ID: <slrna5s35r.h88.grante@isis.visi.com>

In article <PSi78.6815$3E5.522096 at newsread2.prod.itd.earthlink.net>, wrbt wrote:

> Also - I work for a company that tracks vehicles on the internet using GPS .

Cool.  I keep hearing about "on ramps" for the 'net, but I've never been
able to get my car to go up one of them.  How _do_ you get a vehicle up the
ramp onto the 'net?

-- 
Grant Edwards                   grante             Yow!  Did you find a
                                  at               DIGITAL WATCH in YOUR box
                               visi.com            of VELVEETA??


From justin at iago.org  Wed Feb  6 09:56:46 2002
From: justin at iago.org (Justin Sheehy)
Date: Wed, 06 Feb 2002 09:56:46 -0500
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
	<slrna60fur.6sf.philh@comuno.freeserve.co.uk>
	<75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com>
	<slrna61344.8s1.philh@comuno.freeserve.co.uk>
Message-ID: <x73d0ezmj5.fsf@ra.iago.org>

philh at comuno.freeserve.co.uk (phil hunt) writes:

> And what if I decide that for my particular application, that's just
> the behaviour I want?

I wonder why your reaction is so strong here.

Mr. Kraska has clearly stated that this is an experiment.  It certainly
isn't something that is about to be shoved down anyone's throat.  If
your particular application doesn't want to behave as he describes,
just don't get involved with the experiment.

As an aside: while many people think that purely hygienic macros are
lacking in an important sort of power, they are a well-researched
topic and not a bad place to start if implementing a macro system from
scratch for experimentation purposes.

> Or do you think that you know better than all the users of your
> macro system, even though you've not met them and don't know
> anything about the software they are writing?

At this point, he does know that, since he is presumably the only user.

What is the problem here?

>>>   do:
>>>      dosomething()
>>>      while i <10                                                                                        
>>
>>Correct. I'm not quite sure how to handle that one.
>
> It doesn't need handling; python does it OK already.

Many people have asked for a loop-and-a-half construct before.  This
looks like it could be a simple expression of that concept.  I
personally don't want to use it, but that doesn't mean it is worthless.

>>But of course macros themselves significantly erode simplicity, making
>>them _per se_ unpythonic, wouldn't you say? :)

> Any dogmatic pronouncement that a particular programming technique (such as
> macros) is inherently good or bad is just plain brain dead.

I saw no such claim of goodness or badness.  A claim was made
about whether macros are Pythonic, and it seemed to be a justified one.

-Justin

 





From jason at jorendorff.com  Mon Feb 11 15:22:25 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 11 Feb 2002 14:22:25 -0600
Subject: Global Variables in Modules
In-Reply-To: <20020211111025.63f4c956.jsc-news@bluesine.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIENHDBAA.jason@jorendorff.com>

Jeff Clement wrote:
> What I'm actually doing is running a code block in an exec block 
> and declaring a bunch of globals for that exec.  The code block 
> imports some modules and I need those to be able to see the 
> global variables.  I can make this work by adding them to 
> __builtins__ but that doesn't seem right somehow :)

The best thing is to make a config.py, and have
everyone just "import config".  All the code for setting up those
globals should be inside config.py.  You can probably avoid exec
entirely (which is for the best).

Failing that, you can have an empty config.py and do:

  # test.py
  # Import the empty module and populate it using exec.
  import config
  exec "...huge string of globals and imports..." in config.__dict__

  # test_mod.py
  import config
  print config.A

Worst case, you can just:

  # test_mod.py
  import __main__
  print __main__.A

## Jason Orendorff    http://www.jorendorff.com/



From SBrunning at trisystems.co.uk  Wed Feb 20 10:52:17 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 20 Feb 2002 15:52:17 -0000
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <31575A892FF6D1118F5800600846864DCBCCC3@intrepid>

> From:	Russell Nelson [SMTP:nelson at crynwr.com]
>  > Assignments and expressions are different things in Python...
> 
> Indeed, and that's a good thing.  But I'm proposing an alternate
> syntax for the if statement, specifically:
> 	"if" <variable> "=" <expression":.
 
So the assignment (the <variable> "=" <assignment> bit) looks to see if it's
in an 'if' or 'while', and becomes an becomes an expression if it is. Sorry,
but that's ugly. *And* it would allow the exact error which the
assignment/expression thing was introduced to prevent.

Thanks, but no thanks.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From hmartires at cultalg.pt  Mon Feb 25 10:32:10 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 15:32:10 -0000
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de> <vRre8.603$9l1.7321620@newsserver.ip.pt>
Message-ID: <gMse8.611$GQ1.7493742@newsserver.ip.pt>

Sory...
I found it ;)
These are object methods.
Problem solved.


"Hugo Martires" <hmartires at cultalg.pt> wrote in message
news:vRre8.603$9l1.7321620 at newsserver.ip.pt...
> Where are the method's ?? (array.tostring() array.fromstring() )
> I import the array module, but i don't have them.
> I'm using python 1.5
>
> "Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
> news:j4bsedlnai.fsf at informatik.hu-berlin.de...
> > "Hugo Martires" <hmartires at cultalg.pt> writes:
> >
> > > How can i count the number of bytes of:
> > > --> List []
> > > --> array('d',[])
> > >
> > > There is any function that return's the number of bytes ?
> >
> > For the list, no. For the array: len(arr.tostring()) will give a rough
> > estimate. The actual memory consumption is higher, due to Python and
> > malloc overheads.
> >
> > Regards,
> > Martin
> >
> >
>
>




From cbrown at met.co.nz  Tue Feb  5 00:12:19 2002
From: cbrown at met.co.nz (Colin Brown)
Date: Tue, 5 Feb 2002 18:12:19 +1300
Subject: EDF scheduler
Message-ID: <eTJ78.14206$N31.668446@ozemail.com.au>

Hi

I am trying to implement a simple scheduler in Python where the deadline
times of (non-prioritised) entries as entered are not necessarily in
chronological order. From what I see on the Web this is best described as an
"Earliest Deadline First" scheduler.

The "sched" module [using timefunc of time.time and delayfunc of time.sleep]
does not appear to do this since a subsequent entry at an earlier schedule
time does not replace the non-completed next entry but is activated after
it. There does not appear to be a (documented) way of finding the currently
active entry so it can be cancelled and requeued thus allowing the earlier
entry to action at the right time.

Here is a test script that demonstrates sched behaviour:
!---------------------------------------------------------------------
import Queue
import sched
import threading
import time

class skd(threading.Thread):

    def __init__(self,outq):
        self.outq=outq
        self.myq=Queue.Queue()
        self.schd=sched.scheduler(time.time,time.sleep)
        threading.Thread.__init__(self)
        self.start()

    def reply(self,arg):
        self.outq.put(arg)

    def enter(self,entry):
        return self.schd.enter(entry['secs'],1,self.reply,(entry['data'],))

    def run(self):
        while 1:
            self.schd.run()
            time.sleep(0.1)

oq=Queue.Queue()
bks=skd(oq)
bks.enter({'secs': 11, 'data': 'eleven'})
bks.enter({'secs': 5, 'data': 'five'})
bks.enter({'secs': 3, 'data': 'three'})
for sec in range(20):
    time.sleep(1)
    print sec+1
    if sec == 6:
        bks.enter({'secs': 2, 'data': 'nine'})
        bks.enter({'secs': 1, 'data': 'eight'})
        bks.enter({'secs': 8, 'data': 'fifteen'})
    if oq.empty() == 0:
        print oq.get()
!---------------------------------------------------------------------------
-

produces output:
1
2
3
three
4
5
five
6
7
8
9
10
11
eight
12
nine
13
eleven
14
15
fifteen
16
17
18
19
20
!---------------------------------------------------------------------------
--

Here is my understanding of the operation. After the thread is created, the
schedule queue is empty so the scheduler.run() falls through and the thread
hits the 100ms sleep. During this period the 3 sec, 5 sec, 11 sec entries
are inserted in non-chronological order. When scheduler.run() next executes
in the thread they are sorted so the 3 & 5 second entries are actioned.
After 7 seconds 3 more entries are added (for 8, 9 & 15 seconds) but the 11
second entry is currently being processed. However the internal queue
entries have been put in the sorted order of 8,9,11 & 15 seconds. At 11
seconds, the next scheduler queue entry, "8 secs" is returned. I would have
expected to have seen 9 & eleven occur in the next interval together but I
guess the time/scheduler function are rounding things up. The "15 secs"
appears as expected.

I would be obliged if someone can suggest an easy solution to my problem.
There may be hundreds of queue entries but I will be working to only 1
minute resolution so the implementation should not be too demanding of
resources.

Thanks
Colin Brown
PyNZ





From cstrickland at surry.net  Tue Feb 19 22:32:40 2002
From: cstrickland at surry.net (Charles W. Strickland)
Date: Wed, 20 Feb 2002 03:32:40 GMT
Subject: cursor control
Message-ID: <3c731781.5352364@nntp.surry.net>

Is it possible to control cursor positioning in python without using
TKinter?  I wanted to write a program that required it, but could not
find anything anywhere to explain how to do it.  There were a couple
references awhile back, but I never did see a definitive answer.  I
finally wrote the program in C - what a mess!


From tim.one at home.com  Wed Feb 13 02:31:45 2002
From: tim.one at home.com (Tim Peters)
Date: Wed, 13 Feb 2002 02:31:45 -0500
Subject: photo id ?!?!
In-Reply-To: <04a501c1b40c$7d17bf00$0201a8c0@josiah.dalkescientific.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEODNLAA.tim.one@home.com>

It's curious that the FAA doesn't require photo ids (see
http://www/faa.gov/).  I've read that the airlines made up that rule, not
for security, but to make it harder for person A to use a cheap ticket
obtained by person B.  Is that true?  Beats me, but my cynical half can't
not believe it <wink>.

but-in-the-end-everything's-guido's-fault-ly y'rs  - tim




From James_Althoff at i2.com  Fri Feb  8 14:31:27 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 8 Feb 2002 11:31:27 -0800
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <OF8ADB5008.67C7DB4D-ON88256B5A.006A70C1@i2.com>

[Aahz]
>Haven't figured out a good/cheap way to put more than two monitors
>on a desk, particularly when both are 17" or larger.  It's hard
>enough stuffing *one* monitor on most crapola office desks.

Fry's Electronics has been selling a *really* nice 18-inch flat panel
display by NEC.  Same amount of actual display area as many 20 or 21 inch
video monitors and totally flat, lightweight, space saving, etc.  The
prices have come down remarkably over the past year.  I bought 4 of them
for my house over the holidays (yes, I *have* been accused of going
overboard on occasion, thank you. <wink>).

Jim




From deathtospam43423 at altavista.com  Mon Feb 11 18:12:00 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 11 Feb 2002 23:12:00 GMT
Subject: Help: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com>
Message-ID: <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254>

The example at http://www.devshed.com/Server_Side/Python/CGI/page3.html is 
supposed to produce the text in my browser "Hello, Python", but it doesn't 
do a thing.

Instead, using the Xitami personal server, I get the error 

>Internal error: your request was unsuccessful
>Cannot create CGI process - program not found

Here is the text of the example...

#!/usr/bin/python

# Tell the browser how to render the text
print "Content-Type: text/plain\n\n"

print "Hello, Python!"  # print a test string


I had a theory that the first line was the problem. The tut says that it's 
used by Unix and Linux servers to tell the OS to initiate Python. I'm using 
Win98, btw. I tried changing it to the following......

>#path = c:/Python22/Python.exe

and then directing to the file location on my HD ...

>#path = d:/Docs/Web/cgi-bin/index.py

... but in both cases the browser merely gave me the option of downloading 
the file, or opening it using Python.

So what am I missing? 


From callistos at t-online.de  Fri Feb 22 17:13:20 2002
From: callistos at t-online.de (Georg Bremer)
Date: Fri, 22 Feb 2002 23:13:20 +0100
Subject: qtbindings wont work >:-( etc.
Message-ID: <a56fg4$7ph$00$1@news.t-online.com>

Hi

System:
SuSE 7.3 (i386), all Python related stuff came from originally SuSE
7.3 rpm's. Python is version 2.1.1, sid is installed, qt and pyKDE.

hello.py is just a little test, if Python finds the modules.


After I found out that PYTHONPATH was just EMPTY, I set the pathes
manually. Now, Python finds qt BUT then:

  File "hello.py", line 4, in ?
    from qt import *
  File "/usr/lib/python2.1/site-packages/qt.py", line 44, in ?
    import libqtc
ImportError: /usr/lib/python2.1/site-packages/libqtcmodule.so:
undefined symbol: PyUnicode_FromUnicode

So what?!

And Python still doesnt find kdecore and kdeui.

PYTHONPATH =

/usr/lib/python2.1/
/usr/lib/python2.1/plat-linux2/
/usr/lib/python2.1/lib-dynload/
/usr/local/lib/python2.0/site-packages/
/usr/local/lib/site-python/
/usr/lib/python2.1/site-packages/
/usr/lib/site-python/

So where are the f***** kde modules...

Sorry, Im sick and tired after the last weeks working with linux for
the first time. It seems there's nearly NO PROGRAM that works at once
on Linux. Every and every time I get errors like this and have to work
hours to solve them. - I just want to learn python. Im enthusiastic
but this p***** me off.

I will now go to bed (CET) and sleep a while to calm down. Please help
me if you can. I really want get it to work and start learning python.

If you're a german like me maybe it is more practical to talk via
email, reply to sender then.

I want to apologise the words I used. GN8



From whoRunBartertown at masterBlaster.net  Sun Feb  3 17:31:11 2002
From: whoRunBartertown at masterBlaster.net (wrbt)
Date: Sun, 03 Feb 2002 22:31:11 GMT
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net> <MPG.16c760cdaa8ee3639896d5@newnews.dircon.co.uk>
Message-ID: <PSi78.6815$3E5.522096@newsread2.prod.itd.earthlink.net>

C/C++, java, vb, and the scripting twins (vbscript and javascript) have put
the bread on my table.

I've read things about python in the past but never took the plunge until a
couple months ago. I got ActiveState installed on our server at work and
have been busily converting older stuff and writing newer stuff in python. I
love being able to include scripts without dancing the ssi dance or fooling
with DLL hell. I also love the python language, especially for web apps
since 99% of the time a function used by an active server page is going to
return some combination of a list/hash/tuple. Clean syntax (well the
__whatever__ is somewhat funky) real exception handling and the
aforementioned namespace module resolution have made it a joy to learn and
work with.

I've encountered all the standard "newbie" questions (why the hell doesn't
this method scope my global as global? :)) and found answers. I suspect as I
get more hardcore in the future I'll be posting questions to you fine
people. I've still not dived (dove?) into threading or the socket libraries,
I'm looking forward to it.

Also - I work for a company that tracks vehicles on the internet using GPS .





From bkelley at wi.mit.edu  Wed Feb  6 10:31:11 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Wed, 06 Feb 2002 09:31:11 -0600
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de>
Message-ID: <3C614C3F.7000508@wi.mit.edu>

MDK wrote:

>I need to convert a list into a list of characters.
>
>For example:
>
>("cat",5,['dog',[3,3,1]],"zoo")
>
>Would become:
>
>('c','a','t',5,'d','o','g',3,3,1,'z','o','o')
>
>Any information would be appreciated.
>
>Thanks
>
>
It's an interesting problem.  A particularly slow solution follows. 
 Interestingly, it purposely uses the maximum recursion error to keep 
the code simpler and not have to do any type checking.  Given this fact, 
it's faster than I expected :)

def _delve(s, target):
    """(s, target)->recursively split elements of s and
    append each element to the target list
    Warning, purposely uses maximum recursions
    for simplicity!"""
    try:
        for data in s:
            _delve(data, target)
    except TypeError:
        target.append(s)
    except RuntimeError:
        # catch maximum recursion errors
        # this occurs when trying to split a character
        # string like "s"
        target.append(s)

def delve(input):
    """(input)->recursively split input
    i.e.
    ('cat', 5, ['dog', [3, 3, 1]], 'zoo')
    will reult in
    ('c', 'a', 't', 5, 'd', 'o', 'g', 3, 3, 1, 'z', 'o', 'o')

    warning, purposely uses maximum recursions!
    """   
    target = []
    _delve(data, target)
    return tuple(target)

print delve(("cat",5,['dog',[3,3,1]],"zoo"))

Brian Kelley
Whitehead Institute for Biomedical Research



From schneida at seiu.org  Thu Feb 28 13:32:54 2002
From: schneida at seiu.org (Anders Schneiderman)
Date: 28 Feb 2002 10:32:54 -0800
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]> <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com> <54abe19d.0202271315.4e699ed0@posting.google.com> <b7yVwiC1Cff8EwBv@[127.0.0.1]>
Message-ID: <54abe19d.0202281032.77cceb54@posting.google.com>

Chris <nospam@[127.0.0.1]> wrote in message news:<b7yVwiC1Cff8EwBv@[127.0.0.1

> Can you do ordinary 2D graphics stuff without VPython - simply using the 
> ordinary Python?

Other folks have replied re: VPython, but one more toy you might check
out is Alice, at http://www.alice.org/.  Alice is a tool that lets you
create interactive 3D graphics, and it uses Python as its scripting
language.  I've never got the chance to play w/ it, but I've heard
people rave about how much fun it is.  Might be worth a look.

Thanks,
Anders


From tim at vegeta.ath.cx  Wed Feb 13 06:55:00 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Wed, 13 Feb 2002 11:55:00 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a46uqm$gc7$1@peabody.colorado.edu> <mailman.1013399900.24051.python-list@python.org> <slrna6f2ff.ugb.tim@vegeta.ath.cx>
Message-ID: <slrna6kl1m.is0.tim@vegeta.ath.cx>

Tim Hammerquist <tim at vegeta.ath.cx> graced us by uttering:
>> On Sun, 10 Feb 2002, Fernando P?rez wrote:
>>> Indeed, it's just the Spanish form of the 'ist' English suffix (pianist,
>>> violinist, etc.) It simply means 'who performs an activity'.
>> 
>> Could be from a number of languages; I always assumed quasi-Esperanto:
> 
> This may be very likely, but the limit of my Esperanto knowledge is Red
> Dward episodes...

That was "Red Dwarf".

>> Of course, in either Spanish or Esperanto it would more commonly be
>> "Pythonisto" anyway, so who knows.
> 
> No idea about Esperanto, but Spanish ends in -ista, even when masculine.
> eg: dentist (English) --> /dentista/ (Spanish).

I could've explained that better, too.  =)

> HTH
> Tim Hammerquist

Tim Hammerquist
-- 
$ echo `find /mnt/windows | wc -l` wasted directory entries.


From claird at starbase.neosoft.com  Wed Feb 13 10:41:00 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 13 Feb 2002 09:41:00 -0600
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com> <mailman.1013544745.1018.python-list@python.org>
Message-ID: <30037B42640ABBB5.85B6210185167959.6076E8759BE1AED1@lp.airnews.net>

In article <mailman.1013544745.1018.python-list at python.org>,
Wayne Pierce  <wayne at mishre.com> wrote:
>
>What is your target platform?
>
>If you are only working with Windows, you can use MFC (sorry, had to put 
>that in there...).  There is a list at Cameron Laird's site[1], I don't 
>know how current it is though
			.
			.
			.
Not particularly (current), but whenever someone agitates me
personally, I stir into action.  So:  the general procedure
is to read what's there, and, if you find something missing/
confusing/indecipherable/..., just write, and I'll make a
change.
-- 

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


From cfelling at iae.nl  Thu Feb 14 16:52:46 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 14 Feb 2002 22:52:46 +0100
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com> <1013641215.358571@yasure> <a4f29e$bhm$1@animus.fel.iae.nl> <a4h039$2r2q$1@nntp6.u.washington.edu>
Message-ID: <a4hbje$cuq$1@animus.fel.iae.nl>

Donn Cave <donn at u.washington.edu> wrote:
> Quoth Carel Fellinger <cfelling at iae.nl>:
...
> | I've reread the fine manuals, but can't find a clue hinting at this
> | interesting behaviour.

> I guess I should read the manuals some time.  I wonder if they explain
> why os.popen2 reverses the order of popen2.popen2's returns.

It says that os.popen2 uses a more intuitive order and especially for
popen3 this is true.  Compare

  popen2.popen3  ->  (child_stdout, child_stdin, child_stderr)

  os.popen3      ->  (child_stdin, child_stdout, child_stderr)

Besides in the os.popen* case the index in the result tuple reflexes
the unix file handle better.
-- 
groetjes, carel


From jerome at cortex.unice.fr  Mon Feb  4 07:34:56 2002
From: jerome at cortex.unice.fr (Jerome Alet)
Date: Mon, 4 Feb 2002 13:34:56 +0100
Subject: [ANN] PDFMap 1.50 is out
Message-ID: <Pine.LNX.3.96.1020204133407.17533G-100000@cortex.unice.fr>

Hi,

I'm pleased to announce the availability of PDFMap v1.50.

PDFMap is a GPLed Python module and command line tool which intensively
uses the ReportLab toolkit (http://www.reportlab.com) to allow people to
create high quality maps in the PDF format. 

Objects can be represented on maps using different predefined shapes in
user defined colors, or images, and are correctly scaled and oriented
according to each object's dimensions and orientation wrt the North.

Each object can be made clickable, at least in Adobe's Acrobat Reader,
allowing you to make back-links to any web enabled database.

The new version changes are :

	- PDFMap now works as a CGI script too, allowing you to use it as
          a simple map server. A default HTML form is included, adapt it
          to your own needs. 

	- Image shapes can now be downloaded from the Net at runtime :
	  just put double-quoted urls in the configuration file to use
	  this feature.

	- The complete module documentation, generated with HappyDoc, is
	  now included.

        - Margins can now be set in centimeters (default), inches, or
	  points.

	- a .bat script is now included for Windows users. It was
	  successfully tested against Win98 + ReportLab Demo 1.11

You can download PDFMap or see some samples and screenshots, from :

	http://pdfmap.sourceforge.net/

A mailing list and the CVS tree are available from :

	http://sourceforge.net/projects/pdfmap/

Commercial support is available from : contact at librelogiciel.com
(website is currently down for heavy maintainance but email works)

Comments are very welcome.

Thank you for reading.

Jerome Alet - alet at unice.fr



From ulrich.pfisterer at rmb.co.za  Mon Feb  4 10:57:04 2002
From: ulrich.pfisterer at rmb.co.za (Ulrich Pfisterer)
Date: Mon, 04 Feb 2002 17:57:04 +0200
Subject: working with files with the same extenstion
Message-ID: <3C5EAF50.80004@rmb.co.za>

Can anyone please tell me how I would go about doing the following.
I would like to write a script that would take in a command line 
parameter "*.txt". The script would then loop through every file with a 
"txt" extension and something with it.

I am brand new to python (I started using it 2 days ago) so I dont know 
if there is a module that does already this, or whether I have to do 
all the wildcard matching etc myself.

Any help would be much appreciated



From sholden at holdenweb.com  Sun Feb 10 13:08:16 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 10 Feb 2002 13:08:16 -0500
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a43due.3vv9a5r.1@kserver.org> <8852d33c.0202100034.16f1b47b@posting.google.com> <a4580e.3vsjs5j.1@kserver.org>
Message-ID: <9Gy98.138220$KM2.5966121@atlpnn01.usenetserver.com>

"Sheila King" <usenet at thinkspot.net> wrote ...
> On 10 Feb 2002 00:34:45 -0800, wookiz at hotmail.com (wooks) wrote :
>
> > Sheila King <usenet at thinkspot.net> wrote ...
>
> > > I recommend:
>
> > > (2) Core Python Programming by Wesley Chun
> > > (not for beginning programmers...aimed at someone with some
programming
> > > experience...addresses the OO points and many features of the language
> > > quite well)
> >
> > Yes. Thats the one that all the reviews seem to point to. If I
> > remember rightly the copy in the bookshop I went to was rather dated.
> > Will investigate further.
>
> I thought there was only one edition of that book??? Mine is copyright
> 2001. If it based on Python 1.5.2 but has remarks and appendices for
> 2.0. The only book I know of that assume more than Python 2.0 is the
> Python 2.1 Bible (which I also have...and recommend only as a reference
> for someone who already knows Python). Most of the recent ones are
> either based on 1.5.2 with appendices referring to changes in 2.0, or
> they are based on 2.0. That's as recent as you're going to get.
>
Well, if I might be allowed to blow my own trumpet, "Python Web Programming"
is firmly based on 2.0, but mentions known futures and includes Andrew
Kuchling's "What's New in Python 2.1" and an almost-final "What's New in
Python 2.2" as an appendix (for which I am very grateful to Andrew). I
imagine that any recent books will also similarly include at least a nod
towards 2.1/2.2.

Broadly speaking you should expect anything based on 1.5.2 to be revised to
2.0 or later (like Lutz's "Programming Python" has been). If no revision
takes place, perhaps either the book inquestion was rather late to the 1.5.2
market, or hasn't sold well enough to justify a new edition.

One of the problems for Python authors, if my experience was typical, is
that recently the language has been developing pretty dynamically, and it's
difficult (given the nature of the publishing process) to track the changes
accurately. The 2.1 and 2.2 releases have introduced significant new
features into the language.

Fortunately the number of available books has increased recently, and while
there are some mediocre tomes there are also some very good ones. One useful
bibliography, although getting a little long in the tooth right now, can be
found at

    http://www.amk.ca/bookstore/python.html

Wooks, from everything you've written it appears as though you have all the
right skills to benefit from Python, and you just started into the Win32
specifics a bit too soon to feel entirely comfortable with the rest of the
language. With one of the more advanced intriductory texts, or even just by
interaction with this newsgroup, I'm sure you'll soon get where you want to
be.

hoping-for-chance-to-write-a-3.0-edition-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jgardn at alumni.washington.edu  Tue Feb 19 18:31:44 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Wed, 20 Feb 2002 08:31:44 +0900
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56>
Message-ID: <G5Bc8.1$Az6.364@news.hananet.net>

Lutz Schroeer wrote:
> Jumping from one script to the other I need to remember at least the
> user's name. The W3C recommends using cookies (RFC 2109). Unfortunately
> many people don't like cookies and filter them using WebWasher or a
> personal firewall.
> 
> Is there any other technique for keeping the user's data while he is using
> my site without installing any large suite as Zope or the like? I've read
> an article about utilizing soft links but I didn't quite get the point.
> 

A technique that I've tried and has been successful is to use some kind of 
temporary UID. You have to be sure and pass it as a parameter to each 
script the guy accesses however. You also have to have some kind of 
expiration (within hours) and silently replace it as they progress.

The site had a login page. As soon as you logged in, it would set a cookie 
and add the uid= param. It would always try to read the cookie, but failing 
that, it would revert to the information connected to the uid. After a few 
hours, it would ask the guy to log in again, but I found that this was rare.

There are security problems. I tried to force the guy to be coming from one 
unique IP address, but I found that too many places have firewalls or 
something that change the IP address of the user each time they access the 
page, so I had to abandon that.

As with any method, it involves a fair bit of experimentation.

The easy way is still cookies. Tell the use to turn on cookies, tell them 
what you are storing on the cookie, tell them it is temporary, or whatnot, 
but you end up using cookies as the most reliable way of keeping state.

Jonathan


From martin at v.loewis.de  Wed Feb 27 04:05:02 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 10:05:02 +0100
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3c7b9d52.5730880@news.verizon.net>
Message-ID: <m3y9hf1eip.fsf@mira.informatik.hu-berlin.de>

nick.collier at verizon.net (Nick Collier) writes:

> I now have access to the actual error message and seems as if the
> problem is with argument. So the error messages look like:
> 
>   LL1 warning in argument: NAME is  start & successor of deletable
> structure

With "argument" being

argument: [test '='] test

Apparently, when it sees a NAME, it cannot determine whether that
belongs to the first test or the second one. Since those parts of the
production are not non-terminals in themselves, there is no ambiguity;
there would be an ambiguity if the grammar read

argument: kwargument | simpleargument
kwargument: test '=' test
simpleargument: test

In that case, NAME would be in the firstsets of both kwargument and
simpleargument. If your parser generator allows you to associate
semantic actions with each "deletable structure", or creates
distinguished abstract syntax for deletable structures (instead of
creating them just for non-terminals), it would effectively rewrite
the grammar to the second form (*), making it ambiguous.

> The same errors are repeated in arglist, picking up on the errors from
> argument.  Coco\R does create a parser here, so I guess I'm whether it
> will work, after all the python one certainly works!

Notice that this is a warning only: *if* you'd attach semantic actions
in unacceptable ways, you'ld get problems.

Regards,
Martin

(*) More precisely, the rewrite would read

argument: opt_kw test
opt_kw: | test '='

in which case NAME would live both in the firstset and the followset
of opt_kw, which is an ambiguity since opt_kw is not epsilon-free.


From aahz at panix.com  Mon Feb  4 18:52:37 2002
From: aahz at panix.com (Aahz Maruch)
Date: 4 Feb 2002 15:52:37 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <a3n3q9$n1i$1@panix2.panix.com> <7xg04gon8f.fsf@ruckus.brouhaha.com>
Message-ID: <a3n6s5$515$1@panix2.panix.com>

In article <7xg04gon8f.fsf at ruckus.brouhaha.com>,
Paul Rubin  <phr-n2002a at nightsong.com> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>>Paul Rubin, who doesn't know how to use attributions, wrote:
>>>
>>>  a) If I put together a 1.5.2 bugfix release and submit it, will it go
>>>     up on Python.org with a link on the front page like the current
>>>     1.5.2 link?
>> 
>> In theory, yes.  In practice, you'll have document your work so
>> carefully that it probably won't be worth it.  I don't think anyone will
>> commit to making such a release available in the absence of existing
>> work to be submitted.  Have you read PEP 6 yet?
>
>That's ok, it's just a theoretical question.  The other question
>was whether a 1.2.1 bugfix release would also go up, since 1.2 is
>not linked from the front page now.  Versions from 1.5.2 to 2.2 are
>currently linked from the front page.  Versions older than 1.5.2 are
>not.  That reinforces the impression that 1.5.2 is still current while
>versions like 1.2 are ancient history.

This is Python.  We don't care much about theory, except where it
intersects with useful practice.  I don't see any point in even starting
a discussion about 1.2 until someone proves that zie is willing to start
*and* finish the job of creating a full 1.5.2 bugfix release.  (Note that
the release would likely be called 1.5.2.1 to make it compatible with the
PEP6 naming system.)

Finally, you have been told several times that your notion of "current"
is just plain wrong.  You're free to keep using your idiosyncratic
definition, of course -- but we're also free to ignore you.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From emile at fenx.com  Wed Feb 27 15:05:35 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 12:05:35 -0800
Subject: writing efficiently a numpy array
References: <MPG.16e755abd178b85e989682@news.wanadoo.fr>
Message-ID: <a5jege$7r6k2$1@ID-11957.news.dfncis.de>

"Stephan" <stephan.schumacher at andra.fr> wrote in message
news:MPG.16e755abd178b85e989682 at news.wanadoo.fr...
>
> Hi,
>
> I want to write a large numpy array on an ascii file. There must be N
> numbers per line (except for the last line...).
>
> Does anyone knows something more efficient than the function below :
>

In what area are you looking to be more efficient?  Memory, time, or
something else?  Do you need it in ascii form for another program, or
are you saving it to be read back in?

Things you might consider are tostring if your saving for Numeric use
only, or concatenate and reshape by N and writing out row by row.

--

Emile van Sebille
emile at fenx.com

---------



> from Numeric import *
> from string import *
>
> def writeArray(file, my_array, N) :
>     my_string_array=map(repr,my_array.tolist())
>     tmp = []
>     ind = 0
>     for i in my_string_array :
>     tmp.append(i)
> if ind == N-1 :
>        tmp.append('\n')
> ind = (ind + 1)%N
>     file.write(join(tmp," "))
>
>
> # exemple :
>
> a = arange(300000)
> file = open("exemple","w")
> writeArray(file, a, 8)
> file.close()
>
>
>
>



From markabarclay at mediaone.net  Mon Feb 25 18:55:52 2002
From: markabarclay at mediaone.net (Mark Barclay)
Date: Mon, 25 Feb 2002 18:55:52 -0500
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au>
Message-ID: <3C7ACF08.6BB8AA11@mediaone.net>

Thanks Mark. I have your book, and it is invaluable. I should have credited you
personally.

Is there a way of coding a COM server that can be running before it is accessed
by a COM client? I'm not sure if this makes sense, but I'd like to have a program
running, that can be connected to, perhaps multiple times, without it having to
create multiple instances of teh server, just have multiple clients using it.

Mark

Mark Hammond wrote:

> dsavitsk wrote:
> > just for the sake of giving credit where it is due, and not giving it where
> > it is not, Mark Hammond is responsible for the excellent COM groundwork.
> > ActiveState is responsible for laying him off.
>
> LOL - thanks for the clarification.  I was going to post something
> similar, but it would not have sounded as good coming from me! :)
>
> Mark.

--
Mark Barclay (markabarclay at mediaone.net)




From rmunn at pobox.com  Fri Feb 15 13:59:13 2002
From: rmunn at pobox.com (Robin Munn)
Date: 15 Feb 2002 18:59:13 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
Message-ID: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>

Reading the first few chapters of _How to Think Like a Computer
Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/ for those who haven't
heard of this book), I was struck by how Python resembles BASIC in some
ways that make it a very good language for teaching:

* Interpreted, so there's immediate feedback. Students can just type
  'python' at the command prompt and start coding, receiving *immediate*
  feedback on the effects of what they type.

* Non-scary assignment and output statements. In fact, Python and
  BASIC's output statments look identical. "Print" is a natural choice
  for ease of understanding. Using sys.stdout.write can come later, once
  students begin to realize that they don't always *want* a newline or
  space appended to their output.

* No need to declare variables. Use them and they are immediately
  created for you, ex nihilo.

All three of these things make for a very low barrier-to-entry for
people who've never done any programming before. I'm sure there are
other similarities as well, that maybe someone else will point out. Of
course, Python goes beyond BASIC in many ways:

* NO LINE NUMBERS! Enough said.

* Rich control-flow statements. Of course, this is available in more
  modern versions of BASIC, but back when I was first learning to
  program (at age six, using BASIC), all I had was conditional GOTO. You
  know, IF condition GOTO line-number. Or, for the advanced students,
  IF condition GOSUB line-number (I never did understand the necessity
  for that kind of statement until well after I'd stopped using BASIC
  entirely and moved to more "modern" languages like... Pascal).

* Batteries included. This is the big one, which makes Python more than
  just a toy language suitable for learning. The richness of the
  standard library makes Python incredibly useful, and I for one would
  like to see it continue to grow. I understand the concern that people
  developing Python-based programs for distribution would like their
  users to be able to run a 100k script without downloading a 10M
  distribution, but I don't want to see that cause a reduction in the
  standard library. Instead, we could make make two Python
  distributions, a "full" install containing everything and a "lite"
  install containing only the interpreter and maybe -- *maybe* -- a
  couple of essential modules like os and sys. Then people developing
  Python programs like, say, a pygame-based game, could distribute a
  single package containing the "Python lite" distribution, their own
  code, and any other modules their code depended on -- all in one
  easy-for-the-end-used-to-install package.

Anyhow, I hope this helps develop some ideas.

-- 
Robin Munn
rmunn at pobox.com


From frederic.giacometti at arakne.com  Fri Feb 22 23:12:17 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Sat, 23 Feb 2002 04:12:17 GMT
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <a50hd2$6vt$1@panix2.panix.com> <a50im3$bf1$1@news.island.liu.se> <slrna7dipm.5et.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <BEEd8.66441$dG.27583658@news1.rdc1.sdca.home.com>

"Quinn Dunkan" <quinn at regurgitate.ugcs.caltech.edu> wrote in message
news:slrna7dipm.5et.quinn at regurgitate.ugcs.caltech.edu...
> On Wed, 20 Feb 2002 17:23:22 +0100, Anders Dahlberg
<andda715 at student.liu.se>
> wrote:
> >
> >"Aahz Maruch" <aahz at panix.com> skrev i meddelandet
> >news:a50hd2$6vt$1 at panix2.panix.com...
> >> In article <m2wux85hud.fsf at desk.crynwr.com>,
> >> Russell Nelson  <nelson at crynwr.com> wrote:
> >> >Henrik Motakef <henrik.motakef at web.de> writes:
> >> >> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> >> >>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> >> >>>>
> >> >>>> Blig matro en cordiolios, plur tanto lestanto forgolumnia.
Plachero
> >> >>>> czignat?  Classifero en bragna!
> >> >>> Woas da Deifi wos des hoasn mog.
> >> >> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> >> >
> >> >Even better: we could switch languages within threads, to *really*
> >> >keep the monolinguists confused.
> >>
> >> B'vakasha
> >
> >Ja, varf?r inte?
>
> My s?ster was bitten by a m??se once.

Probably one should remind our Nordic friends that, without regard to the
language they choose to express themselves, this is c.l.p, and
python-related conversation is expected to take place here; provided, of
course, that Nordic languages be sufficiently evolved to permit such a thing
:))

So, you may use your village dialect, but please use it to discuss python
matter.

Thanks in advance ...






From martin at v.loewis.de  Mon Feb 25 00:20:20 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: Mon, 25 Feb 2002 06:20:20 +0100
Subject: PEP 263 comments
Message-ID: <200202250520.g1P5KKD01484@mira.informatik.hu-berlin.de>

To make some progress on PEP 263, I suggest that some of the open issues
are resolved as follows:

- Comment syntax: I suggest to use the form
  -*- coding: <coding name> -*-
  Emacs already recognizes this syntax, as does patch #508973 
  on IDLEfork. The other proposed syntaxes should be removed from the
  PEP.

- In addition, to simplify usage on Windows, Python recognizes the
  UTF-8 file signature (e.g. as generated by notepad). Any file
  starting with \xef\xbb\xbf is treated as being UTF-8; a coding
  comment different from "utf-8" in such a file is an error.

- identifiers remain restricted to ASCII

- Implementation strategy: I believe the proposed strategy (change the
  tokenizer) is overly complicated, and likely inefficient. Instead, I
  suggest that the encoding directive applies only to Unicode literals.
  It will still be formally an error if comments or string literals do
  not follow the declared encoding, but the Python parser won't detect
  this error. 

  For use in Unicode literals, the parser will continue to work as it
  does now, except that it applies the declared coding in compile.c.
  To do so, PyUnicode_DecodeRawUnicodeEscape and
  PyUnicode_DecodeUnicodeEscape will expect an additional flag
  indicating whether they operate on a char* or a Py_UNICODE*.

  The only problem with this approach is that encodings where " or '
  could be the second byte of a multi-byte character cannot be
  supported as a source encoding. Python supports no such encoding
  in the standard library at the moment, anyway, so this should not
  be a problem.

- Backwards compatibility: I'm in favour of leaving mostly everything
  as-is, i.e. if there is no declared encoding, it should be possible
  to put arbitrary bytes in string literals and comments; the proposed
  implementation strategy supports that. However, I think that Unicode
  literals which use the Latin-1 fallback should be deprecated, and that
  the implementation should raise a DeprecationWarning: Anybody relying
  on that feature should declare that the encoding is Latin-1.

- Changes to IDLE: When IDLE opens a file, it shall look for the UTF-8
  signature. If no UTF-8 signature is found, it shall look for the
  coding comment. If none is found, it shall apply the locale's
  coding, which is determined as follows:
  - on windows, it is "mbcs"
  - on Unix, it is the one returned by nl_langinfo(CODESET)
  Otherwise, it is the system default encoding.

  When saving a file, IDLE shall preserve the UTF-8 signature if there
  was one. If not, and if there is a coding comment, that should be
  used to encode the file. If there is none, the locale's encoding
  should be used. If encoding fails (whether the coding was found in
  the comment or in the locale), the file shall be UTF-8 encoded, and
  an UTF-8 signature added.

Regards,
Martin



From scarblac at pino.selwerd.nl  Thu Feb 28 06:21:23 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 28 Feb 2002 11:21:23 GMT
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu>
Message-ID: <slrna7s4fc.4pv.scarblac@pino.selwerd.nl>

Fernando P?rez <fperez528 at yahoo.com> wrote in comp.lang.python:
> Python is a general purpose language, not one to be used only by number 
> theorists. Putting at a level as basic as the integers a convention which 
> allows things like 'for i in 3' to be valid iteration code really sounds 
> bizarre and counter intuitive to me.
> 
> But maybe it's just me.

It's not just you.

I believe syntax like 'for i in 5:' is confusing (at first sight I'd think
that i would get one value, 5), and has no merit except saving a few key
strokes compared to 'for i in range(5):', which is clearer as well.

Compare:

for i in 6: print i
for i in 6,7: print i
for i in 6,7,8: print i

You wouldn't think the first line does something different, would you?


-- 
Remco Gerlich


From sjmachin at lexicon.net  Fri Feb  8 18:15:06 2002
From: sjmachin at lexicon.net (John Machin)
Date: 8 Feb 2002 15:15:06 -0800
Subject: isIterable (was Re: FixedPoint)
References: <mailman.1013119093.20033.python-list@python.org> <GJS88.49612$3J1.1396553@news2.tin.it>
Message-ID: <c76ff6fc.0202081515.57d3bfb3@posting.google.com>

Alex Martelli <aleax at aleax.it> wrote in message news:<GJS88.49612$3J1.1396553 at news2.tin.it>...
[snip]
> Unfortunately, if we're in Python 2.2 and x is an iterator, isIterable
> "consumes" x's first item.  This isn't hard to fix:
> 
> def isIterable(x):
>     try: x=iter(x)
>     except: pass
>     else: return 1
>     try:
>         for i in x: return 1   # iterable
>         else: return 1         # iterable, although empty
>     except:
>         return 0
> 
> This assumes function iter, if defined, is "sensible" -- it returns an
> iterator, or else raises an exception.  Builtin function iter does meet
> this requirement (it raises a TypeError if __iter__ is defined but
> returns a non-iterator), and (as usual in Python) it seems sensible
> to assume that if somebody has replaced a builtin function then that 
> somebody is responsible for the overall results still being sensible.

Ignoring the possibility that someone has implemented a botched
replacement for the builtin iter() function, I am having some
difficulty finding an example of an x that would raise an exception in
"iter(x)" but be acceptable in "for i in x:". All sorts of things that
I've tried either pass the first stage of your isIterable() or fail
both stages.

The 2.2 manual says about iter():
"""
Without a second argument, o must be a collection object which
supports the iteration protocol (the __iter__() method), or it must
support the sequence protocol (the __getitem__() method with integer
arguments starting at 0). If it does not support either of those
protocols, TypeError is raised.
"""

Looks to me like iter() is already doing what is in your second stage.

Can you supply an example which requires your second stage, please?


From cliechti at gmx.net  Mon Feb 18 17:58:20 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 18 Feb 2002 23:58:20 +0100
Subject: New to python
References: <yxfc8.145803$Ar5.5568410@typhoon.austin.rr.com>
Message-ID: <Xns91BA45C6117cliechtigmxnet@62.2.16.82>

"Avi Homes" <Moo at houston.rr.com> wrote in
news:yxfc8.145803$Ar5.5568410 at typhoon.austin.rr.com: 
> Items('IName')=temp

try the other brackets: []
Items['IName']=temp

-- 
Chris <cliechti at gmx.net>



From shalehperry at attbi.com  Wed Feb 20 17:56:15 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 20 Feb 2002 14:56:15 -0800 (PST)
Subject: A bundle of questions from a Python newbie
In-Reply-To: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <XFMail.20020220145615.shalehperry@attbi.com>

> 
> 1. The way I thought that an assignment like
> 
> var += 1 
> 
> worked was that the interpreter would go to the place pointed by var and
> add 1. In particular, id(var) should remain constant. But this is not
> what happens as a simple test shows it. So what is wrong in my
> perception of what += does? Is this an implemetation issue?
> 

without looking at the Python code, my assumption is that a new variable is
created for each call to '='.

> 4. I have read the FAQ about the "self controversy" - why one has to
> declare it as an argument and then use it explicitly in the body of the
> method - but I really feel that there must be a better solution. I
> thought something like this could be useful:
> 
> a. self doesn't need to be declared.
> 
> b. Distinguishing data members from local variables could be achieved by
> instead of using self.whatever use .whatever - a single dot in front of
> the name "whatever".
> 
> I don't expect that this solution is adopted (it would break all the
> existing code - ahhh every Nero-like language designer's dream - wreak
> havoc in the citadel of its users and start afresh), what I would like
> to know is how bad/good is this idea.
> 

the problem is the period in front of the word is VERY easy to miss. 
Especially after a long day of coding or for those with poor eye sight.

> 
> 6. Is there a reason why 3.__abs__() does not work? After all 3 is an
> object (in the wider sense of the word). And it does work if we assign
> it to a variable, eg, 
> 
> var = 3
> var.__abs__()
> 
> I know that code like the above (methods applied to number literals) is
> probably rare, but consistence is also a good thing.
> 

because not everything is an object (yet)

> 10. I have a class that works like a list in the sense that supports the
> method __getitem__ and its relatives. But I'm at a loss as to what I
> have to do to support the slice notation.
> 

>>> class Foo:
...   def __getslice__(*args):
...     print args
... 
>>> f = Foo()
>>> f[1:1]
(<__main__.Foo instance at 0x80905e4>, 1, 1)




From logiplexsoftware at earthlink.net  Mon Feb 11 13:10:57 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 11 Feb 2002 10:10:57 -0800
Subject: XP (was Re: Defending the Python lanuage... )
In-Reply-To: <002201c1b301$10d63ee0$0101010a@local>
References: <3C62EE4A.5090305@destiny.com>
	<20020207140755.364d5d77.logiplexsoftware@earthlink.net>
	<200202081311.g18DBQGo020260@ratthing-b246.strakt.com>
	<20020208103714.11a2ae01.logiplexsoftware@earthlink.net>
	<200202110940.g1B9eTGo023834@ratthing-b246.strakt.com>
	<002201c1b301$10d63ee0$0101010a@local>
Message-ID: <20020211101057.208c3f1b.logiplexsoftware@earthlink.net>

On Mon, 11 Feb 2002 07:35:32 -0600
Chris Gonnerman wrote:

> ----- Original Message ----- 
> From: "Laura Creighton" <lac at strakt.com>
> > 
> > Don't formalise the process too much.  You are not trying to bring 
> > order to your establishment by building rules that work top-down.  
> > You are trying to foster order that works from the bottom up.
> 
> Here is the ideal of XP which I think is being lost on managers 
> trying to implement it for buzzword compliance.  They don't really
> understand what they are doing, so they will *mandate* all parts
> of XP even if they are incompatible with the people doing the work.
> 
> XP is only going to work if you adapt it together with your people.
> Each side may have to give; if the programmers you are working with
> are hardheaded (and which of us aren't), forcing them to pair-program
> or do other XP things is *not* going to improve anything.
> 
> I've lived through too many "paradigm shifts" already.  I remember
> the fiasco of TQM being applied to a scientific laboratory, where
> commitment to quality was already so high that the technicians were
> insulted.  Managers blindly applying buzzwords like Band-Aids makes
> me curse... whether it's TQM, or Java, or XP, *no* "new way" will
> work for everyone.
> 
> It's just not a one-size-fits-all world.

Perhaps this is the problem I was having with these ideas myself.  When
someone proposes ideas like this, I look at it from a "mandated" point of
view (since that seems to be what happens) and so end up playing the
devil's advocate, looking for situations where it won't work (or thinking
of times where something else worked better).  I've backed up from my
earlier stance, given the arguments forwarded by Laura and others in this
thread (and also, knowing a little more specifics has helped too - but
that's the danger of arguing about things you know too little about ;)

Anyway, I think that perhaps these techniques, if not adhered to too
tightly, could be very beneficial, and am willing to admit that perhaps I
was talking out the wrong end.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From martin_byrne2 at yahoo.co.uk  Fri Feb 22 03:33:55 2002
From: martin_byrne2 at yahoo.co.uk (B)
Date: 22 Feb 2002 00:33:55 -0800
Subject: Python ActiveX Scripting Engine error
References: <508cb8ae.0202210917.6d97e157@posting.google.com> <3C752D64.9090100@mxm.dk>
Message-ID: <508cb8ae.0202220033.1fd230c7@posting.google.com>

Max M <maxm at mxm.dk> wrote in message news:<3C752D64.9090100 at mxm.dk>...
> Hmm I sometimes experience those. I think it is when I have PythonWin 
> open at the same time an I try to run a script with errors. Then it sort 
> of "gets stuck" on the iis.
> 
> A restart of IIS solves it for me.
> 
I dont have PythonWin open and restarting IIS makes no difference, the
error just keeps happening randomingly it seems.
I did read something about redirecting before a html page is build can
cause this but if anyone has any other ideas I'd like to hear them.

Cheers


From codetube at yahoo.com  Mon Feb  4 02:48:55 2002
From: codetube at yahoo.com (CodeTube)
Date: 3 Feb 2002 23:48:55 -0800
Subject: A Python console in a Visual Basic application????
Message-ID: <5d65cf44.0202032348.79395fea@posting.google.com>

Hello Everyone,

I wonder if any one knows some thing about implementing the PythonWin 
console and other capabilities from with in a Visual Basic 
application.
I am writing an application in which I propose to have a console 
similar to PythonWin and also be able to run python files etc.
This might be a "big" question, so I would appreciate any clues or 
leads in that direction.
I had looked into some examples given by Mark Hammond in his 
book "Programming Python" but then I am looking for a more tight 
integration into the VB application.

Any help would be appreciated...

Best Regards,
Code Tube


From max at alcyone.com  Sun Feb 10 16:48:09 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 10 Feb 2002 13:48:09 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com> <7xlme1hv0y.fsf@ruckus.brouhaha.com>
Message-ID: <3C66EA99.B22478E3@alcyone.com>

Paul Rubin wrote:

> It seems to me like the principle that programs should be able to "do
> nothing" gracefully.  The argument is that split should be the inverse
> of join, even when the splitting string is empty.
> 
> Imagine if "a+0" raised an error.  You wouldn't say "just use 'a'
> instead".

That is simply not comparable.  The purpose of split is to divide up a
string according to a delimiter.  Asking for it to split a string given
a null delimiter doesn't make much sense at all; if you're calling
string.split with the intent of dividing it up and you're passing in ''
as a delimiter, that's almost certainly a mistake.  If you really _do_
want the equivalent of list(string), then you might as well have just
done that in the first place.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From dseaman at seaman.cc.purdue.edu  Tue Feb  5 12:43:53 2002
From: dseaman at seaman.cc.purdue.edu (Dave Seaman)
Date: 5 Feb 2002 12:43:53 -0500
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com> <60018f02.0202050547.4836c5ff@posting.google.com>
Message-ID: <a3p5kp$elc@seaman.cc.purdue.edu>

In article <60018f02.0202050547.4836c5ff at posting.google.com>,
Marc Laymon <laymon at crd.ge.com> wrote:

>I can duplicate the error message with a perl script by changing
>"#!/usr/bin/perl" to "#!/usr/bin/foo", so it seems that for some
>reason, Unix can't find python, but since /usr/bin is in my path
>and I am explicitly giving the command "usr/bin/python",

You need the command /usr/bin/python, not usr/bin/python.

Or use

	#! /usr/bin/env python

as suggested elsewhere.

-- 
Dave Seaman			dseaman at purdue.edu
Amnesty International says Mumia Abu-Jamal decision falls short of justice.
<http://www.amnestyusa.org/news/2001/usa12192001.html>


From threeseas at earthlink.net  Sat Feb  9 10:33:15 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 09 Feb 2002 15:33:15 GMT
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu>
Message-ID: <558.804T2090T11654864threeseas@earthlink.net>

http://mail.gnome.org/archives/gnome-hackers/2002-February/msg00072.html

I kinda like the one on ethics, amoung some of the others.

But then group voting is like a conspiracy huh?


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From emile at fenx.com  Tue Feb  5 13:51:42 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 5 Feb 2002 10:51:42 -0800
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de> <a3p2fg$49g$1@mail1.wg.waii.com> <a3p9b6$1a7qtb$1@ID-98166.news.dfncis.de>
Message-ID: <a3p9uh$19q6me$1@ID-11957.news.dfncis.de>

"MDK" <mdk at mdk.com> wrote in message
news:a3p9b6$1a7qtb$1 at ID-98166.news.dfncis.de...
> >>> class MyList:
> ...     def __init__(self):
> ...             self.list=[]
> ...     def append(self, item):
> ...             self.list.append(item)
> ...
> > >
> >>> import mylist
> >>> x = mylist()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
> TypeError: object of type 'module' is not call
>
> Why is this happening?  My list object must be a class in its own,
> importable file.

s/b

x = mylist.MyList()

--

Emile van Sebille
emile at fenx.com

---------



From andre.risnes at tenpipes.no  Wed Feb 20 04:46:36 2002
From: andre.risnes at tenpipes.no (André Risnes)
Date: Wed, 20 Feb 2002 10:46:36 +0100
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <G5Bc8.1$Az6.364@news.hananet.net> <3C7368A0.5050702@bath.ac.uk>
Message-ID: <a4vrdm$j98$1@lise.netcom.no>

"Simon Willison" <cs1spw at bath.ac.uk> wrote in message
news:3C7368A0.5050702 at bath.ac.uk...
>
> It's pretty important to provide some kind of checking mechanism like
> that to avoid people copying/pasting the URL of the page they are on
> into an e-mail / instant message and inadvertantly giving their session
> to someone else.
>

That can be avoided by embedding the session ID in a hidden
field in a form instead of the URL (if forms are used, that is).

--
Regards
Andr? Risnes




From jimd at vega.starshine.org  Thu Feb 28 05:00:41 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 28 Feb 2002 10:00:41 GMT
Subject: dirwalk.py generator version of os.path.walk
References: <a5d36e$1daf$1@news.idiom.com> <ac677656.0202271621.5a134d44@posting.google.com>
Message-ID: <a5kv49$ae8$2@news.idiom.com>

In article <ac677656.0202271621.5a134d44 at posting.google.com>, Tom Good wrote:

>jimd at vega.starshine.org (Jim Dennis) wrote in message news:<a5d36e$1daf$1 at news.idiom.com>...
>> This function could probably use a bit of polishing,
>>  and it certainly could use some enhancement (some options to
>>  control if, and how we follow symlinks, to how to handle 
>>  exceptions on listdir(), whether to be depth first, and an 
>>  option to avoid crossing mount boundaries with os.path.ismount(), 
>>  etc).

>>  However, it seems to work.  

>>  dirwalk() simply takes an optional top level directory/path name
>>  as an argument and instantiates a generator which will walk down
>>  that tree and return every filename that it can access.  

>>  It's late and I need sleep.  So I'm just going to post this in
>>  it's rough (and probably buggy) form and let y'all thrash on it
>>  a bit.

>>  I guess there's some sort of statcache module that might let me
>>  cache the stat() tuples.  I guess I'm implicitly incurring a stat()
>>  system call for each node by checking islink() and isdir() on it
>>  so it seems like I ought to cache that and make it available to 
>>  my caller (without forcing them to make an additional stat system
>>  call).

>>  I hope that something like this (a simple dirwalk() or other 
>>  greatly simplified alternative to os.path.walk()) makes it into 
>>  Python 2.3 or later.

>> #!/usr/bin/env python2.2
>> from __future__ import generators 
>> import os

>> def dirwalk(startdir=None):
>> 	if not startdir:
>> 		startdir="."
>> 	if not os.path.isdir(startdir):
>> 		raise ValueError ## Is this the right exception?
>> 	stack = [startdir]
>> 	while stack:
>> 		cwd = stack.pop(0)
>> 		try:
>> 			current = os.listdir(cwd)
>> 		except (OSError):
>> 			continue	# Skip it if we don't have access
>> 		for each in current:
>> 			each = os.path.join(cwd,each)
>> 			if os.path.islink(each): 
>> 				pass
>> 			elif os.path.isdir(each):
>> 				stack.append(each)
>> 			yield(each)

>> if __name__ == "__main__":
>> 	# import unittest?
>> 	# test suite should consist of:
>> 	# 	dirwalk() vs. os.listdir()
>> 	# 	dirwalk("/") vs. os.path.walk()
>> 	# 	dirwalk("/etc/passwd") (should raise exception)
>> 	import sys
>> 	for i in sys.argv[1:]:
>> 		for j in dirwalk(i):
>> 			print j
>> 	# should compare this to os.popen("find ....") and
>> 	# or to os.path.walk(...)

>Hi,

> I wrote a different implementation of this general concept at:

> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/105873

> You don't really need to keep a stack of directories and push/pop
> things, because with generators you can recurse instead.

>Tom

 But recursion is likely to cost more.  The only state I need
 to keep is my current "todo" list of directories.  A recursion 
 would store functional state (unless Python supported tail-end
 recursion).  So the append/pop (total cost, 3 lines of code)
 seems like the lightest weight way to do this.


From carifio.nospam at nospam.usys.com  Fri Feb  8 21:08:21 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Sat, 09 Feb 2002 02:08:21 GMT
Subject: What's the equivalent in Python for Perl's CPAN?
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com> <a3v903$kgk$1@panix2.panix.com> <msU88.19568$zJ3.1529583@typhoon.maine.rr.com> <3C6430A6.4DD9316E@ccvcorp.com>
Message-ID: <pw%88.19654$zJ3.1539938@typhoon.maine.rr.com>

> Well, if you'd tried searching Google's archives of this newsgroup, you'd
have
> doubtless found the answer, as it's been discussed many times. ...

Yup. Forgot about that part of google. I stand corrected.





From ron.l.johnson at cox.net  Mon Feb 18 23:06:12 2002
From: ron.l.johnson at cox.net (Ron Johnson)
Date: Tue, 19 Feb 2002 04:06:12 GMT
Subject: confusion regarding os.path.walk()
References: <20020218180623.086f6c58.ron.l.johnson@cox.net>
	<Xns91BA1077BDD93cliechtigmxnet@62.2.16.82>
Message-ID: <20020218220610.167f71d5.ron.l.johnson@cox.net>

On 19 Feb 2002 01:33:52 +0100 Chris Liechti <cliechti at gmx.net> wrote:

> Ron Johnson <ron.l.johnson at cox.net> wrote in 
> news:20020218180623.086f6c58.ron.l.johnson at cox.net:
> > In the visit callout function, it looks as though after the top
> > level is processed, os.path.isfile() and os.path.isdir() both return 
> > 0 when passed elements from the names parameter.
> ...
> >     for n in names:
> >    	    	print n,
> >    	    	if os.path.isdir(n):
> >          ...
> 
> in "names" there are only filenames with no path. to access the files 
> you must join the path back to the name:
> 
> for n in names:
>     	file = os.path.join(path, n)
>     	print file,
>     	if os.path.isdir(file): 
>     	...

Thinking that that might be the problem, I used os.path.realpath().
However, even that didn't work.  Why wouldn't testing an absolute
filename work?

os.path.join() worked.  Thanks.

-- 
+------------------------------------------------------------+
| Ron Johnson, Jr.        Home: ron.l.johnson at cox.net        |
| Jefferson, LA  USA      http://ronandheather.dhs.org:81    |
|                                                            |
| 484,246 sq mi are needed for 6 billion people to live, 4   !
! persons per lot, in lots that are 60'x150'.                |
! That is ~ California, Texas and Missouri.                  !
! Alternatively, France, Spain and The United Kingdom.       |
+------------------------------------------------------------+


From gtalvola at nameconnector.com  Thu Feb 21 08:58:55 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Thu, 21 Feb 2002 08:58:55 -0500
Subject: HTTP state management without cookies?
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E41240DF@MAILBOX.nameconnector.com>

Paul Rubin wrote:
> "Steve Holden" <sholden at holdenweb.com> writes:
> > > http://foo.bar/_SID_=2374820394/myscript.cgi/
> > >
> > > into:
> > >
> > > http://foo.bar/myscript.cgi?_SID_=2374820394
> > >
> > > so the same trick will work for CGI?
> > >
> > Geoff:
> > 
> > I appreciate the slightly unusual nature of thios scheme, 
> but I don't
> > see how it isn't included under the heading "pass session 
> id through all
> > URLs" as mentioned in my original post.
> 
> Why would you want to pass the session ID to off-site URL's?

I didn't express myself properly.  What I meant is, in a CGI application if
you're including the session ID in the query string, then you have to make
sure all of your relative links within the site are constructed to include
that session ID.  In other words, your links will have to be written like:

'<a href="SomeOtherScript.cgi?_SID_=%s">Link to Some Other Script</a>' %
sessionID

Whereas if you use the mod_rewrite trick, your relative links within the
site can just be static strings:

'<a href="SomeOtherScript.cgi">Link to Some Other Script</a>'

which is much cleaner.  Since the session ID is "higher up" in the URL, it
is preserved between requests.


- Geoff



From gerhard at bigfoot.de  Mon Feb 25 14:14:08 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 25 Feb 2002 19:14:08 GMT
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com>
Message-ID: <slrna7l380.16u.gerhard@lilith.my-fqdn.de>

Bob Lancaster wrote in comp.lang.python:
> I am in a bit of a quandry. 
> I am writing a python module which converts some data from various
> python variables, lists, etc. into SQL, which in turn is inserted to
> or selected from a database.
> 
> I need to support three combinations.
> (1) PostgreSQL and Linux, for academic customers.  No serious problems
> here.  In fact, I have learned why folks love PostgreSQL so much.
> 
> (2) Oracle 9i and Linux.  I haven't tried this one yet, but it seems
> to be at least as easy as the third combination.
> 
> (3) Oracle 9i and Windows 2000, for industrial customers.  Trouble!
> I have tried two approaches.  
>   (a) DCOracle2:  builds, connects, but then does not work properly. 
> I see two possible sources of the problem.

Or:

(4) Use a commercially supported DB module like mxODBC.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From mlh at vier.idi.ntnu.no  Wed Feb  6 09:22:28 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 6 Feb 2002 14:22:28 +0000 (UTC)
Subject: CSV revisited
Message-ID: <slrna62f14.pef.mlh@vier.idi.ntnu.no>

A little while ago I started a thread about adding CSV support to the
standard library. There was (as expected) some controversy about this,
but I think the idea had enough support that it might be turned into a
formal proposal... However, I'm not sure wheter this is the sort of
thing that should be formalised as a PEP... Any opinions in the
matter? (Not pro/con the proposal itself... ;)

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From robin at jessikat.fsnet.co.uk  Mon Feb  4 13:48:38 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Feb 2002 18:48:38 +0000
Subject: Win32 Zope 2.5 BUG
Message-ID: <M57ypKAGetX8Ewjj@jessikat.demon.co.uk>

I've been having a few problems with Zope 2.5 running as a service under win2k.

I have Python 2.1.2 already installed in C:\Python. Win32all 145 is installed in the normal way so
C:\Python\win32 is the extensions directory for win32all. I have installed Zope-2.5.0 in

c:\Python\ZOPE_2_5_0

Using ProcExp 5.23 (www.sysinternals.com) I see the DLLs that
are being used by c:\Python\Pythonservice.exe (see below).
All looks OK except that I'm getting mixed DLLs/pyds. In particular
1) C:\WINNT\system32\PyWinTypes21.dll
2) C:\Python\win32\win32api.pyd 
3) C:\Python\ZOPE_2_5_0\bin\lib\win32\PyWinTypes21.dll
    etc

This seems like a potential security hole and/or a source of very hard
to figure out bugs. I'm not sure if there's any real harm, but I have problems
starting/stopping cleanly.

NB the same problem is happening in the python.exe started by Pythonservice that
is the actual zope server.

Process: PYTHON~1.EXE Pid: 736

Base       Size       MM    Description                                      Version          Time                Path
0x240000   0x16000    *                                                                       07/12/1999 12:00    C:\WINNT\system32\unicode.nls 
0x260000   0x2F000    *                                                                       07/07/2000 19:05    C:\WINNT\system32\locale.nls 
0x290000   0x41000    *                                                                       07/12/1999 12:00    C:\WINNT\system32\sortkey.nls 
0x2E0000   0x4000     *                                                                       07/07/2000 19:05    C:\WINNT\system32\sorttbls.nls 
0x8E0000   0x2000     *                                                                       07/12/1999 12:00    C:\WINNT\system32\ctype.nls 
0xAF0000   0x10000          Common Python 2.1 types for win32                2.01.0000.0145   18/01/2002 15:29    C:\WINNT\system32\PyWinTypes21.dll 
0x1E100000 0xAF000          Python Core                                      2.01.0017.1010   25/01/2002 13:42    C:\Python\ZOPE_2_5_0\bin\lib\win32\python21.dll 
0x1E200000 0xF000           Python Win32 API extensions                      2.01.0000.0145   18/01/2002 15:29    C:\Python\win32\win32api.pyd 
0x1E600000 0x10000          Common Python 2.1 types for win32                2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\PyWinTypes21.dll 
0x1E700000 0x8000           Python Win32 Thread and Process extensions       2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32process.pyd 
0x1E800000 0x7000           Python Win32 Event extensions                    2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32event.pyd 
0x1EA00000 0x7000           Python NT Service extensions                     2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\win32service.pyd 
0x1EB00000 0x8000           Python Service Dispatcher Exe                    2.01.0000.0140   22/06/2001 10:47    C:\Python\ZOPE_2_5_0\bin\lib\win32\PythonService.exe 
0x75E60000 0x1A000          Windows 2000 IMM32 API Client DLL                5.00.2195.2821   04/05/2001 19:05    C:\WINNT\system32\imm32.dll 
0x779B0000 0x9B000                                                           2.40.4517.0000   04/05/2001 19:05    C:\WINNT\system32\OLEAUT32.DLL 
0x77A50000 0xF6000          Microsoft OLE for Windows                        5.00.2195.2887   04/05/2001 19:05    C:\WINNT\system32\OLE32.DLL 
0x77B50000 0x89000          Common Controls Library                          5.81.3103.1000   04/05/2001 19:05    C:\WINNT\system32\COMCTL32.DLL 
0x77C70000 0x4A000          Shell Light-weight Utility Library               5.00.3315.1000   04/05/2001 19:05    C:\WINNT\system32\shlwapi.dll 
0x77D40000 0x70000          Remote Procedure Call Runtime                    5.00.2195.2832   04/05/2001 19:05    C:\WINNT\system32\rpcrt4.dll 
0x77DB0000 0x5B000          Advanced Windows 32 Base API                     5.00.2195.2867   04/05/2001 19:05    C:\WINNT\system32\ADVAPI32.DLL 
0x77E10000 0x64000          Windows 2000 USER API Client DLL                 5.00.2195.2821   04/05/2001 19:05    C:\WINNT\system32\USER32.DLL 
0x77E80000 0xB5000          Windows NT BASE API Client DLL                   5.00.2195.2778   04/05/2001 19:05    C:\WINNT\system32\KERNEL32.DLL 
0x77F40000 0x3C000          GDI Client DLL                                   5.00.2195.2778   04/05/2001 19:05    C:\WINNT\system32\GDI32.DLL 
0x77F80000 0x7B000          NT Layer DLL                                     5.00.2195.2779   04/05/2001 19:05    C:\WINNT\system32\NTDLL.DLL 
0x78000000 0x46000          Microsoft (R) C Runtime Library                  6.01.8924.0000   04/05/2001 19:05    C:\WINNT\system32\msvcrt.dll 
0x782F0000 0x242000         Windows Shell Common Dll                         5.00.3315.2902   04/05/2001 19:05    C:\WINNT\system32\SHELL32.DLL 

-- 
Robin Becker


From joonas at olen.to  Fri Feb 15 18:04:16 2002
From: joonas at olen.to (Joonas Paalasmaa)
Date: Sat, 16 Feb 2002 01:04:16 +0200
Subject: How can I make a screenshot using BitBlt or something else???
References: <6ac1ecad.0202151338.15d3c4b@posting.google.com>
Message-ID: <3C6D93F0.8F83FA9E@olen.to>

"I.C." wrote:
> 
> I am new in PythonWin programming so I have a simple question -
> How can I make a screenshot of the whole screen and then get the colors
> of some pixels?

Use Fredrik Lundh's grabscreen utility for PIL.
http://effbot.org/pil/index.htm


From python at chajadan.net  Tue Feb 19 03:41:46 2002
From: python at chajadan.net (chajadan)
Date: Tue, 19 Feb 2002 00:41:46 -0800
Subject: help - inserting values into database tables
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A199@natasha.auslabs.ava
 ya.com>
Message-ID: <5.1.0.14.0.20020219004046.00a76c60@63.118.78.184>

AAaaargh!

I've been moving my websites around WAY to much! Domain Name Server vs. 
Data Source Name oops!

But thanks for the good catch ;)

At 04:10 PM 2/19/02 +1100, you wrote:
> > From: chajadan [mailto:python at chajadan.net]
> >
> > It seems I have ISDN on the brain, because I changed the
> > correct "DNS" to
> > the incorrect "DSN" partway through and never went back ;)
>
>Actually, you changed the incorrect "DNS" to the correct "DSN" (Data Source
>Name) and never went back :)
>
>Tim Delaney
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From huaiyu at gauss.almadan.ibm.com  Thu Feb 28 13:46:48 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Thu, 28 Feb 2002 18:46:48 +0000 (UTC)
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com> <mailman.1014861466.23611.python-list@python.org>
Message-ID: <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com>

On Wed, 27 Feb 2002 19:57:21 -0600, Skip Montanaro <skip at pobox.com> wrote:
>
>    Huaiyu> Is there a way to do this
>    Huaiyu> for k, v in dict.items():
>    Huaiyu>     dict[k] = f(v)
>
>    Huaiyu> without involving a Python for loop?
>
>Check out the operator module.  In particular, operator.setitem will
>probably do what you want:
>
>    import operator
>    keys = dict.keys()
>    vals = map(f, dict.values())
>    map(operator.setitem, [dict]*len(keys), keys, vals)

This works.  Thank you.  It speeds up by more than a factor of four againt
for loop iteration.  The disadvantage is that it builds at least five
temporary lists, so it is not usable if memory is constrained.

>
>    Huaiyu> I'd imagine that a function implemented in C could speed things
>    Huaiyu> up quite a bit.  
>
>It all depends on how expensive f() is related to the overhead of a for loop
>and the cost of getting map ramped up.  Don't rely on your instincts.
>Benchmark it.

Well, I can't benchmark it until it is available.  :-( 

What I was alluding to is that a built-in iteration would loop over all the
values held in the dict without touching the keys at all, and certainly
without building temporary lists.  I imagine this would also bypass all the
the hash calculation and conflict resolution.   The fact that your solution
speeds up so much despite building temporary lists shows that a C loop over
values could save even more.

Huaiyu

PS. The test and result:

import operator
from time import time

def iteritems(f, d):
	for k, v in d.items():
		d[k] = f(v)

def mapvalue(f, d):
	keys = d.keys()
	map(operator.setitem, [d]*len(keys), keys, map(f, d.values()))
		
def timeit(test, f, d):
	_time = time()
	test(f, d)
	print time() - _time

if __name__ == '__main__':
	d = {}
	for i in range(1000000):
		d[i] = None
	def f(x): pass

	timeit(iteritems, f, d)
	timeit(mapvalue, f, d)

# result
	
19.8180559874
4.17512202263


From logiplexsoftware at earthlink.net  Thu Feb 14 12:46:18 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 14 Feb 2002 17:46:18 GMT
Subject: socket server interrupt?
References: <mailman.1013690723.25180.python-list@python.org>
Message-ID: <20020214.093631.260152959.25986@logiplex1.logiplex.net>

In article <mailman.1013690723.25180.python-list at python.org>,
"maximilianscherr" <MaximilianScherr at t-online.de> wrote:
> how can i make my socket server interrupt on ctr-c(i know with the
> keyboardinterrupt exception, but how?)?

The problem you will have with this approach is that your server is
multi-threaded; you won't know which thread will receive the interrupt.
A better alternative is to start yet another thread that accepts console
input and can shut the server down using a command, say 'q' or 'x'.

- Cliff


From eppstein at ics.uci.edu  Wed Feb  6 12:44:52 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 06 Feb 2002 09:44:52 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de>
Message-ID: <eppstein-95B064.09445206022002@news.service.uci.edu>

In article <a3rnd1$1a44oq$1 at ID-11957.news.dfncis.de>,
 "Emile van Sebille" <emile at fenx.com> wrote:

> def flatten(ary, rslt):
>     try:
>         flatten(ary[0], rslt)
>         flatten(ary[1:], rslt)
>     except:
>         if ary: rslt.append(ary)
> 
> rslt = []
> flatten(("cat",5,['dog',[3,3,1]],"zoo"), rslt)
> print 'rslt = ',tuple(rslt)

Isn't all that list-copying (ary[1:]) going to make for a very inefficient 
algorithm?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From QnickQm at alum.mit.edu  Tue Feb 19 23:38:31 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Wed, 20 Feb 2002 04:38:31 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <slrna74ssf.21s.QnickQm@localhost.localdomain> <24c39b2c.0202191658.3fd74528@posting.google.com>
Message-ID: <slrna76a8d.4kl.QnickQm@localhost.localdomain>

[message reformatted]

In article <24c39b2c.0202191658.3fd74528 at posting.google.com>, 
                    Alexander Jerusalem wrote:
> QnickQm at alum.mit.edu (Nick Mathewson) wrote in message 
>          news:<slrna74ssf.21s.QnickQm at localhost.localdomain>...
>> In article <24c39b2c.0202181531.187fad4c at posting.google.com>, 
>>         Alexander Jerusalem wrote:
>> > Has anyone heard of plans to add static type checking to Python?
>> 
>> Someone else has mentioned an old proposal of Guido's for adding
>> optional type declarations in order to help the compiler catch errors.
 
 [... an example that eventually arrives at a type like ...]

>> And you aren't using _all_ of the sequence protocol: just iteration...
>> 
>>      (A, B) -> str   s.t. A has __iter()__ -> C 
>>                      s.t. C has next() -> D (or raises StopIteration)
>>                      s.t. D has __eq__(B) and repr() -> str
>> 
>> IMNSHO, some kind type inference is the only tractable answer.
>> 
>> And-what-if-repr-doesn't-return-a-string?-ly yrs,
> 
> You argue that there are cases where static typing gets in your way,
> where it makes programs less flexible.

Well, not quite.  I argue that assigning types to even trivial Python
programs is a delicate and tricky business, and that the static types
implicit in a typical Python function are tremendously complex.

This _isn't_ an argument against optional static types for Python: I
believe in them too.  This is just an explanation of why they haven't
been done _yet_:
    1) Typical type declarations for Python will be more complex than
        those of most other popular languages.
    2) Therefore, type inference is almost certainly necessary.
    3) Because of 1 and 2, optional static typing for Python will 
        involve a lot of work. :)

Of course, this isn't to say that it won't or shouldn't get done: just
that the cost portion of the cost/benefit here is very high---and the
free time among the people who know enough about Python in particular
and type systems in general is quite low.

Of-course-I'd-love-to-be-proven-wrong-ly yrs,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From mishre at hushmail.com  Mon Feb  4 07:25:42 2002
From: mishre at hushmail.com (Mishre)
Date: 4 Feb 2002 04:25:42 -0800
Subject: Jython question concerning actionPerformed
Message-ID: <f6b73ec8.0202040425.4c3e0bb4@posting.google.com>

I have the following code in Jython running on JDK 1.4, which looks
like it should work, but isn't:

def actionPerformed(self,event):
    if event.getActionCommand() == "Exit":
        java.lang.System.exit(0)
    elif event.getActionCommand() == "comboBoxChanged":
        print event

Then later on I have:

self.catCombo = javax.swing.JComboBox(actionPerformed=self.actionPerformed)

I'm trying to update a JList when the selected item in the JComboBox
is changed.  If I use the code above I get the following error when I
run the program:

"TypeError: can't assign to this attribute in java instance:
actionPerformed"

If I use:

self.catCombo = javax.swing.JComboBox(actionPerformed=self.actionPerformed())

I get:

TypeError: actionPerformed() takes at least 2 arguments (1 given)

This is because I have "def actionPerformed(self, event)" but I need
both of them or I can't look at the event.

I've also tried this with the actionPerformed function declared as a
class inheriting java.awt.event.ActionListener, which works a little
better..but then I can't manipulate the self.catCombo item in the main
function (or at least if I can I don't know how).

Does anyone have any ideas on what to try next?

Thanks,

-M


From alex at newicons.com  Mon Feb 25 19:12:27 2002
From: alex at newicons.com (Alex Chavez)
Date: Mon, 25 Feb 2002 17:12:27 -0700
Subject: Python 2.2 on SunOS5.7 install errors
Message-ID: <alex-723057.17122625022002@iruka.swcp.com>

I'm getting the error msg below when I attempt to install python 2.2 on 
a sun solaris 7 os5.7 sparc box, using gcc 3.0.3


*** Error code 1
make: Fatal error: Command failed for target `Parser/tokenizer.o'



I hope I'm not duplicating the Jeff Bauer's post. I read the thread but 
I cant make heads or tails of it.


Alex Chavez


From kalle at gnupung.net  Sun Feb 17 04:36:27 2002
From: kalle at gnupung.net (Kalle Svensson)
Date: Sun, 17 Feb 2002 10:36:27 +0100
Subject: Have to try
In-Reply-To: <20020217001138.49b8ca05.ron.l.johnson@cox.net>
References: <004501c1b763$78bca9c0$b278c53f@pacbell.net> <mailman.1013918127.17499.python-list@python.org> <20020217001138.49b8ca05.ron.l.johnson@cox.net>
Message-ID: <20020217093627.GA14380@sandra.lysator.liu.se>

[Ron Johnson]
> On Sat, 16 Feb 2002 19:53:30 -0800 Wayne Pierce <wayne at mishre.com> wrote:
> > Rick Hamilton wrote:
> > > OK - I know I am going to get fried here but - I am out of options. I'm 
> > > a programmer (Java, VB and C++) and am taking a class on Operating 
> > > systems. The nut teaching this course is asking us to do a ton on 
> > > programming in Python. OK -
> > 
> >    You've come to the wrong newsgroup if you want to get flamed, sorry. 
> >   Is the professor a nut because he's using Python?  Or does he go 
> > galloping around the room with two coconut halves talking about swallows?
> 
> Just hypothesizing, but maybe Wayne is casting aspersions upon his
> teacher for using Python in an OS class?  Assemply, C, C++, Fortran, 
> even Bliss, but not Python.

Why?  As long as you're not implementing a real operating system, but
just illustrating principles and trying things out in some sort of
testing scaffold, Python is (IMO, of course) a better choise.
The advantages of Python still apply in OS class.  Low-level
programming should be learned elsewhere.

Peace,
  Kalle
-- 
Kalle Svensson (kalle at gnupung.net) - Laziness, impatience, hubris: Pick two!
English: http://www.gnupung.net/  Svenska: http://www.lysator.liu.se/~kalle/
Stuff: ["http://www.%s.org/" % x for x in "gnu debian python emacs".split()]



From writeson at earthlink.net  Mon Feb  4 08:05:53 2002
From: writeson at earthlink.net (Doug Farrell)
Date: 4 Feb 2002 05:05:53 -0800
Subject: wxPython book
Message-ID: <88bc63c6.0202040505.2a9c3016@posting.google.com>

Hi all,

I'm busy evangilizing my office co-workers on the wonders of Python
and wxPython, to which they are responding pretty well. We are
currently working on a test project that uses both, and the most
frequent complaint I get is "there's no book on wxPython..... wah wah
wah". I have seen mention of a book being written, presumably by Robin
Dunn. Is this true and if so, any idea when it might be coming out?

Thanks,
Doug


From phr-n2002a at nightsong.com  Sun Feb 17 08:46:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 05:46:17 -0800
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com> <m37kpcuu31.fsf@mira.informatik.hu-berlin.de>
Message-ID: <7x4rkgw79y.fsf@ruckus.brouhaha.com>

martin at v.loewis.de (Martin v. Loewis) writes:
> > I'm wondering how difficult it would be to add an editable (mutable)
> > string class to Python 2.2 implemented in C under the new type/class
> > unification scheme.
> 
> There is already an efficient mutable vector type, the array type,
> which can be used to implement both mutable byte strings and mutable
> character strings.

Very interesting.  I was about to answer that:

  1) There are functions to insert and delete single values into the array
     but not multi-element slices.  

  2) I don't think the existing string functions work on byte arrays.
     In particular, I don't think you can search for a regular expression in one.

But it turns out that both operations are supported!  I discovered
this by looking in the arraymodule.c source to see how the buffer
interface was implemented, and then experimenting.  The docs for the
array module gave no clue at all that these operations were supported.

> > Java has a StringBuffer class which Java programmers use idiomatically
> > whenever they want to build a big string from smaller ones.  Python
> > surprisingly lacks a counterpart to StringBuffer.
> 
> The Python counterpart of the Java StringBuffer is the StringIO module.

OK, though it looks like the C version needs to be extended to handle Unicode.

Anyway, it looks like the array module may do what I need.  So I guess
I'll submit a doc bug saying the above items should be described.

Thanks!!

Paul


From quinn at vomit.ugcs.caltech.edu  Mon Feb 25 13:39:34 2002
From: quinn at vomit.ugcs.caltech.edu (Quinn Dunkan)
Date: 25 Feb 2002 18:39:34 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu>
Message-ID: <slrna7l176.3al.quinn@vomit.ugcs.caltech.edu>

On Sun, 24 Feb 2002 19:34:34 -0800, David Eppstein <eppstein at ics.uci.edu> wrote:
>In article <a5cb54$27c2$1 at agate.berkeley.edu>,
> Daniel Yoo <dyoo at hkn.eecs.berkeley.edu> wrote:
>
>> : But what is the functional equvalent of:
>> 
>> : for x in items:
>> :     x.f()

There really is no functional equivalent of that, because it's not a functional
concept.  If your function doesn't have any side-effects, calling the function
only to throw away its value does nothing but suck up CPU time.

If you're calling the method for its side-effects, I'd write:

for x in items:
    x.f()

>I'd prefer
>[x.f() for x in items]
>
>It's not functional syntax, but so what?

Sure it is.  It's an expression, works best when x.f() has no side-effects, and
provides nicer syntax for mapping and filtering, two popular functional
concepts.  It's also borrowed from haskell, which, although functional, has a
lot of syntax sugar that turns into function application.


From wrbt at email.com  Thu Feb 28 17:00:00 2002
From: wrbt at email.com (Larry)
Date: 28 Feb 2002 14:00:00 -0800
Subject: fun with ADO
References: <2ec1bc1c.0202221202.1e0ea6aa@posting.google.com> <f20ea932.0202230533.6a01ac4b@posting.google.com> <Xns91C1AA4A0ABC6jimpublishingresourc@209.249.90.101>
Message-ID: <2ec1bc1c.0202281359.4981d443@posting.google.com>

Gracias for the responses. On your kind advice(s) I've already
reworked my Records class to use getRows and map a "pretty field"
function (that deals with cases like the money type in sql server that
can cause drama) onto the results.

Works even faster now. I'm learning every day here.


From cbbrowne at acm.org  Sun Feb  3 20:11:56 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Sun, 03 Feb 2002 20:11:56 -0500
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net>
Message-ID: <m3n0yqt5ib.fsf@chvatal.cbbrowne.com>

"Timothy Rue" <threeseas at earthlink.net> wrote:
> In reviewing over many messages in the autocoding and related
> threads and seeing the evidence of contridictions, hypocracy,
> irrational and illogical responses......

> Its damn clear that there is a need to remove the sort of people
> posing to be professional who are in essence just good liars,
> abstraction manipulators, from the industry.

Why don't you simply put us all in your killfile so you don't see what
we write?

I'm sure an increasing number of readers of these newsgroups are doing
this to you...
-- 
(reverse (concatenate 'string "ac.notelrac.teneerf@" "454aa"))
http://www3.sympatico.ca/cbbrowne/
"Bother,"  said Pooh,  "Eeyore, ready  two photon  torpedoes  and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"


From fperez528 at yahoo.com  Tue Feb  5 11:30:45 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 05 Feb 2002 16:30:45 +0000
Subject: Python 2.2 dynamic library problem on Compaq Tru64
References: <mailman.1012925289.31030.python-list@python.org> <eecbdf57.0202051509.f3fd8b5@posting.google.com>
Message-ID: <a3pq1c$nru$1@peabody.colorado.edu>

Anthony Allen wrote:

>> > ? ? import cx Oracle, os, sys, smtplib

This line is extremely suspicious, see the missing comma between cx and 
Oracle. Modules can't have spaces in their names. Double check that first.

Cheers,

f.


From jkraska at san.rr.com  Sun Feb  3 12:34:35 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 17:34:35 GMT
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org>
Message-ID: <v0tq5u8qfdhg7mi08bpa3523564fbpju9u@4ax.com>

>I've been wondering, lately, about lambda-expressions. Most programming
>languages support and...

Most do? Not that I know of. I suppose Java has the anonymous class
instantiation, which is close to -- but not quite -- a lambda expression.

Anyway, lambda expressions are a convenience for the lazy. If care
is not taken, they yield unmaintainable code. They discourage
reusability of that tidbit of code. But it's sometimes useful,
particularly considering that it associates small bits of code
most closely with their use.

Consider if you could do the following in C...

	f ( arg1, ((int)(int i))
		{
			return i+1;
		}
	);

... here we are, in effect, defining a function which returns an
int and take an int argument and then passing it on the fly to the
invocation of f. To have done this otherwise, we would have had to
fully define this function somewhere else, and pass in a function
pointer.

In languages which make frequent use of lambda expressions, the
most frequent use is often in passing small bits of code to
functions which do regular transformation of container objects.
In lisp, for example, we have the classic map/lambda combination,
where a new list is derived from a lambda applied to each
element of the list with the new list being composed of the
individual transformation implied by each lambda call.

C//



From pawelczyk at lucent.com  Fri Feb 15 04:26:37 2002
From: pawelczyk at lucent.com (Lukasz Pawelczyk)
Date: Fri, 15 Feb 2002 10:26:37 +0100
Subject: Layout management in tkinter.
References: <3C6B9001.E5568D2C@lucent.com> <a4g7lr$oqb$1@norfair.nerim.net> <a4gj46$1g4g$1@mail1.wg.waii.com>
Message-ID: <3C6CD44D.8CE9E71@lucent.com>

Yes Martin,

This is exactly what I have had. 
As I wrote I try to change the geometry manager pack into grid. :)
I think it is better because you can have more control.
but thanks it is probably this what also wrote Eric.
This combine ScrolledText use probably pack() method.
I have to use Scrollbar and Text widgets separately.

Thanks for your help,
Lukasz


From philh at comuno.freeserve.co.uk  Sun Feb  3 20:43:41 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 01:43:41 +0000
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <iJU5VJA6wdX8Ewk4@jessikat.fsnet.co.uk>
Message-ID: <slrna5rpqd.19d.philh@comuno.freeserve.co.uk>

On Mon, 4 Feb 2002 00:56:26 +0000, Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>In article <922.799T2387T11875593threeseas at earthlink.net>, Timothy Rue
><threeseas at earthlink.net> writes
>>In reviewing over many messages in the autocoding and related threads and
>>seeing the evidence of contridictions, hypocracy, irrational and
>>illogical responses......
>>
>>Its damn clear that there is a need to remove the sort of people posing to
>>be professional who are in essence just good liars, abstraction
>>manipulators, from the industry.
>>
>>Who knows, maybe that stench will be enough to motivate me see that it
>>gets done, gets removed.
>>
>>Maybe I should be more appriciative of the stench if it so motivates me.
>>
>>Anyone want to add anything to this?
>>
>all your ideas are belong to you

No, that's:

all your (tim's) idea are belong to *us*.

<evil chuckle> he he he

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From martin at v.loewis.de  Tue Feb 12 19:57:33 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 01:57:33 +0100
Subject: how to compare two strings?
References: <7xofivazj1.fsf@ruckus.brouhaha.com> <slrna6i71s.597.scarblac@pino.selwerd.nl>
Message-ID: <m38z9yw64i.fsf@mira.informatik.hu-berlin.de>

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

> >   def scmp(a,b):
> >     if a<b: return -1
> >     if b<a: return 1
> >     return 0
[...]
> Uhm. scmp() is the same as the builtin cmp(), as far as I can see.

Almost, yes. It performs two comparisons, though, whereas the builtin
performs only one (in general).

Regards,
Martin


From phd at phd.pp.ru  Mon Feb  4 09:25:54 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 4 Feb 2002 17:25:54 +0300
Subject: WANTED: IBM 1401 AUTOCODING WAREZ [was: Autocoding - the resistance  to expect from psuedo programmers]
In-Reply-To: <3c5e885b.12615968@news.t-online.de>; from gerson.kurz@t-online.de on Mon, Feb 04, 2002 at 01:12:14PM +0000
References: <mailman.1012829589.12111.python-list@python.org> <3c5e885b.12615968@news.t-online.de>
Message-ID: <20020204172554.C11454@phd.pp.ru>

On Mon, Feb 04, 2002 at 01:12:14PM +0000, Gerson Kurz wrote:
> On Mon, 4 Feb 2002 08:32:34 -0500 (EST), Steven Majewski
> <sdm7g at Virginia.EDU> wrote:
> 
> >HEY DUDEZ !
> > I WANT TO GET INTO THIS AUTOCODING THING.
> > ANYONE KNOW WHERE I CAN GET SOME IBM 1401 WAREZ ?
> >-- MULTICS-MAN

   ROFL! :))))))))))

> DUDE, I THINK YOU GOT IT ALL WRONG: AUTOCODING MEANS CODING WHILE
> DRIVING IN AN AUTOMOBILE, KIND OF DRIVE-BY-SHOOTING FOR GEEKS.

   This wonderful thread is continueing to bring sweet fruits! :) Please go
on! :)))

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



From fperez528 at yahoo.com  Mon Feb  4 13:21:01 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 04 Feb 2002 18:21:01 +0000
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu>
Message-ID: <a3nc45$ett$1@peabody.colorado.edu>

John wrote:

> Hello,
> 
> Is there any random number generator in Python, that allows users to specify
> "skewness"? Thanks!

Look at the rng's in Numeric, there are various distributions available with 
many control parameters.

f


From fperez528 at yahoo.com  Mon Feb 25 14:44:41 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 25 Feb 2002 19:44:41 +0000
Subject: Can I import an arbitary file?
References: <9d5509fa.0202251811.11837b04@posting.google.com>
Message-ID: <a5esub$oau$1@peabody.colorado.edu>

Peter Ballard wrote:

> I've written a program (an SRAM generator, if you must know) which
> has some configuration info (for different SRAM sizes etc.) in a separate
> file. This configuration file is python code.
> 
> What I want to do is import this file. The problem is, this file could
> be any name at all. What I want to do is parse my command line, and
> find out the name of my configuration file, and then import the file of
> that name. Is that possible?

look at __import__
 
> (Obviously I can read in this arbitrary file and parse it,
> but to actually execute it would be much neater).
> 
> Or is there any other way I can read in a file of an arbitary name,
> and execute its code?

look at execfile




From sholden at holdenweb.com  Mon Feb 11 07:56:40 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 07:56:40 -0500
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com> <7xlme1hv0y.fsf@ruckus.brouhaha.com> <3C66EA99.B22478E3@alcyone.com> <roy-F1119D.17441210022002@news1.panix.com>
Message-ID: <zoP98.149397$KM2.6703559@atlpnn01.usenetserver.com>

"Roy Smith" <roy at panix.com> wrote in message
news:roy-F1119D.17441210022002 at news1.panix.com...
> Erik Max Francis <max at alcyone.com> wrote:
> > That is simply not comparable.  The purpose of split is to divide up a
> > string according to a delimiter.  Asking for it to split a string given
> > a null delimiter doesn't make much sense at all;
>
> I think it makes plenty of sense.
>
I don't .

[legal null constructs]
>
> You could make rational arguments that none of them "make much sense at
> all", and that they are all likely to be errors, but you'd get arguments
> that to outlaw them would destroy the orthogonality of the language.
>
> But I think the strongest argument in favor of allowing '' as an argument
> to split() is the join() symmetry argument.

In that case, since a*0 is legal, presumably you want to make division by
zero legal for symmetry, too?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From simon.budig at unix-ag.org  Wed Feb  6 17:19:19 2002
From: simon.budig at unix-ag.org (Simon Budig)
Date: 6 Feb 2002 23:19:19 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net> <3c5fe63f$1@si-nic.hrz.uni-siegen.de> <upu3kovyv.fsf@python.net> <3c5ff6e8$1@si-nic.hrz.uni-siegen.de> <uheowosxs.fsf@python.net> <3c60000a$1@si-nic.hrz.uni-siegen.de> <u8za6pskh.fsf@python.net>
Message-ID: <3c61abe7$1@si-nic.hrz.uni-siegen.de>

In article <u8za6pskh.fsf at python.net>, Michael Hudson wrote:
> Simon Budig <simon at ping-pong.budig> writes:
>> In article <uheowosxs.fsf at python.net>, Michael Hudson wrote:
>> > Well, it can't go into 2.2.  If you mean 2.2.1, that's a little
>> > greyer; there might be people out there who rely on the current
>> > behaviour.
>> 
>> Hmm. Maybe it is possible to export this variable as an optional
>> parameter to readline.set_completer with the default on the
>> current behaviour?
> 
> But then that's a new feature...
> 
>> I think it is really annoying, that readline in the current
>> python-module implementation insists on completing filenames when
>> the completer returns no matches.
> 
> Sorry.  But I don't think you can claim it's a pure bug, and so it's
> not 2.2.1 material.

Actually I think, that it is a bug, because it makes it impossible
to write a completer function, that avoids the filename completion.
Imagine the rlcompleter: It is an convenient way to have completion
on the Python commandline. But if you start typing a name that is
not (yet) in the global dict and type <TAB> to verify this fact,
readline decides to complete with a filename, which is absolutely
meaningless in this context - AND YOU CANNOT SWITCH THIS OFF!

However, I am not familiar with the python guidelines/procedures for
adding new features and certainly dont think that this is important
enough to break these guidelines.

I am a little bit astonished, that it seems possible to add this to
2.1.x, but not to 2.2.x. Sounds weird to me. For Gimp development
we don't fix some misbehaviours in old versions, but are happy to
fix them in new versions...

Oh, well. Maybe I am missing something obvious here...   :-)

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From hfoffani at yahoo.com  Thu Feb  7 15:55:11 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Thu, 7 Feb 2002 21:55:11 +0100
Subject: IPX through Python
References: <mailman.1013104994.13630.python-list@python.org>
Message-ID: <20020207155523.526$x9@news.newsreader.com>

> Does enybody knows or have an example of  how to implement IPX using
Python?
> I need to send some simple packets using IPX.

You want to search google with "IPX" and "library".
It seems to be several free/commercial libraries available. You'll
need
to make a wrapper for Python.

-Hernan





From eric.brunel at pragmadev.com  Wed Feb 20 10:52:14 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Wed, 20 Feb 2002 15:52:14 +0000
Subject: How to hide Tk root window when using only canned dialogs?
References: <f89f6348.0202200642.7d38ae20@posting.google.com>
Message-ID: <a50cvo$2k00$1@norfair.nerim.net>

<posted & mailed>

Paul Soper wrote:
> The example script below works fine but the Tk window is visible
> throughout.  Is there a way to make it invisible?

Yep: create it explicitely:

from Tkinter import Tk
root = Tk()

and then make it disappear:

root.withdraw()

It should do the trick.

HTH
 - eric -


From mgash at trisystems.co.uk  Fri Feb 22 04:24:21 2002
From: mgash at trisystems.co.uk (Mark Gash)
Date: Fri, 22 Feb 2002 09:24:21 -0000
Subject: Adding GUI to application
Message-ID: <31575A892FF6D1118F5800600846864DCC76B5@intrepid>

	Simon Brunning wrote:

	> Starting the threads up isn't too complicated - see the effbot's
"Python
	> Standard Library" for an example. (You can look at my copy - I'm
only up the
	> stairs, after all. <wink>)

	Yes, the threading module is pretty simple to use.


	> But then you need to get your two threads to communicate. I don't
know how
	> to do this, but I'd have a look at the Queue module if I were you.

	Queue makes inter-thread communication a snap.  In order for two
threads to send
	messages back and forth, you simply set up two Queues, one in each
direction.
	Drop 'message objects' (which can be anything that makes sense for
your
	application) into one end, pull them out the other in a different
thread, and
	everything pretty much just works--the Queue takes care of almost
all of the
	synchronization issues.

Many Thanks for the pointer... Has anybody got any working examples of
Queues, for this situation???

Kind Regards,

Mark Gash
Consultant
TriSystems Ltd.
020 7264 0440
mgash at trisystems.co.uk






-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From tim at vegeta.ath.cx  Mon Feb  4 04:55:41 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Mon, 04 Feb 2002 09:55:41 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx> <mailman.1012707428.2108.python-list@python.org>
Message-ID: <slrna5smo9.srn.tim@vegeta.ath.cx>

dman <dsh8290 at rit.edu> graced us by uttering:
> On Sun, Feb 03, 2002 at 01:40:12AM +0000, Tim Hammerquist wrote:
[ snip ]
> I have trouble with that too.  My preference (FWIW, probably nothing) :
> 
> if ( some_condition() )
> {
>     for (i = first_i();i <= 10; i++)
>     {
>         initialize_some_stuff();
>         while ( some_other_condition() )
>         {
>              do_some_more_stuff()
>         }
>     }
> }
> 
> 
> Notice how the matching braces are horizontally aligned and at the
> same level as the line that initiates the bock?  If the braces are
> removed, and colons added it will look like python (with some
> extraneous parens).

Ah yes. BSD-style. My second choice, and preferred by one of my college
professors.

>| ...yet I have no trouble parsing
>| 
>| if ( &some_condition ) {
>|     for $i (&first_i .. 10) {
>|         &initialize_some_stuff;
>|         &do_some_more_stuff while &some_other_condition;
>|     }
>| }
> 
> This is hard for me because I can't see what the '}' goes with.
> Combine this with leaving out the braces on one-line compound
> statements, and I'll have real problems.

A habit I got out of...

ie:

    /* BAD! */
    if(expression)
        stuff();

    /* Better */
    if(expression) {
        stuff();
    }

...much easier and less error-prone to add one+ more statements to the
if() block. I like that Perl removed that particularly error-prone piece
of syntactic sugar...

Tim Hammerquist
-- 
guru, n: a computer owner who can read the manual.


From mlh at vier.idi.ntnu.no  Tue Feb 12 17:29:39 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Tue, 12 Feb 2002 22:29:39 +0000 (UTC)
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com>  <mailman.1013544745.1018.python-list@python.org>
Message-ID: <slrna6j5qj.gku.mlh@vier.idi.ntnu.no>

In article <mailman.1013544745.1018.python-list at python.org>, Wayne Pierce wrote:
>
>What is your target platform?
[...]
>I typically use Qt for most of my development then if necessary use 
>Gordon McMillan's Installer[2] to package it up.  I've also seen 
>reference to AnyGUI[3], which aims to allow you to program to any GUI 
>with a common API (..at least that's what I think it does..).

Correct. Or *almost* any gui, that is ;)

The latest version doesn't really have any support for graphics,
though. (There is some undocumented support for Jython and Tkinter
graphics, but...)

All that will improve in release 0.2, of course :)

> I haven't tried it yet though.

>[3] http://www.anigui.org/

That should be

     http://www.anygui.org

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From robin at jessikat.fsnet.co.uk  Sat Feb 16 06:17:31 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 16 Feb 2002 11:17:31 +0000
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
 <b16e4ef7.0202130950.67f4004d@posting.google.com>
 <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4>
 <52e5ab5f.0202141045.5867bfbb@posting.google.com>
 <Xns91B71A6CA36C3RASXnewsDFE1@130.133.1.4>
Message-ID: <tnkaeJAL$jb8EwGo@jessikat.fsnet.co.uk>

.
.....
>
>Also... what's this 'filename completion' you speak of? Are you talking 
>about windows? (Though i s'pouse it may have been added past 98, i've 
>never known nor can find a key that does this...)
>
well in cmd.exe (WinNT Win2k etc) you can configure a key to auto-
complete filenames just like bash (I use the tab key). In the standard
file dialogs entering a prefix usually results in a dropdown list of
files/directories beginning with that prefix; you can then use the down
arrow key to select and carriage return to finish.
-- 
Robin Becker


From dkahle at interactdirect.com  Fri Feb 22 16:38:40 2002
From: dkahle at interactdirect.com (Darcy Kahle)
Date: Fri, 22 Feb 2002 16:38:40 -0500
Subject: hep needed with pipes (popen4)
Message-ID: <iUyd8.18492$X2.205012@nnrp1.uunet.ca>

I need some help with pipes in python 2.1.

I am running a comand using os.popen4().  The command produces output until
the user hits control-C, at which time, it prints a summary, and exits.  I
want to send that character down the pipe to it without the user doing so
(control-c kills the script, by default).  I tried sending "\x003", which to
the best of my knowledge, is the hex code for it, but it didn't work.

If this is not possible, then I could do the same by redirecting the stdin
to /dev/null, but if I do that at the start, the command will stop
immediately.

Does anyone know the correct character sequence to send control-C to the
command or how to redirect the command's stdin to /dev/null after it is run?

I was toying with os.dup2(), but I suspect that that will redirect the file
handle inside my program to /dev/null, leaving the stdin of the command
intact.  I was hoping that it would link the two file handles together,
thereby creating a bridge between the two.

Any assistance that you can provide, will be greatly appreciated.

Darcy




From gh_pythonlist at gmx.de  Tue Feb  5 16:47:42 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 5 Feb 2002 22:47:42 +0100
Subject: distutils, C extensions, Borland compiler, wxPython
In-Reply-To: <C3Y78.38682$Ni2.219590@news-server.bigpond.net.au>
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk> <mailman.1012939329.5851.python-list@python.org> <C3Y78.38682$Ni2.219590@news-server.bigpond.net.au>
Message-ID: <20020205214740.GA1434@lilith.hqd-internal>

Le 05/02/02 ? 21:23, Neil Hodgson ?crivit:
> Gerhard H?ring:
> 
> > I think I have heard something to the effect, that on Windows, DLLs
> > can't link back to .EXEs, only to other DLLs.
> 
>    No, DLLs can link back to EXEs although you may have to force (link
> /force) the link to complete and/or link the DLL again after linking the
> EXE. [...]

So it would be possible to get rid of the python{major}{minor}.dll and
still be able to dynamically load .pyd extension modules?

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.5 ?C      Wind: 1.5 m/s



From NOSpawnPAM at lightspawn.org  Fri Feb  1 18:26:25 2002
From: NOSpawnPAM at lightspawn.org (pawn)
Date: 1 Feb 2002 15:26:25 -0800
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com>
Message-ID: <4b4c5d99.0202011526.23585e16@posting.google.com>

bash$ chmod 755 test.cgi
bash$ lynx -dump http://(domain)/cgi-bin/test.cgi
   65534 65533
bash$ chmod 4755 test.cgi
bash$ lynx -dump http://(domain)/cgi-bin/test.cgi
   65534 65533

as for the OS:
bash$ -uname
FreeBSD addr12.addr.com 4.4-STABLE FreeBSD 4.4-STABLE #0: Thu Nov  8
17:07:35 PST 2001
root at addr24.addr.com:/usr/src/sys/compile/ADDRKERN  i386

I was first alerted to the problem by the CGI's refusal to read/write
certain files of mine which are not world read/writeable.

PS: nothing to do with this, but I'm even more outraged by the fact
that addr.com seem to have a support policy that states 'if you can't
solve the problem, don't answer at all'. I've seen that a couple of
other places too, never in writing of course.

Jonathan Hogg <jonathan at onegoodidea.com> wrote in message news:<B88081C5.40F9%jonathan at onegoodidea.com>...
> 
> Are you sure the scripts aren't running as you? How are you testing this?
> 
> Note that setting an executable's setuid bit will change the "effective" uid
> of the resulting process, not the uid (see 'os.get*e*uid()' above). This
> means that the process may use that user's permissions.
> 
> The webserver doesn't contain a Python interpreter does it? Perhaps it's
> directly executing the scripts instead of 'exec'ing them?
> 
> Jonathan


From akuchlin at dust.mems-exchange.org  Mon Feb 25 18:19:03 2002
From: akuchlin at dust.mems-exchange.org (A.M. Kuchling)
Date: 25 Feb 2002 23:19:03 GMT
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com> <a5dia2.3vs4847.1@kserver.org> <3c7abd92_1@corp.newsgroups.com>
Message-ID: <slrna7lhj5.qon.akuchlin@dust.mems-exchange.org>

In article <3c7abd92_1 at corp.newsgroups.com>, Joshua Muskovitz wrote:
> They are the tech faculty for the local tech school.  It is meant to
> demonstrate my ability to teach (something).  I have some ideas, but was
> looking for a few suggestions.  I expect that I won't be able to determine

How about choosing some interesting, moderately complicated algorithm?
Say, graph traversal, perfact hashing, or something like that?

--amk                                                  (www.amk.ca)
Power is the only freedom that I seek.
    -- Omega, in "The Three Doctors"


From mhammond at skippinet.com.au  Mon Feb  4 02:42:24 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 07:42:24 GMT
Subject: win32api.RegQueryValue fails
References: <a3gv20$fv8$05$1@news.t-online.com>
Message-ID: <3C5E3BB0.3010603@skippinet.com.au>

Achim Domma wrote:
> Hi,
> 
> still not having solved my problem importing win32com in an embedded
> interpreter , I tried to debug where the error comes from. My real problem
> is that win32api.RegQueryValue fails.
> 
> The following code works without problems:
> 
> import win32api
> print win32api.GetFullPathName(".")
> 
> but
> 
> import win32api
> keyName = "SOFTWARE\\someKey"
> key = win32api.RegOpenKey(win32con.HKEY_LOCAL_MACHINE , keyName, 0,
> win32con.KEY_ALL_ACCESS)
> win32api.RegQueryValue(key, "Root")
> 
> fails with a 'TypeError: this constructor takes no arguments'. If I print
> the handle returned by RegOpenKey it is not Null, so it seems to be ok. A
> look at the implementation of RegQueryValue (head-revision in cvs) showed
> that it is implemented straight forward and I have no idea what should go
> wrong.
> 
> I have to use ADO from my embedded python, which will not work without
> win32com, so every little hint is welcome.

Works for me, I'm afraid:

 >>> keyName = "SOFTWARE\\Python\\PythonCore\\2.2\\PythonPath"
 >>> key = win32api.RegOpenKey(win32con.HKEY_LOCAL_MACHINE , keyName, 
0,win32con.KEY_ALL_ACCESS)
 >>> win32api.RegQueryValue(key, "")
'f:\\src\\python-2.2\\lib;f:\\src\\python-2.2\\pcbuild'
 >>>

Can you send a complete script that fails, using a registry key that I 
am likely to have (eg, a Python, standard windows, IE, Office etc setting)

Mark.



From gcordova at hebmex.com  Fri Feb 22 17:45:29 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 16:45:29 -0600
Subject: is their any limit on the size of the data passed to a progra
	   m via popen2?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE6@SRM201002>

Yeah, Jeff Shannon wrote something about
pipes hanging in a .read() operation. :-/

So... any other clues yet?

-gus

> -----Original Message-----
> From: husam [mailto:h.jehadalwan at student.kun.nl]
> Sent: Viernes, 22 de Febrero de 2002 04:48 p.m.
> To: python-list at python.org
> Subject: Re: is their any limit on the size of the data passed to a
> progra m via popen2?
> 
> 
> Gustavo Cordova wrote:
> 
> >>  pdbdir = os.popen('ls /home/data/pdbs') # if this directory 
> >>contains 
> >>10000 files, the code does not work.
> >>
> >>files=pdbdir.readlines()
> >>pdbdir.close()
> >>(o,i) = popen2.popen2("/home/programs/profit ")
> >>
> >>i.write("reference /usr/local/3D_Dock/progs/Complex_1g.pdb ") 
> >># This the 
> >>reference molecule.
> >>
> >>for PDB in files:
> >>     i.write(" mobile /home/data/pdbs/" + PDB + "\n") # A new 
> >>molecule.
> >>     i.write("fit\n")		# compares reference 
> and new molecules.
> >>
> >>i.close()
> >>result= o.read()
> >>o.close()
> >>
> >>
> > 
> > You read from the input pipe only upon writing all your data,
> > but maybe the pipe is full?
> > 
> > How about changing the loop with this:
> > 
> > 
> > log = StringIO.StringIO()
> > 
> > for PDB in files:
> >   i.write("mobile /home/data/pdbs/%s\n" % PDB)
> >   i.write("fit\n")
> >   log.write(o.read())
> > 
> > i.close()
> > o.close()
> > 
> > 
> > So now your results are in the log object.
> > 
> > -gus
> > 
> > 
> 
> hi Gustaveo,
> You'r solution did not work either. The code hangs at the first loop, 
> specifically at "log.write(o.read())".
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 



From fperez528 at yahoo.com  Sun Feb 17 15:20:52 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 17 Feb 2002 20:20:52 +0000
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com>
Message-ID: <a4ps2l$5jo$1@peabody.colorado.edu>

sameer wrote:

> sameer_ at email.com (sameer) wrote in message
> news:<6bd9f01b.0202160255.77244c9e at posting.google.com>...
>> Here's the situation:
>> I have a python package (or two) to be installed on a windows machine
>> (for now) which has python installed.  I am using distutils to create
>> a windows installer which installs my python packages just fine.  Now,
>> I want to also create some shortcuts on the Desktop as well as the
>> Start Menu.  As of now, distutils doesn't allow me to do either of
>> those things.  What are my options here to do either or both of these
>> things.  Can the distutils package be modified easily to provide this
>> kind of support on windows?  Or is there another python package
>> available that would allow me to create the desktop and Start Menu
>> shortcuts?
> 
> it's a pity no one has replied.  I suppose there are no sane options for
> this.

Sorry, I hadn't seen this.

Go to http://www-hep.colorado.edu/~fperez/ipython and download the program. 

You'll find besides the standard setup.py a setup_Windows.py which does 
exactly what you're looking for. Feel free to use it/modify it. 
Changes/fixes/suggestions welcome.

Cheers,

f.


From bougle at gluce.com  Wed Feb 27 12:35:47 2002
From: bougle at gluce.com (bougle gluce)
Date: Wed, 27 Feb 2002 09:35:47 -0800
Subject: Unintelligible Lambda Tricks
References: <3c74c211.20012718@news.isar.de> <mailman.1014338076.13501.python-list@python.org> <qnbb7u0hoqpi5s7vg8hg9k9fh0ufijvts3@4ax.com>
Message-ID: <wijf8.151$io1.168438@news.uswest.net>

> Really? I was thinking of the person who wrote it.
>
> I vote for getting some of your homies and going midieval on him.

I'm down wit' dat', dawg! :)




From michael.amrhein at t-online.de  Thu Feb  7 13:10:19 2002
From: michael.amrhein at t-online.de (Michael Amrhein)
Date: Thu, 7 Feb 2002 19:10:19 +0100
Subject: money class?
References: <mailman.1013051771.7427.python-list@python.org>
Message-ID: <a3ufso$rna$07$1@news.t-online.com>

"Mark McEahern" <mark at mceahern.com> schrieb im Newsbeitrag
news:mailman.1013051771.7427.python-list at python.org...
> Without starting an endless thread on how many floating point angels can
> dance on the head of a pin, I'd like to ask:
>
> Does anybody know of a nice Money class they'd be willing to share, point
> to?
>
> (I did a little bit of searching, but found nothing obvious in the 5-10
> minutes I spent.)
>
> I don't think it'd be too hard to write one myself, but I figured I'd see
> what's out there first.  I don't need something that handles different
> currency.  US Dollars is good enough.  I suppose I'll be taken to task (or
> ignored) for not specifying exactly what I need or why I need it.  What
can
> I say?  I'm lazy.  ;-)  Does this suffice?
>
> It should "Do the right thing."
>
> Thanks,
>
> // mark
>
>
Hi Mark,
I've written a class named 'fixed' with portions adopted from
Tim Peter's FixedPoint.py, but a slightly different constructor
(no default precision; precision calculated from given value,
if no precision given) and some additional functions.
Find fixedpoint.py enclosed and use it if you like.
Cheers,
Michael

The news server seems to refuse any posting with attachments.
So, I'll send it to you via e-mail. Sorry.




From eddieborder2001 at yahoo.co.uk  Wed Feb  6 15:19:13 2002
From: eddieborder2001 at yahoo.co.uk (Ed)
Date: 6 Feb 2002 12:19:13 -0800
Subject: get all paths
Message-ID: <d5534716.0202061219.60f2af7e@posting.google.com>

Hi
I'm trying to create a treeview widget, but i am having problems
getting alll directories and files from a specified directory

i.e

getList("C:/") should return all directories and files from C:/
onwards whereas
getList("C:/Python") should return all from that onwards

any help would be appreciated

Ed


From wwaldo4 at cox.net  Thu Feb 21 11:58:02 2002
From: wwaldo4 at cox.net (william)
Date: Thu, 21 Feb 2002 16:58:02 GMT
Subject: Intalling Python 2.1 and Image Lib 1.1.2
Message-ID: <uG9d8.5157$Ep1.182834@news2.east.cox.net>

I am running on RedHat 7.2
I have installed Python2.1 from the rpm.  The rpm installes it in
/ur/lib/python2.1

Installing the Image Library 1.1.2:

1.  I copy the tar file to /usr/lib/python2.1  I untar it to
/usr/lib/python2.1/Imaging-1.1.2

2.  I change directory to /usr/lib/python2.1/Imaging-1.1.2/libImaging, and
then run './configure'

3.  I then run 'make'.   No problems so far.

4.  After compilation, I run   'make check'  and get:
     ./coretest
     >>> processing lena.ppm
     >>> elapsed time: 10000
     >>> as far as we tested, everything seems to be ok....
     xv test.ppm
     make:  xv: Command not found
     make: [check] Error 127 (ignored)

5.  I then save /usr/lib/python2.1/Imaging-1.1.2/setup.ini as
/usr/lib/python2.1/Imaging-1.1.2/setup

6.  Next I run   'make -f Makefile.pre.in boot'   from
/usr/lib/python2.1/Imaging-1.1.2/  - no errors noted.

7. Then 'make'.  Get error:
    /usr/bin/Id: cannot find -Itc18.0
   collect2: Id returned 1 exit status
   make: *** [_imagingtk.so] Error 1
  Which i assume to be no probelm.

8.  Now the readme file statest to run:
     $  PYTHONPATH=.:./PIL ; export PYTHONPATH
     $  python  (this command executes python1.5.2 which i don't want to
use)
     >>>  import_imaging
     >>>  import Image
     No errors noted.

9.  Now when I try to run it using python two :
     $  PYTHONPATH=.:./PIL ; export PYTHONPATH
     $  python2
     >>>  import_imaging
     WARNING: Python C API version mismatch for module _imaging:
        This Python has API version 1010, module _Imaging has version 1007.

After this long post, anyone see where i went wrong?





From deathtospam43423 at altavista.com  Mon Feb 11 18:32:08 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 11 Feb 2002 23:32:08 GMT
Subject: Yeeeeesss!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254> <FrY98.156468$KM2.7108035@atlpnn01.usenetserver.com>
Message-ID: <Xns91B366E5B8D93AusYourStandingInIt@210.49.20.254>

Thankyou!!!!!

"Steve Holden" <sholden at holdenweb.com> wrote in
news:FrY98.156468$KM2.7108035 at atlpnn01.usenetserver.com: 

........
> So I'm guessing if
> you use 
> 
> #! C:/Python22/python
> 
> all will be well. Xitami's a great little server (with a big heart).
> Enjoy! 
> 
> regards
>  Steve
> --
> Consulting, training, speaking: http://www.holdenweb.com/
> Author, Python Web Programming: http://pydish.holdenweb.com/pwp/
> 
> "This is Python.  We don't care much about theory, except where it
> intersects with useful practice."  Aahz Maruch on c.l.py
> 
> 
> 
> 
> 



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From duncan at NOSPAMrcp.co.uk  Fri Feb  1 04:21:15 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Fri, 1 Feb 2002 09:21:15 +0000 (UTC)
Subject: confused by isinstance()
References: <3c5717bb$0$23518$626a54ce@news.free.fr> <a3b9ib$61c$1@panix2.panix.com> <3c5a3352$0$4604$626a54ce@news.free.fr>
Message-ID: <Xns91A85C660FC3Bduncanrcpcouk@127.0.0.1>

"Francois Petitjean" <littlejohn.75 at free.fr> wrote in 
news:3c5a3352$0$4604$626a54ce at news.free.fr:

> (1) It is trivial as it has :
> only one module
Two modules, Foo and __main__
> only one class
Two classes, one in each module. Foo.Foo and __main__.Foo

> only one instance
> and it uses only builtins and doesn't use 2.2 features.
> 

When you run a script, it executes the code in the context of the module 
called __main__. When you import a module it has a separate context even if 
the script and the module came out of the same source file. Importing a 
module executes each line in the file, even if you already ran most of them 
as a script.

Every time you execute a class statement, even if the class name is the 
same, you create a new class.


-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From fredrik at pythonware.com  Sun Feb 17 03:24:15 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 17 Feb 2002 08:24:15 GMT
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <m34rkhbvw3.fsf@mira.informatik.hu-berlin.de> <3C6E943B.BC7CCF9@earthlink.net>
Message-ID: <PMJb8.26248$l93.4947655@newsb.telia.net>

Hans Nowak wrote:
> Either way, my question is, is there a way to parse the XML
> *and* somehow preserve the newlines in the string?

as Paul pointed out, XML parsers are supposed to normalize
whitespace in attribute values.

if you want non-standard XML behaviour, you have to use
lower-level toolkits.  I suggest building your own DOM builder
based on the sgmlop scanner:

    http://effbot.org/sgmlop.htm

</F>




From precision at devrandom.net  Fri Feb  8 14:14:47 2002
From: precision at devrandom.net (Uriah Welcome)
Date: 8 Feb 2002 11:14:47 -0800
Subject: SourceForge Python foundry ????
References: <3c6301ac.188163695@news.laplaza.org> <a40hh0$1bdtaj$1@ID-7529.news.dfncis.de>
Message-ID: <65c17385.0202081114.14d2a605@posting.google.com>

"Karl M. Syring" <syring at email.com> wrote in message news:<a40hh0$1bdtaj$1 at ID-7529.news.dfncis.de>...
> "Mats Wichmann" <mats at laplaza.org> schrieb
> > What's up with the Python foundry on SourceForge? It's not listed in
> > any of their lists of foundries, although if you plug in "python" in
> > the standard path they use for a foundry it comes up fine...
> 
> Perhaps, that is because SF still has a grubby v1.5.2 installed?

As soon as debian 'stable' includes a newer version we'll be happy to include it.


From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 19:03:22 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 18:03:22 -0600
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <Xns91B352E74A030AusYourStandingInIt@130.133.1.4>
Message-ID: <002a01c1b358$b19af120$0101010a@local>

----- Original Message -----
From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>
> Thanx. I'm working out how to write a script at the moment, following the
> tutorial at http://www.devshed.com/Server_Side/Python/CGI/ . May I get
back
> to you if I have any difficulties?

Best to reply on the list (as you have) so the largest number of
Pythoneers (never liked Pythonista myself) can reply to you.

You are fortunate that you chose Python, not only because it is
a most excellent language, but also because this is the nicest
newsgroup/mailing list around.  We don't mind newbies, not like
some lists...





From hmartires at cultalg.pt  Thu Feb 21 09:38:21 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Thu, 21 Feb 2002 14:38:21 -0000
Subject: Suggestion for best intro to Python
References: <c510a321.0202210608.1acc4d29@posting.google.com>
Message-ID: <IB7d8.202$IZ4.2654278@newsserver.ip.pt>

Book: Core Python Programming (Chun)

"Bill Hackmann" <hack at socket.net> wrote in message
news:c510a321.0202210608.1acc4d29 at posting.google.com...
> What book/web site/online document/something else would be best as a
> source to learn Python for an experienced programmer?
>
> Thanks,
> Hackmann
>




From philh at comuno.freeserve.co.uk  Sun Feb 10 20:42:36 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 11 Feb 2002 01:42:36 +0000
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <eppstein-A87A6C.14430710022002@news.service.uci.edu>
Message-ID: <slrna6e8cc.sme.philh@comuno.freeserve.co.uk>

On Sun, 10 Feb 2002 14:43:07 -0800, David Eppstein <eppstein at ics.uci.edu> wrote:
>In article <eSz98.4759$zb.260473 at news1.east.cox.net>,
> "Jeff Hinrichs" <jlh at cox.net> wrote:
>
>> Pythonista - Pie thun ee sta
>> I took it as a unintended play on Sandanista, which is a group of  "freedom
>> fighting" rebels in Nicaragua.
>
>I think it's a normal Spanish suffix,

Indeed so, -ista is the Spanish equivalent of English -ist.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From loewis at informatik.hu-berlin.de  Mon Feb 18 11:50:47 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 18 Feb 2002 17:50:47 +0100
Subject: Subclassing a read-only class
References: <mailman.1014046530.13105.python-list@python.org>
Message-ID: <j4ofimybrs.fsf@informatik.hu-berlin.de>

Gerhard H?ring <gh_pythonlist at gmx.de> writes:

> For compatibility testing, I'd like to put an additional method into
> window object I get from the curses module. Unfortunately, if I try to
> do so, I get:
> 
>     TypeError: object has read-only attributes
> 
> I'm pretty sure there is a solution, probably with a proxy object. But
> currently I just don't get it.
> 
> Any ideas?

You haven't described the compatibility testing in detail. However, it
might be that deriving from Bastion.BastionClass could do the trick:
Just add the additional functions in the subclass, and implement a
_get_ that delegates everything to the window object.

Regards,
Martin



From gcordova at hebmex.com  Tue Feb 19 10:37:22 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 19 Feb 2002 09:37:22 -0600
Subject: newbie: Some file object tips?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AB0@SRM201002>

> 
> So does this command ....
> 
> > lines = open("log.txt").readlines()
> 
> ... actually load the whole log into RAM.
> 

Yes it does.

But then, what's actually flawed is the algorithm,
not the program, or the tools.

You need a new way to store the message which will
let you read only part of the log into memory.
Right?

Well, sorry. Maybe "flawed" is a too harsh qualification,
but overly simple. Ok.

Hmmm... What would I do?

I wouldn't read the "last 'n' lines", as you put it,
but something easier to calculate, "the last 'n' K of text",
which is more "file-like".

So, to read the lines in the last "n" Kb of text:

====== SOME CODE ======
def GetLog(filename, maxSize=None):
	log = open(filename)
	if maxSize:
		# Seek to end of file.
		log.seek(0,2)

		# Obtain current position index.
		fileLen = log.tell()

		# If file is larger than maxSize,
		# seek to (EOF - maxSize).
		if maxSize < fileLen:
			log.seek(fileLen-maxSize)

			# Skip one line.
			log.readline()

		# Or return to the file beginning.
		else:
			log.seek(0)

	# Return all lines.
	return log.readlines()
====== SOME CODE ======

So now you only get the last maxSize Kb lines-worth of
lines from the log file.

Maybe it's a little clumsy, I bet someone can clean it
up a bit.

Good luck :-)

-gustavo



From neal at metaslash.com  Tue Feb 26 20:40:50 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Tue, 26 Feb 2002 20:40:50 -0500
Subject: ANN: PyChecker Mailing List
Message-ID: <3C7C3922.5CDA0B5E@metaslash.com>

A mailing list for PyChecker has been created:

	pychecker-list at lists.sourceforge.net

This list is meant for general discussion of PyChecker.
It can also be used for general discussion of static analysis
issues for python.

If you would like to subscribe to the list, go to this page:

	http://lists.sourceforge.net/lists/listinfo/pychecker-list

Neal



From aahz at panix.com  Wed Feb 20 15:54:30 2002
From: aahz at panix.com (Aahz Maruch)
Date: 20 Feb 2002 12:54:30 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com>
Message-ID: <a512e6$dqk$1@panix2.panix.com>

In article <de3fc1ef.0202201233.2228e5b1 at posting.google.com>,
William Tanksley Google <wtanksley at bigfoot.com> wrote:
>rmunn at pobox.com (Robin Munn) wrote:
>>
>> * Batteries included. This is the big one, which makes Python more than
>
>My friend was griping about one battery that Basic comes with but
>Python doesn't (at least not that I've seen): graphics.  Yes, I know
>there's no single way to draw in Basic, but every version has a
>platform-specific graphics library.
>
>Does Python?  If so, I've never seen it.  I'd like to.

Python pretty much comes with Tkinter.  What's wrong with that?  Why do
you need a platform-specific library for "batteries included"?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"Not everything in life has a clue in front of it...."  --JMS


From dgallion1 at yahoo.com  Mon Feb 11 00:28:56 2002
From: dgallion1 at yahoo.com (Darrell)
Date: Mon, 11 Feb 2002 05:28:56 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com> <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com>
Message-ID: <sEI98.68031$dh.16756602@typhoon.nyroc.rr.com>

Not sure what you want so here's what I have.

You can zero pad the hex digits instead of using hex()
>>> '%0.2x'%12
'0c'

>>> s1='ab'
>>> s2='abccc'
>>> map(max,zip(min(s1,s2).ljust(len(max(s1,s2))),max(s1,s2)))
['a', 'b', 'c', 'c', 'c']

--Darrell

dsavitsk wrote:

> the truncate part is the problem.  I am building VB Hex strings to send to
> a
> VB server.  so
> 
>>>> red = hex(56)[2:4]
>>>> green = hex(156)[2:4]
>>>> blue = hex(12)[2:4]
>>>> thestring = '&H' + blue + green + red
> 
> but here, since the blue string is only 1 character it throws the rest
> off.
> 
> -d
> 
> "Joshua Muskovitz" <joshm at taconic.net> wrote in message
> news:3c674cbd_2 at corp.newsgroups.com...
>> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>> >>> ''.join(map(max,zip('00','AA')))
>> 'AA'
>> >>> ''.join(map(max,zip('00','A')))
>> 'A'
>> >>> ''.join(map(max,zip('00','')))
>> ''
>>
>> Damn, I thought I had it.  :-)  Is there an variant of zip which doesn't
>> truncate?  BTW, this does work, provided that the input strings are the
> same
>> length, and even allows for more than two strings!
>>
>> >>> s = ['abte','scrt','sdfw','cxvs']
>> >>> ''.join(map(max,zip(*s)))
>> 'sxvw'
>>
>> --
>> # Joshua Muskovitz
>> # joshm at taconic.net
>> def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
>> lyyrs('Hire me!  I need the work!')
>>
>>
>>
>>
>> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
>> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
>> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----



From jkraska at san.rr.com  Sat Feb  9 13:07:09 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 09 Feb 2002 18:07:09 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <2106.802T789T14243611threeseas@earthlink.net> <3c647f02$0$22477$3c090ad1@news.plethora.net> <3665.805T682T7313596threeseas@earthlink.net> <3c6562fb$0$79557$3c090ad1@news.plethora.net>
Message-ID: <tkpa6u0djajha91t13qsal7e2kaqm1shs5@4ax.com>

>I object to the allegation that all moderated newsgroups are "censored" in
>any meaningful sense.  The ones I'm familiar with don't post porn spam,
>but that's about it.

Correct. A goodly number of moderated newsgroups use "verified opt-
in" moderation where after your first post or two, the moderators
put you on the automatic approval list. This magically manages to
exclude virtually all commercial spams, get rich quick schemes, and
desperate women in St. Petersburg looking for an American husband.

On these same groups, if you are a regular poster, you have to be quite
a bad boy to get censured.

C//



From bokr at oz.net  Fri Feb 15 15:06:27 2002
From: bokr at oz.net (Bengt Richter)
Date: 15 Feb 2002 20:06:27 GMT
Subject: ANN: PSF Contribution Documents
References: <mailman.1013769204.16499.python-list@python.org>
Message-ID: <a4jpo3$ilt$0@216.39.172.122>

On Fri, 15 Feb 2002 11:32:10 +0100, "M.-A. Lemburg" <mal at lemburg.com> wrote:

>The Python Software Foundation (PSF; http://www.python.org/psf/) 
>was created to (eventually) hold the intellectual property (IP)
>for the Python programming language.
>
>In order to be able to succeed in this, contributions to the
>Python programming language will have to be covered by legally
>valid contribution agreements. To make this process as simple
>as possible, the PSF members have come up with three different
>contribution agreements.
>
>The current versions of those agreements have been posted to
>a new mailing list at:
>
>    http://mail.python.org/mailman/listinfo/psf-forms-discuss
>
>If you want to participate in the final discussion of these
>documents, please sign up to the list.
>
I skimmed the agreements. Committers' agreements are a bit special
in that they cover access etc., but for declaring the legal copyright
status of a chunk of text, I wish there was something as apparently
simple as saying "I hereby put this in the public domain," but for
defining and transferring rights in detail.

What I'd like to see is an agreement/license/whatever-boilerplate
repository, kept as files with names in the form

<license version iso-date>.<small ID>.<MD5 digest of file read in binary>.txt

together with a law that would make it possible to include them
by reference and have the thus included text have the same legal
effect as if literally pasted in in the place of the referring line.

The law would let you say "All rights governed by "<repository file name>"
In place of the usual "All rights reserved."

If desired, the MD5 digest can easily be encoded as all alpha ([a-zA-Z])
in 24 characters by a simple base 52 trick on groups of hex bytes making
3 chars from 4. The '.txt' extension is to make it easy in that world.

But the main idea is a repository (with enough mirrors and random copies
that google can find the exact file by its unique name, which also
guarantees by the MD5 part of the name a way to check content integrity,
which in turn would seem to be a decent basis for legal effect, though
IANAL.

I just hate seeing source code with more boilerplate than substance.
I would think with the right wording in a repository file, a developer
should be able to confer ownership of copyright and licensing terms all
with one line (ok, it's a bit longish - maybe make "All rights ..." legal
in the immediately following line) in a posted source file, e.g.

# Copyright (C) 2002 by author.
# All rights governed by "2001-05-01.PSFLic2a.asdasdasdasdasdasdasdasd.txt"

Signing the referring source file with pgp would guarantee that the reference
line was an unaltered part of the whole. (That problem exists whether the
full copyright terms are literally pasted in or included by reference).

Making it easy to do a standard thing might make it more common and uniform.

The government could have an official boilerplate registry site at
the Library of Congress, so a particular mirror would not have to be mentioned.

Any IAALs out there care to comment?

Regards,
Bengt Richter


From tripps81 at yahoo.com  Sat Feb 16 02:27:27 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Sat, 16 Feb 2002 14:27:27 +0700
Subject: basic books/guides on multithreading programming?
In-Reply-To: <20020215152403.179d014b.logiplexsoftware@earthlink.net>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
 <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
Message-ID: <5.1.0.14.2.20020216142523.04abee78@202.95.131.154>

thanks cliff, i played with the threading module yesterday, as 
well as sockets, making time service and little things like 
that. i too can say that it's simple and straightforward. not to mention fun!

t

At 16/02/2002 06:24, Cliff Wells wrote:
>I suggest implementing something simple, say a thread-safe queue, hammer it
>with dozens/hundreds of threads and see what you learn.  Believe it or not,
>the standard Python documentation will probably be sufficient.




From donn at u.washington.edu  Tue Feb 12 12:31:58 2002
From: donn at u.washington.edu (Donn Cave)
Date: 12 Feb 2002 17:31:58 GMT
Subject: Shell command
References: <3c68e1d0$0$12225$4d4efb8e@news.be.uu.net>
Message-ID: <a4bjie$tna$1@nntp6.u.washington.edu>

Quoth "DelPiccolo Ivano" <ivano.delpiccolo at degroof.be>:
| How can I do to launch a command in my python script ?
| example : I want to launch the command : rm test.txt

The handful of people who have already recommended
os.system('rm test.txt') are correct, and os.unlink('test.txt')
too.  But if you really wanted to execute something more like
'rm %s' % file - in other words, your command parameters are
determined partly at run time - then there's a grave risk with
os.system(), and you should consider os.spawnv() instead.

The problem is that os.system invokes the UNIX shell, which
interprets the command, and any shell punctuation that happens
to get into your data can radically change the command.  This
is particularly important when the data comes from some external
source, such as user input.

exitcode = os.spawnv(os.P_WAIT, '/bin/rm', ['rm', 'text.txt',])

Another advantage is that you'll get the exit code's integer
value, instead of a bit mask that has to be analyzed to get
that value.  0 is success, anything else is failure (and a
negative number is the terminating signal.)  And it doesn't
invoke the shell, so it ought to be more efficient.

	Donn Cave, donn at u.washington.edu


From frr at ThouShallNotSpam.EasyJob.NET  Mon Feb 25 07:16:31 2002
From: frr at ThouShallNotSpam.EasyJob.NET (Fernando Rodríguez)
Date: Mon, 25 Feb 2002 13:16:31 +0100
Subject: Threads in python?
Message-ID: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com>

I'm going to start coding a little utility that needs to retrieve webpages and
process them. 

However, I worried that the whole app might stall if one server takes too long
to respond or is offline.

Does Python support threads? "Learning Python" doesn't mention it... 



| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| Fernando Rodr?guez   frr at EasyJob.NET 
| http://www.EasyJob.NET/
| Expert resume and cover letter creation system. 
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


From sholden at holdenweb.com  Wed Feb 13 09:49:36 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 09:49:36 -0500
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk>
Message-ID: <X1va8.207034$KM2.8517807@atlpnn01.usenetserver.com>

"Christopher Saunter" <Christopher.Saunter at durham.ac.uk> wrote in message
news:a4dt3d$6us$1 at sirius.dur.ac.uk...
> Greetings All!
>
> I have been using Python now for arround a year, and have found it
> incredably easy to learn the language, primarily because things are very
> obvious cf other languages, and until the other day, I had never had a
> 'Gotcha' - a bug caused by me entering some code wrongly and not realising
> it.
>
> It all stems to this, I was assigning to numbers to variables:
>
> var1 = 1000
> var2 = 0100
> etc.
>
> When I ran the code, things did not go as they sould.  Some investigation
> (running the code line by line in the interpreter...) revealed the problem
> to be with asigning a value of 0100 to a variable.  (I just put the 0 in
> there to ballance the screen.  Daft I know...)
>
> From the interpreter:
> >>> 0100
> 64
>
> I missed the bit in the docs where prefacing a number with 0 means you are
> entering in octal...
>
> Personally, I would like to see octal number entry migrated from 0--- to
> something like 0o---.  This would bring things more in line with the entry
> of hex numbers (0x---)
>
> I know it's only a minor thing, but it'd be nice if my numbers behaved as
> I expected...
>
> Opinions?
>
Chris: Welcome to c.l.py! Your point is not unreasonable, but this octal
representation is widely used in the Unix world. Unfortunately, making 0oNNN
the *only* way to represent octal would break far too many programs in which
0100 is *supposed* to mean 64.

It might be possible to introduce it as a redundant alternative, but
generally speaking there's resistance to having more than one way to
represent something (confuses learners), and this wouldn't anyway have saved
you from the glitch you experienced.

Serves you right for being anal about alignment when only the interpreter
and you were going to read the code <0.9 wink>.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From maxm at mxm.dk  Thu Feb 21 12:24:52 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 21 Feb 2002 17:24:52 +0000
Subject: Python ActiveX Scripting Engine error
References: <508cb8ae.0202210917.6d97e157@posting.google.com>
Message-ID: <3C752D64.9090100@mxm.dk>

Hmm I sometimes experience those. I think it is when I have PythonWin 
open at the same time an I try to run a script with errors. Then it sort 
of "gets stuck" on the iis.

A restart of IIS solves it for me.

regards Max M

B wrote:
 > I'm getting the following error intermittently in my ASP scripts.  Does
 > anybody know why it is happening and how to sort it out?
 >
 >
 > Python ActiveX Scripting Engine error '80020009'
 > Traceback (innermost last): KeyboardInterrupt:
 >
 > /Default.asp,line 0
 >



From dalsan at sparcs.org  Thu Feb 21 20:52:40 2002
From: dalsan at sparcs.org (Jinkuk Kim)
Date: 21 Feb 2002 17:52:40 -0800
Subject: [Q] Problem in installing 4XSLT
Message-ID: <761b16e3.0202211752.5b255cac@posting.google.com>

Hello, I'm Kim

 I want to use 4Suite to use XSLT, but i faced a problem installing
it.

 first, I installed python2.2 and it worked well.
 (I downloaded src, and configure, make, make install ..)

 second, I installed PyXML-0.7 and i think it works well.
 because when i typed "from xml.dom.ext.reader import PyExpat"
 no error occured.
 (I downloaded src, and "python setup.py build", "python setup.py
install" ..)

 third, I installed 4Suite-0.11.1 and i thought it was installed
successfully,
 but when i typed "from xml.xslt.Processor import Processor"
 following error message was printed
 (I downloaded src, and "python setup.py install" ..)

Traceback (most recent call last):
 File "<stdin>", line 1, in ?
 File "/usr/local/lib/python2.2/site-packages/_xmlplus/xslt/Processor.py",
line 24, in ?
  from xml.xslt import StylesheetReader, ReleaseNode
 File "/usr/local/lib/python2.2/site-packages/_xmlplus/xslt/StylesheetReader.py",
line 53, in ?
  from xml.xslt.Stylesheet import StylesheetElement
 File "/usr/local/lib/python2.2/site-packages/_xmlplus/xslt/Stylesheet.py",
line 22, in ?
  from xml.xslt import XsltElement, XsltException, InternalException,
Error
ImportError: cannot import name InternalException

 I checked the source file of Stylesheet.py,
 in that, xml.xslt.xpath.InternalException was imported,

 but when I checked site-packages/_xmlplus/xpath
 I couldn't find the definition of InternalException.

 I'm getting very difficulty solving this problem.
 please help solving my problem.


 * This installation was done on both Solaris8 and Debian Linuix,


From logstx at bellatlantic.net  Tue Feb 26 17:50:50 2002
From: logstx at bellatlantic.net (logistix)
Date: Tue, 26 Feb 2002 22:50:50 GMT
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com> <5uGe8.4184$LN5.796@nwrddc01.gnilink.net> <7xwux04snb.fsf@ruckus.brouhaha.com>
Message-ID: <ejUe8.1601$E11.1000@nwrddc02.gnilink.net>

>
> Of course it's possible in a dynamic language, and it helps compilers.
> Lisp compilers have worked that way since the 60's or even earlier.
>
> Obviously that example can't be checked statically by the compiler.
> It can be checked dynamically at runtime (debug mode) or the compiler
> can generate code that assumes the advice is correct and which crashes
> if the advice is wrong (optimized mode).

Okay, I exaggerated, it is possible to type in a dynamic language if you
don't mind increasing the number of opportunities for the program to crash.
I'd say 98% of the Python community is probably more interested in REDUCING
the number of crashes than a potential speed increase (especially since
there's not even a native compiler to optimize yet)

> Why would I want to use the builtin type() function everywhere I want to
> validate the type, if Python can do the check for me automatically based
> on the type advice?

Why would you change the entire semantics of the Python language just so
that you can do something in one line of code that can already be done in
one line of code?




From wolfson at uchicago.edu  Tue Feb 19 01:24:34 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Tue, 19 Feb 2002 00:24:34 -0600
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
References: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com> <mailman.1014024089.4625.python-list@python.org>
Message-ID: <pan.2002.02.19.00.24.33.166041.3802@uchicago.edu>

On Mon, 18 Feb 2002 03:20:33 -0600, Tim Legant wrote:

> From (ancient) memory, isn't it:
> 
> 'Twas _brillig_, and the _slithy_ _toves_
> Did _gyre_ and _gimble_ in the _wabe_.
> All _mimsy_ were the _borogoves_ (borogroves?)
> And the _mome_ _raths_ _outgabe_.

Close: the last word is "outgrabe" (past tense of "outgribe", naturally).

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From jeff at ccvcorp.com  Thu Feb 28 15:04:24 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 12:04:24 -0800
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz> <3C7CFAF8.1080108@mxm.dk> <Xgaf8.10287$pC1.1854469@newssrv26.news.prodigy.com>
Message-ID: <3C7E8D47.21BE30B3@ccvcorp.com>


dsavitsk wrote:

> "Max M" <maxm at mxm.dk> wrote in message news:3C7CFAF8.1080108 at mxm.dk...
>
> > btw: if you really want to store it in a list you can get dict-like
> > behaviour with list speed like this:
>
> is this a python idiom that people often use? cause i've not seen it before,
> but it is really clever.

No, it's not something that people often use.  Generally, if you want dict-like
behavior, you just use a dict.  The times that it's worth the effort to fake the
behavior with something else, are pretty unusual.  The speed difference between
lists and dicts isn't all that significant, anyhow, and which one has the
advantage depends on how you're using it.  Python dicts are pretty amazingly
nifty, really; there's no reason to be shy about using them.

Jeff Shannon
Technician/Programmer
Credit International




From donn at drizzle.com  Mon Feb 18 00:26:52 2002
From: donn at drizzle.com (Donn Cave)
Date: Mon, 18 Feb 2002 05:26:52 -0000
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org>
Message-ID: <1014010011.94628@yasure>

Quoth Mark Mitchell <mark at codesourcery.com>:
...
| This code is supposed to make sure that the file descriptors are
| not leaked; when we exit this scope, the descriptors will be closed,
| even if an exception occurs.  [Note that I am assuming that os.close
| will never thrown an exception when given a valid file descriptor.
| That's not strictly speaking true on some operating systems, but
| let's pretend that it is true.  For the purposes of this discussion,
| the particular functions involved don't matter.]

I guess it's just a matter of perspective, but instead of pretending
that close(2) would never throw an exception directly, I'd just as
soon imagine that it will.  If your object is to write very robust
code, that's the only tenable assumption, right?  The point of this
is to suggest that your problem with the finally: block isn't really
unique to programs that inflict signal handling on themselves, they're
just vulnerable in a particularly indiscriminate way.

The only thing I can suggest is nested finallys.

   try:
       # Do stuff with the pipe.
   finally:
       # Close the ends of the pipe.
       try:
           os.close(pipe[0])
       finally:
           os.close(pipe[1])

Elsewhere in your post you mention the possibility that a signal-raised
exception could abort the finally block before its first statement
executed.  I don't know either way, but my bet would be that the finally
block's first statement is guaranteed to start executing.

And of course if that's true, you could just use sigprocmask() as you
proposed to.  I don't recall what Python signal handling looks like
exactly, but I bet it would be worth looking at signalmodule.c before
you just write a simple wrapper around sigprocmask(), in case you need
to account for the latent signal that has been trapped by the interpreter
but not yet delivered to the handler.

If it isn't true (if an exception from a signal handler can interrupt a
finally block even before its first statement), then my preference would
be to get that fixed, rather than add provisions to finally just to account
for signal-handling folly.

	Donn Cave, donn at drizzle.com


From robin at jessikat.fsnet.co.uk  Fri Feb  8 14:10:42 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Feb 2002 19:10:42 +0000
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
 <a413k9$s8h$1@panix2.panix.com>
Message-ID: <fWIGSWAyKCZ8Ewif@jessikat.demon.co.uk>

In article <a413k9$s8h$1 at panix2.panix.com>, Aahz Maruch <aahz at panix.com>
writes
>In article <mailman.1013181915.301.python-list at python.org>,
>Skip Montanaro  <skip at pobox.com> wrote:
>>
>>Guido posted a bunch of talk slides to the Python website in PowerPoint
>>format.  I understand his reasons for not creating HTML versions (mostly
>>because the HTML generated by PPT sucks).  Still, that leaves people without
>>access to PPT sort of out in the cold.  What non-MS tools are out there
>>(open source or otherwise) for reading, displaying or converting PPT slides
>>to other, more accessible, formats?  If we can get a reasonable (but fairly
>>short) list of tools, I can probably see they get referenced on the relevant
>>page of the Python website.
>
>Just figured I'd make a plug for ReportLab, which avoids this problem by
>creating PDF instead of PowerPoint.
beat me to it :)

pythonpoint is really neat.
-- 
Robin Becker


From laymon at crd.ge.com  Wed Feb  6 10:08:38 2002
From: laymon at crd.ge.com (Marc Laymon)
Date: 6 Feb 2002 07:08:38 -0800
Subject: directly executing a python script in Unix - RESOLVED !
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com> <60018f02.0202050547.4836c5ff@posting.google.com> <uvgdcowgq.fsf@python.net>
Message-ID: <60018f02.0202060708.5e65a7c9@posting.google.com>

The dreaded DOS CR is the problem !  I initially wrote the script in a
Windows editor, then moved it to Linux.  I was not aware the CR's
would
make a difference on the #!line, plus I could not even see them.  I
was looking at the file in Emacs, which used to show CR's as ^M.  I
recently upgraded to
a new version of Linux and Emacs apparently no longer shows the ^M,
which I was not aware of until I ran the head command you suggested. 
This command showed that the file did contain the CR's.

Thanks for your help. 

Marc Laymon

Michael Hudson <mwh at python.net> wrote in message news:<uvgdcowgq.fsf at python.net>...
> laymon at crd.ge.com (Marc Laymon) writes:
> 
> > Several people suggested that the script needs to be
> > executable.  However, python is in
> > /usr/bin/python and the script is executable:
> > 
> > (laymon) [23] which python
> > /usr/bin/python
> > (laymon) [24] ls -l retest.py
> > -rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*
> > 
> > If the script is not executable, you get a "Permission denied"
> > error, not "Command not found"
> > 
> > (laymon) [25] chmod -x retest.py
> > (laymon) [26] ./retest.py
> > ./retest.py: Permission denied.
> > (laymon) [27] chmod +x retest.py
> > (laymon) [28] ./retest.py
> > ./retest.py: Command not found.
> > (laymon) [29] ls -l retest.py
> > -rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*
> > 
> > I can duplicate the error message with a perl script by changing
> > "#!/usr/bin/perl" to "#!/usr/bin/foo", so it seems that for some
> > reason, Unix can't find python, but since /usr/bin is in my path
> > and I am explicitly giving the command "usr/bin/python",
> > I can't figure out why.  Anybody have any other ideas ?
> 
> Have carriage returns snuck into your script?
> 
> try
> 
> $ head -n1 retest.py | cat -v
> 
> if the result is
> 
> #!/usr/bin/python^M
> 
> then that's your problem.
> 
> Cheers,
> M.


From billludden at email.msn.com  Sun Feb 10 14:33:04 2002
From: billludden at email.msn.com (billludden)
Date: Sun, 10 Feb 2002 12:33:04 -0700
Subject: Newbie needs references
Message-ID: <u9bkUtmsBHA.388@cpimsnntpa03>

Hi<

I have not programmed since 1975, and that was in BASIC.
I'd like to learn Python, but everything I find on the web is
    way above my head.
Does anyone have books or tutorials they would recommend
    for an illiterate to teach himself Python?

Thank you,  Bill




From dinu at reportlab.com  Wed Feb 13 11:15:01 2002
From: dinu at reportlab.com (Dinu C. Gherman)
Date: Wed, 13 Feb 2002 17:15:01 +0100
Subject: Compiling PIL on OS X
Message-ID: <dinu-8BC811.17150113022002@news.t-online.com>

Hi,

does a well-defined way exist to compile PIL 1.1.2 on 
MacOS X, and if so, does anybody know where it is to
be found?

Thanks,

Dinu


From andy47 at halfcooked.com  Sun Feb 10 17:28:29 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Sun, 10 Feb 2002 22:28:29 +0000 (UTC)
Subject: Are there any Python books or sites with good exercises?
References: <g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r@4ax.com>
Message-ID: <Xns91B2604B33F77andy47halfcookedcom@203.109.250.24>

Stefan Franke <franke at ableton.com> wrote in 
news:g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r at 4ax.com:

> (WRT your personal definition of 'good', of course). I'm going to 
> prepare a Python introductory course for experienced programmers
> and I'm looking for handy exercises which hopefully provide them
> numerous Aha! effects.
> 
> Thanks,
> Stefan
> 
> 
> 

http://diveintopython.org/

Regards,
Andy
-- 
Contents free posts a speciality


From zomba at dtmf.com.ar  Mon Feb  4 12:47:49 2002
From: zomba at dtmf.com.ar (Zomba)
Date: Mon, 4 Feb 2002 14:47:49 -0300
Subject: ioctl and stuff
Message-ID: <004401c1ada4$0fc205a0$020a0a0a@cotelviso.com.ar>

I make a 
s=socket(AF_INET, SOCK_DGRAM)
res= fcntl.ioctl(s.fileno(), 0xc020690d, 'le1' + '\0' *29)

the 0xc020690d call, work pretty well, i can take from res (with struct.unpack) the ip addr of my device, but the problem is:

fcntl.ioctl(s.fileno(), 0xc020690c, res)
and it said: "Errno 22 - Invalid Argument" 
 
(I try this, because i make my own 'res' with other information, but it had the same error)

Im working on OpenBSD, the 0xc020690d is OSIOCGIFADDR and 0xc020690c is SIOCSIFADDR

Ideas??!

Thanks in advance ppls
Zomba

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020204/e06d4838/attachment.html>

From robin at jessikat.fsnet.co.uk  Thu Feb 14 04:05:27 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 09:05:27 +0000
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
 <b16e4ef7.0202130950.67f4004d@posting.google.com>
 <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4>
Message-ID: <pzib74AX33a8EwPJ@jessikat.fsnet.co.uk>

In article <Xns91B4EC515D335RASXnewsDFE1 at 130.133.1.4>, Philip
Swartzleonard <starx at pacbell.net> writes
>Stephen Ferg || Wed 13 Feb 2002 09:50:00a:
>
>>> Is there a simple and easy way to allow the end-user to select a
>>> directory?  
>> 
>> I'm working on a package that I'd like to polish some more, called
>> easygui.  Go to www.ferg.org/easygui.  The function you want is called
>> dirbox().
>> 
>> If you try it, let me know how you get on... :-)
>> -- Steve Ferg
>
>This doesn't really have anything to do with this thread, but i'd like to 
>make this comment: =PLEASE= don't use the (apparently) standard directory 
>selcetion dialog in windows if you can avoid it. It can be truly 
>aggrivating when you know exactly how to type your way there, but -have- to 
>use the box... and on a reasonably crowded file system, this means 'open 
>node, scroll down, open node, scroll down' repeat-until-insane. 
>
>Ok, enough random ranting =)
>
well you are allowed to type directly into that entry widget. My problem
has always been the auto-completion character is not TAB and dosen't
cycle through possibilities.
-- 
Robin Becker


From James_Althoff at i2.com  Wed Feb 27 18:52:34 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Wed, 27 Feb 2002 15:52:34 -0800
Subject: Status of PEP's?
Message-ID: <OF97FD42A8.314B2082-ON88256B6D.0082F36F@i2.com>

Below is a revision (not yet submitted) of PEP 276.

Rather than restart the entire debate on PEP 276, please send me
suggestions for specific changes to the PEP.

Thanks,

Jim

===============================================================

PEP: 276
Title: Simple Iterator for ints
Version: $Revision: 1.1 $
Last-Modified: $Date: 2001/11/13 20:52:37 $
Author: james_althoff at i2.com (Jim Althoff)
Status: Draft
Type: Standards Track
Created: 12-Nov-2001
Python-Version: 2.3
Post-History:


Abstract

    Python 2.1 added new functionality to support iterators[1].
    Iterators have proven to be useful and convenient in many coding
    situations.  It is noted that the implementation of Python's
    for-loop control structure uses the iterator protocol as of
    release 2.1.  It is also noted that Python provides iterators for
    the following builtin types: lists, tuples, dictionaries, strings,
    and files.  This PEP proposes the addition of an iterator for the
    builtin type int (types.IntType).  Such an iterator would simplify
    the coding of certain for-loops in Python.


Specification

    Define an iterator for types.intType (i.e., the builtin type
    "int") that is returned from the builtin function "iter" when
    called with an instance of types.intType as the argument.

    The returned iterator has the following behavior:

    - Assume that object i is an instance of types.intType (the
      builtin type int) and that i > 0

    - iter(i) returns an iterator object

    - said iterator object iterates through the sequence of ints
      0,1,2,...,i-1

    Example:

        iter(5) returns an iterator object that iterates through the
        sequence of ints 0,1,2,3,4

    - if i <= 0, iter(i) returns an "empty" iterator, i.e., one that
      throws StopIteration upon the first call of its "next" method

    In other words, the conditions and semantics of said iterator is
    consistent with the conditions and semantics of the range() and
    xrange() functions.

    Note that the sequence 0,1,2,...,i-1 associated with the int i is
    considered "natural" in the context of Python programming because
    it is consistent with the builtin indexing protocol of sequences
    in Python.  Python lists and tuples, for example, are indexed
    starting at 0 and ending at len(object)-1 (when using positive
    indices).  In other words, such objects are indexed with the
    sequence 0,1,2,...,len(object)-1


Rationale

    A common programming idiom is to take a collection of objects and
    apply some operation to each item in the collection in some
    established sequential order.  Python provides the "for in"
    looping control structure for handling this common idiom.  Cases
    arise, however, where it is necessary (or more convenient) to
    access each item in an "indexed" collection by iterating through
    each index and accessing each item in the collection using the
    corresponding index.

    For example, one might have a two-dimensional "table" object where one
    requires the application of some operation to the first column of
    each row in the table.  Depending on the implementation of the table
    it might not be possible to access first each row and then each
    column as individual objects.  It might, rather, be possible to
    access a cell in the table using a row index and a column index.
    In such a case it is necessary to use an idiom where one iterates
    through a sequence of indices (indexes) in order to access the
    desired items in the table.  (Note that the commonly used
    DefaultTableModel class in Java-Swing-Jython has this very protocol).

    Another common example is where one needs to process two or more
    collections in parallel.  Another example is where one needs to
    access, say, every second item in a collection.

    There are many other examples where access to items in a
    collection is facilitated by a computation on an index thus
    necessitating access to the indices rather than direct access to
    the items themselves.

    Let's call this idiom the "indexed for-loop" idiom.  Some
    programming languages provide builtin syntax for handling this
    idiom.  In Python the common convention for implementing the
    indexed for-loop idiom is to use the builtin range() or xrange()
    function to generate a sequence of indices as in, for example:

     for rowcount in range(table.getRowCount()):
         print table.getValueAt(rowcount, 0)

    or

     for rowcount in xrange(table.getRowCount()):
         print table.getValueAt(rowcount, 0)

    From time to time there are discussions in the Python community
    about the indexed for-loop idiom.  It is sometimes argued that the
    need for using the range() or xrange() function for this design
    idiom is:

    - Not obvious (to new-to-Python programmers),

    - Error prone (easy to forget, even for experienced Python
      programmers)

    - Confusing and distracting for those who feel compelled to understand
      the differences and recommended usage of xrange() vis-a-vis range()

    - Unwieldy, especially when combined with the len() function,
      i.e., xrange(len(sequence))

    - Not as convenient as equivalent mechanisms in other languages,

    - Annoying, a "wart", etc.

    And from time to time proposals are put forth for ways in which
    Python could provide a better mechanism for this idiom.  Recent
    examples include PEP 204, "Range Literals", and PEP 212, "Loop
    Counter Iteration".

    Most often, such proposal include changes to Python's syntax and
    other "heavyweight" changes.

    Part of the difficulty here is that advocating new syntax implies
    a comprehensive solution for "general indexing" that has to
    include aspects like:

    - starting index value

    - ending index value

    - step value

    - open intervals versus closed intervals versus half opened intervals

    Finding a new syntax that is comprehensive, simple, general,
    Pythonic, appealing to many, easy to implement, not in conflict
    with existing structures, not excessively overloading of existing
    structures, etc. has proven to be more difficult than one might
    anticipate.

    The proposal outlined in this PEP tries to address the problem by
    suggesting a simple "lightweight" solution that helps the most
    common case by using a proven mechanism that is already available
    (as of Python 2.1): namely, iterators.

    Because for-loops already use "iterator" protocol as of Python
    2.1, adding an iterator for types.IntType as proposed in this PEP
    would enable by default the following shortcut for the indexed
    for-loop idiom:

     for rowcount in table.getRowCount():
         print table.getValueAt(rowcount, 0)

    The following benefits for this approach vis-a-vis the current
    mechanism of using the range() or xrange() functions are claimed
    to be:

    - Simpler,

    - Less cluttered,

    - Focuses on the problem at hand without the need to resort to
      secondary implementation-oriented functions (range() and
      xrange())

    And compared to other proposals for change:

    - Requires no new syntax

    - Requires no new keywords

    - Takes advantage of the new and well-established iterator mechanism

    And generally:

    -  Is consistent with iterator-based "convenience" changes already
       included (as of Python 2.1) for other builtin types such as:
       lists, tuples, dictionaries, strings, and files.


Backwards Compatibility

    The proposed mechanism is generally backwards compatible as it
    calls for neither new syntax nor new keywords.  All existing,
    valid Python programs should continue to work unmodified.

    However, this proposal is not perfectly backwards compatible in
    the sense that certain statements that are currently invalid
    would, under the current proposal, become valid.

    Tim Peters has pointed out two such examples:

    1) The common case where one forgets to include range() or
       xrange(), for example:

        for rowcount in table.getRowCount():
            print table.getValueAt(rowcount, 0)

       in Python 2.2 raises a TypeError exception.

       Under the current proposal, the above statement would be valid
       and would work as (presumably) intended.  Presumably, this is a
       good thing.

       As noted by Tim, this is the common case of the "forgotten
       range" mistake (which one currently corrects by adding a call
       to range() or xrange()).

    2) The (hopefully) very uncommon case where one makes a typing
       mistake when using tuple unpacking.  For example:

           x, = 1

       in Python 2.2 raises a TypeError exception.

       Under the current proposal, the above statement would be valid
       and would set x to 0.  The PEP author has no data as to how
       common this typing error is nor how difficult it would be to
       catch such an error under the current proposal.  He imagines
       that it does not occur frequently and that it would be
       relatively easy to correct should it happen.


Issues:

    Extensive discussions concerning PEP 276 on the Python interest
    mailing list suggests a range of opinions: some in favor, some
    neutral, some against.  Those in favor tend to agree with the
    claims above of the usefulness, convenience, and simplicity of
    a simple iterator for integers.

    Issues with PEP 276 include:

    - Some feel that iterating over the sequence "0, 1, 2, ..., n-1"
      for an integer n is not intuitive.  "for i in 5:" is considered
      (by some) to be "non-obvious", for example.

      Response: This seems to be largely a matter of preference.
      Some like the proposed idiom and see it as simple and
      elegant.  Some are neutral on this issue.  Others, as noted,
      dislike it.

    - Is it obvious that iter(5) maps to the sequence 0,1,2,3,4?

      Response: Given, as noted above, that Python has a strong
      convention for indexing sequences starting at 0 and stopping at
      (inclusively) the index whose value is one less than the length
      of the sequence, it is argued that the proposed sequence is
      reasonably intuitive to a Python programmer while being useful
      and practical.  More importantly, it is argued that once learned
      this convention is very easy to remember.

    - Possible ambiguity

          for i in 10: print i

      might be mistaken for

     for i in (10,): print i

      Response: This is exactly the same situation with strings in
      current Python (replace 10 with 'spam' in the above, for
      example).

    - Might dissuade newbies from using the indexed for-loop idiom when
      the standard "for item in collection:" idiom is clearly better.

      Response: The standard idiom is so nice when it fits that it
      needs neither extra "carrot" nor "stick".  On the other hand,
      one does notice cases of overuse/misuse of the standard idiom
      (due, most likely, to the awkwardness of the indexed for-loop
      idiom), as in:

       for item in sequence:
           print sequence.index(item)

    The majority of disagreement with PEP 276 came from those who
    favor much larger changes to Python to address the more general
    problem of specifying a sequence of integers where such
    a specification is general enough to handle the starting value,
    ending value, and stepping value of the sequence and also
    addresses variations of open, closed, and half-open (half-closed)
    integer intervals.  Many suggestions of such were discussed.

    These include:

    - adding Haskell-like notation for specifying a sequence of
      integers in a literal list,

    - various uses of slicing notation to specify sequences,

    - changes to the syntax of for-in loops to allow the use of
      relational operators in the loop header,

    - creation of an integer-interval class along with methods that
      overload relational operators or division operators
      to provide "slicing" on integer-interval objects,

    - and more.

    It should be noted that there was much debate but not an
    overwhelming concensus for any of these larger-scale suggestions.

    Clearly, PEP 276 does not propose such a large-scale change
    and instead focuses on a specific problem area.  Towards the
    end of the discussion period, several posters expressed favor
    for the narrow focus and simplicity of PEP 276 vis-a-vis the more
    ambitious suggestions that were advanced.  There did appear to be
    concensus for the need for a PEP for any such larger-scale,
    alternative suggestion.  In light of this recognition, details of
    the various alternative suggestions are not discussed here further.


Implementation

    An implementation is not available at this time but is expected
    to be straightforward.


References

    [1] PEP 234, Iterators
    http://python.sourceforge.net/peps/pep-0234.html

    [2] PEP 204, Range Literals
    http://python.sourceforge.net/peps/pep-0204.html

    [3] PEP 212, Loop Counter Iteration
    http://python.sourceforge.net/peps/pep-0212.html


Copyright

    This document has been placed in the public domain.



Local Variables:
mode: indented-text
indent-tabs-mode: nil
fill-column: 70
End:





From shredwheat at attbi.com  Mon Feb 25 11:24:06 2002
From: shredwheat at attbi.com (Pete Shinners)
Date: Mon, 25 Feb 2002 16:24:06 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com>
Message-ID: <3C7A66A7.8060503@attbi.com>

Donnal Walter wrote:
> But what is the functional equvalent of:
> 
> for x in items:
>     x.f()

if all "items" are the same class, you could get away with this,

map(class_of_x.f, items)



From tismer at tismer.com  Fri Feb 15 03:27:15 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 15 Feb 2002 09:27:15 +0100
Subject: stackless python
References: <9vpjds$due$1@usc.edu> <1008824522.772261@yabetcha.sttl.drizzle.com> <mailman.1009810787.26750.python-list@python.org> <7x3d1r1k9j.fsf@ruckus.brouhaha.com>
Message-ID: <3C6CC663.4010309@tismer.com>

Paul Rubin wrote:

 > Christian Tismer <tismer at tismer.com> writes:
 >
 >>I'm pretty sure Scheme can. It has raw continuations, so you can
 >>implement everything with them, why not microthreads?
 >>Continuations can do even more, they are able to create new
 >>control structures for your language. And exactly that it the point
 >>which makes them insuitable for Python: They are too powerful.
 >>Python has its own control structures, and we don't need a construct
 >>with the power to build some. This is oversized, and very oversized
 >>construct turns out to be a drawback at some future.
 >>What Python needs is a secure mechanism to switch frame changes
 >>at certain times. This is not continuations, but microthreads with
 >>explicit or implicit switching.
 >>
 >
 > The thing I don't fully understand is that several Scheme
 > implementations are both smaller and faster than Python.  I've been
 > wondering for a while whether it's time to graft a Python parser onto
 > a Scheme compiler/evaluator.  The limitations of "simple generators"
 > also seem kind of artificial--maybe Stackless for 2.2 can support
 > calling a generator from multiple places.


Python's approach at generators is the cheapest possible
way to get generators. And this implies that the generator
frame is bound to the current context.
Stackless new dirty little brother goes a completely different
way and can of course generate general generators which do
run without any limitation -- at some cost of course.

Python generators will run at maximum speed, but "stack-based".

Stackless generators will exist as a special case of Stackless
tasklets, which can act as microthreads, coroutines or
generators, at a slightly higher computational cost, but
only limited by the programmer's imagination.

These festures will be available by the end of February 2002,
the latest. A way to pickle program state will be implemented
until end of March 2002.

Kind regards - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
       where do you want to jump today?   http://www.stackless.com/






From robinjim at earthlink.net  Thu Feb 28 21:18:01 2002
From: robinjim at earthlink.net (robin and jim)
Date: Fri, 01 Mar 2002 02:18:01 GMT
Subject: super confused !
Message-ID: <txBf8.24714$ZC3.1996215@newsread2.prod.itd.earthlink.net>

I'm confused about the behavior of "new-style" classes in Python 2.2.

Here is a sample session that illustrates my confusion.

>>> class Int_a(int):
...  def __init__(self, spec):
...   super(Int_a, self).__init__(spec)
...
>>> Int_a(1)
1
>>> class Int_b(Int_a):
...  def __init__(self, spec):
...   super(Int_b, self).__init__(spec)
...
>>> Int_b(2)
2
>>> class Int_c(Int_a):
...  def __init__(self):
...   super(Int_c, self).__init__(99)
...
>>> Int_c()
0
>>> class Int_d(Int_a):
...  def __init__(self):
...   Int_a.__init__(self, 99)
...
>>> Int_d()
0

Why are Int_c() and Int_d() zero (the default for int()) rather than 99?

How can I make them 99?

Thanks for your interest.




From buzzard at urubu.freeserve.co.uk  Mon Feb 25 21:07:32 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Tue, 26 Feb 2002 02:07:32 -0000
Subject: dlg.Destroy() behaviour
References: <a5e561$q8v$1@news5.svr.pol.co.uk> <mailman.1014680321.7184.python-list@python.org>
Message-ID: <a5eqpf$8fn$1@news5.svr.pol.co.uk>

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
news:mailman.1014680321.7184.python-list at python.org...
> On Mon, 25 Feb 2002 19:58:42 -0000
> Duncan Smith wrote:
>
> > Just a quick question about whether or not I need to destroy dialogues
> > (wxPython) with dlg.Destroy().  When creating a number of
> wxTextEntryDialogs
> > in succession only the first dialogue gets the focus.  I need a mouse
> click
> > to set the focus on subsequent dialogues.  However, if I remove the
usual
> > 'dlg.Destroy()', then subsequent dialogues do get the focus.  Is this
> > advisable?  If not, what is the neatest way of setting the focus on the
> > subsequent dialogues?  Thanks in advance.
>
> I don't know if this works (never tried it on a dialog), but
dlg.SetFocus()
> ?

Thanks, but no.  A computer scientist mate of mine initially suggested that.
It didn't work.  He then suggested I just got rid of dlg.Destroy(), which
works, but he couldn't tell me whether it was a safe thing to do or not.  He
suggested that as the 'deleted' dialogue wouldn't be destroyed immediately
with dlg.Destroy() it might retain the focus.  I don't know.

Duncan

>
> --
> Cliff Wells, Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308  (800) 735-0555 x308
>




From magnus at thinkware.se  Thu Feb 21 14:50:15 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 20:50:15 +0100
Subject: Adding static typing to Python
References: <mailman.1014225213.11001.python-list@python.org> <a52rbc$p1i$1@thorium.cix.co.uk>
Message-ID: <3C754F77.3020206@thinkware.se>

gbreed at cix.compulink.co.uk wrote:

> It looks like "strong typing" 
> doesn't have a universally agreed definition, so I suppose the best thing 
> is to avoid the term.


But we have two orthogonal dimensions here, and it is
meaningful to explain the differences between C and
Python for instance.

Clearly the definition you showed suffers from trying
to describe two distinct qualities with one name. "C and
C++ are sometimes described as strongly typed, but are
perhaps better described as weakly typed." This certainly
contradict the statement that it's the static quality that
"gives strength".

In C, the type of a variable name is decided at compile time,
and in Python, it's decided at runtime. That is ONE difference,
and dynamic / static typing probably describes that best.

The OTHER aspect of typing is that while C types are static,
they aren't very rigid, or whatever you'd like to call it.
You can add integers to memory pointers, and through casting
you can make the compiler do the most meaningless thing without
as much as a slight complaint. In Python you can't. There are
a few well defined ways of coercing types. Unicode containing
only ASCII can be coerced into strings, since there is no doubt
about interpretations, and numerical types can be coerced in a
sensible way, but in general Python will refuse to do meaningless
operations on incompatible types.

I think it's good to push for the terms "static" and "dynamic"
when that's what we mean. But we have to describe the other aspect
too. It's unfortunate that strong/weak typing has come to be used
for two different things. Just like "hacker". We could of course
invent new terms like robust / brittle typing, but I don't think
that is less difficult than to re-educate people about what strong
and weak typing really is.

But I'm open to suggestions...



From SKuzminski at FairIsaac.com  Mon Feb 25 19:25:32 2002
From: SKuzminski at FairIsaac.com (Stefan Kuzminski)
Date: Mon, 25 Feb 2002 16:25:32 -0800
Subject: advocacy help ( Python vs VB )
Message-ID: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>

I need to convince my co workers and erstwhile managers to build a
production system in Python.  They're coming from a VB world and are fearful
to detach from the Redmond mothership.  I need hard facts, not warm and
fuzzies about syntax but names of big companies using Python and things VB
just *cannot* do ( threads, regular expressions.. ? ).

Thanks,
Stefan



From aahz at panix.com  Sat Feb 16 16:12:29 2002
From: aahz at panix.com (Aahz Maruch)
Date: 16 Feb 2002 13:12:29 -0800
Subject: Why Python is like BASIC ...  java "vs" python
References: <mailman.1013881166.27493.python-list@python.org>
Message-ID: <a4mhvt$ga8$1@panix2.panix.com>

In article <mailman.1013881166.27493.python-list at python.org>,
john coppola  <john_coppola_r_s at yahoo.com> wrote:
>[Fernando Perez wrote ]
>>
>> (And by the way, a whale is a mammal, not a fish).
>
>That was my whole point!  It's dumb to compare them.

Whales are mammals.

Mammals have hair.

Shave the whales.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

[[Credit to Scott Adams.]]


From jason at jorendorff.com  Sat Feb 16 10:29:28 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 09:29:28 -0600
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <926k4a.3a.ln@127.0.0.1>
Message-ID: <HFEKILOLEFEFMKAECNDLGEAJDDAA.jason@jorendorff.com>

Carl Banks wrote:
> Cliff Wells wrote:
> > Gillou wrote:
> > > * NO LINE NUMBERS! Enough said.
> > > ...
> > >   IF condition GOSUB line-number (I never did understand the necessity
> > >   for that kind of statement until well after I'd stopped using BASIC
> > >   entirely and moved to more "modern" languages like... Pascal).
> >
> > Only old men like you and me remember this
>
> Hi.
>
> I remember using calculated GOSUB, something like this, in GW-BASIC:
> 220 ON LEVEL GOSUB 1000,2000,3000,4000,5000,...
> It was for a crude level-based video game.  I'm 25 now.

Feh.  You youngsters these days with yer GW-BASIC and yer "video games".
I remember using the Commodore 64 two-character shortcuts for the
keywords.  C64 BASIC = executable WingDings.  (I'm 24.)

## Jason Orendorff    http://www.jorendorff.com/




From sholden at holdenweb.com  Tue Feb 12 07:18:27 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Feb 2002 07:18:27 -0500
Subject: just a moron trying to get python working on win98 and personal web  server
References: <3C685DDE.AA8AFD10@kc.rr.com>
Message-ID: <fK7a8.171629$KM2.7541471@atlpnn01.usenetserver.com>

"james" <jdaily at kc.rr.com> wrote in message
news:3C685DDE.AA8AFD10 at kc.rr.com...
>
> RE: just a moron trying to get python working on win98 and personal web
> server
>
> i am trying to get python to work on my MS personal web server ( which
> is
> supposed to support python in a web url ).
>
> what do i need to do?
>
The most obvious thing that's easy to overlook is that the scripts have to
be called *.asp.

> i have tried setting the registry.
>
With the win32 extensions, it should work as standard.

> i have set the PATH in autoexec.bat.
>
> is there an apache alternative.
>
>
Apache will work on 95 and 98. Xitami is easier, and better-suited to the
lightweight platform.

not-as-daft-as-you-look-'cos-at-least-you-asked-for-help-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jonathan at onegoodidea.com  Sat Feb  9 09:24:06 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Sat, 09 Feb 2002 14:24:06 +0000
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org>
Message-ID: <B88AE186.48CD%jonathan@onegoodidea.com>

On 9/2/2002 0:56, in article
mailman.1013216835.14351.python-list at python.org, "Cliff Wells"
<logiplexsoftware at earthlink.net> wrote:

> Aahz Maruch wrote:
> 
>> I have to admit that I'm seriously thinking about an iMac, though.
> 
> Yikes.  That's an expensive Linux box ;)

Hah! all smileys aside, you'd have to be mad to run Linux on it when it
comes with Mac OS X and all of Apple's free iApps. I get to run commercial
applications including the Adobe and Microsoft range, it's got full Java
support, runs all the UNIX apps I need including Apache, Python and
PostgreSQL, and is fun to use.

Stuff cygwin, there's no comparison when you have a full BSD under the hood.
And when Linux runs Photoshop, MS Office, IE, and Palm Desktop, I'll
consider using it as my main work OS.

I'd whole-heartedly recommend getting an iBook with Airport though unless
you need the extra speed of the new iMac or the DVD burner. It's cheaper,
has the same size screen, works on the sofa and my local coffee shop, gets 4
hours of battery life, and sleeps and wakes in an instant.

You don't realise how cool the latter is until you've used one for a while.
I've never seen a PC portable that got this licked. I close the lid and go
make a cup of tea. Come back, open the lid and continue where I left off.

Jonathan



From info at jandries.be  Mon Feb 11 04:26:31 2002
From: info at jandries.be (Jan Dries)
Date: 11 Feb 2002 09:26:31 -0000
Subject: cgi authoriszation
In-Reply-To: <q4Sr5IAZ4cZ8EwGa@jessikat.fsnet.co.uk>
Message-ID: <20020211092631.25686.qmail@mail2.dcube-resource.be>

[Robin Becker]
> Is there any way to get the cgi module to do authorization. I can make
> scripts demand authorization by including headers [...]
> but I don't seem to see either 'authorization' or 'http_authorization'
> in the
> client's response. Is apache filtering out the credentials?

Apache indeed filters out the "HTTP_AUTHORIZATION", but you can force it to 
pass it to a cgi-script as en environment variable using the following rewrite-
rule (assuming you have mod_rewrite enabled):

 RewriteEngine on
 RewriteCond %{HTTP:Authorisation} ^(.*)$ [NC]
 RewriteRule /.* - [E=HTTP_AUTHORISATION:%1]

But of course beware, after this the information will be available to all 
scripts. You could further restrict it by adding extra RewriteCond-lines.

Regards,
Jan

From elflord at panix.com  Tue Feb 26 14:21:10 2002
From: elflord at panix.com (Donovan Rebbechi)
Date: Tue, 26 Feb 2002 19:21:10 +0000 (UTC)
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <slrna7no18.h4v.elflord@panix2.panix.com>

In article <3C7283E4.4050604 at physics.ucsb.edu>, Craig Maloney wrote:
> 
> Hi all.
> 
> I'm doing some scientific computing and have
> amassed a moderate set of classes.  Due to the
> heavy computational demands, I don't think I
> can justify rewriting everything in an interperated
> language; nice as that would be now that I'm
> playing around with what I've created.
> 
> I'm in a situation now where I've written about
> a dozen of quickie, throw away, utility routines
> that I call from the command line, and in essence,
> BASH has become my scripting environment.
> 
> Unfortunately, the number of throw-away routines
> is exploding, and I feel like I'm loosing the advantage
> of having constructed a nice set of classes with
> which to play.
> 
> I would, of course, like to have some sort of scripted
> environment (e.g. Python) in which to "play"
> with my objects.
> 
> That said, I'd like to solicit opinions as to the relative
> merits of the various possible approaches.
> 
> I'm making essential use of templates and polymorphism.
> 
> I have no experience with Boost.Python, but from what I
> understand, it would not involve writing wrapper definitions
> (ala SWIG or IDL), right?  Are there other options?

I've been in a similar situation to you -- I'm writing library code and
utilities that use a lot of numerical algorithms. Having an interpreted 
environment is indeed extremely useful and convenient. Python also has the
added advantage that the user shell is first rate -- it includes online
help, runtime module loading, and the same sort of command history and command
line editing features as bash has.

One approach I found useful was functional programming -- you hand a range 
and a function object to C++, and use the speed of C++ to your advantage.

The tool I've used is Sip. It automatically (or close to) produces C++ 
bindings for Python, in much the same way as SWIG produces C bindings. From
my point of view, the main advantage of Sip is that it's also the 
implementation kit for PyQt, and I'm using a lot of Qt in my C++ work.

Cheers,
-- 
Donovan


From h.jehadalwan at student.kun.nl  Sat Feb 23 09:58:54 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Sat, 23 Feb 2002 15:58:54 +0100
Subject: is their any limit on the size of the data passed to a progra         m via popen2?
References: <mailman.1014421598.16958.python-list@python.org>
Message-ID: <3C77AE2E.3020301@student.kun.nl>

Gustavo Cordova wrote:

> That doing a read() on pipes hangs, so you have to
> define a timeout or something, so that you can take
> control back.
> 
> But I dinnae kno how ta do such a thang.
> 
> :-/
> 
> -gus
> 
> pd: I *think* it was jeff.
> 
> 

Ya, I red that one. But, now the problem is that running the code 
succefully really depends on the number of the files passed to the 
application in one "popen2" session. I found that the number of the 
files must not exceed 250. I worked around this by splitting the 10000 
files into 50 lists each of them contains 200 files and for each list 
opening a "popen2" session and close it at the end in a "for" loop. Now 
the code works fine. I have searched the internet for informations 
related to the size of the pipe and how to modify it, but I found 
nothing except "select.select()". At the moment I really don't have time 
to implement this.

any way, thanks for the contribution.



From peter at engcorp.com  Thu Feb  7 23:53:36 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 07 Feb 2002 23:53:36 -0500
Subject: ActivePython and wxPython
References: <RiH88.37389$Jq.1973082@news2.calgary.shaw.ca>
Message-ID: <3C6359D0.FD72884E@engcorp.com>

Andrew wrote:
> 
> Sorry if this question is inappropriate for this newsgroup, but I was
> wondering if there are known problems using wxPython with ActivePython?  I
> try to run the first example (from www.wxpython.org) in the wxPython
> tutorial but get a crash when I attempt to create a wxFrame object.  I
> stepped into the code and it's failing on an 'apply' call.

It's a very appropriate question, and my answer is that there
shouldn't be, provided the versions are compatible.

What versions of those are you using?  Did you have any previous
versions installed?  What does "crash" mean?  (Exception raised?
If so, post the traceback here for better responses.  If you 
run it from a DOS prompt it should print the trackback to the 
console.)

-Peter


From aseem at neurobs.com  Tue Feb 12 17:46:38 2002
From: aseem at neurobs.com (A M)
Date: 12 Feb 2002 14:46:38 -0800
Subject: IPython 0.2.5
References: <a4ainp$gts$1@peabody.colorado.edu> <fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com> <a4bre5$8ok$1@peabody.colorado.edu>
Message-ID: <ed6d1c1a.0202121446.2966c58e@posting.google.com>

> hard disk, but with 256Mb of ram. And it's just fine. I turned off most of 
> the eye candy and it runs quite well. It feels barely slower than it did 
> under 98, and the gains in stability are priceless.

Unfortunately most users dont know how to turn off the eye-candy..... :)
-A


From mowens at bigfoot.com  Wed Feb 27 01:21:27 2002
From: mowens at bigfoot.com (mike)
Date: 26 Feb 2002 22:21:27 -0800
Subject: [Python-Help] advocacy help ( Python vs VB )
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01> <mailman.1014706963.7898.python-list@python.org> <9cem7uc521ni47v2mo0qiv1dfougdciqp6@4ax.com>
Message-ID: <9588af6.0202262221.41b4ae9d@posting.google.com>

Courageous <jkraska at san.rr.com> wrote in message news:<9cem7uc521ni47v2mo0qiv1dfougdciqp6 at 4ax.com>...
> While this is academic, things like this are going to come up again
> in the future. What would really sell them is a Python which compiled
> to CLR and gave them full access to the suite of COM/.NET objects.
> Many "VB shops" require this. He should take a long gander at win32all
> and make sure that it fits their need.
> 
> C//

Python.NET does exactly this. This is available from Activestate.


From clpy at snakefarm.org  Sun Feb 17 12:44:13 2002
From: clpy at snakefarm.org (Carsten Gaebler)
Date: Sun, 17 Feb 2002 18:44:13 +0100
Subject: cStringIO module broken by design?
Message-ID: <3C6FEBED.40704@snakefarm.org>

Hi there!

It seems like the cStringIO module doesn't provide a write() method for 
StringIO objects that were initialized with a string:

Python 2.2 (#8, Jan 11 2002, 20:10:20)
[GCC 3.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
 >>> import StringIO
 >>> s = StringIO.StringIO()
 >>> s.write("foobar")
 >>> s = StringIO.StringIO("foo")
 >>> s.write("bar")
 >>>
 >>> import cStringIO
 >>> s = cStringIO.StringIO()
 >>> s.write("foobar")
 >>> s = cStringIO.StringIO("foo")
 >>> s.write("bar")
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
AttributeError: write
 >>>

Why is that? I find it quite annoying.

cg.



From jkraska at san.rr.com  Thu Feb  7 20:42:51 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 01:42:51 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk>
Message-ID: <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com>

>It's just that being forced -- "for my own good" -- not to use certain
>features because someone has decreed them intrinsically bad, doesn't
>appeal to me at all.

Sure. You realize that a hygienic macro doesn't prevent you from
modifying things by consequence, right? All it does is require for
you to mean it; this caters to Python's phenonomenon of least
surprise. Imagine the following

>>> i = 3
>>> j = 4
>>> mymacro i
>>> print j
5

That would vex some folks. But that's exactly what a non hygienic
macro can do. Granted, in some cases non hygeniec macros can be
quite useful. And the design I'm targeting is only _pseudo_ hygienic
in any case. It's hygienic for the local namespace. It is not,
however, hygienic for the class or the instance. A macro could
therefor modify the class or instance it is associated with, just
not the local namespace. Don't ask how I'm doing that yet, however,
it would make some of the Python implementers squeak. :)

C//



From smk at goeci.com  Fri Feb 22 13:15:13 2002
From: smk at goeci.com (news.primushost.com)
Date: Fri, 22 Feb 2002 13:15:13 -0500
Subject: MySQLdb
References: <XD7d8.203$zZ4.2649220@newsserver.ip.pt>
Message-ID: <HVvd8.687$Mp1.57843@news.shore.net>

For starters, you could try
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/65235

"Hugo Martires" <hmartires at cultalg.pt> wrote in message
news:XD7d8.203$zZ4.2649220 at newsserver.ip.pt...
> Where i can find an "how to" to learn working with MySQLdb module.
> I have downloaded from sourceforge and install it, but there is no much
> information on how to use it.
> Any one knows where i can get it ?
>
> Tanks in advance
>
>




From lac at strakt.com  Fri Feb  8 07:39:44 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 08 Feb 2002 13:39:44 +0100
Subject: encapsulation, typing & other questions 
In-Reply-To: Message from Jeff Shannon <jeff@ccvcorp.com> 
   of "Thu, 07 Feb 2002 12:32:28 PST."
Message-ID: <200202081239.g18CdiGo020002@ratthing-b246.strakt.com>

Design Patterns in Python are a joy to write and use.  But please
buy a copy of the Design Patterns Smalltalk Companion so your code
does not end up as 'Thinking in C++, transliterated into python'.
Python is incredibly more expressive than that.

But everybody brings the poison of the last language they were working
with to python, which is accomodating at lets you write 'that language
in python.'  

And your python patterns should end up more like Smalltalk ones than
like C++ ones if you are doing it right.  It will feel really good
when you get it to start happening.

Take what I say with some salt though -- I am trying to overcome
a tendancy to think in Smalltalk rather than in Python.

Laura Creighton




From chris.gonnerman at newcenturycomputers.net  Mon Feb 18 21:12:54 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 18 Feb 2002 20:12:54 -0600
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com>
Message-ID: <009b01c1b8ea$f2f22ca0$0101010a@local>

----- Original Message -----
From: "Joshua Muskovitz" <joshm at taconic.net>


> The Ruebot wrote:

Who, Joshua, it's not right to grace Rue by calling him a "bot".  Only the
great minds of the Python community deserve that honor!

(I'll probably never qualify... can't even handle those
multiple-dashes-sigs...)





From Lutz.Schroeer at kybernetik-manufaktur.de  Sun Feb 10 16:09:14 2002
From: Lutz.Schroeer at kybernetik-manufaktur.de (Lutz Schroeer)
Date: 10 Feb 2002 21:09:14 GMT
Subject: Is Parrot dead?
References: <B88C9617.532F%Roman.Fischer@teleweb.at> <a46me3$1ch6pm$1@ID-7529.news.dfncis.de>
Message-ID: <Xns91B1E185DCEFFLatzikatz@139.174.2.56>

"Karl M. Syring" <syring at email.com> wrote in
news:a46me3$1ch6pm$1 at ID-7529.news.dfncis.de: 

> Uhm, I think you have been the victim of last year's April joke
> (http://www.python.org/parrot.html).

Hey, being aware of it being a joke I did like the idea... getting rid of 
all the Perl stuff :) <flame war declared open>


From s.schwarzer at ndh.net  Thu Feb 14 15:00:33 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 14 Feb 2002 21:00:33 +0100
Subject: preference tree like in jdk 1.4
References: <mailman.1013712805.16911.python-list@python.org>
Message-ID: <3C6C1761.6BB16ADA@ndh.net>

Hello Michael

Michael Chermside wrote:
> Instead of using an empty function, use something designed specifically
> for holding fields: a class!
> 
>  >>> class Empty:
>         pass
>  >>> pref = Empty()
>  >>> pref.personalData = Empty()
>  >>> pref.personalData.firstName = 'John'
>  >>> pref.personalData.lastName = 'Cleese'
>  >>> pref.windowSettings = Empty()
>  >>> pref.windowSettings.windowA = Empty()
>  >>> pref.windowSettings.windowA.x = 200

Actually, you use instances of the class to store the values. I think,
you meant that anyway, didn't you? :-)

Sometimes classes are used for such a purpose. However, this may cause
problems if you need more than one "container" (like above). Using
class instances is more common, I assume.

Stefan


From timr at probo.com  Sun Feb 17 01:25:22 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 16 Feb 2002 22:25:22 -0800
Subject: Why Python is like BASIC ...  java "vs" python
References: <3136.203.177.73.44.1013825599.squirrel@ns1.bizlinks.net.ph> <mailman.1013826279.13649.python-list@python.org>
Message-ID: <i4iu6ug73jr2lmpblj2e70e1tci043st0l@4ax.com>

john coppola <john_coppola_r_s at yahoo.com> wrote:
>
>OK fellas, I'm gonna break it down for you.  Python is
>nothing like BASIC or perl.  Neither is Java like C++.

It is silly to argue that Python is NOTHING like BASIC or Perl.

   i = 3
   print i * i

That program is legal and functional in both Python and BASIC.  I daresay
you could not write a similar program that works in BASIC and C, or Perl
and Java.  In this one aspect, Python is like BASIC.

(Actually, I just realized that this program:
   ;
is valid in both Perl and C, so perhaps I need to modify my statement to
say "...a single non-trivial program...".)

>I will admit that I used BASIC.  A friend turned me
>on to Python.  I never looked back.  You cannot
>compare things on basis of how they appear or the
>syntax of the language or certain similarities.

Of course you can.  There are as many ways to compare things as there are
things.  More, actually.

>You will begin to see that basic and python are
>nothing alike even though they are both interpreted. 
>In what way is basic interpreted?  In what way is perl
>interpreted?

They are interpreted in the same way that Python is interpreted, in that
the usual implementations of all three languages are as interpreters that
do not translate to machine language.  The difference between interpreted
and compiled languages is important; it is perfectly valid to say that
"BASIC, Python and Perl are usually implemented with interpreters, whereas
Java, C and C++ are usually implemented with compilers".  In that one way,
at least, BASIC is like Python, and Java is like C++.

I just don't see the usefulness of the the rather strict position you are
taking.  Perhaps you could expand your thoughts for us a bit more.

>Java is not at all similar to C++.  Different.

This is also just silly.  "Similar" is not the same as "identical".  A
competent C++ programmer can pick up a typical Java program and read it
without further training.  There will certainly be aspects that require
explanation, but it's not a complete relearning process.  Java is not at
all similar to RPG, but it certainly IS similar to C++.

>Sorry to be so harsh.  Long day..

In my opinion, your position is not defensible, although I would certainly
invite you to post more on the topic.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From jimd at vega.starshine.org  Thu Feb 28 17:41:11 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 28 Feb 2002 22:41:11 GMT
Subject: dirwalk.py generator version of os.path.walk
References: <a5d36e$1daf$1@news.idiom.com> <ac677656.0202271621.5a134d44@posting.google.com> <a5kv49$ae8$2@news.idiom.com> <just-BED145.11365928022002@news1.xs4all.nl>
Message-ID: <a5mbm7$18vj$1@news.idiom.com>

In article <just-BED145.11365928022002 at news1.xs4all.nl>, Just van Rossum wrote:
>In article <a5kv49$ae8$2 at news.idiom.com>,
> jimd at vega.starshine.org (Jim Dennis) wrote:

>>> I wrote a different implementation of this general concept at:

>>> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/105873

>>> You don't really need to keep a stack of directories and push/pop
>>> things, because with generators you can recurse instead.
>>> Tom
 
  BTW: Tom, I did come across your version in the ASPN a day or so
  after I wrote mine.  I agree it looks cleaner and simpler.
  I might still use a list (FIFO or stack) and while loop instead
  of the recursion. (see below).

>>  But recursion is likely to cost more.  The only state I need
>>  to keep is my current "todo" list of directories.  A recursion 
>>  would store functional state (unless Python supported tail-end
>>  recursion).  So the append/pop (total cost, 3 lines of code)
>>  seems like the lightest weight way to do this.

> Why don't you try it out? Setting up a generator is as expensive as 
> calling a function, but resuming or yielding a generator is _very_ cheap.

> I personally find the recursive version clearer.
> Just

 I guess I'm just wary of recursion, particularly when I've read
 that Python doesn't support tail-end recursion.  

 However, I agree that a good test is worth infinitely more than
 any amount of theoretical speculation.

 I've read that there's some sort of profiling module for Python;
 how would I profile each of these approaches in terms of footprint
 CPU utilization and wall-clock time?  
 
 As I say, I'm still concerned that the generator will have lots 
 of stack frame overhead in recursion while the interative approach 
 just incurs the size of the list.

 I'm considering playing with my dirwalk() a bit more (possibly 
 adding additional options to it, or implementing it as a class and
 writing a set of Decorators to add some of the features that I 
 mentioned in my OP.  I suspect I need to design it as a hybrid; 
 options that affect the depth and traversal (i.e. not crossing
 mount points, setting max depth, etc) should probably be added
 in, things that affect the yield()'d results (anything returned
 by stat(), filename or regex patterns, contents scans, etc) should
 probably be done in Decorators).
 
 I'm just trying to learn how to be a "real programmer" so this is
 mostly for personal edification.  However, it would be nice if I
 created a useful Python module which provided all of the power
 for Python that the "find" command provides to the shell.  (Of course
 we already *can* do most of that with os.path.walk, and we could 
 use popen() on a find command, or we could use either of these
 dirwalk() generators and do our own stats, etc.  However, I'd like
 to write something that makes it a bit easier and more concise.  
 



From niemeyer at conectiva.com  Tue Feb 19 17:10:50 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Tue, 19 Feb 2002 19:10:50 -0300
Subject: Recursively traverse linked list -- help!
In-Reply-To: <1014136534.2891.169.camel@matijek>
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org> <7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek> <mailman.1014126336.32560.python-list@python.org> <7x7kp9blab.fsf@ruckus.brouhaha.com> <1014136534.2891.169.camel@matijek>
Message-ID: <20020219191050.A3778@ibook.distro.conectiva>

> Fair enough! Actually if I pit the LinkedList against the builtin list
> in the case you're talking about (where you never have to do any work to
> find the insertpoint) the LinkedList will come out as a winner already
> at 5000 - 10000 inserts.

Hey!! That's not fair:

> inserts at start of LinkedList
> 20000 0.457922101021
[...]
> inserts in middle of builtin list
> 20000 0.787240982056

You're comparing oranges and bananas here... ;-)

I've repeated your test (using time.clock() instead of time.time()).
Here is the result:

inserts at start of LinkedList
00000 -> 10000  0.64
10000 -> 20000  0.64
20000 -> 30000  0.66

inserts at start of builtin list
00000 -> 10000  0.07
10000 -> 20000  0.07
20000 -> 30000  0.07

inserts in middle of LinkedList
00000 -> 10000  448.46
10000 -> 20000  1365.52
20000 -> 30000  (too long)

inserts in middle of builtin list
00000 -> 10000  0.34
10000 -> 20000  1.04
20000 -> 30000  1.62

This means at least two things: first, the list builtin object is
written in C; and second, the builtin list is not just a dumb array
asking for more memory at every insert.

--
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From amuys at shortech.com.au  Mon Feb 25 18:45:41 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 25 Feb 2002 15:45:41 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com> <a5cfpe$67voi$1@ID-11957.news.dfncis.de> <7xbseeayxk.fsf@ruckus.brouhaha.com>
Message-ID: <7934d084.0202251545.5fb4bbf3@posting.google.com>

Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7xbseeayxk.fsf at ruckus.brouhaha.com>...
> "Emile van Sebille" <emile at fenx.com> writes:
> 
> > "Paul Rubin" 
> > > Both of those build up a new list of the results, instead of
> > > discarding the values.  If the f function takes an integer and
> > > computes a 20-megabyte structure, you've got a problem.  
> > 
> > If you know it's coming, throw it away first:
> > 
> > [x.f() and None for x in items]
> 
> OK, but that buidls a list of None as long as the item list.  Not
> so bad, since it gets reclaimed right away, but still unsatisfying.
> 
> > filter(None,[x.f() and None for x in items])
> > 
> > returns an empty list vs 0 for the reduce.  
> 
> Again there's this intermediate list that's as long as the original list.
> 
> Do you see a way around that?
> 

filter(lambda x:x.f() and 0, items) ?

It's identical to the the reduce method posted previously, but avoids
the tuple indexing required by reduce's use of a binary operator vs.
filter's unary.

Of course my first preference would be the list comprehension,
discarding the redundant list, and if that was prohibitively expensive
falling back to the original imperative style.

Andrae Muys


From sholden at holdenweb.com  Mon Feb 11 15:00:17 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 15:00:17 -0500
Subject: mySQL access howto?
References: <mi6g6uk490gnfrs3lm69i1ogbauu71vf5n@4ax.com>
Message-ID: <bpV98.154026$KM2.6952436@atlpnn01.usenetserver.com>

"Jan Johansson" <spamfilter2 at mupp.net> wrote in message
news:mi6g6uk490gnfrs3lm69i1ogbauu71vf5n at 4ax.com...
> Examples / classes / howtos on the subject gotta be arpund somewhere..
> Pointers anyone?
One fairly terse source of information:

    http://sourceforge.net/projects/mysql-python/

Then you could take a look at

    http://www.python.org/topics/database/

because MySQLdb sticks fairly closely to the Python DB-API v2.0

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From edream at tds.net  Mon Feb 25 16:18:59 2002
From: edream at tds.net (Edward K. Ream)
Date: Mon, 25 Feb 2002 21:18:59 GMT
Subject: Making breakpoints functional in IDLE on Windows
Message-ID: <3C7AAA43.82FEC0D0@tds.net>

Hello all,

This morning Guido confirmed that breakpoints do not, in fact, work in
IDLE on any Windows machine(!)  The problem has to do with the case of
file names.

Here is how make breakpoints functional in IDLE.  The fix required 2
additional lines of code.  Both lines are marked with # EKR.

Insert 1 line in Python\Tools\idle\Debugger.py.  Change:

def set_break(self, filename, lineno, temporary=0, cond = None):
    import linecache # Import as late as possible
    line = linecache.getline(filename, lineno)
    ...

to:

def set_break(self, filename, lineno, temporary=0, cond = None):
    import linecache # Import as late as possible
    filename = self.canonic(filename) # EKR
    line = linecache.getline(filename, lineno)
    ...
    
Insert 1 line in Python\Lib\bdb.py.  Change:

def canonic(self, filename):
        canonic = self.fncache.get(filename)
        if not canonic:
            canonic = os.path.abspath(filename)
            self.fncache[filename] = canonic
        return canonic
To:

def canonic(self, filename):
        canonic = self.fncache.get(filename)
        if not canonic:
            canonic = os.path.abspath(filename)
            canonic = os.path.normcase(canonic) # EKR
            self.fncache[filename] = canonic
        return canonic

This works for me on Windows XP and Python 2.1.  I'll let you know if
any
problems come up.

Edward
--------------------------------------------------------------------
Edward K. Ream   email:  edream at tds.net
Leo: Literate Editor with Outlines
Leo: http://personalpages.tds.net/~edream/front.html
--------------------------------------------------------------------


From joekm at earthlink.net  Tue Feb 26 13:22:11 2002
From: joekm at earthlink.net (Joe)
Date: Tue, 26 Feb 2002 18:22:11 GMT
Subject: Stumped by PhotoImage issue
Message-ID: <3c7bcefb.501573@news.earthlink.net>

I've actually seen that there are others with a similar problem that
have posted on the web.  I've read G.R's FAQ and looked at the
solutions others have used but I am still confused.....

In a program that I am writing, I have an "about" dialog box that is
written as a class that is in turned called from the "help : about"
pulldown menu item in my "menubar" class.  This "about" dialog has
three .gif images that are handled by PhotoImage.  The problem is that
they appear to be displaying transparant.

Now, if I pull the class out of my main program and wrap a minimalist
mainloop around it, everything works fine.  Also, if I intentionally
put an error in the part of the class that draws the button, the gif's
are drawn without a problem (but I get an error message and the button
does not draw).  

I've tried setting instance variables, global variables, etc. and I'm
running out of ideas.  Any suggestions would be appreciated.  Just so
you know, this is my first original python program and my first real
intro into object oriented programming.  OTOH, I have read everything
I could download and purchased a couple of books on the subject
including "Python and Tkinter Programming".

..and no, I do not wish to use PMW at this time, I wish to gain
experience with Tkinter first.

The class follows (needs to be un-word-wrapped):

class About:

    def __init__(self, parent, GraphicsPath):

        self.parent = parent

        self.deltaLogo = PhotoImage(file=GraphicsPath+'DeltaLogo.gif')
        self.pythonLogo =
PhotoImage(file=GraphicsPath+'PythonPowered.gif')
        self.tkLogo = PhotoImage(file=GraphicsPath+'TCL-TK_logo.gif')

        AboutDialog = Toplevel(self.parent, relief=RAISED,
borderwidth=3)
        AboutDialog.geometry('+300+150')
        AboutDialog.overrideredirect(1)
        AboutDialog.tkraise()
        self.AboutBorder = Frame(AboutDialog, relief=GROOVE,
borderwidth=2)

        title = Label(self.AboutBorder, text='Simple Airframe Fracture
Evaluation Tool [SAFE-T]',
                             font=('Ariel', 11, 'bold'))
        title.pack(side=TOP, padx=2, pady=5)
        
        global Revision_Type, Revision
        version = Label(self.AboutBorder, text=Revision_Type+'
'+Revision)
        version.pack(pady=5)
        
        copyright1 = Label(self.AboutBorder, text='Copyright 200X by
Joseph K. Mooney and')
        copyright1.pack(pady=5)

        self.delta = Label(self.AboutBorder, image=self.deltaLogo)
        self.delta.pack() 

        copyright2 = Label(self.AboutBorder, text='ALL RIGHTS
RESERVED\nSee License Agreement for Further Information\n')
        copyright2.pack()
        
        description = Label(self.AboutBorder, text="Programed in
PYTHON - GUI  built with Tkinter")
        description.pack(pady=8)

        self.python = Label(self.AboutBorder, image=self.pythonLogo)
        self.python.pack(side=LEFT) 

        self.tclTk = Label(self.AboutBorder, image=self.tkLogo)
        self.tclTk.pack(side=RIGHT) 

        self.AboutBorder.pack(fill=X, expand=YES)

        self.OK = Button(self.AboutBorder, text='OK', font=('Arial',
9, 'bold'), command=AboutDialog.destroy)
        self.OK.place(relx=.48, rely=.8, anchor=NW)
#


From wookiz at hotmail.com  Sat Feb  9 05:22:44 2002
From: wookiz at hotmail.com (wooks)
Date: 9 Feb 2002 02:22:44 -0800
Subject: Perhaps I am just dumb
Message-ID: <8852d33c.0202090222.34be7da2@posting.google.com>

New to the language, new to OO but I do understand the basic concepts,
not at all new to programming, trying to do the right thing by
searching archives before asking questions. Yet I feel I need to be
doing a university degree course in philosophy or mathematics to
follow almost any discussion. If that's the way it is fair enough but
Python is supposedly simple to learn (thus far I've had an easier time
of it learning Perl). If someone comes and says I've had the following
understanding of concept X since I was in grade 2 why is it not so in
Python, the answers seem to be heavy on the philosophical
justifications fo the Python way and light on the practical benefits.
The answers also tend to be in a totally different intellectual
context.

True I haven't been around here that long but it's a 1st impression
and it counts. Perhaps it's because so many come to Python with
baggage from another language so that issues they raise are addressed
in that context rather than conceptually.

Bought Python for Win32 book and already been caught out (Part 1 was a
breeze part 2 and I shuddered to a grinding halt). For perspective I
am a tester who wants to learn Python because of it's excellent unit
test framework and it's scripting glue like abilities (not interested
in building applications). If vbunit was as easy to figure (or had as
good a tutorial) as PyUnit (thank you Mark Pilgrim) I probably never
would have come here.  As it is perhaps getting to grips with vbunit
might be the lesser evil.

The benefits of using of Python as a scripting glue like language are
widely touted but don't seem (I am probably wrong but it's a 1st
impression) to be well catered for. As I said I am not interested in
building applications. Is there a forum for the dumber less demanding
Python user?


From fredrik at pythonware.com  Wed Feb 13 12:05:59 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 17:05:59 GMT
Subject: Wow.
References: <mailman.1013536760.2921.python-list@python.org> <slrna6j64f.gku.mlh@vier.idi.ntnu.no>
Message-ID: <X1xa8.25539$l93.4756472@newsb.telia.net>

Magnus Lie Hetland wrote:
> I think I've heard the term before -- but Google only gives 5 hits.
> Not in much use, obviously :)

search for "wackyparse" instead, and you'll find some
more hits, including a definition.

(searching usenet groups instead of the web also helps)

goggle-wacking-ly yrs /F




From martin at v.loewis.de  Sat Feb 23 18:17:09 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 00:17:09 +0100
Subject: Is there a python parser generator that is similar to JavaCC from webgain?
References: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com>
Message-ID: <m38z9jpz4a.fsf@mira.informatik.hu-berlin.de>

"Shen Wang" <shenwang at ameritech.net> writes:

> As title

There are a number of Python parser generators. Since I don't know
JavaCC (and don't want to spend time learning it), I cannot tell you
which of those is similar to JavaCC.

Why don't you just describe the features you like in JavaCC? Then
people can better tell which generator would meet these requirements.

Regards,
Martin


From store_li at sina.com  Sat Feb 23 09:14:02 2002
From: store_li at sina.com (Ricky)
Date: Sat, 23 Feb 2002 22:14:02 +0800
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones>
Message-ID: <a58885$5acoc$1@ID-12869.news.dfncis.de>

Yes, as you guess, I haven't call .write method, because I was confused by
the usage of write method. It always required a fp object, which I totally
don't what it means. Is it a object create by a=ConfigParser.ConfigParser()
or a file name or a section name?

Please help me! Thank you!


"Graham Ashton" <graz at mindless.com> wrote in message
news:fUJd8.418$PE2.43612 at stones...
> On Fri, 22 Feb 2002 16:36:22 +0000, Ricky wrote:
>
> > I intend to use ConfigParser module to manage my INI file. I can use get
> > to retrieve data from my ini file, and use set to change option value in
> > Python, just in python, after I closed python, I found that all changes
> > I made in python lost. There is nothing altered in my ini file!
> >
> > How can I make my ini file to be changed?
>
> As you haven't posted any code I can't be sure, but I suspect from what
> you've said that you've not called the write() method on your
> ConfigParser object.
>
> Read the module's docs.
>
> --
> Graham




From christoforou at gaaros.com  Wed Feb 13 14:16:40 2002
From: christoforou at gaaros.com (christoforou at gaaros.com)
Date: Wed, 13 Feb 2002 19:16:40 GMT
Subject: Python License Issues
Message-ID: <20020213191640.13956.qmail@mail.yifansoft.com>

Hello from rainy London

We have an issue with the python license which basically boils down to this: 

If we ship an application that uses python, is the user of our application 
bound by the python license agreement (PSF)? It seems so given clause 8 of 
the PSF: 

   8. By copying, installing or otherwise using Python 2.1.2, Licensee
      agrees to be bound by the terms and conditions of this License
      Agreement. 

If such is the case we will encounter problems with clients that do not like 
clause 4 of the PSF which states: 

 4. PSF is making Python 2.1.2 available to Licensee on an "AS IS"
    basis.  PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
    MPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND
    DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS
    FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON 2.1.2 WILL NOT
    INFRINGE ANY THIRD PARTY RIGHTS. 

Some clients demand a certain level of 'protection' from situations of gross 
misconduct, say python has stolen major chunks of code from other apps and 
thus exposing them to huge legal risks. We can consider accepting liability 
on all third party code we use, for such gross misconduct situations but is 
it technically possible to do so given the PSF license? Ideally we would 
like to negotiate a single license agreement with these clients (our license 
agreement) which provides additional liability over and above to that which 
PSF provides. However given clause 4 it seems that the client *must* 
directly be bound by the PSF license. 

Any comments? 

Pavlos



From aleax at aleax.it  Fri Feb  1 12:36:07 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 17:36:07 GMT
Subject: calling upper constructor
References: <3C5A04E9.3040502@sunet.ru> <l23k5ucbs5f1k0v3ppgr67svhkcvoghol3@4ax.com>
Message-ID: <bmA68.16936$om6.396945@news1.tin.it>

Courageous wrote:

>>class abc(base):
>>   def __init__(self):
>>     # how do I call base's init here?
> 
> class derive (base):
>     def __init__ ( self, x ):
>         base.__init__( self, x )

This correctly answers the literal question as posed.

However, the intention behind the question may be to
"upcall to the 'right' superclass constructor".  Multiple
inheritance means that base.__init__ may not be the
'right' superclass-constructor to call at this point, e.g.:

class base:
    def __init__(self):
        print 'base init'
 
class derive1(base):
    def __init__(self):
        print 'derive1 init'
        base.__init__(self)
 
class derive2(base):
    def __init__(self):
        print 'derive2 init'
        base.__init__(self)
 
class doublyderive(derive1, derive2):
    def __init__(self):
        print 'doublyderive init'
        derive1.__init__(self)
        derive2.__init__(self)
 
dd = doublyderive()

[alex at lancelot alex]$ python dd.py
doublyderive init
derive1 init
base init
derive2 init
base init

This leads to running base init twice -- sometimes
innocuous, sometimes dangerous.  As you can see,
if derive1 and derive2 each call base.__init__ directly,
there's really no way doublyderive can ensure each
of its direct bases is initialized but the bases' base is
only initialized once.

Python 2.2 to the rescue!  Here's how you can code
this in Python 2.2:

class base(object):
    def __init__(self):
        print 'base init'
 
class derive1(base):
    def __init__(self):
        print 'derive1 init'
        super(derive1, self).__init__()
 
class derive2(base):
    def __init__(self):
        print 'derive2 init'
        super(derive2, self).__init__()
 
class doublyderive(derive1, derive2):
    def __init__(self):
        print 'doublyderive init'
        super(doublyderive, self).__init__()
 
dd = doublyderive()


[alex at lancelot alex]$ python dd.py
doublyderive init
derive1 init
derive2 init
base init


Voila -- thanks to the magic of super(), every initializator
runs once, and only once.  Note that base needs to derive
from object (directly or indirectly) for this to work, i.e., it
must be a "new-style class"; a classic class, one without
such derivation, closely mimics 2.1's semantics and thus,
alas, cannot be used in a super() call.


Alex



From aleax at aleax.it  Mon Feb 18 11:21:42 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 16:21:42 GMT
Subject: Subclassing a read-only class
References: <mailman.1014046530.13105.python-list@python.org>
Message-ID: <qS9c8.83559$3J1.2671693@news2.tin.it>

Gerhard H?ring wrote:

> Hello,
> 
> For compatibility testing, I'd like to put an additional method into
> window object I get from the curses module. Unfortunately, if I try to
> do so, I get:
> 
>     TypeError: object has read-only attributes
> 
> I'm pretty sure there is a solution, probably with a proxy object. But
> currently I just don't get it.

You can generally substitute automatic-delegation in lieu of
inheritance (as long as nobody is so uncouth as to typetest,
e.g. with isinstance -- in which case, you're in trouble).

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52295 gives
a simple example ('inheriting' [via automatic delegation] from a
file object back in Python 2.0 / 2.1 when you couldn't actually
just subclass file as you can today) -- I hope you find the idea
easy to adapt to your specific needs.


Alex




From phd at phd.pp.ru  Wed Feb 20 03:42:30 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 20 Feb 2002 11:42:30 +0300
Subject: =?koi8-r?Q?R=E9f=2E_:_Re:_strange_libtk8=2E3=2Eso:_open_failed?=
In-Reply-To: <"020219214906Z.WT25538. 28*/PN=Francois-regis.Chalaoux/O=RESEARCH/PRMD=SANOFI/ADMD=ATLAS/C=FR/"@MHS>; from Francois-regis.Chalaoux@sanofi-synthelabo.com on Tue, Feb 19, 2002 at 03:23:08PM +0100
References: <"020219214906Z.WT25538. 28*/PN=Francois-regis.Chalaoux/O=RESEARCH/PRMD=SANOFI/ADMD=ATLAS/C=FR/"@MHS>
Message-ID: <20020220114230.D15274@phd.pp.ru>

On Tue, Feb 19, 2002 at 03:23:08PM +0100, Francois-regis Chalaoux wrote:
> * The sun machine is:
> SunOS lads01 5.8 Generic_108528-03 sun4u sparc SUNW,Ultra-60
> 
> * On command line Dynamic linker find libtk8.3.so and the librairie is there:
> 
> lads01 /tmp/Python-2.1.1 % ldd /usr/local/bin/python
>         libtk8.3.so =>   /usr/local/lib/libtk8.3.so

   Looks good. What is the difference between this command line and the way
you tried to use Tk? What is $LD_LIBRARY_PATH in your command line
environment?

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



From dsavitsk at e-coli.net  Mon Feb 11 13:18:27 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 18:18:27 GMT
Subject: [ANN] ecp_printer : ActiveX wrapper for VB Printer object
Message-ID: <TVT98.3387$pC1.698927@newssrv26.news.prodigy.com>

This isn't strictly Python related, but that's what I use it for.

ecp_printer is both a wrapper for VB6's built-in Printer object and a print
previewer for displaying would be printed stuff to the screen.  The
previewer is more intended to avoid the wasting of paper that typically
precedes using the Printer object successfully, but it could be used in a
program as well. It is a quick solution to print simple documents.  To use
it, one does something like the following:

>>> import win32com.client
>>> x = win32com.client.Dispatch('ecp_printer.Printer')
>>> x.ecpPrint('Hello World!')
>>> x.EndDoc()

A windows installer and the source are available from
http://www.e-coli.net/software. There are some test scripts and examples
there as well.




From threeseas at earthlink.net  Sun Feb 10 13:22:59 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 10 Feb 2002 18:22:59 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <3c66ac80.2522937@127.0.0.1>
Message-ID: <2022.806T252T8483762threeseas@earthlink.net>

On 10-Feb-02 12:19:18 (Five Fresh) Fish <root at 127.0.0.1> wrote:
>On Sun, 10 Feb 2002 15:51:51 GMT, "Timothy Rue" <threeseas at earthlink.net>
>wrote:

>>Funny how Hitler chased Einstein and Fermi out of his territory, their
>>homeland.....
>
>Does this mean this thread can be ended, now?


-From: "(Five Fresh) Fish" <root at 127.0.0.1>
-Subject: Re: (Long) Re: Autocoding project proposal.
-Date: 10 Feb 2002 17:19:18 GMT
-Message-id: <3c66ac80.2522937 at 127.0.0.1>
-Path: [snipped]
-Newsgroups: gnu.misc.discuss,comp.lang.python
-Cancel-Lock: sha1:vC9l4Qkqan7XZyQOCKvfUSBRruY=
-References: [snipped]
-X-Newsreader: Forte Agent 1.5/32.452

-X-No-Archive: yes

-MIME-Version: 1.0
-X-NFilter: 1.2.1-b1
-NNTP-Posting-Host: 142.179.60.23
-X-Trace: news1.telusplanet.net 1013361558 142.179.60.23
 (Sun, 10 Feb 2002 10:19:18 MST)
-NNTP-Posting-Date: Sun, 10 Feb 2002 10:19:18 MST
-X-Received-Date: Sun, 10 Feb 2002 09:19:19 PST
 (newsmaster1.prod.itd.earthlink.net)
-Xref: [snipped]


Since you are not existing in the public archives...

Einstien and Fermi leaving their homeland didn't end the war.

Yesterday I read thru all my posts/responses in this newsgroup (just my
words). Quite a contridiction in comparison the words chosen and used by
others regarding myself and what I've said. Main reason for doing so was
to see all this negativity others have been claiming about me. The only
thing I saw was my getting angry about such claims.

If you really existed, I'd suggest you do so, to see the contridicion for
yourself. But then you are apparently a contridiction yourself.

you just can't stop physics. So how about helping to make better use of
it?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From chris.gonnerman at newcenturycomputers.net  Wed Feb 13 10:32:03 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Wed, 13 Feb 2002 09:32:03 CST
Subject: ANN: anycons - Simple Universal Console I/O
Message-ID: <200202131535.JAA12067@scotland.k12.mo.us>

Somebody on the list asked about generic console I/O which works in
Windows and Unix/Linux both.  (Sorry, can't remember who...)  So, I sat
down this morning and wrote it.

The OP listed a small number of basic operations he wanted support for;
such a small list, in fact, that it took only about an hour to implement for
both platforms.  On Windows, my anycons package needs WConio or
a working curses; on Unixoids you need a working curses (only thing that
works).  The following calls are supported:  Init(), Close(), Clear(), 
Write(stringval), Read(), and Goto(x, y).  Pretty primitive but I think it meets
the OP's requirements.

Future development (whenever I feel like it) may include special key 
handling, a line or field editing function, and possibly a pure-ANSI backend.

Get it (if you care) at:

	http://newcenturycomputers.net/projects/anycons.html




From ivano.delpiccolo at degroof.be  Fri Feb 22 05:41:39 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Fri, 22 Feb 2002 11:41:39 +0100
Subject: fork problem
Message-ID: <3c76205c$0$7111$4d4efb8e@news.be.uu.net>

I have the following error when i try to fork :

Traceback (most recent call last):
  File "explorer_proxy.py", line 33, in ?
    child = os.fork()
AttributeError: 'module' object has no attribute 'fork'

Here is my code :
child = os.fork()
 if child == 0 :
        print "\nEntering in child process"

Where is my error ?????

Thanks




From starx at pacbell.net  Fri Feb  8 17:38:56 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 8 Feb 2002 22:38:56 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4>

Aahz Maruch || Fri 08 Feb 2002 11:10:47a:

> I know some of y'all still keep roots to the Perl community, so I was
> wondering if there's any kind of definitive nickname for Perl
> programmers, the way we call ourselves Pythonistas?

Hm, i don't seem to remeber seeing this Pythonistas thing till maybe two 
days ago. Anyway, where did it come from-- based on some word? And HTF do 
you pronouce it? :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From philh at comuno.freeserve.co.uk  Sun Feb  3 20:41:31 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 01:41:31 +0000
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net>
Message-ID: <slrna5rpmb.19d.philh@comuno.freeserve.co.uk>

On Mon, 04 Feb 2002 00:00:53 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 30-Jan-02 06:52:48 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>>On 30 Jan 2002 00:48:40 GMT, Peter Seebach <seebs at plethora.net> wrote:
>
>>>Anyway, it looks to me like Python is the wrong language for what you want
>>>to do.  C# is almost certainly the best choice for this; it's designed for
>>>the kinds of tasks you're looking at,
>
>>Is Tim trying to lock people in to the Microsoft monopoly, then?
>
>Why do you make up things by taking things out of context?

Why do you misinterprete a joking question as a personal attack on you?
You seem to have very fragile self-esteem...

>Are you naturally dishonest or do you work at it?

Yes. I mean no. Oops I've been caught lying again.

>Autocoding will be a nice thing to have, it'll get rid of all the arrogant
>jerks.

Then I suggest you create it.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From matthias_j at lycos.com  Wed Feb 13 17:30:03 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 13 Feb 2002 14:30:03 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <slrna6lbrp.guj.mlh@vier.idi.ntnu.no>
Message-ID: <d7d5ebdf.0202131429.16a21cca@posting.google.com>

Hi Steve Ferg ;

I tried the link to get more info but it just displayed page not found.

Please could you let me know where to get more info for EASYGUI

Thanks


From jacek.generowicz at cern.ch  Thu Feb 14 11:12:41 2002
From: jacek.generowicz at cern.ch (Jacek Generowicz)
Date: 14 Feb 2002 17:12:41 +0100
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013700446.9798.python-list@python.org>
Message-ID: <tyf3d04f3eu.fsf@pb-s-513-1-26-6-1.cern.ch>

Gustavo Cordova <gcordova at hebmex.com> writes:

> Espa?ol correcto es una modalidad de escritura que pocos aprenden.

Lo mismo se puede decir de qualquier idioma.


From sholden at holdenweb.com  Mon Feb  4 10:40:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 10:40:04 -0500
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com> <7xzo2pqopt.fsf@ruckus.brouhaha.com>
Message-ID: <WZx78.40291$KM2.1247277@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote ...
> "Steve Holden" <sholden at holdenweb.com> writes:
> > > If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> > > it isn't backwards compatible.
> > >
> > Erm, don't you seem to be using a different version of backwards
> > compatibilty from most people? As far as I'm concerned, if I run version
> > 1.5.2 of a language, and all my programs continue to run, then the new
> > release provides backwards compatibility.
>
> Yes, but they DON'T continue to run, then the new release DOESN'T provide
> backwards compatibility.

But surely your example above is about asking 1.5.2 or 2.1 to do a "//"
operation? Anyway, we can omit all further discussion on *that* topic since
it's clearly not what's foremost in your mind.
>
> If I write in 2.2 or later 2.x's, I have to choose between two
> alternatives, both bad:
>
>  1) use "/" the traditional way.  My programs will then work under 2.1
>     and earlier, but they'll break under 3.0 and later.
>
>  2) Use "from __future__ import division" and use // everywhere I
>     want traditional integer division.  That means that in the Python
>     versions that most people today are actually using, my programs
>     won't even parse, let alone run correctly.
>
Only because you've already decried *the* forward-compatible alternative:

3) Use divmod()[0]

You said "yerch" (or something similar to this), but surely from an
engineering point of view it gets the pragmatic result you want.

[what does "current" mean...]
>
> But Tim has said there might be a 1.5.3 release.  That means that the
> 1.5 series is still being "issued" and is not obsolete.  The 2.1
> series is definitely still being issued--a new 2.1 release came out
> just a few weeks ago, and I believe Tim himself recommended it as the
> most stable Python version currently available.  Why on earth would
> anyone develop a new 2.1 bugfix release if they thought everyone
> should switch to 2.2?
>
Tim admitted this as a remote possiblity rather than suggesting it was
probable. As in "there's nothing to stop anyone doing it". Certainly 2.1.2
is the most stable release, but it doesn't implement the current Python
language, whose definition is now implemented by the 2.2 release.

> > > If they're unlikely to ever release another 1.2.x or 0.9.x, then
> > > those versions are not current, even if I release a 1.2.x myself.
> > >
> > Nonsense. Do you understand the meaning of the word "currency"? And stop
> > relying on the "official Python maintainers" - one of these is Tim
Peters,
> > and you continue to disagree with him about this matter...
>
> Tim is the one who said there might be a 1.5.3 someday, which is part
> of why I refer to the 1.5 series as still current, just like the 2.1
> series.  Of course he said it like it (1.5.3) was just a theoretical
> possibility rather than something actually likely to happen, so I
> won't hold out for that part of "current".  The other part--that lots
> of systems still run it--is valid and compelling.
>
Indeed. So just write in the forwards-compatible subset of 1.5.2, and your
problem goes away.

> > Coding in a way that requires 2.2 inconveniences *only* those users who
a)
> > want to use my code *and* b) won't move forward to 2.2.
>
> Even if they're willing to switch to 2.2, making them do so
> inconveniences them.  But why should someone be willing to switch to
> 2.2 today, unless they need the new features for something?  It has a
> lot of major rework and I'd rather let the maintainers get the bugs
> out before I run production code under it.  I might play with 2.2 at
> home, but if I was installing a new production system today, I'd use
> 2.1.1.

I'd use 2.1.2, but I presume you've just overlooked this January 16 release.
I must be missing something. If I'm not using 2.2 features, I have no need
to switch. So, surely, if I *want* to use those features then I will switch
regardless of the inconvenience.

>
> If, someday, there are just a few throwbacks using <= 2.1 (say less
> than 10% of all the users), then it will be reasonable to casually use
> 2.2 features in programs for general distribution.  Right now, it's
> probably more like 90% using <= 2.1.  Publishing a script that
> requires 2.2 is like publishing a web page that only works with IE6
> (or whatever the latest browser is).  It's simply obnoxious to do that
> without a good reason.
>
Aah, "for general distribution". In that case, you are stuck with either
remaining in the forwards-compatible 1.5.2 subset, or distributing the
interpreter with your applications. We can probably agree that while the
latter is practical for large applications (via py2exe of the McMillan
installer, or in the same way Zope does), for smaller programs it simply
carries too much overhead. But I see you wanting the best of both worlds.

> > I suppose that our writing these posts in English inconveniences
> > many of the group's readers for whom English is not the first
> > language. They persist in reading them because the benefits outweigh
> > the disadvantages (in most cases, anyway, though I'm beginning to
> > have my doubts about this particular thread). So, do you want 2.2,
> > or 1.5.2. And please *don't* say you want to be able to program 2.2
> > features in 1.5.2.
>
> I don't want to use 2.2 features in 1.5.2.  I also don't always get to
> choose what version I run, and I almost never get to choose what
> version other people run.  On my personal machine I use 2.1.  On
> Ruckus (which isn't mine), 1.5.2 is installed and I use that.  Some
> other machines I use have even older versions (1.2 or 1.3 from old
> Debian releases).
>
> > I still don't see why you want to be able to use "from __future__" in
1.5.2
> > programs. What's the point of importing future behavior when you can if
you
> > also want to run in environments where such behavior isn't available?
>
> I don't think I asked to use from __future__ in 1.5.2.  Maybe you
> didn't understand my request.  I wanted to put something like
>
>    try: from __backward_compatibility__ import old_division
>    except ImportError: pass
>
> in all my current scripts.  This is perfectly fine and won't break
> anything in 1.5.2 and 2.1.x.  The import will fail and the caught
> exception will turn it into a do-nothing.  The request is that in 3.0,
> the import should succeed and make my scripts keep working the same
> way the did under older versions.
>
> > > I'd non-rhetorically say that for most users, 1.5.2 or 2.1 is current
> > > and 2.2 is bleeding edge.  (I'm still using 2.1 and plan to get around
> > > to upgrading to 2.1.1).
> > >
> > This *is* purely rhetoric, despite your protestations, until you provide
> > evidence.
>
> The evidence is that 2.2 has been out (non-beta) for only about 1 month,
> while the older versions have been around for years.  Lots more people
> are using the older versions.
>
I was talking about the language definition, you about the interpreter.

> > So, what about 1.5.2 and 2.1 programs won't run in 2.2? THOSE features
are
> > the ones that break backwards compatibility (isn't this where we came
in?)
> > not the new ones that won't run in older versions.
>
> That is a problem too.
>
But a (relatively) minor one. RedHat have stuck with 1.5.2 for rather longer
that was strictly necessary for Anaconda support, they could have updated it
relatively easily but presumably backed off for engineering reasons.

> > No. It has been PEP'd for a long time that division breaks in 3.0+,
> > and you'd better get used to the idea if you want to move forward
> > [in a year or two].
>
> I don't agree with the notion of changing the meaning of basic arithmetic
> operators in a supposedly mature language, but at least I can understand
> the motivation for it.  I don't see at all why some mechanism can't be
> provided to turn off the change ("import from past") just like there's
> now a mechanism to turn on the change ("import from future").
>
Because Guido doesn't feel it will be good for your sanity.

> > > I want my programs to run the same way across the different Python
> > > versions that people are using now and will be using in the near
future.
> > > Am I really the first person to want that?
> > >
> > No, you're not even the first person to be so wrong-headed in the way
you
> > decry the chosen path for the language's development <0.8 wink>. Most
unlike
> > you, if I might make a personal observation. Just the same, I claim that
> > Python supports a *remarkably high* degree of backwards compatibility.
Very
> > few 1.5.2 programs broke in 2.0, and most of those (from memory) were
using
> > a non-documented "feature" of the sockets library.
>
> The rand module also disappeared, completely gratuitously as far as I
> can tell.  That's what broke the Redhat scripts.
>
I had overlooked that.

> > Hey, under 3.0 you'll be able to redefine __div__ for everything so you
get
> > exactly the behavior you want. Not sure I'd like to maintain that code,
> > though ;-)
>
> Is that for real?  Is there a bad performance hit?  If that makes a
> way to keep my scripts working without slowing them down too much,
> that might be a reasonable solution.

Well, I can't see __div__ being removed in 3.0, and it's present in the
language right now. If everything's going to inherit from object in 3.0
maybe there'll be some way to use "rubintegers" instead of the ordinary kind
everywhere. Woudn't bet the farm on it, though: the compiler won't know you
want them in the code it compiles. I was being kind of tongue-in-cheek,
sorry.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From robert_kuzelj at yahoo.com  Thu Feb 21 10:04:52 2002
From: robert_kuzelj at yahoo.com (Robert Kuzelj)
Date: 21 Feb 2002 07:04:52 -0800
Subject: pythonisten stammtisch in muenchen
Message-ID: <eaef2e43.0202210704.1e6509d5@posting.google.com>

hi,

nachdem also auch die berliner ihren stammtisch gehalten haben,
dachte ich, es waere doch schoen eine aehnliche veranstaltung
auch in muenchen zu organisieren. 

deshalb hier der aufruf an den eventuellen muenchner phytonista
"bitte meldet dich!"

wenn es ein paar interessenten gaebe, wuerde ich mich bereit
erklaeren die organisatorische handhabung zu uebernehmen.

bei interesse also bitte eine mail mit
terminvorschlaegen und ort an mich.

ciao robertj


From mwh at python.net  Thu Feb 21 07:12:50 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 12:12:50 GMT
Subject: Why this behavior? (was: Stupid Lambda Tricks)
References: <3c74c211.20012718@news.isar.de> <js5d8.116$Az6.466@news.hananet.net>
Message-ID: <uwux7kp3h.fsf@python.net>

Jonathan Gardner <jgardn at alumni.washington.edu> writes:

> I've always wondered:
> 
> >>> "abc".upper()
> "ABC"
> >>> [3,1,2].sort()
> < nothing >
> 
> Why is that? Is it just easier to implement it this way, or is there some 
> sound philosophical reason for this?

Well, strings are immutable, for one thing...

Cheers,
M.

-- 
  I've reinvented the idea of variables and types as in a
  programming language, something I do on every project.
                                          -- Greg Ward, September 1998


From maxm at mxm.dk  Tue Feb  5 05:35:33 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 05 Feb 2002 10:35:33 +0000
Subject: WANTED: IBM 1401 AUTOCODING WAREZ [was: Autocoding - the resistance to expect from psuedo programmers]
References: <mailman.1012829589.12111.python-list@python.org> <a3ndqq$h6c$1@newshost.accu.uu.nl>
Message-ID: <3C5FB575.8050005@mxm.dk>

Martijn Faassen wrote:

> FOR IMMEDIATE DISPATCH TO ALL PSU MEMBERS -- TOP SECRET

That was a mistake. Now the secret



From aahz at panix.com  Sat Feb  2 13:18:52 2002
From: aahz at panix.com (Aahz Maruch)
Date: 2 Feb 2002 10:18:52 -0800
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012661825.25959.python-list@python.org>
Message-ID: <a3haic$kuo$1@panix2.panix.com>

In article <mailman.1012661825.25959.python-list at python.org>,
Jason Orendorff <jason at jorendorff.com> wrote:
>Aahz wrote:
>>
>> [...] nine times out of ten, if you think you disagree with Guido's
>> intuition, you'll change your mind later.  As Alex (I think it was Alex)
>> pointed out a few months ago, Guido's ability to do good design work
>> vastly outstrips his ability to *explain* his thought processes.
>> 
>> Therefore, as long as you emphasize "*seemingly* arbitrary", you won't
>> get flamed by me.
>
>I agree with all that.  Not to take any credit away from the BDFL,
>but I also think that about 4 times out of 10, what's important is not
>his intuition, but simply that one (any) of the alternatives be chosen
>and endorsed, so we can get on with it.

Oh.  Okay, fair enough.  And I'm sure you'll agree that this is a
*necessary* skill in a project leader.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From cfelling at iae.nl  Sat Feb 23 12:37:58 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 23 Feb 2002 18:37:58 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl> <a56o39$7dg$1@animus.fel.iae.nl> <3C7761D4.7050100@student.kun.nl>
Message-ID: <a58k1m$8t1$1@animus.fel.iae.nl>

husam <h.jehadalwan at student.kun.nl> wrote:
> Carel Fellinger wrote:
>> husam <h.jehadalwan at student.kun.nl> wrote:
...
>> var, what about trying it with `os.popen("/home/programs/profit", "w)'

> Off course, it works this way. But, then I have to execute another 
> script to process the output. Therefor I chosen for "popen2" so that I 
> can process the output in the same script.

Okee, so it works with lots of files and popen and fails for popen2.

A way to circumvent the problem then might be to use popen and redirect
the output to a temp file like:

    os.popen("/home/programs/profit >/tmp/tempfile", "w")

and read this file in the same srcipt when `profit' is done.

But this would deprive us from understanding what went wrong:)


>>     i = os.popen("/home/programs/profit ", "w")
>> 
>> Should this ---------------------------^ space be there?

> The space is needed here, otherwise the command and it's arguments 
> passed to the progam, get passed like this: "CommandArguments" instead 
> of " Command Arguments"

I don't get this.  All arguments to `/home/programs/profit' have to be
given in the popen* call.  I don't see arguments in that call so I
don't see need of the space here.  Or are you using a different script
than what you showed us here?  Even slight differences might be
relevant, so if you're running different sripts how about showing
them?

>>         i.write(" mobile /home/data/pdbs/" + PDB + "\n")
>> 
>> Should this -----^ space be there?

> The same explanation as above!

When you write to `i' you don't add arguments to the commandline used
to start `profit',  you're writing to the inputstream of `pdbs', quit a
different thing. So I still don't see the need for this space.



-- 
groetjes, carel


From kfarmer at thuban.org  Mon Feb  4 05:32:33 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 04 Feb 2002 10:32:33 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com> <3C5B90F3.3020503@skippinet.com.au> <CkN68.6145$F6.3643599033@newssvr14.news.prodigy.com> <3C5E34BA.6030702@skippinet.com.au>
Message-ID: <5rt78.7926$Da3.4244083767@newssvr14.news.prodigy.com>

Still isn't working -- it's printing "HTBORDER" as before, so it's
presumably returning it as well.  Win's still changing the cursor and
allowing resize as if it were still getting HTBOTTOM, etc.

 def _OnWMNCHitTest(self, params):
  wParam = params[2]
  lParam = params[3]
  lr = win32gui.DefWindowProc(self._GetAppBarWnd().GetSafeHwnd(),
      WM_NCHITTEST,
      wParam,
#      0,
      lParam
#      self._LowHighWordsToLparam(win32api.GetCursorPos())
     )
  if (lr == HTBOTTOM or lr == HTBOTTOMLEFT or lr == HTBOTTOMRIGHT):
   print "HTBORDER"
   return HTBORDER
  else:
   return lr

--
----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C5E34BA.6030702 at skippinet.com.au...

> The major difference I see if the call to DefWindowProc.  You seem to be
> passing 0 for wParam, and GetCursorPos() for lParam.
>
> Try passing wParam and lParam exactly as passed to you, as per the C++
code.





From fredrik at pythonware.com  Sat Feb  9 18:34:03 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 09 Feb 2002 23:34:03 GMT
Subject: box on soap
References: <mailman.1013283562.647.python-list@python.org>
Message-ID: <Lli98.24849$l93.4538901@newsb.telia.net>

Mark McEahern wrote:
> Don Box on the Importance of Being WSDL

David McCusker on the Importance of Ignoring Don Box:

http://www.treedragon.com/ged/map/ti/newFeb02.htm#06feb02-box

(I'll just give you a summary. I won't argue.  David's right
and Don's wrong ;-)

</F>




From quinn at regurgitate.ugcs.caltech.edu  Fri Feb 22 17:25:59 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Feb 2002 22:25:59 GMT
Subject: Adding static typing to Python
References: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu> <a55qob$ebu$1@thorium.cix.co.uk>
Message-ID: <slrna7dhbm.5et.quinn@regurgitate.ugcs.caltech.edu>

On Fri, 22 Feb 2002 16:14:03 +0000 (UTC), gbreed at cix.compulink.co.uk
<gbreed at cix.compulink.co.uk> wrote:
>Quinn Dunkan wrote:
>
>> >But you can do that in Python:
>> >
>> >>>> def area(shape):
>> >...   if isinstance(shape, Circle):
>> >...     return math.pi*shape.radius**2
>> >...   if isinstance(shape, Rectangle):
>> >...     return shape.width*shape.height
>> >...   if isinstance(shape, Square):
>> >...     return shape.width**2
>> >...   raise TypeError, "unknown shape"
>> >...
>> 
>> This kind of giant typecase is exactly the sort of thing that OO is 
>> supposed
>> replace.
>
>Yes, but then are overloaded functions "OO" either?

Of course, that's what type-based dispatching, aka data driven whatever, is.
Consider the typical scheme OO reinvention, where you have a procedure that
accepts messages and returns functions, so you
(define (message obj) ((obj 'message) args)).  The point of all the trickiness
is that you can now call MESSAGE on any object that responds to 'message.
Effectively, MESSAGE is overloaded to have different definitions depending
on what (obj 'message) returns.

>> >You could make them all subclasses of Shape.  In which case, you have 
>> >functions defined to use a polymorphic object that depend on function 
>> >overloading by the real type.  Not all statically typed languages will 
>> >allow that.
>> 
>> C++ requires special keywords for it, but all statically typed OO 
>> languages I
>> know of do dynamic dispatch.  I certainly don't know every static 
>> language out
>> there, but just because there's a limited implementation somewhere 
>> doesn't
>> mean the concept is unworkable :)
>
>In Java, you need the same kind of "giant typecase" as in the Python 
>example you objected to.

Can't you just say 'shape.area()' and expect the right version of 'method' to
be called?  The typecase occurs automatically in the runtime vtable-or-whatever
lookup.

>                          I think C++ works the same way, unless you use 
>templates.  If I understand "dynamic dispatch" correctly, that's dynamic 
>typing by the back door, so not much of an argument for static typing.

No, dynamic dispatch doesn't have much to do with dynamic typing.  At least
that's my understanding.  Many implementations of static types (where "type"
is only a compile-time concept) make dynamic dispatch difficult, though, since
to do it you have to keep some type info around (hence c++'s vtables).
"Dynamic dispatch" is when you type "f(Shape *o) { o->area() }" and expect
that the actual 'area' function which called will depend on what type 'o'
happens to be at run time.

C++'s notion of an overloaded function leaves out the "at run time" part, so
in my view, it's just syntactic sugar for a different function name (usually
with lots of underscores and the types encoded somehow).

>> Sure you are.  area will reject objects that don't have the proper 
>> methods
>> defined for them.  How it works in haskell is that each class has a 
>> minimum
>> interface that you must implement.  If you don't implement all the 
>> methods
>> of the Shape class, Shape-accepting functions won't accept your objects.
>> This is less flexible than having to implement *only* the methods the
>> function actually uses, as in python, but c++ templates are sorta like 
>> that.
>
>If the types have to be resolved at runtime, you aren't getting static 
>type checking.

Sure.  In the description above, though, they're resolved at compile time
(what I meant by "Shape-accepting functions won't accept your objects" is
"your pogram won't compile").

>                What you describe is called "genericity" right?  I wasn't 

No.  What I describe (maybe not clearly) is called "type checking" :)

>thinking of that, and I did specify "explicit" above to exclude it.  AIUI, 
>Haskell does static, implicit typing and so long as everything's 
>statically compiled that gets you the same convenience as dynamic typing.

Not at all.  Haskell (and the MLs and various other hindley-milner relatives)
infers signatures automatically, but it's still statically typed.  It merely
saves you the trouble of typing up all those signatures (and many people
consider it good practice to type them explicitly anyway).  You're not going to
get a type error at run time (unless you consider a failed pattern match in
function application a type error).

>The proposed language changes for "optional static typing" all specify 
>types explicitly.  A more productive approach may well be "how do we alter 
>the language so that type inference can work?"  That may be why slots got 
>added.

No argument on that one :)  It's a little less intrusive.

>> >Wouldn't it be better to make area a method of the shapes?  And for 
>> static >typing, make it abstract in the Shape class?
>> 
>> In your above example, that's what it is.  You're just spelling 
>> 'circle.area()'
>> as 'area(circle)'.
>
>No, circle.area() and area(circle) mean very different things in Python, 
>so none of your comments on circle.area() really apply to area(circle).

What they mean depends on the definition of area().  len(circle) and
circle.__len__() mean the same thing in python, right?  (post 2.2 at least)

I also thought we were discussing a hypothetical non-python (hence the
"but this is not python" above).  

>That's a matter of taste, but you can get area() to work much the same way 
>Python implements len(), as
>
>def area(shape):
>  return shape.__area__()

That only works if area() has one argument.

>which still uses polymorphism rather than operator overloading.  (Time for 
>a definition of "polymorphism"?)

I'd call them the same.  "operator overloading" is when one function has
different implementations for different types, and "polymorphism" is when
you have a generic function that can work sensibly with different types.

>> I'm not even sure why the smalltalk style 'object followed by method' 
>> notation
>> caught on, especially since 'function followed by args' has always been
>> firmly entrenched.
>
>It happens to match the usual English word order: "the class's method" 
>rather than "the method of the class".  If the field had been dominated by 
>French speakers, perhaps it would have been the other way round.

Yeah, I've heard postfix function application described as "subject-verb" and
prefix application as "verb-subject", which indicates that people are thinking
linguisticly.  But I still think English has at least a strong orientation
around "the area of the circle" or "rotate the circle" or "the sorted list" all
of which are verb-subject.  Another hypothesis is that OO-heads wanted to
emphasize that their programs are conceptually data oriented, rather than
function oriented.  Could be the same reason they invented all that funny
"message" and "responds to" terminology.

I'm sure this has been discussed to death so I'll shut up now.  And it's the
sort of endless semantic debate that never gets anywhere :)


From mlh at vier.idi.ntnu.no  Tue Feb 12 17:34:55 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Tue, 12 Feb 2002 22:34:55 +0000 (UTC)
Subject: Wow.
References: <mailman.1013536760.2921.python-list@python.org>
Message-ID: <slrna6j64f.gku.mlh@vier.idi.ntnu.no>

In article <mailman.1013536760.2921.python-list at python.org>,
James_Althoff at i2.com wrote:
>
>[Meowbot]
>>In a few weeks of skimming those threads i'd consistently
>>wackyparsed
>
>"wackyparse"  -- I like that.  Is it original?

I think I've heard the term before -- but Google only gives 5 hits.
Not in much use, obviously :)

>(Passing on the "Meowbot" handle, though. <wink>)
>
>Jim

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From grante at visi.com  Thu Feb 21 12:00:01 2002
From: grante at visi.com (Grant Edwards)
Date: Thu, 21 Feb 2002 17:00:01 GMT
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de> <Xns91BC7AB03699duncanrcpcouk@127.0.0.1>
Message-ID: <lI9d8.1181$N7.63227@ruti.visi.com>

In article <Xns91BC7AB03699duncanrcpcouk at 127.0.0.1>, Duncan Booth wrote:
> gerson.kurz at t-online.de (Gerson Kurz) wrote in 
> news:3c74c3d6.20466328 at news.isar.de:
> 
>> I found an easier solution:
>> 
>> sort = lambda x:(x.sort(),x)[1]
>> 
> If you are heading that way:
> sort = lambda x:x.sort() or x

Assuming you never run into an "x" that returns a true value
from its sort method...

-- 
Grant Edwards                   grante             Yow!  Look DEEP into the
                                  at               OPENINGS!! Do you see any
                               visi.com            ELVES or EDSELS... or a
                                                   HIGHBALL??...


From cbrown at met.co.nz  Wed Feb  6 15:51:30 2002
From: cbrown at met.co.nz (Colin Brown)
Date: Thu, 7 Feb 2002 09:51:30 +1300
Subject: EDF scheduler
References: <eTJ78.14206$N31.668446@ozemail.com.au>
Message-ID: <FJg88.14875$N31.722467@ozemail.com.au>

Thanks Joshua for your reply. Unfortunately I do not think it handles the
problem I have which is the addition of new entries which should be actioned
before the currently timed event. Because your solution is waiting on a
sleep, an entry that comes in with a very short delay time may not be seen
before the current sleep has timed out.

Thanks Jeff for your reply. "sched" does priority sorting, that is not the
problem. See above.

I have implemented a scheduler on a mainframe that does exactly what I want.
I just did not want to re-invent another wheel in Python. My mainframe
solution had an (EDF "earliest-deadline-first") sorted queue with a single
timer on the head entry. However on every new entry, the head entry deadline
time was compared with the entry deadline time and if the new entry was
earlier than the current head, the current timer is cancelled and the next
head entry (ie the new entry) is put in the timer.

Unless someone can suggest a better way, I have decided that since I will
only be entering groups of entries on minute intervals and the longest
period should not exceed 3 hours then I will live with up to 180 threads,
see example code below (not elegant but it is simple).

Thanks again
Colin Brown
PyNZ

!---------------------------------------------------------------------------
------
'''
Multi-thread EDF scheduler

Notes:
1    use simple thread; it terminates after function completion
2    entries put into sched before it is run are EDF sorted
3    separate groups run in parallel and are thus EDF
4    use Queue for thread-safe result return to mainline
5    in practise new threads will probably be created once per
      minute in my application and not all at once as here but 3
      above applies anyway.
'''
import Queue
import sched
import thread
import time

def thdq(q,line):
    q.put(line)

def thd(q,entries):
    s=sched.scheduler(time.time,time.sleep)
    for item in entries:
        s.enter(item[0],1,thdq,(q,item[1]))
    s.run()

q=Queue.Queue()
for ii in range(20):
    a1='Thread '+str(ii)+' @'+str(37-ii)+' secs'
    a2='Thread '+str(ii)+' @'+str(30-ii)+' secs'
    a3='Thread '+str(ii)+' @'+str(34-ii)+' secs'
    thread.start_new_thread(thd,(q,[(37-ii,a1),(30-ii,a2),(34-ii,a3)]))

time.sleep(0.1)
for ii in range(50):
    time.sleep(1)
    print ii+1
    while q.empty()==0:
        print q.get()
!---------------------------------------------------------------------------
------
Python 2.1 on Win NT4 produces:
1
2
3
4
5
6
7
8
9
10
11
Thread 19 @11 secs
12
Thread 18 @12 secs
13
Thread 17 @13 secs
14
Thread 16 @14 secs
15
Thread 19 @15 secs
Thread 19 @15 secs
16
Thread 18 @16 secs
Thread 14 @16 secs
17
Thread 17 @17 secs
Thread 13 @17 secs
18
Thread 19 @18 secs
Thread 16 @18 secs
Thread 12 @18 secs
19
Thread 18 @19 secs
Thread 15 @19 secs
Thread 11 @19 secs
20
Thread 14 @20 secs
Thread 17 @20 secs
Thread 10 @20 secs
21
Thread 16 @21 secs
Thread 13 @21 secs
Thread 9 @21 secs
22
Thread 12 @22 secs
Thread 15 @22 secs
Thread 8 @22 secs
23
Thread 14 @23 secs
Thread 11 @23 secs
Thread 7 @23 secs
24
Thread 10 @24 secs
Thread 13 @24 secs
Thread 6 @24 secs
25
Thread 12 @25 secs
Thread 9 @25 secs
Thread 5 @25 secs
26
Thread 11 @26 secs
Thread 8 @26 secs
Thread 4 @26 secs
27
Thread 7 @27 secs
Thread 10 @27 secs
Thread 3 @27 secs
28
Thread 2 @28 secs
Thread 9 @28 secs
Thread 6 @28 secs
29
Thread 1 @29 secs
Thread 5 @29 secs
Thread 8 @29 secs
30
Thread 0 @30 secs
Thread 7 @30 secs
Thread 4 @30 secs
31
Thread 3 @31 secs
Thread 6 @31 secs
32
Thread 2 @32 secs
Thread 5 @32 secs
33
Thread 1 @33 secs
Thread 4 @33 secs
34
Thread 0 @34 secs
Thread 3 @34 secs
35
Thread 2 @35 secs
36
Thread 1 @36 secs
37
Thread 0 @37 secs
38
39
40
41
42
43
44
45
46
47
48
49
50
>>>





From niemeyer at conectiva.com  Wed Feb 20 08:58:07 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Wed, 20 Feb 2002 10:58:07 -0300
Subject: Recursively traverse linked list -- help!
In-Reply-To: <1014197390.1203.7.camel@matijek>
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org> <7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek> <mailman.1014126336.32560.python-list@python.org> <7x7kp9blab.fsf@ruckus.b <20020219191050.A3778@ibook.distro.conectiva> <1014197390.1203.7.camel@matijek>
Message-ID: <20020220105807.B1470@ibook.distro.conectiva>

> Well actually it is fair, because the insertion time for the linked list
> is constant, as long as you have a handle to the insertion point. That's

That indeed would be true if you always have a handle to the insertion
point. But 136500% of overhead in the usual case is too much for me.

> pretty much the reason for using it. Inserting at the start will be just
> as fast/slow as inserting in the middle. I just didn't feel like
> rewriting the class so that it would take an arbitrary insertion point.

That's the "point". It's not an arbitrary insertion point at all. It's a
point where you have the handle.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From lac at strakt.com  Thu Feb 14 11:20:12 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 14 Feb 2002 17:20:12 +0100
Subject: Using tabs instead of spaces in IDLE? 
In-Reply-To: Message from Tom Bryan <tbryan@python.net> 
   of "Thu, 14 Feb 2002 13:13:50 GMT." <iKOa8.102499$A51.43433850@typhoon.southeast.rr.com> 
References: <mailman.1013661023.30875.python-list@python.org>  <iKOa8.102499$A51.43433850@typhoon.southeast.rr.com> 
Message-ID: <200202141620.g1EGKCGo002969@ratthing-b246.strakt.com>

> Tom Bryan
> 
> That's what I was afraid of.  I think that Jython could do good things for 
> my group at work, but I really don't look forward to the stupid whitespace 
> discussions.  It's bad enough on c.l.py.  I can't wait to hear it in person.

This is now fixed for c.l.py
http://python.sourceforge.net/peps/pep-0666.html

Good luck with your co-workers.

Laura Creighton



From loewis at informatik.hu-berlin.de  Thu Feb 28 12:59:25 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 18:59:25 +0100
Subject: Simple threading thing
References: <Rlsf8.136$5vh.170729984@news.telia.no>
Message-ID: <j4vgchebcy.fsf@informatik.hu-berlin.de>

"Thomas Weholt" <thomas at gatsoft.no> writes:

> I got a webserver based on BaseHTTPServer using threads. When it starts I
> want to create a seperate process, a thread, that goes into the background
> and checks the server on given intervals, clearing expired sessions etc. I
> cannot get the created thread to pass control back to the server and let it
> continue. It just sleeps the specified interval and then does it thing then
> sleeps again. The server never gets into action at all.
> 
> Any clues?

What operating system? What do you mean by "create a separate process"?

Regards,
Martin


From klaatu at evertek.net  Sat Feb 23 15:22:11 2002
From: klaatu at evertek.net (Mike Dean)
Date: Sat, 23 Feb 2002 14:22:11 -0600
Subject: string substitutions
In-Reply-To: <c4e6b17d.0202231152.5f87765d@posting.google.com>; from bobnotbob@byu.edu on Sat, Feb 23, 2002 at 11:52:10AM -0800
References: <c4e6b17d.0202231152.5f87765d@posting.google.com>
Message-ID: <20020223142211.A9610@evertek.net>

* Bob Roberts <bobnotbob at byu.edu> [2002-23-02 11:52]:
> What would be a good way to replace every one or more spaces (" ") in
> a string with just one space?  Or replace any number of newlines with
> just one?

This is a perfect job for Python's regular expression (RE) library. For
example:

import re

# Replace all occurances of one or more spaces with a single space
re.sub(' +', ' ', mystring)
# Ditto for newlines
re.sub('\n+', '\n', mystring)
# And, to combine the two into one operation:
re.sub('(\n+| +)', '\1', mystring)

If you perform this operation more than once, precompile your RE with
re.compile, like so:

spnl = re.compile('(\n+| +)')
spnl.sub('\n', mystring)

That will speed up the substitution, quite a bit for more complicated
expressions (don't know how much of a factor there is with a simple
expression like this, but there will be some speed improvement).

HTH,
Mike



From dreed at capital.edu  Fri Feb 22 13:32:28 2002
From: dreed at capital.edu (Dave Reed)
Date: Fri, 22 Feb 2002 13:32:28 -0500
Subject: PyGTK problem
In-Reply-To: <a560ch$i8q@r02n01.cac.psu.edu> (message from Rajarshi Guha on
	Fri, 22 Feb 2002 12:50:56 -0500)
References: <a560ch$i8q@r02n01.cac.psu.edu>
Message-ID: <200202221832.g1MIWSm27894@localhost.localdomain>

> From: Rajarshi Guha <rxg218 at psu.edu>
> 
> Hi,
>   I'm not sure whether this is the right place to ask this....
> I have a PyGTK app built using libglade which basically consists of a 
> window containiong a text entry box.
> 
> I would like to take an action when the Return key is pressed. I made a 
> signal handler for the 'changed' even in GtKEditable. However the callback 
> is triggered whenever any text is entered into the box. Which is OK - but 
> the callback does'nt get called when a Return is entered.
> 
> I have got around it by adding a button - but it seems kludgy compared to 
> just pressing the Return key.
> 
> Does anybody have any suggestions?

Use the "activate" signal, not the "changed" signal.

Dave



From doug at hellfly.net  Tue Feb  5 07:53:40 2002
From: doug at hellfly.net (Doug Hellmann)
Date: Tue, 5 Feb 2002 07:53:40 -0500
Subject: ANN: HappyDoc 2.0.1
Message-ID: <200202051253.HAA07845@branagan.hellfly.net>

Announcing the latest version of HappyDoc, a Python documentation extraction 
tool.

  HappyDoc is a tool for extracting documentation from Python source
  code.  It differs from other such applications by the fact that it
  uses the parse tree for a module to derive the information used in
  its output, rather that importing the module directly.  This allows
  the user to generate documentation for modules which need special
  context to be imported.

  More details are available on the HappyDoc home page at
  http://happydoc.sourceforge.net.

  !!! NOTICE !!!

    Before installing HappyDoc 2.0+, any version earlier than 2.0 must
    be removed.  This includes the file 'happydoc.pth' and directory
    'happydoc' in site-packages, as well as the executable 'happydoc'
    in the 'bin' directory.

  Version 2.0.1 -- 

    The primary reason for this release is to resolve several bugs
    which prevented Win32 support from working properly.  Support for
    Win32 operating systems should now be restored.

    - **New Features**

      - Added limited support for CGI programs by expanding the docset
        processing to look for files ending in '.cgi' as well as
        '.py'.

    - **Bug Fixes**

      - Resolved defect #505456 and 498204, a problem with
        'happydocwin.py' that prevented it from working properly on
        Win32 systems.

      - Resolved defect #501240 so that path handling under Win32
        systems works properly and HappyDoc can generate output.  

      - Resolved defect #505188 so that using the '-o' option will
        properly send output to stdout.

    - **Other Changes**

      - Added more unit tests for formatters, especially
        'fileformatterbase.py' and 'formatter_HTMLFile.py'.

      - Updated the StructuredTextNG docstring converter so that if an
        exception is generated during parsing of the ST, we fall back
        to ClassicStructuredText.  The two syntaxes appear to be
        incompatible, and this should protect backwards compatibility
        for most users.


Download

  Download the latest version of HappyDoc from the home page on
  SourceForge:

     http://sourceforge.net/projects/happydoc




From threeseas at earthlink.net  Sun Feb 24 20:45:53 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 25 Feb 2002 01:45:53 GMT
Subject: Busy with better things to do??? >Re: Guido wins FSF Award
References: <mailman.1014220773.23411.python-list@python.org> <3c73e173.10676151@127.0.0.1>
Message-ID: <1263.820T1799T12905345threeseas@earthlink.net>

On 20-Feb-02 12:47:01 (Five Fresh) Fish <root at 127.0.0.1> wrote:
>On Wed, 20 Feb 2002 15:58:47 -0000, Simon Brunning
><SBrunning at trisystems.co.uk> wrote:

>>> From:	grante at visi.com [SMTP:grante at visi.com]
>>> > what are revealing the Nine Pythonic Actions underlying all that we do.
>>> > can you explain what that means ??????
>>>
>>> Bhuddism allusion?
>>
>>Autocoding allusion, I think.
>
>Speaking of which, the culprit seems to have vanished.
>
>Hope he's okay.  Even net loons shouldn't come to harm.  (Preferably, they
>should be medicated, get rid of the delusions, help them lead constructive
>lives...)



The reasons of my not being here more can't possibly be:

My work load is non-computer non-programming.

You are not being productive or helpful to the project.

<rolls eyes>

http://groups.google.com/groups?hl=en&selm=5098.813T2333T10665859threeseas%40earthlink.net&rnum=10

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From ehoute at zeelandnet_nl  Thu Feb 14 12:26:10 2002
From: ehoute at zeelandnet_nl (Ewald van Houte)
Date: 14 Feb 2002 17:26:10 GMT
Subject: Color database in Tkinter
References: <veqn6ucitupfrt3jaujtd7rhrq0sl39r2g@4ax.com>
Message-ID: <Xns91B5BB561C7DCehoutezeelandnetnl@212.115.192.220>

Lawrence Oluyede <rhymes at myself.com> wrote in
news:veqn6ucitupfrt3jaujtd7rhrq0sl39r2g at 4ax.com: 

> 
> Where i could find the whole database for color names
> built in Tkinter? I need a list of colors
> 

http://tcl.activestate.com/man/tcl8.4/TkCmd/colors.htm
these are for Tk but i think Tkinter knows them too.


From djmitchell at optushome.com.au  Mon Feb 11 03:53:23 2002
From: djmitchell at optushome.com.au (djmitchell)
Date: Mon, 11 Feb 2002 19:53:23 +1100
Subject: Zope cookbook?
Message-ID: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au>

Is anyone working on a Zope cookbook, or know of one that already exists?

I'm pretty comfortable with Python (+ Perl, VB, C++, Java/J2EE, ...) and 
building Web apps in general, but making the leap to Zope seems pretty 
tough due to an apparent lack of "here's how to do simple, common task X" 
type of documentation.  I've got an app where a Zope solution looks like a 
perfect fit, but churning out code feels like pulling teeth - I'm agonizing 
over every line, and convincing myself there's easier/simpler/faster ways 
to do just about everything.  I'm also finding myself reading through other 
peoples' source code almost constantly - usually enlightening, but a REALLY 
slow way of learning.

I've give a specific example of one thing I need to do: I want to build 
something analogous to an ActiveX control or a JavaBean, that can be used 
to input a date.  It needs to check for valid data being input (e.g. no 
"31-February-2002"); I might want it to be displayed using some cute DHTML 
if the browser is IE6, or maybe something dumber/simpler like pull-down 
text boxes if the browser is Netscape 2; it might also need to only accept 
dates in a certain range (e.g. between "12-January-2002" and 
"15-January-2002") that I specify; it might need to display a default date; 
based on the user's geographic location, I might want entered dates to be 
displayed on screen as "15-January-2002", "15-1-2002", "2002-1-15", or 
"1/15/2002".  It needs to be a logical "function" that I can call from 
within an HTML form, and assign the (validated) date entered by the user to 
a variable of some sort.  Ignoring the distinct possibility that I might be 
able to download exactly this "ZopeBean" from somewhere, I'm sure there's a 
"normal" way of coding this within Zope (DTML method?  Python script?), but 
I just don't know where I should be looking.  Oh, and I don't want to 
resort to either ActiveX or Java to get the job done ;->

Any tips, pointers, advice, useful sources of info, etc.?

Thanks in advance

Dave Mitchell


From john_coppola_r_s at yahoo.com  Sat Feb 16 12:38:02 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Sat, 16 Feb 2002 09:38:02 -0800 (PST)
Subject: Why Python is like BASIC ...  java "vs" python
In-Reply-To: <a4km09$6d9$1@peabody.colorado.edu>
Message-ID: <20020216173802.56774.qmail@web11801.mail.yahoo.com>

[Fernando Perez wrote ]
> (And by the way, a whale is a mammal, not a fish).

That was my whole point!  It's dumb to compare them.

-john



__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From deathtospam43423 at altavista.com  Sat Feb 23 14:23:44 2002
From: deathtospam43423 at altavista.com (A.Newby)
Date: 23 Feb 2002 19:23:44 GMT
Subject: Puzzled: Y am I ending up with extra bytes?
Message-ID: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4>

Why is this happening? I read large chunks of data from a text file, 
according to byte locations specified on another file, and for some reason, 
this function (below), spits out a few extra bytes.

Here's the code, as entered into the Python shell...... 


    	index = map(string.rstrip, open('D:\cgi-bin\indx.txt').readlines())
    	#this opens the index file, which has precise byte locations of each 
    	#chunk of data I want to extract from the log.txt file, and turns it 
    	#into a list.


    	def fish(end, start, deduct):
	    	chat = open('D:\cgi-bin\log.txt', 'r')
	    	g = int(index[end]) - int(index[start])
	    	chat.seek(int(index[start]))
	    	print chat.read(g - deduct)



Now, if I enter the following into the command line ...

fish(205, 204, 0)

,,, I get about four extra characters. God knows why! So that's why I try 
...

fish(205, 204, 4) 

... And it seems to work perfectly. I can even "fish" up to about about ten 
"index" lines with it. But as soon as I try and fish out any more than 
that, I get the dreaded extra 4 bytes of code again. Why?

Now I know what you're thinking. You're thinking that perhaps my index file 
is corrupted, and hasn't got an accurate account of precisely what's in the 
log file. I suspected that might be the case myself, but ... when I try 
fishing out each chunk of data individually, it works fine! I can even do 
this ...

for x in range(1, 90):
	fish(211+x, 210 + x, 4)

...... without ending up with that extra data I don't want. However, this 
method is too slow for my purposes. Plus, I really wanna know what it is 
that's going wrong.

Can anyone spot it?

BTW, if you're curious, this is part of a chat script I'm putting together. 


cdewin at dingoblue.net.au


From usenet at thinkspot.net  Sun Feb 24 01:14:02 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 22:14:02 -0800
Subject: [Tutor] unrelated bug issue
References: <3C78529D.F4E7D754@mail.verizon.net> <mailman.1014530559.26764.python-list@python.org>
Message-ID: <a5947b.3vsa2bl.1@kserver.org>

On Sun, 24 Feb 2002 01:00:59 -0500, Tim Peters <tim.one at comcast.net>
wrote in comp.lang.python in article
<mailman.1014530559.26764.python-list at python.org>:

> I got a better one:  suppose you shoved a banana down your throat, then a
> carrot in each nostril.  Would it then be the banana's or the carrots' fault
> that you couldn't breathe anymore?  That's right:  it would be the carrots',
> because you didn't have any trouble breathing before they wound up in your
> nose.  Or maybe it was the banana's fault after all, since if it weren't
> there, the carrots' presence wouldn't matter.  It's a puzzler.  Just so long
> as it's clear that it's never ever your fault <wink>.

LMAOAPIMP


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


From sholden at holdenweb.com  Wed Feb 13 07:29:24 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 07:29:24 -0500
Subject: Having a mare with Classes
References: <3C6A4BC4.4060409@unite.com.au>
Message-ID: <w_sa8.205934$KM2.8425463@atlpnn01.usenetserver.com>

"Graeme Matthew" <graeme.matthew at unite.com.au> wrote in ...
> Please can someone help, im having an absolute mare, I have Wesley Chuns
> book on core python programming and I cannot get casses to work, im just
> using a simple test from the examples in the book.
>
> OK I HAVE A CLASS PERSON
>
> class Person:
>
>      def printName(self):
>          print 'You printed it'
>
> I HAVE A SCRIPT:
>
> #!/usr/local/bin/Python2.2
>
>
> import sys
>
> sys.path.append('/www/cgi-bin')
>
> import Person
> person = Person() #### THIS IS THE PROBLEM
>
> person.printName()
>
> I keep on getting this error:
>
> #! /usr/local/bin/Python2.2
>
>
> TypeError: call or non function (type module)
>
Because you define the Person class in a Person module, which you then
import into your main script, you must refer to it as Person.Person (module
name qualified by class name: each module is a Python namespace). So the
error message is arising because you are trying to call a module (Person).

Replace the line I marked with

    person = Person.Person()

and all should be (relatively) well. Now go, my son, and sin no more :-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jimd at vega.starshine.org  Sun Feb 17 15:36:02 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 17 Feb 2002 20:36:02 GMT
Subject: lambda to Sort Dictionary.items() by Value
Message-ID: <a4p47i$13ed$1@news.idiom.com>

 I'm playing with text and I decided to write a simple histogram
 function (calculating letter occurence frequencies, in this case).

 In the course of that I found two simple uses for lambda one of 
 which seems like it would be a very common idiom in Python.

 Here's the function:

def histogram(s):
    """ Calculate the histogram for letters in a sequence
        returns histogram dictionary and a sequence sorted by  
        decreasing frequency (rarest last) (case insensitive)"""
    letters = string.letters
    h = {}
    for i in s:
        if i in letters:
            i = i.lower()
            if h.has_key(i): h[i] += 1
            else: h[i] = 1
    i = h.items()
    i.sort(lambda (k1,v1),(k2,v2): cmp(v2,v1) ) # sort by value, not by key
    i=string.join(map(lambda (x,y):x, i),"")  # extract letters from each
    return (h,i)

 The useful idiom seems to be in the comparison function for the sort().  
 When sorting the items of a dictionary by value (rather than by key) it 
 seems like using the lambda (k1,v1),(k2,v2): v2 - v1 to decompose
 the two tuples and return a comparison is the most concise way to 
 express it.

 It seems like one might need to sort dictionary items by value
 pretty often.

 The other use of lambda here is less interesting; I'm just extracting
 items from tuples.

 I guess these are slightly more readable as:

def cmpvalr((a,b),(c,d)):
	"""Compare two tuples by their 2nd (value) elements: reverse"""
	return cmp(d,b)

 ...

items = h.items()
items.sort(cmpval)

 Incidentally, in occurs to me that using cmp() allows this
 idiom to use any comparable types --- so it would would with
 any dictionary so long as the values were comparable.  
 Initially I'd written it as v2 - v1 --- which works for numbers
 but not for other types.
  


From rasmussn at lanl.gov  Tue Feb 26 13:03:33 2002
From: rasmussn at lanl.gov (rasmussn at lanl.gov)
Date: Tue, 26 Feb 2002 11:03:33 -0700
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <6bf54244.0202260920.5920efc6@posting.google.com>
Message-ID: <25BFA948-2AE3-11D6-A367-0030657EBC3E@lanl.gov>

On Tuesday, February 26, 2002, at 10:20 AM, Craig Maloney wrote:

> Hi David.
>
> Since I'm not a "computer guy" per se, I was wondering if you could
> clear up some more of my confusion.  I don't know much about how
> compilers work, but I would presume that any compiler (eg g++) would
> create some internal representation, similar to the .cil file, after
> parsing but before compiling.

Parsing is just one phase of the entire process of compiling.  From Aho, 
Sethi,
and Ullman, the phases of a compiler are: lexical analysis, syntax 
analysis,
semantic analysis, intermediate code generation, code optimization, and
code generation.

> Presumably the EDG front end is geared toward producing this information
> in a format more suited to being used by the "wrapper generators" in 
> contrast
> with a compiler which produces this information only for itself and with
> the sole goal of compiling an object file?

I can't really speak for the EDG people, but I assume that the .cil files
(or equivalent information, stored in memory) are used as part of the
compilation process, although, it may not always be written to disk.

> What I'm getting at is that I'm confused as to why people wouldn't have
> used information from the compiler *itself* to automate the process of
> wrapper generation.  Obviously the compiler writer people would have to
> collaborate with the wrapper generator people... but there must be 
> technical
> obstacles?

We are using the compiler, but only the first three stages.

Craig




From syring at email.com  Fri Feb  8 14:31:06 2002
From: syring at email.com (Karl M. Syring)
Date: Fri, 8 Feb 2002 20:31:06 +0100
Subject: SourceForge Python foundry ????
References: <3c6301ac.188163695@news.laplaza.org> <a40hh0$1bdtaj$1@ID-7529.news.dfncis.de> <65c17385.0202081114.14d2a605@posting.google.com>
Message-ID: <a41971$1b8tsa$1@ID-7529.news.dfncis.de>

"Uriah Welcome" <precision at devrandom.net> schrieb im Newsbeitrag
news:65c17385.0202081114.14d2a605 at posting.google.com...
> "Karl M. Syring" <syring at email.com> wrote in message
news:<a40hh0$1bdtaj$1 at ID-7529.news.dfncis.de>...
> > "Mats Wichmann" <mats at laplaza.org> schrieb
> > > What's up with the Python foundry on SourceForge? It's not listed in
> > > any of their lists of foundries, although if you plug in "python" in
> > > the standard path they use for a foundry it comes up fine...
> >
> > Perhaps, that is because SF still has a grubby v1.5.2 installed?
>
> As soon as debian 'stable' includes a newer version we'll be happy to
include it.

Google stalking at work?
Nonetheless, this is a good message, at least we wont get a prompt saying
something like "AOL-Linux -- you have got snails!"

Karl M. Syring




From sjmachin at lexicon.net  Sun Feb 24 01:28:50 2002
From: sjmachin at lexicon.net (John Machin)
Date: 23 Feb 2002 22:28:50 -0800
Subject: Correction: Re: string substitutions
References: <mailman.1014504639.21135.python-list@python.org>
Message-ID: <c76ff6fc.0202232228.201f829e@posting.google.com>

Mike Dean <klaatu at evertek.net> wrote in message news:<mailman.1014504639.21135.python-list at python.org>...
> * Mike Dean <klaatu at evertek.net> [2002-23-02 14:22]:
> > import re
> > 
> > # And, to combine the two into one operation:
> > re.sub('(\n+| +)', '\1', mystring)
> 
> This last line doesn't work as intended - sorry! (didn't proofread it
> carefully enough :-()
> It won't convert those chars to single characters
> (in fact, it does absolutely nothing except consume processor time)...

Try testing instead of proofreading. Post actual examples of input and
output, like this:

>>> mystring = "aaaa\n\n\n\nbbbbbbb\n\n\nbb\nbbbb"
>>> re.sub('(\n+| +)', '\1', mystring)
'aaaa\x01bbbbbbb\x01bb\x01bbbb'
>>> re.sub('(\n+| +)', r'\1', mystring)
'aaaa\n\n\n\nbbbbbbb\n\n\nbb\nbbbb'
>>>

Your posted code replaces matched substrings with "\x01".
What you *meant* to post does nothing but chew up CPU time.


From sjmachin at lexicon.net  Tue Feb 26 17:09:50 2002
From: sjmachin at lexicon.net (John Machin)
Date: 26 Feb 2002 14:09:50 -0800
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com>
Message-ID: <c76ff6fc.0202261409.36d2d7e1@posting.google.com>

amrit040 at yahoo.com (Amrit) wrote in message news:<71c0a06e.0202260513.4b67841f at posting.google.com>...
>   /*  Make the output tuple */
>   retvec=(PyObject *) malloc(2*sizeof(PyObject));
>   retvec=Py_BuildValue("(ff)",min,max);
>   return(retvec);

An afterthought:

You should lose the malloc() -- not only pointless (some compilers
would even give you a warning like 'retvec is assigned a value which
is not used') but also leaks memory.


From nils at paragon.no  Tue Feb 12 20:54:52 2002
From: nils at paragon.no (Nils R Grotnes)
Date: Wed, 13 Feb 2002 01:54:52 +0000
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de> <mailman.1013521581.26252.python-list@python.org>
Message-ID: <a4cqiv$1d91p7$1@ID-79093.news.dfncis.de>

Chris

> Ur.  What's the Win32 equivalent of ldd (dumps linked dll's)?  I ran strings
> on it and filtered out the .dll files (case-insensitive match) and came up
> with this:

<http://www.dependencywalker.com>

-- 
Nils R Grotnes
_______________________________________________________
Support everyones right to speak up about crap!
PC-Help is under legal attack from Lockdown 2000
More information at <http://www.pchelpers.org>


From netzapper at magicstar.net  Wed Feb 27 22:18:01 2002
From: netzapper at magicstar.net (A. Jones)
Date: Thu, 28 Feb 2002 03:18:01 GMT
Subject: Peer review: Python encyphering script
Message-ID: <3c7c1a91.150875561@menuhin.netfront.net>

It's my second day Pythoning and I've written this code to encypher
text (or files, ostensibly) into two pieces, neither of which can be
used to determine the message.

Any problems pointed out, clunky lines spotted, or security problems
would be most appreciated.  Also, if you've seen something like this,
if you'd tell me, please.

For now, no interface, just C&P between sessions of IDLE.

ASCII converts to numbers because of future security measures (such as
inserting chunks of random data to pad small messages, which would
then automatically delete themselves in the process of decrypting
using multiplication by 0)

Source below.

---

#encyphering tool 
plaintext = "Isn't it awfully nice to have a penis; isn't it
frightfully good to have it on!" #The variable makes it obvious.
blarg = [] #To hold translation of ASCII to num
for b in plaintext:
#    print b #Was here for development purposes.
    blarg.append(ord(b)) #Turn it into a list of ASCII numbers
                         #nums make sorting easier&append/ord makes a
nice list
plaintext = "Text purged."

blurb = {} #To hold 1:1 value/offset references
b = 0 #reuse variable so's to destroy evidence/make my life easier
for z in blarg:
    if z in blurb: #duplicate key fixer
            if type(blurb[z]) == "<type 'list'>": #Stops nesting
strings
                blurb[z].append(b)
            else:
                temp = blurb[z] #Used to create first list entry
                temp.append(b)
                blurb[z] = temp
    else: blurb[z] = [b] #sets dictionary with values as keys, offsets
as objects
    b = b+1 #Next step in the offset counter... clunky.

valuelist = blurb.keys() #extract keys of dictionary
valuelist.sort() #to be randomized later(date)

finval = [] #to hold final value list
finoff = [] #to hold final offset list
for r in valuelist:
    z = 0
    while z < len(blurb[r]): #Writes number of values equal to number
of offsets
        finval.append(r)
        z = z+1
    for q in blurb[r]: #Decomposes my clunky nesting lists/dicts
        finoff.append(q)

blurb = {}  #Destroy the evidence!
valuelist = []

--
#decyphering tool
text = [] #finval
key = [] #finoff
numdict = {}
z = 0
for r in key: #This here is the heart of the decryption process.
Simply coalates the lists.
    numdict[r] = text[z]
    z = z + 1
readlist = numdict.keys()
readlist.sort() #puts the keys/offsets in order
plaintext = ""
for r in readlist:
    plaintext = plaintext + chr(numdict[r])
print plaintext #prints plaintext
---

Aubrey Jones
______________________________________________________________________________
Posted Via Binaries.net = SPEED+RETENTION+COMPLETION = http://www.binaries.net


From fperez528 at yahoo.com  Tue Feb 26 09:11:15 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 26 Feb 2002 14:11:15 +0000
Subject: Coredumps with Python 2.1.1-3
References: <20020226.185425.1884661237.5746@ic.ac.uk>
Message-ID: <a5gtp4$sin$1@peabody.colorado.edu>

Phil Mayers wrote:

> WTF? Even weirder, if I remove the line of input on which the crash
> happens, it will get a lot further, but die somewhere else... Very
> strange...
> 

That's a tell-tale sign of a corrupted pointer. Most likely a C extension 
module you're using messed up the memory earlier. Now your code has become a 
time bomb, and it will go off at seemingly random times.

Python code can't do that, but ultimately Python sits on top of C code, so 
it's not 100% bulletproof.

Good luck,

f.


From s.keim at laposte.net  Tue Feb 12 04:15:57 2002
From: s.keim at laposte.net (sebastien)
Date: 12 Feb 2002 01:15:57 -0800
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com>
Message-ID: <7da8d8f8.0202120115.c52526e@posting.google.com>

joost_jacob at hotmail.com (J.Jacob) wrote in message news:<13285ea2.0202080524.e2e7948 at posting.google.com>...
> We would like a -portable- module that provides 4 functions:
> setupTextUI(width, height)  # returning a text UI object, top left is
> (0,0)
> writeXY(textWindow, x, y, string)   # print string on position (x,y)
> readXY(textWindow, x, y)    # waiting for user input with cursor
> positioned on
>                          # (x,y) and returning a string from user
> clearWindow(textWindow)
(...)
> What we need is something in the Python distribution -itself- that is
> portable.

I would have a few more request for this kind of module:
readXY should have an optional argument for the maximal length of the
input

I'd really like to have a portable function getkey() able to return
the value of the key pressed or "" if no key have been pressed.


From rdsteph at earthlink.net  Sun Feb 10 06:16:19 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 10 Feb 2002 11:16:19 GMT
Subject: Linux and Python scripts, help needed
Message-ID: <3C665762.550F7605@earthlink.net>

I am back from a trip, I have manually updated the mygale generated
Pythonic web articels page on my web site. There are quite a few new
links of interest.

However, the automated updating from my Linux box still did not work
while I was gone. I will manually update once per day but I would sure
like to get the automaton working.

I am embarrassed to come back once more to ask for help on this subject.
I have used crontab to set up 
a my_cron_file that runs five Python scripts to automatically run
mygale.py and associated programs to 
update the output.html file listing new web articles about Python once
per day.

My crontab file reads as suchas checked using crontab -l:

30 4 * * * /usr/bin/python /home/ron/mygale.py get all
30 5 * * * /usr/bin/python /home/ron/urls.py
30 6 * * * /usr/bin/python /home/ron/urldb.py
00 7 * * * /usr/bin/python /home/ron/mygale.py -n 200 display all
30 7 * * * /usr/bin/python /home/ron/mygale-uploader.py

crond is running , as checked using
  /etc/rc.d/init.d/crond  status which shows a running cron process with
a pid number.

Also,  ps aux | grep cron shows a /usr/sbin/cron process.

Finally, the  cron daemons log check /var/log/cron shows all 5 Python
scripts properly called at their 
appointed times each morning. No errors are at all are recorded in the
cron error log files.

Now, when I run these scripts exactly as typed above, from the Bash
shell,
everything works as planned and that's how I have been updating the
online HTML page that shows the Pythonic articles on my web site.

The mystery is that, even though the log files show the Python scripts
called each day, no update occurs 
on my web site . Indeed, when I check the output.html file in my
/home/ron directory, it is not updated 
at all by these cron processes. Even though the log file shows the
scripts as being ran, it is as if they did not run at all, in that no
output occurs.

Yet when I run the same scripts manually, all is updated.

So this morning at 4:30 AM I sat up and ran a "top" process from the
Bash shell, that shows all 
running processes. Sure enough, at exactly 4:30 AM, a new Python process
was spawned. The new 
Python process was matched with a new entry in the cron log file of the
mygale.py program being 
called.

I watched carefully, and the mygale.py process lasted for about 30
seconds and then disappeared from 
the "top" screen. I assume it terminated normally. Except that when I
run the mygale.py get all script 
manually  in the BASH shell, it takes about two minutes to complete.
This is the only sign of any 
problem, but still no error messages are generated anywhere I can find.

Well, it seems to me that a series of processes on Linux or any
operating system should be 
deterministic. That is, things should make sense, when the exact same
steps are taken, the result should 
always be the same.

This procedure I am trying to do does not seem to make sense to me, it
seems mysterious. Linux should 
not be mysterious.

Can anyone help me?

By the way, shouldn't it be possible to cut and paste from the BASH
shell to a text editor? It would save 
me a lot of typing ;-))))

The clipboard in general is a frustration; I can't yet make it work
properly in many situations. Is it me 
or does the KDE clipboard lack a lot of functionality compared to
Windows clipboard? I can't even 
copy and paste from K-EDIT, nor K-WORD, to Netscape Composer. Oh, well.

One last frustration: Linux does hang, at least to the point that this
newbie can't recover from it. My 
box hung when I tried to open a web presentation by Guido van Rossum
about Python Conference 10 
using K Presenter.  I could not even open a console nor close the
offending KDE shell to get back to a 
command line. The box also hangs when I try to print using cups on my HP
T45 printer. (I wish I could 
use my printer.) My downloaded Open Office also hangs my computer just
by trying to open it.

Nonetheless, these hang-ups and crashes are not of pivotal importance ot
me. But mysterious behavior 
that not does not allow me to run daily scripts to update my web site
does bother me. Computers 
should not be mysterious. They should be deterministic. Do A, then B,
then C, get X for output.

Help ;-))))


From ajeru at vknn.org  Wed Feb 20 16:10:04 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 20 Feb 2002 13:10:04 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <slrna74ssf.21s.QnickQm@localhost.localdomain> <24c39b2c.0202191658.3fd74528@posting.google.com> <slrna76a8d.4kl.QnickQm@localhost.localdomain>
Message-ID: <24c39b2c.0202201310.26267bb8@posting.google.com>

> > You argue that there are cases where static typing gets in your way,
> > where it makes programs less flexible.
> 
> Well, not quite.  I argue that assigning types to even trivial Python
> programs is a delicate and tricky business, and that the static types
> implicit in a typical Python function are tremendously complex.
> 
> This _isn't_ an argument against optional static types for Python: I
> believe in them too.  This is just an explanation of why they haven't
> been done _yet_:
>     1) Typical type declarations for Python will be more complex than
>         those of most other popular languages.
>     2) Therefore, type inference is almost certainly necessary.
>     3) Because of 1 and 2, optional static typing for Python will 
>         involve a lot of work. :)
> 
> Of course, this isn't to say that it won't or shouldn't get done: just
> that the cost portion of the cost/benefit here is very high---and the
> free time among the people who know enough about Python in particular
> and type systems in general is quite low.
> 
> Of-course-I'd-love-to-be-proven-wrong-ly yrs,

I'm afraid you won't be proven wrong very soon. The types-SIG doesn't
seem to be very alive. The papers are rather old and the mailing list
archives don't contain much for the recent months.

Alexander


From ballabio at mac.com  Mon Feb 25 04:43:57 2002
From: ballabio at mac.com (Luigi Ballabio)
Date: Mon, 25 Feb 2002 09:43:57 +0000
Subject: functional programming with map()
In-Reply-To: <0qej7u8uoepjg377elqo1msgklfc4t5eed@4ax.com>
References: <918bc22f.0202241902.2160b75a@posting.google.com>
 <a5cb54$27c2$1@agate.berkeley.edu>
 <eppstein-69765A.19343424022002@news.service.uci.edu>
Message-ID: <5.1.0.14.0.20020225092426.00a5ee70@mail.mac.com>

At 04:12 AM 2/25/02 +0000, Courageous wrote:

> >[x.f() for x in items]
>
>Yeah, whoever it was that decided this was the syntax
>for a list comprehension was just a plain genius. It's
>one of those linguistic thingies where you take one
>look at it, and you say, "of course".

It's from Haskell, I think, which in turn took it from plain old mathematics.
Haskell syntax is : [ f(x,y) | x <- items, y <- moreItems, x+y < 5  ]
Math notation is the same without brackets and with the "belongs to" symbol 
(the one you get with $\in$ in TeX). Well, you might have : instead of |, too.

Of course Guido might have travelled back in time and suggested it so that 
it was there for Python to borrow afterwards :)

Bye,
         Luigi




From westernsam at hotmail.com  Fri Feb  1 11:14:25 2002
From: westernsam at hotmail.com (sam)
Date: 1 Feb 2002 08:14:25 -0800
Subject: While loop error
References: <32c382ad.0201312208.166d5de3@posting.google.com> <slrna5kd7c.bd0.QnickQm@localhost.localdomain>
Message-ID: <292c8da4.0202010814.ab1c798@posting.google.com>

len() gives you the length of a list (or string etc)
the following list has a length of three

list = [1,2,3]
#therefore 
len(list) == 3

you reference list starting at zero
therefore

list[0] == 1
list[1] == 2
list[2] == 3
list[3] ==

index out of range exception ...

to make you code work make list_length = len(list) -1

this is the same with every languages implemetation of list or array
that i have ever seen. You just gotta get used to it.

sam


From thomas.heller at ion-tof.com  Fri Feb  1 05:03:37 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Feb 2002 11:03:37 +0100
Subject: ANN: py2exe version 0.3.0 - does now windows NT services
Message-ID: <a3dp5u$179scg$1@ID-59885.news.dfncis.de>

py2exe is a distutils extension to convert python scripts
into windows exe-files (plus maybe a few dlls), able to run
on computers without requiring a python installation.


New in release 0.3.0:

  The first release which can create windows NT services:
  A sample service with detailed instructions is provided.
  Builds version resources into the executables.
  Other small bug fixes as well.


Download from the usual location:
http://starship.python.net/crew/theller/py2exe/


Thomas





From emile at fenx.com  Mon Feb 25 17:51:46 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 14:51:46 -0800
Subject: 15 minute presentation on python?
References: <3c7ab974_4@corp.newsgroups.com>
Message-ID: <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de>

"Joshua Muskovitz" <joshm at taconic.net> wrote in message
news:3c7ab974_4 at corp.newsgroups.com...
> Hey all,
>
> If you had to give a 15 minute introduction to Python, what would you
cover?
> What is a good example of a simple, but useful, text-based app for a
Wintel
> environment, that has good "wow" quality?
>

Hmm... text-based... "wow" appeal... oxymoron?  ;-)

Otherwise, what Sheila said.

--

Emile van Sebille
emile at fenx.com




From usenet at thinkspot.net  Wed Feb 20 20:53:00 2002
From: usenet at thinkspot.net (Sheila King)
Date: Wed, 20 Feb 2002 17:53:00 -0800
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org>
Message-ID: <a50npt.3vscd1j.1@kserver.org>

On Wed, 20 Feb 2002 19:08:59 -0500, William Park <opengeometry at yahoo.ca>
wrote in comp.lang.python in article
<mailman.1014250235.21771.python-list at python.org>:

> With the money you pay to ISP, why don't you get DSL with static IP, and
> run your own server?

Well, for one thing, he may not have a machine to dedicate to using as a
server, that is as reliable as a good hosting service. He may not always
be around to ensure that his server is up. Also, a good web host will
provide support and help you with things you aren't sure of.

On 20 Feb 2002 15:51:24 -0800, kdahlhaus at yahoo.com (Kevin Dahlhausen)
wrote in comp.lang.python in article
<283adf56.0202201551.7026d548 at posting.google.com>:

> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

Well, I'm not sure what you have in mind by "inexpensive", but I highly
recommend FutureQuest.net. They have recently installed Python2.2. Bruce
Guenter (well known author of Qmail add-ons) is on staff and supports
the Python installation. http://em.ca/~bruceg/ 
The FutureQuest.net community is fabulous. http://aota.net/forums/

Here is a link to their homepage:
<a href="http://www.qksrv.net/click-533635-805701">FutureQuest.net</a>

I have a basic plan, which is $20/month on a month-to-month basis,
$15/month if paid yearly. I write and run not only Python cgi-scripts,
but also python utilities and python mail scripts on my site.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From store_li at sina.com  Thu Feb  7 03:38:27 2002
From: store_li at sina.com (Ricky)
Date: Thu, 7 Feb 2002 16:38:27 +0800
Subject: Questions  about pyscript
Message-ID: <a3tegq$1ahish$1@ID-12869.news.dfncis.de>

I make a try a pyscript. But there is one thin confuse me. When I use print
statement intent to send something to browse. It doesn't work, while
document.write can make it. What make me confused is what part is controlled
under python(such as flow control, functions, package etc) and which part is
controlled by DHTML(output?).

Could some one give me an over view option of what pyscript is and the
object tree map of pyscript is a plus.

Thanks!




From store_li at sina.com  Mon Feb 25 22:38:07 2002
From: store_li at sina.com (Ricky)
Date: Tue, 26 Feb 2002 11:38:07 +0800
Subject: MSWord and Python
References: <3C7A4EF8.4010404@thomas-guettler.de>
Message-ID: <a5f00p$6udv4$1@ID-12869.news.dfncis.de>

The client side document is include in Python online documents. While the
COM method and properties can only be found in Microsoft's documents. VB's
object browser is a good source. PythonWin also come with a Object Browser,
but I don't think it work fine enough now.

"Thomas Guettler" <st-newsgroups at thomas-guettler.de> wrote in message
news:3C7A4EF8.4010404 at thomas-guettler.de...
> Hi!
>
> I am new to Python and COM.
>
> 1. Where can I find a introduction on the net?
>
> 2. Where can I find the documentation of methods that MSWord
>   exposes by the use of COM?
>
> 3. I want take a word template. Replace text in it, and save it again.
>   Has someone code examples fot something like this?
>
>   Thank you, thomas
>




From sdm7g at Virginia.EDU  Wed Feb 27 16:25:07 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Wed, 27 Feb 2002 16:25:07 -0500 (EST)
Subject: full closures
In-Reply-To: <%rbf8.71116$X64.25961788@news1.rdc2.pa.home.com>
Message-ID: <Pine.OSX.4.43.0202271604120.4269-100000@d-128-61-180.bootp.virginia.edu>



> "Michael P. Soulier" <msoulier at nortelnetworks.com_.nospam> wrote in
> message news:slrna7q9m0.4pb.msoulier at pmerd071.ca.nortel.com...
> >     While I know about bound methods in Python, I'm wondering if a
> > full closure can be implemented. Can it?
> >

Python is *FULL* of ways to build closures:
	class instances
	exec expr in dict
	new.function( oldfunc.func_code, globals )
	generators
	... and maybe a few more I haven't though of.

Which to choose depends on what you need to do.
( What *do* you need to do ? )

On Wed, 27 Feb 2002, Terry Reedy wrote:
>
> Default arg works as long as function is called properly:
>
> >>> def prcnt(counter = [0]):
> ...   print "The current count is %d" % counter[0]
> ...   counter[0] = counter[0] + 1
> ...

However, I wouldn't count Terry's method (above) among them.

A closure is a expression + environment|bindings, but I think
it's implicit in the definition that the two are independent,
and you can easily construct a new closure with a different
set of bindings.

You *could* use new.function to build one with a different default
value, but those values are by position rather than the by-name
globals dict, so that seems more awkward than
	 new.function( func.func_code, newGlobals )

By itself, I'ld just call the above an example of python's strange
syntax for a local static variable.


-- Steve Majewski





From chris.gonnerman at newcenturycomputers.net  Sat Feb 16 13:45:42 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sat, 16 Feb 2002 12:45:42 -0600
Subject: Safe to use asynchat?
References: <slrna6t7f8.n2k.mlh@vier.idi.ntnu.no>
Message-ID: <007e01c1b71a$24cf90a0$0101010a@local>

----- Original Message ----- 
From: "Magnus Lie Hetland" <mlh at vier.idi.ntnu.no>


> I just wondered... Is asynchat "here to stay"? The fact that it is
> undocumented made me worry that maybe it was deprecated or that it
> might suddenly disappear from the distribution... What's its status?

I sure hope it's staying... I just learned how it works from Steve 
Holden's book.





From martin at v.loewis.de  Wed Feb 13 14:31:46 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 20:31:46 +0100
Subject: PullDOM docs?
References: <slrna6l7i2.8hs.mlh@vier.idi.ntnu.no>
Message-ID: <m33d05jhzx.fsf@mira.informatik.hu-berlin.de>

mlh at vier.idi.ntnu.no (Magnus Lie Hetland) writes:

> I just wondered if perhaps some more documentation could be added to
> the standard docs about PullDOM? 

Contributions of more documentation are always welcome; please submit
them to the SF patch tracker.

> It seems to me that for people who want simple XML processing (like
> me), but who might at times find minidom a bit inefficient

I think the solution to that problem would be to speed up the minidom
implementation. If they find it significantly too slow, pulldom won't
help, either, and they may need to use 4Suite's cDomlette, or perhaps
PIRXX (dunno what performance characteristics that has).

Regards,
Martin



From s.schwarzer at ndh.net  Tue Feb 12 12:58:25 2002
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Tue, 12 Feb 2002 18:58:25 +0100
Subject: searching backwards in a string
References: <mailman.1013525800.15169.python-list@python.org> <a4bc9e$ope$06$1@news.t-online.com>
Message-ID: <3C6957C1.8954DB64@ndh.net>

Hello Mark

"Mark Z?ltsch" wrote:
> > Short answer:  Reverse the string first.
> 
> I'm new to this python stuff and to this newsgroup too..

Welcome to Python :-)

> but why you don't use 'rfind' instead of 'find'?
> I think that's much easier...
> 
> somthing like:
> 
> import string
> 
> str = 'ac ab ab ac'
> res = string.rfind(str, 'ac', 0, len(str))
> 
> print 'last position :', res

string.rfind works, as the module name suggests, with constant strings.
However, Paul asked for a search of a _regular expression pattern_
which is not possible with rfind.

I think the answer from Steve is the most appropriate for the question.

Btw, I would recommend to use rfind as string method:

example_string = 'ac ab ab ac'
result = example_string.rfind('ac')

Two more notes:

- str might be a bad name because it's the name for the string type in
  Python 2.2:

  >>> s = 'abc'
  >>> type(s)
  <type 'str'>

  Naming an object str would also make it impossible to use the builtin
  str as factory:
  
  >>> str(1)
  '1'
  >>> str = 'abc'
  >>> str(1)
  Traceback (most recent call last):
    File "<stdin>", line 1, in ?
  TypeError: 'str' object is not callable

  (unless you qualify it explicitly:

  >>> __builtins__.str(1)
  '1'

  which can probably be considered to be bad style. ;-) )

- You can omit 0 and len(str) because these are the defaults anyway.

Stefan


From maxm at mxm.dk  Mon Feb 25 20:51:12 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 26 Feb 2002 01:51:12 +0000
Subject: call me
References: <u7fcig3gep8k40@corp.supernews.com> <slrna7fh30.a72.gerhard@lilith.my-fqdn.de> <wgue8.3490$N7.975881@ruti.visi.com> <3C7ADA75.43CE5F3F@visi.com>
Message-ID: <3C7AEA10.10901@mxm.dk>

Grant's Mother wrote:


> Oh you kids...  And remember to write more often!
> 
> Love, Mom
> 
> P.S.: Are you still dating that nice girl?  You don't bring her 
> by any more.

LOL!

maxm



From lellinghaus at yahoo.com  Sat Feb  2 19:50:36 2002
From: lellinghaus at yahoo.com (Lance Ellinghaus)
Date: Sat, 2 Feb 2002 16:50:36 -0800 (PST)
Subject: Windows CE vs PalmOS version
Message-ID: <20020203005036.44013.qmail@web20904.mail.yahoo.com>

Hello everyone...
I need to write a program in Python that requires a GUI and to run on a
hand held device. This would be very easy if it was a desktop
application, but it is not.

So the question is:
    Should I purchase a Windows CE or Palm compatible device and run
Python on it?

    What GUI interfaces are available for the Windows CE environment?

Thank you!

Lance Ellinghaus


=====
--
Lance Ellinghaus

__________________________________________________
Do You Yahoo!?
Great stuff seeking new owners in Yahoo! Auctions! 
http://auctions.yahoo.com



From tjreedy at home.com  Sat Feb  2 19:04:17 2002
From: tjreedy at home.com (Terry Reedy)
Date: Sun, 03 Feb 2002 00:04:17 GMT
Subject: Standing Objects ... possible?
References: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4> <Pine.SOL.4.30.0202022157430.597-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <58%68.21348$X64.6920411@news1.rdc2.pa.home.com>

>  BTW, _name is only a convention: it
> is treated by the interpreter in the same way as any other symbol
not
> starting with a *double* underscore.  Double underscore __names are
hidden
> to some extent.

_name is not imported by

from module import *

Terry J. Reedy





From jkraska at san.rr.com  Tue Feb 19 02:42:06 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 19 Feb 2002 07:42:06 GMT
Subject: Adding static typing to Python
References: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com> <mailman.1014092911.31096.python-list@python.org> <gft37u8csm1ukd923nr933iq7meimqu5i7@4ax.com> <a4stbu$ptq$1@peabody.colorado.edu>
Message-ID: <ur047u8lrm308jbivi5h8g7ld4dugfb5i7@4ax.com>

>> One of the things I like about C sharp is their use of attributes,
>> where you can define getters and setters on an attribute, but
>> otherwise the attribute is simply used as if it were a public one.
>> An objects attributes can thereby be used as a published aspect of
>> its interface safely, but no one will be the wiser if you manipulate
>> the getter and setter code.
>
>Just curious, isn't there something quite similar to what you describe in 
>py2.2?

Well, one has been able to roll one's own version of this for a long
time; simply override getattr and settatr and on a per instance basis
do what you need to do. Basically, it's all syntactic sugar. For the
C++ "family" of language, it's quite an innovation, however, as this
is the first time one has been able to use an attribute but get a method
as a consequence.

I don't know if 2.2 offered anything like this. It's not obvious in
the documenation. The docs are in sorry shape, though.

C//



From db3l at fitlinxx.com  Mon Feb 25 23:16:13 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 23:16:13 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu> <mailman.1014687041.32731.python-list@python.org> <a5eqvu$79a$1@news.cis.ohio-state.edu>
Message-ID: <u3czox4he.fsf@ctwd0143.fitlinxx.com>

"Locke" <brown.2053 at osu.edu> writes:

> so you are saying it sends the data in groups? how do i know how many calls
> to recv i need to do before the server is done sending everything it needs
> to send? because if i call recv after the server is done sending everything,
> the program will just deadlock waiting. any suggestions? anybody know
> anything about pop3?

Although I'm always a fan of knowing what is really going on under the
covers with the network protocols and what not, have you tried looking
at the poplib module to see if might suit your purpose?  POP3
certainly isn't the most complicated protocol, but as with most
networking protocols, you'll have to deal properly with CRLF issues,
error handling and what not, and there might not be a need to reinvent
the wheel in this case.

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


From jason at jorendorff.com  Wed Feb 20 16:18:26 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 15:18:26 -0600
Subject: Recursively traverse linked list -- help!
In-Reply-To: <7xit8s868e.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEBJDEAA.jason@jorendorff.com>

Paul Rubin wrote:
> "Terry Reedy" <tjreedy at home.com> writes:
> > def f(val):
> >   global f
> >   f = val and f1 or f2
> >   return f(val)
> 
> This is not tail recursion.  The "global f" declaration shadows
> the "f" that names the function.

That's not true.  The f that names the function is global anyway.
It's the assignment that causes the problem.

## Jason Orendorff    http://www.jorendorff.com/



From oren-py-l at hishome.net  Sat Feb  2 14:10:30 2002
From: oren-py-l at hishome.net (Oren Tirosh)
Date: Sat, 2 Feb 2002 14:10:30 -0500
Subject: Enhanced Generators - the difference between sources and iterators
In-Reply-To: <83g04kzoy2.fsf@panacea.canonical.org>
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org> <83g04kzoy2.fsf@panacea.canonical.org>
Message-ID: <20020202191029.GA11306@hishome.net>

On Fri, Feb 01, 2002 at 07:51:01PM -0500, Kragen Sitaker wrote:
> Oren Tirosh <oren-py-l at hishome.net> writes:
> > My implementation of xmap, xfilter and xzip returns a source, not an
> > iterator.  It can be iterated multiple times, just like the output of map,
> > filter or zip.
> 
> That won't work when passed iterators as arguments, right?

Sure, but as long as iterators are kept as invisible temporary objects and
are never referenced explicitly everything can remain functional, lazy and
restartable.  See also my posting about re-iterable generator comprehensions.

> > The logical opposite of a source is a destination.  Destinations should
> > be able to produce a temporary object mirroring an iterator.  I call such 
> > an object a sink (PLUG: see http://tothink.com/python/dataflow)
> 
> Well, if you want to have two pieces of code with independent local
> control state connected by a stream of values, in which the sink runs
> until it needs a new value, and then the source runs until it produces
> a new value, after which the sink runs again etc., generators give you
> that already.  

A transformation looks about the same whether it is a generator function
taking an upstream iterator as an argument or a consumer function taking a
downstream sink as an argument.  The need for sinks becomes more apparent 
when you try to construct a complete dataflow:  you soon find out that you 
must switch from pull to push somewhere in the middle. 

The pull side is a chain of tranformations, each one reading from its
upstream, until the first transformation which reads from the source.  
It is passive.  It won't do anything by itself.  It only responds to 
someone trying to read from its loose end, propagates the request up 
all the way up to the source and returns the data to the puller.

The push side is also passive.  It reponds to someone pushing data to 
its loose end and propagetes it through the transformations all the way 
down to the destination at the end of the chain.

If both sides are passive nothing happens unless some active element reads
from the pull side and writes it to the push side.  The simplet case is a 
for loop reading from the pull side and writing data to the push side.
If the program is juggling multiple dataflows without threads this will 
be some kind of central event-driven loop (asyncore, for example).

> Having "sinks" just gives you another way to do the same thing.  

You are correct in theory, but not in practice.  In the course of a 
dataflow it can switch only once from pull to push without the use
of threads or queues (try it!).  I like generators because they let you 
go much further without resorting to the use of threads.  (Thanks Guido!).

The 'canonical' form of transformations is pull form (with generators taking 
an upstream iterator) but it's also useful to be able to use the push form 
(a consumer taking a downstream sink as argument).  This way you can 
apply some transformation to a push-mode destination and get an object with 
the same type of push-mode interface.

>
  ... examples of the same filter (uniq) implemented 
  ... as generator and as a consumer - deleted
> That's also not any simpler.  Are there examples where the second pair
> is noticeably simpler than the first?  I doubt it.

Sure, the two forms are equivalent.  But *using* them is very different.
One of them exposes a push-mode interface and the other exposes a pull-mode
interface.  Try to write code that wraps one of them so that it exposes 
the other type of interface.  You'll soon see what I mean.

	Oren




From usenet at thinkspot.net  Sat Feb 23 19:33:03 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 16:33:03 -0800
Subject: [Tutor] Working with Email question  [new 'email' module in Python 2.2]
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIENHNOAA.tim.one@comcast.net>
References: <Pine.LNX.4.21.0202231312380.666-100000@hkn.eecs.berkeley.edu> <LNBBLJKPBEHFEDALKOLCIENHNOAA.tim.one@comcast.net>
Message-ID: <6AD3F866286@kserver.org>

[this message has been posted to both the python-list and the
python-tutor list, and bcc'd to my co-developers]

On Sat, 23 Feb 2002 18:40:18 -0500, Tim Peters <tim.one at comcast.net>
wrote about RE: [Tutor] Working with Email question  [new 'email' module
in Python 2.2]:

> Unfortunately, way too many people have bumped into a bug in the 2.2 email:
> if a MIME message claims to be multipart, but actually contains only one
> part, Mailman raises a bogus exception in some contexts.  I think Barry
> never bumped into this because the mail composers he uses never produce MIME
> that lame (*claiming* that a one-part message is multipart isn't exactly
> smart).  This is fixed in current CVS, and should ship in the 2.2.1 bugfix
> release, which Michael Hudson is winding down on (so I guess will happen
> within a few weeks).

There are other problems with the email module as well.

For example: If one tries to read an email from stdin and there is a
problem with the boundaries in an email message, the whole instantiation
of the email message fails.

When I mentioned this previously, in comp.lang.python, the ONLY response
I got was this:
http://mail.python.org/pipermail/python-list/2002-February/085521.html

At the time I was intimidated by that response, but as time has gone on
and I've talked about it with my development partners, my spouse, and
had time to reflect on it, I simply disagree.

The docs for the email module state:
http://www.python.org/doc/2.2/lib/module-email.html

"It subsumes most of the functionality in several older standard modules
such as rfc822, mimetools, multifile, and other non-standard packages
such as mimecntl"

But this is not true. I can read ALL of the messages which cause a
problem with email module in through rfc822 and parse as much as
possible, given the poor email boundaries, and at least have some data
to work with.

Furthermore, ALL of the messages for which I was unable to instantiate
an email instance with the email module in my testing, could be read
into my email client (Forte Agent) and the clients my co-developers were
using (such as Outlook and Eudora).

Here's the deal:

When writing software that COMPOSES an email message, one should stick
to standards. But when writing software that READS email messages, one
should be as fault tolerant as possible. (These are, I believe, general
principles of software authoring, right? Not specific to email clients.)

If I wanted to write an email client, say, using some GUI package and
Python, I could never use the email module as it stands right now,
because it would not accept some email. I, as a developer, could
certainly not tell my customers, "Oh, well, sorry you were not able to
get some of your email. But those were badly formed messages, and so you
should just consider those people who sent them lame-o's."

In fact, I've found that most such mail appears to be spam. But I don't
think this would be very convincing to my customers either. "You lost
mail? Well, that's because it was badly formed. It was probably spam."

Right now, I'm using the email module in a script I'm working on. It
reads a message on stdin as the mail is delivered to the pop mailbox by
Qmail. And I LOVE a lot of the functionality that the email module
provides. Wonderful. Love the object oriented-ness of it, and the way
you can "walk" through the parts of the message. Great. *BUT* in order
to get around this problem with the email module, I'm currently
capturing the stdin as a string, then attempting to instantiate an
instance of an email object, and if that fails, then I resort to parsing
with the rfc822 module. I've been considering re-writing the whole thing
without the email module and just using rfc822 and multifile/mimetools.
This method I'm using now uses too much memory, but if I don't do it
this way, and the instantiation of the email object fails, then I've
already consumed the stdin input and have no way to get that email any
more (it is lost).

I hope this doesn't sound like sniping or whining. I do really love the
email module (excepting for these instantiation problems), and was
drooling over the release of 2.2 partly for this very module. But I
believe that my complaints are valid, and that these are problems that
need to be corrected in the email module.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From philh at comuno.freeserve.co.uk  Thu Feb  7 07:00:29 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 12:00:29 +0000
Subject: Autocoding project proposal.
References: <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net> <slrna61kma.57s.sholden@pgrad.cs.usyd.edu.au> <1873.802T1682T14394596threeseas@earthlink.net>
Message-ID: <slrna64r2t.ahr.philh@comuno.freeserve.co.uk>

On Thu, 07 Feb 2002 04:28:50 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>
>Perhaps that party was simply weak minded enough to seek a negative within
>what I had said in my last group of posts, and found my use of deminsional
>things to be reference to the same type of things?
>
>Fact is, I have had more trouble communicating the VIC to programmers than
>to non-programmers. In fact non-programmers often pick up quite quickly
>when I'm communicating to them

Or perhaps they are just too polite to say they don't understand.

>There continues to be this persistance to preceive the VIC as some sort of
>"language" no matter how many times I say it's not. And this follows wit
>the idea that it's an idea, which it is not.

<paxman>Riiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiight</paxman>

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From claird at starbase.neosoft.com  Wed Feb 13 10:44:54 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 13 Feb 2002 09:44:54 -0600
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de>
Message-ID: <7512146ADFC7C5D2.12C740DBCB3315BF.2E2B43736E1AC6BE@lp.airnews.net>

In article <a4c2ge$1dognu$1 at ID-98166.news.dfncis.de>, MDK <mdk at mdk.com> wrote:
>In PythonWin I entered this script from a book:
>
>from Tkinter import Label
>widget = Label(None, text='Hello GUI world!')
>widget.pack()
>widget.mainloop()
>
>When I run it in PythonWin it runs fine.
>
>When I double click on it in Windows explorer I get a message that tcl80.dll
>can not be found.  That file does not exist on my PC.  How then is it
>running under PythonWin if that file does not exists and how can I make it
>run when double-clicked?
>
>My path includes:
>
>PATH=C:\Python22\;c:\python22\dlls\;C:\Python22\Lib;C:\Python22\Lib\lib-tk
>
>Thanks.
>
>

What name does the script have?  Did you call it
something like
  mytest.tcl
?  
-- 

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


From starx at pacbell.net  Sun Feb  3 18:02:57 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 3 Feb 2002 23:02:57 GMT
Subject: Executing User Entered Code?
References: <8efc7149.0202031326.72806bda@posting.google.com>
Message-ID: <Xns91AA9914DF8ADRASXnewsDFE1@130.133.1.4>

Bob || Sun 03 Feb 2002 01:26:18p:

> Hi,
> I'm writing a network simulator using Python and Tkinter.  I'd like to
> be able to allow the user to enter the code for the routing algorithm
> in Python and then execute it for each node during the simulation.
> 
> Can anyone point me in the right direction here?  I've looked through
> most of the docs and haven't been able to find what I'm looking for.
> 
> Thanks,
> -Bob
> 

Well, theres exec() and eval() built in functions (i think) ... but if 
you want to continually feed in statements and stuff, you might want to 
look at the interpeter-builing tools in the code module. 

There are also some issues about sandboxing and restricted execution of 
untrusted code, some of which may be a part of these modules, but I 
don't know much about that and if you are doing this for debugging or 
simulation and you aren't worried about picking up a string from 
somewhere along the lines of "import os;os.system("deltree /y c:\*")" 
and killing some end-user's box... you probably don't need to worry 
about that stuff.

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From jolsen at mailme.dk  Mon Feb 25 07:46:35 2002
From: jolsen at mailme.dk (Jesper Olsen)
Date: 25 Feb 2002 04:46:35 -0800
Subject: Numeric arrays - initialising with binary data
Message-ID: <cf0ad9fb.0202250446.3df07a08@posting.google.com>

Hi,

The build in array type has the methods

array.tostring()
array.fromstring()

which make it easy to initialise arrays with binary data
read from a file.

The Numeric array, however, apperantly only has the tostring method.

How can I initialise a Numeric array (eg Float32), with binary data?

I can off course do this by using the Numeric C-API, but I am sure
there must be an easier way.

I am surprised the Numeric array type does not have the same methods
as the build in type, if the intention is in time to replace this
module...

Cheers
Jesper


From gerson.kurz at t-online.de  Sun Feb  3 11:28:57 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 03 Feb 2002 16:28:57 GMT
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org> <j47kpuhbuu.fsf@informatik.hu-berlin.de> <3c5d4c21.38617375@news.t-online.de> <mailman.1012752906.26372.python-list@python.org>
Message-ID: <3c5d624d.44293625@news.t-online.de>

On Sun, 3 Feb 2002 17:14:55 +0100, Bas van Gils <bas.vangils at home.nl>
wrote:

>Having read the recent thread on timing how fast code exectes, though, I
>wonder: are lambda-statments _faster_ ?

I think not, especially if the lambda-algorithm is more complicated
than the straightforward solution.

And, of course, my examples were not taken "from real life". There is
however one nice usage of lambda: passing context. 

Say, you have a convention that allows you to specify a callback
function, but no context to be passed to that function. For example,
say someone has written a newton-approximation function:
--------------
# newton-approximate function(x) in intervall range [start,stop]
def newton(function,start,stop):
	...
--------------
now, in this example "function" expects exactly one argument, x, so
"function" must look something like this:
--------------
def sample_function(x):
	# do something with x
...
newton(sample_function,0,pi)
--------------
If you want to pass context parameters to this function, lambda comes
in handy:
--------------
def sample_function(x,context):
	# do something with x given context
...
newton(lambda x:sample_function(x,context),0,pi)
--------------
the expression "lambda x:sample_function(x,context)" creates a new
function "on-the-fly", that will return the value of
sample_function(x,context). This is a very nice feature not found (or
rather: not easily worked around) in languages like C and C++ which
miss lambda:

If you know windows, you know WNDPROC, and you know that you cannot
easily pass context to a WNDPROC function. If you know linux, you know
signal(), and you know that you cannot easily pass context to a signal
handler. 



From logiplexsoftware at earthlink.net  Thu Feb 21 14:46:37 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 21 Feb 2002 11:46:37 -0800
Subject: Newbie: Crazy, but Quick
In-Reply-To: <u7ag65cbc0q3e@corp.supernews.com>
References: <u7ag65cbc0q3e@corp.supernews.com>
Message-ID: <20020221114637.2eab3e11.logiplexsoftware@earthlink.net>

On Thu, 21 Feb 2002 10:49:52 -0800
Virginia O wrote:

> Hi all.
> 
> I need to check if I'm heading in the right direction.
> 
> I'm trying to write a 'protype' of a program. I'll give a quick example
of
> what I'm trying to do:
> 
> Say I spend a month training for Speed Skating.
> I open up my Speed Skating Training Buddy Program on my computer. I just
> bought the program and it should be cross-platform compatible, to work on
my
> PC at home and the Mac's at work.
> 
> I enter the date and time and my race times.
> 
> Then I choose the weather conditions for each day. There should already
be a
> DB with all the info, so I don't have to enter each time.
> 
> I would like to hit a button that calculates the correlations between
> weather and my race times.
> ------
> So far, this is what I understand I need to do:
> 
> Build my database. (I will be using Access,unless convinced otherwise.)

Let me try to convince you right now ;)  I've used Access, MySQL,
PostgreSQL, dBase (and maybe a couple of others I can't remember) and
Access is by far the worst database I've ever encountered.  Much of the
problem is in the "documentation", but it also has problems with concurrent
access and corruption.  Plus, it's not really portable (although you
probably won't have much problem between PC's and Macs).  To top it all
off, you have to pay for it, when superior solutions are available for
free.

I'd suggest MySQL for this particular project, since your database
requirements are fairly simple and it's fairly easy to setup and learn
MySQL.  Plus Python support for MySQL has been around for some time and is
well-tested.  There are also good books available if you need them.

> 
> Learn more Python, so I can write the functions necessary to calculate
> correlations. (There would be more than what I've shown here.)
> 
> Setup the Python database module, to be able to access the DB.

You can use either MySQLdb or mxODBC to do this if you use MySQL.

> Then, do I want wxPython, Tkinter or the like to create an interface? I
want
> graphics and all the fancy stuff.....not just a buncha form fields.

wxPython support on the Mac isn't quite up to speed (last I heard - anyone
correct me?) otherwise that would be my recommendation.  Tkinter should
work well on both platforms, but will look like all Tk-based apps: a bit
out-of-place.

> Then I see py2exe helps create the executable program.

Does this work on Mac?  I thought py2exe only worked for Windows and
McMillan's Installer only worked for Windows and Linux (but maybe I'm
wrong?).  Anyway, I don't think this should really be an issue unless
you're going to distribute to a lot of people.

> ---------
> So, am I way out in left field? At this point, I don't want to hire a
> developer. I'll be happy if I can get the works to function in Python,
> before I make it a standalone program.

Good luck with your project.  

BTW, you'd only be crazy if you planned on doing it in C.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From phr-n2002a at nightsong.com  Sun Feb  3 00:28:27 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 02 Feb 2002 21:28:27 -0800
Subject: Standing Objects ... possible?
References: <mailman.1012709585.15293.python-list@python.org>
Message-ID: <7x665fgmmc.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> Philip Swartzleonard writes:
> > What i want is a construct like:
> > 
> > object Foo( ancestor ):
> >     	pass
> 
> Yep, it exists.

This has gone past me somehow.  Can someone explain it?  Thanks.


From nas at python.ca  Mon Feb 18 12:08:00 2002
From: nas at python.ca (Neil Schemenauer)
Date: Mon, 18 Feb 2002 09:08:00 -0800
Subject: List unchanged in function--why?
In-Reply-To: <fa7108b0.0202180838.6557992a@posting.google.com>; from bit_bucket5@hotmail.com on Mon, Feb 18, 2002 at 08:38:13AM -0800
References: <fa7108b0.0202180838.6557992a@posting.google.com>
Message-ID: <20020218090800.A24359@glacier.arctrix.com>

Chris wrote:
> This must have something to do with references etc, but I'm not sure
> what's going on.

Reset your brain:

    http://www.effbot.org/guides/python-objects.htm

Cheers,

  Neil



From martin at v.loewis.de  Thu Feb 14 03:51:37 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 14 Feb 2002 09:51:37 +0100
Subject: undefined symbols when embedding python
References: <slrna6mm3i.2aj.landshark@burrito.verticaladdiction.net>
Message-ID: <m3y9hwpht2.fsf@mira.informatik.hu-berlin.de>

landshark at verticaladdiction.net (Dave Marotti) writes:

> However, when I try to run my C++ program which contains the embedded python
> interpretor, any call which imports my wrapped C++ class causes undefined
> symbols:
> 
> ImportError: ./libtestwrapc.so: undefined symbol: Py_InitModule4

I take it that the extension module is in a shared library? (which is
kind of surprising, because in an embedded application, you'ld
normally want all extension modules to be linked into the application,
instead of having them as separate files on disk).

I also take it that you use GCC on Linux? In that case, you need to
export the symbols from the executable to the shared library, using
-Wl,--export-dynamic (see ld(1) for details).

Of course, this problem would also go away if you would incorporate
the extension into the application.

Regards,
Martin



From marc.engel at recif.com  Wed Feb  6 04:07:49 2002
From: marc.engel at recif.com (Marc ENGEL)
Date: 6 Feb 2002 01:07:49 -0800
Subject: Catch COM events generated by a Python COM server with Visual Basic
Message-ID: <6411841c.0202060107.773ffe68@posting.google.com>

I am trying for more than one week to use a Python COM server I made
inside Visual Basic. I encounter a problem with Connection Points

Here are the step I already tried:

a - I made a python com server deriving from ConnectableServer.

b - I tested my python com server with a python client. Everything is
working fine including Events. The callback are called correctly.

c - I used typelibutil to generate a Type library for this COM server.
Then I testes my server from VB. As typelibutil does not manage
Connection Points, I was only able to call the methods exposed by my
COM server.

d - I extracted the IDL from this TLB and manually added support for
Connection Points. I created the Event interface:
[.....]
interface IMyServer_Event
{
  [id(1), helpstring("method MyEvent")] HRESULT MyEvent();
};

and added in the coclass the line:

[default, source] interface IMyServer_Event;

e - I recompiled this IDL with MIDL and get a new TLB.

f - From VB, I added a reference to this new TLB. The object viewer
shows me correctly the methods and the event

g - I declared in VB my COM object using the syntax:
Dim WithEvents MyServInstance As MyServComponent
Then VB automatically created the handler for my event:
MyServInstance_MyEvent where I just added a message box

The problem occurs here: I can still call the methods but the event is
never called!

I have tested several things including:

When I call in the server:
interface.Invoke(1, 0, pythoncom.DISPATCH_METHOD, 0, arg)
nothing happens.

I know the VB client call the Advise method of ConnectableServer
because I traced it.

If I call:
id = interface.GetIDsOfNames('MyEvent')
I got a unexpected result: id = 0x60030005

Then if I call invoke on this id nothing happens on the client

I have generated the python stub class for early-binding using makepy
on the TLB generated. I needed to change my interface to a
dispinterface. It worked properly

I managed also to receive event in VB using a small C++ server.

Now I don't know what to do.

If someone has any idea, he is very welcome. The best would be a
working example

Thanks in advance,

--
Marc


From jason at jorendorff.com  Mon Feb 18 14:38:18 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 13:38:18 -0600
Subject: Subclassing a read-only class
In-Reply-To: <20020218153451.GA7162@lilith.hqd-internal>
Message-ID: <HFEKILOLEFEFMKAECNDLKEHODDAA.jason@jorendorff.com>

Gerhard H?ring wrote:
> I'm pretty sure there is a solution, probably with a proxy object.

You are exactly correct.  The basic pattern is like this:

class Proxy:
    def __init__(self, obj):
        self.__obj = obj
    def __getattr__(self, name):
        return getattr(self.__obj, name)
    def __setattr__(self, name, value):
        setattr(self.__obj, name, value)

    ... your extra methods here ...

## Jason Orendorff    http://www.jorendorff.com/




From djc at object-craft.com.au  Sat Feb  9 08:44:42 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 10 Feb 2002 00:44:42 +1100
Subject: Sybase module 0.33pre2 released
Message-ID: <m3g04a93cl.fsf@ferret.object-craft.com.au>

What is it:
The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

WARNING:

Since the last release I have come to the realisation that the Sybase
dynamic SQL functionality is not nearly general enough to use for
correct DB-API functionality.  I have thrown up my hands and gone back
to the drawing board with the high level Sybase.py module.

Once the module is stable again there will be three differences between
the new and old functionality:

1) The paramstyle is now 'format'.  This means that where you
   previously did something like this:

     c.execute("select * from titles where title like ?", ['The %'])

   you now have to do this:

     c.execute("select * from titles where title like '%s'", ['The %'])

   or this:

     c.execute("select * from titles where title like 'The %'")

2) Since the cursors are no longer using dynamic SQL there will no
   longer be an extra result set which reported the status of the
   temporary stored procedure.  This should remove some of the
   confusing behaviour of cursors.

   There should be fewer restrictions on the use of cursors.

3) Threading is enabled in the CT library wrapper by default.

PLEADING:

Please download the pre-release version, try it out, and then send me
some mail telling me what you found.  You would be surprised at how
little mail I get regarding the Sybase module, I really do need your
help.

The module is available here:

    http://www.object-craft.com.au/projects/sybase/sybase-0.33pre2.tar.gz

The module home page is here:

    http://www.object-craft.com.au/projects/sybase/

- Dave

-- 
http://www.object-craft.com.au


From claird at starbase.neosoft.com  Mon Feb 11 17:36:34 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 11 Feb 2002 16:36:34 -0600
Subject: Is Parrot dead?
References: <B88C9617.532F%Roman.Fischer@teleweb.at> <3C66CDEE.EDA48DC2@ix.netcom.com>
Message-ID: <F587A95C94FFF0C8.D4EE006E2A0315BF.810AE72837EF6B68@lp.airnews.net>

In article <3C66CDEE.EDA48DC2 at ix.netcom.com>,
Antaeus Feldspar  <feldspar at cryogen.com> wrote:
>"Ing. Roman Fischer" wrote:
>> 
>> For a long time I do not find informations about Parrot (the programming
>> language). O Reilly still estimates July 2001 for the "Programming in Parrot
>> in a Nutshell"-book and even Amazon (which is not very shy to offer books
>> long before they are available) does not have it in its offer. I also did
>> not find any real complete definitions of this language.
>> 
>> So is this language dead?
>> 
>> Roman
>
>I think I can clear up this confusion.  ^_^  It stems from a huge,
>elaborate April Fool's joke in April of 2001, in which Guido van Rossum
>and O'Reilly books and many other open-source leaders participated.  The
>full story can be found here:
>http://www.oreilly.com/news/parrotstory_0401.html .  
>
>However, that is not the end of the story.  There wasn't ever a real
>language named Parrot that combined the features of Perl and Python. 
>However...  there is now a virtual machine in development that should be
>able to run both Perl and Python code, as well as other interpreted
>languages.  Since the lead developer of this project is the very man who
>orchestrated the Parrot-the-programming-language hoax, Simon Cozens, the
>obvious name for this virtual machine was Parrot.
>
>	-jc

See <URL: http://www.parrotcode.org/ > for details.
-- 

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


From sholden at holdenweb.com  Fri Feb  1 13:34:22 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 13:34:22 -0500
Subject: Equivalent of Perl chomp?
References: <mailman.1012521782.11272.python-list@python.org>
Message-ID: <ffB68.107390$8b.4046814@atlpnn01.usenetserver.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.1012521782.11272.python-list at python.org...
> Now that Steve Holden has determined sliceIt is quicker than endsWith in a
> semi-believable <wink> way, note the timings his program gives are the
same
> whether the string is of length 1000000 or 100 (provided you have enough
> RAM).  For even more fun, see what happens if you check for a 2-character
> suffix instead of a 1-character suffix.  Hmm!  In either case, how much of
> the difference is due to what endswith *does*, and how much is due to the
> overhead of simply building the s.endswith bound method object?
>
> there's-no-end-to-this-kind-of-fun-ly y'rs  - tim
>
1,000,000 iterations.

Testing for length 1:
Endswith 0.00000505
sliceIt  0.00000209

Testing for length 2:
Endswith 0.00001137
sliceIt  0.00001148

Length 3:
Endswith 0.00001055     [!]
sliceIt  0.00001264

Of course, these tests are with the tests succeeding. I'm presuming failure
would shorten the timings, but I'm not going to bother checking that
assumption.

semi-believab-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From timr at probo.com  Thu Feb 14 01:23:41 2002
From: timr at probo.com (Tim Roberts)
Date: Wed, 13 Feb 2002 22:23:41 -0800
Subject: Help! IIS and Python
References: <8f8ffe67.0202131245.375fe55d@posting.google.com>
Message-ID: <eqlm6ugbbtsu4i1d61pek0c4g49as7j119@4ax.com>

sjoshi at ingr.com (Sunit Joshi) wrote:
>
>I'm trying to use IIS5.0 with Python 2.1 and whenever I try to import
>modules I get an error. The test.asp and error are below.
>thanks

It wouldn't have solved your problem, but it might help you to know that
you do not need to import the string module any more.  All of the functions
in the string module are now methods of string objects.  That is, instead
of:
   x = string.upper( "xyz" )
you can say:
   x = "xyz".upper()

And, in fact, the code in the string module now does just that:

   def upper(s)
     return s.upper()

Since we're on the subject of Python in ASP programming, how do you use the
<%=xxx%> trick with indentation?  This fails with an indentation error, for
example:

<%
for i in range(5):
%>
  <%="aaa%d" % i %>

--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From hmartires at cultalg.pt  Wed Feb  6 06:32:18 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Wed, 6 Feb 2002 11:32:18 -0000
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org>
Message-ID: <st888.7971$oI4.33917985@newsserver.ip.pt>

but i don?t need all the numbers, i only need 0.6


"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1012956250.23108.python-list at python.org...
> Hugo Martires writes:
> > suposing that n=10 we get: myArray[0] = 0.600000023 ("something
> > like this")
> >
> > but if i do: myArray[i] + (1.0/n) it give's me 0.6
> >
> > what's the problem of the array, and how can i solve this ?
>
> It's only half of the normal precision.  From the documentation:
>
>     The following type codes are defined:
>         Type code   C Type             Minimum size in bytes
>         'f'         floating point     4
>         'd'         floating point     8
>
> Try using 'd' instead of 'f'.
>
> ## Jason Orendorff    http://www.jorendorff.com/
>
>




From jhorneman at pobox.com  Thu Feb 28 04:52:09 2002
From: jhorneman at pobox.com (Jurie Horneman)
Date: 28 Feb 2002 01:52:09 -0800
Subject: Trouble importing / using a custom text codec
Message-ID: <9ac02e81.0202280152.5bb1bf25@posting.google.com>

I'm having an irritating problem with a text codec I've written.

- The codec is in site-packages/MBCS_codecs/cp932.py.

- site-packages/MBCS_codecs contains an empty __init__.py.

- from MBCS_codecs import cp932 works fine.

- If I put cp932.py in the same place as a script that decodes a text
using encoding "cp932", there is no problem.

However, I want to have all my encodings somewhere else...

- With cp932.py in site-packages as described above, if I decode a
text using encoding "MBCS_codecs.cp932", I get the error message
"LookupError: unknown encoding". This happens even if I import cp932
by hand before decoding the text.

- The code that fails is in encodings/__init__.py :

    modname = encoding.replace('-', '_')
    modname = aliases.aliases.get(modname,modname)
    try:
        mod = __import__(modname,globals(),locals(),'*')
    except ImportError,why:
        # cache misses
        _cache[encoding] = None
        return None

- Executing the following code in one of my own scripts works:

    modname = "MBCS_codecs.cp932"
    modname = modname.replace('-', '_')
    try:
        mod = __import__(modname,globals(),locals(),'*')
    except ImportError,why:
        print "FAILED"

i.e. the module is imported.

- The aliases in encoding/__init__.py do not modify the module name.
In other words, it appears that the same code works or does not work
depending on where it is executed.

I cannot explain this behavior. Any help is appreciated.

Jurie Horneman


From lac at strakt.com  Mon Feb 18 08:44:57 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 18 Feb 2002 14:44:57 +0100
Subject: tab order for Tkinter widgets
Message-ID: <200202181344.g1IDivma012130@ratthing-b246.strakt.com>

I thought that the order in which you tabbed through widgets was the
order in which you gridded them (assuming you are using the grid manager).
I have a bunch of widgets that know what row and what column they are
to be gridded in, but I cannot guarantee what order I can get them out
of the database.

So in the middle of a loop I have:

	decorated.append((d['managerArgs']['row'],
                          d['managerArgs']['column'],
                          d['managerArgs'], widget ))

then at the bottom I can do

        decorated.sort()
        for junk, junk, args, widget in unordered:
            widget.grid(args)

They are all gridding nicely.  The sort works great as well.  And
tabbing gets you randomly flopping from one entry field to another
same as always.  Clearly there is something else I need to do to get
my widgets to tab in the correct order.  What?

Thanks very much,
Laura Creighton



From djc at object-craft.com.au  Wed Feb 13 05:01:24 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 13 Feb 2002 21:01:24 +1100
Subject: Major league corruption - anyone seen anything like this?
References: <mailman.1013557302.9749.python-list@python.org> <7xzo2ekv5g.fsf@ruckus.brouhaha.com> <mailman.1013572042.18876.python-list@python.org> <3C69ED33.CE0100DF@pizero.colorado.edu> <15465.63503.142011.246239@12-248-41-177.client.attbi.com> <Pine.LNX.4.21.0202122224100.12912-100000@higgs.colorado.edu>
Message-ID: <m33d05u2dn.fsf@ferret.object-craft.com.au>

>>>>> "Fernando" == Fernando P?rez <fperez at pizero.colorado.edu> writes:

Fernando> One idea: why don't you remount it as ext3 and try to
Fernando> trigger the condition again? ext3 is supposed to have a lot
Fernando> of internal safeguards that go off when 'wrong' things
Fernando> happen to the filesystem. Of course, it might be that as far
Fernando> as the filesystem is concerned nothing 'wrong' is happening
Fernando> (that is, your process is somehow writing junk but well
Fernando> within the boundaries of its permissions). But it might
Fernando> help, or provide some extra info.

Use strace to track system calls.  If it is your code you will see the
system calls.  If it is hardware you might see things which cannot be
explained by the system calls.

- Dave

-- 
http://www.object-craft.com.au


From winelight at spamcop.net  Sat Feb 23 02:18:14 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Sat, 23 Feb 2002 07:18:14 -0000
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com> <nkUc8.19043$kX6.133584@NewsReader> <655abd4a.0202220114.21447213@posting.google.com> <slrna7dqet.32.missive@whave.frontiernet.net>
Message-ID: <WmHd8.6828$jM1.56405@NewsReader>

I am certainly using 2.2, and I get the urllib hang.

Andrew Ward





From idot at vt.edu  Thu Feb 21 12:00:03 2002
From: idot at vt.edu (Carl Banks)
Date: Thu, 21 Feb 2002 12:00:03 -0500
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net>   <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de> <7x7kp7krjl.fsf@ruckus.brouhaha.com> <a52ml7$42r9l$1@ID-7529.news.dfncis.de> <7x8z9n6nwl.fsf@ruckus.brouhaha.com> <a52os1$45k2v$2@ID-7529.news.dfncis.de>
Message-ID: <j2935a.0b.ln@127.0.0.1>

Karl M. Syring wrote:
> But you really should to it. I have floating point code that was running for
> a long time without a hitch and then the horrible happened. The ugly thing
> is, the code often will not crash but produce wrong results that seem to be
> plausible. If you don't have any results to check the output, the worst
> nightmares might come true.

I'm sorry I answered this troll.

I he weren't a troll, I might ask him to try updating his rational
number 100 times per second with a differential equation and see what
happens to the denominator.  But he is a troll, so it wouldn't matter.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."



From ajeru at vknn.org  Tue Feb 19 19:46:00 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 16:46:00 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com>
Message-ID: <24c39b2c.0202191646.776c1993@posting.google.com>

Daniel Klein <danielk at aracnet.com> wrote in message news:<ghp47u02vceavgohdod9c8vbs9u5po91bt at 4ax.com>...
> Let me reverse the question: What do you think the response would be if Sun
> was asked to drop static type checking (as well as checked exceptions).
> 
> When you can answer this, then you'll have _your_ answer.
> 
> Daniel Klein

Well, my hope is that open source projects are more flexible than
bureaucracies like Sun.


Alexander


From tbarry at rpl-ca.com  Wed Feb 13 18:40:50 2002
From: tbarry at rpl-ca.com (Tristan Barry)
Date: Wed, 13 Feb 2002 23:40:50 GMT
Subject: Simple Question
Message-ID: <6QCa8.7212$zj7.1668243@news1.telusplanet.net>

Is there a way to pass references or pointers as parameters to functions in
python?




From guy at wyrdrune.com  Tue Feb  5 14:47:29 2002
From: guy at wyrdrune.com (Guy Gascoigne - Piggford)
Date: Tue, 5 Feb 2002 11:47:29 -0800
Subject: python and windows short file names
References: <3C5F5711.990519D@mindspring.com>
Message-ID: <3c6036c0$1_2@news.nwlink.com>

I have to say that I cheat and shell out to the already installed windows
scripting host.  Under jython I use the following.

------

def isJython():
    return os.name == 'java'

def isNT():
    return ( isJython() and javaos._osType == 'nt' ) or os.name == 'nt'

if isJython(): import javaos

def shortPath( filespec ):
    if ( isJython() and javaos._osType == 'nt' ) or os.name == 'nt':
        # the real reason for doing this is to avoid problems with spaces in
filename,
        # so if there arn't any in the first place then skip this somewhat
expensive routine.
        if re.match( '\s*', filespec ):
            # monumental hack warning :-(
            # We shell out to the Windows Script Host to get the
            # short name of the passed file name
                script='''
                    function ShortPath( filespec )
                    {
                        var f;
                        var fso = new ActiveXObject(
"Scripting.FileSystemObject" );
                        try {
                            f = fso.GetFile( filespec );
                        }
                        catch( exception ) {
                            f = fso.GetFolder( filespec );
                        }
                        return f.ShortPath;
                    }
                    WScript.Echo( ShortPath( "%s" ) );
                ''' % pathToUnix( filespec )

                fileName=mktemp(".js")
                f = open( fileName, "w" );
                f.write( script );
                f.close();

                try:
                    if isJython():
                        cmdline = "cscript /nologo %s" % fileName
                        p = javaos._shellEnv.execute( cmdline )
                        filespec =
javaos._shellEnv._readLines(p.getInputStream())[0]
                    else:
                        filespec = os.popen( "cscript /nologo %s" %
fileName ).readlines()[0]

                    if filespec[-1] == '\n':
                        filespec = filespec[:-1]

                except Exception, e:
                    print "Caught exception :-( ", e

                try:
                    os.remove( fileName )
                except OSError, e:
                    pass

    return pathToUnix( filespec )

def pathToUnix( path ):
    return re.sub( r'\\', "/", path );

------

"Tarle" <savithari at mindspring.com> wrote in message
news:3C5F5711.990519D at mindspring.com...
> Hello All,
>                 I think this question must have been asked a lot of
> times before.  I could not find in the searches I did.
>
> I want to be able to use java to python ( thru jython is fine) to be
> able to manipulate the full file name and the long dir structures etc.,
>
> I would like to be able short file names and other useful file i/o's
> including ability to delete directory, create directory, rename, move
> etc., etc.,
>
> How can I accomplish this without COMMERCIAL SOFTWARE ?
>
> Thanks for your help.
>
> -Narahari.
>




From db3l at fitlinxx.com  Fri Feb  1 19:19:34 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Feb 2002 19:19:34 -0500
Subject: Question regarding win32comport_demo.py
References: <a3bsm7$16jf7i$1@ID-98166.news.dfncis.de> <3C59CC8E.8AC2E485@engcorp.com> <a3e8ip$17ckca$1@ID-98166.news.dfncis.de>
Message-ID: <uu1t0spk9.fsf@ctwd0143.fitlinxx.com>

"MDK" <mdk at mdk.com> writes:

> The reason behind this is I am trying to uderstand where I am getting the
> data from as the example uses a continusly looping thread to get the data
> which, in light of what you are saying, is not necessary.  Once the
> CreateFile() is done then I can just access the memory buffer.

Using ReadFile() is _how_ you access the memory buffer.  The buffer
Peter was referring to is inside the serial port driver under NT (or
possibly also the filesystem layer, but still within the OS domain).
You can't directly access it because it needs to be managed and
interlocked between retrieval and new insertions (during interrupt
processing).

The serial driver on the system is arbitrating between you and the
physical hardware.  On the receive path, this ensures that no data is
lost, as the driver buffers received information until your
application is ready to accept it (with ReadFile()).

To your original question about the "serial ports memory buffer", most
serial port hardware at best has an approximately 16 byte FIFO queue
of data it has received.  Data has to be drained from there rapidly or
you'll lose new incoming information, so access at that level is
handled by the hardware device driver during interrupt processing (the
serial port chip will interrupt when the FIFO is near full - or at
selectable fill levels).

You need far more buffering than that on most multitasking systems to
ensure data is held until an application gets around to looking at it,
thus the serial port driver has its own buffers, and there may even be
additional buffering being done at higher layers where the port is
behaving like a file handle.  The good news is that you don't really
need to worry about any of this - just open the port as a file, and
then use standard NT file I/O operations (ReadFile, WriteFile) to
perform the I/O.

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


From eric.brunel at pragmadev.com  Mon Feb  4 05:30:14 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 4 Feb 2002 11:30:14 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de>
Message-ID: <a3lnl7$5nq$1@norfair.nerim.net>

Hi,

> Laura, I don't think,
>
> widget=widgetType(master, **mydict)
>
> is necessary. I have tried
>
> widget=widgetType(master, mydict)
>
> with mydict={'text':'hello', 'relief':'sunken'}. It worked without the
> two stars.

The second form is Tkinter specific. All methods/constructors are defined
like that:
def m(parent, optionDict={}, **options)
So it allows to pass the options either as a regular dictionary, or as named
parameters.

> Aren't the two stars just necessary for function calls with variable
> length key parameters?

I admit I didn't know about the m(**dict) syntax, but in the general case,
it is indeed required. Using the "apply" stuff also works.

HTH
 - eric -




From mdk at mdk.com  Thu Feb  7 10:46:53 2002
From: mdk at mdk.com (MDK)
Date: Thu, 7 Feb 2002 10:46:53 -0500
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <3C629FF9.CF4670FF@metaslash.com>
Message-ID: <a3u7hl$1an8dd$1@ID-98166.news.dfncis.de>

"Neal Norwitz" <neal at metaslash.com> wrote in message
news:3C629FF9.CF4670FF at metaslash.com...
> MDK wrote:
> >
> > I have this:
> >
> > # myprog.py
> >
> > from mymod import *
> >
> > def dosomething(x):
> >     dosomething(1000)
> >
> > # End code
> >
> > I want the dosomething(1000) run from the function in mymod.  However,
> > instead it is running dosomething(x) in myprog.py.
> >
> > Note: Changing the names of dosomething() in either module is not an
option.
> > (The names depicted here are not the real names in the program but have
been
> > changed for clarity.)
> >
> > I know that one way to fix this is to change
> >
> > from mymod import *
> >
> > to
> >
> > import mymod
> >
> > But I would like to know if, without changing the import line, there is
a
> > way to tell Python to use the dosomething() from mymod.
>
> from  mymod import *
> mymod_dosomething = dosomething
>
> def dosomething(x):
>     mymod_dosomething(1000)
>
> Neal

Neat!  Thanks, Neal.




From amrit040 at yahoo.com  Tue Feb 26 08:13:20 2002
From: amrit040 at yahoo.com (Amrit)
Date: 26 Feb 2002 05:13:20 -0800
Subject: Strange problem with C API
Message-ID: <71c0a06e.0202260513.4b67841f@posting.google.com>

Hi 

I'm trying to write a C function callable from python. I'm getting
some strange results :

Here's a typical run:
"""
Python 2.0 (#2, Jun 18 2001, 15:29:44) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import lt
>>> lt.__file__
'lt.so'
>>> dir(lt)
['__doc__', '__file__', '__name__', 'fmm']
>>> lt.fmm(range(15),-41)
(0.0, -1.0)
>>> lt.fmm(range(15),-41)
TypeError: function requires at least one argument
"""

On calling the same function again with the same argument gives me a
type error. This alternates. I get a typeerror and then a correct
output and so on. Any ideas what could be wrong. My C code is pasted
below

Thanks 
Amrit
/*

File: lt.c 

I Compile using 
gcc -c lt.c -I/usr/local/include/python2.0/
ld -shared lt.o -o lt.so
*/

#include "Python.h"
#include <stdio.h>
#define N 20 // Maximum size of the input 

static PyObject * findminmax(PyObject * self, PyObject * args);

static PyMethodDef TestMethods[] = {
  {"fmm",findminmax,METH_VARARGS}, 
  {NULL,NULL}
};

void initlt(){
  (void) Py_InitModule("lt",TestMethods);
} 


static PyObject *  findminmax (PyObject * self, PyObject * args)
{
  /*
    A function to find the maximum and minimum of a LT function
    Call from python as (a,b) = fmm(range(15),-41) 
  */
  PyObject *vec, *obj[N], *newobj,*retvec;
  float x[N];
  int a[N];
  int f[N+1];
  int i, n,j;
  float Thr;
  float sum;
  float  max,min;
  
  if (!PyArg_ParseTuple (args, "Of",&vec,&Thr)){
    printf("Error in arg parse\n");
    return NULL;
  }
  if (!PySequence_Check (vec)){
    printf("Error in arg parse as a vec\n");
    return NULL;
  }
  n = PyObject_Length (vec);
  if (n>N) {
    /* This shouldn't be happening */
    /* so who cares what we do now */
    /* look up a better way of gracefully returning */
    return(NULL);
  }
  //  printf ("Starting with thr=%f and n=%d\n",Thr,n);
  for (i = 0; i < n; i++)
    {
      obj[i] = PySequence_GetItem (vec, i+1);
      PyArg_Parse (obj[i], "f", &x[i]);
      a[i]=0;
      f[i]=i;
    }

  /* End Parsing code ... start computation */

  for (i=n;i<N;i++) f[i]=i;
  
  f[n]=n;
  sum=Thr;
  j=0;
  min=1<<30; // min is the min positive value
  max=1<<31; // max is the max negative value
  /* Go through all 2^n possible combinations of a[i] and compute the
max and min
My python implemtation of the same algo works just fine !
*/
  while(1){
    if (sum>=0) min=sum<min?sum:min;
    else max=sum>max?sum:max;
    j=f[0];
    if (j==n)
      break;
    f[0]=0;
    f[j]=f[j+1];
    f[j+1]=j+1;
    if (a[j]==0){
      a[j]=1;
      sum+=x[j];
    }
    else{
      a[j]=0;
      sum-=x[j];
    }
  }
  /*  printf("Done Computation\n"); */
  /*  Make the output tuple */
  retvec=(PyObject *) malloc(2*sizeof(PyObject));
  retvec=Py_BuildValue("(ff)",min,max);
  return(retvec);
}
/* End lt.c */


From starx at pacbell.net  Mon Feb 18 05:13:05 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 18 Feb 2002 10:13:05 GMT
Subject: Faster than we think...
Message-ID: <Xns91B91688165CBRASXnewsDFE1@130.133.1.4>

Philip Swartzleonard || Fri 15 Feb 2002 09:17:30p:

> Well, it certainly is used for applications, although i can't think
> of any off the top of my head. Games too, see pygame for that. I'm
> actually working on a 3d roguelike completely in python, and it's
> only a little bit slow (and thats only because in my current setup
> every frame drawn involves either 1200 or 1600 function calls, i can
> do better =). (There are some screenshots at
> rubydragon.com/image/sproj -- not very good looking yet though :).

Just a little update on this. I just got a GEforce4 based card, which
is capable of doing hardware accelerated graphics in a window, and now
my program is is running perfectly and with no per-frame lag
whatsoever--it's at least 30fps as far as i can tell (and i have some
experience with this :)

So we have those 1200 or 1600 function calls per frame causing NO
IMPACT... it does it ALL thirty plus times a second. Interpreted,
clunky, unoptimized, heavily function dependent code running at
apparently 48000 or so function calls per second without a problem.
And fast.

I'm sure the card isn't optimizing my function calls out of the code
... and i thought they were the slowest thing in the show.

Now i have a different problem, which i had back before i started with
textures and things got complicated. Apparently the WX input queue
freezes after i receive forty or so messages in a very short span...
Weird.

Anyway, the point is... well, had i tried a hardware accelerated
version of this sooner, i would probably have seen it... but it seems
that less and less is fit these days for being written in a lower
level language than python =)


From mlh at vier.idi.ntnu.no  Sat Feb 16 12:59:04 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 16 Feb 2002 17:59:04 +0000 (UTC)
Subject: Safe to use asynchat?
Message-ID: <slrna6t7f8.n2k.mlh@vier.idi.ntnu.no>

I just wondered... Is asynchat "here to stay"? The fact that it is
undocumented made me worry that maybe it was deprecated or that it
might suddenly disappear from the distribution... What's its status?

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From DeepBleu at DeepBleu.org  Fri Feb  1 14:24:26 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Fri, 1 Feb 2002 13:24:26 -0600
Subject: telnetlib read_until()( question
Message-ID: <u5lqt76m0ogfd8@corp.supernews.com>

For telnetlib's read_until(expected [,timeout])

a.  Does anyone know the default value of the timeout when no timeout is
given, and
b.  When a timeout is given:  Is the numeric value in seconds or mintues?
Thanks,
DeepBleu




From jason at jorendorff.com  Mon Feb 11 17:15:41 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 11 Feb 2002 16:15:41 -0600
Subject: help in converting c code needed
In-Reply-To: <a49dk0+n16v@eGroups.com>
Message-ID: <HFEKILOLEFEFMKAECNDLEEOADBAA.jason@jorendorff.com>

maximilianscherr wrote:
> can anybody help me convert this c code to python:

Well, actually it's C++.  And one or two pieces are missing.
But sure.

First, use zlib instead, if you can.

>>> import urllib   # ...just so we have something to compress
>>>
>>> data = urllib.__doc__   # a big string to play with
>>> print len(data)
955
>>> zd = data.encode('zlib')
>>> print len(zd)
522
>>> data2 = zd.decode('zlib')
>>> print len(data2)
955
>>> assert data == data2


But if you really have to do this, then you can pretty much
translate this code line-for-line into Python.  But many little
changes are needed:

  - Turn pointer variables into index variables
  - Turn "pdest" into an array, and don't make the user pass you
    an empty array as an argument; create one automatically instead.
    (By "array", I mean this:
     http://www.python.org/doc/current/lib/module-array.html )
  - Don't require a src_size argument; you can just do len(src)
  - Instead of "--x" you must do "x -= 1"
    and so forth.

If you don't know C or C++, you have little chance of getting
it all exactly right.  Get a friend to do it.  :)

## Jason Orendorff    http://www.jorendorff.com/





From owen at astrono.junkwashington.emu  Tue Feb  5 13:00:21 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Tue, 05 Feb 2002 10:00:21 -0800
Subject: question about subclassing dict
References: <mailman.1012873998.10011.python-list@python.org>
Message-ID: <a3p6jl$102i$1@nntp6.u.washington.edu>

In article <mailman.1012873998.10011.python-list at python.org>,
 "Delaney, Timothy" <tdelaney at avaya.com> wrote:

>> From: Russell E. Owen 
>> 
>> I'm writing a subclass of dict (OrderedDict, keeps items in the order 
>> entered; you can get it from http://www.astro.washington.edu/owen/ if 
>> interested.).
>
>I would prefer to call this a "ChonologicalDict" or "ChronoDict" to better
>describe the order that it stores things.
>
>"OrderedDict" implies that the objects will come out in the order that you
>would get by calling sort() on a list.

I would call that a SortedDict. I've contemplated writing one, but have 
not found a use for it yet. (In contrast, I've been wanting an ordered 
dictionary -- or chronological if you prefer -- for a long time. Now 
that I have it, I can go back and clean up some of my existing code 
quite nicely.)

In general, I wish Python had a more complete set of collection classes 
(like Smalltalk). Writing OrderedDict was not all that difficult, but 
there were a lot of finicky details to get right and I'm not sure even 
now that it offers the full dictionary interface.

-- Russell

P.S. I uploaded a few more improvements as of today. Keys() was 
returning the internal list instead of a copy, and I made all the iter 
methods depend on iterkeys(), for less reliance on internal details and 
easier subclassing. I plan to stop messing with it now and start using 
it.


From jjl at pobox.com  Wed Feb 27 17:26:04 2002
From: jjl at pobox.com (John J. Lee)
Date: Wed, 27 Feb 2002 22:26:04 +0000
Subject: 15 minute presentation on python?
In-Reply-To: <fa7108b0.0202271256.57e27d82@posting.google.com>
References: <3c7ab974_4@corp.newsgroups.com> <a5ef3h$6p6g1$1@ID-11957.news.dfncis.de> <3C7ADE12.C6D895A0@engcorp.com> <fa7108b0.0202271256.57e27d82@posting.google.com>
Message-ID: <Pine.SOL.4.30.0202272224160.15823-100000@mimosa.csv.warwick.ac.uk>

On 27 Feb 2002, Chris wrote:

> Peter Hansen <peter at engcorp.com> wrote in message news:<3C7ADE12.C6D895A0 at engcorp.com>...
[...]
> I agree--if you have Internet access, do something that grabs
> something from a web page.  I also did the above (based on an example
> in Learning Python to grab the temp for a city from a weather site)
> and it wowed me.
[...]

The 'wow' isn't so much the fact of being able to do it, as the length of
the xxxlib modules that implement the various internet protocols.


John



From shelton at custommachine.com  Fri Feb 15 18:09:07 2002
From: shelton at custommachine.com (Jeffrey N. Shelton)
Date: Fri, 15 Feb 2002 18:09:07 -0500
Subject: win32com and Passing SafeArrays of Doubles
Message-ID: <005201c1b675$c8662210$0a01a8c0@DELL530>

In the process of using win32com to hook Python into SolidWorks (a 3D
modeler), I've run into the same problem as others before me -- namely, some
strange things happen when passing SafeArrays of doubles into SolidWorks.
Having created the appropriate gen_py file, I'm usually successful when I
pass in a tuple with the appropriate values. But when passing SafeArrays of
doubles into SolidWorks, it thinks I've inserted an extra value of 0.0
between each parameter I've passed. So it thinks that an array (1, 1, 1) is
instead (0, 1, 0, 1, 0, 1).

Past discussion at:
http://groups.google.com/groups?hl=en&selm=39943819.65368645%40news.onlineis
p.com

Mark Hammond's reply (back in August of 2000) indicated that he thought the
problem was with the SolidWorks code, and I'm quite willing to accept that.
However, I've noted from the SolidWorks API documentation (quoted below)
that, even when using VB, some special massaging is required.

So, do I have to do any special 'wrapping' to get SolidWorks to accept my
arrays, or does this still look like its just a shortcoming in the
SolidWorks implementation?

Any suggestions would be appreciated!

Jeff

>From SolidWorks API Documentation:
>
> Passing array data to SolidWorks in Visual Basic requires the use of a
> Variant variable to hold the data. The procedure is to put the data
> into an array then to assign a Variant to the array. The variant is then
> passed to an API function with parentheses round it:
>
>      Dim varArray As Variant
>      Dim DataArray(9) As double
>
>      ' Assign values
>      DataArray(0) = 0.1
>      ...
>
>      ' Pass the data to SolidWorks
>      varArray = DataArray
>      Result = Object.Method( (varArray) )  ' Note the extra parentheses
are required




From jblaine at mitre.org  Mon Feb  4 11:15:32 2002
From: jblaine at mitre.org (Jeff Blaine)
Date: Mon, 4 Feb 2002 16:15:32 +0000 (UTC)
Subject: Modules/makesetup failure
Message-ID: <Xns91AB72884BB13jblainemitreorg@129.83.20.55>

Any Python version seems to exhibit this behavior at build
time (1.5 through 2.1.x at least).

If I put a line (which I understand to be valid) like:

    -L/my/tcl/lib -Xlinker -rpath /my/tcl/lib \ 

in Modules/Setup, then the makesetup script fails when it
hits the second '/foo' and says it's a 'bad word'.

If I add a special case in makesetup at around line 137 like
this:

    /*)    libs="$libs $arg"; skip=libs;;

it works fine and Python + Modules compile fine.

What's up with that?  I'm surprised that this has not been
crossed/fixed yet and it makes me think, against all instinct
in this case, that I am doing something wrong.  That's a very
common gcc+ld convention to use... (-Xlinker -rpath /foo)


From mwh at python.net  Mon Feb 11 09:23:18 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 11 Feb 2002 14:23:18 GMT
Subject: signal through popen2
References: <de6c123c.0202092047.424c8091@posting.google.com>
Message-ID: <u4rkohz7g.fsf@python.net>

llchen223 at hotmail.com (Lei Chen) writes:

> Hi,
> 
> I would like to send a CTRL-C on a process opened by popen2.  Is
> this possible?

Not as such, as you don't know it's PID.  You could try:

def popen2pid(cmd, bufsize=-1, mode='t'):
    inst = Popen3(cmd, 0, bufsize)
    return inst.fromchild, inst.tochild, inst.pid

(Assuming unixoid systems).

There might be some ioctl for finding out which process owns the other
end of a pipe, but I'm not sure.

Cheers,
M.

-- 
  Programming languages should be designed not by piling feature on
  top of feature, but by removing the weaknesses and restrictions
  that make the additional features appear necessary.
               -- Revised(5) Report on the Algorithmic Language Scheme


From seebs at plethora.net  Fri Feb  8 20:44:34 2002
From: seebs at plethora.net (Peter Seebach)
Date: 09 Feb 2002 01:44:34 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1372.801T1070T13994028threeseas@earthlink.net> <3c60c05c$0$79564$3c090ad1@news.plethora.net> <2106.802T789T14243611threeseas@earthlink.net>
Message-ID: <3c647f02$0$22477$3c090ad1@news.plethora.net>

In article <2106.802T789T14243611threeseas at earthlink.net>,
Timothy Rue <threeseas at earthlink.net> wrote:
>C# is a subtopic of .NET / Mono, as is many other programing languages,
>including Python.

Unless, of course, you use that to build *your* framework.  They need
options.  Give them some.

>The community you speak of is not inclusive of the end user/consumer, but
>rather a community of programmers that are likley to follow MS mindset
>programming.

I think it's safe to say they're the community most likely to accept your
vision.

>If you have never done it all the way right, how will you know when you
>do?

How would you know?  You never really know; you simply observe that your
current model covers everything you've seen.

-s
-- 
   Copyright 2001, all wrongs reversed.  Peter Seebach / seebs at plethora.net
   $ chmod a+x /bin/laden      Please do not feed or harbor the terrorists.
     C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/


From m.faassen at vet.uu.nl  Thu Feb  7 13:57:11 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Feb 2002 18:57:11 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org>
Message-ID: <a3uim7$dqe$1@newshost.accu.uu.nl>

Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
[snip]
> Ah, perhaps you are right.  I was in a particularly bad mood at that point.

> How the heck to you "enjoy" an idiot like Rue though?

I'm just weird like that. It's like a voyage of exploration to see
strange things. Or like an interest in alien life forms. Or human beings
for that matter. I am interested in the occasional unusual personality.

I was not suggesting we all enjoy him or something. We all should just
go our merry way. Besides, you're just responding to him like that as
you're a pseudo-programmer feeling threatened by his superior auto-coding
technology! I mean, it's all obvious that everybody arguing with him on
it is just running scared. Rue's brilliant psychological insight in the matter
gave me one of those zen moments. :)

Anyway, I agree with people that I haven't made a lot of sense out of 
his auto-coding proposals yet. On the surface the desire is simple; 
being able to write more complicated software without doing so much 
detailed work. Using a computer for it, instead! That's really what most
good programmers do; climb up onto ever higher levels of abstractions (in
any direction that seems 'up') to make problems easier and quicker to solve.

Presumably we'll keep on climbing. Where Rue loses me is in that he 
says he has a general solution to this that can do it all automatically.
Of course theoretically we could construct a human-scale AI that can do it,
but until then...writing software is a human activity, requiring a large
range of human knowledge. I don't see a computer do that just yet, though
of course software (languages, libraries and other things) is likely to 
continue to become increasingly helpful.

Of course Rue might likely pop in and say I'm all wrong. I'd enjoy that,
so far he doesn't seem to have responded to me yet. Probably doesn't know
what to make of me.. :)

Regards,

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


From fperez528 at yahoo.com  Tue Feb  5 11:37:59 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 05 Feb 2002 16:37:59 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <mailman.1012925350.31242.python-list@python.org> <7kr06u8dctsk6b54qbthbligih0np0v31t@4ax.com>
Message-ID: <a3pqeu$nru$3@peabody.colorado.edu>

Courageous wrote:

> This is a case
> of "implement it to find out."

An example our recent 'troll du jour' could learn lots from. Not mentioning 
names out of fear of waking up the unsightly beast...

f.


From dogansmoobs at NOSPAM.prexar.com  Mon Feb 18 14:38:05 2002
From: dogansmoobs at NOSPAM.prexar.com (Mik Mifflin)
Date: Mon, 18 Feb 2002 14:38:05 -0500
Subject: PyKDE
Message-ID: <u72ll0kis0p201@corp.supernews.com>

I'm trying to learn Qt programming and decided to use my new favorite 
language (Python, of course).  I'm using SuSE 7.3, and there is an rpm 
called PyQt installed, but it looks like I need PyKDE as well?  This is the 
program that I'm trying to run, and the error messages it gives:

!/usr/bin/python
from qt import *
from kdecore import *
from kdeui import *

a = KApplication (sys.argv)
w = QWidget()
a.setMainWidget (w)
a.setTopWidget (w)
w.resize (200, 200)
w.show()
a.exec_loop()

> ./qt1
  File "./qt1", line 3, in ?
    from kdecore import *
ImportError: No module named kdecore

If I do need PyKDE, does anyone know of an rpm that will work with SuSE?  
Or am I just better off installing from source?

--
 - Mik Mifflin
.doog era syeknoM
AIM:  dogansmoobs
EMAIL:dogansmoobs at prexar.com


From m.faassen at vet.uu.nl  Mon Feb  4 20:51:22 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Feb 2002 01:51:22 GMT
Subject: WANTED: IBM 1401 AUTOCODING WAREZ [was: Autocoding - the resistance to expect from psuedo programmers]
References: <mailman.1012829589.12111.python-list@python.org>
Message-ID: <a3ndqq$h6c$1@newshost.accu.uu.nl>

Steven Majewski <sdm7g at virginia.edu> wrote:
> HEY DUDEZ !
> I WANT TO GET INTO THIS AUTOCODING THING.
> ANYONE KNOW WHERE I CAN GET SOME IBM 1401 WAREZ ?
> -- MULTICS-MAN

FOR IMMEDIATE DISPATCH TO ALL PSU MEMBERS -- TOP SECRET

Recently a discussion has erupted on comp.lang.python on 'autocoding'.
While the resulting flamewar has lengthened the total age of the universe
by a significant amount, the Python Secret Underground still asks any and
all of its members to refrain from posting in this thread.

'autocoding' is in fact a secret PSU project. Using autocoding technologies,
it is expected the Earth can be saved from the dreaded alien whitespace
eating nanoviruses with just nine (9) commands. This is possible because
of advanced input and output redirection technology, which allows the
right sequences of commands to be generated using a simulated accomplishment
algorithm. This gives us many of the benefits of time travel without the
necessity of importing the timetravel module, allowing in effect a 
true restricted execution environment for the 'grok' command.

Recently an unfortunate accident involving large quantities of an alcoholic
beverage and Guido's time machine has resulted in the creation of a rogue
mutant version of the timbot self-designated as 'Tim Rue', the name being
an obvious subtle clue as to its origins. The PSU member responsible for
this accident has been removed from the time stream and does not exist
anymore. Any lingering memories of a person called Steven Fredrikelli will
soon disappear.

The rogue heavily mutated 'Tim Rue' timbot immediately started to leak 
details about the supremely secret 'autocoding' project to usenet.
Luckily its transcendent post-singularity knowledge is virtually 
incomprehensible to pre-singularity humans. Thus, the PSU fully 
expects that with a few simple measures lasting harm to this timeline
can be avoided, and that the secrecy of the autocoding project can be
preserved.



From sholden at holdenweb.com  Wed Feb 13 07:42:11 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 07:42:11 -0500
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <CQka8.189960$KM2.8132548@atlpnn01.usenetserver.com> <7x1yfq2gdf.fsf@ruckus.brouhaha.com>
Message-ID: <uata8.206064$KM2.8431588@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in ...
> "Steve Holden" <sholden at holdenweb.com> writes:
> > Paul, this thread's probably now old enough for you to tell us what the
real
> > problem is! Why exactly do you need to search backwards from the
50,000th
> > character to find the beginning of an HTML tag?
>
> Suppose I'm parsing the file and I see a </table> tag and I want to
> find the matching <table> tag.  It could be pretty far back in the file.
> That's what I was doing when I encountered this question.  But searching
> backwards is a normal thing to want to do in general--for example it's
> a standard command in any decent text editor.
>
> Anyway, I just entered a sourceforge bug about it being missing from
> Python's re module.
>
I'd be very surprised if this meets with any response other than "this is
not a bug".

Frankly, if you see a </table> tag and you have no idea where the matching
<table> tag appears then whatever you are doing to the HTML file you
certainly aren't parsing it!

Don't know whether this will help: it's an example from "Python Web
Programming" that shows you how to extract the table structure from an HTML
file.

import htmllib, urllib, formatter, sys

def Usage():
    print """
Usage: python showtbls.py URL
"""

class myHTMLParser(htmllib.HTMLParser):

    def __init__(self, f):
        htmllib.HTMLParser.__init__(self, f)
        self.tblindent = 0

    def start_table(self, attrs):
        sys.stdout.write("%s<table" % ("    " * self.tblindent, ))
        for k, v in attrs:
            if k in ("width", "cellspacing"):
                sys.stdout.write(' %s="%s"' % (k, v),)
        print ">"
        self.tblindent += 1

    def end_table(self):
        self.tblindent -= 1
        print "%s</table>" % ("    " * self.tblindent, )

def parse(url, formatter):
    f = urllib.urlopen(url)
    data = f.read()
    f.close()
    p = myHTMLParser(formatter)
    p.feed(data)
    p.close()

if len(sys.argv) != 2:
    Usage()
else:
    fmt  = formatter.NullFormatter()
    parse(sys.argv[1], fmt)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From m.faassen at vet.uu.nl  Thu Feb  7 20:02:49 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Feb 2002 01:02:49 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <slrna650nr.ahr.philh@comuno.freeserve.co.uk> <a3v1fc$jmc$1@newshost.accu.uu.nl> <3C63246F.135516E@engcorp.com>
Message-ID: <a3v83p$jms$1@newshost.accu.uu.nl>

Peter Hansen <peter at engcorp.com> wrote:
> Martijn Faassen wrote:
[I explain the gospel of auto-coding]
> My god, it's infectious!  Martijn, did the PSU put you up to

The PSU has nothing to do with it. I've genuinely become convinced 
programming is a racket and with just 9 commands we would be able to
overthrow this racket. What's so hard to believe about that? No need
to come up with any conspiracy theories... Besides, if the PSU were
behind it, it would've told me to


From tdelaney at avaya.com  Thu Feb 28 21:04:37 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 1 Mar 2002 13:04:37 +1100 
Subject: Status of PEP's?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1CB@natasha.auslabs.avaya.com>

> From: Tim Peters [mailto:tim.one at comcast.net]
> [Delaney, Timothy]
> > ...
> > Are the following two equivalent though?
> 
> Not necessarily, but the possible difference isn't where you think:
> 
> > if 1 in obj:
> >     pass
> >
> > if 1 in iter(obj):
> >     pass
> >
> > i.e. is iter(obj) called implicitly in the first case?
> 
> Maybe.  First type(obj) is checked to see whether it 
> implements __contains__
> (at the C level, whether obj->ob_type->tp_as_sequence->sq_contains is
> reachable and non-NULL).  If it is, iter() gets involved if 
> and only if the
> __contains__ method decides to use iter() (and, for example, 
> dicts certainly
> do not).  If type(obj) doesn't implement __contains__, then 
> "in" tickles
> iter(obj) repeatedly.

That's pretty much what I thought - I was assuming the case that
__contains__ wasn't defined. It answers the real question, which was "In the
general case, is '1 in iterable' a valid expression?".

The answer is "Yes, but you can do something to prevent it."

Tim Delaney



From mhammond at skippinet.com.au  Fri Feb 15 08:04:56 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 15 Feb 2002 13:04:56 GMT
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk> <3C6C4C37.6020209@skippinet.com.au> <t4oqZKA4$Mb8Ew3k@jessikat.fsnet.co.uk>
Message-ID: <3C6D07E2.5030005@skippinet.com.au>

Robin Becker wrote:
> sounds right to me, they would also have to fix pythonservice as well.

I don't see why pythonservice needs to be fixed - it just uses the 
python (if indeed it even needs that registry key.

> Incidentally I notice that the path used by pythonservice is the 8.3 dos
> mangled version. Inside the Zope startup script they're using the normal
> long filename version to possibly correct the path. Will that not cause
> problems? 

I presume you are talking about the sys.path entries?  I have no idea 
why pythonservice would use the short name - I don't think I have ever 
seen it do that.  Can't see why it would be a problem though - both 
names are equivilent.

Mark.





From James_Althoff at i2.com  Fri Feb 15 14:52:54 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 15 Feb 2002 11:52:54 -0800
Subject: Java vs Python ... ???
Message-ID: <OFDB7E6D66.B4C7A4BD-ON88256B61.006C9DFB@i2.com>

[Steve Orr]
>For instance, with the Oracle Application Server (not the database)
>you supposedly can have all your Java apps load balanced "for free"
>without having to write load balancing code. So I was wondering the
>other day, since Jython compiles to bytecode, will it run under
>applications servers (such as Oracle) and enjoy the same Java
>features those servers provide?

[Jim]
>Yes.

[Steve]
>Thanks but could you make your answers a little more brief. :-)

Si.




<wink>

As Jython is in some sense "just a Java program" -- it (Jython, that is)
being written entirely in Java -- a Jython "program" is really just an
ordinary Java program that happens to read in text files with .py
extensions and happens to use a lot of the Java reflection API.  So
assuming some details like the fact that Java security has to be set up to
enable the writing of files (normal for Java applications) your Jython
programs can run in J2EE containers just like any Java programs can.  And,
yes, this is a *very* nice feature of Jython.

Jim




From martin at v.loewis.de  Wed Feb 27 03:55:34 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 09:55:34 +0100
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu>
Message-ID: <m33czn2tix.fsf@mira.informatik.hu-berlin.de>

Fernando Pereira <pereira at cis.upenn.edu> writes:

> > arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**'
> > test)
> 
> This production cannot be disambiguated top-down with *any* lookahead, for
> the reason you suggest. 

There is no reason to disambiguate anything here: this is a single
non-terminal. It has a first set of firstset(argument) \/ {'*', '**'}.

Notice that, because it just uses a regular syntax on the right-hand
side, there definitely is a deterministic finite automaton that
accepts precisely the words defined by the regular expression
(assuming, for the moment, that argument and test are tokens). For the
same reason, this production, in itself, defines a LL(1) language.

> Thus, it doesn't seem Python's grammar is LL(1). 

You definitely cannot conclude this statement from the above
production.

Regards,
Martin


From starx at pacbell.net  Fri Feb  8 20:07:28 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 01:07:28 GMT
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <a40e2r.3vvegs9.1@kserver.org> <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net> <tbf14a.ep7.ln@wombat.san-francisco.ca.us> <a41ob2$9k3$1@panix2.panix.com>
Message-ID: <Xns91AFAE30B3A22RASXnewsDFE1@130.133.1.4>

Aahz Maruch || Fri 08 Feb 2002 03:52:02p:

> In article <tbf14a.ep7.ln at wombat.san-francisco.ca.us>,
> Keith Keller <kkeller at speakeasy.net> wrote:
>>
>>Aahz, are you sorry you asked yet?  :)
> 
> Not really.  Someone referred to "Perlies" in an e-mail to me (which I
> actually misread as "Perlers", because I'd heard that term before), and
> when I asked another Perl programmer of my acquaintance, zie didn't much
> care for "Perlers" (and had never heard of it, to boot).  So between all
> the responses I've received in public and private, I think I'll stick
> with "Perl programmers".

Maybe shortened to Perlpro(s)? That sounds respectful and short, though 
probably silly. (Add pypro, forpro, cpro, japro, adepro, etc :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From timr at probo.com  Wed Feb 27 02:36:48 2002
From: timr at probo.com (Tim Roberts)
Date: Tue, 26 Feb 2002 23:36:48 -0800
Subject: Class functions
References: <mailman.1014764563.7542.python-list@python.org> <QGUe8.213$en5.1614@typhoon.sonic.net>
Message-ID: <i33p7u03h6ptkr3rmv4c5ljeolaol5hseh@4ax.com>

"Chad Maine" <chad.maine at broadlink.com> wrote:

>Yeah, my function runs a loop looking for updates and posting them to a
>StringIO file object.  I want access to that file object in real time in my
>client instead of waiting for the function to "return".

Is there any reason why you can't spin off a second thread?  That seems
like the ideal solution.  Python has a relatively simple-to-use threading
module.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From shenwang at ameritech.net  Sat Feb 23 13:49:17 2002
From: shenwang at ameritech.net (Shen Wang)
Date: Sat, 23 Feb 2002 18:49:17 GMT
Subject: Is there a python parser generator that is similar to JavaCC from webgain?
Message-ID: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com>

As title

thanks





From ron.l.johnson at cox.net  Sun Feb 17 01:11:40 2002
From: ron.l.johnson at cox.net (Ron Johnson)
Date: Sun, 17 Feb 2002 06:11:40 GMT
Subject: Have to try
References: <004501c1b763$78bca9c0$b278c53f@pacbell.net>
	<mailman.1013918127.17499.python-list@python.org>
Message-ID: <20020217001138.49b8ca05.ron.l.johnson@cox.net>

On Sat, 16 Feb 2002 19:53:30 -0800 Wayne Pierce <wayne at mishre.com> wrote:

> Rick,
> 
> Rick Hamilton wrote:
> > OK - I know I am going to get fried here but - I am out of options. I'm 
> > a programmer (Java, VB and C++) and am taking a class on Operating 
> > systems. The nut teaching this course is asking us to do a ton on 
> > programming in Python. OK -
> 
>    You've come to the wrong newsgroup if you want to get flamed, sorry. 
>   Is the professor a nut because he's using Python?  Or does he go 
> galloping around the room with two coconut halves talking about swallows?

Just hypothesizing, but maybe Wayne is casting aspersions upon his
teacher for using Python in an OS class?  Assemply, C, C++, Fortran, 
even Bliss, but not Python.

-- 
+------------------------------------------------------------+
| Ron Johnson, Jr.        Home: ron.l.johnson at cox.net        |
| Jefferson, LA  USA      http://ronandheather.dhs.org:81    |
|                                                            |
| 484,246 sq mi are needed for 6 billion people to live, 4   !
! persons per lot, in lots that are 60'x150'.                |
! That is ~ California, Texas and Missouri.                  !
! Alternatively, France, Spain and The United Kingdom.       |
+------------------------------------------------------------+


From gh_pythonlist at gmx.de  Mon Feb 11 14:23:11 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 11 Feb 2002 20:23:11 +0100
Subject: Python MySQL and MAC OS X
In-Reply-To: <a487t7$qhb$1@highway.leidenuniv.nl>
References: <3C656ACD.1020907@inter.nl.net> <mailman.1013294776.15973.python-list@python.org> <3C663807.AA608724@xs4all.nl> <a487t7$qhb$1@highway.leidenuniv.nl>
Message-ID: <20020211192310.GC1686@lilith.hqd-internal>

Le 11/02/02 ? 10:54, Peter H.J. v.d. Kamp ?crivit:
> >> > According to the MAC OS X documentation of Python 2.0 I have
> >> > to drop the setup.py file onto the interpreter and pass the
> >> > build argument by setting a Unix style command line (which can
> >> > be set by holding the option key during interpreter startup).
> >> 
> >> Sounds frightening ;-) Don't you have a shell or terminal window on
> >> MacOS X?
> >
> >Sure there is. I think maybe he's not yet aware that there are still two
> >very different version of Python available for OSX: MacPython, which
> >is still backwards compatible with MacOS 9 and earlier, and then there's
> >plain vanilla "unix Python". It sounds like he's trying with MacPython
> >where he really needs unix Python.
> 
> 
> Sorry, that was an omission in my original posting. I'm indeed working with
> MacPython and not with the plain vanilla "Unix Python".

I have almost zero knowledge of Macs, but AFAIK MacPython is built with
CodeWarrior, so you'll probably need this to compile MySQLdb.

I strongly suggest you use the unixoid Python for MacOS X, as it will
allow you to use GNU autoconf et al. for building the MySQL client
libraries and gcc for building the C part of the MySQLdb Python
extension.

I used to see some MacOS X folks around here, but there's also a Python
for Macintosh SIG with a quite active mailing list:

    http://www.python.org/sigs/pythonmac-sig/

I've quickly browsed around at http://www.cwi.nl/~jack/macpython.html
and it looks like you can easily install the developer tools from your
MacOS X distribution.

I'd also suggest you install a commandline version of Python from
http://tony.lownds.com/macosx/ then continue with building MySQLdb from
the commandline.

Ah well. maybe you won't need to build MySQL proper, looks like a binary
distribution for MacOS X is available at
http://www.mysql.com/downloads/mysql-3.23.html

Just recognized that I have access to MacOS X on Sourceforge's compile
farm. I'll try to build a MySQLdb.

later,

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.6 ?C      Wind: 6.2 m/s



From martin at v.loewis.de  Tue Feb 12 05:09:27 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 12 Feb 2002 11:09:27 +0100
Subject: Python 2.2 class methods examples?
References: <e840346c.0202120154.6b41f449@posting.google.com>
Message-ID: <m3pu3b6mg8.fsf@mira.informatik.hu-berlin.de>

kalath at lycos.com (Mark Brady) writes:

> Hi. I've just moved to 2.2 and I think I've got my head around most of
> the cool new features but I can't see what class methods would be used
> for. I understand the difference between them and static methods and I
> know where static methods are useful but it would be great if someone
> could post an example of class method usage or some links. Thanks.

One application is to use them as factory methods:

class Base:
  _allinstances = {}
  def byName(cls, name):
    try:
      return cls._allinstances[(cls,name)]
    except KeyError:
      return cls._allinstances.setdefault((cls,name), cls())
  byName = classmethod(byName)

class Derived(Base):
  pass

Then, Derived.byName("foo") will give you a Derived instance, namely
the same over and over again, but that will be different from
Base.byName("foo"), which is a Base instance.

Regards,
Martin



From altis at semi-retired.com  Tue Feb 12 16:53:01 2002
From: altis at semi-retired.com (Kevin Altis)
Date: Tue, 12 Feb 2002 13:53:01 -0800
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com>
Message-ID: <g7ga8.58$Hk.91173@news.uswest.net>

Depending on the number of simultaneous boids you want to support, you might
need to use NumPy and do array calculations to speed up the math.

Anyway, I almost wrote a boids script using the Turtle graphics code in
PythonCard http://pythoncard.sourceforge.net/

You can have as many turtle objects as you want and move them relative to
the current turtle position with forward, back, left, right and there are
distance and heading methods to simplify your calculations towards the other
boids. Of course, you can also just do all the calculations with cartesian
coordinates if you want. There are many sample turtle scripts included with
PythonCard.

Do some google searches for boids, flocking, etc. for algorithms and example
code.

ka

"WEC7108" <WEC7108 at boeing.com> wrote in message
news:3C6968A4.1CA9C62E at boeing.com...
> I would like to develop a GUI program similar to the ever-popular BOIDS
> using Python instead of Java.  Are there URL references or books on this
> kind of development?  :o)  Bill
>




From synthespian at uol.com.br  Sun Feb 10 21:09:00 2002
From: synthespian at uol.com.br (synthespian)
Date: 11 Feb 2002 00:09:00 -0200
Subject: No explanation for weird behavior in re module!
References: <a470re$1cl30l$1@ID-78052.news.dfncis.de> <c76ff6fc.0202101814.5d18209d@posting.google.com>
Message-ID: <87pu3c7osj.fsf@uol.com.br>

sjmachin at lexicon.net (John Machin) writes:

> synthespian <synthespian at uol.com.br> wrote in message news:<a470re$1cl30l$1 at ID-78052.news.dfncis.de>...
> > Hi-
> > 
> > 	I'm really intrigued by this behavior:
> > 
> > >>> import re
> > >>> p = re.compile('^(der|die|das(\s\w+))')
> > >>> m = p.match('die T?r, T?ren')
> > >>> n = p.match('das Auto, Autos')
> > >>> m.group(0)
> >  'die'
> > >>> m.group(1)
> >  'die'
> > >>> m.group(2)
> >  [nothing!!!!]
> > >>> n.group(0)
> >  'das Auto'
> > >>> n.group(1)
> >  'das Auto'
> > >>> n.group(2)
> > 'Auto'
(snip)
> see that this matches just 'die' also.
> 
> Anticipating the next raft of problems: (1) You will need to use the
> re.UNICODE flag when you call re.compile(), otherwise \w will not
> recognise the Unicode alphabetics (this *is* documented) (2) You may
> need to give it an input whose Python type is 'unicode' -- being able
> to see the umlaut on your screen is not sufficient evidence of this
> :-) (3) You should get into the habit of using the raw string notation
> with your regexes whether it is necessary or not, else you will be
> bitten in the future.
> 
> Anyhow, the following works for me:
> 
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> >>> p = re.compile(r'^((der|die|das)(\s\w+))',re.UNICODE)
> >>> p.match('das Auto').groups()
> ('das Auto', 'das', ' Auto')
> >>> z = u'die T\u00FCr, T\u00FCren'
> >>> p.match(z).groups()
> (u'die T\xfcr', u'die', u' T\xfcr')

Hi-
	Thanks for helping. Unfortunately, the solution did not yet solve the problem at hand.
	You see, I must read lines forma a plain text file, which has the form "das Auto, Autos".
	The original program has to ask if "Auto" has "der, die or das" as the definite article.
	The problem is that I can't make Python read anything with non-ASCII character set.
	The output you have suggested, "T\xfcr", is, for all practical purposes, unreadable.
	As I understood it from the other posts, the "\w+" on the regex will depend on my locales.

	I have, in /etc/enviroment. the following set-up (native language: Portuguese):

	LANG=C
	LANGUAGE=pt_BR
	LC_ALL=pt_BR
	LC_CTYPE=latin1
	LESSCHARSET=latin1
	NLSPATH=/var/catman
	MM_CHARSET=ISO-8859-1

	Seems like non-ASCII is a real bother in Python...I'd expected this to have been better looked
after in Python-2.0...

	I appreciate any help...

	Thank you.

	H


-- 
(??.?(?*?.?     ?.?*?)?.??)
 ?.???. synthespian .???.?
(?.??(?.??*    *??.?)??.?)
	     @
	   u o l
	   c o m
	    b r


From jamesk at ifm-services.com  Mon Feb 25 03:36:51 2002
From: jamesk at ifm-services.com (James Knowles)
Date: Mon, 25 Feb 2002 01:36:51 -0700
Subject: Newbie Q: Import code at run-time?
References: <mailman.1014621160.14089.python-list@python.org>
Message-ID: <3C79F7A3.774BCD72@ifm-services.com>

> check out execfile().

This works like a charm. 

Out of curiosity, I tried the other. The module appears to load, but it
doesn't seem to import the symbols into the global namespace. 

Thank you very much for putting up with the newbie question.

-- 
Realizing that, since people will never cease trying to interfere with the
liberties of others in pursuing their own, the state can never wither away. 
- Wystan H. Auden


From phr-n2002a at nightsong.com  Sun Feb 24 23:18:53 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 24 Feb 2002 20:18:53 -0800
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu>
Message-ID: <7xd6yu42j6.fsf@ruckus.brouhaha.com>

David Eppstein <eppstein at ics.uci.edu> writes:
> > : But what is the functional equvalent of:
> > 
> > : for x in items:
> > :     x.f()
> > 
> > 
> > 
> > Here's one way to do it:
> > 
> > ###
> > map(lambda x: x.f(), items)
> > ###
> 
> I'd prefer
> [x.f() for x in items]
> 
> It's not functional syntax, but so what?

Both of those build up a new list of the results, instead of
discarding the values.  If the f function takes an integer and
computes a 20-megabyte structure, you've got a problem.  This really
calls for a generator comprehension as discussed a few weeks back.

   reduce(lambda x,y:(y.f(),0)[1], items, 0)

almost works, but invokes f an extra time at the end.

Someone see how to fix it?


From skip at pobox.com  Fri Feb  8 19:31:56 2002
From: skip at pobox.com (Skip Montanaro)
Date: Fri, 8 Feb 2002 18:31:56 -0600
Subject: [OT] PPT reader?
In-Reply-To: <a41o1m$8uv$1@panix2.panix.com>
References: <mailman.1013181915.301.python-list@python.org>
        <slrna689o0.cn1.akuchlin@ute.mems-exchange.org>
        <a41o1m$8uv$1@panix2.panix.com>
Message-ID: <15460.28156.897610.796120@12-248-41-177.client.attbi.com>

    amk> [html slides ref]

    Aahz> I tried that for OSCON 2000, only to discover that the Linux
    Aahz> version of Netscape by default could only handle a maximum 24pt
    Aahz> font size.  Was anyone who was using HTML doing so under Linux?

Yes, both Jeremy and I were.  I think Jeremy used Netscape.  I used Opera.

Skip



From nomail at laimon.moc  Wed Feb 13 19:08:24 2002
From: nomail at laimon.moc (MegaHurts)
Date: Wed, 13 Feb 2002 16:08:24 -0800
Subject: Python on Windows CE
Message-ID: <a4ev4g012jn@enews4.newsguy.com>

I've loaded the 1.5.2 Python for Windows CE
onto my MobilePro with CE 2.0.

Now, can I really write small applications with
this?

(To me, this means: Can I compile to bytecode
and run a small app which does not visibly invoke
the Python interpreter?  Or is one limited to running
text scripts with all the visual baggage?

It also means: Is there any way to use Python to
develop small, professional-looking and acting
applications on the old CE machines?)

I can find precious little on Google from folks who
have used Python on CE.  Any other throwbacks like
me out there?

Thanks




From jason at jorendorff.com  Fri Feb 22 13:58:22 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 22 Feb 2002 12:58:22 -0600
Subject: Python Gotcha with Octal Numbers
In-Reply-To: <4Jud8.1483$N7.287257@ruti.visi.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEIKDEAA.jason@jorendorff.com>

Grant Edwards wrote:
> I just got bit by the leading-zero thing again about 10 minutes
> ago when I changed the minor version number of something from 07
> to 08.  Had to skip from 1.07 to 1.10...

Serves you right.  Version numbers like 1.07 are silly.
1.0.7 or 1.7 would be more sensible.

A tuple is particularly swell for version numbers.
See sys.version_info.

## Jason Orendorff    http://www.jorendorff.com/




From starx at pacbell.net  Mon Feb 11 16:46:51 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 11 Feb 2002 21:46:51 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <3C682A16.1D6C800D@ccvcorp.com>
Message-ID: <Xns91B28C2B6EBE0RASXnewsDFE1@130.133.1.4>

Jeff Shannon || Mon 11 Feb 2002 12:31:18p:

> 
> 
> Jeff Hinrichs wrote:
> 
>> I took it as a unintended play on Sandanista, which is a group of 
>> "freedom fighting" rebels in Nicaragua. 
> 
> cf. also Zapatistas, Peronistas, etc... it's common for followers of a
> specific leader to be referred to thusly, just as in English we have
> Marxists, Leninists, Maoists, etc...  nothing specific about the
> "-ista" suffix having anything to do with "freedom
> fighters"/"revolutionaries"/"terrorists" ... 
> 
> (Though using Python *could* perhaps be equated to "freedom fighting"
> in other ways...  ;) )

Then it becomes 'Guidoista' perhaps, for following a specific leader? ;)
-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From jason at jorendorff.com  Wed Feb  6 12:19:27 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 11:19:27 -0600
Subject: Problem while uploading a binary file using cgi module
In-Reply-To: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU>
Message-ID: <HFEKILOLEFEFMKAECNDLKEMDDAAA.jason@jorendorff.com>

J. Wang writes:
> import cgi
> 
> print 'Content-type: text/html\n'
> form = cgi.FieldStorage()
> if not form:
>     [...]
> elif form.has_key("filename"):
> 	item = form["filename"]
> 	if item.file:
> 		while 1:
> 			data = item.file.readline()
> 			fp = file('1.jpg','wb')
> 			if data:
> 				fp.write(data)
> 			else:
> 				break
> 			fp.close()

Every time through the loop you are completely overwriting your file.

Do this instead:

elif form.has_key("filename"):
    item = form["filename"]
    if item.file:
        fp = file('1.jpg','wb')
        while 1:
            data = item.file.readline()
            if data:
                fp.write(data)
            else:
                break
        fp.close()

## Jason Orendorff    http://www.jorendorff.com/



From akuchlin at ute.mems-exchange.org  Sun Feb 10 09:18:29 2002
From: akuchlin at ute.mems-exchange.org (A.M. Kuchling)
Date: 10 Feb 2002 14:18:29 GMT
Subject: box on soap
References: <mailman.1013283562.647.python-list@python.org>
Message-ID: <slrna6d095.sc5.akuchlin@ute.mems-exchange.org>

In article <mailman.1013283562.647.python-list at python.org>, 
	Mark McEahern quoted:
> publication. Luckily, the W3C just launched a WSDL working group and it
> looks like the community at large has the will to clean it up, ...

<splorf> Since when has a W3C working group, or any committee for that
matter, cleaned something up and simplified it as opposed to making
it even more baroque and complicated?

--amk                                                  (www.amk.ca)
You cannot speak as though reality is a one-dimensional concept.
    -- The Valeyard, in "The Ultimate Foe"


From laz at strakt.com  Mon Feb  4 04:54:32 2002
From: laz at strakt.com (Fredrik Juhlin)
Date: Mon, 4 Feb 2002 10:54:32 +0100
Subject: ICU wrapper for Python?
Message-ID: <20020204095432.GA3733@strakt.com>

Hello,

I was fiddling around with a Python wrapper for the ICU libraries this
weekend, when it occured to me to check if anyone else is/has been working
on the same thing. I looked around a bit on the net but couldn't find
anything, so I figured I'd drop a question here too :)

I started working on it to see if it was a feasible way to get collation
to work with Unicode. It seems to work OK, although it's a bit rough
around the edges (to be mild :)

If you don't know what ICU is (and care to find out), check out:
http://oss.software.ibm.com/icu/

//FJ



From mdk at mdk.com  Wed Feb 13 16:42:05 2002
From: mdk at mdk.com (MDK)
Date: Wed, 13 Feb 2002 16:42:05 -0500
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de> <7512146ADFC7C5D2.12C740DBCB3315BF.2E2B43736E1AC6BE@lp.airnews.net>
Message-ID: <a4emjd$1e4u4v$1@ID-98166.news.dfncis.de>

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:7512146ADFC7C5D2.12C740DBCB3315BF.2E2B43736E1AC6BE at lp.airnews.net...
> In article <a4c2ge$1dognu$1 at ID-98166.news.dfncis.de>, MDK <mdk at mdk.com>
wrote:
> >In PythonWin I entered this script from a book:
> >
> >from Tkinter import Label
> >widget = Label(None, text='Hello GUI world!')
> >widget.pack()
> >widget.mainloop()
> >
> >When I run it in PythonWin it runs fine.
> >
> >When I double click on it in Windows explorer I get a message that
tcl80.dll
> >can not be found.  That file does not exist on my PC.  How then is it
> >running under PythonWin if that file does not exists and how can I make
it
> >run when double-clicked?
> >
> >My path includes:
> >
>
>PATH=C:\Python22\;c:\python22\dlls\;C:\Python22\Lib;C:\Python22\Lib\lib-tk
> >
> >Thanks.
> >
> >
>
> What name does the script have?  Did you call it
> something like
>   mytest.tcl
> ?
> --
>
> Cameron Laird <Cameron at Lairds.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html

I called it GUI_test.py





From piet at cs.uu.nl  Fri Feb  1 05:51:05 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 01 Feb 2002 11:51:05 +0100
Subject: PyAppBar -- bug fix, nifty widget
References: <7OP58.4001$KC5.2845539457@newssvr14.news.prodigy.com> <3C5804B4.5597CF4C@computer.org> <WVZ58.4176$C27.2949742694@newssvr14.news.prodigy.com> <kL_58.4198$mn1.2958984294@newssvr14.news.prodigy.com> <3c599fe0.14487672@127.0.0.1> <R_j68.5701$lS4.3236040866@newssvr14.news.prodigy.com> <mailman.1012519205.28575.python-list@python.org>
Message-ID: <wz4rl18oh2.fsf@nono.cs.uu.nl>

>>>>> Cliff Wells <logiplexsoftware at earthlink.net> (CW) writes:

CW> IE is definitely one of the better browsers.  I'll give it a try as soon as
CW> it runs on one of the better OS's ;)

It already does. (Mac OSX)
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From michael.amrhein at t-online.de  Thu Feb  7 12:40:00 2002
From: michael.amrhein at t-online.de (Michael Amrhein)
Date: Thu, 7 Feb 2002 18:40:00 +0100
Subject: FixedPoint
References: <mailman.1013070672.23308.python-list@python.org>
Message-ID: <a3ue3t$c5o$01$1@news.t-online.com>

"Mark McEahern" <mark at mceahern.com> schrieb im Newsbeitrag
news:mailman.1013070672.23308.python-list at python.org...
> In the FixedPoint module found here:
>
>
>
ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoin
> t.py.Z
>
> there seems to be a bug comparing to None like this:
>
>   if f == None:
>
> Demo:
>
>     $ python
>     Python 2.2 (#1, Dec 31 2001, 15:21:18)
>     [GCC 2.95.3-5 (cygwin special)] on cygwin
>     Type "help", "copyright", "credits" or "license" for more information.
>     >>> import FixedPoint
>     >>> f = FixedPoint.FixedPoint(1.0)
>     >>> if f == None:
>     ...     print "that's not true!"
>     ...
>     Traceback (most recent call last):
>       File "<stdin>", line 1, in ?
>       File "FixedPoint.py", line 272, in __cmp__
>         xn, yn, p = _norm(self, other)
>       File "FixedPoint.py", line 415, in _norm
>         y = FixedPoint(y, x.p)
>       File "FixedPoint.py", line 216, in __init__
>         raise TypeError("can't convert to FixedPoint: " + `value`)
>     TypeError: can't convert to FixedPoint: None
>
> This doesn't happen if you use either of the following:
>
> if f is None:
> ...
>
> if not f:
> ...
>
> I ran into this because PyPgSql (v2.0) uses the first method:
>
>     File "/usr/lib/python2.2/site-packages/pyPgSQL/PgSQL.py", line 1434,
in
>   _quote
>
>       if value == None:
>
> I'm not sure what the best fix is here but one that works is to modify the
> __cmp__ method for FixedPoint to anticipate comparison to None.  Here's
the
> current __cmp__:
>
>     def __cmp__(self, other):
>         xn, yn, p = _norm(self, other)
>         return cmp(xn, yn)
>
> The fix would look like this:
>
>     def __cmp__(self, other):
>         ## To avoid crashing on if x == None and the like:
>         if other is None:
>             return 1
>         xn, yn, p = _norm(self, other)
>         return cmp(xn, yn)
>
> Or perhaps it'd be better to muck with __eq__ and __ne__?  Well, anyway,
> that mod to __cmp__ works.
>
> Cheers,
>
> // mark
>

Hi Mark,
your patch fixes the problem and makes FixedPoint more compatible with the
built-in numeric types.
However, as a hint to the author(s) of PyPgSQL: it's always more safe to use
'if x is None' to test if x is None ;-)

There is another bug in FixedPoint:

>>> FixedPoint(3, 0) / FixedPoint(2, 0)
FixedPoint('2.', 0)
>>> FixedPoint(5, 0) / FixedPoint(2, 0)
FixedPoint('2.', 0)

This is due to a bug in the helper routine _roundquotient.
A fixed and more general (negative quotient) version would look like this:
def _roundquotient(x, y):
    i, r = divmod(x, y)
    c = cmp(abs(2*r), abs(y))
    # r > y/2 or r = y/2 and i not negative:
    if c > 0 or (c == 0 and i >= 0):
        return i + 1
    else:
        return i

Note that this will break the test routine because there's a corresponding
error.
You will have to change
    assert o == fp(".998", 10)
into
    assert o == fp(".999", 10)
two times.

Cheers,
Michael

P.S.: I've written a simular class named 'fixed' with portions adopted from
Tim's FixedPoint, but a slightly different constructor (no default
precision; precision calculated from given value, if no precision given) and
some additional functions. I'll post it in response to your original
request.







From BPettersen at NAREX.com  Wed Feb 27 16:00:11 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Wed, 27 Feb 2002 14:00:11 -0700
Subject: Printing to a "log screen" in a multi-module QT app
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158D45@admin56.narex.com>

> From: Anders Schneiderman [mailto:schneida at seiu.org] 
> 
> I'm writing an application using QT.  Right now when I want 
> to use the equivalent of "print", I call a method called 
> "log" that's defined in the main class in my main module:
> 
> def log(self, msg):
>    try:
>       self.logArea.insertline(msg, -1)
>    except:
>       self.logArea.insertline(str(msg), -1)
> 
> To call it in other classes that are in other modules of the 
> app, I pass "log" to the new class.  For ex:
> 
> class CodeEditor:
>    def __init__ (self, log, ...)
>        self.log = log
> 
> Is there a cleaner way to do this?  For example, is there a 
> way to redefine the "print" command?  Whatever solution I 
> come up with, the command to print has to be short;  
> "self.logObject.log('File successfully updated')" is too much 
> of a pain.  Any suggestions?

You could do something like:

  class Logger:
      def __init__(self, logArea):
          self.logArea = logArea

      def write(self, msg):
          try:
             self.logArea.insertline(msg, -1)
          except:
             self.logArea.insertline(str(msg), -1)

Then you can pass an instance of the Logger class around and simply do:

   print >> logger, 'hello world'

-- bjorn



From sholden at holdenweb.com  Fri Feb  1 18:32:40 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 18:32:40 -0500
Subject: comparing DateTime instances
References: <mailman.1012602663.18405.python-list@python.org>
Message-ID: <XCF68.111054$8b.4223878@atlpnn01.usenetserver.com>

"Mark McEahern" <mark at mceahern.com> wrote in message
news:mailman.1012602663.18405.python-list at python.org...
> I think I just need a sanity check.  I'm trying to compare DateTime
> instances (from the eGenix mx.DateTime package).
>
> $ python
> Python 2.2 (#1, Dec 31 2001, 15:21:18)
> [GCC 2.95.3-5 (cygwin special)] on cygwin
> Type "help", "copyright", "credits" or "license" for more information.
> >>> expected = "2003-01-31 14:13:13.65"
> >>> actual = "2003-01-31 14:13:13.65"
> >>> from mx.DateTime import DateFrom, cmp
> >>> e = DateFrom(expected)
> >>> a = DateFrom(actual)
> >>> cmp(a, e)
> 0
>
> I think this is telling me that a and e are equal.  I'm still trying to
get
> my brain to instantly recognize the seemingly flipped around semantics of
> cmp.  I understand that __cmp__ returns 0 for equality--and that 1 and -1
> are used to indicate less-than, greater-than ordering.  So too for cmp,
eh?
>
It is indeed telling you the DateTime values are equal. Did you think of
trying the equality operator? The point about __cmp__ is that [absent rich
comparisons] implementing it tends to make objects do the Right Thing when
queried by the usual comparison operators. So you would also discover:

>>> e < a
0
>>> e == a
1
>>> e > a
0

> So that would mean, if I were comparing two DateTime instances in a unit
> test and I wanted the test to fail if the dates were NOT equal, I might
say
> something like this:
>
> self.failIf(cmp(date1, date2))
>
Well, you might. Bit you should at least consider as an alternative
something like

self.failIf(date1 != date2)

> If date1 equals date2 (I know I can use accuracy to account for
fuzziness),
> cmp will return 0, which is the same as false, and therefore failIf passes
> because it only fails if the argument is true.
>
> The documentation for mx.DateTime's cmp seems to assume the reader
> understands __cmp__.

Well, cmp is actually just an interface to __cmp__, so that's probably true.
But the foremost thing to understand about cmp() is that it often isn't
encessary to use it!

> This is all we find there:
>
> "cmp(obj1,obj2,accuracy=0.0)
> Compares two DateTime[Delta] objects.
> If accuracy is given, then equality will result in case the absolute
> difference between the two values is less than or equal to accuracy."
>
> Perhaps this could expand on the actual set of results:
>
> -1 : obj1 is less than obj2
>  0 : obj1 is equal to obj2
>  1 : obj1 is greater than obj2
>
> It wouldn't seem to hurt and it might avoid someone else wondering the
same
> silly things I'm wondering.
>
It would be even nicer if it suggested that you could simply compare
DateTime objects like numerical values, which in fact you can. Then
intelligent newcomers such as yourself wouldn't find themselves
contemplating the bowels of the language when all that's required is a
straight smack on the head! I suspect you are going to enjoy Python.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, "Python Web Programming": http://pydish.holdenweb.com/pwp/







From phr-n2002a at nightsong.com  Thu Feb 21 03:13:14 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 00:13:14 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de>
Message-ID: <7xpu2zdzhh.fsf@ruckus.brouhaha.com>

"Karl M. Syring" <syring at email.com> writes:
> But this is really dilettantic work and NASA today is an organization run by
> amateurs. If you look at real computational geometry programs like CGAL, you
> can see, how it is done.
> Anyway, it is not necessary to appeal to authorities to get a grasp of the
> problem.: Imagine, you have a polygon in the plane and you do a series of
> rotations and translations on it. Now, you want to know whether the position
> is exactly identical to another polygon. If you do this with floating point,
> you will never get an exact result, while in rational coordinates you always
> will.

Physical simulations and computational geometry don't have much in common.


From claird at starbase.neosoft.com  Fri Feb 15 18:44:00 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Feb 2002 17:44:00 -0600
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net> <20020215225431.7bdef34e.risnes@alfanett.no>
Message-ID: <A7E1D67D0B325B81.DFE9DE5C58C8BD56.EBE9D6BF2EA2B597@lp.airnews.net>

In article <20020215225431.7bdef34e.risnes at alfanett.no>,
Andr? Larsen Risnes  <risnes at alfanett.no> wrote:
>Artur Skura <arturs at iidea.pl> wrote:
>[...]
>>
>> Now can you suggest an easy, lazy way of doing this (with any
>> toolkit) so things don't pile up and everything looks nice?
>> 
>Sure (with tkinter):
			.  .  .
>for part in text.split("%"):
>    if part == "text":
>        Text(root, width=8, height=1).pack(side=LEFT)
>    else:
>        Label(root, text=part).pack(side=LEFT)
			.  .  .
Seconded.  This is precisely the attraction of
Tkinter's "pack".  People accustomed to use of
Tkinter in this way are aghast at the difficulties
VB-style programmers have with this task.
-- 

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


From Peter.Milliken at GTECH.COM  Mon Feb 18 19:59:57 2002
From: Peter.Milliken at GTECH.COM (Milliken, Peter)
Date: Mon, 18 Feb 2002 19:59:57 -0500
Subject: Adding static typing to Python
Message-ID: <0DFA7478F1E5D311B90900508B1208CF565712@ausmail02.mis.gtech.com>

You have drawn a false assumption there Brian, you assume that 95% of my
bugs are due to misnamed variables - I wished! :-) That wasn't what I was
trying to imply or state with that statement.

Strong typing requires that you design before you code (Extreme Programmers
beware! :-)) - therefore by the time a clean compile is achieved 95% of the
bugs *are* out :-) That is one of the primary advantages to strong typing,
it *forces* you to do the design up front otherwise you will never get a
clean compile.

You should try it one day :-)

Peter

-----Original Message-----
From: Brian Quinlan [mailto:brian at sweetapp.com]
Sent: Tuesday, February 19, 2002 12:04 PM
To: 'Peter Milliken'; python-list at python.org
Subject: RE: Adding static typing to Python


Peter Milliken wrote:
> When I compile a Ada program cleanly I know that 95% of the bugs are
out
> of it - before I even start testing! :-) Of course, it takes me longer

> to write up front but overall development effort (in my subjective
opinion > - since I have no metrics :-)) is that it is much lower.

If 95% of your bugs are of such a trivial nature, then you are either
very sloppy in your code production, you are an absolutely brilliant
designer or you are working on simple problems.

I would say that 99% of my bugs are due to design mistakes (e.g.
inappropriate algorithm selection, poorly designed interface) or
semantically incorrect algorithms (e.g. boundary errors, incorrectly
identified complexity/memory usage).

And finding these errors is substantially more difficult than finding
misnamed variables.

Cheers,
Brian



From hans at inenco.no  Tue Feb 12 10:04:40 2002
From: hans at inenco.no (Hans Kristian Ruud)
Date: Tue, 12 Feb 2002 15:04:40 GMT
Subject: Drawing in "Slow motion".
References: <3C692698.C9DAD820@inenco.no> <a4ba6i$14bg$1@norfair.nerim.net>
Message-ID: <3C692F69.6402555B@inenco.no>


Eric Brunel wrote:

> Hi Hans,
>
>
> Two solutions come to my mind:
> - keep your time.sleep calls, but call the update_idletasks method on the
> canvas after each sleep. The call should redraw the screen.
> - a bit trickier: use the delayed callback facility in Tk via the "after"
> method to draw while your application is running. This is a bit more
> complicated and you would certainly have to rewrite your code differently.
> But it's the only way to have a "reacting" application while the drawing
> takes place (i.e. user is able to click on buttons, call menus, etc...).
> Here is a simple example, just drawing random lines:
>
>

Thanx a lot; these solutions were very helpful.

>
> HTH
>  - eric -

- hans kristian -





From martin at v.loewis.de  Wed Feb 20 17:44:02 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 23:44:02 +0100
Subject: No space left on device
References: <mailman.1014239691.10707.python-list@python.org>
Message-ID: <m3pu2zixjh.fsf@mira.informatik.hu-berlin.de>

sag at hydrosphere.com (Sue Giller) writes:

> I have a program that generates a LOT of small files in a 
> Windows2000 subdirectory.  It is my understanding (often faulty) 
> that there is no limit to the number of files such a subdirectory can 
> contain.  However, I am getting an error when I hit about 21,843 
> files.

What file system? I believe there *is* a limit to the number of files
you can have in a directory on a FAT32 file system. There is also
actually a limit to the number of files you can have in a directory
residing on an NTFS volume, but that is much larger.

IOW, there is no file system in wide use that is without limits.

> I have over 5 G of space on the drive, and the test files are 0 len 
> anyway.  (It still happens if the files are not 0 len)

I think it maxed out the directory size, then.

> I can then create additional files in that directory from within a DOS 
> window, but I cannot rerun my file generator to make even one more 
> file.   So I am wondering if the limit comes from Windows, from 
> Python 'open...' or some other innards of Python.

It should be a Windows limit; Python is passing the request straigth
to Windows. My guess is that there is a maximum directory size,
counting the number of bytes that the directory consumes on disk; I
would not be surprised if this is 2GB on a FAT32 system (or perhaps
4GB). You may need to turn off 8.3 file name generation (ask MSDN for
details), to reduce the space consumed by a single directory entry.
You may also try to use shorter file names.

Regards,
Martin


From martin at v.loewis.de  Tue Feb 26 17:15:01 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 23:15:01 +0100
Subject: what I would like to see in python to make a better "glue" language (Newbie)
References: <899f842.0202261325.6ff3dd54@posting.google.com>
Message-ID: <m3sn7nubyy.fsf@mira.informatik.hu-berlin.de>

anthony_barker at hotmail.com (Anthony_Barker) writes:

> Very Important:
> 1. LDAP module should be included in the base distro. Active Directory
> is already available via com (nice)

Adding modules is typically a two-stage process. First, a specific
module should establish itself as the monopoly implementation on a
certain field. Then, it can be considered for inclusion in the
standard library.

If you have any specific LDAP module in mind, I'd expect, as a
prerequisite, that it is a distutils package. That should make
installation very easy, and open doors to inclusion in "foreign"
distributions. E.g. Linux distributors may include it. Also, there
always was the plan of a Sumo distribution which includes all packages
that atleast build successfully - volunteers to make this come true
are still needed.

> 2. DNS really should be included in the base library, I emailed
> Anthony Baxter and he replied, saying it was almost done. 

I completely agree.

> 3. Telnet Library is great - including an SSH module would be a nice
> addition.

This is not trivial. To my knowledge, there is no SSH library that you
could link. I don't know how difficult it is to implement a SSH module
on top of OpenSSL.

> 4. Include an SNMP implementation, there are several around, but a
> pure python one as a library would be nice

All such libraries are quite large, right? If so, chances for
inclusion in the standard library are low, since I expect the user
community for such a library is quite small. Nevertheless, bug the
maintainers of those libraries for distutils support.

Regards,
Martin


From syring at email.com  Mon Feb  4 11:48:59 2002
From: syring at email.com (Karl M. Syring)
Date: Mon, 4 Feb 2002 17:48:59 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <13285ea2.0202030404.5dfe9b66@posting.google.com> <a3j9du$181gf1$1@ID-7529.news.dfncis.de> <13285ea2.0202031557.244bb70d@posting.google.com> <mailman.1012787226.16823.python-list@python.org> <a3lo1p$18b48u$1@ID-7529.news.dfncis.de>
Message-ID: <a3me38$19ig5h$1@ID-7529.news.dfncis.de>

"Karl M. Syring" <syring at email.com> schrieb im Newsbeitrag
news:a3lo1p$18b48u$1 at ID-7529.news.dfncis.de...
> "Gerhard H?ring" <gh_pythonlist at gmx.de> schrieb
> > Le 03/02/02 ? 15:57, J.Jacob ?crivit:
> > > [Karl M. Syring]>
> > > > This is utter nonsense. Pdcurses simply works on Windows.
> >
> > True, but this doesn't say anything about building the Python curses
> > module againsat it.
> >
> > In fact, I've tried and I can tell you that this isn't plug-and-play.
> > After seeing lots of undefined symbols, I gave up. It simply wasn't
> > worth the effort to me.
> >
> > But you seem to be pretty sure that it's pretty easy, so ...
> >
> > > Maybe -you- can get it to work?
> >
> > ... I'd also be interested if you get this to work.
>
> I will give it a try, time to udate to 2.2.

More difficult than I thought. It compiles with PDcurses 2.5, but the
following functions are not implemented by PDCurses:

_wresize
_getattrs
_mvwgetnstr
_putwin
_filter
_noqiflush
_getwin
_has_key
_setupterm
_putp
_qiflush
_tigetflag
_tigetnum
_tigetstr
_tparm
_use_env

I think, most of them are imported from term.h

Karl M. Syring



From DeepBleu at DeepBleu.org  Thu Feb 14 15:45:59 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Thu, 14 Feb 2002 14:45:59 -0600
Subject: zlib.so and libz.so and Python
References: <u6niisr909gtd6@corp.supernews.com>        <mailman.1013703624.30857.python-list@python.org>        <u6nuqr108os0ae@corp.supernews.com> <mailman.1013714244.25257.python-list@python.org>
Message-ID: <u6o8ii2horrvf7@corp.supernews.com>

Thanks Skip
Right after I posted my second message I started looking and it is part of
zlib package.  Apprently, zlib is included now with Sparc 7 and 8 but not
with 2.6.  So of course libz.so is not on the machine that runs sparc 2.6
Thanks again,
DeepBleu

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.1013714244.25257.python-list at python.org...
>
>     DeepBleu> Where do I get the libz.so from?
>
> http://www.gzip.org/zlib/
>
> Cheers,
>
> Skip
>




From seeger at sitewaerts.de  Fri Feb 22 06:33:51 2002
From: seeger at sitewaerts.de (Felix Seeger)
Date: Fri, 22 Feb 2002 12:33:51 +0100
Subject: running system command in cgi problem
Message-ID: <a55a6d$4orlk$1@ID-79019.news.dfncis.de>

Hi

I've writen a small cgi/python script that starts a command on the server.
This command may take about 30 minutes to finish. So my browser becomes a 
timeout.

Is there any way to run the command, and while it works send the browser 
every 30 secs a please wait string + refresh.

If the command is finnished send the browser the output from command.

P.S. I'm new in programming and I think threads are too hard for me.
P.P.S Are there any small programs which do this, so I can look at the 
source and see how it works.



thanks
have fun
HAL


From sdm7g at Virginia.EDU  Tue Feb 19 12:55:35 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 12:55:35 -0500 (EST)
Subject: confusion regarding os.path.walk()
In-Reply-To: <20020218220610.167f71d5.ron.l.johnson@cox.net>
Message-ID: <Pine.OSX.4.43.0202191250520.1674-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, Ron Johnson wrote:

>
> Thinking that that might be the problem, I used os.path.realpath().
> However, even that didn't work.  Why wouldn't testing an absolute
> filename work?
>

Testing an absolute filename *does* work.

If you print out what os.path.realpath() returns, you'll see that
it's returning an absolute path to a file that sometimes doesn't
exist. It isn't a DWIM function: os.path.realpath( 'XXXX' )
is just going to return "XXXX" appended onto the current directory
path. You still need to to explicitly join the directory path.

-- Steve Majewski






From carifio.nospam at nospam.usys.com  Fri Feb  8 13:06:10 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Fri, 08 Feb 2002 18:06:10 GMT
Subject: What's the equivalent in Python for Perl's CPAN?
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com> <a3v903$kgk$1@panix2.panix.com>
Message-ID: <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>

> That's an impressive post.  Try a Google for Parnassus.

I'm aware of Parnassus, which appears to be a "jumpstation" to various
modules.
Certainly a centralized directory has value, but it also forces me to
manually fetch
code in various formats as .py files, .tgz files or even in "Distutil
format" (which I'm
reading about).

CPAN has a mirrored repository where the modules all appear to be packaged
in the
same format, perhaps "bundled together" where it makes sense. And there's a
CPAN
client side (a Perl module) that can fetch stuff from a named repository and
install it
locally. This includes chasing module dependencies and installing those
also.
It also includes knowing which modules have already been downloaded and
installed.

The closest thing I seem to be able to find is ActiveState's Package Manager
(http://www.activestate.com/Products/ActivePython/more_information.plex).
Is this what I should use or is something else preferred?

So, what's the equivalent in Python for Perl's CPAN?







From gcordova at hebmex.com  Tue Feb 12 19:09:54 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 12 Feb 2002 18:09:54 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A54@SRM201002>

> >It's common for little children to mistakenly use "-isto" as well as
> >"-ista" when talking about boys or girls, that's why it's funny, 
> >because it's an infantile mistake.
> 
> Is -ista always feminine? E.g. is it always _la dentista_ not 
> _el dentista_?

It's used both ways: "El Dentista" for a male dentist, "La Dentista"
for a female dentist.

On the other hand, "Doctor", can have male and female usage:
	El Doctor
	La Doctora

But, there's other words, like "radio", which seems male but it's
actually a female subject: "La radio".

And also "agua" which seems female, but it's a male subject: "el agua".

-gustavo



From loewis at informatik.hu-berlin.de  Thu Feb 28 10:18:18 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 16:18:18 +0100
Subject: Trouble writing custom text codec
References: <9ac02e81.0202280228.5161a43c@posting.google.com>
Message-ID: <j4n0xtfxdx.fsf@informatik.hu-berlin.de>

jhorneman at pobox.com (Jurie Horneman) writes:

> Why does my custom text codec, which I have based on an existing,
> working codec as recommended by the documentation, work with
> codecs.open() but not with MyString.decode(), where as the built-in
> codecs do?

Can you report the return value of codecs.open? It looks like you are
returning the wrong type of things.

Regards,
Martin



From starx at pacbell.net  Tue Feb  5 23:28:15 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 6 Feb 2002 04:28:15 GMT
Subject: To all Higher Level....A True Beginner
References: <mailman.1012966280.4535.python-list@python.org>
Message-ID: <Xns91ACD03DB48A4RASXnewsDFE1@130.133.1.4>

Stephen L Arnold || Tue 05 Feb 2002 07:30:51p:

> On 5 Feb 02, at 18:21, David Pham wrote:
>> [Asking for intro material]
>
> Not surprising...  But this topic rightly belongs on the python-
> tutor list, rather than the regular python list.
> 
> [Go to www.python.com, etc]

Well, he was posting though google to comp.lang.python ... which is the only 
python newsgroup even if there are more specific lists. Though it must not be 
obvoius who's using usenet from the email system... over here, mail-list 
people's responses don't seem to thread the same way as normal messages.

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From ullrich at math.okstate.edu  Mon Feb 11 13:51:28 2002
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 11 Feb 2002 18:51:28 GMT
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <mailman.1013382318.2151.python-list@python.org>
Message-ID: <3c68118e.3802800336@news>

On Sun, 10 Feb 2002 18:05:01 -0500, "Tim Peters" <tim.one at home.com>
wrote:

>[Bengt Richter]
>>  >>> 'abc'.split('')
>>  Traceback (most recent call last):
>>    File "<stdin>", line 1, in ?
>>  ValueError: empty separator
>>
>> Wouldn't it make sense to return list('abc') ?
>
>[Neil Schemenauer]
>> It would also make sense to return ['a', 'b', 'c'].
>
>Well, that's what list('abc') does return, so you're in violent agreement.
>However, this still stands:
>
>> Since it's not obvious what you want Python raises an error.
>
>Indeed, my first thought was "OK, if it has to return *something*, then
>since we're asking it to split on nothing, it shouldn't split at all:
>
>    ['abc']
>
>is what it should return." 

Makes more sense to me for it to be an error - deciding
what it "should" return seems like deciding what 1/0
should return.

I'm suprised nobody's said that what it should return
is an infinite list of empty strings (which would be
inconvenient, so an error is better). That seems to
me to be "obviously" what it "really is" - when I
think about how to write a "split" in the "obvious"
way that's what comes out:

def split(data, sep):
  res = []
  pos = 0
  tok = ''
  while pos < len(data):
    if data[pos:pos+len(sep)] == sep:
      res.append(tok)
      tok = ''
      pos = pos + len(sep)
    else:
      tok = tok + data[pos]
      pos = pos + 1
  res.append(tok)
  return res

I don't know what re.split does. What's "natural"
code for a split function that returns something
reasonable when the separator is ''? Without
cheating.

> That's also compatible with
>
>    ''.join[['abc']] == 'abc'
>
>People who think '' should split "everywhere" instead of "nowhere" should
>really be arguing for 'abc'.split('') to return
>
>    ['', 'a', 'b', 'c', '']
>
>instead, since in any sense that '' could be said to "match", it matches at
>4 slice positions in 'abc', not 2.
>
>


David C. Ullrich


From jkraska at san.rr.com  Mon Feb 11 12:41:35 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 11 Feb 2002 17:41:35 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net>
Message-ID: <mt0g6ukklms8q21qsrb2kkjoohblgq3bn8@4ax.com>

>    "People tend to hold overly favorable views of their
>    abilities in many social and intellectual domains. The
>    authors suggest that this overestimation occurs, in
>    part, because people who are unskilled in these
>    domains suffer a dual burden: Not only do these
>    people reach erroneous conclusions and make un-
>    fortunate choices, but their incompetence robs
>    them of the metacognitive ability to realize it"

It's a wise enough observation. The concept of
both individual and group assessment has been
extensively studied. As it turns out, feeling one
is better than other people on average is a normal,
healthy aspect of human psychology. Only depressed
people routinely rank themselves as average. Which
is kind of funny; it appears that a healthy state
of mind is somewhat delusional in this particular way.

With all that said, I know only one thing for sure:

I FUCKING ROCK AND YOU ALL SUCK!

:) :)

C//



From gh_pythonlist at gmx.de  Mon Feb 18 16:18:54 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 22:18:54 +0100
Subject: Newbie question...
In-Reply-To: <3C716B27.3090408@bc90021redundantdot.net>
References: <3C716B27.3090408@bc90021redundantdot.net>
Message-ID: <20020218211853.GB9549@lilith.hqd-internal>

Le 18/02/02 ? 21:00, bc90021 ?crivit:
> X-No-Archive: Yes
> Hi all,
> 
> Sorry if this has been asked a thousand times already... I tried 
> searching google with very little luck.  I'm a newbie to Python, but not 
> to Linux.  (Though of course, RedHat has to make everything difficult 
> with their RPMS and such... ;)  )
> 
> I am trying to install Python 2.2 RPMS on a RedHat 7.2 system.  I can 
> get the docs and the devel packages to install, as there are no 
> dependencies for those.  However, when I try and install the Python 2.2 
> rpm itself, I get the following:
> 
> error: failed dependencies:
> 	rpmlib(PartialHardlinkSets) <= 4.0.4-1 is needed by python-2.2-7
> 	libcrypto.so.3   is needed by python-2.2-7
> 	libdb-4.0.so   is needed by python-2.2-7
> 	libssl.so.3   is needed by python-2.2-7

I had more luck recompiling the source rpms with rpm --rebuild.

> I have been unable to find out what rpmlib is or what the 
> "PartialHardLinkSets" are.

No idea.

> Libcrypto.so.3 and libssl.so.3 are both related to OpenSSL, but as far 
> as I can find on RPMFind.net, there is nothing above libcrypto.so.2 and 
> libssl.so.2.*

Creating the necessary symlinks might help.

> * These are really just symlinks, I learned from searching google. 
> Creating symlinks that point to the same thing with .so.3 didn't work.
> 
> I also have no idea what libdb-4.0.so is...

Berkeley DB 4.0. These Python 2.2 RPMs seem to be built against some
pretty recent libraries btw.

> [...] 
> If I get frustrated enough with this, I will just compile from source so 
> that I can run my Python programs.

That's usually easiest. Easier than the DLL^w RPM dependency hell for
sure. 

> However, since I uninstalled all the old Python stuff, all of my
> RedHat tools are broken.

I'd reinstall Python 1.5.2 from the Redhat installation media.

> I figure it will be easier to fix the Python dependencies than that
> RedHat tools,

I've had enough of Redhat once after approx. 10 minutes, so I'm not in a
position to judge, but I think it's better to not try to mess with its
Python 1.5.2. Look into the Python 2.2 SRPM - you can build it both
ways, as "primary" Python or as a "secondary" Python. The "primary"
version creates the /usr/bin/python symlink. The secondary version does
not.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.2 ?C      Wind: 3.2 m/s



From starx at pacbell.net  Sat Feb 16 19:38:01 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 17 Feb 2002 00:38:01 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <Xns91B7A92CBB55ARASXnewsDFE1@130.133.1.4>

Robin Munn || Fri 15 Feb 2002 10:59:13a: 

> [Lots of stuff] 
> 
> * Batteries included. This is the big one, which makes Python more
> than just a toy language suitable for learning. The richness of the
> standard library makes Python incredibly useful, and I for one would
> like to see it continue to grow. I understand the concern that
> people developing Python-based programs for distribution would like
> their users to be able to run a 100k script without downloading a
> 10M distribution, but I don't want to see that cause a reduction in
> the standard library. Instead, we could make make two Python
> distributions, a "full" install containing everything and a "lite"
> install containing only the interpreter and maybe -- *maybe* -- a
> couple of essential modules like os and sys. Then people developing
> Python programs like, say, a pygame-based game, could distribute a
> single package containing the "Python lite" distribution, their own
> code, and any other modules their code depended on -- all in one
> easy-for-the-end-used-to-install package.

Well, using Installer, i've got something a little like this 'lite' 
design you're talking about. I've got a exe that contains all of the 
modules and dlls and stuff I need (4 megs with onefile), and a script 
which runs the real script via execfile. So i can distribute my script 
app, and say 'either get python 2.2, pyopengl+context, wxpython, 
numeric, pil, and maybe some other stuff, or download this exe, put it 
in the app directory, and run IT'. It can also read a config file to get 
the name of the file to exec, so you don't have to recompile it to 
change that.

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com 
"and-i'll-prolly-put-something-that-uses-this-up-later-today... GACK, 
been reading this group to long =)"


From v.wehren at home.nl  Sun Feb 24 15:01:57 2002
From: v.wehren at home.nl (v.wehren)
Date: Sun, 24 Feb 2002 21:01:57 +0100
Subject: .exe files
References: <jpae8.66140$X64.23604369@news1.rdc2.pa.home.com>
Message-ID: <vtbe8.27305$Wm3.1515495@zwoll1.home.nl>

Search starship Python for the PY2EXE distribution  and download/install it.

If you in addition use the following script A. Semenov that imports
py2exe.py, compiling your code becomes even more painless:

from distutils.core import setup
import sys, os, py2exe

name = sys.argv[1]
sys.argv[1] = 'py2exe'
sys.path.append(os.path.dirname(os.path.abspath(name)))

setup(name=name[:-3], scripts=[name])




Regards
Davinci

"Stephen Smith" <ssmith619 at hotmail.com> schrieb im Newsbeitrag
news:jpae8.66140$X64.23604369 at news1.rdc2.pa.home.com...
> I'm new to Python, and have created a number of small programs that users
> might find interesting, althought they aren't really geeky and the people
> that use them would probably not have the Python IDLE or anything even
> related to Python installed.  Is there anyway to turn a .py file into a
.exe
> file that can be executable on Windows?  It doesn't call from any other
> modules, it's completely stand alone.  Thank you for any information!
>
>




From mcherm at destiny.com  Mon Feb 18 11:54:41 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 18 Feb 2002 11:54:41 -0500
Subject: List unchanged in function--why?
Message-ID: <3C7131D1.4080305@destiny.com>

> This must have something to do with references etc, but I'm not sure
> what's going on.


Yes... It has to do with references. Let me explain.

> def changeList(theList):
>   theList = ['a']
> 
> aList = ['x', 'y']
> print aList
> changeList(aList)
> print aList
> 
> 
> Running the above results in
> 
> ['x', 'y']
> ['x', 'y']
> 
> Can someone explain?  I was expecting the second line printed to be
> ['a']


Sure. The argument theList is passed "by reference" (the ONLY way that 
args are passed in Python), so you can mutate the object if you like. 
But you didn't... your line "theList = ['a']" CHANGES what the local 
variable 'theList' refers to, instead of modifying the object it now 
refers to.

This should achieve what you wanted:

def changeList(theList):
   theList[:] = ['a']

It's not a very useful idiom... if you wanted to replace the entire 
contents of the list the BEST way to do it is like this:

def makeNewList():
    return ['a']
aList = makeNewList()

But of course you were really only experimenting. Your real goal is 
probably not to replace the entire list but modify the list... append 
new items onto it perhaps. And those kinds of operations work really 
well the way you started out:

def addToList(theList, newItem):
    theList.append(newItem)


Have fun!

-- Michael Chermside






From gerson.kurz at t-online.de  Wed Feb  6 12:18:28 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Wed, 06 Feb 2002 17:18:28 GMT
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
Message-ID: <3c6164f0.50243062@news.t-online.de>

Well, it sure doesn't look like that in their programming contest:

http://www.google.com/programming-contest/index.html#rules

Or was the Google-Python relation just an "urban legend"?


From aahz at panix.com  Fri Feb  8 14:10:47 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 11:10:47 -0800
Subject: OT: Perl programmers?
Message-ID: <a417rn$ge8$1@panix2.panix.com>

I know some of y'all still keep roots to the Perl community, so I was
wondering if there's any kind of definitive nickname for Perl
programmers, the way we call ourselves Pythonistas?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From m.faassen at vet.uu.nl  Fri Feb  8 15:23:50 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Feb 2002 20:23:50 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <a41c4m$bf4$1@newshost.accu.uu.nl>

Donn Cave <donn at u.washington.edu> wrote:
> So is sci.math choked with posts from people who normally seem
> to have their heads screwed on, endlessly replying this guy as
> if they sincerely think there's some purpose in it?  People who
> take time several times every day to post about how useless his
> ideas are and how they have no interest in helping?  Or are
> they saner there than we are?

I hope you're not including me. I've posted in those threads but
only because I'm so amused.

Besides, even if I were arguing with the fellow I might've been
doing it for amusement. But this type of person doesn't seem quite
what to make of me most of the time, so they just don't respond..

> The existence of people with unusual ideas isn't surprising.
> I'm kind of taken aback by the response here, though.

Don't worry, most newsgroups seem to exhibit this phenomenon. It is just
comp.lang.python's immune system seems to be a little bit disabled
sometimes recently.

Regards,

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


From sag at hydrosphere.com  Wed Feb 20 18:57:41 2002
From: sag at hydrosphere.com (Sue Giller)
Date: Wed, 20 Feb 2002 16:57:41 -0700
Subject: No space left on device
In-Reply-To: <A7B5C22F20EE3C4E95744E2F03F1887F55B1@MAYOR.retzlaff.com>
Message-ID: <20020220235338622.AAA41@mail.climatedata.com@SUEW2000>

Ok, more info.

My Win2000 machine harddrive seems to be FAT32 format.  I can 
successfully generate (extremely quickly) the 100K files if I create a 
set of subdirectories and put files in there (my test case put 50 files 
in each subdir).  This would indicate that the slow downs are indeed 
due to mktemp doing its checking.

I can get the full 100K files to be created, relatively quickly, on 
another drive on my network (NT 4.0), with NTFS formatting, with no 
problems.  This is much quicker than creating the 21K files on the 
FAT32 disc, in one directory, even over the network.

So, I am assuming there is some problem with the FAT32 system 
when a directory gets lots of files in it.  Since I can make the 100K 
files if I don't use mktemp, I am also assuming there is some 
problem with mktemp in this scenario.

If anyone figures this out, it would be interesting to know what is 
going on.  Meanwhile, I guess I'll need to redesign my file 
generation algorithm to split things up.

thanks for the input.

sue





From logiplexsoftware at earthlink.net  Fri Feb  8 16:22:07 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 13:22:07 -0800
Subject: Please stop responding to Rue
In-Reply-To: <a41c4m$bf4$1@newshost.accu.uu.nl>
References: <b03e80d.0202060227.368f6ea7@posting.google.com>
	<3c62a989.3448437947@news>
	<a3ucsk$12h8$1@nntp6.u.washington.edu>
	<a41c4m$bf4$1@newshost.accu.uu.nl>
Message-ID: <20020208132207.16013fec.logiplexsoftware@earthlink.net>

On 8 Feb 2002 20:23:50 GMT
Martijn Faassen wrote:

> 
> Besides, even if I were arguing with the fellow I might've been
> doing it for amusement. But this type of person doesn't seem quite
> what to make of me most of the time, so they just don't respond..

Arguing with irrational oddballs on usenet is today's equivalent of the
Rubik's Cube.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From wurmy at earthlink.net  Fri Feb  1 23:10:52 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Sat, 02 Feb 2002 04:10:52 GMT
Subject: Why has 'dir()' changed in Python 2.2?
References: <mailman.1012618323.27059.python-list@python.org>
Message-ID: <3C5B66CC.27A32175@earthlink.net>

"Dr. David Mertz" wrote:
> 
> I just noticed a change in Python 2.2... that turns out to break my
> [xml_pickle] module.  It shouldn't be hard to fix (although I *do* have
> a decision to ponder about the best fix)... but I mostly wonder what the
> thoughts behind the change were (it seems like it will break a lot of
> code that is moderately introspective).

The descrinto document explains why:

  http://www.python.org/2.2/descrintro.html#introspection

"In classic Python, the method names of lists were available as 
the __methods__ attribute of list objects, with the same effect 
as using the built-in dir() function. [...] Under the new proposal, 
the __methods__ attribute no longer exists. [...] Instead, you can 
get the same information from the dir() function, which gives more 
information."

Apparently, it was expected that not much code would break.
Personally, I only use dir() in interactive sessions, to quickly
inspect a module or (other) object. For introspection in
programs, I usually inspect the object's __dict__.

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From sholden at holdenweb.com  Fri Feb  8 14:30:14 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 14:30:14 -0500
Subject: Usenet fascism
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <kk186ugk4hpo7m0eca81bv2lm2d0smnjiv@4ax.com>
Message-ID: <1HV88.108504$KM2.4230280@atlpnn01.usenetserver.com>

"Courageous" <jkraska at san.rr.com> wrote in message
news:kk186ugk4hpo7m0eca81bv2lm2d0smnjiv at 4ax.com...
>
> >1. Nobody or nothing gives you that right.
>
> That's the _one_ thing you're mistaken about. I can
> entertain the idea of everything else you said, but
> the notion that people don't have the right to do this
> is just plain wrong. Plainly is the right not only held,
> but it can be exercised.
>
I suspect you are confusing rights and abilities.

If we are talking, and I am carrying a gun when something you say gets up my
nose, I have the ability to shoot you in the head just to make you feel
sorry you uttered inflammatory words.

I don't have the *right* to do that, however. Most civilized countries would
have some legal sanction if I exercised that ability.

not-that-you'd-be-any-less-dead-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From boud at rempt.xs4all.nl  Wed Feb 27 16:11:22 2002
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 27 Feb 2002 21:11:22 GMT
Subject: Printing to a "log screen" in a multi-module QT app
References: <54abe19d.0202271213.5e6ef2b0@posting.google.com>
Message-ID: <3c7d4b79$0$18499$e4fe514c@dreader4.news.xs4all.nl>

Anders Schneiderman <schneida at seiu.org> wrote:
> I'm writing an application using QT.  Right now when I want to use the
> equivalent of "print", I call a method called "log" that's defined in
> the main class in my main module:

> def log(self, msg):
>    try:
>       self.logArea.insertline(msg, -1)
>    except:
>       self.logArea.insertline(str(msg), -1)

> To call it in other classes that are in other modules of the app, I
> pass "log" to the new class.  For ex:

> class CodeEditor:
>    def __init__ (self, log, ...)
>        self.log = log

> Is there a cleaner way to do this?  For example, is there a way to
> redefine the "print" command?  Whatever solution I come up with, the
> command to print has to be short;  "self.logObject.log('File
> successfully updated')" is too much of a pain.  Any suggestions?

As a random, untested, and wild idea: create an object that allows
reading and writing. Have this object be a class member in a module:
import that class. Have the apps write to the object; have your log
window read from that object periodically, using QTimer.

Another equally wild idea: add a log function to the QApplication class,
which you can access with qApp. Have that function emit a signal, which
is connected to a slot in your log displayer.

You could, of course, also have a log function in _all_ your classes
(perhaps via an add-in class), which once establishes the ultimate parent
QMainWindow, and writes to the single logwindow of that object.

Or you could emit a log signal from all your classes, with the message as
the payload, and connect your objects to the logwindow when creating them.

But passing the logwindow to all your classes should be quite unnecessary,
since, in essence, it's an object shared by all your other objects, much
like the configuration object in the example application in my book. (Which
is now available in paper format from opendocs.org -- yay!)

-- 

Boudewijn Rempt  | http://www.valdyas.org 


From phr-n2002a at nightsong.com  Sun Feb 17 02:45:51 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 16 Feb 2002 23:45:51 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com>
Message-ID: <7xr8nkzh3k.fsf@ruckus.brouhaha.com>

Tim Roberts <timr at probo.com> writes:
> And the original poster should understand that this is true because such
> systems are of little practical use.  There are very, very few physical
> processes where floating point is useful AND precision of more than 6
> significant digits is really required.  Large integers can be useful (and
> Python supports them), but a requirement for high precision floats is
> usually a sign that the requirer does not understand his problem space.

If that were really true, computers wouldn't bother implementing 
double and extended precision.  Those features are there for a reason,
which is that they are needed.

If you try to (say) solve a near-singular matrix with single precision
and a numerically unstable algorithm, your answers can come out
totally bogus (all digits incorrect) even if the inputs started with 6
accurate significant figures.  You can then either spend years of your
life studying numerical analysis so you can design algorithms with
ultra-careful error control and get good answers with single
precision, or you can just run your program with double or extended
precision arithmetic and get good answers right away by just burning a
little more computer time.

If all you want is to get good answers, it's often easier to crank up
the precision and let the machine do the work.

See for example 

   http://www.cs.berkeley.edu/~wkahan/JAVAhurt.pdf

which mostly is a rant against Java for not implementing IEEE extended
precision arithmetic, but discusses the precision issue from a number
of perspectives.


From spencer at efn.org  Thu Feb 28 15:48:55 2002
From: spencer at efn.org (Spencer Ernest Doidge)
Date: Thu, 28 Feb 2002 12:48:55 -0800
Subject: Pointers for solving windows GUI problem.
In-Reply-To: <f72dac0d.0202280747.2360c608@posting.google.com>
References: <f72dac0d.0202280747.2360c608@posting.google.com>
Message-ID: <Pine.GSU.4.21.0202281241260.23516-100000@garcia.efn.org>

If in the end you do have to use VC++, you might find as I have that it is
actually easier not to use MFC. Charles Petzold's book and Jeffrey
Richter's book opened the door into Win32 API for me. You don't have to go
near MFC in VC++, and you can still build resources using the GUI. So you
don't have to write code, for example, to create a fancy dialog box.

Like you, I am looking for a pure Python/WinGUI solution. AS I just
learned from this group, http://www.activestate.com has Visual Python, but
you have to have Visual Studio .NET for that. That means many $$$ :-(

Spencer Doidge
www.spencerdoidge.com

On 28 Feb 2002, Alan Kennedy wrote:

> Greetings all,
> 
> I'm posting this question in the hope that someone has already done
> this, or something similar, and can perhaps direct me some code or a
> tutorial that will shorten the path to a solution.
> 
> I have a Windows COM object which renders a series of images into a
> window. I can control the COM object using provided API methods. I
> don't have the source for the COM object.
> 
> I want to do screen grabs of the display of this window at various
> stages, so that I can dump them to files, in GIF, JPEG or PNG format.
> 
> What I'm looking for is perhaps some sample code that does this or
> something similar. Failing that, a pointer to a good MFC tutorial
> would be most helpful.
> 
> I am totally unfamiliar with MFC and Win32 GUI programming, although
> I've developed some X-Windows stuff in the distant past.
> 
> However, given the excellent tools available, namely Python and Mark
> Hammond's excellent Windows extensions, I have every faith that the
> job can be done without having to go near VC++ (which I can do, but
> really don't want to :-)
> 
> TIA for any pointers.
> 
> Cheers,
> 
> Alan.
> 



From jason at jorendorff.com  Fri Feb  1 20:39:50 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 19:39:50 -0600
Subject: comparing DateTime instances
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBKEFPKAAA.mark@mceahern.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMENDCPAA.jason@jorendorff.com>

As some icing on this topic, you can remember the
meaning of what cmp() returns like this:

  cmp(a, b) < 0   means  a < b
  cmp(a, b) == 0  means  a == b
  cmp(a, b) != 0  means  a != b
  cmp(a, b) > 0   means  a > b
  ...

Get the pattern?  Works for C's strcmp() too.

## Jason Orendorff    http://www.jorendorff.com/



From James_Althoff at i2.com  Thu Feb 28 18:52:38 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 28 Feb 2002 15:52:38 -0800
Subject: Status of PEP's?
Message-ID: <OFB0AA6551.3FEB33B9-ON88256B6E.0081ABFE@i2.com>

[Tim Delaney]
> I never think of it though as
>
> for i in 5:
>     print i
>
> but always as
>
> for i in len(seq):
>     print i
>
> or something similar, which has obvious semantic logic.

And I think such context makes the example easy to understand.  (And, yes,
abuse is possible).

> However, I could see myself happily using for i in 5: in a quick script.

Me too.  :-)

> One important thing to remember, which I think a few people have missed,
> is that an integer will not be an iterator. An integer will be iterable -
> that is, to get this behaviour, iter(integer) must be called at some
stage.
> This is done transparently in a for: loop.

Very good point.

> Are the following two equivalent
> though?
>
> if 1 in obj:
>     pass
>
> if 1 in iter(obj):
>     pass
>
> i.e. is iter(obj) called implicitly in the first case?

Good question.  I was wondering that, myself.  I'm guessing that the answer
is no because of the following:

>>>
>>> 1 in 'spam'
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: 'in <string>' requires character as left operand
>>>
>>> 1 in iter('spam')
0
>>>

Jim




From mwh at python.net  Tue Feb 19 11:06:18 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 19 Feb 2002 16:06:18 GMT
Subject: Python crash
References: <993ae4ca.0202190754.6ccb79d2@posting.google.com>
Message-ID: <u664tcv2n.fsf@python.net>

AdrianLeu at kelseus.com (Adrian Leu) writes:

> Hi !
> I have written a script which I am calling from inside a C++
> application using:
> 
>  module = PyImport_ImportModule( ... ) and 
>  method = PyObject_CallMethod( ... ) 
> 
> methods. 

I think knowing which version of Python you have would be useful.

Cheers,
M.

-- 
  I hate leaving Windows95 boxes publically accessible, so shifting
  even to NT is a blessing in some ways.  At least I can reboot them
  remotely in a sane manner, rather than having to send them malformed
  packets.      -- http://bofhcam.org/journal/journal.html, 20/06/2000


From tdelaney at avaya.com  Tue Feb  5 17:57:41 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 6 Feb 2002 09:57:41 +1100 
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A169@natasha.auslabs.avaya.com>

> From: Frederick H. Bartlett [mailto:fbartlet at optonline.net]
> 
> Well, this is fun. Perhaps the source of Python's appeal ... 
> snakes are
> surely ambidextrous if they're any kind of dextrous at all.

But snakes have nothing to do with Python. I shall now have to slap you with
a fish.

Tim Delaney



From gh_pythonlist at gmx.de  Tue Feb 12 00:44:28 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 12 Feb 2002 06:44:28 +0100
Subject: Possible dumb query
In-Reply-To: <000801c1b385$97747aa0$0100000a@ahimsa>
References: <000801c1b385$97747aa0$0100000a@ahimsa>
Message-ID: <20020212054427.GB1648@lilith.hqd-internal>

Le 12/02/02 ? 07:24, ahimsa consulting ?crivit:
> Greeting Pythonites
> 
> This might qualify as one of those "really dumb questions", but:

There is no such thing as a dumb question :-)

> Under Linux RH 7.2 when doing a search for which python ("$ which
> python") I have discovered that I have one entry for python as
> /usr/bin/python and one entry for python2.2 as /usr/bin/python2.2. My
> question is this: if I were to delete the directory /usr/bin/python
> what impact - if any - will it have on any of the libraries/ etc. that
> I would require to use python2.2

/usr/bin/python is a symlink to then "preferred" Python version, i. e.
the one that will be started when you enter "python" on a shell.

As you're using Redhat, which is terribly up-to-date in Python matters,
/usr/bin/python is a symlink to /usr/bin/python1.5 which is Python
1.5.2.

Deleting the symlink or changing it to the Python executable of your
Python 2.2 installation is not recommended. It will certainly break
things, as some Redhat system tools depend on Python 1.5, yet have a
hardcoded path "/usr/bin/python" in their first "shebang" line.

In short: Just invoke your Python 2.2 with "python2.2" and you will be
fine. Messing with the tools of the base system (like Python on Redhat)
can always be done, but you should be prepared to fix things that break.

So, when will Redhat finally make a symlink "/usr/bin/python-redhat" and
just use this one in their scripts!?

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.9 ?C      Wind: 7.1 m/s



From ivo at nospamAmaze.nl  Mon Feb  4 10:36:47 2002
From: ivo at nospamAmaze.nl (ivo at nospamAmaze.nl)
Date: 4 Feb 2002 15:36:47 GMT
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <a3m9qf$hrq$1@news1.xs4all.nl>

Fred <c at tech.usd253.org> wrote:
: Hello, I am just starting out with Python. I have a very basic book
: that is not helping me much. I have done some simple C programming in
: the past and I think if I could see the below C code done in Python it
: would get me over the beginners hump. I tried to do it in Python but
: got stuck right at the start, Python did not like it when I tried to
: say: for i = 1

: #include <stdio.h>
: int main (void)

: {

: int i;
: int j;

: for (i = 1; i < 13; i++)
: 	{
: 	for (j = 1; j < 13; j++)
: 	  printf("%4d", i*j);
: 	        printf("\n");
:    }
: return 0;
: }

I think you shoud work on your indentation first before trying python. :)

	Ivo


-- 
Drs. I.R. van der Wijk                              -=-
Brouwersgracht 132                      Amaze Internet Services V.O.F.
1013 HA Amsterdam                                   -=-
Tel: +31-20-4688336                          Linux/Web/Zope/SQL
Fax: +31-20-4688337                           Network Solutions
Web:     http://www.amaze.nl/                    Consultancy
Email:   ivo at nospamAmaze.nl                         -=-



From jkraska at san.rr.com  Mon Feb 25 23:29:25 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 04:29:25 GMT
Subject: DLL/Embeded interpreter wierdness
References: <qltl7uoh7q01drhb37b48bsmipr7h8kaam@4ax.com> <uy9hgvpi2.fsf@ctwd0143.fitlinxx.com>
Message-ID: <164m7ug4ramg6hjfiac8mb4inajc59mr9h@4ax.com>

>Typically when I've done this (both embedded and extended in the same
>DLL or executable), what I've done is automatically define my module
>internally before calling out to the embedded interpreter.  The
>differences between initialization in the two cases are very slight.

I don't really know what you mean here. This module in question is
actually masquerading as a package, so extended import commands look
more like...

from lramm import task

...and so on. So the explicit initialization is required from the embedded
code, which is otherwise unrestricted Python.

I seem to have some sort of knowledge failure here. Outside of the
usual init[blah] hookups, what does one do to "automatically define
my module internally"?

C//



From mcherm at destiny.com  Wed Feb 13 10:31:51 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Wed, 13 Feb 2002 10:31:51 -0500
Subject: defining functions
References: <89.135d5d70.2999c21f@aol.com>
Message-ID: <3C6A86E7.8070601@destiny.com>

Patio87 at aol.com wrote:

>        Thanks for your reply, but I had another question. I dont 
> understand how factorial(n) gets the value of two? And does The 'n' have 
> 1 assigned to by default?

Okay, try this simpler example:

 >>> def threeTimes(x):
	print 'x is now %i' % x
	if x == 0:
		return 0
	else:
		return 3 + threeTimes(x-1)

After writing this, what is the value of 'x'?
	
 >>> print x
Traceback (most recent call last):
   File "<pyshell#9>", line 1, in ?
     print x
NameError: name 'x' is not defined

As you can see, x is not defined... it doesn't have ANY value. But if we 
try CALLING the function, it'll print out x DURING the call:

 >>> threeTimes(0)
x is now 0
0

Great... so x was 0 then. But what about now, AFTER we called it?

 >>> print x
Traceback (most recent call last):
   File "<pyshell#11>", line 1, in ?
     print x
NameError: name 'x' is not defined

Okay... what's going on here is that 'x' took on a value TEMPORARILY, 
while the function was being evaluated, but then it went back to being 
undefined. Actually, a more precise definition would be that the 
variable 'x' was "local" to the function, and so the 'x' that had been 
set to 0 was destroyed when the function finished.

What will happen if we call threeTimes() with a bigger number? Let's 
start with 1:

 >>> threeTimes(1)
x is now 1
x is now 0
3

First, it starts up the function "threeTimes", which means that it has 
to create an 'x' -- this gets set to 1 since that's what you used when 
you called it. Then it prints it out, so you get the line "x is now 1". 
It tries the 'if', and decides to go with the 'else' clause... so it 
needs to return "3 + threeTimes(x-1)", or "3 + threeTimes(0)". But NOW 
it has to work out what "threeTimes(0)" is.

So it starts up ANOTHER COPY of the function "threeTimes". The old copy 
is still there (in which x=1), but the new copy has to set x=0. It 
prints out x (so we get the line "x is now 0"), and then it tries the if 
and finds out that all it has to do is return an answer of 0.

So the SECOND copy of "threeTimes" has finished, and the FIRST copy can 
now continue. It now knows that "3 + threeTimes(0)" is just "3 + 0", and 
so it returns "3", which then gets printed out.

I imagine you can now guess what happens when we call "threeTimes(6)"... 
right?

 >>> threeTimes(6)
x is now 6
x is now 5
x is now 4
x is now 3
x is now 2
x is now 1
x is now 0
18

There are many situations for which this technique (recursion) is very 
powerful and useful, so it's worth learning even if it warps your brain 
a little bit at first.

-- Michael Chermside




From BPettersen at NAREX.com  Mon Feb  4 13:23:39 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Mon, 4 Feb 2002 11:23:39 -0700
Subject: working with files with the same extenstion
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158CF6@admin56.narex.com>

> From: Ulrich Pfisterer [mailto:ulrich.pfisterer at rmb.co.za] 
> 
> Can anyone please tell me how I would go about doing the 
> following. I would like to write a script that would take in 
> a command line 
> parameter "*.txt". The script would then loop through every 
> file with a 
> "txt" extension and something with it.
> 
> I am brand new to python (I started using it 2 days ago) so I 
> dont know 
> if there is a module that does already this, or whether I have to do 
> all the wildcard matching etc myself.

Welcome to Python.

You're looking for the glob module. (and you'll probably need to
remember that on *nix the command line arguments are expanded by the
shell unless quoted)

-- bjorn



From st-newsgroups at thomas-guettler.de  Tue Feb  5 04:53:53 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Tue, 05 Feb 2002 10:53:53 +0100
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <7x665cz2xl.fsf@ruckus.brouhaha.com>
Message-ID: <3C5FABB1.5010703@thomas-guettler.de>

Paul Rubin wrote:

> Thomas Guettler <st-newsgroups at thomas-guettler.de> writes:
> 
> 
>>I have two python version installed:
>>with the cygwin version "print 2**32-1" works
>>but with the zope version I get OverFlowError.
>>
>>Background:
>>I want to get a random 32-Bit integer like this randrange(0, 2*32-1)
>>
> 
> Values of sys.maxint must be different.  Try: randrange(0,2147483647).
> 

The module random contains a lot of functions. I think getting a random 
32 random number is needed often.

randrange(0,2147483647) is not a 32 Bit random number because
  2**32-1 is 4294967295.






From gcordova at hebmex.com  Fri Feb 22 18:08:53 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 17:08:53 -0600
Subject: help: Problem with cgi form
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE8@SRM201002>

> 
> I'm a beginner. How do I do this? It's probly very simple ......
> 
> I have a hidden input on a cgi form... 
> 
>     print '<INPUT TYPE=HIDDEN NAME= "last_pos" 
> VALUE=',last_byte_pos,'>'
> 
> I need the VALUE to be the  variable "last_byte_pos", but 
> unfortunately 
> when I run the script for the first time the script crashes 
> because it's 
> undeclared. 
> 
> What subroutine do I use to check if last_byte_pos is undeclared, and 
> assign it a default value if it is? 
> 

How can you not know if "last_byte_pos" is undeclared?
Have you assigned any values to it previously?
no?

then it's undeclared.

-gus



From sholden at holdenweb.com  Thu Feb  7 10:12:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 10:12:04 -0500
Subject: Is anyone capable of explaining ?
References: <mailman.1012997891.14129.python-list@python.org>
Message-ID: <XOw88.95050$KM2.3453687@atlpnn01.usenetserver.com>

"Scherer, Bill" <Bill.Scherer at VerizonWireless.com> wrote in message
news:mailman.1012997891.14129.python-list at python.org...
> On Tue, 5 Feb 2002, A wrote:
>
> >
> > Can anybody please answer the following question?
> >
> > Question: How does web browser know what cookies should be
> > sent to server as a response?
>
> The browser determines which cookies to include with a request by
> comparing the domain of the server it is about to contact with
> the domain of the cookies.   I didn't see any domain in the
> cookies you are setting; there should be one.
>
> The template http cookie header:
>
> Set-Cookie: name=value; expires=date; path=pathname; domain=domainname
>
> See:
> http://www.webreference.com/js/column8/http.html
>
Let's not forget that the browser will also be doing path comparisons. In
general a server can issue a cookie that applies to any subset of

a) the server's domain or a subdomain
b) a subset of all possible paths

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jjl at pobox.com  Tue Feb  5 13:45:53 2002
From: jjl at pobox.com (John J. Lee)
Date: Tue, 5 Feb 2002 18:45:53 +0000
Subject: [repost] win32 debugging-- python21_d.dll not found (fwd)
In-Reply-To: <mailman.1012826230.26184.python-list@python.org>
References: <Pine.SOL.4.30.0202022147160.597-100000@mimosa.csv.warwick.ac.uk> <mailman.1012826230.26184.python-list@python.org>
Message-ID: <Pine.SOL.4.30.0202051844000.26065-100000@mimosa.csv.warwick.ac.uk>

On Mon, 4 Feb 2002, Gerhard [iso-8859-15] H?ring wrote:

> Le 02/02/02 ? 21:49, John J. Lee ?crivit:
> > [problem with mingw compiled Python extensions in debugging mode]
>
> If all you want to do is debug a Python extension compiled with mingw32,
> the following HOWTO from me might help you (watch out: *one* line):
>
> http://groups.google.com/groups?q=python+howto+debugging+open+source&hl=en&selm=849vl9.6b1.ln%40gargamel.hqd-internal&rnum=1
[...]

Thanks -- I'd already found your howto, not sure if it helps yet.

> Btw. I'm currently porting Python to a 100% pure mingw build process
> using MSYS. I've already a running python.exe :-) The rest is getting
> rather tricky, though :-/

Sounds useful.


John



From tjreedy at home.com  Thu Feb 21 00:46:58 2002
From: tjreedy at home.com (Terry Reedy)
Date: Thu, 21 Feb 2002 05:46:58 GMT
Subject: Recursively traverse linked list -- help!
References: <mailman.1014240283.12938.python-list@python.org> <7xadu3wwr5.fsf@ruckus.brouhaha.com>
Message-ID: <mR%c8.57438$X64.20569489@news1.rdc2.pa.home.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xadu3wwr5.fsf at ruckus.brouhaha.com...
> "Jason Orendorff" <jason at jorendorff.com> writes:
> > > > def f(val):
> > > >   global f
> > > >   f = val and f1 or f2
> > > >   return f(val)
> > >
> > > This is not tail recursion.  The "global f" declaration shadows
> > > the "f" that names the function.

I picked about the simplest example I could that would illustrate the
point that a tail call that looks like tail recursion, when viewed by
itself, is not necessary so in Python due to dynamic name-object
binding.  The rebinding could occur in any function on the call chain
under f.  It could also take place after some calls that really are
recursive.

> > That's not true.  The f that names the function is global anyway.
> > It's the assignment that causes the problem.
>
> Oh yes, you're correct.  However, the tail call should still be
> optimized, compiled as an indirect jump to f.

I can accept 'maybe could' (see below) but 'should'?  What sense of
shouldness are you using?  "It would be nice if"?  The passive 'be
optimized' glides over the 'who' that 'should' do the work and
maintain it for the next decade.  Automated object code optimization
seems to be a tricky process.  Other suggestions have been made, and
some volunteer work done, but the last report I have read from Tim
Peters is that the CPython interpreter sticks to straightforward
tranlation to PyCode.

As for 'maybe could': by 'indirect', I presume you mean after looking
up the name to find its current binding.  If so, then it really does
not matter what the name is that is looked up.  So, are you suggesting

A. general tail call optimization?

  If so, 'jump to' is problematic, since there is no frame object to
jump to, the avoidance of creating such being part of the purpose of
the optimization.  If the call is not recursive, I could imagine
replacing the contents of the current frame. But I do not know how
feasible this is with the current implementation.

or B. just tail recursive call optimization (where the call is
recursive)?

  If so, there is still the two tasks of 1) determining whether the
target of the jump is the same as the currently executing code body,
and 2) doing the jump properly.

Terry J. Reedy






From tim at vegeta.ath.cx  Fri Feb 15 22:26:56 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Sat, 16 Feb 2002 03:26:56 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013803495.27900.python-list@python.org>
Message-ID: <slrna6rkg9.atu.tim@vegeta.ath.cx>

Gustavo Cordova <gcordova at hebmex.com> graced us by uttering:
>> 
>> Take a latin alphabet. Take a lot of exceptions. Throw in some weird
>> letters. Take some English words. Change a lot of spelling. 
>> Mix in some more spooky letters.
>> Make rules. Make exceptions to these rules.  Make
>> even more exceptions. Shuffle. The resulting language will be ....
>>
> 
> PERL?????

What's "PERL"?

No. He limited his add-ons to actual letters and rules. As far as c.l.py
is concerned, Perl has far too few of either.  <wink>

>> German :)
>> 
> 
> Ah. Oh.

Tim Hammerquist
-- 
Anyone who thinks that the act of programming is worth
less than $50/hr should leave the industry.
    -- Randal L. Schwartz (clpm, 27 Jul 2000)


From rony.steelandt at bucodi.com  Wed Feb 20 04:22:50 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 20 Feb 2002 01:22:50 -0800
Subject: cursor control
References: <3c731781.5352364@nntp.surry.net>
Message-ID: <b34d3835.0202200122.40f826cf@posting.google.com>

cstrickland at surry.net (Charles W. Strickland) wrote in message news:<3c731781.5352364 at nntp.surry.net>...
> Is it possible to control cursor positioning in python without using
> TKinter?  I wanted to write a program that required it, but could not
> find anything anywhere to explain how to do it.  There were a couple
> references awhile back, but I never did see a definitive answer.  I
> finally wrote the program in C - what a mess!

Have a look at the curses module

Rony


From cimarron+google at taylors.org  Thu Feb 28 15:03:06 2002
From: cimarron+google at taylors.org (Cimarron Taylor)
Date: 28 Feb 2002 12:03:06 -0800
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com> <29e28c51.0202270014.61dbd1d0@posting.google.com> <a5ji37$ceb$2@peabody.colorado.edu>
Message-ID: <29e28c51.0202281203.d4cf6ee@posting.google.com>

Fernando P?rez <fperez528 at yahoo.com> wrote in message news:<a5ji37$ceb$2 at peabody.colorado.edu>...
> I sincerely hope this was a joke (and as such, a really good one :). I can't 
> seriously imagine you actually *doing* those things. 1 is fine, maybe 2.
> 
> Anyway, thanks for the good laugh!
> 
> f.

I did actually use most of these tricks on one Java project although not all 
were used equally frequently: all the code was obfuscated, half the modules
used dispatch tricks (which actually had another purpose), a lot of bulk
was added by third party libs, some classes were loaded from a decrypted
license file (those classes were responsible for setting up the dispatchers
so without a license the code would not operate, not because of an explicit
license check, but because instances would not be properly initialized),
the license was distributed separately from the program and had a 
separate installation procedure.

Cim


From tim.one at comcast.net  Thu Feb 21 15:46:46 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 21 Feb 2002 15:46:46 -0500
Subject: Large integers from C
In-Reply-To: <c76ff6fc.0202211211.4f33f947@posting.google.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEFENOAA.tim.one@comcast.net>

[John Machin]
> ...
> So why not grab the source of _PyLong_FromByteArray() from the 2.2
> distribution and put in your own code?

That's what I'd do.  Do note the leading underscore:  C functions in the
core whose names begin with a leading underscore are deliberately
undocumented in the manuals, and nothing about them (not even their
existence) is guaranteed across releases.  They're internal implementation
details, not part of the public C API.  Use of them is entirely at your own
risk.




From xunningy at ece.arizona.edu  Tue Feb 12 02:40:24 2002
From: xunningy at ece.arizona.edu (Xunning Yue)
Date: Tue, 12 Feb 2002 00:40:24 -0700
Subject: isNumberType? a bug or a feature?
Message-ID: <20020212004024.035d005e.xunningy@ece.arizona.edu>

Here is my situation with Python2.2 in Linux:
>>> from operator import *
>>> class dummy:
	pass

>>> t = dummy()
>>> type(t)
<type 'instance'>
>>> isNumberType(t)
1

Is this a bug or a feature of Python? Can anyone answer this?
Thanks


From Francois-regis.Chalaoux at sanofi-synthelabo.com  Tue Feb 19 09:23:08 2002
From: Francois-regis.Chalaoux at sanofi-synthelabo.com (Francois-regis Chalaoux)
Date: Tue, 19 Feb 2002 15:23:08 +0100
Subject: Réf. : Re: strange libtk8.3.so: open failed
Message-ID: <"020219214906Z.WT25538. 28*/PN=Francois-regis.Chalaoux/O=RESEARCH/PRMD=SANOFI/ADMD=ATLAS/C=FR/"@MHS>

* The sun machine is:
SunOS lads01 5.8 Generic_108528-03 sun4u sparc SUNW,Ultra-60

* On command line Dynamic linker find libtk8.3.so and the librairie is there:

lads01 /tmp/Python-2.1.1 % ldd /usr/local/bin/python
        libpthread.so.1 =>       /usr/lib/libpthread.so.1
        libsocket.so.1 =>        /usr/lib/libsocket.so.1
        libnsl.so.1 =>   /usr/lib/libnsl.so.1
        libdl.so.1 =>    /usr/lib/libdl.so.1
        libthread.so.1 =>        /usr/lib/libthread.so.1
        libtk8.3.so =>   /usr/local/lib/libtk8.3.so
        libtcl8.3.so =>  /usr/local/lib/libtcl8.3.so
        libX11.so.4 =>   /usr/lib/libX11.so.4
        libm.so.1 =>     /usr/lib/libm.so.1
        libc.so.1 =>     /usr/lib/libc.so.1
        libmp.so.2 =>    /usr/lib/libmp.so.2
        libXext.so.0 =>  /usr/lib/libXext.so.0
        /usr/platform/SUNW,Ultra-60/lib/libc_psr.so.1

FR





phd at phd.pp.ru%internet le 18/02/2002 17:05:14
Veuillez r?pondre ? phd at phd.pp.ru%internet at WTFR
Pour :	Francois-regis Chalaoux/FR-LABEGE/RESEARCH/SANOFI at Research
cc :	python-list at python.org%internet at WTFR 

Objet :	Re: strange libtk8.3.so: open failed

Mail-Followup-To: copter24 <francois-regis.chalaoux at sanofi-synthelabo.com>, python-list at python.org
User-Agent: Mutt/1.2.5i

On Mon, Feb 18, 2002 at 07:51:03AM -0800, copter24 wrote:
> ld.so.1: /usr/local/bin/python: fatal: libtk8.3.so: open failed: No
> such file or
>  directory

   Dynamic linker cannot find libtk8.3.so or *.so referenced by
libtk8.3.so. What OS? Where is your libtk8.3.so in the filesystem?

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






From pbienst at MIT.EDU  Tue Feb 26 11:47:27 2002
From: pbienst at MIT.EDU (Peter Bienstman)
Date: 26 Feb 2002 11:47:27 -0500
Subject: Precompiled regular expressions slower?
In-Reply-To: <31575A892FF6D1118F5800600846864DCBCD08@intrepid>
References: <31575A892FF6D1118F5800600846864DCBCD08@intrepid>
Message-ID: <1014742048.11444.12.camel@hunsmac.mit.edu>

Sticking with readlines rather than xreadlines for the moment to compare
apples to apples, these are the timings:

precompiled as global var: 4.7 sec
precompiled as local var : 4.5 sec
not precompiled          : 3.9 sec

It helps somewhat, but doesn't solve the problem.

On Tue, 2002-02-26 at 11:31, Simon Brunning wrote:
> > From:	Peter Bienstman [SMTP:pbienst at mit.edu]
> > I've a Python script a want to optimise. It basically loops over all the 
> > lines in a file and does some regular expression matching:
> > 
> > line = inputfile.readline()
> > while line:
> >   s = search(r".....", line)
> >   ...
> >   line = inputfile.readline()
>  
> You might find something like this faster:
> 
> for line in inputfile.xreadlines():
>     s = search(r".....", line)
>     ...
>  
> > There are several reg ex in the loop. Since the Python version runs 2 to 3
> > 
> > times as slow as its Perl counterpart, I thought I'd try speeding up the 
> > script by precompiling the reg ex:
> > 
> > re1 = compile(r"....") 
> > ...
> > line = inputfile.readline()
> > while line:
> >   s = search(re1, line)
> >   ...
> >   line = inputfile.readline()
> > 
> > 
> > Strangely enough this new version seems to run about 30% *slower* than the
> > 
> > old version.
>  
> Try putting the whole lot in a function - global access is slower than local
> access.
> 
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> 
> 
> 
> 
> -----------------------------------------------------------------------
> The information in this email is confidential and may be legally privileged.
> It is intended solely for the addressee. Access to this email by anyone else
> is unauthorised. If you are not the intended recipient, any disclosure,
> copying, distribution, or any action taken or omitted to be taken in
> reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
> accept liability for statements made which are clearly the senders own.





From michael at stroeder.com  Fri Feb 22 08:36:53 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 22 Feb 2002 14:36:53 +0100
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org> <7xlmdmfl6a.fsf@ruckus.brouhaha.com> <3C763007.78FEA7FF@stroeder.com> <7xk7t5lmrq.fsf@ruckus.brouhaha.com>
Message-ID: <3C764975.3D8E9D72@stroeder.com>

Paul Rubin wrote:
> 
> Michael Str?der <michael at stroeder.com> writes:
> > 1. Detect session hijacking by cross-checking CGI-BIN vars you can
> > assume to be constant throughout the whole session for *each* hit.

> Any particular cgi vars you recommend?

See the list in
http://www.stroeder.com/pylib/PyWebLib/pydoc/pyweblib.session.html

You can configure the env vars you want to include in cross-checking
by setting parameter crossCheckVars of WebSession.__init__(). The
vars actually cross-checked are determined when
WebSession.newSession() is called and might be a subset of the
crossCheckVars.

My standard disclaimer regarding PyWebLib:
I don't promote PyWebLib to be used by others. It's just published
for informational purpose. If you decide to use it be warned that
documentation will stay very sparse and the API might be still
subject to incompatible changes without prior announcement and
discussion.

> > If you want to be more secure use SSL on a server properly handling
> > SSL session renegotiation and exposing the SSL session ID in the
> > CGI-BIN env vars (e.g. Apache with mod_ssl -> env var
> > SSL_SESSION_ID).
>
> You can't count on the SSL session ID staying constant.
> Some browsers seem to like re-opening the SSL connection every now and then.

Yes, it depends on the server *and* the browser whether SSL session
renegotiation is properly done or not. My experiences with Apache
and mod_ssl and various mainstream browsers are fairly good. I
stated that you can't always rely on this - somewhat implicit
though.

> > If you want to be really secure or you can't rely
> > on proper SSL session renegotiation deploy client certificates and
> > use the subject DN and issuer DN exposed in the env vars.
>
> And the idea of making people
> enroll client certificates in order to spare them the indignity of
> accepting browser cookies sounds like something from another planet

I mentioned client certificates for describing the high-end of
security mechanism available.

> (no offense intended).

No offense taken.

> > 2. Have a URL redirector in place *within* your application which
> > outputs HTML with redirect URL in <head> section. In this case the
> > browser sends the redirector's URL as referer URL.

> > it's not feasible for inline-displaying of images via
> > <img> tag.
> 
> Interesting!  You're saying if I have a page called abc.html,
> containing an html meta redirection tag to xyz.html in the head
> section, then the browser will immediately navigate to xyz.html before
> trying to get any inline images, and also, even after the redirection,
> abc.html will show up as the referer on any new links or images?

IMHO it does not work with redirecting <img> links as already stated
in my prior posting. I think the image won't get loaded at all. Feel
free to test that.

> And
> that's true in general, not an implementation kink of some particular
> browser?  I didn't know that--thanks!

I tested it with some mainstream browsers. Feel free to contribute
more test results.

Ciao, Michael.


From jkraska at san.rr.com  Sat Feb  2 19:11:43 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 00:11:43 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org>
Message-ID: <if0p5u8lu6t4qnms5bneu808lnh745997u@4ax.com>

>	} else {
>      }
>

I _always_ put braces like this

something
{
}
somethingelse
{
}

Why? So that when I hit the % key in vim, it takes me to the exact same
indentation level brace, giving me a clue whether or not I matched the
right one or not (which can get screwed up if the wrong thing is commented
out, dontcha know).

C//



From phd at phd.pp.ru  Thu Feb 21 08:37:52 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 21 Feb 2002 16:37:52 +0300
Subject: Bug in e-mail package?
In-Reply-To: <3fq97uoqvtbq4mkf9p6hjl5vereg30n93u@4ax.com>; from paul@nlhas.no on Thu, Feb 21, 2002 at 12:51:32PM +0000
References: <3fq97uoqvtbq4mkf9p6hjl5vereg30n93u@4ax.com>
Message-ID: <20020221163752.A393@phd.pp.ru>

On Thu, Feb 21, 2002 at 12:51:32PM +0000, paul at nlhas.no wrote:
> AttributeError: 'int' object has no attribute 'lower'

   There was a bug in the email package, and it was fixed in the CVS.
Download latest code from mimelib.sf.net.

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



From greg at cosc.canterbury.ac.nz  Wed Feb 13 23:39:13 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 14 Feb 2002 17:39:13 +1300
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <eppstein-A87A6C.14430710022002@news.service.uci.edu> <a46uqm$gc7$1@peabody.colorado.edu> <eppstein-D7569F.18475010022002@news.service.uci.edu>
Message-ID: <3C6B3F71.EFCEACD6@cosc.canterbury.ac.nz>

Fernando P?rez <fperez528 at yahoo.com>:

> Indeed, it's just the Spanish form of the 'ist' English suffix (pianist,
> violinist, etc.) It simply means 'who performs an activity'.

Something about the sound of the Spanish form to me
suggests passion and enthusiasm, in a way the English
form doesn't. So Pythonista is quite appropriate!

Steven Rumbalski <srumbalski at prodigy.net>:

> 'Guidoista': three syllables or four? GWEE-DOY-STAH or GUI-DOH-EE-STAH?

Not three, please! It makes it sound Russian, which
seems rather brutal to my ears.

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


From robin at jessikat.fsnet.co.uk  Sun Feb 17 06:15:53 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 17 Feb 2002 11:15:53 +0000
Subject: Directed Acyclic Graph Representation
References: <j9du6uckr4g1am3780qtig50jdhcaf6fcr@4ax.com>
Message-ID: <Rds1QLApD5b8Ewia@jessikat.fsnet.co.uk>

In article <j9du6uckr4g1am3780qtig50jdhcaf6fcr at 4ax.com>, Tim Roberts
<timr at probo.com> writes
>Is anybody aware of a reasonable package for representing a directed
>acyclic graph in Python, and for rendering that graph in some meaningful
>way?  I need to build a dependency tree analyzer for, for example,
>determining the necessary link order for a set of object libraries.
>--
>- Tim Roberts, timr at probo.com
>  Providenza & Boekelheide, Inc.
GvR did an essay called "Python Patterns - Implementing Graphs". It
should be somewhere at www.python.org. As I recall he did stuff using
edgelists implemented using dictionaries.
-- 
Robin Becker


From jim.vickroy at noaa.gov  Mon Feb  4 13:32:01 2002
From: jim.vickroy at noaa.gov (jim.vickroy)
Date: Mon, 04 Feb 2002 11:32:01 -0700
Subject: working with files with the same extenstion
References: <3C5EAF50.80004@rmb.co.za>
Message-ID: <3C5ED3A1.EE2DF0B7@noaa.gov>

Welcome, I hope you enjoy Python as much as I have.

You may wish to look at the the listdir() procedure in the os (operating
system) module to generate a list of files.

If you are using Python 2.1 or later (not sure about Python 2.0), string
data types have an associated endswith() procedure.  For example:

> filename = 'anything.txt'
> filename.endswith('.txt')
1  # true
> filename endswith('.doc')
0 # false

String slicing is also an option.  For example:
> filename[-4:]
'.txt'
> filename[:-4]
'anything'

For more sophisticated pattern matching, look at the re (regular
expressions) module.

Ulrich Pfisterer wrote:

> Can anyone please tell me how I would go about doing the following.
> I would like to write a script that would take in a command line
> parameter "*.txt". The script would then loop through every file with a
> "txt" extension and something with it.
>
> I am brand new to python (I started using it 2 days ago) so I dont know
> if there is a module that does already this, or whether I have to do
> all the wildcard matching etc myself.
>
> Any help would be much appreciated



From gtalvola at nameconnector.com  Fri Feb  1 14:26:11 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Fri, 1 Feb 2002 14:26:11 -0500 
Subject: pickle security
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E4124087@MAILBOX.nameconnector.com>

I'm trying to understand how to safely unpickle untrusted strings.  I'll
summarize what I think I've learned from Google searches, and someone tell
me if I'm wrong.

If you want to unpickle a string s but you only want built-in types to be
created during the unpickling process, never classes or extension types,
here's the best way to do it:

import cPickle, cStringIO
unpickler = cPickle.Unpickler(cStringIO.StringIO(s))
unpickler.find_global = None
result = unpickler.load()

And if you want to allow particular classes to be unpickled, for example if
you want to allow the class "bar" in the module "foo" to be unpickled, you
can use something like:

import cPickle, cStringIO
def find_global(module, klass):
    if module == 'foo' and klass == 'bar':
        import foo
        return foo.bar
    else:
        raise cPickle.UnpicklingError, \
              "can't unpickle a %s.%s" % (module, klass)
unpickler = cPickle.Unpickler(cStringIO.StringIO(s))
unpickler.find_global = find_global
result = unpickler.load()

Did I get it right?  Are there any other security issues I need to be aware
of, or does this cover them?

-- 

- Geoff Talvola
  gtalvola at NameConnector.com



From jlh at cox.net  Thu Feb 21 00:08:05 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Thu, 21 Feb 2002 05:08:05 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu>
Message-ID: <Vg%c8.4179$XT1.116344@news1.east.cox.net>

see:
http://mail.python.org/pipermail/python-list/2001-August/060976.html
for some info.   Unless you have total control over what is being eval'd you
are at risk.
-Jeff

"Rajarshi Guha" <rxg218 at psu.edu> wrote in message
news:a50udo$1dou at r02n01.cac.psu.edu...
> On Wednesday 20 February 2002 01:24 in comp.lang.python Jeff Hinrichs
wrote:
>
> > If you wanted to get away from the dangerous eval, you could put your
> > functions inside of a class and then,
>
> Why is eval dangerous?
>
> --
> -------------------------------------------------------------------
> Rajarshi Guha                  | email: rajarshi at presidency.com
> 417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
> Dept. Of Chemistry             | ICQ  : 123242928
> Pennsylvania State University  | AIM  : LoverOfPanda
> -------------------------------------------------------------------
> GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A
> Public Key     : http://pgp.mit.edu/
> -------------------------------------------------------------------
> There is no truth to the allegation that statisticians are mean.
> They are just your standard normal deviates.
>




From Scott.Daniels at Acm.Org  Fri Feb  1 21:30:30 2002
From: Scott.Daniels at Acm.Org (Scott David Daniels)
Date: Sat, 02 Feb 2002 02:30:30 GMT
Subject: comparing DateTime instances
References: <NCBBLFCOHHDIKCAFGCFBKEFPKAAA.mark@mceahern.com> <mailman.1012605243.26055.python-list@python.org>
Message-ID: <1103_1012617030@news.dsl-only.net>

My technique for remembering what's what is:

	cmp(a,b) ? 0   <=>   a ? b
So:
	cmp(a,b) == 0  <=>  a == b
	cmp(a,b) < 0   <=>   a < b
	cmp(a,b) <= 0   <=>   a <= b
..






From martin at v.loewis.de  Wed Feb 27 04:09:58 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 10:09:58 +0100
Subject: PEP 263 comments
References: <mailman.1014762060.30054.python-list@python.org>
Message-ID: <m3u1s31eah.fsf@mira.informatik.hu-berlin.de>

"Jason Orendorff" <jason at jorendorff.com> writes:

> Counter-proposal:
> 
>  - Comment syntax: none.
>  - UTF-8 file signature: not supported.
>  - Python source code encoding: must always be UTF-8.
>  - Implementation: within the parser, everything's just
>    ordinary UTF-8 bytes.
>  - IDLE: always save UTF-8 unless otherwise directed.

Do you seriously want to pursue this route? If so, how do you want to
deal with backwards compatibility? Currently, you can put arbitrary
bytes in character strings, and people make use of this opportunity
(even though the documentation says this is undefined).

Will you reject a source module just because it contains a latin-1
comment?

Regards,
Martin


From Nick_Thomas at AdSQ.com  Sun Feb  3 21:55:19 2002
From: Nick_Thomas at AdSQ.com (Nick Thomas)
Date: Mon, 4 Feb 2002 02:55:19 -0000
Subject: Advertising Square Closing down - Cheap 4 Letter domain available!
Message-ID: <E16XZHr-0006pc-00@rhenium>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020204/c1c41436/attachment.html>

From jmsun at bioeng.ucsd.edu  Thu Feb 14 18:51:27 2002
From: jmsun at bioeng.ucsd.edu (Jeffrey)
Date: 14 Feb 2002 15:51:27 -0800
Subject: Highlights & Cursors
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com>
Message-ID: <3daa1aa8.0202141551.1dae3651@posting.google.com>

Lawrence Oluyede <rhymes at myself.com> wrote in message news:<0jun6ukqqtq58t1dh7ustpu06patd71d2s at 4ax.com>...
> I'm just learning Tkinter but I have a little problem.
> Can anyone of you explain me how to use the highlight
> for the focus in a widget and the cursor option?
> 
> Thanks in advance.

  I'm not quite sure I understand your question but I'll try to answer
your question.  The highlight on a widget occurs whenever one
mouses-over the widget.  The color that you what the highlight to be
depends on what you configure it to be .  For example if you have a
widget called some_widget, you would do something like
some_widget.configure( highlightcolor = 'red' ).

  In terms of cursors, that's a different story.  Note that this is
off the top of
my head.  But I'll assume that you are tyring to change the cursor? 
For the same widget, what you do is (this is off the top of my head)
some_widget.bind("Any-Enter", change_cursor_enter ) where
change_cursor is a function that changes the cursor to some other
special cursor.  Then you have to bind the other way around too like
some_widget.bind("Any-Leave", change_cursor_leave )....

  I hope I haven't been too confusing, and that I've been somewhat
helpful.
        
     Jeff


From peter at engcorp.com  Mon Feb  4 01:25:27 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 04 Feb 2002 01:25:27 -0500
Subject: what about python ??
References: <832e3897.0202032107.25b3e70f@posting.google.com>
Message-ID: <3C5E2957.E3B927F6@engcorp.com>

lennoncao wrote:
> 
> pls,tell me what about python and where it can been used!

Have you visited http://www.python.org yet?  Looking there,
or doing a search on google, will provide lots of answers to your
question.

Do you have some specific needs, or was this just a general
inquiry out of curiosity?

-Peter


From threeseas at earthlink.net  Mon Feb 11 00:40:46 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 11 Feb 2002 05:40:46 GMT
Subject: Three primary user Interfaces >Re: Please stop responding to Rue
References: <a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com> <5659.806T1220T5243918threeseas@earthlink.net> <YsC98.16409$Zu6.75381@news-server.bigpond.net.au>
Message-ID: <2082.807T1534T304260threeseas@earthlink.net>

On 10-Feb-02 17:27:04 Neil Hodgson <nhodgson at bigpond.net.au> wrote:
>Timothy Rue:

>> MS DOS began with a Command Line Interface, then GUI but never did it
>> evolve it's Command Line Interface to have what is considered standard
>> functionality like command line history

>   Command line history became available with the DOSKEY command in MSDOS
>5.0 released 1991.

I was probably using something better then, like 4dos.

>> The Mac began it's user interfaces with the GUI, never really having a
>> Command Line Interface - Though Now it might have one in OSX.

>   There was the MPW Shell available to programmers. Even back in 1984 there
>were primitive command line shells for the Mac although the OS did not
>support many of the concepts, such as redirectable streams, that make
>command lines useful.

<shrug> only counts if user oriented. User oriented is one of the
requirements to being a User Interface....

>> Its third
>> user interface is limited or constrained to applescript.

>   It can be used from C or Python or UserLand Frontier or any other OSA
>(Open Scripting Architecture) supporting language.

Don't really know much about Mac, cept for having run an Mac emulator on
Amiga and I do recall having downloaded and tried out Frontier. I believe
I found it then not so user friendly.

>   Neil

But the target is of course having all three in a manner that they are
compatable with each other and the typical user.

Enough with the claim that users are to stupid to use such things when
they have proven to do much more complex things. Enough so that if given
the tools to automate what they do in a repeatable manner, they can
certainly automate their repeatable actions, and would.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From philh at comuno.freeserve.co.uk  Sun Feb  3 21:00:10 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 02:00:10 +0000
Subject: Attention Phil Hunt
References: <mailman.1012327189.31750.python-list@python.org> <585.799T42T11732924threeseas@earthlink.net>
Message-ID: <slrna5rqpa.19d.philh@comuno.freeserve.co.uk>

On Mon, 04 Feb 2002 00:00:59 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 29-Jan-02 12:58:27 Laura Creighton <lac at strakt.com> wrote:
>>Every minute you spend arguing here is a minute that the release of
>>Herbivore is delayed.
>
>protecting once psuedo position can be very important.

ITYM "protecting one's psuedo position can be very important"

In what way is my psotion "pseudo"? Are you suggesting I didn't
originate the herbivore projevct or write Herbrip? If so, who did?

Or did you mean something else entirely?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>






From logiplexsoftware at earthlink.net  Fri Feb 15 14:43:46 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 15 Feb 2002 11:43:46 -0800
Subject: certification (Brainbench)
In-Reply-To: <dsVb8kKkXgXB092yn@bellatlantic.net>
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>
	<a4j5hd$l3j2$1@ID-11957.news.dfncis.de>
	<dsVb8kKkXgXB092yn@bellatlantic.net>
Message-ID: <20020215114346.04afd84c.logiplexsoftware@earthlink.net>

On Fri, 15 Feb 2002 14:01:49 -0500
Dr. David Mertz wrote:

> |"Trung Hoang" <thoa0025 at mail.usyd.edu.au> wrote in message
> |> i was wondering if i can get certified with python from an
> |> internationally recognised company?

> That said, the original question was from someone who wanted a
> certificate.  You can certainly take the Brainbench exam, and get a
> piece of paper out of the process.  There's not necessarily a need to
> forward my criticisms to your potential employer :-).

But then, how many certificates are really worth the paper they're printed
on?  They don't make good candidates better and they don't screen the poor
ones.  In short, they do nothing but generate revenue for companies who
issue certificates.  

My personal feeling is that employers who demand these certificates are
usually the type who demand that projects be developed using "standard"
tools, i.e. whatever tool they've heard a lot about or gotten a press
release on.  If they don't want to hire me because I don't have a
certificate then I don't want to work for them because they haven't got a
clue ;)

I guess maybe certifications *do* work in some sense: they help me screen
retarded employers ;)

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From djc at object-craft.com.au  Thu Feb 14 07:43:56 2002
From: djc at object-craft.com.au (Dave Cole)
Date: 14 Feb 2002 23:43:56 +1100
Subject: Sybase module 0.33pre3 released
Message-ID: <m3r8no8c8j.fsf@ferret.object-craft.com.au>

What is it:
The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

Note:

Since the 0.32 release I have come to the realisation that the Sybase
dynamic SQL functionality is not nearly general enough to use for
correct DB-API functionality.  I have thrown up my hands and gone back
to the drawing board with the high level Sybase.py module.

Status of 0.33pre2:

The 0.33pre2 release seems to be quite solid.  I have not received any
complaints from any of the 10 (count them) people who downloaded the
module.  Ben Gertzfield managed to implement a multi-threaded whois
server on top of the 0.32pre2 release.

Changes for 0.33pre3:

1) Release 0.32pre2 changed the paramstyle to 'format'.  Shai Berger
   sent me some valuable advice on how to continue to send binary
   parameters to the server without using ct_dynamic() so the
   paramstyle is now 'named'.  This means that you can now do this:

     c.execute("select * from titles where title like @arg",
               { '@arg': 'The %' })

2) You can send arguments to Cursor.callproc() the old way:

     c.callproc('sp_help', ['titles'])

   or the new way:

     c.callproc('sp_help', { '@objname': 'titles' })

3) Some internal cursor state constants were privatised (via leading
   '_').  Sadly no money was raised in the process and the conditions
   of the privatisation contract are commercial-in-confidence.

4) You can now compile the extension module without bulkcopy support
   by doing this:

     python setup.py build_ext -U WANT_BULKCOPY
     python setup.py install

5) The default build does not do threading any more so if you want
   threading enabled you will have to do this:

     python setup.py build_ext -D WANT_THREADS
     python setup.py install

6) There is some extra debugging detail for CS_DATAFMT which means you
   now get messages like this:

      >>> import Sybase
      >>> db = Sybase.connect('SYBASE', 'sa', '', 'pubs2')
      >>> db._conn.debug = 1
      >>> c = db.cursor()
      ct_cmd_alloc(conn0, &cmd) -> CS_SUCCEED, cmd1
      >>> c.callproc('sp_help', {'@objname': 'titles'})
      ct_command(cmd1, CS_RPC_CMD, "sp_help", CS_NULLTERM, CS_UNUSED) -> CS_SUCCEED
      ct_param(cmd1, &databuf0->fmt=[name:"@objname" type:CS_CHAR_TYPE status:CS_INPUTVALUE format:CS_FMT_NULLTERM count:1 maxlength:7], databuf0->buff, 6, 0) -> CS_SUCCEED
      ct_send(cmd1) -> CS_SUCCEED
      ct_results(cmd1, &result) -> CS_SUCCEED, CS_ROW_RESULT
      ct_res_info(cmd1, CS_NUMDATA, &value, CS_UNUSED, NULL) -> CS_SUCCEED, 3
      ct_describe(cmd1, 1, &fmt) -> CS_SUCCEED, datafmt0=[name:"Name" type:CS_CHAR_TYPE status:CS_UPDATABLE format:CS_FMT_UNUSED count:0 maxlength:30]
      ct_bind(cmd1, 1, &datafmt0->fmt=[name:"Name" type:CS_CHAR_TYPE status:CS_UPDATABLE format:CS_FMT_UNUSED count:1 maxlength:30], databuf1->buff, databuf1->copied, databuf1->indicator) -> CS_SUCCEED, databuf1
      ct_describe(cmd1, 2, &fmt) -> CS_SUCCEED, datafmt1=[name:"Owner" type:CS_CHAR_TYPE status:48 format:CS_FMT_UNUSED count:0 maxlength:30]
      ct_bind(cmd1, 2, &datafmt1->fmt=[name:"Owner" type:CS_CHAR_TYPE status:48 format:CS_FMT_UNUSED count:1 maxlength:30], databuf2->buff, databuf2->copied, databuf2->indicator) -> CS_SUCCEED, databuf2
      ct_describe(cmd1, 3, &fmt) -> CS_SUCCEED, datafmt2=[name:"Type" type:CS_CHAR_TYPE status:48 format:CS_FMT_UNUSED count:0 maxlength:22]
      ct_bind(cmd1, 3, &datafmt2->fmt=[name:"Type" type:CS_CHAR_TYPE status:48 format:CS_FMT_UNUSED count:1 maxlength:22], databuf3->buff, databuf3->copied, databuf3->indicator) -> CS_SUCCEED, databuf3
           
7) There is some initial work to support FreeTDS.  One of the problems
   with FreeTDS is that it does not support inline error handling.
   The callback error handling works well if you do not compile for
   threads support, but causes problems when compiled for threading.

   The problem occurs when the extension module releases the global
   interpreter lock when calling the Sybase CT library.  A Python
   callback causes the interpreter to be reentered by the callback
   which leads to bad things when the following warning from the
   Python include file ceval.h is violated:

     WARNING: NEVER NEST CALLS TO Py_BEGIN_ALLOW_THREADS AND
     Py_END_ALLOW_THREADS!!!

   At the moment the module will compile for FreeTDS (at least it does
   on my machine) and some things do work.  All you have to do is
   compile like this:

     python setup.py build_ext -D HAVE_FREETDS
     python setup.py install

   There is no error handling and reporting for FreeTDS yet.

8) The was a minor change to setup.py for:

   a) Solaris.  It no longer staticly links the libcomn.a Sybase
      library.

   b) Linux.  No longer looks for both libsybtcl and libtcl.  Now
      looks for libinsck.

Extra Pleading:

Please download the pre-release version, try it out, and then send me
some mail telling me what you found.  You would be surprised at how
little mail I get regarding the Sybase module, I really do need your
help.  If more than 10 people tried the module we would all benefit...

The module is available here:

    http://www.object-craft.com.au/projects/sybase/sybase-0.33pre3.tar.gz

The module home page is here:

    http://www.object-craft.com.au/projects/sybase/

- Dave

-- 
http://www.object-craft.com.au


From mstrayer1 at attbi.com  Tue Feb 26 19:46:40 2002
From: mstrayer1 at attbi.com (Michael Strayer)
Date: Wed, 27 Feb 2002 00:46:40 GMT
Subject: Python 2.2 Socket Test Failure
Message-ID: <Q%Ve8.8688$Dv1.29388@rwcrnsc53>

Hello..
I'm new to both Python and Linux so I'm not quite
sure what the following errors are trying to tell me.

I've built Python 2.2 on a Red Hat 7.2 system and when
I run make test I get the following error:

test test_socket crashed -- socket.gaierror: (-2, 'Name or service not 
known')
test_socket_ssl
test test_socket_ssl skipped -- Use of the `network' resource not enabled
test_socketserver
test test_socketserver skipped -- Use of the `network' resource not enabled

Could someone explain how to fix the socket crash.  Also how do I
enable to network resource?

Thanks
Mike Strayer



From bokr at oz.net  Sat Feb  2 18:36:17 2002
From: bokr at oz.net (Bengt Richter)
Date: 2 Feb 2002 23:36:17 GMT
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu> <B8809EB2.411E%jonathan@onegoodidea.com> <a3fisr$9j6$0@216.39.172.122> <B881642D.41AD%jonathan@onegoodidea.com>
Message-ID: <a3ht5h$2o0$0@216.39.172.122>

On Sat, 02 Feb 2002 09:38:53 +0000, Jonathan Hogg <jonathan at onegoodidea.com> wrote:

>On 2/2/2002 2:28, in article a3fisr$9j6$0 at 216.39.172.122, "Bengt Richter"
><bokr at oz.net> wrote:
>
>> Rule 17 says maybe the above would work if you *don't* setuid it, but I don't
>> think
>> you need it, unless maybe mod Python could otherwise get inbetween? I haven't
>> set that
>> up myself.
>
>Yes, perhaps I should have made that more clear. I meant using 'suexec'
>*instead* of randomly setuiding executables. With mod_python it's even
>harder as the Apache process itself needs to be running as the user
>(requires the new 2.0 per-user pooling magic), but I don't think that's the
>problem.
>
>'suexec' is a much more secure mechanism for running cgis as a particular
>user. The user is set in the config file (per virtual host) and suexec has a
>number of rules that it checks before it will run the script as that user
>(as Bengt refers to) - such as that it is owned by the user, isn't writeable
>by anyone else, is within a particular cgi directory (this is the main
>problem with suexec as it is only one path per installation, so I've had to
>set it to something useless like '/' in the past to make it work with
>disparate cgi directories).

IME, you can use ScriptAlias to set suexec cgi directories on a per-virtual
host basis. I have one test VH running cgi out of '/', another with cgi
directory named other than 'cgi-bin', and others from the usual '/cgi-bin',
all running as account owner. See

    http://httpd.apache.org/docs/mod/mod_alias.html#scriptalias

Perhaps one path per installation applies to another version or mode of use?

Regards,
Bengt Richter



From paul at zope.com  Mon Feb 11 08:55:27 2002
From: paul at zope.com (Paul Everitt)
Date: Mon, 11 Feb 2002 08:55:27 -0500
Subject: Zope cookbook?
References: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au>
Message-ID: <3C67CD4F.3070503@zope.com>

Hello Dave.  I think you are looking for the recipes at ZopeLabs:

   http://www.zopelabs.com/

--Paul

djmitchell wrote:
> Is anyone working on a Zope cookbook, or know of one that already exists?
> 
> I'm pretty comfortable with Python (+ Perl, VB, C++, Java/J2EE, ...) and 
> building Web apps in general, but making the leap to Zope seems pretty 
> tough due to an apparent lack of "here's how to do simple, common task X" 
> type of documentation.  I've got an app where a Zope solution looks like a 
> perfect fit, but churning out code feels like pulling teeth - I'm agonizing 
> over every line, and convincing myself there's easier/simpler/faster ways 
> to do just about everything.  I'm also finding myself reading through other 
> peoples' source code almost constantly - usually enlightening, but a REALLY 
> slow way of learning.
> 
> I've give a specific example of one thing I need to do: I want to build 
> something analogous to an ActiveX control or a JavaBean, that can be used 
> to input a date.  It needs to check for valid data being input (e.g. no 
> "31-February-2002"); I might want it to be displayed using some cute DHTML 
> if the browser is IE6, or maybe something dumber/simpler like pull-down 
> text boxes if the browser is Netscape 2; it might also need to only accept 
> dates in a certain range (e.g. between "12-January-2002" and 
> "15-January-2002") that I specify; it might need to display a default date; 
> based on the user's geographic location, I might want entered dates to be 
> displayed on screen as "15-January-2002", "15-1-2002", "2002-1-15", or 
> "1/15/2002".  It needs to be a logical "function" that I can call from 
> within an HTML form, and assign the (validated) date entered by the user to 
> a variable of some sort.  Ignoring the distinct possibility that I might be 
> able to download exactly this "ZopeBean" from somewhere, I'm sure there's a 
> "normal" way of coding this within Zope (DTML method?  Python script?), but 
> I just don't know where I should be looking.  Oh, and I don't want to 
> resort to either ActiveX or Java to get the job done ;->
> 
> Any tips, pointers, advice, useful sources of info, etc.?
> 
> Thanks in advance
> 
> Dave Mitchell
> 






From phd at phd.pp.ru  Sun Feb 10 10:53:36 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sun, 10 Feb 2002 18:53:36 +0300
Subject: cgi authoriszation
In-Reply-To: <q4Sr5IAZ4cZ8EwGa@jessikat.fsnet.co.uk>; from robin@jessikat.fsnet.co.uk on Sun, Feb 10, 2002 at 01:34:17AM +0000
References: <q4Sr5IAZ4cZ8EwGa@jessikat.fsnet.co.uk>
Message-ID: <20020210185336.A17561@phd.pp.ru>

On Sun, Feb 10, 2002 at 01:34:17AM +0000, Robin Becker wrote:
> client's response. Is apache filtering out the credentials? 

   Yes. Apache filter out "Authorization" header, but passes "REMOTE_USER"
environment variable. Though I'm sure it's passed by mod_auth_* modules, so
you may need to stick to Apache authorization modules, or write your own.

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



From dsh8290 at rit.edu  Tue Feb  5 22:25:41 2002
From: dsh8290 at rit.edu (dman)
Date: Tue, 5 Feb 2002 22:25:41 -0500
Subject: Defending the Python lanuage...
In-Reply-To: <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>
References: <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>
Message-ID: <20020206032541.GA11067@dman.ddts.net>

On Mon, Feb 04, 2002 at 08:43:00AM -0500, Steve Holden wrote:
| "Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
| [editor stuff]
| >
| > I suppose it might be a brain thing. I use my left hand for writing and
| > drawing while I use my right hand for more coarse-grained tasks, like
| > throwing a ball and such.
| 
| Strange. I don't come across many other people who do this.

I am "left handed" -- that is, I write better and more naturally with
my left hand -- but I use both hands for various things and sometimes
switch back and forth.  I always use a mouse with my right hand, but
scissors with my left.  I can use a hammer or a hole punch (yeah, I
worked in a video store with those annoying punch cards) or a
handgun-style barcode scanner with either hand.  The oddest thing wrt
to my handedness is that I dribble a basketball with my right hand,
but shoot with my left.  It's natural that way.

(not that many, if any, of you are interested in this line noise ;-))

-D

-- 

A man of many companions may come to ruin,
but there is a friend that sticks closer than a brother.
        Proverbs 18:24




From DeepBleu at DeepBleu.org  Thu Feb 28 16:06:13 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Thu, 28 Feb 2002 15:06:13 -0600
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net>
Message-ID: <u7t7793khgum56@corp.supernews.com>

Deitel and Deitel book on Python?  I personally would not touch it.  Give me
a book written by Lutz, Holden or Hammond any time.  But not from the
establishment of Deitel.  I bet you they do not even know how to program in
Python.  All their books are churned out by what resembles an
industry-complex with standard issue.
Long ago I discovered a mistake in their old C++ How to Program.  I sent
them an e-mail.  The answer still brings vivd memories of when you feel like
throwing up.  Arrogant bunch of oppurtunists who are not worth the price of
the paper their books are printed on.
You want books on Python:  Buy Beazley, Lundh, Holden, Hammond, Greyson
etc....
Leave Deitel alone!
DeepBleu

"Ron Stephens" <rdsteph at earthlink.net> wrote in message
news:3C7D8DCA.119D52EC at earthlink.net...
> Sorry but I can't wait to give a brief "preview review" of the new
> Deitel and Deitel book, "Python How to Program" (I also posted this at
> my website at http://www.awaretek.com/plf.html ).
> ...
> Well, let's see, I bought this 1292 page book from Amazon a week ago for
> $72. This is the most I have ever paid for a computer book. Heck, it may
> well be the most I have ever paid for *any* book ;-))).
>
> I am leaving on a long business trip to Asia in two days, and I just
> can't wait until I get back to tell you how much I am enjoying this
> book. I have read the first 340 pages, which cover the core language; I
> intend to carefully read the remaining chapters, which cover various
> more advanced topics, as I fly around the world from
> New York City to Amsterdam, to Singapore, to Taiwan, to Japan, to Korea,
> on to Shanghai, China, and finally back home to New York. I am looking
> forward to the book more than the trip ;-))))
>
> This book is one of the most enjoyable I have ever read. The pedagogical
> approach that the authors use is superb; they have obviously learned
> their educational craft well; for me, it actually works. The quality of
> the paper is the finest in any Python book yet; the text is large and
> clear. The way actual code is highlighted and separated from the text is
> extremely
> helpful. Everything is first class.
>
> I actually sort of expected the book to be a second rate effort in the
> sense that the Deitels have written several best sellers about other
> languages, and I sort of expected them to half-heartedly do a me-too
> book by just applying the same old formula to a Python version.
>
> But the book is fresh and clearly thought out. Since I have never read
> any other Deitel book, it is possible that those who have done so may
> have a different experience.
>
> But I think even Deitel veterans will be pleased. More than anything
> else, the book reminds me of a great college textbook, and reading such
> a text is something I haven't done in way too many years. But this
> educational quality of the text brings it to a whole new level of
> sophistication, compared to most computer programming books.
>
> The book includes, at the end of each chapter,  extensive questions and
> answers, review lists, and summaries. It actually feels like you are
> studying for credit in a college course. The end result is that, after
> carefully reading the chapters covering the core language, I feel like I
> have consolidated my Python knowledge considerably and that I am both
> ready to do some really serious coding (once I get back from Asia ;-)))
> and also to successfully attack the advanced chapters while I am on the
> planes.
>
> In addition to excellent coverage of the core language, the book covers
> CGI, XML, Databases, XHTML (quite extensive) , Data Structures,
> Multimedia, OpenGL, Tkinter, CSS, Python Server, Networking, Threading,
> Regular Expressions, and Unicode. There are even excellent and extensive
> appendixes reviewing Octal, Hexadecimal, and Binary Number Systems;
> HTML; and Python version 2.2 Additions. Most of these topics are not
> just skimmed over but are covered in great detail. yet the book really
> flows well. The exercises and questions and answers are equal to the
> very best college text books  I have encountered. Even the summaries
> after each chapter are useful; I think I will be reviewing the chapter
> summaries on the plane and later again after a few weeks, in order to
> consolidate my learning even more. It seem like a good use of my time,
> because the summaries are well thought out and very helpful.
>
> I am excited.
>
> I will do a detailed and very thorough review of the whole book as soon
> as I return home from my trip. But even based on my reading of only the
> first 340 pages, this book is a winner and I think any
> Pythonista who is willing to part with the (admittedly huge) sum of $72,
> will be richly rewarded. One could easily get engrossed in this fine
> book and wind up absorbing all 1292 pages in 10 days of blissful mind
> mongering. (Well, what the hell, its only a preview review and I can
> coin an asinine phrase like "mind mongering " if I want to ;-))))
>
> ...
> By the way, I have studied the code of Hans Nowak's mygale.py well
> enough to be confident in making modifications and additions, and I am
> really looking forward to doing that as soon as I return  home. I intend
> to add more starting point sources for the article searches, to add the
> capability to search for any given topic, and also trying to fine tune
> the searching mechanisms. I even have some other ideas for the code, but
> I'll keep them quiet for now. Quite basic coding of course, the real
> work has already been done by Hans and I really appreciate his
> open-sourcing of this fine program.
>
> I'm still searching for more tutorials and I'm up to 24 links now. I
> have a lot of other stuff I want to add when I get back in mid March.
>
> Ron Stephens
> < a href = "http://www.awaretek.com/plf.html">Python City</a>




From phr-n2002a at nightsong.com  Fri Feb 22 07:32:57 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 22 Feb 2002 04:32:57 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org> <7xlmdmfl6a.fsf@ruckus.brouhaha.com> <3C763007.78FEA7FF@stroeder.com>
Message-ID: <7xk7t5lmrq.fsf@ruckus.brouhaha.com>

Michael Str?der <michael at stroeder.com> writes:
> 1. Detect session hijacking by cross-checking CGI-BIN vars you can
> assume to be constant throughout the whole session for *each* hit.
> If you want to be more secure use SSL on a server properly handling
> SSL session renegotiation and exposing the SSL session ID in the
> CGI-BIN env vars (e.g. Apache with mod_ssl -> env var
> SSL_SESSION_ID). If you want to be really secure or you can't rely
> on proper SSL session renegotiation deploy client certificates and
> use the subject DN and issuer DN exposed in the env vars.

Any particular cgi vars you recommend?  You can't count on the SSL
session ID staying constant.  Some browsers seem to like re-opening
the SSL connection every now and then.  And the idea of making people
enroll client certificates in order to spare them the indignity of
accepting browser cookies sounds like something from another planet
(no offense intended).

> 2. Have a URL redirector in place *within* your application which
> outputs HTML with redirect URL in <head> section. In this case the
> browser sends the redirector's URL as referer URL. It's not
> sufficient to redirect by HTTP status codes 301 or 302 and Location:
> header! Off course it's much work to keep track of all links you're
> displaying and it's not feasible for inline-displaying of images via
> <img> tag.

Interesting!  You're saying if I have a page called abc.html,
containing an html meta redirection tag to xyz.html in the head
section, then the browser will immediately navigate to xyz.html before
trying to get any inline images, and also, even after the redirection,
abc.html will show up as the referer on any new links or images?  And
that's true in general, not an implementation kink of some particular
browser?  I didn't know that--thanks!

> You're invited to play with the on-line demo of web2ldap at
> http://sites.inka.de:8002/web2ldap which implements these kind of
> things and places the session ID in PATH_INFO (because personally I
> do not like cookies although I see some of their advantages, e.g.
> single cross-site sign-on).
> 
> Feedback welcome.

I'll take a look at it!  

Paul


From logiplexsoftware at earthlink.net  Fri Feb 15 18:24:03 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 15 Feb 2002 15:24:03 -0800
Subject: basic books/guides on multithreading programming?
In-Reply-To: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
Message-ID: <20020215152403.179d014b.logiplexsoftware@earthlink.net>

On Sat, 16 Feb 2002 05:57:09 +0700
Tripp Scott wrote:

> which ones do you recommend? examples in any language is fine, 
> though i prefer python, perl, or ruby. what i would like is 
> guidance on what is thread safety, how to achieve it, what are 
> the issues in multithreading programming, the global interpreter 
> lock, and other basics.

It's been many years since I've read a book on multi-threading, so I can't
suggest a good title, but it's a lot simpler than you might think (given
the fact you're looking for a book ;)  This is especially true with Python,
which has a remarkably easy-to-use threading library (use the threading
module, not the thread module).

Thread-safety boils down to this: lock any /shared/ resource that might be
/modified/ by a thread.  End of story.  You can get into all sorts of
debates about when these locks are necessary, but 99% of the time they are,
and the other 1% isn't worth the effort to decide whether they really are
or not.

I suggest implementing something simple, say a thread-safe queue, hammer it
with dozens/hundreds of threads and see what you learn.  Believe it or not,
the standard Python documentation will probably be sufficient.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From jimd at vega.starshine.org  Thu Feb 21 04:35:42 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 21 Feb 2002 09:35:42 GMT
Subject: confusion regarding os.path.walk()
References: <mailman.1014078032.30409.python-list@python.org> <Xns91BA65E93C0ECandrewdarwinwarscom@194.168.222.19> <3C72244A.5704081A@imk.fraunhofer.de>
Message-ID: <a52f1e$111q$1@news.idiom.com>

In article <3C72244A.5704081A at imk.fraunhofer.de>, Joachim Kaeber wrote:

>Hi,

>Andrew Brown wrote:
>> The practical question I was left with was "how do you identify broken
>> symlinks with python?" I'd have liked a script that rm-ed broken and only
>> broken symlinks, and I can't figure out how to test for them.
>> os.path.islink(file) will tell me whether it's a symlink. But is there a
>> call to say what it's supposed to point at? Then I can test whether that
>> exists.

>Maybe os.stat is your friend:

>% ln -s /dev/null a
>% ln -s /dev/xxxx b

 Maybe.  That will let the underlying OS attempt
 to follow the chain of symlinks to it's terminus
 and (obviously, as below) raise and exception
 (from the ENOENT, EPERM or ELOOP or ??? error returned 
 by the system call on your particular OS).

 However, you might also want to use the os.path.os.readlink() 
 method (I think os.path.os looks odd, but I understand the idea, 
 it a subset of OS dependent functions which are a subset of the 
 overall os.path domain).  Using readlink() you can follow the
 symlink chain one link at a time.  You can also use os.path.os.lstat()
 to get inode details about the link *rather* than about the target
 of the link.  (Think of stat() as following the link chain and 
 returning the results of the real target, it dereferences the
 pointers, lstat() just returns data on the link (and any inode that
 stores the link's target, perms, timestamps, etc).  Note that 
 lstat() data (ownership and permissions) are mostly ignored by
 most versions of UNIX (including Linux).  So most of the lstat()
 data is totally useless to almost all applications.

 It's also helpful to remember that UNIX was well established before
 symlinks were added to it.  So it was vital that they be mostly 
 transparent to "legacy" applications and utilities of the time. Thus
 it makes perfect sense that they'd had stat(), open(), etc follow
 the symlinks and implement a new system call lstat() to provide 
 lower-level utilities (especially commands like cp, ls, and archivers
 like cpio, tar, and later pax) with the means to discriminate between
 the data (target) and metadata (symlink).

 Of course you probably still want to run that from within
 an exception block --- since there are many fussy reasons why
 the call might fail (following an islink() with a subsequent
 readlink() is inherently a race condition, for example).  

 In general all access to the fs should be run in an exception 
 catching block if you want your application to be gracefully 
 robust in the face of most failures.  Obviously your own little
 utilities can just die and report the problem, but application
 end-users don't like to see tracebacks polluting their pretty 
 UIs (even their text/curses UIs).

>Python 2.1 (#1, May 17 2001, 11:31:53)
>[GCC 2.95.3 19991030 (prerelease)] on linux2
>Type "copyright", "credits" or "license" for more information.
>>>> import os
>>>> os.stat("/tmp/a")
>(8630, 8, 6L, 1, 0, 0, 0, 0, 0, 0)
>>>> os.stat("/tmp/b")
>Traceback (most recent call last):
>  File "<stdin>", line 1, in ?
>OSError: [Errno 2] No such file or directory: '/tmp/b'

>HTH



From mdk at mdk.com  Fri Feb  1 13:22:21 2002
From: mdk at mdk.com (MDK)
Date: Fri, 1 Feb 2002 13:22:21 -0500
Subject: Help with time
References: <a3ecqt$16bk9d$1@ID-98166.news.dfncis.de> <3c5ad0bf$0$2200$7586b60c@news.frii.com>
Message-ID: <a3emd2$17tc99$1@ID-98166.news.dfncis.de>

"Marc Poulin" <mpoulin at verinet.com> wrote in message
news:3c5ad0bf$0$2200$7586b60c at news.frii.com...
> In article <a3ecqt$16bk9d$1 at ID-98166.news.dfncis.de>, "MDK" <mdk at mdk.com>
> wrote:
>
> > I am passing to a function how long in milliseconds I want to wait for
>
> > import time
> >
> > def getdata(ms):
> >     starttime = time.time()
> >     while len(data) < 20:
> >         now = time.time()
> >         print '%.3f seconds %.3f' % (now - starttime, timeout / 1000) if
> >         now - starttime > timeout/1000:
> >             break
> >     return data
> >
> > What am I doing wrong?
>
> Maybe a truncation problem (mixing ints and floats)?
>
> My Python interpreter says
> >>> print 100/1000
> 0
> >>> print 100/1000.0
> 0.1
> >>>

Yes, that was it. Thanks! I never would have thought of that.




From roger.day at globalgraphics.com  Wed Feb 27 08:07:29 2002
From: roger.day at globalgraphics.com (roger.day at globalgraphics.com)
Date: Wed, 27 Feb 2002 13:07:29 +0000
Subject: advocacy help ( Python vs VB )
Message-ID: <80256B6D.0048190F.00@notescam.cam.harlequin.co.uk>

I'm in a company where we are thinking of using Python seriously - mind you,
this company has a bespoke project written in Perl (some 20,000+ lines) which
quite turned people away (me included) from using Perl. So I don't think we're
"normal".

I like your idea of stealth deployment. I will be doing more of that in future.

At 26/02/2002 18:32:53, Stefan Kuzminski <SKuzminski at FairIsaac.com> wrote:
#  I suppose the fact that I get 20 emails in response to this question is a
#  sign, I know you couldn't get 20 positive third party comparisions of VB to
#  Python.  I've tried to sell Python in several organizations over the last
#  few years with varying degrees of success.  The *key* sticking point ( other
#  than general fear of the unknown ) is enevitably the 'well I can just go an
#  hire VB programmers but nobody knows this Python thing'.  Which of course is
#  absurd since the python learning curve is so small.  But the managers aren't
#  in any position to relate since all they control are resources with resumes
#  that say VB, C++ or Java...  ~sigh~  So my strategy so far has to been to
#  build the thing without ever mentioning Python until it's done..
#
#  Stefan
#
#
#
#  -----Original Message-----
#  From: Matthew Sherborne [mailto:miracle at paradise.net.nz]
#  Sent: Monday, February 25, 2002 5:21 PM
#  To: Stefan Kuzminski; python-list at python.org;
#  activepython at listserv.ActiveState.com; python-help at python.org
#  Subject: Re: advocacy help ( Python vs VB )
#
#
#  Google uses python to script their cluster(s).
#
#  http://www.google.com/jobs/openings.html
#
#  (Search for python on that page, I counted 10)
#
#  Also check out these URLs:
#
#  http://home.rmi.net/~lutz/advocacy.html
#
#  http://py-howto.sourceforge.net/advocacy/advocacy.html
#
#  GBU
#  Matthew Sherborne
#
#  ----- Original Message -----
#  From: "Stefan Kuzminski" <SKuzminski at FairIsaac.com>
#  To: <python-list at python.org>; <activepython at listserv.ActiveState.com>;
#  <python-help at python.org>
#  Sent: Tuesday, February 26, 2002 1:25 PM
#  Subject: advocacy help ( Python vs VB )
#
#
#  > I need to convince my co workers and erstwhile managers to build a
#  > production system in Python.  They're coming from a VB world and are
#  fearful
#  > to detach from the Redmond mothership.  I need hard facts, not warm and
#  > fuzzies about syntax but names of big companies using Python and things VB
#  > just *cannot* do ( threads, regular expressions.. ? ).
#  >
#  > Thanks,
#  > Stefan
#  > _______________________________________________
#  > ActivePython mailing list
#  > ActivePython at listserv.ActiveState.com
#  > To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
#  >
#  _______________________________________________
#  ActivePython mailing list
#  ActivePython at listserv.ActiveState.com
#  To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
#
Roger





From owen at astrono.junkwashington.emu  Mon Feb 11 13:44:58 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Mon, 11 Feb 2002 10:44:58 -0800
Subject: Add iterator to gzip files?
Message-ID: <a493fa$11ei$1@nntp6.u.washington.edu>

Anyone know of plans to add an iterator to the gzip module, so one can 
write:
for line in gzipfile:
    ...

It's trivial to change the class code or code up a subclass, e.g.:

class FixGZip(gzip.GzipFile):
    """the GzipFile class normally does not have an iterator; add one"""
    def __iter__(self):
        return self

    def next(self):
        line = self.readline()
        if line:
            return line
        else:
            raise StopIteration

def gzopen(filename, mode="r"):
    return FixGZip(filename, mode)

Similarly, I'm wondering if sockets will, or do, support iteration to 
read lines of data. (I've also wondered why they don't directly support 
read and write as unix sockets do; one can create a file from a socket, 
but that seems a bit indirect.)

-- Russell


From wurmy at earthlink.net  Wed Feb 20 23:45:00 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Thu, 21 Feb 2002 04:45:00 GMT
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <3C747268.6C756B9D@earthlink.net> <7xvgcrfosh.fsf@ruckus.brouhaha.com>
Message-ID: <3C747B81.367B5D50@earthlink.net>

Paul Rubin wrote:
> 
> Hans Nowak <wurmy at earthlink.net> writes:
> > > Why doesn't sort() return the sorted list.  I would like to chain it
> > > to other operations:
> > >         b=[x for x in a].sort()
> >
> > The other replies told you why list.sort() doesn't return
> > the sorted list. You can easily roll your own function,
> > though:
> >
> > >>> def sort2(lst):
> >       z = lst[:]
> >       z.sort()
> >       return z
> > ...
> Just write it like this:
> 
> def sort3(lst):
>   lst.sort()
>   return lst
> 
> The list comprehension already makes a temporary list.  You don't need
> to copy it around again.

True... in this case. He may not always use it with a
listcomp though, and then sort3 will sort the list in
place and return it.

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From tdelaney at avaya.com  Mon Feb 11 21:29:52 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 12 Feb 2002 13:29:52 +1100
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A17B@natasha.auslabs.avaya.com>

> From: Steven Rumbalski [mailto:srumbalski at prodigy.net]
> Philip Swartzleonard wrote:
> 
> > Then it becomes 'Guidoista' perhaps, for following a 
> specific leader? ;)
> 
> 'Guidoista': three syllables or four? GWEE-DOY-STAH or 
> GUI-DOH-EE-STAH?

Definitely 4 (how could it be 3???) but it's GUI-DOH-IS-TAH.

Tim Delaney



From db3l at fitlinxx.com  Mon Feb 25 23:25:09 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 23:25:09 -0500
Subject: DLL/Embeded interpreter wierdness
References: <qltl7uoh7q01drhb37b48bsmipr7h8kaam@4ax.com>
Message-ID: <uy9hgvpi2.fsf@ctwd0143.fitlinxx.com>

Courageous <jkraska at san.rr.com> writes:

> So. Let me see if I can phrase this right:
> 
> If I have an embedded interpreter sitting in a .dll and this
> self-same dll wants to be a module for import in scripts
> executed by the embedded interpreter, will this work?

Typically when I've done this (both embedded and extended in the same
DLL or executable), what I've done is automatically define my module
internally before calling out to the embedded interpreter.  The
differences between initialization in the two cases are very slight.

That way, if the embedded interpreter does do an "import xxxx" the
module will already be available and it won't try to load me again.
To the embedded code, the module appears as a built-in module.

Thus you can write scripts that use "import xxxx" and will resolve the
module properly whether started as an embedded script, or run to use
the module as an extension module.

Actually, in many cases, I execute the "import xxxx" automatically on
behalf of the embedded script before running it just to make it
simpler for the script writers (since the embedded cases are normally
simple extension scripts), but that's just a nicety.

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


From ajeru at vknn.org  Tue Feb 19 21:42:32 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 18:42:32 -0800
Subject: Adding static typing to Python
References: <mailman.1014129930.16624.python-list@python.org>
Message-ID: <24c39b2c.0202191842.63e7b9cf@posting.google.com>

Michael Chermside <mcherm at destiny.com> wrote in message news:<mailman.1014129930.16624.python-list at python.org>...
> No discussion of adding static typing to Python would be complete 
> without a reference to Guido's thoughts on the subject:
> 
>     http://www.python.org/~guido/static-typing/index.htm
> 
> I suspect that these may be a bit dated (~2 yrs old), but a few key 
> ideas should still be relevent:
> 
>     * Static typing should be optional.
>     * The 'purpose' of static typing is to ensure that type errors are
>       caught at compile time, not runtime (in checked code).
>     * Don't expect huge performance gains -- catching certain kinds of
>       errors earlier is the main gain.
> 
> Not that I fully agree, or want to tie Guido to old ideas, but those who 
> come to this idea fresh should realize it has been considered before. 
> The main results of the previous consideration are:
> 
>     * Performance isn't really the issue (though people often think it
>       is).
>     * Unit testing could (if used well) catch the same errors that static
>       type checking catches.
>     * Unit testing is a good thing.
>     * Python is a REALLY dynamic language, and one which tends to trust
>       the programmer.
> 
> -- Michael Chermside

That answers my initial question if there is some effort to include
static typing into Python. Thanks. What I miss in the definition of
the purpose is type based dispatching and multi methods. I wonder just
what happened to that type-SIG that's mentioned somewhere.

Alexander


From andrea at proceryon.at  Tue Feb  5 02:04:46 2002
From: andrea at proceryon.at (Andrea)
Date: 4 Feb 2002 23:04:46 -0800
Subject: Get current cursor position in ScrolledText
Message-ID: <4feefc55.0202042304.62cc5cc0@posting.google.com>

I've got following problem:

I have a ScrolledText in which i can insert different characters
(a,b,c, etc.).
On every seperate insert i need the cursor-position and the current
length of all characters together. But if i bind an event to every
possible key (e.g. <Key-A>) and i make a <ScrolledTextName>.get() i
get the content of the ScrolledText before the 'A'-insertion happens.
So the problem is if i paste a longer text i've no possibility to get
the real length after insertion.
Do anybody know how i can evade this problem?

TIA, Andrea


From m.faassen at vet.uu.nl  Tue Feb 19 10:23:04 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Feb 2002 15:23:04 GMT
Subject: My Spring project!  [A dictionary that allows substrings for key lookup]
References: <mailman.1014071250.31662.python-list@python.org>
Message-ID: <a4tqko$e60$1@newshost.accu.uu.nl>

Danny Yoo <dyoo at hkn.eecs.berkeley.edu> wrote:
[snip snip]
> I've written a wrapper around Dan Gusfield's 'strmat' Suffix Tree library,

>    http://www.cs.ucdavis.edu/~gusfield/strmat.html

> and I think it's all quite crazy.  *grin* If anyone's interested, I've
> collected my notes and code here:

>    http://hkn.eecs.berkeley.edu/~dyoo/programming/current.html

> As a warning: there are definitely memory leaks in the wrapper.  Also,
> parts of the 'strmat' library haven't been wrapped yet because they look a
> little... well, fragile.  But it's still very interesting, and perhaps
> even useful!

[example snipped]

Neat! I've become interested in doing quick full text searches like this
lately, though in my case things like this would be stored in the 
ZODB so I'm not sure exactly how useful your code is for me right now.

That said, it's pointing me in the right direction for further study. Thanks!

Regards,

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


From MaximilianScherr at T-Online.de  Thu Feb 14 18:48:16 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Thu, 14 Feb 2002 23:48:16 -0000
Subject: socket server interrupt?
In-Reply-To: <20020214.093631.260152959.25986@logiplex1.logiplex.net>
Message-ID: <a4hic0+jf3c@eGroups.com>

is there any possibiliy to close the server thread(s) (im using 
SocketServer)?

--- In python-list at y..., "Cliff Wells" <logiplexsoftware at e...> wrote:
> In article <mailman.1013690723.25180.python-list at p...>,
> "maximilianscherr" <MaximilianScherr at t...> wrote:
> > how can i make my socket server interrupt on ctr-c(i know with the
> > keyboardinterrupt exception, but how?)?
> 
> The problem you will have with this approach is that your server is
> multi-threaded; you won't know which thread will receive the 
interrupt.
> A better alternative is to start yet another thread that accepts 
console
> input and can shut the server down using a command, say 'q' or 'x'.
> 
> - Cliff
> -- 
> http://mail.python.org/mailman/listinfo/python-list




From shalehperry at attbi.com  Wed Feb 13 21:10:02 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 13 Feb 2002 18:10:02 -0800 (PST)
Subject: Opening random line from file with random.shuffle()
In-Reply-To: <87bses6g1p.fsf@uol.com.br>
Message-ID: <XFMail.20020213181002.shalehperry@attbi.com>

On 14-Feb-2002 synthespian wrote:
> 
> 
> Hello-
> 
>       I need to get a random line from a file. I read in thins NG about a
"poorly"
> documented random.shuffle(), but I can't get it to work:
> 

>>> f = open('/home/shaleh/apt.howto')
>>> lines = f.readlines()
>>> import random
>>> random.choice(lines)
'Apt is a suite of utilities that assist the admin in maintaining an\n'

random.choice() is really cool.  Give it any sequence (list, tuple, things that
act like a sequence, etc) and it chooses one item and returns it.  You could
ask for a random card from a list of cards, name in a list of names, etc.



From rhymes at myself.com  Sun Feb 17 09:37:52 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sun, 17 Feb 2002 15:37:52 +0100
Subject: Highlights & Cursors
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com> <3daa1aa8.0202141551.1dae3651@posting.google.com> <2apq6ucuh9n7s446rrcsdi2fu7eis0qaqu@4ax.com>  <13285ea2.0202151501.47080e91@posting.google.com> <mailman.1013954129.31154.python-list@python.org>
Message-ID: <40gv6u8ka3s0hsa667hjjemkapfeg1gfou@4ax.com>

On Sun, 17 Feb 2002 14:50:53 +0100, Laura Creighton <lac at strakt.com>
wrote:

> Get John Shipman's 84 page guide.
>http://www.nmt.edu/tcc/help/lang/python/tkinter.html
>This is on page 9

thanks i got it

----
Lawrence "Rhymes" Oluyede
rhymes at myself.com


From ayinger1 at pacbell.net  Sun Feb  3 20:14:40 2002
From: ayinger1 at pacbell.net (andrew)
Date: 3 Feb 2002 17:14:40 -0800
Subject: parsing an xml document with funky ascii characters
Message-ID: <be58dbe.0202031714.3ebee2df@posting.google.com>

Hi.

I am using sax parser in python 2.1.
How do I deal with xml documents with characters like '?'?

I have tried:
	- setting encoding="ISO-8859-1 in the xml doc itself
	- setting the InputSource encoding via:
source.setEncoding('ISO-8859-1')
	- escaping the character in the doc: ('\x84')
	- and, finally, encoding the parsed strings that have this character:
myString.encode("ISO-8859-1")

What I have found is that the default parser (appears to be expat,
retrieved from sax.make_parser) seems to store every element as
unicode strings.  It appears to store them incorrectly (so, '?'
appears in the unicode string as '\xe4' instead of '\x84').  The
result is that if I try to encode the unicode string that i get back
from the parser, the character in question incorrectly appears as 'E'
(sum).

Any ideas?  Am I doing something wrong here?

Thanks,
Andrew


From starx at pacbell.net  Sun Feb 17 17:09:20 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 17 Feb 2002 22:09:20 GMT
Subject: Ascii Art
References: <mailman.1012327189.31750.python-list@python.org> <585.799T42T11732924threeseas@earthlink.net> <slrna5rqpa.19d.philh@comuno.freeserve.co.uk> <a3ltbd$3oi$1@verence.demon.co.uk>
Message-ID: <Xns91B88FF6B244ERASXnewsDFE1@130.133.1.4>

Paul Wright || Mon 04 Feb 2002 04:03:57a:

>                       ______
>                      /      \
>                    .' PLEASE `.
>                    |  DO NOT  |      _____
>                    | FEED THE |    ,'.....`.
>                    `. LOONY  ,'  ,'........ )
>                      \_    _/   |........ ,'
>                        |  |     `. .... _/
>                        |  |      ,'.,'-'
>                        |  |     /../
>                        |  |   ,'.,'
>                        |  |  /../
>                      . |  | /..'
>                    .\_\|  |/_/,
>                    ___ |  | ___
>                      . `--' .
>                       .    .
> 
> It was fun at first but now it's tiresome. 
> 

I like it, even if it's uncertain weather it's a sign or a Paddle of 
Punishment =)


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From jdoak at lanl.gov  Tue Feb 19 18:47:38 2002
From: jdoak at lanl.gov (Justin Doak)
Date: 19 Feb 2002 15:47:38 -0800
Subject: sys.path Problem with 2.2?
Message-ID: <3d7957d4.0202191547.1f13d03b@posting.google.com>

I add a directory to my PYTHONPATH, but when I fire up the
interpreter, sys.path is getting the old value for PYTHONPATH. I ran
2.1, and it seems to be loading up the new PYTHONPATH just fine. 2.2
also seems to know when my PYTHONPATH is set or not; when I unset it,
nothing from the PYTHONPATH shows up in sys.path. This is decidedly
strange behavior; is there some caching of the PYTHONPATH going on
somewhere/somehow?

JD

# sys.path doesn't get all the directories from PYTHONPATH.

134 doak->doak% echo $PYTHONPATH
.:/n/toaster/u/doak/cvs/computer_vision:/n/toaster/u/doak/python/lib
135 doak->doak% python
Python 2.2 (#1, Jan  4 2002, 14:14:33)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', '/nh/toaster/u/doak', '/n/toaster/u/doak/cvs/computer_vision',
'/packages/lib/python2.2', '/packages/lib/python2.2/plat-linux2',
'/packages/lib/python2.2/lib-tk',
'/packages/lib/python2.2/lib-dynload',
'/packages/lib/python2.2/site-packages',
'/packages/lib/python2.2/site-packages/Numeric']

# But sys.path knows when I've unset PYTHONPATH.

136 doak->doak% unsetenv PYTHONPATH
137 doak->doak% python
Python 2.2 (#1, Jan  4 2002, 14:14:33)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', '/packages/lib/python2.2', '/packages/lib/python2.2/plat-linux2',
'/packages/lib/python2.2/lib-tk',
'/packages/lib/python2.2/lib-dynload',
'/packages/lib/python2.2/site-packages',
'/packages/lib/python2.2/site-packages/Numeric']

# 2.1 reads the PYTHONPATH correctly.

141 doak->doak% source ~/.cshrc
142 doak->doak% echo $PYTHONPATH
.:/n/toaster/u/doak/cvs/computer_vision:/n/toaster/u/doak/python/lib
143 doak->doak% /packages/.packages/python/python-2.1/bin/linux/python
Python 2.1 (#1, Jun  8 2001, 09:50:58)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', '/nh/toaster/u/doak', '/n/toaster/u/doak/cvs/computer_vision',
'/n/toaster/u/doak/python/lib', '/packages/lib/python2.1',
'/packages/lib/python2.1/plat-linux2',
'/packages/lib/python2.1/lib-tk',
'/packages/lib/python2.1/lib-dynload',
'/packages/lib/python2.1/site-packages',
'/packages/lib/python2.1/site-packages/Numeric']


From mlh at vier.idi.ntnu.no  Mon Feb 11 11:58:39 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 11 Feb 2002 16:58:39 +0000 (UTC)
Subject: New PEP (281): Loop Counter Iteration with range and xrange
Message-ID: <slrna6fu1v.goj.mlh@vier.idi.ntnu.no>

I've submitted a draft PEP describing the idea of letting range and
xrange accept sequences (or possibly, for xrange, any iterable) and
return/yield one index per element. It's closely related to PEP 212,
but instead of proposing new functions/syntax, it proposes extending
the capabilities of existing functions.

The PEP is available at:

  http://www.python.org/peps/pep-0281.html

Comments are (of course) welcome.

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From max at alcyone.com  Sun Feb 10 14:44:36 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 10 Feb 2002 11:44:36 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122>
Message-ID: <3C66CDA4.8E38F4CA@alcyone.com>

Bengt Richter wrote:

>  >>> 'abc'.split('')
>  Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
>  ValueError: empty separator
> 
> Wouldn't it make sense to return list('abc') ?

If that's what you really wanted, it would make far more sense to just
use list directly rather than calling string.split on the string.  The
error seems reasonable to me -- you shouldn't be using string.split if
that's what you really want.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From printers at sendme.cz  Tue Feb  5 17:25:14 2002
From: printers at sendme.cz (A)
Date: Tue, 5 Feb 2002 23:25:14 +0100
Subject: Is anyone capable of explaining ?
Message-ID: <3C6069DA.18895.92952C@localhost>

Can anybody please answer the following question?

Question: How does web browser know what cookies should be 
sent to server as a response?
For example I found out that a server sets 2 cookies

__cookie_client_data__=Uglv3rXE+d73YaYUuxoh/sDKsJ
t98itBb/hg2fFwFvHeEabkpKICibilujHZm

and

__cookie_temp_client_data__=aZNjw12D5JMrduWikHGk+
2pythOqr


but browser, as a response,sends only 
__cookie_temp_client_data__ 

Here is a part of communication between server 
and a client
##################first  SERVER #################

HTTP/1.1 200 OK - [54-68]
 Date: Mon, 04 Feb 2002 15:14:45 GMT - [71-105]
 Server: Apache/1.3.19 (Unix) Resin/1.1 - [108-
145]
 Cache-Control: no-cache="set-cookie,set-
cookie2" - [148-195]
 Expires: Thu, 01 Dec 1994 16:00:00 GMT - [198-
235]
 Set-Cookie: 
cookie_check=null;Path=/bin/user;Version=1 - 
[238-291]
 Set-Cookie: 
__cookie_client_data__=Uglv3rXE+d73YaYUuxoh/sDKsJ
t98itBb/hg2fFwFvHeEabkpKICibilujHZm ... - [294-
483]
104 - [294-484]
 Set-Cookie: 
__cookie_temp_client_data__=aZNjw12D5JMrduWikHGk+
2pythOqr

###########and CLIENT#################
----- HTTP Client Request ----- - [54-589]
[536 byte(s) of data]
 GET 
/html.ng/Site=English&Column=home&Category=/&BizN
ature=1&Province=&Position=1&Country=CZ&Ima ... -
 [54-189]
 Accept: image/gif, image/x-xbitmap, image/jpeg, 
image/pjpeg, application/vnd.ms-powerpoint, appl 
... - [192-332]
 Referer: http://www.alibaba.com/bin/user/signin 
- [335-381]
 Accept-Language: cs - [384-402]
 User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; 
Windows 98; Win 9x 4.90) - [405-475]
 Host: ad3.alibaba.com - [478-498]
 Connection: Keep-Alive - [501-522]
65 - [501-523]
 Cookie: 
__cookie_temp_client_data__=aZNjw12D5JMrduWikHGk+
2pythOqr
################# end#################

Thank you for explanation.
Ladislav




From christoforou at gaaros.com  Fri Feb 15 05:23:38 2002
From: christoforou at gaaros.com (christoforou at gaaros.com)
Date: Fri, 15 Feb 2002 10:23:38 GMT
Subject: Python License Issues
Message-ID: <20020215102338.759.qmail@mail.yifansoft.com>

Thanks for the comments

First some clarification ... 

There was no intention whatsoever to suggest that the PSF should be  
modified in any way and I believe it is clear in the post. 


Indeed we have hired IP lawers but until they figure it out I decided to ask 
the python community hoping that other python companies/developers who might 
have encountered similar issues have come up with a clevel solution. 

What worries me is that the 'legal climate' of our times will hurt open 
source acceptance and small soft shops in general. For it seems to me that 
the recent escalation in legal actions against third party IP infringments 
are driving coorporate clients to seek indemnity from IP violations. In 
their (reasonable) view the cost of software is: 

 Cost= ActualCostOfApplication  
+(ProbabilityOfLegalActionRelatingToApplication*SettlementAmount) 


The higher the second part of the equation the less important the first part 
becomes and certainly a small development shop cannot provide the level of 
indemnity that IBM can. 


>>> from BasicUtilities import deleteFromGlobalConsiousness
>>> deleteFromGlobalConsiousness(category='SoftwarePatents',level='Total') 
>>> 
 

Pavlos 




From rdacker at pacbell.net  Sun Feb  3 17:54:05 2002
From: rdacker at pacbell.net (rdack)
Date: 3 Feb 2002 14:54:05 -0800
Subject: [OSX] milter won't build
Message-ID: <644f6688.0202031454.51623317@posting.google.com>

trying to build milter module 0.3.9 on OSX 10.1.2 i get this error:
cc -bundle -undefined suppress build/temp.darwin-5.2-Power
Macintosh-2.1/miltermodule.o -lmilter -lsm -o
build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
/usr/bin/ld: -undefined error must be used when -twolevel_namespace is
in effect
error: command 'cc' failed with exit status 1

is it telling me i need  an '-undefined error' option? or that it
doesn't like the 'undefined suppress' option?
where is the 'cc -bundle' command set up? can i edit it?
is it possible that it finished and the module is ready to go? 
where would the milter module end up after install?
what will its name be?
are python modules '.pyo' files? do they end up in the
/usr/local/lib/python2.1 directory?
help?


From wookiz at hotmail.com  Sun Feb 10 03:40:24 2002
From: wookiz at hotmail.com (wooks)
Date: 10 Feb 2002 00:40:24 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <3C6596B6.20008@sneakemail.com>
Message-ID: <8852d33c.0202100040.732fbbfb@posting.google.com>

Joshua Macy <l0819m0v0smfm001 at sneakemail.com> wrote in message news:<3C6596B6.20008 at sneakemail.com>...
> 
>    Yeah, I came to Python from Perl, too, and I used to think that all 
> that context sensitivity was the cat's pyjamas--but after the second 
> two-day debugging session caused by Perl guessing wrong (or my guessing 
> wrong about what Perl would guess), I wasn't so thrilled.  I actually 
> tried Python because I bought the O'Reilly book on Advanced Perl 
> Programming, and in the author's intro he said really nice things about 
> Python.  I went out and looked at Python online, then bought Programming 
> Python, and never looked back.  The Perl book is still sitting unread on 
> my shelf....
> 

I think you can get into trouble with any language. Seems I am coming
to Python with a motivation for prevention via the unit test tools
available whereas your motivation was cure.


From emile at fenx.com  Sat Feb 16 17:52:36 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 16 Feb 2002 14:52:36 -0800
Subject: "implemented by subclass"
References: <slrna6t9de.nt2.philh@comuno.freeserve.co.uk>
Message-ID: <a4mo62$1egpb$1@ID-11957.news.dfncis.de>

"phil hunt" 
> Is this the most pythonic way to do it? If not, what would be
> better? How about rasing a ShouldBeImplementedBySublass exception?
> 
> How do other people do it?
> 

I've used:

raise RuntimeError, "Method not implemented"

-- 

Emile van Sebille
emile at fenx.com

---------



From phd at phd.pp.ru  Sat Feb  9 06:53:31 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sat, 9 Feb 2002 14:53:31 +0300
Subject: I wish.............
In-Reply-To: <slrna6a1ql.jm3.mlh@vier.idi.ntnu.no>; from mlh@vier.idi.ntnu.no on Sat, Feb 09, 2002 at 11:26:14AM +0000
References: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com> <mailman.1013246955.13570.python-list@python.org> <slrna6a1ql.jm3.mlh@vier.idi.ntnu.no>
Message-ID: <20020209145331.A28246@phd.pp.ru>

On Sat, Feb 09, 2002 at 11:26:14AM +0000, Magnus Lie Hetland wrote:
> In article <mailman.1013246955.13570.python-list at python.org>, Oleg
> Broytmann wrote:
> >   Python is the (best) languge. CPython is one (and not the best)
> >implementation.
> 
> Why is it not the best implementation? Which is better in your
> opinion?

   Currently there are no "best" implementations. Vyper promised to be one,
but it stale now :(

   It is slow. There are thoughts that using Jython and compiling byte code
to native code may give performance boost. My be we (at my job) will try to
experiment with the approach.

   Current CPython implementation stores bytecode files (*.py[co]) along
with the *.py files. This does not allow to use *.py[co] files with
different versions of python. That's why. for exmple, shared
/usr/local/lib/site-python is mostly meaningles; one needs to use
/usr/local/lib/python$VERSION/site-packges.

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



From hugomartires at hotmail.com  Mon Feb 25 13:41:53 2002
From: hugomartires at hotmail.com (Hugo Martires)
Date: 25 Feb 2002 10:41:53 -0800
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com> <3c7a278a.13899093@news.isar.de>
Message-ID: <d0380df9.0202251041.2d9ca447@posting.google.com>

This only receive twice.
If i need to receive more than twice that is a problem.
See this:

class Client2:
     def __init__(self, file):
         self.file = file
     def read(self):
         # assume size repr is less than 20 bytes
         x = os.read(self.file, 20)
         pos = x.find('.')
->=8000  size_expected = int(x[:pos])
         if size_expected > 20:
             # read rest of data
->sockets    more_data = client.socket.read(self.file, size_expected -
pos )
             return cPickle.loads(x[pos+1:] + more_data)
         else:
             return cPickle.loads(x[pos+1:])
         

I want to send a big vector it's size could be more or less than 8000.
The client to receive the entire vector must receive several times
since BUFFER SIZE it's not so big. (do you know it's maximum value ?)
So the server send :
# Send: size of vector; separator (.) ; vector in a string
CliSock.send( "%d.%s" %(vp_size, vp.tostring() ) )       

My client code:
------------------------

HOST='10.10.22.245'
PORT=21567
ADDR=(HOST , PORT)
BUFSIZE=1024
c=socket(AF_INET,SOCK_STREAM)
c.connect(ADDR)
data_vector=array.array('d')

data = c.recv(BUFSIZE)
sep = data.find('.')    # find the separator in string received
size = int ( data[:sep] )       # get the vector size
while (1):
        if size > BUFSIZE:
            #read rest of data
            data=data + c.recv(BUFSIZE)
        else:
            print data
data = data[sep+1:]
print data
data_vector.fromstring( data )

That's not working

Tanks for help


gerson.kurz at t-online.de (Gerson Kurz) wrote in message news:<3c7a278a.13899093 at news.isar.de>...
> Here is a suggestion:
> ------------------------------------------------
> import cPickle
> import StringIO
> import struct
> import os
> 
> # your classes
> class Client:
>     def __init__(self, file):
>         self.file = file
>     def read(self):
>         # read length and load object
>         x = os.read(self.file, 4)
>         length = struct.unpack('i', x)[0]
>         buf = os.read(self.file, length)
>         f = StringIO.StringIO(buf)
>         return cPickle.load(f)
>         
> class Server:
>     def __init__(self, file):
>         self.file = file
> 
>     def write(self, obj):
>         # write length of serialozed object, write serialized object
>         f = StringIO.StringIO()
>         cPickle.dump(obj, f, 1)
>         val = f.getvalue()
>         length = len(val)
>         plen = struct.pack("i", length)
>         os.write(self.file, plen)
>         os.write(self.file,val)
> 
> # my classes        
> class Client2:
>     def __init__(self, file):
>         self.file = file
>     def read(self):
>         # assume size repr is less than 20 bytes
>         x = os.read(self.file, 20)
>         pos = x.find('.')
>         size_expected = int(x[:pos])
>         if size_expected > 20:
>             # read rest of data
>             more_data = os.read(self.file, size_expected - pos )
>             return cPickle.loads(x[pos+1:] + more_data)
>         else:
>             return cPickle.loads(x[pos+1:])
>         
> class Server2:
>     def __init__(self, file):
>         self.file = file
> 
>     def write(self, obj):
>         test = cPickle.dumps(obj)
>         # interestingly, this seems to be slower:
>         #os.write(self.file,"%d." % len(test) + test)
>         os.write(self.file,"%d.%s" % (len(test),test))
> 
> # test script
> if __name__=='__main__':
>     rp, wp = os.pipe()
> 
>     import time
> 
>     test_string = u"this is a test string that is more than 20 bytes
> long"
> 
>     c = Client(rp)
>     s = Server(wp)
>     start = time.clock()
>     for i in xrange(3000):
>         s.write(test_string)
>         assert(test_string==c.read())
>     print time.clock() - start
> 
>     c = Client2(rp)
>     s = Server2(wp)
>     start = time.clock()
>     for i in xrange(3000):
>         s.write(test_string)
>         assert(test_string==c.read())
>     print time.clock() - start
> 
> 
> # results:
> 0.518367439587 (your)
> 0.228796436055 (mine)
> -------------------------------------------------------
> 
> But, your version seems reasonably fast already :). My "Full-Blown
> RPCServer" that uses dedicated threads for function calls (so that one
> RPC call can issue another RPC call while not blocking other threads
> making their RPC calls etc) is *a lot* slower...


From jbauer at rubic.com  Wed Feb 20 08:56:22 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 20 Feb 2002 07:56:22 -0600
Subject: MySQLdb build on Solaris 7
References: <F203anSs21Ud9mTP2vZ000016c2@hotmail.com>
Message-ID: <3C73AB06.F573840A@rubic.com>

Asim Beg wrote:
> Here's an excerpt from setup.py
> 
> elif sys.platform == "sunos5": # Solaris 2.8 + gcc
>     include_dirs = ['/usr/local/mysql/include']
>     library_dirs = ['/usr/local/mysql/lib']
>     libraries = [mysqlclient, "z"]
>     runtime_library_dirs = ['/usr/local/lib:/usr/openwin/lib:/usr/dt/lib']
>     extra_compile_args = ["-fPIC"]

Looks reasonable.  You've checked that the above directories
actually have mysql include files and libraries present, right?

Here's mine:

  elif sys.platform == "sunos5": # Solaris 2.8 
      include_dirs = ['/usr/local/include/mysql'] 
      library_dirs = ['/usr/local/lib/mysql'] 
      libraries = [mysqlclient, "z"] 
      runtime_library_dirs =
['/usr/local/lib:/usr/openwin/lib:/usr/dt/lib'] 
      extra_objects = [] 

And:

  LD_LIBRARY_PATH=/usr/local/lib:/usr/local/lib/mysql

Also, in your situation, your administrator may not permit you
to install files in /usr/local/lib/python2.1/site-packages/ --
though there isn't any real reason she shouldn't -- and it would
make it more convenient for running cgi scripts.  But assuming
she doesn't, you'll need those files (in the particular
relative path structure I mentioned in my earlier post) to 
be present in your importable path, e.g. PYTHONPATH or
(ugh) manually appended to sys.path.

I have worked with ISP admins whom I was able to convince
that installation into /site-packages/ was a good thing to
do for packages such as the MySQL interface; it would not
compromise security or maintenance and benefit other
Python users on their system.

Good luck.

Jeff Bauer
Rubicon Research



From hdiwan420 at speakeasy.net  Sat Feb  2 20:18:42 2002
From: hdiwan420 at speakeasy.net (hdiwan420 at speakeasy.net)
Date: Sun, 03 Feb 2002 01:18:42 -0000
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <u5p3vihkcm5764@corp.supernews.com>

Fred:
	Here you are:
for i in range (1..12):
  for j in range (1..12):
     print i*j,
  print

c at tech.usd253.org (Fred) writes:

>#include <stdio.h>
>int main (void)

>{

>int i;
>int j;

>for (i = 1; i < 13; i++)
>	{
>	for (j = 1; j < 13; j++)
>	  printf("%4d", i*j);
>	        printf("\n");
>   }
>return 0;
>}
-- 
Hasan =)
GPG KeyID: 0xBE42DCA6


From ol1 at v10a.com  Wed Feb 20 04:35:58 2002
From: ol1 at v10a.com (O Polite)
Date: 20 Feb 2002 10:35:58 +0100
Subject: calling a function indirectly
In-Reply-To: <Pine.LNX.4.44.0202200112010.14394-100000@mercury.sabren.com>
References: <Pine.LNX.4.44.0202200112010.14394-100000@mercury.sabren.com>
Message-ID: <1014197758.2088.10.camel@matijek>

On Wed, 2002-02-20 at 07:13, Michal Wallace wrote:
> 
> locals()[funcvar]("whatever")
> 

nifty!

op






From emile at fenx.com  Wed Feb 27 10:37:47 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Feb 2002 07:37:47 -0800
Subject: Telnet connection
References: <a5iskt$ge8$1@wanadoo.fr>
Message-ID: <a5iuq9$7701i$1@ID-11957.news.dfncis.de>

"Riss Nicolas"
> Hello,
> I must do a program who can connect with telnet. But I don't know how
to
> make a such connection with python, can somebody help me please.
>

Start with telnetlib

--

Emile van Sebille
emile at fenx.com

---------



From shalehperry at attbi.com  Wed Feb 20 15:15:46 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 20 Feb 2002 12:15:46 -0800 (PST)
Subject: a question regarding conciseness
In-Reply-To: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <XFMail.20020220121546.shalehperry@attbi.com>

> 
> for i in d.keys().sort():
>   do something
> 
> I get:
> 
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: loop over non-sequence
> 
> But I know that d.keys() returns a list to which the sort() function should 
> be applicable.

Yes keys() returns a sequence.  However sort() is a method of that sequence and
does not return anything.

mylist = alist.sort() # also an invalid statement

You could write a sort() function so that

for i in sort(d.keys()):
  do something

would work.



From sholden at holdenweb.com  Thu Feb  7 09:07:18 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 09:07:18 -0500
Subject: Python 2.2 dynamic library problem on Compaq Tru64
References: <mailman.1012925289.31030.python-list@python.org> <eecbdf57.0202051509.f3fd8b5@posting.google.com> <a3pq1c$nru$1@peabody.colorado.edu>
Message-ID: <cSv88.94610$KM2.3418966@atlpnn01.usenetserver.com>

"Fernando P?rez" <fperez528 at yahoo.com> wrote in message
news:a3pq1c$nru$1 at peabody.colorado.edu...
> Anthony Allen wrote:
>
> >> > ? ? import cx Oracle, os, sys, smtplib
>
> This line is extremely suspicious, see the missing comma between cx and
> Oracle. Modules can't have spaces in their names. Double check that first.
>
Fernando:

Maybe your news server hadn't shown you the original post when you sent this
reply ... there was an underscore between "cx" and "Oracle", giving the
correct module name.

Almost certainly the cron job somehow needs to establish the correct
sys.path to find the cx_Oracle module. Whether you do this by manipulating
the shell environment (either PYTHONPATH or LD_LIBRARY_PATH) or directly
modifying sys.path in Python is a matter of taste and local administration.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From wurmy at earthlink.net  Mon Feb 18 00:20:05 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Mon, 18 Feb 2002 05:20:05 GMT
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no>  <3C706035.6AC11606@earthlink.net> <slrna711lc.4oc.mlh@vier.idi.ntnu.no>
Message-ID: <3C708F33.514A527B@earthlink.net>

Magnus Lie Hetland wrote:
> 
> In article <3C706035.6AC11606 at earthlink.net>, Hans Nowak wrote:
> >Magnus Lie Hetland wrote:
> [...]
> >> No, no! This isn't simple enough. The *real* truth (and it's physics,
> >> so you can't contradict me) is that there is only one true action:
> >>
> >> DS - Do Stuff
> >>
> >> Now deal with it and go somewhere else with your false theory.
> >
> >Do you also have a model with zero actions?
> 
> How could that possibly cover everything? Sorry, one action is
> minimum.

Yeah, but think of the drastically reduced debugging times! In
fact, it seems that the PSU has


From philh at comuno.freeserve.co.uk  Mon Feb  4 09:18:43 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 14:18:43 +0000
Subject: Status of Python for .NET
References: <3c580a60.6586500@news.t-online.de> <d7fcf8b4.0201311356.704a8cfa@posting.google.com>
Message-ID: <slrna5t623.2jt.philh@comuno.freeserve.co.uk>

On 31 Jan 2002 13:56:51 -0800, Samuel Schulenburg <samschul at pacbell.net> wrote:
>gerson.kurz at t-online.de (Gerson Kurz) wrote in message news:<3c580a60.6586500 at news.t-online.de>...
>> Two questions: 
>> 
>> - The current version of VisualPython .NET seems not to generate CLR
>> code, but just provide a fancy IDE for "Python as we know and love
>> it". Will Python support the CLR, and specifically the
>> language-interoperability thing? (I'm asking because I want to be
>> spared the ... bad language-lookalike that is C#). 
>> 
>> - With the leave of Mark Hammond from Activestate, does anybody have a
>> guess about the future of Visual Python .NET?
>
>I have run into the following catch 22 between what Microsoft say's
>and what they do.
>
>I recently attened their .NET seminiar where they proudly proclamed
>that .NET will support various languages, and
>Python,Fortran,Cobal,Effel,and others were mentioned. They did not say
>who would be porting these languages to .NET.
>
>Now at the company I work for, (A Microsoft House) we use Windows2000
>as the OS for our product. Microsoft's EULA that we have to follow
>states that we can not use open source, or GPL'd software. i.e I can
>not use Python at work/

Is this the EULA for Windows 2000? Or some other EULA?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From redskin at tut.by  Sat Feb 23 19:57:19 2002
From: redskin at tut.by (Andrey Sidorenko)
Date: 23 Feb 2002 16:57:19 -0800
Subject: Tkinter + Entry + cp1251 cyrillic
Message-ID: <26428104.0202231657.639cec52@posting.google.com>

I use Active Python 2.2 on Windows 2000 and have trouble - "Entry"
class can't display cyrillic letters (cp1251 encoding) correctly in
entry field. Can anybody help me with this weird problem?

Thanks in advance.


From logiplexsoftware at earthlink.net  Tue Feb  5 20:26:32 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 5 Feb 2002 17:26:32 -0800
Subject: ASPI from Python
In-Reply-To: <20020205172150.22ffdafd.logiplexsoftware@earthlink.net>
References: <a3putt$9sp$1@news.wplus.spb.ru>
	<20020205172150.22ffdafd.logiplexsoftware@earthlink.net>
Message-ID: <20020205172632.147b9345.logiplexsoftware@earthlink.net>

On Tue, 5 Feb 2002 17:21:50 -0800
Cliff Wells wrote:

> On Wed, 6 Feb 2002 03:55:25 +0300
> Dmitry Rozmanov wrote:
> 
> > Hi.
> > 
> > How can I interface to ASPI from python in Windows2000? Is it possible?
> > 
> > How can I talk with a scanner from python in Linux?
> 
> print "hello, scanner"
> 
> There is supposed to be a module that supports SANE under Linux:
> 
> About: http://www.linuxjournal.com/article.php?sid=2395
> SANE: 	http://www.mostang.com/sane/
> Python API: http://starship.skyport.net/crew/amk/projects/

Sorry, the third link appears broken.  Try

http://www.amk.ca/python/code/sane.html

> I haven't checked this out so YMMV
> 
> > Thank you.
> > 
> >     ---Dmitry.
> > 
> > 
> > 
> > -- 
> > http://mail.python.org/mailman/listinfo/python-list
> 
> 
> -- 
> Cliff Wells, Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308  (800) 735-0555 x308
> 
> "Then with your new power you'll accomplish all sorts of cool stuff 
>  in no time, and We'll All Be Sorry.  At that point you can either 
>  gloat a bit, and then relent, or go ahead and send the robot army 
>  after us." - Quinn Dunkan
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From sholden at holdenweb.com  Mon Feb 11 07:14:12 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 07:14:12 -0500
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com> <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu> <8852d33c.0202110117.374f062@posting.google.com>
Message-ID: <dAO98.148833$KM2.6668696@atlpnn01.usenetserver.com>

"wooks" <wookiz at hotmail.com> wrote in message
news:8852d33c.0202110117.374f062 at posting.google.com...
[Quinn explains polymorphism]
> ... if I get magic methods right, they are special functions which
> if present enable the object to apply a unique interpretation of a
> python operator.
> So + has standard behaviour but will  adopt whatever is specified in
> the __add__ function if one is present for the class.
>
> Yes?

Correct

by-jove-he's-got-it-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From tessonec at inifta.unlp.edu.ar  Thu Feb 21 15:11:07 2002
From: tessonec at inifta.unlp.edu.ar (Claudio Tessone)
Date: 21 Feb 2002 12:11:07 -0800
Subject: Suggestion for best intro to Python
References: <c510a321.0202210608.1acc4d29@posting.google.com>
Message-ID: <2567f8cd.0202211211.529c799c@posting.google.com>

"dive into python" is great, but also (for an intro to the language)

"thinking as a computer scientist in python" is a good starting point.
I even recommend this before the previous


From usenet at thinkspot.net  Sat Feb 23 17:33:24 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 14:33:24 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com> <a582kl.3vsjht5.1@kserver.org> <mailman.1014499369.29333.python-list@python.org>
Message-ID: <a5897k.3vsgfhl.1@kserver.org>

On Sat, 23 Feb 2002 16:20:54 -0500, "Mike C. Fletcher"
<mcfletch at geocities.com> wrote in comp.lang.python in article
<mailman.1014499369.29333.python-list at python.org>:

>  >>> import string
>  >>> data = 'this\n\nthose\n\n\nthem\ntheir\nthose'
>  >>> string.join(filter( None, string.split( data, '\n')), '\n')
> 'this\nthose\nthem\ntheir\nthose'
>  >>>

Thanks for the suggestion. I really should have thought of that myself.
But there are so many different aspects to this script that I'm working
on, that sometimes something simple like this, I just want to "get done"
and work on the other "harder" stuff.

Your method is much better. I happen to be working on this script right
now, and will incorporate this change. 

Thanks for the suggestion,

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From aa8vb at yahoo.com  Thu Feb 28 13:36:55 2002
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 28 Feb 2002 13:36:55 -0500
Subject: Python DB that supports numbers
In-Reply-To: <3C7E80DD.3655.6F5638C@localhost>; from gs@styrax.com on Thu, Feb 28, 2002 at 07:11:25PM +0000
References: <3C7E7E34.24356.6EB01AA@localhost>; <20020228131104.A2852698@vislab.epa.gov> <3C7E80DD.3655.6F5638C@localhost>
Message-ID: <20020228133655.A2853090@vislab.epa.gov>

Garry Steedman:
 |>  http://www.chordate.com/gadfly.html
 |>            Connection Failed 
 |
 |http://sourceforge.net/projects/gadfly/
 |
 |should work...

Ok, but there are no docs, no project files, and the CVS repository appears
empty.  

What am I missing?

Randall

-- 
Randall Hopper
aa8vb at yahoo.com



From jblaine at mitre.org  Mon Feb  4 10:36:27 2002
From: jblaine at mitre.org (Jeff Blaine)
Date: Mon, 4 Feb 2002 15:36:27 +0000 (UTC)
Subject: Python 2.2 Modules/Setup lies?
References: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net> <mailman.1012768207.25312.python-list@python.org> <slrna5rc6q.dtp.jon+usenet@snowy.squish.net> <m3sn8hkbqh.fsf@mira.informatik.hu-berlin.de> <slrna5t5up.mr.jon+usenet@snowy.squish.net>
Message-ID: <Xns91AB6BE7C93C1jblainemitreorg@129.83.20.55>

I agree completely here with Jon.  I was just mentioning this last week
to a friend of mine.

> The problem appears to be that the install procedure has changed
> radically, without the install instructions changing at all.


From john_coppola_r_s at yahoo.com  Thu Feb 14 16:29:56 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Thu, 14 Feb 2002 13:29:56 -0800 (PST)
Subject: Convert String to Dictionary question
In-Reply-To: <HFEKILOLEFEFMKAECNDLKEJODCAA.jason@jorendorff.com>
Message-ID: <20020214212956.94856.qmail@web11807.mail.yahoo.com>

even better to use to use cPickle.

--- Jason Orendorff <jason at jorendorff.com> wrote:
> > I saved my dictionary {'hello1': [1, 0, 0]} to a
> file.
> > When I do a readline the I get back a string.
> > How do I convert this string to a dictionary?
> 
> It's better to save data using pickle.
>  
>
http://www.python.org/doc/current/lib/module-pickle.html
> 
> import pickle
> mydata = {'hello1': [1, 0, 0]}
> 
> # How to store data
> f = open('mydata.dat', 'wb')
> pickle.dump(mydata, f, 1)
> f.close()
> 
> # How to load data back
> f = open('mydata.dat', 'rb')
> mydata = pickle.load(f)
> f.close()
> 
> ## Jason Orendorff    http://www.jorendorff.com/
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list


__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com



From logiplexsoftware at earthlink.net  Tue Feb 19 12:36:51 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 19 Feb 2002 09:36:51 -0800
Subject: basic books/guides on multithreading programming?
In-Reply-To: <5.1.0.14.2.20020220002245.0a4ade80@202.95.131.154>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
	<mailman.1013816158.26552.python-list@python.org>
	<5.1.0.14.2.20020220002245.0a4ade80@202.95.131.154>
Message-ID: <20020219093651.70c08cba.logiplexsoftware@earthlink.net>

On Wed, 20 Feb 2002 00:26:00 +0700
Tripp Scott wrote:

> On 18/02/2002 11:09, Andrae Muys wrote:
> >"End of Story" is a rather interesting way to summarise the issues of
> >Deadlock, Lock-ordering, Livelock, Starvation, Convoying, and
> >Lock-Contention to name a few of the problems you may face moving from
> >serial to parallel algorthims (although the last two only affect
> >perfomance, not correctness).
> 
> ok, i understand more or less about deadlock or lock ordering 
> (from the database equivalent terms), though have not heard 
> about livelock or the others. suppose i decide to be ignorant 
> about these issues "for now" and go about building my 
> multithreading project. and then suppose these issues cause 
> serious problems. would it be difficult to resolve it? would i 
> have to rethink/review all the threaded code?

Just type "livelock" into Google and you'll find plenty of references.  The
same is true for any of the terms mentioned.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From sjmachin at lexicon.net  Fri Feb 22 00:05:04 2002
From: sjmachin at lexicon.net (John Machin)
Date: 21 Feb 2002 21:05:04 -0800
Subject: Large integers from C
References: <mailman.1014324517.14965.python-list@python.org>
Message-ID: <c76ff6fc.0202212105.56aa14f8@posting.google.com>

Tim Peters <tim.one at comcast.net> wrote in message news:<mailman.1014324517.14965.python-list at python.org>...
> [John Machin]
> > ...
> > So why not grab the source of _PyLong_FromByteArray() from the 2.2
> > distribution and put in your own code?
> 
> That's what I'd do.  Do note the leading underscore:  C functions in the
> core whose names begin with a leading underscore are deliberately
> undocumented in the manuals, and nothing about them (not even their
> existence) is guaranteed across releases.  They're internal implementation
> details, not part of the public C API.  Use of them is entirely at your own
> risk.

In this case as the OP in essence pointed out,
PyLong_FromSomeFormatLongerThanCLongLong()would be very useful for
punters who somehow are generating such beasties themselves.
_PyLong_FromByteArray() exposes no implementation details that are
private to longobject.c. Do you have any objection to removing the _
and getting the comments in the .h file copied into the C API manual?

Regards,
John


From thomas.heller at ion-tof.com  Thu Feb  7 02:38:35 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 7 Feb 2002 08:38:35 +0100
Subject: scripted install of Python-2.2 and win32all?
References: <961e341e.0202062153.f976988@posting.google.com>
Message-ID: <a3taug$1a58fo$1@ID-59885.news.dfncis.de>

"David Sloyer" <dsloyer at ieee.org> wrote in message news:961e341e.0202062153.f976988 at posting.google.com...
> Is it possible for one to install Python-2.2 and win32all without any
> user intervention?  I'd like to install these on a set of machines as
> part of a script, and would like to do this automatically.  The
> installation would take place on Windows XP and Win2K boxes.
>

Appended is the relevant section from Wise Installbuilder 7.0 online help.
Note that Python 2.2 is built with Wise 8.0, win32all may  be built with earlier
versions, so YYMV.

Thomas
----------
Command Line Switches for Wise and Wise Installations

Command Line Switches for WISE32.EXE

? /c file.wse creates executable file from wise script

? /r opens Wise in SetupCapture window

? /u runs the upgrade wizard that checks for any upgrades online

? /d CompVar=Value defines a single compiler variable in Enterprise Edition. Multiple definitions require multiple /d

? /d=filename uses filename to define compiler variables. Each line in file is CompVar=Value

Command Line Switches for install files created in wise, i.e. SETUP.EXE

? /S silent mode, automatic mode with no user choices

? /T test mode

? /X pathname extracts files into pathname

? /Z pathname extracts files into pathname and reboots

? /M=filename specifies a value file similar to /d above, but for standard variables

? /M1 causes Wise to step through the installation and display each file that is flagged for self-registration, allowing the author
to identify each file as it is being self-registered.

? /M2 reserved for internal use by Wise during debugger sessions

Command Line Switches for UNWISE.EXE and UNWISE32.EXE

? /Z remove empty directories, including one with unwise itself in it

? /A automatic mode, no user choices other than cancel

? /S silent mode, automatic mode with no user choices

? /R rollback mode, selects option to rollback on uninstall

? /U like automatic mode, but gives all choices other than custom/automatic

? title - window title can be at the end of the command line

Copyright? 1998 Wise Solutions, Inc. All rights reserved.




From dalke at dalkescientific.com  Wed Feb 13 06:06:01 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Wed, 13 Feb 2002 04:06:01 -0700
Subject: photo id ?!?!
References: <mailman.1013585541.16266.python-list@python.org>
Message-ID: <a4dh6m$196$1@slb6.atl.mindspring.net>

Uncle Tim:
>It's curious that the FAA doesn't require photo ids (see
>http://www.faa.gov/).

All I could find was http://www.faa.gov/apa/tipbroch.htm
] Adult passengers must bring a government-issued photo ID. The FAA
] requires that air carriers request government-issued identification,
] such as a driver?s license or draft card, if the passenger appears
] old enough to have an ID. If a government-issued photo ID is not
] available, bring two pieces of ID, one of which must be from a
] governmental authority.

But so help me, I can't find the actual rule which says that photo
id is really required, or what counts as "government-issued".  (Could
I get my city to issue me a photo ID?)  The closest is

http://frwebgate.access.gpo.gov/cgi-bin/DBSearch.cgi?DBSearchDB=ecfr&DB=1&AC
TION=View&QUERY=108.201&RGN=BSEC&OP=and&QUERY=14&RGN=BTI&QUERY=3389&RGN=BSEC
CT&SUBSET=SUBSET&FROM=1&ITEM=1

which only talks about searches for explosives, arms, etc.

So I sent email to the FAA asking them to point me to the regulation
requiring ID and, if possible, justification for that requirement.

I also sent email to my local major airport asking what I need to
do in case while travelling I lose my ID or it gets stolen.

I did like this one, from Nov. 18, 1999 http://cas.faa.gov/ig5.pdf says
] Our criterion for testing was the ability to penetrate a secure
] area through any access point.  In 32 percent of our tests,
] we were *not* able to penetrate a secure area.

I want to see the Office of the Inspector General document on the same
topic a year from now.


>I've read that the airlines made up that rule, not
>for security, but to make it harder for person A to use a cheap ticket
>obtained by person B.  Is that true?  Beats me, but my cynical half can't
>not believe it <wink>.

I've heard that as well.  I'll wait for my response from the FAA since
I heard that about 10 years ago, when the requirement for photo id first
went into place.  Any bets on how long it will take for the FAA to
respond?

                    Andrew Dalke
                    dalke at dalkescientific.com





From Christian.Huehne at dlr.de  Mon Feb 18 03:57:16 2002
From: Christian.Huehne at dlr.de (Christian Hühne)
Date: Mon, 18 Feb 2002 09:57:16 +0100
Subject: List-Box
Message-ID: <a4qfje$26flu$1@ID-96133.news.dfncis.de>

PROGRAMMING IN PYTHON: LIST-BOX

In the following script I use 2 list-boxes. But I am only able to mark an
entry in one list-box.

Thank you
Chris.
--------------------------------------------------------------------
import re
import string
import Tkinter
import Pmw
from Tkconstants import *

tk = Tkinter.Tk()
frame = Tkinter.Frame(tk)
frame.pack(fill=BOTH,expand=1)
tk.wm_geometry("+100+45")

subframegeo = {}
geolistbox = {}

subframegeo['Delete'] = Tkinter.Frame(frame)
subframegeo['Delete'].pack(side=TOP,anchor=NW,fill=BOTH,expand=1)
geolistbox['Delete'] = Tkinter.Listbox(subframegeo['Delete'])
geolistbox['Delete'].grid(row=1,column=0, sticky=W, padx=20, pady=10)
geolistbox['Delete'].insert(END,'Height Radius Angle')
geolistbox['Delete'].insert(END,'Height Radius Angle')
geolistbox['Delete'].insert(END,'Height Radius Angle')
geolistbox['Delete'].insert(END,'Height Radius Angle')
geolistbox['Delete'].insert(END,'Height Radius Angle')
geolistbox['Delete'].selection_set(3)

subframegeo['Select'] = Tkinter.Frame(frame)
subframegeo['Select'].pack(side=TOP,anchor=NW,fill=BOTH,expand=1)
geolistbox['Select'] = Tkinter.Listbox(subframegeo['Select'])
geolistbox['Select'].grid(row=1,column=0, sticky=W, padx=20, pady=10)
geolistbox['Select'].insert(END,'Height Radius Angle')
geolistbox['Select'].insert(END,'Height Radius Angle')
geolistbox['Select'].insert(END,'Height Radius Angle')
geolistbox['Select'].insert(END,'Height Radius Angle')
geolistbox['Select'].selection_set(1)

tk.mainloop()





From timr at probo.com  Wed Feb 27 02:33:30 2002
From: timr at probo.com (Tim Roberts)
Date: Tue, 26 Feb 2002 23:33:30 -0800
Subject: HTTP_COOKIE variable not set
References: <3C79CAF7.97EA935F@dstc.edu.au>
Message-ID: <cf2p7uo09qgjvvn217otpfdlspsu8defvu@4ax.com>

Bevan Koopman <bevank at dstc.edu.au> wrote:
>
>When I run the script below the 'HTTP_COOKIE' environment variable is
>not set, i.e. "HTTP_COOKIE" does not appear in the scripts output:

Of course not.  The Cookie module doesn't set any environment variables.
HTTP_COOKIE is set by the web server (for example, Apache or IIS) when it
launches a CGI script in response to a form request from a browser.

The Cookie module is just a shortcut for creating a Set-Cookie header line
to be sent back to the web browser.  The next time that browser sends a
request, it will include the cookie in ITS headers; Apache will read those
headers and create the HTTP_COOKIE environment variable for you.

Are you trying to use cookies to communicate between two normal (non-CGI)
Python scripts?  That ain't gonna work.  There's no easy way to change the
environment of the shell that called you.

>#####################
>import cgi
>import os
>import Cookie
>
>print "Content-Type:text/html\n\n"
>
>cookie = Cookie.SimpleCookie()
>cookie["user"] = the_user

Change this to:

print "Content-Type: text/html"
cookie = Cookie.SimpleCookie()
cookie["user"] = the_user
print cookie
print

That prints the appropriate Set-Cookie header as part of the set of HTTP
headers.  It STILL doesn't set the environment variable; that won't happen
until the NEXT request you get from that browser.

>Hence when I run the following script the key is not found
>
>#####################
>import cgi
>import os
>import Cookie
>
>c = Cookie.SimpleCookie()
>if(os.environ.has_key('HTTP_COOKIE')):
>	c.load(os.environ['HTTP_COOKIE'])
>	print c["user"].value
>else:
>	print "no user"
>#####################
>
>How do I correctly create the cookie so I can retrieve in the second
>script.

If that second script is called as a CGI script from the same browser that
receives the page from the FIRST script, it will work.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From aahz at panix.com  Thu Feb 28 20:50:11 2002
From: aahz at panix.com (Aahz Maruch)
Date: 28 Feb 2002 17:50:11 -0800
Subject: listing threads?
References: <mailman.1014930527.20065.python-list@python.org>
Message-ID: <a5mmoj$2ji$1@panix2.panix.com>

In article <mailman.1014930527.20065.python-list at python.org>,
maximilianscherr <MaximilianScherr at T-Online.de> wrote:
>
>in the docs theres the information, that i can list all active 
>threads with threading.enumerate(), but i always just get the 
>Mainthread displayed in the interpreter console.

DON'T DO THAT.  I'm serious.  You really do NOT want to use
threading.enumerate(); you'll only confuse yourself.  I've pointed you
at my web page on threading; is there some reason you haven't read it,
or is there something you don't understand?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"He's Quebecois. He puts mayonnaise on *everything*."  --siobhan at virulent.org


From mark at mceahern.com  Fri Feb  1 17:29:56 2002
From: mark at mceahern.com (Mark McEahern)
Date: Fri, 1 Feb 2002 14:29:56 -0800
Subject: comparing DateTime instances
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEFPKAAA.mark@mceahern.com>

I think I just need a sanity check.  I'm trying to compare DateTime
instances (from the eGenix mx.DateTime package).

$ python
Python 2.2 (#1, Dec 31 2001, 15:21:18)
[GCC 2.95.3-5 (cygwin special)] on cygwin
Type "help", "copyright", "credits" or "license" for more information.
>>> expected = "2003-01-31 14:13:13.65"
>>> actual = "2003-01-31 14:13:13.65"
>>> from mx.DateTime import DateFrom, cmp
>>> e = DateFrom(expected)
>>> a = DateFrom(actual)
>>> cmp(a, e)
0

I think this is telling me that a and e are equal.  I'm still trying to get
my brain to instantly recognize the seemingly flipped around semantics of
cmp.  I understand that __cmp__ returns 0 for equality--and that 1 and -1
are used to indicate less-than, greater-than ordering.  So too for cmp, eh?

So that would mean, if I were comparing two DateTime instances in a unit
test and I wanted the test to fail if the dates were NOT equal, I might say
something like this:

	self.failIf(cmp(date1, date2))

If date1 equals date2 (I know I can use accuracy to account for fuzziness),
cmp will return 0, which is the same as false, and therefore failIf passes
because it only fails if the argument is true.

The documentation for mx.DateTime's cmp seems to assume the reader
understands __cmp__.  This is all we find there:

"cmp(obj1,obj2,accuracy=0.0)
Compares two DateTime[Delta] objects.
If accuracy is given, then equality will result in case the absolute
difference between the two values is less than or equal to accuracy."

Perhaps this could expand on the actual set of results:

-1 : obj1 is less than obj2
 0 : obj1 is equal to obj2
 1 : obj1 is greater than obj2

It wouldn't seem to hurt and it might avoid someone else wondering the same
silly things I'm wondering.

Thanks,

// mark




From philh at comuno.freeserve.co.uk  Sun Feb  3 20:58:22 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 01:58:22 +0000
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <mailman.1012781585.24995.python-list@python.org>
Message-ID: <slrna5rqlu.19d.philh@comuno.freeserve.co.uk>

On Sun, 3 Feb 2002 18:11:07 -0600, Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
>Remember... no responses to Rue.  We all need to go cold (on this) turkey.

You'll Rue the day you said that. :-)

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jason at jorendorff.com  Sun Feb  3 23:22:10 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 22:22:10 -0600
Subject: Making a list-inhereted object return standard repr?
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A15B@natasha.auslabs.avaya.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOECLDAAA.jason@jorendorff.com>

Timothy Delaney wrote:
> This brings up the interesting question of whether there should be a an
> object method which returns the "canonical repr" of an object i.e. the
> <module.Class object at address> or possibly even a builtin method (so it
> would work with classic classes too).

object.__repr__() works for everything.

If you are working with classic classes, you might prefer this:

def basic_repr(x):
    t = type(x)
    try:
        cls = x.__class__
    except AttributeError:
        pass
    else:
        if t is not cls:
            return t.__repr__(x)  # classic instance
    return object.__repr__(x)    # anything else

This picks up the name of the class for classic instances;
object.__repr__() in that case would return something like
'<instance object at 0x00903888>' which isn't always helpful.

## Jason Orendorff    http://www.jorendorff.com/



From Ed.Smirnov at mail.ee  Thu Feb 21 03:00:56 2002
From: Ed.Smirnov at mail.ee (Eduard Smirnov)
Date: Thu, 21 Feb 2002 10:00:56 +0200
Subject: Python vs. C/C++/Java: quantitative data ?
Message-ID: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>

I?m writing a diploma paper on the difference between scripting (e.g. in
Python or Tcl) and system programming (e.g. in C, C++ or Java), and one of
the claims I want to make is that you can often develop applications 10x or
more faster in a scripting language.
I apologue that more applications can be written or re-written in scripting
language faster and with more complex functionality then in system
programming language.
If you have any experience in writing applications in system programming
language (e.g. in C, C++) and re-writing it in scripting language (e.g.
Python) or vice versa, please compare your work in both implementations,
describe your work/application?s functionality, approximate number of lines
and time of development of each implementation, skill?s level needed.
Any more comments about are welcome.

Additional thanks for copy your answer to my e-mail.

Eduard.Smirnov at mail.ee
Tallinn Technical Univercity
Estonia




From emile at fenx.com  Fri Feb  8 11:34:22 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 8 Feb 2002 08:34:22 -0800
Subject: [NEWBIE] List issue aka: "Why doesn't this code work?"
References: <mailman.1013184014.10227.python-list@python.org>
Message-ID: <a40v0h$1aahs0$1@ID-11957.news.dfncis.de>

"Bruce Dykes" <bkd at graphnet.com> wrote in message
news:mailman.1013184014.10227.python-list at python.org...
> Okay, here's my function:
>
> def process_numbers(source_list):
>     for number in range(len(source_list)):
>         entry = string.split(string.strip(source_list[number]))
>         print number, len(entry), entry
>
> Now this function is doing *something*...source_list is a list of
lines
> retrieved via ftp.retrlines(), and this should break each line into
it's own
> list, and it looks like that's what's being printed out:
>
> ['line1', 'header1', 'header2']
> ['line2', 'datum1', 'datum2']
>
> Notice what's missing? I'm expecting to see:
>
> 0 3 ['line1', 'header1', 'header2']
> 1 3 ['line2', 'datum1', 'datum2']
>
> What the heck am I doing wrong here?


If I had to guess, I'd change

def process_numbers(source_list):

to

def XXXprocess_numbers(source_list):

then track down what function is producing the results you're seeing.

--

Emile van Sebille
emile at fenx.com

---------



From mlh at vier.idi.ntnu.no  Sun Feb 17 20:43:11 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 18 Feb 2002 01:43:11 +0000 (UTC)
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net>
Message-ID: <slrna70n1f.som.mlh@vier.idi.ntnu.no>

In article <8867.813T2365T9245607threeseas at earthlink.net>, Timothy Rue wrote:
[...]
>Certainly everyone does understand in reading and responding to post here,
>they actually make use of all nine action. It's physics, but they may
>think it's me and that somehow their reactions will cause a dismissal
>of.... well physics.
>
>Lets see now:
[...]

No, no! This isn't simple enough. The *real* truth (and it's physics,
so you can't contradict me) is that there is only one true action:

DS - Do Stuff

Now deal with it and go somewhere else with your false theory.

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From mlh at vier.idi.ntnu.no  Sat Feb 16 20:52:32 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Sun, 17 Feb 2002 01:52:32 +0000 (UTC)
Subject: win32gui
References: <a4monr$hmg$1@tyfon.itea.ntnu.no>  <u3d01kjy8.fsf@online.no>
Message-ID: <slrna6u370.86f.mlh@vier.idi.ntnu.no>

In article <u3d01kjy8.fsf at online.no>, Syver Enstad wrote:
>"Jorg R?dsj?" <jorgelr at stud.ntnu.no> writes:
>> Can anyone give us, or point us to some resources when it comes to
>> using
>> 
>> win32gui?
>
>The book python win32 programming by Hammond and Robinson has a
>chapter on this. It's a book you need if you are going to use Python
>on win32, plus it's a whole lot of fun to use/read.

Jorg: Du kan l?ne denne av meg :)

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From mwh at python.net  Tue Feb 26 12:22:52 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 26 Feb 2002 17:22:52 GMT
Subject: Time to rename Stackless?
References: <mailman.1014739915.24754.python-list@python.org>
Message-ID: <uheo4xihc.fsf@python.net>

PoulsenL at capecon.com writes:

> How about High Performance Python?

How about sticking to the original name?  Even if it is no longer
accurate, it has considerable mind-share already, and the user visible
effects are similar.  Plus it saves CT from having to buy a new domain
:)

Cheers,
M.

-- 
    -Dr. Olin Shivers,
     Ph.D., Cranberry-Melon School of Cucumber Science
                                           -- seen in comp.lang.scheme


From francis.meyvis at sonycom.com  Thu Feb 28 08:05:04 2002
From: francis.meyvis at sonycom.com (Francis Meyvis)
Date: Thu, 28 Feb 2002 13:05:04 GMT
Subject: HTTP chunk Transfer-Encoding
Message-ID: <4Wpf8.13$zd.865@nreader2.kpnqwest.net>

Hoi,

Is there some python source that sends chunked packets (RFC 2616) in
the SimpleHTTPServer or BaseHTTPServer?
Or do you know where to find snippet of code that can split up
file to send them in chunk format?

Thx

--
---
Kind regards, best wishes,
francis meyvis




From donnal at donnal.net  Wed Feb 27 21:45:59 2002
From: donnal at donnal.net (Donnal Walter)
Date: 27 Feb 2002 18:45:59 -0800
Subject: functional programming with map()
References: <mailman.1014747703.26001.python-list@python.org>
Message-ID: <918bc22f.0202271845.16f75b16@posting.google.com>

I wish to thank everyone that contributed to this thread. I have
been overwhelmed at the response, but this discussion has been
extremely helpful. Thank you

As a result, I attached a dozen lines of code to my XML database
management module so that I can now generate sophistocated reports
from an HTML template with embedded string formatting operators (%) .
These refer to an embedded dictionary at the end of the HTML file
something like this:

<html>
  <head><title>Prism: Progress Note</title></head>
<body>
<table width="100%%" cellpadding="5"><tr>
<td width="70" valign="top"><p>%(date)s</p></td>
<td>
<p><strong>Assessment and Plan:</strong></p>
<ol>
<li>%(ga)s wk newborn infant %(order)s, birthweight %(bw)s kg,
yesterday %(wt)s kg,<br />
chronologic age %(age)s, post-conceptual age %(pca)s wk. <br />
</li>
<li>Temperature control: %(heat)s
</li>
<li>Respiratory support: %(resp)s <br />
</li>
<li>Nutrition and Fluids:<br /> %(fen)s <br />
</li>
%(pList)s
</ol><br /><br />
<br /><br /><br />_______________________________________<br />
Donnal Walter, M.D., Attending Neonatologist
</td>
</tr></table>
<!-- <vars>{
  'date':  self.today.GetDate(),
  'ga':    self.birthGA.Get(),
  'order': self.birthorder.Get(),
  'bw':    self.birthweight.Get(),
  'wt':    self.weight.Get(),
  'age':   self.age.Get(),
  'pca':   self.pca.Get(),
  'heat':  self.warmer.Get(),
  'resp':  self.respiratory.Get(),
  'fen':   (
            IF(self.formula.Get(),
               ('____ cc/kg/day enteral of ' + 
                self.formula.Get() + '<br />'),
                '') +
            '____ cc/kg/day total fluids, ____ kcal/kg/day' + 
            '<br /> ____ cc/kg/hr urine output;' +
            IF(self.lytes.Get(),
               '<br />Na____, K____, Cl____, CO____,',
               '')
           ),
  'pList': CONCAT([('<li>' + x.label.Get() +
             IF(len(x.meds),
               ('<ul>' +
                 CONCAT([('<li>' + y.label.Get() + 
                   IF(y.dose.Get(),
                     (', ' + 
                       y.dose.Get() + ' ' +
                       y.units.Get() + ' (' +
                       y.scaled.Get() + ' ' +
                       y.units.Get() + '/kg) ' +
                       y.route.Get() + ' every ' +
                       y.interval.Get() + ' hr; '),
                     '') +
                   '</li>'
                 ) for y in x.meds]) +
               '</ul>' ),
               '') +
             '</li>'
           ) for x in self.problems])
}</vars>
-->
</body></html>

Note especially the loops in the definition for pList (problem list).
The code that interprets this template is simply:

    def Generate(self, filename):
        """filename = template file"""
        IF = lambda a,b,c:(a and [b] or [c])[0]
        CONCAT = lambda z:reduce((lambda x,y: x+y), z, ' ')
        try:                                 # try this
            io = open(filename, 'r')         #   open read only
            t = io.read()                    #   read full text
            io.close()                       #   close file
            i = string.find(t, '<vars>')     #   start of vars
            j = string.find(t, '</vars>')    #   end of vars
            s = t[:i] + t[j+7:]              #   remove "vars"
            v = eval(t[i+6:j])               #   make dictionary
            return s % v                     #   return string
        except KeyError:                     # unless key error
            return 'Key error in template'   #   then warning

    def Write(self, template, output):
        io = open(output, 'w')
        io.write(self.Generate(template))
        io.close()

Thanks again,
Donnal Walter


From michael at rcp.co.uk  Thu Feb  7 05:36:19 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Thu, 7 Feb 2002 10:36:19 +0000 (UTC)
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com>
Message-ID: <Xns91AE6BE28FCD8michaelrcpcouk@194.238.50.13>

aahz at panix.com (Aahz Maruch) wrote in news:a3sk2i$67m$1 at panix2.panix.com:

> will attest.  However, they don't AFAIK use Python for their core
> high-performance code, and it appears that this contest is slated for
> that role.
> 
> What does surprise me is that they think Java can do better than Python.

Well, I think the most enthusiastic supporter of Python has to acknowledge 
that Python has a performance problem when systems are pushed to their 
limits.  The fact that Python cannot be effectively compiled to machine 
code (while Java can) makes a substantial difference when it comes down to 
raw machine cycles, and Python's use of memory is not particularly 
efficient.

I reckon to pay a cost of perhaps a factor of 100 in raw CPU for the 
privilege of using Python.


From mlh at vier.idi.ntnu.no  Sun Feb 17 23:44:28 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Mon, 18 Feb 2002 04:44:28 +0000 (UTC)
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no>  <3C706035.6AC11606@earthlink.net>
Message-ID: <slrna711lc.4oc.mlh@vier.idi.ntnu.no>

In article <3C706035.6AC11606 at earthlink.net>, Hans Nowak wrote:
>Magnus Lie Hetland wrote:
[...]
>> No, no! This isn't simple enough. The *real* truth (and it's physics,
>> so you can't contradict me) is that there is only one true action:
>> 
>> DS - Do Stuff
>> 
>> Now deal with it and go somewhere else with your false theory.
>
>Do you also have a model with zero actions?

How could that possibly cover everything? Sorry, one action is
minimum.

> This one seems kind of difficult to me.

Hah - it's all a conspiracy. ;)

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From jason at jorendorff.com  Thu Feb 28 11:09:08 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 28 Feb 2002 10:09:08 -0600
Subject: string title() method does not work on Danish characters
In-Reply-To: <3C7E4F69.5080806@mxm.dk>
Message-ID: <HFEKILOLEFEFMKAECNDLIEACDGAA.jason@jorendorff.com>

Max M wrote:
> Thanks for the suggestion, but it does not solve my problem. I am 
> pulling the data from an older MS SQL server that returns standard ascii.

If it contains characters >127, it's not standard ASCII.
It's probably ISO-8859-1, aka Latin-1, aka Western Europe.

If you're getting this data back as 8-bit strings, then you need
to convert them to unicode first:

ustr = s.decode('latin-1')
ustr = ustr.title()
...

## Jason Orendorff    http://www.jorendorff.com/



From gimbo at ftech.net  Thu Feb 21 05:31:28 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Thu, 21 Feb 2002 10:31:28 +0000
Subject: why does popen2 silently ignore IOError?
In-Reply-To: <20020221005340.32346.qmail@web10304.mail.yahoo.com>
References: <20020221005340.32346.qmail@web10304.mail.yahoo.com>
Message-ID: <20020221103128.GA5505@andy.tynant.ftech.net>

On Wed, Feb 20, 2002 at 04:53:40PM -0800, Graham Guttocks wrote:

> I thought what I wanted was simple.  I'm sending mail by opening a
> pipe to /usr/sbin/sendmail -- if that fails for whatever reason (e.g,
> sendmail program is missing) I want an exception raised with some
> indication of the problem, instead of the current silent failure.

Have you looked at using smtplib to send mail, rather than calling
sendmail?  It's very nice.

-Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From matt at mondoinfo.com  Thu Feb 14 17:10:35 2002
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Thu, 14 Feb 2002 22:10:35 GMT
Subject: Tkinter Resize
References: <3daa1aa8.0202141329.63fdc057@posting.google.com>
Message-ID: <vBWa8.37225$Wf1.7066948@ruti.visi.com>

On 14 Feb 2002 13:29:52 -0800, Jeffrey <jmsun at bioeng.ucsd.edu> wrote:

Dear Jeff,

>Hi Everyone,

Hi!

>I'm working with a Tkinter Text Widget.  How do you bind a function
>to the Text Widget whenever it resizes?  Plus the function I want to
>bind needs to detect the new size of the Text Widget.  Is it possible
>to get the new width by doing: text_widget.cget( 'width' )?

You can bind a handler to the <Configure> event. The event will give
you the new width and height, but in dots, not characters.

The <Configure> event is documented in Fredrik Lundh's excellent An
Introduction to Tkinter at (wrapped for line length):

http://www.pythonware.com/library/tkinter/
  introduction/events-and-bindings.htm

Fredrik's documentation is enormously valuable if you ask me. I almost
always have a local copy open when I'm doing Tkinter programming.

Alas, it seems that the width and height options of the Text widget
aren't automatically updated. If you need the new size in characters,
it may work to find the size of your characters using the tkFont
module and do some division.

>>> from Tkinter import *
>>> import tkFont
>>> r=Tk()
>>> f=tkFont.Font(font="7x13")
>>> f.measure("x")
6
>>> f.metrics()
{'ascent': 10, 'linespace': 11, 'descent': 1, 'fixed': 1}

But sometimes that seems not to work quite right for me.

Regards,
Matt


From peter.milliken at gtech.com  Sun Feb 17 23:37:00 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 18 Feb 2002 15:37:00 +1100
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org>
Message-ID: <a4q0f4$akj4@news1.gtech.com>

Why not use a language that provides your requirements of signal safety etc
and do a Python API for it? (assuming one doesn't exist) i.e. I find it
somewhat difficult to reconcile your statement about responding to time
critical signals when you propose use of an interpretive language..... :-)
Sounds like you would ideally like real-time language features in a language
that isn't designed for it (well, I'm not sure Guido had visions of Python
being used in real-time applications :-)).

Peter

"Mark Mitchell" <mark at codesourcery.com> wrote in message
news:mailman.1013998768.20770.python-list at python.org...
> Python provides synchronous signals, even though signals are
> asynchronous events at the operating system level.  In many ways,
> this is a Good Thing.  For example, in C you can't do much of anything
> from within a signal handler since you don't know much about the
> current state of the application, but in Python you can do just
> about anything.
>
> In particular, you can throw an exception from a Python signal handler,
> which makes handling signals simple.  In C, you generally have to
> set a flag from within the signal handler, and then poll the flag
> elsewhere; in Python you simply have the signal handler thrown an
> exception and then provide an appropriate exception handler.
>
> But, I'm having a hard time figuring out how to write signal-safe
> code in the presence of synchronous signals.  Consider, for example,
> the following:
>
>   pipe = os.pipe()
>
>   try:
>     # Do stuff with the pipe.
>   finally:
>     # Close the ends of the pipe.
>     os.close(pipe[0])
>     os.close(pipe[1])
>
> This code is supposed to make sure that the file descriptors are
> not leaked; when we exit this scope, the descriptors will be closed,
> even if an exception occurs.  [Note that I am assuming that os.close
> will never thrown an exception when given a valid file descriptor.
> That's not strictly speaking true on some operating systems, but
> let's pretend that it is true.  For the purposes of this discussion,
> the particular functions involved don't matter.]
>
> This code, however, is not signal-safe.  For example, if a signal
> occurs between the two calls to os.close, we will end up not closing
> the second descriptor.  The equivalent C++ code *would* be signal safe;
> an exception cannot be thrown from the signal-handler, so if the program
> is still executing we can be sure that both calls to os.close will
> occur.
>
> One solution, of course, is to make sure that the signal-handlers do
> not throw exceptions.  But that's not a reasonable solution in library
> code; the library can't go around deciding what signal handlers should
> be doing.
>
> Another solution is to install an alternate signal handler around the
> entire try-block.  This alternate signal handler would remember the
> signal; we could then regenerate it at the end of the try-block.  This,
> however, is not very robust; what about multiple signals occurring?  And
> if the signal that occurs is important, we're ignoring it until the end
> of the try-block, which isn't polite.
>
> Another solution would be to provide a module that provides access to
> sigprocmask.  Then, one could explicitly block signals during parts
> of the code where these problems could occur.  For example:
>
>   pipe = os.pipe()
>
>   try:
>     # Do stuff with the pipe.
>   finally:
>     # Block signals.
>     sigprocmask(...)
>     # Close the ends of the pipe.
>     os.close(pipe[0])
>     os.close(pipe[1])
>     # Unblock signals.
>     sigprocmask(...)
>
> This does not work either; there are still at least two race conditions.
> A signal that arrives after we have entered the finally clause, but
> before we have called sigprocmask, would still result in a failure to
> close the descriptors.  Similarly, a signal that occurs after the call
> to os.pipe, but before the try-block is entered would result in us
> skipping the finally clause.  (And moving the try earlier is not valid;
> we can't clean up the descriptors unless we know that the pipe has been
> created.)
>
> Here is another try:
>
>   # Block signals.
>   sigprocmask(...)
>
>   try:
>     pipe = os.pipe()
>
>     try:
>       # Do stuff with the pipe.
>     finally:
>       # Close the ends of the pipe.
>       os.close(pipe[0])
>       os.close(pipe[1])
>   finally:
>     # Unblock signals.
>     sigprocmask(...)
>
> This variant is robust; we are now guaranteed that if the pipe is
> created, the descriptors will be closed.
>
> However, it is still unsatisfactory in that it results in the signal
> being delayed for an arbitrary amount of time; if a time-critical
> signal happens during the try-block, we will not process it until
> much later.
>
> Another solution is to require that the code be run in a thread other
> than the main thread; since only the main thread receives signals,
> the signal-safety problem does not occur.  However, it seems excessive
> to require threads simply to write signal-safe code!  And if you have
> code that depends on, say, getting SIGPIPE, then you have to have some
> way of communicating the signal from the main thread to the thread
> running the code above.
>
> What do Python programmers that need to write truly robust code do
> about this problem?  Is there a solution that I have missed?
>
> If not, would the Powers That Be be amenable to making changes to the
> language to support signal-safety?  One possible fix would be
> except/finally clauses that blocks signals on entry.  For example,
> if there were a finally_block_signals clause, you could do:
>
>   # Block signals.
>   sigprocmask(...)
>
>   try:
>     pipe = os.pipe()
>   except:
>     # Unblock signals.
>     sigprocmask(...)
>     # Reraise the exception.
>     raise
>
>   try:
>     # Unblock signals.
>     sigprocmask(...)
>     # Do stuff with the pipe.
>   finally_block_signals:
>     # Close the pipe ends.
>     os.close(pipe[0])
>     os.close(pipe[1])
>     # Unblock signals.
>     sigprocmask(...) # There needs to be a way of getting the signal
>                      # mask on entry to the finally clause.
>
> This is not particularly pretty, but it would work.
>
> Thoughts?
>
> --
> Mark Mitchell                mark at codesourcery.com
> CodeSourcery, LLC            http://www.codesourcery.com
>




From bokr at oz.net  Fri Feb 22 15:52:33 2002
From: bokr at oz.net (Bengt Richter)
Date: 22 Feb 2002 20:52:33 GMT
Subject: Reloading modules
References: <a51tt1$msq$1@news1.ucsd.edu> <u664rm510.fsf@python.net> <u1yffm45s.fsf@python.net>
Message-ID: <a56b2h$isr$0@216.39.172.122>

On Thu, 21 Feb 2002 12:02:08 GMT, Michael Hudson <mwh at python.net> wrote:

>Michael Hudson <mwh at python.net> writes:
>
>> Jeff Davis <jdavis at empires.org> writes:
>> 
>> > Is there a way to say that all calls to import should check for a newer 
>> > version, if a newer one exists then completely drop the old one and import 
>> > the new one?
>> 
>> An import hook could do this.  Wouldn't be trivial to write, but not
>> too hard, either.
>
>This "seems to work".  It's probably still a bit fragile, but might be
>handy during development, I guess.
>
>import __builtin__, os, time
>
>old_import = __import__
>
>def my_import(*args):
>    mod = old_import(*args)
>    file = getattr(mod, "__file__", "")
>    if file.endswith("c"):
>        file = file[:-1]
>    if not file.endswith(".py"):
>        return mod
>    mtime = os.path.getmtime(mod.__file__[:-1])
>    itime = getattr(mod, "__importtime__", 0)
>    if itime < mtime:
>        reload(mod)
>    mod.__importtime__ = time.time()
>    return mod
>
>__builtin__.__import__ = my_import
>
In general, I'd be wary of using time() and getmtime() to detect file changes
programmatically if there's the faintest chance of a change happening in
less than two seconds (the file time stamp resolution of, e.g., Windows 95).

You can sometimes get bit even in a fast IDE edit-build situation.

Slower, but reliable, might be to check an md5 digest of the file for change.
Of course, then you could not expect to trigger this with touch.

Regards,
Bengt Richter



From jonathan at onegoodidea.com  Sat Feb  9 16:36:01 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Sat, 09 Feb 2002 21:36:01 +0000
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org> <B88AE186.48CD%jonathan@onegoodidea.com> <a43ipn$s37$1@panix2.panix.com>
Message-ID: <B88B46C1.48F3%jonathan@onegoodidea.com>

On 9/2/2002 16:29, in article a43ipn$s37$1 at panix2.panix.com, "Aahz Maruch"
<aahz at panix.com> wrote:

> My Apple-pushing partner told me that MacAddict says that G3 isn't
> sufficient for running OS X.  I guess you feel differently?

A couple of friends and I were swapping timings the other day and we got
this table for a 'make -j2 all' of Python plus all the Mac modules:

   What         CPU        Time
   ----------------------------
   iBook        600Mhz     7:23
   TiBook       500Mhz     6:27
   G4       2 x 800Mhz     3:32
   G4       2 x   1Ghz     2:45

which seems to suggest that a G4 is about 40% faster clock-for-clock than a
G3 running a fairly mixed task. All the OS X eye candy is optimised for the
Altivec unit in the G4, so it generally feels snappier.

But still, I've got three iBooks ranging from 300MHz to 600MHz and all run
OS X well enough for me. The main thing that OS X loves is RAM - it just
can't get enough of it.

> The other
> thing is that Apple's notebooks have traditionally been no more reliable
> than PC notebooks, but the first generation iBooks were built like
> tanks.  How true is it of the current generation?

There's no comparison between my two white iBooks and my old orange
warhorse, but the new ones still feel pretty solid. They're very rigid in
comparison to something like Sony's ultraportables (polyurethane case and
magnesium frame) and the drive is rubber mounted. I had to send one back
after the latch broke, but generally I'm pretty happy with them.

> Also, the screen sizes aren't exactly the same, but I'd say that 14.1"
> is probably better for 1024x768 than 15" (I'd want 1152x864 for the
> latter).  Looking at the Apple web site, actually, I think I'd just get
> the 12" screen and use one of my existing external monitors at home.

Sorry, I meant in terms of pixels rather than inches. I had a quick play
with one of the new 14" iBooks in a shop, but having used the 12" one it
feels out of proportion. The screen on the 12" is very very sharp and really
quite beautiful to use. Unless you have difficulties with such a high
resolution, I'd stick to the 12" one (though the 14" one gets a bit longer
battery life).

I think the new iMac is a lovely piece of kit and, compared to Apple's pro
machines, very good bang for buck. I just find the ability to work anywhere
much more compelling.

Jonathan



From hans at inenco.no  Tue Feb 12 09:27:03 2002
From: hans at inenco.no (Hans Kristian Ruud)
Date: Tue, 12 Feb 2002 14:27:03 GMT
Subject: Drawing in "Slow motion".
Message-ID: <3C692698.C9DAD820@inenco.no>

Hello.

I have written a small script concerned with some drawing into a Canvas
widget
(The actual drawing procedure is triggered by an event bound to a
Button).

My problem is that I would like the program to draw one line at a time;
then
pause, then draw the next line... and so on.

I have tried to insert pauses by calling time.seep,
however, the drawing operations are realized only
when the drawing procedure ends and program control is resumed by

root.mainloop()

Any suggestions?


Thanx in advance

- hans kristian -




From aries at bitstream.net  Sat Feb 23 04:14:14 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Sat, 23 Feb 2002 03:14:14 -0600
Subject: pythonwin using COM import pythoncom error
Message-ID: <3c775bc7$0$79558$6536d21b@news.bitstream.net>

Hey Pythonaholics,

Heres my situation on trying to use COM.

after installing the python ver 2.0 from the book core
 python programming, then installed the pythonwin
 from the same CD area.

I wanted to monkey around with the COM stuff and my
first attempt went badly when I typed this

>>> import pythoncom
Traceback (innermost last):
File "<interactive input>", line 1, in ?
ImportError: DLL load failed: A device attached to the
 system is not functioning.
>>>

---- so then I looked in the python-install log files
and found this:
.......
RegDB Val: C:\Python20\win32\Help\Python Win32 Extensions.hlp
RegDB Root: 2
File Copy: C:\Python20\win32\Help\Python Win32 Extensions.fts
File Overwrite: C:\WINDOWS\SYSTEM\MSVCIRT.DLL
Date/Time: Following file not copied.
File Overwrite: C:\WINDOWS\SYSTEM\MSVCRT.DLL
Made Dir: C:\Python20\win32\demos
File Copy: C:\Python20\win32\demos\cerapi.py
.....etc

--the file that not copied MSVCIRT.DLL
I see 2 diff files here MSVCIRT.DLL, MSVCRT.DLL

what should I do ????
how can I get this file onto my system  ?????
is this my problem then when I try to import pythoncom ????

sega,
Joe F.






From gs at styrax.com  Thu Feb 28 14:11:25 2002
From: gs at styrax.com (Garry Steedman)
Date: Thu, 28 Feb 2002 19:11:25 +0000
Subject: Python DB that supports numbers
In-Reply-To: <20020228131104.A2852698@vislab.epa.gov>
References: <3C7E7E34.24356.6EB01AA@localhost>; from gs@styrax.com on Thu, Feb 28, 2002 at 07:00:04PM +0000
Message-ID: <3C7E80DD.3655.6F5638C@localhost>

randall,

http://sourceforge.net/projects/gadfly/

should work...

hth,

garry

On 28 Feb 2002, at 13:11, Randall Hopper wrote:

Date sent:      	Thu, 28 Feb 2002 13:11:04 -0500
From:           	Randall Hopper <aa8vb at yahoo.com>
To:             	Garry Steedman <gs at styrax.com>
Copies to:      	python-list at python.org
Subject:        	Re: Python DB that supports numbers

> Garry Steedman:
>  |> Just a simple DB for quick projects without the setup headache of
>  |> server-based database systems. | |gadfly:
>  http://www.chordate.com/gadfly.html
> 
> Thanks.  I'd actually tried to check that one out before mailing,
> but...:
> 
>      ERROR
>      The requested URL could not be retrieved
>            Connection Failed 
>      The remote host or network may be down. Please try the request
>      again. 
> 
> so I wasn't sure if it was still maintained.  I'll check back later to
> see if the site recovers.
> 
> Randall
> 
> -- 
> Randall Hopper
> aa8vb at yahoo.com


+-------------------------------------------+
  Garry Steedman      mailto:gs at styrax.com  
  Styrax Associates  http://www.styrax.com/ 

  "The Good Man has no shape."            
+-------------------------------------------+



From cfelling at iae.nl  Wed Feb 27 22:16:35 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 04:16:35 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org>
Message-ID: <a5k7ej$d5a$1@animus.fel.iae.nl>

James_Althoff at i2.com wrote:

> Below is a revision (not yet submitted) of PEP 276.

> Rather than restart the entire debate on PEP 276, please send me
> suggestions for specific changes to the PEP.

I'm sorry not to obey, but I find this PEP so hard to appreciate and
you being such a heavy weight that I fear Guido might be lured into
thinking this to be a good idea that I felt compelled to react in
public.

I admit up front that this is indeed a very minimalistic solution.
But I fail to understand how you can propose to range over a number.
Maybe it's my flaky understanding of the deeper levels of math, but
to me a number is a singular thing, not like a sequence at all.
So please explain why you think it natural to loop over a number.


...
>     - Unwieldy, especially when combined with the len() function,
>       i.e., xrange(len(sequence))

I miss in your discussion mentioning of the proposal to enhance (x)range
to accept a sequence as parameter and generate the indici for it.

...
>     Finding a new syntax that is comprehensive, simple, general,
>     Pythonic, appealing to many, easy to implement, not in conflict
>     with existing structures, not excessively overloading of existing
>     structures, etc. has proven to be more difficult than one might
>     anticipate.

And I don't see convincing arguments against having to use a function
call to get the indici sequence derived from a single number like in:

     for rowcount in range(table.getRowCount()):

Your proposed:

>      for rowcount in table.getRowCount():

really makes me wonder, what the heck is going on here.  I would
expect table.getRowCount() to return an integer, but instead it
returns a sequence, better look up what the code does.  Wonder,
wonder, it *does* return an integer. Wonder, wonder.  Oh wait, someone
better versed in math seemed it natural to view an integer as the
sequence of integers that precede it.

Sorry that the above sounds pissed, think I am, know I shouldn't:(

>From the discussions at the time, I understood there were more
that really didn't understand how you could loop over a single
number, and here in your PEP you make it look like some minority
of nitwits objected, that it was merely a matter of preference.
I obviously don't agree with that.  It's not a matter of preference,
it's lack of understanding what it means to loop over a single number.

So please add a section in your PEP explaining why it is natural
to loop over a number.  Trow in some math theory or whatever,
educate us, but don't treat it as a matter of taste.

You see, I would understand things like
    for room in house:
or
    for tree in forest:
but I fail to understand things like
    for atom in atom:


>     - Some feel that iterating over the sequence "0, 1, 2, ..., n-1"
>       for an integer n is not intuitive.  "for i in 5:" is considered
>       (by some) to be "non-obvious", for example.

>       Response: This seems to be largely a matter of preference.
>       Some like the proposed idiom and see it as simple and
>       elegant.  Some are neutral on this issue.  Others, as noted,
>       dislike it.


>     The majority of disagreement with PEP 276 came from those who
>     favor much larger changes to Python to address the more general

At the time I really thought it was your april 1st joke.  And I wouldn't
be surprised when others thought the same.  How sad we didn't react
there at that moment, so the counts would be more balanced:)

-- 
groetjes, carel


From sab at NOSPAM.freeuk.com  Thu Feb 28 13:29:17 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Thu, 28 Feb 2002 18:29:17 -0000
Subject: Can python handle operating a modem?
Message-ID: <a5ltoj$m59$1@newsg2.svr.pol.co.uk>

Can python handle operating a modem? i.e. can i say, lift the hook on the
line to see if there is a dial tone? and test this? are there any modules
available for this?

Thanks,

G. Willoughby




From jason at jorendorff.com  Mon Feb 18 00:38:01 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 17 Feb 2002 23:38:01 -0600
Subject: Editable strings anyone?
In-Reply-To: <7xr8njv243.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEFIDDAA.jason@jorendorff.com>

Paul Rubin wrote:
> Jason Orendorff wrote:
> > There's no gap in arrays.
>
> OK.  I guess the simplest approach to editable strings is to have
> a new type of array
> that has multiple segments and implements the buffer interface.
> Then the re module can be extended to handle the buffer interface,
> if it doesn't handle it already.

Someone may have mentioned this before, but I think you would have
to sneakily close the gap whenever the re module asks you for your
buffer pointer, because re needs a contiguous buffer.  See:

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/python/dist/src/Module
s/_sre.c

About halfway down, you'll see a function getstring() which returns
a pointer to the searchable data in an object.  This is where the
requirement of a *contiguous* buffer is enforced.

  if (... buffer->bf_getsegcount(string, NULL) != 1)
          ... then raise an exception.

## Jason Orendorff    http://www.jorendorff.com/




From idot at vt.edu  Wed Feb 20 16:56:43 2002
From: idot at vt.edu (Carl Banks)
Date: Wed, 20 Feb 2002 16:56:43 -0500
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de>
Message-ID: <r2615a.ii.ln@127.0.0.1>

Karl M. Syring wrote:
> "Carl Banks" <idot at vt.edu> schrieb>
> <snip>
>> Flight simulation: The earth's radius is about 20 million feet (about
>> 7 million meters for my SI using friends), and the simulator must
>> resolve, at a minimum, distances of about 0.1 feet.  That's a
>> difference of 9 decimal places.  We then need several more places for
>> numerical robustness.
> <snip>
> 
> You surely are joking, Mr. Banks?
> This is exactly the kind of misuse of floating point that was referred to in
> a previous posting. Every computational geometry software would use a
> rational data type here, although you might get away with simple point data
> binning in this case.


I suspect you don't know what I'm talking about, since what I'm
referring to is nothing like computational geometry.  Nevertheless,
I'll answer this.

No, I'm not joking.

I have seen the source code of maybe a dozen flight simulators, and
I've written several myself (not in Python, though, yet).  Several of
the simulators are used in research, one came out of NASA, and one was
proprietary.  I have yet to see anything stored as a rational number.

High-fidelity, long-period spacecraft simulations might require
something more precise than 64-bit floating point, but for atmospheric
flight, unmodeled disturbances make the extra precision pointless.



sorry-that-this-is-straying-off-topic-but-it-could-be-worse-as-I-
could-be-replying-to-Timothy-Rue-so-I-hope-you'll-forgive-me-and-
is-this-closing-thingie-some-obscure-Monty-Pythonism-or-what'ly yours,

-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From theller at python.net  Tue Feb 26 15:41:09 2002
From: theller at python.net (Thomas Heller)
Date: Tue, 26 Feb 2002 21:41:09 +0100
Subject: py2exe for Python 2.2 not finding modules? [long]
References: <ebcd2ef6.0202251650.5207cb28@posting.google.com>
Message-ID: <a5gruh$76pch$1@ID-59885.news.dfncis.de>

"Erik Westra" <ewestra at wave.co.nz> wrote in message news:ebcd2ef6.0202251650.5207cb28 at posting.google.com...
> Hi,
>
> Has anyone had any joy getting the latest version of py2exe (0.3.1)
> working with Python 2.2?  I've been using py2exe successfully for
> months with Python 1.5.2, but after upgrading my dev environment to
> Python 2.2 I suddenly find py2exe failing to find some of the modules
> in my rather large wxPython app.

[snip]

> I'm running under win2k, and uninstalled Python 1.5.2 before
> installing Python 2.2.  Everything else seems to work fine -- it's
> just py2exe which is behaving strangely.

Python 1.5.2 and Python 2.2 (at least the Pythonlabs versions)
can coexist on the same PC, so you can try also building with 1.5.2,
probably with an older version of py2exe, and look for differences.

The alternative would be to zip your app (maybe stripped down somewhat)
and mail it to me for inspection.

Regards, and thanks for using px2exe,

Thomas




From mkelly2002NOSPAM at earthlink.net  Mon Feb 18 17:17:05 2002
From: mkelly2002NOSPAM at earthlink.net (Michael Kelly)
Date: Mon, 18 Feb 2002 22:17:05 GMT
Subject: ActiveX Questions
References: <mailman.1014061903.12565.python-list@python.org>
Message-ID: <0uu27ukp4jpnu4o5pjrb8tuqvedtf556s6@4ax.com>

On Mon, 18 Feb 2002 11:48:01 -0800, "Jimmy Retzlaff"
<jimmy at retzlaff.com> wrote:

[snip]

Here's a python script I have for using an
"invisible" ActiveX Control that dynamically
creates and displays Message Dialogs with
cool color backgrounds:

(please forgive newsreader word wrap messing up the script)

# Demonstrates handling an event fired from an .ocx ActiveX Control in
# Python
# November 20, 2001
# Michael Kelly
#
# The ActiveX Control provides dialog boxes with mixed color
#backgrounds.
# If a HelpCtx other than -1 is passed to a method a Help button is
#also
# displayed in the dialog.  The demo handler below handles the event
#for
# the Help button press.

import pythoncom
import win32com.client
import GradDlgX
import win32com.server.util
import os

HelpMsgs = ["Some Help Appropriate to Context 0",
            "Some Other damn Message for Context 1",
            "Okay, you Pressed Help with Context 2 Oberday!",
            "Alright already!  Help Context 3"]

def InternalHelpFunc(HelpCtx):
    if HelpCtx >= 0 and HelpCtx < len(HelpMsgs):
        print HelpMsgs[HelpCtx]

def ExternalHelpFunc(HelpCtx):
    if HelpCtx >= len(HelpMsgs):
        os.startfile("D:\Program Files\Microsoft Visual
Studio\MyProjects\MyActiveX\GradDlgX.hlp")
        
class HelpHandler:
    def OnContextHelp(self, HelpCtx=GradDlgX.defaultNamedNotOptArg):
        #self.__dlg =
win32com.client.Dispatch("GradDlgXControl1.GradDlgX")
        #if HelpCtx >= 0 and HelpCtx < len(HelpMsgs):
        if 0 <= HelpCtx < len(HelpMsgs):
            InternalHelpFunc(HelpCtx)
        else:
            ExternalHelpFunc(HelpCtx)

EDlg = win32com.client.DispatchWithEvents("GradDlgXControl1.GradDlgX",
HelpHandler)
EDlg.InfoDlg("Hello", 1, 1, 2)

=== end of script ===

Most of my ActiveX are of the "invisible" variety so
there may be container issues that don't apply.

The ActiveX Control in this case is one created by
Delphi 5 wizard from a Delphi "TWinControl"
component.

[Note that one fun thing with this is that the
ActiveX uses a local dynamically created
dialog, so in the help handler I can call
the ActiveX Control to make another dialog
to show a msg if I like.]




--

"I don't want to belong to any club that would have me as a member."
    -- Groucho Marx


From jwbaxter at spamcop.net  Thu Feb 14 10:03:59 2002
From: jwbaxter at spamcop.net (John W. Baxter)
Date: Thu, 14 Feb 2002 07:03:59 -0800
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net> <a4geh2$hir$12@bob.news.rcn.net>
Message-ID: <140220020703593280%jwbaxter@spamcop.net>

In article <a4geh2$hir$12 at bob.news.rcn.net>, <jmfbahciv at aol.com> wrote:

> In article 
> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345 at lp.airnews.net>,
>    claird at starbase.neosoft.com (Cameron Laird) wrote:
> >In article <130220020751360934%jwbaxter at spamcop.net>,
> >John W. Baxter <jwbaxter at spamcop.net> wrote:
> >                       .
> >                       .
> >                       .
> >>Did the elves at AT&T pick up the "leading 0 means octal" from
> >>something earlier, or did they invent this stupidity themselves?
> >                       .
> >                       .
> >                       .
> >OK, folks, how far back *can* we trace this?  My
> >mind associates it with DEC systems, going back
> >to the '60s, but I couldn't find any confirmation
> >of that in a quick search.
> 
> You will never find a DEC standard that ever defined
> a number with a leading zero to be octal.  I never
> typed up anything that ever had that kind of standard. 

Well, Digital's hardware was what the AT&T elves worked with, so the
connection is certainly there, but not necessarily coming in from DEC.

I can't remember from my not-extensive messing about with PDP-1 (serial
1, provided at no charge and with support by DEC) at MIT whether that
assembler used a leading 0 to signal octal, or used something else. 
[Octal was natural on that machine...hex wasn't an issue.]  That
assembler pretty much moved to the PDP-1 from the older TX-0
("Transistorized eXperimental" and the round thing is zero) next door.

And then moved on to DEC along with a bunch of people, who were key
parts of the technical staff for a long time.  That was a very fertile
"seeding" by the General.  And produced "SpaceWar."

  --John


From mike at nthwave.net  Fri Feb 15 21:37:57 2002
From: mike at nthwave.net (Michael Mell)
Date: Fri, 15 Feb 2002 18:37:57 -0800
Subject: line break confusion
Message-ID: <3C6DC606.5482C478@nthwave.net>

Using Python 2.2 running on any arbitrary OS (Linux, Win, Mac), I want
to
    1) read several files having an arbitrary mix of line break types
(Linux, Win, Mac)
    2) then standardize the breaks of each of the files to a single type
and
    3) then concatenate them all into a string for writing.

Problems:
Linux and Mac:
    I have not found the magic string which represents the Windows line
break on Linux or Mac. Neither '\r\n', nor '\n\r' works.
Mac:
    It seems that RETURN and NEWLINE are swapped: an '\r' creates Linux
breaks, an '\n' creates Mac breaks.


In the string module, whitespace = ' \t\n\r\v\f' (5 characters here).
Yet, when I run this:
>>> import string
>>> for c in string.whitespace:
>>>     print '.'+ c + '.'

I get 7 results.
What's going on here?

Thanks
Mike


--
mike at nthwave.net
llemekim         YahooIM
415.455.8812     voice
419.735.1167     fax





From gh_pythonlist at gmx.de  Mon Feb 18 04:04:15 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 10:04:15 +0100
Subject: Embedding python with lcc-win32
In-Reply-To: <335764c1.0202170339.8db63a0@posting.google.com>
References: <335764c1.0202161031.425dd81a@posting.google.com> <335764c1.0202170339.8db63a0@posting.google.com>
Message-ID: <20020218090414.GA2258@lilith.hqd-internal>

Le 17/02/02 ? 03:39, decay ?crivit:
> ae98096 at aero.iitm.ernet.in (decay) wrote in message news:<335764c1.0202161031.425dd81a at posting.google.com>...
> > Hello.
> > 
> > I've been trying to embed python in a C program. I use the lcc-win32
> > compiler system. I made an import library from python21.dll using
> > implib. But when I link a simple program, I get "Undefined reference
> > to __imp__Py_Initialize" in test.c. I suppose its got to do with
> > making the import library. Can anyone please enlighten me?
>
> I got my simple python-embedded-in-C program (read below) to compile
> with lcc-win32 by doing the following:
> 
> I used buildlib with the following file as argument:
> -----------------
> python21.dll
> __imp__PyArg_Parse PyArg_Parse
> __imp__PyArg_ParseTuple PyArg_ParseTuple
> ...
> -----------------
> 
> The program now compiles, but when I execute it, it "caused an error
> in <unknown>". Can anyone tell me what's happening?

Out of interest, I've tried to look at lcc-win32 again. It looks like
you didn't convert the MSVC-style python22.lib. Do that (Tools/Import
library or some such) and then just add "python22.lib" to your libraries
list. A minimal test program then runs fine on my machine, except it
crashes somewhere after "return 0" in the main function. I guess the
reason is that lcc-win32 links against the darn crtdll. If you use
msvcrt.dll like Python does, you should be fine.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.6 ?C      Wind: 2.8 m/s



From sabren at manifestation.com  Thu Feb 28 22:52:42 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 28 Feb 2002 22:52:42 -0500 (EST)
Subject: super confused !
In-Reply-To: <txBf8.24714$ZC3.1996215@newsread2.prod.itd.earthlink.net>
Message-ID: <Pine.LNX.4.44.0202282251001.16877-100000@mercury.sabren.com>

On Fri, 1 Mar 2002, robin and jim wrote:

> >>> class Int_c(Int_a):
> ...  def __init__(self):
> ...   super(Int_c, self).__init__(99)
> ...
> >>> Int_c()
> 0
> >>> class Int_d(Int_a):
> ...  def __init__(self):
> ...   Int_a.__init__(self, 99)
> ...
> >>> Int_d()
> 0
> 
> Why are Int_c() and Int_d() zero (the default for int()) rather than 99?
> 
> How can I make them 99?

use __new__ instead of __init__... Guido explains it in 
"Unifying types and classes in Python 2.2":

      http://www.python.org/2.2/descrintro.html#__new__

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From chris at cmb-enterprises.com  Mon Feb 25 14:48:04 2002
From: chris at cmb-enterprises.com (Chris Dutton)
Date: Mon, 25 Feb 2002 19:48:04 -0000
Subject: help: Problem with cgi form
References: <mailman.1014420097.13372.python-list@python.org>
Message-ID: <20020225144728494-0500@news.fltg.net>

In <mailman.1014420097.13372.python-list at python.org> Gustavo Cordova
<gcordova at hebmex.com> wrote:
>> 
>> I'm a beginner. How do I do this? It's probly very simple ......
>> 
>> I have a hidden input on a cgi form... 
>> 
>>     print '<INPUT TYPE=HIDDEN NAME= "last_pos" 
>> VALUE=',last_byte_pos,'>'
>> 
>> I need the VALUE to be the  variable "last_byte_pos", but 
>> unfortunately 
>> when I run the script for the first time the script crashes 
>> because it's 
>> undeclared. 
>> 
>> What subroutine do I use to check if last_byte_pos is undeclared, and >> assign it a default value if it is? 

How about...

import cgi
form = cgi.FieldStorage(keep_blank_values=1)
if form.has_key("last_byte_pos"):
	last_byte_pos = form["last_byte_pos"]
else:	last_byte_pos = default_value


From gmcm at hypernet.com  Sun Feb  3 09:25:30 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Feb 2002 14:25:30 GMT
Subject: Is there a Metakit release for python 2.2?
References: <11e94203.0202010858.17a6bce7@posting.google.com>
Message-ID: <Xns91AA6004B50gmcmhypernetcom@199.171.54.215>

Jose Rivera wrote:

> I just Installed Python 2.2 and I am getting the followin error:
> 
> Traceback (most recent call last):
>   File "D:\scripts\Rev_Si_Hay_Archivos_En_BCKS_Sin_Procesar.py", line
> 2, in ?
>     import metakit
>   File "D:\TOOLS\Python22\lib\metakit.py", line 12, in ?
>     from Mk4py import *
> ImportError: Module use of python21.dll conflicts with this version of
> Python.
> 
> It worked with Python 2.1, and I searched on the WEB and did not found
> a comment about a new release.

Announced day before yesterday:
   http://www.equi4.com/metakit/

-- Gordon
http://www.mcmillan-inc.com/


From jdgorman at ieee.org  Wed Feb  6 00:39:47 2002
From: jdgorman at ieee.org (John Gorman)
Date: Wed, 06 Feb 2002 05:39:47 GMT
Subject: wxPython book
References: <88bc63c6.0202040505.2a9c3016@posting.google.com>
Message-ID: <3C60C1CC.9080002@ieee.org>

Doug,
   There may be a book in the works, but 'til then check out the python 
pit ;^) ... There are several interesting examples there.  Also, read 
through the wxPython demo source.

http://wxpython-pit.sourceforge.net/

John Gorman

Doug Farrell wrote:

> Hi all,
> 
> I'm busy evangilizing my office co-workers on the wonders of Python
> and wxPython, to which they are responding pretty well. We are
> currently working on a test project that uses both, and the most
> frequent complaint I get is "there's no book on wxPython..... wah wah
> wah". I have seen mention of a book being written, presumably by Robin
> Dunn. Is this true and if so, any idea when it might be coming out?
> 
> Thanks,
> Doug
> 



From zeitlin at seth.lpthe.jussieu.fr  Sat Feb 16 16:21:43 2002
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: Sat, 16 Feb 2002 21:21:43 +0000 (UTC)
Subject: wxPython:  EVT_CLOSE question
References: <c0a56015.0202151702.1f29d19e@posting.google.com>
Message-ID: <slrna6tjb6.e10.zeitlin@seth.lpthe.jussieu.fr>

On 15 Feb 2002 17:02:45 -0800, alan scales <alans at killergame.com> wrote:
> i have some controls (in particular wxStyledTextCtrl) - they are NOT
> frames or dialogs.  I want to be able to trap when they are closed (or
> destroyed)(e.g. to save data). I was hoping to use EVT_CLOSE and
> provide a handler.  however it does not seem to work.
> 
>   1. is it supposed to work? (it is only documented for
> frames/dialogs)

 No, it isn't supposed to work for the simple controls. It only works for he
top level windows.

>   2. if not, what can be done to detect when a control is closed?

 The best would be to do it in EVT_CLOSE of the frame containing the control.
You also can use EVT_WINDOW_DESTROY but this could be too late to do anything
meaningful with the control.

 Regards,
VZ
-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From SBrunning at trisystems.co.uk  Thu Feb 21 12:11:02 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 17:11:02 -0000
Subject: Simple regexp problem
Message-ID: <31575A892FF6D1118F5800600846864DCBCCE7@intrepid>

> From:	Shane Hoversten [SMTP:srh232 at nyu.edu]
> Sent:	Thursday, February 21, 2002 5:04 PM
> To:	python-list at python.org
> Subject:	Simple regexp problem
> 
> Hi all,
> 
> 	This question has gotta be real simple.  I want to remove preceeding
> 
> whitespace from a string.  Seems like I should be able to say (using 
> another example to make it easier to see):
> 
> re.sub("^1", "", "11111junk")
> 
> 	and get "junk".  Except it doesn't work.  What I get is:
> 
> '1111junk'
> 
> 	In other words, it replaces only the first "1", and no others.  I've
> read 
> the dox and the howto's and this is such a simple thing.  I can do it 
> another way, but I can't figure out why this doesn't work.  Can someone 
> help?
 
string = string.lstrip()

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From starx at pacbell.net  Sat Feb  9 16:08:15 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 21:08:15 GMT
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4> <3c658a5a_3@corp.newsgroups.com>
Message-ID: <Xns91B085A6FC07FRASXnewsDFE1@130.133.1.4>

Joshua Muskovitz || Sat 09 Feb 2002 12:55:09p:

> My suggestion would be for each type of receiver to simply create its
> own dict which maps message types (strings) to local functions.  If
> it's in the dict, then you can assume its callable, and thus avoid the
> logic per incoming message.  If you set up your constructors correctly,
> you could even inherit the list of existing handlers from the base
> classes.  Thus class BaseReceiver could create the dict which handles
> the MsgA and MsgB types.  A derived class could add a MsgC handler to
> the dict, etc.  This also allows derived classes to "overload" base
> handlers by simply replacing the function reference in the dict.
> 
> something like (coded off the top of my head, if it don't run, sorry
> :-) 

Aside from missing a __ on an init, i sure as hell can't see any errors :)
 
> class Receiver:
>     def __init__(self):
>         self.msgFuncs = {}
> 
>     def message(self, msg):
>         if self.msgFuncs.has_key(msg.type):
>             self.msgFuncs[msg.type](msg)
> 
> class F(Receiver):
>     def __init__(self):
>         Receiver.__init__(self)
>         self.msgFuncs['fighto'] = rec_fighto
> 
>     def rec_fighto(self, msg):
>         ...
> 
> class S(Receiver):
>     def __init(self):
>         Receiver.__init__(self)
>         self.msgFuncs['spello'] = rec_spello
>         self.msgFuncs['fighto'] = rec_fighto
>         self.msgFuncs['none'] = rec_none
> 
>     ...
> 
> There are better ways to handle the dict adds, of course, but you get
> the idea.  The cool thing about this is that on a per receiver basis
> you can have different classes of messages still handled by a single
> common handler (just point them to the same internal function).

Hm, well, i do like the idea of being able to point different messages at 
the same handler... I figured i would do that kind of stuff by modifing 
certain 'message' methods. I think what i want is to have the system as 
transparent as possible... but of course i don't really ever seem to know 
what i want untill i go and do it -some- way :)

Oh, and i don't really need the callable logic i don't think; it would be 
better to just make a project rule: 'dont ever name anything rec_xxx unless 
it's a function taking a message, unless you want to hurt yourself.' There 
really shouldn't be a reason to...

Thanks for the ideas... even if it doesn't change my mind i think it makes 
me realize in less vague terms why i want to do it a different way :)

Ja ne *sneeze*
 
> --
> lyyrs('Hire me!  I need the work!')
llyrs('Sorry just a semi poor college student here') ;)
-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From mjbarber at ascc.artsci.wustl.edu  Wed Feb 27 08:55:23 2002
From: mjbarber at ascc.artsci.wustl.edu (Michael James Barber)
Date: 27 Feb 2002 07:55:23 -0600
Subject: Profiling memory usage?
Message-ID: <a5iogb$laa@ascc.artsci.wustl.edu>

I'm currently looking to optimize a couple of my python programs in terms
of memory usage.  I think I know what I need to improve, but I suspect
that "profile before optimizing" applies to memory usage as well as to
speed.

Is there something analogous to the profile module for describing how much
memory is used by e.g. instances of different classes?  I didn't find
anything using Google or in the Vaults of Parnassus.

-- 
					Michael J. Barber


From excalibor at iname.com  Fri Feb 15 15:02:51 2002
From: excalibor at iname.com (David S. de Lis)
Date: 15 Feb 2002 12:02:51 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org>    <slrna6kkpj.is0.tim@vegeta.ath.cx> <pu6e4a.5ti.ln@wombat.san-francisco.ca.us>
Message-ID: <c131437e.0202151202.62f2a0b7@posting.google.com>

Hi there,

kkeller at speakeasy.net (Keith Keller) wrote in message news:<pu6e4a.5ti.ln at wombat.san-francisco.ca.us>...
> In article <slrna6kkpj.is0.tim at vegeta.ath.cx>,
> 	Tim Hammerquist <tim at vegeta.ath.cx> writes:
> 
> > IIRC, "agua" _is_ feminine, but takes the masculine article when
> > singular. Mi diccionario dice que "agua" es feminina, tambi?n.
> > 
> > OTOH, la palabra "mar" es masculina _y_ feminina. Por ejemplo:
> > 
> >     "el agua en la mar"     # Correct
> >     "las aguas del mar"     # Correct

'Agua' is a feminine word in spanish, as it is in latin, where spanish
evolved from.

The reason the male definite article is used with 'agua' is just for
phonetic reasons, to avoid a cacophonic (i.e. bad sounding) phrasing:

'la agua' -> 'la_agua' -> 'l\'agua'   which sounds horrible in spanish
(thoug it *is* said in certain 'subcultural environments').

'el agua' breaks the cacophony...

The plural, however, doesn't follow the exception, because articles
follow gramatically gender and number of their nouns, thus 'las
aguas'.

It's just the same with 'aguila' (eagle) and most short (one or two
syllabes) words that start with a-... ('aguila' has 3, of course...)

eg. 'el alma' (the soul) but 'las almas' against 'la arena' (the
sand).

While there are many theories, a particular school of thought in Spain
and other spanish-speaking countries consider the definite article as
an intrinsic part of the noun, not as an auxiliar adverbial form.
Effectively, definite articles are never found as language independent
units in spanish, only attached to a noun, so it makes lots of sense.
For many gender-independent words, like the ones ending in -ista we
are talking about in this thread, it is the article the one who
signals the gender. (eg. el pianista, la pianista (piano player, male
and female resp.)

Others, like 'doctor', 'm?dico', 'juez', etc which are all masculine,
are being counterattacked with their perfectly legal feminine forms
(doctora, m?dica, jueza, etc...) as women are socially allowed and
visible in those positions...

Another interesting word, and so signaled, used in the examples above
is 'mar' (sea), which is a masculine word but enjoys both genders in
the singular form: 'el mar' but also 'la mar'. While the latter is
considered a poetic form, it is widely used in many parts of south
Spain and America... It's always 'los mares' (the seas) however...
('las aguas de la mar', the waters of the sea)... being a monosylabic
word, having both genders allowed, among other things, to adjust the
metric of a verse by using the contracted of + the (de + el -> del)
masculine form or the uncontractable feminine 'de la'... This is also
found in other monosyllabic words (eg. 'el sal'* (the salt) incorrect
but used, comes to my mind... also, gender inversion is not uncommon
in the south of Spain...

And another interesting case that comes to my mind: 'cometa'. The
feminine form (la cometa, pl. las cometas) refers to a kite, while the
masculine (el cometa, pl. los cometas) refers to a comet. That's
polymorphism depending on the context, I guess ;-)

> Es mas malo que Perl!  ;-)

ahem... :)

and back to the original question, I've seen and used JAPH, perler,
perlite and other variations... there's more than one way to call
oneself... ;-)

salutations,
david


From mhammond at skippinet.com.au  Mon Feb 25 17:14:26 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 25 Feb 2002 22:14:26 GMT
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com>
Message-ID: <3C7AB746.1030800@skippinet.com.au>

dsavitsk wrote:
> just for the sake of giving credit where it is due, and not giving it where
> it is not, Mark Hammond is responsible for the excellent COM groundwork.
> ActiveState is responsible for laying him off.

LOL - thanks for the clarification.  I was going to post something 
similar, but it would not have sounded as good coming from me! :)

Mark.



From cguardia at aldeasystems.com  Thu Feb 21 20:16:53 2002
From: cguardia at aldeasystems.com (Carlos de la Guardia)
Date: Thu, 21 Feb 2002 19:16:53 -0600 (CST)
Subject: Any Pythonistas in Mexico?
Message-ID: <200202220112.g1M1C0k28307@mail.aldeasystems.com>

Hello,

my company is located in Mexico and we use Python for our development
projects, I am trying to find out how many of us Python enthusiasts live in
Mexico and see if maybe we can form a user group or something.

Thanks,

Carlos de la Guardia
Aldea Systems




From pereira at cis.upenn.edu  Sun Feb 24 08:44:05 2002
From: pereira at cis.upenn.edu (Fernando Pereira)
Date: Sun, 24 Feb 2002 08:44:05 -0500
Subject: Is there a python parser generator that is similar to JavaCC
	from webgain?
References: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com> <m38z9jpz4a.fsf@mira.informatik.hu-berlin.de> <B89D9263.3864%pereira@cis.upenn.edu> <m3it8n9ti8.fsf@mira.informatik.hu-berlin.de>
Message-ID: <B89E5855.38AA%pereira@cis.upenn.edu>

On 2/24/02 3:25 AM, in article m3it8n9ti8.fsf at mira.informatik.hu-berlin.de,
"Martin v. Loewis" <martin at v.loewis.de> wrote:

> Fernando Pereira <pereira at cis.upenn.edu> writes:
> 
>> JavaCC takes grammar specs (lexical patterns and rules) and produces a
>> parser (a Java class) using a LL(k)-type of rule compilation (that is,
>> deterministic recursive descent with lookahead).
> 
> Ok, I guess the closest thing for Python like this is YAPPS:
> 
> http://theory.stanford.edu/~amitp/Yapps/

Close, but more limited. JavaCC allows k lookahead for k > 1, Yapps just k =
1.

-- F



From sholden at holdenweb.com  Fri Feb 22 10:19:48 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 10:19:48 -0500
Subject: FILEDSN
References: <c13db1aa.0202220643.37578e1c@posting.google.com>
Message-ID: <Ujtd8.84991$n73.1594161@atlpnn01.usenetserver.com>

"ilgattodelvicino" <ilgattodelvicino at hotmail.com> wrote in message
news:c13db1aa.0202220643.37578e1c at posting.google.com...
> Can i use a file DSN with odbc module??
>
Sorry, you can anly use file DSNs with ADO components, as far as I know.
ODBC database systems need to pick up the appropriate ODBC driver for the
particular database instance you have to work with.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From python at chajadan.net  Tue Feb 26 02:23:26 2002
From: python at chajadan.net (chajadan)
Date: Mon, 25 Feb 2002 23:23:26 -0800
Subject: gnuplot woes, a second story
Message-ID: <5.1.0.14.0.20020225231647.00a94610@63.118.78.184>

When I try to run gnuplot from my Python2.2 installation, I get an error 
(listed below)

I'm on windows 98, I installed NumPy, I unzipped gp371w32.zip into my main 
python directory, and I tried changing the command to the gnuplot program 
FROM gnuplot_command = 'pgnuplot.exe' TO gnuplot_command = 
'c:\Python20\gp371w32\pgnuplot.exe'

I also changed the import of popen from win32pipe to os, but none of these 
configurations seem to work.

Can you help?

Here's the error:

Traceback (most recent call last):
   File "C:\PYTHON22\Pythonwin\pywin\framework\scriptutils.py", line 301, 
in RunScript
     exec codeObject in __main__.__dict__
   File "C:\Python22\Gnuplot-1.4\demo.py", line 129, in ?
     demo()
   File "C:\Python22\Gnuplot-1.4\demo.py", line 55, in demo
     g.reset()
   File "__init__.py", line 1244, in reset
     self('reset')
   File "__init__.py", line 1095, in __call__
     self.gnuplot(s)
   File "gp_win32.py", line 100, in __call__
     def __call__(self, s):
IOError: [Errno 32] Broken pipe




From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 09:42:15 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 08:42:15 -0600
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net>
Message-ID: <002c01c1b241$234c35e0$0101010a@local>

----- Original Message ----- 
From: "Ron Stephens" <rdsteph at earthlink.net>


> I am back from a trip, I have manually updated the mygale generated
> Pythonic web articels page on my web site. There are quite a few new
> links of interest.
> 
> However, the automated updating from my Linux box still did not work
> while I was gone. I will manually update once per day but I would 
> sure like to get the automaton working.
> 
> I am embarrassed to come back once more to ask for help on this 
> subject.
> I have used crontab to set up 
> a my_cron_file that runs five Python scripts to automatically run
> mygale.py and associated programs to 
> update the output.html file listing new web articles about Python 
> once per day.
> 
> My crontab file reads as suchas checked using crontab -l:
> 
> 30 4 * * * /usr/bin/python /home/ron/mygale.py get all
> 30 5 * * * /usr/bin/python /home/ron/urls.py
> 30 6 * * * /usr/bin/python /home/ron/urldb.py
> 00 7 * * * /usr/bin/python /home/ron/mygale.py -n 200 display all
> 30 7 * * * /usr/bin/python /home/ron/mygale-uploader.py
> 
> crond is running , as checked using
>   /etc/rc.d/init.d/crond  status which shows a running cron process 
> with a pid number.
> 
> Also,  ps aux | grep cron shows a /usr/sbin/cron process.
> 
> Finally, the  cron daemons log check /var/log/cron shows all 5 Python
> scripts properly called at their 
> appointed times each morning. No errors are at all are recorded in 
> the cron error log files.
> 
> Now, when I run these scripts exactly as typed above, from the Bash
> shell,
> everything works as planned and that's how I have been updating the
> online HTML page that shows the Pythonic articles on my web site.
> 
> The mystery is that, even though the log files show the Python 
> scripts called each day, no update occurs 
> on my web site . Indeed, when I check the output.html file in my
> /home/ron directory, it is not updated 
> at all by these cron processes. Even though the log file shows the
> scripts as being ran, it is as if they did not run at all, in that no
> output occurs.
>
> Yet when I run the same scripts manually, all is updated.
> 
> So this morning at 4:30 AM I sat up and ran a "top" process from the
> Bash shell, that shows all 
> running processes. Sure enough, at exactly 4:30 AM, a new Python 
> process was spawned. The new 
> Python process was matched with a new entry in the cron log file of 
> the mygale.py program being called.
> 
> I watched carefully, and the mygale.py process lasted for about 30
> seconds and then disappeared from 
> the "top" screen. I assume it terminated normally. Except that when I
> run the mygale.py get all script 
> manually  in the BASH shell, it takes about two minutes to complete.
> This is the only sign of any 
> problem, but still no error messages are generated anywhere I can 
> find.

cron errors are mailed to your system mailbox.  Have you tried typing
"mail" at the BASH prompt?

> Well, it seems to me that a series of processes on Linux or any
> operating system should be 
> deterministic. That is, things should make sense, when the exact same
> steps are taken, the result should 
> always be the same.

cron jobs run in a bit different environment than the command prompt;
this often leads to problems as you are noting.

Add this to the top of each script:

    import sys
    sys.stdout = open("/home/ron/.error-log","a")
    sys.stderr = sys.stdout
    print "\n(job name) Starting..."

then any error traceback goes to the file /home/ron/.error-log and you
can review them at your leasure.

> This procedure I am trying to do does not seem to make sense to me, 
> it seems mysterious. Linux should not be mysterious.
> 
> Can anyone help me?
> 
> By the way, shouldn't it be possible to cut and paste from the BASH
> shell to a text editor? It would save 
> me a lot of typing ;-))))

Depends on what terminal window you are using.  I'm not a KDE user,
but I think the normal KDE terminal window supports old-fashioned
cut-and-paste; highlight the text you want to copy, switch windows,
and select paste (no cut/copy step).  GNOME's terminal is positively
frustrating this way.

> The clipboard in general is a frustration; I can't yet make it work
> properly in many situations. Is it me 
> or does the KDE clipboard lack a lot of functionality compared to
> Windows clipboard? I can't even 
> copy and paste from K-EDIT, nor K-WORD, to Netscape Composer. Oh, 
> well.

Regarding Netscape Composer, I don't know; I use Mozilla now and
it generally works.

> One last frustration: Linux does hang, at least to the point that 
> this newbie can't recover from it. My 
> box hung when I tried to open a web presentation by Guido van Rossum
> about Python Conference 10 
> using K Presenter.  I could not even open a console nor close the
> offending KDE shell to get back to a command line. 

Using CTRL-ALT-F1 or CTRL-ALT-F2 didn't get you a console?  Sounds
bad.

> The box also hangs when I try to print using cups on my HP
> T45 printer. (I wish I could use my printer.) My downloaded Open 
> Office also hangs my computer just by trying to open it.

This has to be the worst Linux-related hang problem I've ever heard
of; you should be posting to a Linux group about this.

> Nonetheless, these hang-ups and crashes are not of pivotal importance 
> ot me. But mysterious behavior 
> that not does not allow me to run daily scripts to update my web site
> does bother me. Computers 
> should not be mysterious. They should be deterministic. Do A, then B,
> then C, get X for output.

Add the logging steps I've outlined and see what happens.

> Help ;-))))

Good luck.  You can mail me off-list if you need additional help.





From eppstein at ics.uci.edu  Wed Feb 27 23:30:27 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 27 Feb 2002 20:30:27 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <a5k7ej$d5a$1@animus.fel.iae.nl>
Message-ID: <eppstein-F06C24.20302727022002@news.service.uci.edu>

In article <a5k7ej$d5a$1 at animus.fel.iae.nl>,
 Carel Fellinger <cfelling at iae.nl> wrote:

> So please explain why you think it natural to loop over a number.

Maybe it's as natural as Perl's ability to concatenate numbers?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From pinard at iro.umontreal.ca  Fri Feb 15 14:23:02 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 15 Feb 2002 14:23:02 -0500
Subject: Using Python to generate MIDI files (Mac)
In-Reply-To: <a4jh2l$jid$1@thorium.cix.co.uk>
References: <5e8bd451.0202150813.20720fc5@posting.google.com>
	<a4jh2l$jid$1@thorium.cix.co.uk>
Message-ID: <oqy9hu35yh.fsf@carouge.sram.qc.ca>

[gbreed at cix.compulink.co.uk]

> > MIDI (though in general total crap for some musical application (like
> > anything not equal-tempered, which pretty much kills anything i am
> > doing)

One of the synthesisers I have allow the setting of an overall unequal
temperament over the 12 notes (with a resolution of 5 cents if I remember
correctly, which is a bit gross for some temperaments), and I presume
many synthesisers would do something similar.  Once a temperament pushed
in the synthesiser, you can use MIDI nevertheless, to experiment a bit...

> > (ironic isn't it?  MIDI was made for music! [...]

This is pretty debatable.  It depends on how you choose to define "music":
either as the expression of an intent, or the expression of a rendering.
MIDI has virtues on both sides, but is also (severely) lacking on both.

> Now you have a basic library, perhaps I could assist you on improving it.
> Contact me privately if this is getting to specialist.

Count me in! :-) I would also like to have nicer MIDI support from Python.

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




From colinmeeks at rogers.com  Wed Feb 20 04:43:43 2002
From: colinmeeks at rogers.com (Colin Meeks)
Date: Wed, 20 Feb 2002 09:43:43 GMT
Subject: COM file references
References: <a4vq9q$1jsn$1@mimas.salford.ac.uk>
Message-ID: <jdKc8.1337$EYv.1260@news1.bloor.is>

Hi Matt

Please pardon me if I do a spelling mistake, but it's gone 4:30am and I need
to sleep :-)
You only need to pass the base file name, I.e. the bit without the
extension. You would
also make the program a bit more intelligent and check for trailing slashes
in the
paths and double them up where necessary.  This should get you on the right
track though

regards

Colin

P.S.  Sorry if Outlook strews up the formatting.

def ConvertToText(UseFileName, UseInPath, UseOutPath):
    try:
        import win32com.client
        w=win32com.client.Dispatch("Word.Application")
        w.Visible=1
        w.Documents.Open(UseInPath+UseFileName+".doc")
        #save file as text file - number '5' is position of word to text
converter
        #in the word 'file type' dialogue
        w.ActiveDocument.SaveAs (UseOutPath+UseFileName+".txt", 5)
        w.ActiveDocument.Close(SaveChanges=0)
        return(1)
    except:
        return(0)



>
> However, I really need to turn this into a function to which I can pass an
> input file name and path, and an output file name and path.
>
> This is troubling me - however I try, I can't work out how to pass the
path
> and file to the 'Documents.Open' expression (or vice-versa)
>
> If anyone could offer any guidance I would really appreciate it. -
Apologies
> for the astonishingly simple nature of the query!
>
> cheers
>
> Matt
>
>


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.324 / Virus Database: 181 - Release Date: 14/02/2002




From martin at v.loewis.de  Wed Feb 20 17:37:56 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 23:37:56 +0100
Subject: How do you conver output of "popen('ls')" to a list?
References: <3C737245.BC5D7164@macrozon.net>
Message-ID: <m3u1sbixtn.fsf@mira.informatik.hu-berlin.de>

husam <husam at macrozon.net> writes:

> thanks for reply.

readlines()

Martin


From joost_jacob at hotmail.com  Sun Feb 10 10:07:01 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 10 Feb 2002 07:07:01 -0800
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com> <Xns91B074EB2FA5ahmebahhotmailcom@198.99.146.10> <Xgj98.24854$l93.4542796@newsb.telia.net>
Message-ID: <13285ea2.0202100707.5fe84495@posting.google.com>

[Fredrik Lundh]
> http://effbot.org/efflib/console/index.htm
> 
> about a 1000 lines, but that includes quite a few bells and
> whistles.
> 
> the api is relatively clean; writing a curses implementation
> should be trivial (did most of that once, but lost the code).

The link you give has a link to a "downloads page" and that one gives
me a 404.
I did download one of the "old" version mentioned to try it and am
trying to use it to write a portable textui module; that textui module
should use curses when under unix/linux and could use your console
module when under Windows.  Not sure about Mac.  I hope you keep
working on the projects mentioned on your site, they look great.
What we really want is something portable in the Python distribution
itself and i think many more people would love it.  Especially for
programming projects in schools.  But i am not sure how to ask for it,
maybe at the sourceforge Python site under "Feature Requests"? 
Portability is very important; students should be able to develop
things at home at their Mac or Windows pc and demonstrate it using
unix at the university.  We think portability is one of the best
things about Python and makes it very suitable for small programming
assignments in education.


From cokefiend66 at hotmail.com  Wed Feb 13 01:44:04 2002
From: cokefiend66 at hotmail.com (Bob)
Date: 12 Feb 2002 22:44:04 -0800
Subject: Error creating Blt Vector?
Message-ID: <8efc7149.0202122244.5fd404ee@posting.google.com>

Hi,
I'm using a BLT Vector in a program I wrote and it was working fine
until tonight.  I made some changes in my code but nothing near where
I create the vector.  Now when I try to create a vector I get this
error on the console:

Exception TclError: 'invalid command name "::blt::vector"' in <bound
method Vector.__del__ of []> ignored

and this error in the pop up window:

Error: 1
TclError Exception in Tk callback
  Function: <bound method Top.canvasRClicked of <__main__.Top instance
at 0x00906998>> (type: <type 'instance method'>)
  Args: (<Tkinter.Event instance at 0x009D67C0>,)
  Event type: ButtonPress (type num: 4)
Traceback (innermost last):
  File "C:\Python22\lib\site-packages\Pmw\Pmw_0_8_5\lib\PmwBase.py",
line 1690, in __call__
    return apply(self.func, args)
  File "net.py", line 126, in canvasRClicked
    newArc = Arc( self.canvas, self.sim )
  File "net.py", line 479, in __init__
    self.history = Pmw.Blt.Vector()
  File "C:\Python22\Lib\site-packages\Pmw\Pmw_0_8_5\lib\PmwBlt.py",
line 99, in __init__
    self.tk.call(_vectorCommand, 'create', self._name)
TclError: invalid command name "::blt::vector"

================================================
  Event contents:
    char: ??
    delta: 3
    focus: 0
    height: 0
    keycode: 3
    keysym: ??
    keysym_num: 3
    num: 3
    send_event: 0
    serial: 128
    state: 8
    time: 1785684158
    type: 4
    widget: .10320848
    width: 0
    x: 274
    x_root: 410
    y: 194
    y_root: 426

Can anyone help?
Thanks,
-Bob


From tim.one at home.com  Sun Feb 10 19:21:22 2002
From: tim.one at home.com (Tim Peters)
Date: Sun, 10 Feb 2002 19:21:22 -0500
Subject: No explanation for weird behavior in re module!
In-Reply-To: <a470re$1cl30l$1@ID-78052.news.dfncis.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEAMNLAA.tim.one@home.com>

[synthespian]
> >>> import re
> >>> p = re.compile('^(der|die|das(\s\w+))')
> >>> m = p.match('die T?r, T?ren')
> >>> n = p.match('das Auto, Autos')
> >>> m.group(0)
> 'die'
> >>> m.group(1)
> 'die'
> >>> m.group(2)
> [nothing!!!!]
> >>> n.group(0)
> 'das Auto'
> >>> n.group(1)
> 'das Auto'
> >>> n.group(2)
> 'Auto'
>
> 	I'm using Python2.0 on a Debian potato system.
> 	Why didn't m.group(2) produce 'T?r' as the output???

Because 'die' is all the entire expression matched.  I think you're probably
misunderstanding how regexps group:  a|b|cd acts like (a)|(b)|(cd), not like
(a|b|c)d.  Try this instead:

p = re.compile('^((?:der|die|das)(\s\w+))')

You won't get None anymore, but whether or not \w+ matches more than T may
depend on your locale setting.

> 	Python2.0 is supposed to have Unicode support buil-in the
> re module right?

Yes, but these are no Unicode strings in your program.

> 	Other than the fact that 'T?r' has the '?' unicode
> charcater, I fail to see any difference!

Heh.  Leaving this joy to someone else <wink>.




From westernsam at hotmail.com  Wed Feb 27 13:07:21 2002
From: westernsam at hotmail.com (sam)
Date: 27 Feb 2002 10:07:21 -0800
Subject: recursive problem
Message-ID: <292c8da4.0202271007.1b3af9ab@posting.google.com>

A litte while ago somebody on this group asked about a problem at this
site:

http://www.itasoftware.com/careers/programmers.php

I've been looking at the "mystery M function" question, just for fun,
I'm not going for a job there (honest). Below is the code of where
i've got to but it blows becuase it goes too deep into recursion I
think (for i,j,k = 4,4,4 anyway). What I'm doing here is storing in a
Dictionary the result of m where I get one, so I can use it later. As
I said it still blows. Any clues for how to improve my algorithm
appriciated.


Python code:


import sys
import string

dict = {}
count = 0

def m(i, j, k):
    print "i, j, k", i, j, k
    returne = 0
    key = "%i_%i_%i" % (i, j, k)
    if (dict.has_key(key)):
        returne = dict[key]
        global count
        count = count + 1
        print count, " USING DICT"
    else:
        if (i == 0): returne = 1+k
        elif (i ==1 and k == 0):returne = j
        elif (i == 2 and k == 0):returne = 0
        elif (k == 0):returne = 1
        else: returne = m(i-1, j, m(i, j, (k-1)))
        dict[key] = returne
    return returne        
        

if __name__=='__main__':
    print "Start ..."
    print m(4, 4, 4)


From lbates at syscononline.com  Sat Feb  9 14:25:19 2002
From: lbates at syscononline.com (Larry Bates)
Date: Sat, 09 Feb 2002 19:25:19 GMT
Subject: Python libTIFF wrapper
Message-ID: <zIe98.35309$B21.8411830@news1.rdc1.fl.home.com>

I have searched high and low and am still unable to find an existing
interface between Python and Leffler's libTIFF routines.  I need to be able
to read G4 TIFF files and Python Imaging Library still doesn't have that
decoder.  Can anyone point me in the right direction?

Thanks in advance,
Larry Bates





From aahz at panix.com  Fri Feb  1 13:01:31 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 10:01:31 -0800
Subject: isinstance() considered harmful
References: <mailman.1011915337.20073.python-list@python.org>
Message-ID: <a3el5r$2je$1@panix2.panix.com>

In article <mailman.1011915337.20073.python-list at python.org>,
 <kosh at aesaeion.com> wrote:
>
>Well try except is slower then if else and rendering pages to the screen
>does have a time limit to it so speed is a pretty importance factor. 

Short version of Alex's post: try/except is guaranteed faster than
if/else if an exception is *not* raised.  If you expect that the try
clause will almost always not raise an exception (or if speed is
non-critical when an exception results), try/except is better for speed.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From kauphlyn at speakeasy.org  Mon Feb 25 17:02:52 2002
From: kauphlyn at speakeasy.org (Daniel Coughlin)
Date: Mon, 25 Feb 2002 14:02:52 -0800 (PST)
Subject: [Tutor] Re: Is it possible in Python too?
In-Reply-To: <001e01c1be36$2f740a90$08020196@Jonah>
Message-ID: <Pine.LNX.4.44.0202251337140.21176-100000@grace.speakeasy.net>


Hey, iirc, the proxy server setting is set in the registry, and you cant set 
registry items from an internet explorer object. 

so you need to change the registry - heres the link from aspn telling how to do 
just such a thing:
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/66011

the entry you will be needing to modify is
HKEY_CURRENT_USER/Software/Microsoft/Windows/CurrentVersion/Internet Settings

or something like that. there maybe other things you need to modify, but i 
think that should put you in the right direction. 

test this and test a bunch before deploying into anykind of production 
environment. the registry is a very sensitive place. 

OT: A bit of netiquettte: You dont need to send emails to so many lists. Most of 
the people on any one of these lists are prally on at least one other. If you 
cant get your question answered on one list. Then try another. No need to send so many 
emails. 



Daniel

 





On Tue, 26 Feb 2002, Matthew Sherborne wrote:

> Yes but I don't know how.
> 
> Tell me how to do it in another language, and I'll tell you how in python ;)
> 
> GBU
> Matthew
> ----- Original Message -----
> From: "A" <printers at sendme.cz>
> To: <python-list at python.org>; <tutor at python.org>;
> <activepython at listserv.ActiveState.com>; <python-help at python.org>
> Sent: Tuesday, February 26, 2002 4:51 AM
> Subject: Is it possible in Python too?
> 
> 
> >
> > Hi,
> > Is possible to set up a proxy name (address) in MS Explorer
> > browser from a running  Python program?
> > Thank you for your help.
> > Ladislav
> >
> >
> >
> > _______________________________________________
> > ActivePython mailing list
> > ActivePython at listserv.ActiveState.com
> > To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
> >
> 
> 
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> http://mail.python.org/mailman/listinfo/tutor
> 




From sholden at holdenweb.com  Fri Feb  8 09:16:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 09:16:04 -0500
Subject: Why it does NOT work on Linux ?
References: <mailman.1012871110.28776.python-list@python.org> <Vsv88.94274$KM2.3404324@atlpnn01.usenetserver.com> <eEL88.24195$l93.4429576@newsb.telia.net>
Message-ID: <u4R88.105748$KM2.4031421@atlpnn01.usenetserver.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:eEL88.24195$l93.4429576 at newsb.telia.net...
> Steve Holden wrote:
>
> > > Ladislav's use of findall works, but this feature of re.findall()
> > > is not documented and should be avoided, as undocumented features
> > > have been known to disappear.
> >
> > Nonsense. The function interface continues to be a fully-documented part
of
> > the re module.
>
> that's not what Jason was talking about:  he pointed out that the
> original code passed a compiled pattern object, not a string, as the
> first argument to re.findall.
>
> > You are apparently confusing
> >
> >     http://www.python.org/doc/current/lib/node99.html
> >
> > (the re module contents description, which describes the functional
> > interface to regular expressions) with
> >
> >     http://www.python.org/doc/current/lib/re-objects.html
> >
> > (the method descriptions for re objects). It may not be the way *you* do
it,
> > but that doesn't mean it's not documented <wink>.
>
> where does the documentation say that the "pattern" argument to
> re.findall can be either a string or a compiled pattern?
>
And I suppose you want me to listen to you just because you wrote the damned
code, right? Jason, thank you for your tolerance.

me-and-my-size-9-mouth-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From krasna.halopti at virgin.net  Wed Feb 13 10:52:36 2002
From: krasna.halopti at virgin.net (Krasna Halopti)
Date: 13 Feb 2002 07:52:36 -0800
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com>
Message-ID: <a54c54d5.0202130752.356371d5@posting.google.com>

You can also check out

http://www.dma.org/~tw/pyboids/

Best of luck,

KH


From aleax at aleax.it  Fri Feb  8 11:08:06 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 08 Feb 2002 16:08:06 GMT
Subject: FixedPoint
References: <mailman.1013119093.20033.python-list@python.org>
Message-ID: <GJS88.49612$3J1.1396553@news2.tin.it>

Mark McEahern wrote:
        ...
>> Hmm., while we're at it, we should probably change this test too: "in
>> [ListType, TupleType]" Does anybody know how I can reliably test if
>> something is a Sequence or is iterable?
> 
> I'm just thinking out loud here...  How about something like this:
> 
> if hasattr(value, '__getitem__'):
> ...
> 
> This may only work in 2.2:
> 
> if hasattr(value, '__iter__'):
> ...
> 
> I don't really have a good grasp of all the subtleties involved in testing
> for sequence in a way that works across lots of different versions of
> Python.

"something is iterable" if and only if you can iterate on it.  So, try 
iterating on it, and handle the resulting exception should it turn out
you can't.  I'm not sure exactly what exceptions would be raised by
various versions of Python -- probably a TypeError, but one never
knows -- better to experiment on all relevant versions, or maybe, just
for once, a generic try/except is acceptable.  An almost acceptable
version might therefore be:

def isIterable(x):
    try:
        for i in x: return 1   # iterable
        else: return 1         # iterable, although empty
    except:
        return 0

Unfortunately, if we're in Python 2.2 and x is an iterator, isIterable
"consumes" x's first item.  This isn't hard to fix:

def isIterable(x):
    try: x=iter(x)
    except: pass
    else: return 1
    try:
        for i in x: return 1   # iterable
        else: return 1         # iterable, although empty
    except:
        return 0

This assumes function iter, if defined, is "sensible" -- it returns an
iterator, or else raises an exception.  Builtin function iter does meet
this requirement (it raises a TypeError if __iter__ is defined but
returns a non-iterator), and (as usual in Python) it seems sensible
to assume that if somebody has replaced a builtin function then that 
somebody is responsible for the overall results still being sensible.

Of course, this does return different results for similar objects in 
various Python releases -- as it should: a file or dict is iterable in 
Python 2.2, but not in 2.1 or earlier, after all.

As noticed in a following post, "is this iterable" or "is this a sequence" 
may or may not be what you really want to know -- strings and Unicode 
objects are iterable sequences, for example, yet often we want to treat 
them as 'atoms', depending on context.  So you need to determine what 
exactly you DO want to check.  Once again, if you want to exclude string, 
UserString, Unicode objects, and so on, don't typetest -- use try/except:

def isStringLike(x):
    try: x+''
    except TypeError: return 0
    else: return 1

You may want to use different definitions for "is string-like", but 
personally I've found that "lets an empty string be concatenated to it" is 
a pragmatically useful definition.


Alex



From paul at boddie.net  Fri Feb 15 06:26:53 2002
From: paul at boddie.net (Paul Boddie)
Date: 15 Feb 2002 03:26:53 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <b16e4ef7.0202141211.3fb6b1f@posting.google.com>
Message-ID: <23891c90.0202150326.5bdde2b4@posting.google.com>

steve at ferg.org (Stephen Ferg) wrote in message news:<b16e4ef7.0202141211.3fb6b1f at posting.google.com>...
> 
> Kevin pinpointed exactly what is different about EasyGui ... it is
> just putting up dialogs for simple input and output in a linear
> program.
> 
> What makes EasyGui different from other GUIs is that it is NOT
> event-driven.

[...]

> I honestly don't know if there is a need for such an animal as
> EasyGui.  But I think there might be...

There is a class of applications or tools which provide user
interfaces in a similar way to EasyGui. For example, the Red Hat Linux
text-based installation and administration tools lead users through
sequences where users are prompted for information through text boxes,
lists and dialogues. It seems to me that by implementing other display
mechanisms, you would be able to cover such applications, provide
alternative display options for them, and be some kind of
"anygui-light".

Paul


From alwyns at prism.co.za  Sat Feb  2 14:31:59 2002
From: alwyns at prism.co.za (Alwyn Schoeman)
Date: Sat, 2 Feb 2002 21:31:59 +0200
Subject: Signal and thread problems on Solaris revisited -HELP!!
Message-ID: <a3h814$g5v$1@ctb-nnrp1.saix.net>

Hi there,

Awhile back I posted a message about having problems with signal handling 
in a multithreaded application running on Solaris.  I'd like to follow up 
on that problem.

I've seen on the sourceforge forum's that the bug is supposedly fixed in 
Python 2.2.  I've thus upgraded a Solaris 8 and Solaris 7 machine to Python 
2.2.  Both were compiled with --with-signal-module and --with-threads.

The differences between the 2 systems are the following:
1)  The one machine is Solaris 8 and the other Solaris 7.
2) The Solaris 8 Python 2.2 was compiled with gcc.
3) The Solaris 7 Python 2.2 was compiled with Sun Forte C++ 6.

I have 2 versions of my program.  The one uses no implicit signal handling 
but only except KeyboardInterrupt, the other uses signal handling.

On Solaris 7 machine the signal handling work perfect for both versions.  
On Solaris 8 none of them work.

On request this is a pseudo description of how the program works.

The system opens 2 socket connections to 2 different external services.  
Each of these 2 connections has a reader and writer thread associated with 
it that consumes and produces data asynchronously.  Each connection also 
has a management thread associated with it which is responsable for 
managing the connection state among other things.  One of these connections 
also has a number of processor threads responsable for processing data. 
Queues are used for data access synchronization.   Condition objects are 
used to send messages between threads.  Except for this there is also the 
main thread.

I'm not going to go into the KeyboardInterrupt version as that 
implementation is obvious and only differs in the main program/thread.

Main program/thread (unnecessary info excluded):
import signal

exitFlag = 0

def handler(signum,frame):
        global exitFlag
        if signum == 2:
                exitFlag = 1

def main():
        Create Manager1 thread and start.
        Create Manager2 thread and start
        while not exitFlag:
                signal.pause()
        Call setExitFlag method on Manager1.
        Join on Manager1
        Call setExitFlag method on Manager2
        Join on Manager2

Above is the main program.  The manager threads work in a similar fashion.
They initialize the threads.  Enter a processing loop or idling loop, which 
exits on a private exitflag that is set in the setExitFlag method of the 
manager thread.  The manager threads then call the setExitFlag methods of 
all created workerthreads which in turn causes a private exitflag to be set 
in each thread, causing the threads to exit their respective processing 
loops.  These threads then join in the respective Manager thread.  Lastly 
the Manager threads join in the main thread.

E.g. (Still pseudo pseudo)
class Manager(threading.Thread):
        def __init__(self,etc):
                set private exitFlag = 0
                set other stuff
                Call threading.Thread constructor

        def setExitFlag(self):
                self.__exitFlag = 1

        def run(self):
                create and start worker threads
                while not self.__exitFlag:
                        do something to pass the time.  
                call setExitFlag method on all worker threads.
                join on each thread.

Above is the basic Manager thread.

Worker threads are similar except they do not create of join other threads.

If you need more information just let me know.

On Solaris 8 I just see ^C on the console but the signal gets ignored. On 7 
it works find...

I can send the output of truss in both cases, I just do not have fast 
enough access now (One machine is in Manila, other in Johannesburg).

Jason Lowe's code was put into thread_pthread I think.  I'm just not sure 
if perhaps thread_solaris isn't used. (Shot in the dark).

Just let me know what you need....

Regards,
Alwyn Schoeman


From shalehperry at attbi.com  Tue Feb 26 12:49:59 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Tue, 26 Feb 2002 09:49:59 -0800 (PST)
Subject: Precompiled regular expressions slower?
In-Reply-To: <20020226170751.GB28508@andy.tynant.ftech.net>
Message-ID: <XFMail.20020226094959.shalehperry@attbi.com>

> 
> Possible mechanism for what's happening:
> 
>     - Setting up a compiled re object takes time X
> 
>     - Executing one search using compiled re object takes time x
> 
>     - Thus, n searches using compiled object takes X + (n * x)
> 
>     - Executing search using re.search() takes time y
> 
>     - "Setup" is included in that time (basically consists of
>       following reference to re.search())
> 
>     - Thus, n searches using re.search() takes n * y
> 

Actually, it is more odd.  sre.py does compile the regex and stores it
in an internal dictionary if you just do re.search(r'blah', foo).  So either
way the regex is getting compiled.  Compiling the regex and storing it in a
local variable should save you a function call or two and a dictionary lookup.

Perhaps Peter should post the relevant code so we can see what is actually
happening for ourselves?



From brian at sweetapp.com  Fri Feb  8 00:47:57 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Thu, 7 Feb 2002 21:47:57 -0800
Subject: FixedPoint
In-Reply-To: <LNBBLJKPBEHFEDALKOLCGEGKNKAA.tim.one@home.com>
Message-ID: <000601c1b064$29d3a940$445d4540@Dell2>

Tim Peters wrote:
> Note that I said "non-numeric object".  FixedPoint supports a rich set
of
> auto-coercions from types that "make sense".  I'm not a fan of
> semantically senseless comparisons, e.g.
> 
> >>> 5 < '5'
> 1
> >>>

Even if you disagree with that example, I would argue that equivalence
is a special case. If you cannot convert a => b in a meaningful way then
a != b.

Cheers,
Brian




From martin at v.loewis.de  Thu Feb 14 19:37:41 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 15 Feb 2002 01:37:41 +0100
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de> <m3lmdyw6r3.fsf@mira.informatik.hu-berlin.de> <a4gd73$2dp9$1@ID-98166.news.dfncis.de>
Message-ID: <m3664zwpey.fsf@mira.informatik.hu-berlin.de>

"MDK" <mdk at mdk.com> writes:

> > One possible explanation is that you have an additional Python
> > installation whose _tkinter.pyd is linked with an older Tcl version,
> > or that you merely have an older _tkinter.pyd on your system.
[...]
> How do I find out if this is the case and what do I do about it?

Search all your hard drives for files starting with either python,
tcl, tk, or _tkinter, and ending in either .dll, or .pyd. Then remove
those which don't belong to a current installation.

Regards,
Martin



From dgallion1 at yahoo.com  Thu Feb 21 23:20:40 2002
From: dgallion1 at yahoo.com (Darrell)
Date: Fri, 22 Feb 2002 04:20:40 GMT
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1> <7xd6yyfdep.fsf@ruckus.brouhaha.com>
Message-ID: <sGjd8.111580$dh.31131176@typhoon.nyroc.rr.com>

Working with unsigned is a pain.

>>> a=0x80000000
>>> hex(a>>32)
'0xffffffff'

>>> int('0x80000000',16)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ValueError: int() literal too large: 0x80000000
>>> eval('0x80000000')
-2147483648
>>>

And eval doesn't need to be told the number base in advance.
Other than the 0x... that is.

Longs help.

>>> b=0x80000000L
>>> b
2147483648L
>>> hex(b)
'0x80000000L'
>>> hex(b>>32)
'0x0L'
>>> hex(b>>31)
'0x1L'
>>>

But there are problems with this also.

>>> struct.pack("i",b)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OverflowError: long int too large to convert to int
>>> struct.pack("l",b)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OverflowError: long int too large to convert to int
>>> struct.pack("l",int(b))
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
OverflowError: long int too large to convert to int

You can play games to get around this.

>>> bs=b>>1
>>> bs=bs<<1
>>> struct.pack("i",bs)
'\x00\x00\x00\x80'
>>>

--Darrell


Paul Rubin wrote:
> 
> Am I missing something?  Python has the usual bit operations that work
> on ints.
> 
> a = do_some_stuff(byte >> 4)        # upper nibble
> b = do_similar_stuff(byte & 0x0f)   # lower nibble
> new_byte = (a << 4) + b     # assume a and b are nibble results




From tim.one at comcast.net  Sat Feb 23 10:10:28 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sat, 23 Feb 2002 10:10:28 -0500
Subject: HELP! Obscure Python Reentrancy Question
In-Reply-To: <Pine.LNX.4.44.0202230904310.23509-100000@mercury.sabren.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEMHNOAA.tim.one@comcast.net>

[Courageous]
> I didn't think that Python was structured to allow
> multiple interpreters in a single process?

[Michal Wallace]
> I know python can do this because mod_python uses this
> feature.

Yes, and you can read all about it in the Python/C API Reference Manual,
under Py_NewInterpreter().  Note that it's only available at the C level; no
analogue of this function is exposed at the Python level.

Michal, from the rest of your msg, it sounds like you'd profit more from
reading the "Thread State and the Global Interpreter Lock" section of that
manual.  For straightforward cases it's straightforward <wink>.  For nasty
callbacks it can be a nightmare.

A miserable example in the core is _PyPclose() in posixmodule.c, where a
thread known not to be holding the global interpreter lock needs to obtain
the GIL to call back into the Python runtime.  Normally this is easy:  you
just acquire the GIL.  But to acquire the GIL you first need a valid thread
state, and in this case the thread doesn't have one.  But to create a valid
thread state, you first need a valid interpreter state (this is where
multiple interpreters running in a single process intersects), and in this
case the thread doesn't have one of those either.  Etc.

It took hours to figure out how to make this work, and I had the slight
advantage of knowing what I was doing <wink>.  It's worse if you don't know
whether or not the thread holds the GIL, and worse still if you don't even
know whether Python has been initialized yet.  Mark Hammond's Windows
extensions have to deal with all those cases (_PyPclose is as bad as it gets
in the core), so that's where to look if the only alternative is suicide.




From jason at jorendorff.com  Mon Feb  4 21:39:03 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 20:39:03 -0600
Subject: Backwards Compatibility of Python versions
In-Reply-To: <7x8za8dgle.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEGODAAA.jason@jorendorff.com>

Paul Rubin writes:
> I see publishing Python programs as like publishing web pages.  If I
> have to tell 2% of my users to upgrade those ridiculous old browsers
> to view my pages, I don't mind.  If I have to tell 25% of them to
> upgrade, I'm causing hassles for a significant population and I
> shouldn't do that without a good reason.  If I have to tell 80% of
> them to upgrade, I'm probably going to lose a lot of them.

My rule of thumb is:  if you tell X% of them to upgrade,
you'll lose about X% of them.

This applies to Python, too.  If you tell users "you need to
install Python", you lose the vast majority of potential users
before you even start talking about versions.

## Jason Orendorff    http://www.jorendorff.com/



From quinn at hurl.ugcs.caltech.edu  Sun Feb 10 17:12:57 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 22:12:57 GMT
Subject: [NEWBIE] List issue aka: "Why doesn't this code work?"
References: <mailman.1013184014.10227.python-list@python.org>
Message-ID: <slrna6ds39.dm.quinn@hurl.ugcs.caltech.edu>

On Fri, 8 Feb 2002 10:51:55 -0500, Bruce Dykes <bkd at graphnet.com> wrote:
>Okay, here's my function:
>
>def process_numbers(source_list):
>    for number in range(len(source_list)):
>        entry = string.split(string.strip(source_list[number]))
>        print number, len(entry), entry
>
>Now this function is doing *something*...source_list is a list of lines
>retrieved via ftp.retrlines(), and this should break each line into it's own
>list, and it looks like that's what's being printed out:
>
>['line1', 'header1', 'header2']
>['line2', 'datum1', 'datum2']
>
>Notice what's missing? I'm expecting to see:
>
>0 3 ['line1', 'header1', 'header2']
>1 3 ['line2', 'datum1', 'datum2']
>
>What the heck am I doing wrong here?

Forgetting to reload the module?

It works for me.

>>> process_numbers(['line1 header1 header2', 'line2 datum1 datum2'])
0 3 ['line1', 'header1', 'header2']
1 3 ['line2', 'datum1', 'datum2']




From rdacker at pacbell.net  Mon Feb  4 20:28:51 2002
From: rdacker at pacbell.net (bob ackerman)
Date: Mon, 04 Feb 2002 17:28:51 -0800
Subject: [OSX] milter won't build
In-Reply-To:  <Pine.OSX.4.43.0202041601140.415-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <B61483FF-19D7-11D6-BA4F-003065428126@pacbell.net>

that's what i needed to know. libsm.a was in the original sendmail folder,
  but not installed. i moved it to /usr/lib and i was able to build and 
install.
Thank you!

On Monday, February 4, 2002, at 01:10  PM, Steven Majewski wrote:

>
>
> On 4 Feb 2002, rdack wrote:
>
>> now i get:
>> cc -bundle -flat_namespace -undefined suppress
>> build/temp.darwin-5.2-Power Macintosh-2.1/miltermodule.o -lmilter -lsm
>> -o build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
>> /usr/bin/ld: can't locate file for: -lsm
>>
>> what file is it looking for? why can't it find it? what is option
>> '-lsm'? could i just remove it? what file would it be in?
>
> "-lsm"  means it wants to link to  libsm.[ a, so, ... ].
> Without a "-L" switch to add other library search directories, the
> default is to look in /usr/lib.
>
> I have no idea what "libsm" is -- that's not part of Darwin/Mac OS X.
> Some common libs, like libm for example, are merged into the system
> library, so you can just leave off "-lm" and it'll link to the
> math libs without it -- so it's worth trying if you don't know
> what it is or where it comes from.
>
> -- Steve Majewski
>
>




From peter at engcorp.com  Thu Feb  7 00:56:29 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 07 Feb 2002 00:56:29 -0500
Subject: Delphi vs Python
References: <zBm88.333397$pa1.80603034@news3.rdc1.on.home.com>
Message-ID: <3C62170D.5095436D@engcorp.com>

Bo Vandenberg wrote:
> 
> I'm planning learning strategy and I thought I would ask people who use both
> Python and Delphi their opinion.

I used Delphi (back in the 3 and 4 days, I think) extensively for several 
years, and had a lot of respect for the environment and language.  Very
productive for me, relatively few bugs (compared to C++ and C, roughly
similar to my levels with Java).

Python is, for me, a hands down superior language.  The projects on which
we used Delphi could easily have been implemented in Python, and I'm
now convinced it would have taken significantly less time to write and
debug, and the results would have been on par in terms of performance,
and the resulting product would have been *much* more readable and as
a result maintainable.

I wouldn't suggest anyone _not_ look at Delphi, but I'd certainly 
recommend starting with Python, since I believe in many cases one
would not bother moving away from it afterwards, except for very
special cases.

> I've determined that I at least cannot speedily learn both at once. So my
> current strategy is to get up to speed in Python first and then maybe do the
> slowest bits of code in Delphi.

You may be surprised with the performance of Python.  I still have not 
found the need to recode anything in C, and we have optimized only one
routine to date (just improved the algorithm... no need to leave Python)
in over fifteen developer-years of Python development.

Most programmers spend *far* too much time thinking about performance,
and in my opinion it should not even be a consideration at this early
stage, unless you have specific and rather strenous performance 
requirements (and generally even then the various Python extension libraries 
are quite up to the task, since the key parts are often coded in C anyway).

> Has anyone seen any Delphi 6 personal edtn. stuff show up to serve python?

No, sorry, that came after I stopped using it.

-Peter


From codetube at yahoo.com  Wed Feb  6 19:34:09 2002
From: codetube at yahoo.com (codetube)
Date: Thu, 07 Feb 2002 00:34:09 -0000
Subject: How to set an active view in Python?
Message-ID: <a3si21+c85i@eGroups.com>

Hello everyone,

I have subclassed code.py and turned that file into a COM server. In 
that file I have called some methods of the code.py for testing. I 
get a message saying "win32ui: There is no active view".
Can anyone please advice me on how to set up an active view so that I 
can emulate the python console etc.

Best Regards,
CodeTube




From afp_randjohnson at yahoo.com  Sun Feb 24 10:03:23 2002
From: afp_randjohnson at yahoo.com (Mike Johnson)
Date: 24 Feb 2002 07:03:23 -0800
Subject: Retrieve a variable from an embedded interpreter
References: <zo1zjioe.fsf@fido.trinitycapital.com> <sn7r2hu1.fsf@fido.trinitycapital.com>
Message-ID: <adty3osk.fsf@fido.trinitycapital.com>

Hello all (who apparently don't read on the weekends):

I've got something that works now. I would just like to say the faq on
the subject is completely not useful. The section, "How do I extract C
values from a Python object?" doesn't address how to get an arbitrary
variable from Python.

Specifically, this guy's comments should be bronzed somewhere:

http://groups.google.com/groups?q=python+set+variable+c&start=30&hl=en&selm=m0wNSjb-001CneC%40shell.rmi.net&rnum=32

I'd like to add a faq item, just after 5.7, to compile together the
concepts addressed in the preceding sections.

---

In order to share values between C and python, it is useful to create
a module. From C one can add methods and variables to the module, and
in Python one can access and change these variables.

In this example, the module's name is "cpserve". It's a simple text
file that includes these statements:
----
#!/usr/bin/python2.1
version = .1
----

>From C, we're going to add a property "a". First, import the module in
C, using Python's API access:
----
PyObject *a;
char *sa;
PyObject *pdict;
PyObject *pymod;

Py_Initialize ();

pdict = PyDict_New();
PyDict_SetItemString( pdict, "__builtins__", PyEval_GetBuiltins() );
pymod = PyImport_ImportModule ( "cpserve" );
----

Next, we'll add the property "a" and set it's value to the string
"hello":
----
a = PyString_FromString ( "hello" );
PyObject_SetAttrString ( pymod, "a", a );
----

Now we're ready to execute the Python code. Here I use
PyRun_SimpleString() but these calls could easily be placed in a file
and executed that way.
----
PyRun_SimpleString ( "import cpserve" );
PyRun_SimpleString ( "print\"a=\" + cpserve.a" );
PyRun_SimpleString ( "cpserve.a = \"hi\"" );
PyRun_SimpleString ( "print \"a=\" + str(cpserve.version)" );

Finally, we can retrieve the value of "cpserve.a" with a C variable:
----
a = PyObject_GetAttrString ( pymod, "a" );
PyArg_Parse ( a, "s", &sa);
printf ( "sa = %s\n", sa );

Py_DECREF ( pdict );
Py_DECREF ( pymod );
Py_DECREF ( a );
Py_Finalize ();
----

When compiled and run, this produces the output:
----
a=hello a=0.1 sa = hi
----

I'm sure somebody will find something wrong with this. Suggestions
are welcome. I'd just like to see some more documentation on
this sort of thing.

And I'm willing to document as I learn. :-)

Thanks,

Mike Johnson



From kdahlhaus at yahoo.com  Wed Feb  6 08:03:34 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 6 Feb 2002 05:03:34 -0800
Subject: Web Hosting - one report
Message-ID: <283adf56.0202060503.6556b483@posting.google.com>

Time for the weekly where-can-I-cheaply-host-pythonic-web-app posting.

I thought I had found a place at 'Hostdepartment.com'  $5/month for
100Mb and a full shell. They have 1.6 installed.  You can compile your
own python, but many modules do not build.  It turns out that they
have disabled dynamic loading of modules from user accounts.  So if
you have a swigged extension, python can load any shadow files, but
the .so is never found.  That's a shame, as I was hoping to actually
post an *answer* to this reoccuring question.  Personally, I need to
be able to use extension modules - at least PIL, GD charts, etc.

(I was quite annoyed when after 3 days of notices that my trouble
ticket was being looked at, a message came that said 'your ticket is
completely resolved.  you cannot load dynamic modules from user
accounts'.  That's explained, not resolved.)

Anyway, so that's one report.  Has anyone found one of these
inexpensive providers that allows you to use your own native-language
extension modules?


From peter at engcorp.com  Sun Feb 24 15:02:35 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 15:02:35 -0500
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>  <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no> <3C706035.6AC11606@earthlink.net>
Message-ID: <3C7946DB.A7FE6260@engcorp.com>

Hans Nowak wrote:
> 
> Magnus Lie Hetland wrote:
> >
> > In article <8867.813T2365T9245607threeseas at earthlink.net>, Timothy Rue wrote:
> > [...]
> > >Certainly everyone does understand in reading and responding to post here,
> > >they actually make use of all nine action. It's physics, but they may
> > >think it's me and that somehow their reactions will cause a dismissal
> > >of.... well physics.
> > >
> > >Lets see now:
> > [...]
> >
> > No, no! This isn't simple enough. The *real* truth (and it's physics,
> > so you can't contradict me) is that there is only one true action:
> >
> > DS - Do Stuff
> >
> > Now deal with it and go somewhere else with your false theory.
> 
> Do you also have a model with zero actions? This one seems kind of
> difficult to me.

Yes.  Look at the way computers store floating point numbers.  The
left-most one is assumed, so never stored.  Likewise with the 
zero-instruction VIC (ZIVIC) one just assumes the DS instruction
before all parameters.  I posted a script written in ZIVIC a 
few weeks ago, using parameters drawn from running the dictionary
compiler over the Python docs to extract the basic structure of
the language.  The result was quite straight-forward and readable
to most of us here, with one possible exception...

I'll post another sample here:

# ZIVIC code: assumes DS instruction prior to all parameters
file = open('somefile.txt')
length = len(file.readlines())
print 'Lines in file:', length

As this is the first working implementation based on the VIC 
concepts, I trust this will put to rest all the negativism 
expressed here about the viability of the concept.  Clearly it
works, and I suggest we all go back to using it...

-Peter


From sholden at pgrad.cs.usyd.edu.au  Sun Feb  3 23:18:30 2002
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 4 Feb 2002 04:18:30 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1726.795T908T14258threeseas@earthlink.net> <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net>
Message-ID: <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au>

On Mon, 04 Feb 2002 00:00:49 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 30-Jan-02 23:23:31 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>
>>This is a discuss group (gnu,misc.discuss anyway). It is not a place to seek
>>coding help. If you don't want a debate don't post to a discussion forum. If
>>comp.lang.python is like every other comp.lang group I know then it is
>>also not a place to seek coding help - other than specific language problems
>>(such as how do I convert a string into a number?).
>
>>You just don't understand how programmers work. And until you do you
>>won't find many willing to help you. In fact if someone did think your ideas
>>were good they would probably just fork your code and do it themselves
>>without your involvement - since you seem to be going out of your way to turn
>> people off your project.
>
>
>Don't like the idea of autocoding huh?

Actually I have no opinion on it at all. I don't have time to get involved in
a project at the moment, and thuis I haven't tried to understand your long
explanations of what it is.

Since I don't know what it actually is, I don't know if I like or don't
like the idea.

If it is something that will help non-programmers better use the 
computer resources they have access to without having to learn how
to program, then it would actually be something I would be interested
in - since that is an area I am interested in.

If it is that old assembler type thing then I'm not intersted since
that is an area I have no interest in.

If it is to do with autocoder by Smart Software Company, then I'm also
not interested, as automating open ended questions is not something I
personally find interesting.

I only posted to try and help you better communicate with programmers, and
if your reply is anything to go by you didn;t bother reading my advice,
or didn't think it was valid. Since your reply has not generated
any interest in me (a programmer) for your project - but that really
doesn't matter, as as I said, I'm busy at the moment.


Sam Holden



From skip at pobox.com  Thu Feb 14 11:20:51 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 14 Feb 2002 10:20:51 -0600
Subject: zlib.so and libz.so and Python
In-Reply-To: <u6niisr909gtd6@corp.supernews.com>
References: <u6niisr909gtd6@corp.supernews.com>
Message-ID: <15467.58339.806161.172243@beluga.mojam.com>

    DeepBleu> Python comes with zlib.so (for Sun OS).  But I noticed that on
    DeepBleu> the same machine, there is also libz.so.  Can someone please
    DeepBleu> tell me what is the difference between these two and what
    DeepBleu> libz.so is used for?

"zlib" is the name of a Python module.  "zlib.so" is the runtime incarnation
of that module.  My guess is that libz.so is a shared library version of the
C "zlib" library.  The Python module is essentially just a wrapper around
the zlib library.  On my linux machine ldd shows that zlib.so depends on
libz.so:

    % ldd zlib.so
            libz.so.1 => /lib/libz.so.1 (0x40015000)
            libgcc_s.so.1 => /usr/local/lib/libgcc_s.so.1 (0x40024000)
            libc.so.6 => /lib/libc.so.6 (0x4002c000)
            /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)

My guess is that you will see something similar on your Solaris system.

    DeepBleu> Also, I want to find out if zlib.so (the one that comes with
    DeepBleu> the Python distribution) is all we need to work with ZIP files
    DeepBleu> in Python. 

Technically, I suppose you could, but I think you'll find it a heck of a lot
easier to use the zipfile module (which uses zlib underneath the covers) to
work with ZIP files.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From jeff at ccvcorp.com  Tue Feb 26 15:54:39 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 26 Feb 2002 12:54:39 -0800
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com> <3c7a70d2_4@news.bluewin.ch> <gherman-A86A1B.18413225022002@news.t-online.com>
Message-ID: <3C7BF60F.5C84FF88@ccvcorp.com>


Dinu Gherman wrote:


> I'm pretty surprised! I knew the Python test suite is *very* far
> from complete, but list() is an *extremely* crucial function,
> isn't it?

Not in my (admittedly somewhat limited) experience, at least not in the way that
you're using it.  I've used list() to convert strings or tuples to lists; I've
never used it to copy a list, and it wouldn't have even occurred to me that it
*should* copy a list.  (What's the result of int(5)? Why would you use it?  Same
idea.)  Looking at the docs, I suppose that it *is* documented behavior, but
it's still somewhat obscure.  If I want a copy, I'll ask for a copy (using
either copy.copy() or full-slice notation [:]); if I'm using list(), it's
because I want to convert some non-list sequence to a list.  So, I don't find it
at all surprising that nobody would have noticed this bug in Jython before.

Jeff Shannon
Technician/Programmer
Credit International




From magnus at thinkware.se  Wed Feb 20 04:50:52 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 01:50:52 -0800
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com> <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com>
Message-ID: <258fd9b8.0202200150.1b4e53de@posting.google.com>

"Keith Farmer" <kfarmer at thuban.org> wrote in message news:<x7Cc8.2126$RJ4.246925423 at newssvr14.news.prodigy.com>...
> The same behavior was seen with wxPython's PyCrust shell, so I suspect it's

Running a non-ActiveState version of Python?

> not (just) activepython, but rather the way the interpreter module's
> written.

But it doesn't happen in the cygwin Python interpreter (also 2.1.1),
or in Linux (2.1.1 again).

Actually:

>>> help
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
NameError: name 'help' is not defined

But even if I do "from pydoc import help" in cygwin or linux, I
don't get this strange behaviour. ActiveState did something to
the Python interpreter...

8<---

ActivePython changes to Python
------------------------------

This document describes changes made to the core Python sources by
ActiveState Tool Corp. for the ActivePython distribution.

...

3. Add the "help" command which calls Ping's pydoc.help

...

--
ActiveState Tool Corp.
18 April, 2001

8<---

Well, I was getting a bit disappointed about not having access
to the nice 2.2 features anyway, so maybe it's time to dump
ActivePython...


From jgardn at alumni.washington.edu  Thu Feb 28 02:43:53 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Thu, 28 Feb 2002 16:43:53 +0900
Subject: Want to monitor process..
References: <87550ef1.0202272315.15779bbd@posting.google.com>
Message-ID: <f2lf8.385$Az6.1854@news.hananet.net>

Haeyoung Kim wrote:

> I want to run or kill a process(program?) using python.

from os import *
# Start a process
...
pid = fork()
if not pid:
        # I'm a child!
        exec*(what you want to do)
# I'm a parent
...

# Kill a process:
kill(pid)

Check out the documentation on 'os' and especially section 6.1.5 in the lib 
documentation. Forget about anything that says "Windows only" or "for 
Windows". Just focus on anything that says "Unix only".

> Once it's started, I don't want to run the same process any more. (but
> can be killed or restarted, of course)

Then keep track of which files you have started. Maybe a dict with the keys 
being the files you would like to start and the values being the pid?

> At present I use a naive way to write a pid file when initiate the
> process, but in that case I can't restart it when the process has been
> down. (i.e., if that when the process has gone, I should kill the
> processes and remove pid file manually.)

Check out the module "atexit". You can set it up so that when the process 
is dying, it will call the functions. One of the functions should remove 
the pid lock file.

What you are talking about now is daemons that run in the background. Let's 
say you don't want to accidentally start two webservers. First off, if you 
tried to start two webservers, only one of them could listen to the port of 
choice. The other guy would have problems.

The pidfile is useful when you want to send a signal to THE webserver. You 
do something like:
$ kill -SIG_HUP `cat /the/pidfile`
with Bash. (man kill)

You can also test if the daemon is still running by seeing if the pidfile 
is there.

> I'm not so good at such a system programming.. :(

No one really is. You'll have to find an approach and a system that works 
for your needs. There are so many choices on Unix it is crazy. After doing 
sytems programming on Unix, though, you will feel like you are wearing a 
straightjacket and blinders when you are programming on Windows. That is, 
if you are still sane afterwards.

If you are going to do a lot of system programming, you should become 
familiar with C and the way things are done in Unix. (comp.unix.programmers 
is a good group to post this question to.)

> Is there any nice solution?

Look at how Apache handles its children and itself. You don't have to dive 
into the source, just understand the basics. I think Apache is probably one 
of the best out there. You may also want to try to understand how X windows 
works, how sendmail works, and how telnetd, inetd, sshd and stuff work to 
get a better picture of systems programming. You'll find that at least ONE 
of the programs on your system does something very similar to what you want 
to do. Trust the programs that have been around since the beginning of 
time, and don't try something revolutionary unless you really have to.

Jonathan



From mwh at python.net  Tue Feb  5 09:26:26 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 14:26:26 GMT
Subject: distutils, C extensions, Borland compiler, wxPython
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk>
Message-ID: <ulme8ovdf.fsf@python.net>

"Anders J. Munch" <andersjm at dancontrol.dk> writes:

> I have built Python from source on Win2K using cygwin except with a
> Borland compiler (BCB4) instead of gcc.  It took a fair amount of
> hacks to get the configure script and makefiles to work (s/.o/.obj/ is
> just the beginning) but it was worth it.

Which version of Python?  I thought there was at least *some* support
for this in recent versions of python.

Have you looked at google to see if people have tried this before?

> It's all statically linked.  I kinda like monolithic executables, so
> rather than get .dlls to work I have just disabled that part of the
> makefile.
> 
> It works great.  I've been extending and embedding like crazy for a
> while, using Setup.local.
> 
> There's just one hitch: distutils.  I'd like to use wxPython, but
> wxPython uses distutils with C extensions.
> 
> "python setup.py install" fails, unsurprisingly, because it doesn't
> know how to build and import the C++ code.

Try using 

$ python setup.py build --compiler=bcpp install

> I wonder if the configure/make process is supposed to store this
> information somewhere, but even if it did, it probably wouldn't
> work.  I presume distutils relies on the use of .dll/.so's which I'd
> rather not have to mess with,

Hopefully the author of ../distutils/bccpcompiler.py has done the
messing already...

> and also my python.exe is not a cygwin app. and hence doesn't
> understand the cygwin paths from the make process.

Don't really follow you here, but as I've never used the borland
compiler and try to avoid understanding windows issues...

> Now what I'd like to have is for distutils to just tell me which
> .c/.c++ files need to be compiled into python and with which options,
> so that I could add these files to Setup.local manually and rebuild
> python.exe.

This *shouldn't* be that hard; the setup.py file should contain a list
of source files that need compiling, like

setup(...
      extensions = [Extension("wx", ["wxfile1.c", "wxfile2.c"]),
        ...]
      ...)

this translates to lines in Setup.local along the lines of

wx wxfile1.c wxfile2.c 

> The distutils docs I have seen so far don't seem to get beyond
> "press the big green button and everything works automatically".  

That's the hope.

> Are there any docs on how to tell distutils which compiler to use
> and how?

There's

http://python.sourceforge.net/devel-docs/inst/inst.html

but I don't know how thorough this is -- I generally read the source
when I need to understand what distutils is up to...

> TIA for any advice.

HTH.

> the-docs-state-that-"Windows-Python-is-built-in-Microsoft-Visual-C++"-
> which-is-of-course-perfectly-true-(except-when-it-isn't)-ly y'rs, 
>  Anders
> 
> 
> PS: The wxPython setup.py error message that I get:
> ! Traceback (most recent call last):
> !   File "setup.py", line 10, in ?
> !     from my_distutils import run_swig, contrib_copy_tree
> ! File "my_distutils.py", line 125, in ?
> !  ccompiler.default_compiler['nt'] = 'my_msvc'
> ! AttributeError: 'module' object has no attribute 'default_compiler'

Oh, it looks like wxPython plays silly buggers in it's
setup.py... this may make things more complicated.  Or not.

Cheers,
M.

-- 
112. Computer Science is embarrassed by the computer.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From mark at mceahern.com  Sat Feb  9 14:38:41 2002
From: mark at mceahern.com (Mark McEahern)
Date: Sat, 9 Feb 2002 11:38:41 -0800
Subject: box on soap
Message-ID: <NCBBLFCOHHDIKCAFGCFBGENIKBAA.mark@mceahern.com>

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnwsdl/html
/boxwsdl.asp?frame=true

Don Box on the Importance of Being WSDL
Don Box
Architect, Microsoft Corporation

February 2002

Summary: A personal appeal to the Web services community to make type-safety
a priority by writing WSDL for script-based Web services. (3 printed pages)

In watching the various Internet discussion forums, it has become glaringly
obvious that the development community at large is still struggling with how
the Web services fabric will be held together. By the end of 2001, everyone
in the emerging Web service industry had converged on SOAP. Based on the
raging debates held in 1999, this is amazing progress. Unfortunately, SOAP
alone is insufficient, just as XML before it was not enough. SOAP as it
stands today assumes that all parties "just know" what messages can be
exchanged and how they relate. Despite the persistent protests of one of the
original authors (yours truly), SOAP remains completely silent as to how one
becomes aware of this information. I personally believe that the relatively
late development of standardized discovery and metadata formats held SOAP
development back for most of 1999 and 2000.

Do what I mean, not what I say
Whenever two or more parties collaborate, some shared understanding of what
the collaboration will entail is needed. First and foremost, all parties
need to understand what it means to collaborate. What are the benefits for
each party? What are each party's responsibilities and obligations? This is
true in human interaction just as it is true in Web services interaction or
software integration in general.

The meaning and terms of a particular collaboration can be viewed as a
contract. In an ideal world, contracts would have crisp and concise
descriptions of the contract semantics and nothing else. For example, if I
hired you to mow my lawn, the contract might read as follows:

The reader will mow the writer's lawn once per week in exchange for ten
dollars.
This contract is pure semantics. You mow my lawn. I pay you. Clean. Simple.
Done.

Unfortunately, the US legal industry has made a fortune cleaning up the mess
made by contracts that look like this example. Why? Because the intention of
one or more parties is not adequately captured in the language of the
contract. The vagaries of the English language leave far too much room for
interpretation. For example, what if more than one reader were to mow my
lawn? Do I need to pay them all or just you (and you know who "you" are)?
Also, can I pay you in Canadian dollars? Can I pay you once per decade in
arrears? What recourse do I have if you do a bad job? In human contracts as
well as software contracts, the more intention that can be explicitly
stated, the less opportunity there is for misinterpretation.

In the world of Web services, there is a strong argument that contracts
should be stored and transmitted in a machine-readable format.
Machine-readable formats are preferable to non-machine-readable formats for
several reasons. For one, having the contract in machine-readable form
allows any number of features to be enabled, including validation,
acceleration, and deep tool integration. Additionally, once you have a
machine-readable format, you can write a program that transforms the
contract into any other known format. This includes transforming the
contract into human-readable documentation. The converse is not the case;
trying to harvest anything from normal documentation inevitably requires
lots of human intervention.

The idea of having machine-readable contracts is obvious to developers who
work in strongly typed programming languages such as Java, Delphi, or C++.
Unfortunately, the idea of machine-readable contracts is hard to swallow for
some programmers, especially those who have embraced the weakly typed world
of most scripting languages. However, until the entire world converts to
Perl, something needs to be done to enable strongly typed languages to deal
with Web services in a natural way.

A cry for help
I have tried over the years to convince the scripting community of the
merits of machine-readable contract formats. I have tried technical
arguments, but to no avail. I have been tempted to use ad hominem arguments,
but I can't bring myself to going that far. So today, I will try a different
approach based on personal confession and self-deprecation:



Hello, my name is Don.



I use strongly typed languages.



I need your help.



I have tried to use scripting languages exclusively. I really did.

I will admit that working in script made me feel free. It was exciting. It
was even addictive. Unfortunately, my addiction got the better of me.

My first problem was that sometimes my programs didn't always work. Try as I
might, sometimes I forgot to pass a parameter to a function. Other times, I
would pass parameters like "XXVII" or "Twenty Seven" to my math libraries.
My biggest problem was that my spelling is really horrible, and I found that
while many scripting languages are not case-sensitive, they all seem to be
spelling-sensitive.

What was really weird was that a lot of my programming errors never showed
up when I was testing my script programs. Instead, they always seemed to
surface after my users started to apply my programs out in the field. This
made me very unhappy, especially when it happened to users who had my home
phone number. I think my wife hated my use of script more than anyone. She
reminded me that we used to get a lot of the same sorts of calls back when I
worked in K&R C, and that once I switched to ANSI C, we got a lot more
sleep.

I found that by eschewing all alcohol and limiting myself to 30 minutes of
coding per day to reduce fatigue, I was able to get my error counts down
considerably. Unfortunately, I always find myself working with other
programmers, many of whom use strongly typed languages. I tried to get these
guys to use my programs with theirs, but it always led to problems. My
Windows-based scripts were accessible, but only through this thing called
IDispatch, which my C++ friends seemed to find offensive. They would often
write these weird-looking IDL files, often while saying some not so nice
things about me under their breath. I would try to convince them to move
over to script like me, but to no avail. I quickly found out that C++
programmers are the smartest programmers in the world, as each one I met
would quickly point out to me.

I must admit that I am weak. I ultimately succumbed to strongly typed
languages.

I was not strong enough to write code without the crutch of a compiler to
check my program for stupid errors and spelling mistakes.

I was not strong enough to ignore the taunts of my colleagues working in
C++.

And now, I need the help of the scripting community.

Despite the years I spent trying to make SOAP a standard way for programs to
communicate over the Internet; I find that raw SOAP and XML are at odds with
all of these compilers I am now using. I am told that if you give me
machine-readable contract definitions, my compiler can help me talk to your
Web services. A lot.

If you don't give me a machine-readable contract, then I am going to have to
write one of these weird-looking WSDL files by hand, and that always makes
me cranky. I understand that writing WSDL makes you cranky too, but I'll bet
if you wrote the WSDL once and put it on your Web site, everyone else would
just use it, and no one would ever need to write that WSDL again. And if you
wrote a ten-line WS-Inspection or DISCO file to go along with it, I could
find out about all of your other services too.

I know that WSDL isn't perfect. God knows I tried to make it better prior to
publication. Luckily, the W3C just launched a WSDL working group and it
looks like the community at large has the will to clean it up, just as SOAP
was cleaned up once it got the attention of a large community of
practitioners and experts. In fact, SOAPBuilders is running a WSDL bake-off
in February that surely will yield some progress on this front.

I also know that writing WSDL for your script-based Web services is more
work for you, but your suffering would benefit thousands or more developers
anxious to use your stuff. And just think of the nice things they will say
about you once you made their lives easier.

And not under their breath.

Related links:
WSDL 1.1 Specification
SOAP v1.1 Specification
Web Services Activity



Don Box is a leading educator, recognized authority on the Component Object
Model (COM), co-author of the Simple Object Access Protocol (SOAP)
specification, and coiner of the term "COM is Love." He recently joined
Microsoft as an architect in the Microsoft? .NET Developer and Platform
Evangelism Group. Box is the best-selling author of three books: Essential
COM, Effective COM and Essential XML. He is also writing a series of books
examining the .NET strategy entitled Essential .NET. Earlier in his career,
Box co-founded DevelopMentor Inc., a component software think tank aimed at
educating developers on the use of the COM, Java and XML. A popular public
speaker, Box is known for engaging audiences around the world, combining
deep technical insight with often outrageous stunts?not the least of which
was conducting a discussion on SOAP while in a bathtub full of suds at
Tech?Ed 2001 in Barcelona. During his free time, Box enjoys jamming with
fellow MSDN Magazine contributors in Band on the Runtime, which performs
songs on various programming topics. Box is also responsible for the
creation of a fashion line, starting with Don Boxers.




From fperez528 at yahoo.com  Mon Feb  4 07:18:59 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 04 Feb 2002 12:18:59 +0000
Subject: how can i get the screen width and height in pixels?
References: <a3mclk$pjh$1@news5.svr.pol.co.uk> <3c5ec34b.27704203@news.t-online.de>
Message-ID: <a3mmt9$3q4$1@peabody.colorado.edu>

Gerson Kurz wrote:

> On Linux:
> 
> Buy a metric ruler, then construct a lego mindstorm robot to measure
> the screen with that ruler.
> 
> (Sorry folks, couldn't resist, I have nothing against linux, in fact I
> like it :).

Or you can actually do something useful:

[~]> xdpyinfo | grep dimensions
  dimensions:    1400x1050 pixels (474x356 millimeters)

f.


From sakkinen at cs.jyu.fi  Tue Feb 12 02:21:03 2002
From: sakkinen at cs.jyu.fi (Markku Sakkinen)
Date: Tue, 12 Feb 2002 07:21:03 +0000 (UTC)
Subject: Inheritance workshop
Message-ID: <a4afov$mri$5@mordred.cc.jyu.fi>

			       CFP for

		The Inheritance Workshop at ECOOP 2002
		======================================

   - The first ECOOP workshop focusing on inheritance since 1992 -


Call for Papers and Participation
---------------------------------

This workshop is intended as a forum for designers and implementers of
object-oriented languages, and for software developers with an
interest in inheritance. It is hard to come up with a litmus test for
object-orientedness because of the size and diversity of this field,
but one of the most widely accepted ingredients would be
inheritance. There has even been a backlash against inheritance,
focusing on problems like fragile base classes, the so-called
inheritance anomaly, encapsulation between a class and its subclasses,
and more. However, we believe the abundant activity just demonstrates
that inheritance is both hard to avoid, and hard to get right. The
goal of this workshop is to advance the state of the art of designing
inheritance mechanisms, and using inheritance judiciously. Many
subtopics are relevant, including the following:

  - Flaws and anomalies in the way inheritance is currently
    implemented or used. Examples which are hard to express.

  - Inheritance mechanisms: Single, multiple, mixin based, classless,
    composing, first-class or second-class, aspect-oriented.

  - Inheritance and parameterization, inheritance and typing,
    inheritance and encapsulation, inheritance and behavior.

  - Inheritance and software maintenance, inheritance
    vs. analysis/design/implementation, inheritance vs. program
    understanding, refactoring, tools.

  - Conceptual views upon inheritance: As a conceptual modeling tool,
    as a dependency management tool, as a code reuse mechanism.

  - Experience on the benefits and drawbacks of inheritance in real
    projects.

We invite high quality position papers within the topic area.
Experience papers are particularly solicited.  The papers should be at
most 7 pages long (shorter papers are preferred) and in PDF or
Postscript format (other formats, e.g. HTML, can be accepted if
necessary), and they should be submitted as e-mail attachments to the
address iws2002 at cs.jyu.fi.  Final versions of accepted position papers
will be made available from the workshop web page,
http://www.cs.auc.dk/~eernst/inhws/.

The maximum number of participants will be about 30, and the maximum
number of position papers to be presented will be about 12. We will
accept participants in two categories:

  - Constructive: those who present a position paper.
    Note that more papers may be accepted to the workshop proceedings
    than can be actually presented.

  - Critical: every other participant is expected to study one
    or more position papers thoroughly in advance, and present concise
    comments at the workshop.

Every prospective participant is required to submit a position paper,
and every accepted participant is urged to read all the position papers,
such that the presentations, critical comments, and discussions will be
based on a mutual understanding of concepts and terminology.

The organizers will select the participants and the papers to be
presented based on several criteria:

  - the quality and relevance of each position paper
  - a sufficient breadth of topics (even a good paper may get rejected,
    or not selected for presentation, if there are too many, too
    closely related ones)
  - the number of authors of each position paper (in the worst case,
    only the main author will be invited)
  - ``first come first served'', where other criteria are inconclusive


Workshop web site
-----------------

The workshop web site is

  http://www.cs.auc.dk/~eernst/inhws/

which contains the call for papers, the time table, and additional
information about the workshop.  A list of participants and the
accepted papers will also be made available from there.


Workshop schedule
-----------------

Here is a preliminary version of the workshop schedule:

  1.  Short paper presentations with short comments
  2.  Discussion of most relevant topic areas in 2 to 4 subgroups
  3.  General discussion, based on subgroup reports


Publication
-----------

The collection of the final versions of all accepted workshop papers,
including those which are not presented, will be published as a
technical report by the Information Technology Research Institute
(ITRI) of the University of Jyv?skyl?.  The printed report will be
distributed at the workshop.

Springer-Verlag will publish the ECOOP'2002 Workshop Reader as an LNCS
volume.  This volume will include a report for each workshop, which
should not consist simply of a collection of the position papers
presented at the workshop.  The organizers will write the report, in
collaboration with the participants of the workshop. Your active
collaboration will thus be needed even after the workshop itself.

The report will provide a summary of the workshop with the major
issues discussed and the conclusions of the working groups (if
applicable).  The report will also discuss current research being
carried out in the area, and open research directions on the workshop
themes.


Important Dates
---------------

The ECOOP 2002 organizers have requested the same deadlines for all
the workshops, and they are as follows:

  Position papers due: April 8, 2002

  Notification of acceptance: April 29, 2002

  Deadline for early registration: May 6, 2002

  Final version of accepted papers due: May 27, 2002
    (specific to this workshop)

  Workshop: June 11, 2002 (Tuesday)


Workshop Organizers
-------------------

  Andrew P. Black
  OGI School of Science & Engineering
  Oregon Health & Science University
  Beaverton, Oregon, USA
  black at cse.ogi.edu
  http://www.cse.ogi.edu/~black/

  Erik Ernst
  DAIMI / Department of Computer Science
  University of Aarhus
  Denmark
  eernst at daimi.au.dk
  http://www.daimi.au.dk/~eernst/ 

  Peter Grogono
  Department of Computer Science
  Concordia University
  Montreal, Canada
  grogono at cs.concordia.ca
  http://www.cs.concordia.ca/~faculty/grogono/

  Markku Sakkinen
  Information Technology Research Institute
  University of Jyv?skyl?
  Finland
  sakkinenm at acm.org
  http://www.cs.jyu.fi/~sakkinen/


From bokr at oz.net  Mon Feb  4 17:42:45 2002
From: bokr at oz.net (Bengt Richter)
Date: 4 Feb 2002 22:42:45 GMT
Subject: Equivalent of Perl chomp?
References: <mailman.1012626605.22240.python-list@python.org> <a3lfuf$6sj$0@216.39.172.122>
Message-ID: <a3n2p5$sp5$0@216.39.172.122>

On 4 Feb 2002 08:15:11 GMT, bokr at oz.net (Bengt Richter) wrote:
[... a bunch of timing discussion and code ...]

>[23:00] C:\pywk\crdtsc>python endsvslice.py 1 10 850 1000
>10 loops, 1000 avgloops, 850 tadj
> endswith1:  1312, avg  1327
> endswith2:  1335, avg  1366
> endswith3:  1368, avg  1407
> endswith4:  1432, avg  1388
>    slice1:   669, avg   740
>    slice2:  1898, avg  1942
>    slice3:  1937, avg  1934
>    slice4:  1915, avg  1907
>
>IOW, endswith did not take a huge jump measuring my way, but
>the slice was certainly affected from 1 to 2.
>
>The little dip for slice4 is curious. Here are some more trials of that:
>    slice4:  1898, avg  1916
>    slice4:  1909, avg  1913
>    slice4:  1909, avg  1913
>Hm, maybe slice3 above is off?
>    slice3:  1945, avg  1934
>    slice3:  1926, avg  1933
>    slice3:  1928, avg  1942
>    slice3:  1927, avg  1941
>Curiouser and curiouser... anyway, no huge anomalies ;-)
>
I am wondering if the dip might be due to being able to access
aligned chunks better than odd sized ones. Testing that theory:

Strings where len(gtail)%2 == 0 seem to be a little faster. I
pasted in a second run for consistency check. I was looking
for some advantage in %4 but I don't see it here. Is the actual
string data aligned on 4 or 2 bytes in python's strings?

The routine being timed is

def slice_n(s):
    global gnslice, gtail
    return s[-gnslice:] == gtail

If the leftmost printout number below is trial, each trial sets up
    gnslice = trial+1
    gtail = ' '*trial+'\n'
before calling the timing stuff. Note constant 2850 offset
to make number differences stand out more.
Time units are 300mhz ticks.

[13:56] C:\pywk\crdtsc>python endsvslice9.py 9 10 2850 1000 20
10 loops, 1000 avgloops, 2850 tadj 20 ntrials
  0:    slice_n: -1056, avg -1044    |  -1043
  1:    slice_n:   152, avg   164    |    159 <- (maybe not gaining via loop yet?)
  2:    slice_n:   102, avg   124    |    123 
  3:    slice_n:   109, avg   102    |    104 <-
  4:    slice_n:    66, avg   124    |    126 
  5:    slice_n:    96, avg   101    |     99 <-
  6:    slice_n:   101, avg   132    |    123 
  7:    slice_n:    74, avg   106    |    103 <-
  8:    slice_n:   103, avg   127    |    130 
  9:    slice_n:    97, avg   107    |    107 <-
 10:    slice_n:   103, avg   128    |    125 
 11:    slice_n:   126, avg   108    |    109 <-
 12:    slice_n:   134, avg   125    |    124 
 13:    slice_n:    88, avg   108    |    100 <-
 14:    slice_n:   109, avg   128    |    124 
 15:    slice_n:    79, avg   109    |    106 <-
 16:    slice_n:   135, avg   153    |    140 
 17:    slice_n:    78, avg   113    |    115 <-
 18:    slice_n:   116, avg   154    |    148 
 19:    slice_n:   125, avg   123    |    119 <-

Kind of interesting what you can pick up. I wonder what
a spectral analysis of raw timing result sequences would
reveal. I'd expect to pick up any cyclic behavior in
memory pool usage, etc.

Regards,
Bengt Richter



From xhalasa at fi.muni.cz  Thu Feb 14 17:44:27 2002
From: xhalasa at fi.muni.cz (=?ISO-8859-2?Q?J=E1n_Hala=B9a?=)
Date: Thu, 14 Feb 2002 22:44:27 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <GrJoHr.9Hn@news.muni.cz>

Scirious wrote:

> Dear friends;
> 
> I've been learning JAVA for a while, but now I've heard about Python.
> Any coments concerning these two languages?
> 
> TIA;
> Scirious.

Hi,

you can find performance comparison
of these two and many other languages at
http://www.bagley.org/~doug/shootout/

Jan


From phr-n2002a at nightsong.com  Mon Feb 25 08:16:26 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 25 Feb 2002 05:16:26 -0800
Subject: strings and sort()
References: <mailman.1014641682.25067.python-list@python.org>
Message-ID: <7xadtxln11.fsf@ruckus.brouhaha.com>

"Jason Orendorff" <jason at jorendorff.com> writes:
> But:
>   Flat is better than nested.
>   Sparse is better than dense.
>   Readability counts.
> 
> That's why Python is the way it is.

But: dense
             can
                   be
                           more
          readable
  than
                       sparse.

See?


From alien275 at libertysurf.fr  Wed Feb 27 10:09:41 2002
From: alien275 at libertysurf.fr (Riss Nicolas)
Date: Wed, 27 Feb 2002 16:09:41 +0100
Subject: Telnet connection
Message-ID: <a5iskt$ge8$1@wanadoo.fr>

Hello,
I must do a program who can connect with telnet. But I don't know how to
make a such connection with python, can somebody help me please.

Thanks




From jgitomer at erols.com  Mon Feb 18 22:58:26 2002
From: jgitomer at erols.com (Jerry Gitomer)
Date: Mon, 18 Feb 2002 22:58:26 -0500
Subject: Operating system requirements
References: <3C71A556.3E7BF27E@xatt.netx>
Message-ID: <3C71CD62.23BF11AC@erols.com>

John Burke wrote:
> 
> What are the operating system requirements for python to run on a Windows
> pc. Thank you.
> 
> John

Windows 95 or newer.


From peter at engcorp.com  Thu Feb  7 20:18:27 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 07 Feb 2002 20:18:27 -0500
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <Xns91AE6BE28FCD8michaelrcpcouk@194.238.50.13>
Message-ID: <3C632763.4DF779D0@engcorp.com>

Michael Abbott wrote:
> Well, I think the most enthusiastic supporter of Python has to acknowledge
> that Python has a performance problem when systems are pushed to their
> limits.  The fact that Python cannot be effectively compiled to machine
> code (while Java can) makes a substantial difference when it comes down to
> raw machine cycles, and Python's use of memory is not particularly
> efficient.
> 
> I reckon to pay a cost of perhaps a factor of 100 in raw CPU for the
> privilege of using Python.

My rule of thumb, based on informal but extensive measurements over
a year ago, is "between 10 and 100 times slower than C" for simple
problems (i.e. not trying to code up sophisticated dictionaries in
C to compare the performance at that level).  Average roughly 30.

This would *not* be the same as saying 100 times slower than Java
by any stretch of the imagination.  I believe that comparison would
be closer to "3 times slower".

My theory is also that Python code is much more "optimizable" in
the sense of "potential increase in performance for unit optimization
effort".  I can't prove that yet, and it doesn't apply to very
experienced Python developers who might dive for map() and filter()
much sooner than more recent arrivals.

-Peter


From mwh at python.net  Thu Feb 21 11:59:53 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 16:59:53 GMT
Subject: Embedding: How to set globals for PyObject_Call
References: <mailman.1014234635.18641.python-list@python.org> <79b2b0bb.0202210841.457c0b81@posting.google.com>
Message-ID: <uit8qlqdm.fsf@python.net>

xscottgjunk at yahoo.com (Scott Gilbert) writes:

>   // This one breaks
>   Var g = VarEval("globals");
>   g(); // This causes a segfault

Do we get to see the backtrace?

>   // This one works
>   Var g = VarEval("globals()");
> 
>   // As you pointed out, This one works too
>   VarExec("def GLOBALS(): return globals()");
>   Var g = VarEval("GLOBALS");
>   g();
> 
> Many other functions in the __builtin__ module work just fine.  For
> instance:
> 
>   Var hex = VarEval("hex");
>   VarPrint(hex(1234));
> 
> So I think it's a little odd that some builtins are different in this
> regard than others.

I'd agree here.

Cheers,
M.

-- 
  Guido (like us!) is a bit schizophrenic here: he wants to be a
  benevolent dictator, but also wants to treat people like
  grownups. This probably worked better before Python got a large
  American audience <0.9 wink>.             -- Tim Peters, 10 Feb 2000


From mhammond at skippinet.com.au  Thu Feb  7 07:25:15 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 07 Feb 2002 12:25:15 GMT
Subject: Questions  about pyscript
References: <a3tegq$1ahish$1@ID-12869.news.dfncis.de>
Message-ID: <3C62727A.1090306@skippinet.com.au>

Ricky wrote:
> I make a try a pyscript. But there is one thin confuse me. When I use print
> statement intent to send something to browse. It doesn't work, while
> document.write can make it. What make me confused is what part is controlled
> under python(such as flow control, functions, package etc) and which part is
> controlled by DHTML(output?).
> 
> Could some one give me an over view option of what pyscript is and the
> object tree map of pyscript is a plus.

ActiveScripting (ie, pyscript) is a technology that allows an 
application to expose an object model, and allow many languages to use 
scripting.  Python works with IE, ASP, WSH and a number of other 
applications - thus there is no "pyscript" documentation that can help you.

IE is just one such application (and the one you seem to be using).  MS 
documents this object model with IE - however, you are only likely to 
find examples for VBScript or jscript, so will need to convert them to 
Python yourself.

Since pyscript doesn't really know about IE in any special way, it 
doesn't know how to redirect 'print' statements to document.write() 
calls.  That has been my exuse anyway - hard-coding for IE wouldn't 
necessarily be too evil :)

Mark.



From starx at pacbell.net  Fri Feb  8 20:04:02 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 01:04:02 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4>
Message-ID: <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4>

Flavian Hardcastle || Fri 08 Feb 2002 04:23:16p:

> Ello. I'm a newbie programmer using the Python language. I've recently 
> completed a couple of the Python beginner's tutorials out there on the
> web (Josh Cagliati's, and Richard Baldwin's).
> 
> I have this ambitious idea for a project. I want to make my own browser
> based chat room. 
> 
> It seems simple. I basically wrote a little Python subroutine which
> takes input via the Python IDE, and writes a little html document using
> IO commands. Voila! Un chat program! Albeit an extremely simple one at
> first. Now if I could just train the little bugger to take input from a
> web browser I'd be in business. I figured the easiest way to do this
> was to write an html form that writes the input to a text or dat file,
> and my prog can extract the data from that and print it back to web
> page 
> 
> But ... I can't find any easy way to do that, to my suprise. I tried 
> looking through all the html tutorials for information about forms. The
> nearest I could come to, specific to my requirements, was tutorials on
> how to set up your own messageboard. Apparantly, I have to get a web
> server that supports CGI (whatever that is), use a telnet
> client(whatever that is) to connect to it. I also have to learn PERL
> (What? A whole new programming language?), and ideally the server
> should be UNIX (A whole new operating system?).
> 
> I can't believe I have to learn 4 things at once just to get to get
> input from a browser! Surely there must be an easier way to do it?
> 
> If I could,  I would much rather be able to have the script on my PC,
> use my personal web server to host it. Then I'd invite a net buddy or
> two into it to help me perfect the script. And then I might think about
> actually putting the room on a proper server, if I don't lose interest
> in the project by then.

Well, as far as i can remeber, CGI is just an interface, you can use it 
with Perl or Python or C or whatever. You probably will have to learn it, 
but it should just be a modlue and some extra commands IN python... and 
you'll have to configure whatever web server you're using to use CGI 
program, that's probaly the hard part.

As for perl, well, a lot of CGI-based things happen to be written in perl 
(i think), which is probably why those tutorials said 'learn perl'. You 
don't need to.

Telnet is just a program -- all you have to know is how to put a server's 
address and port number into it, after that, it's either something just 
like a dos window for a remote unix system, or a mud interface ;). And for 
what you would have to do to manage your files on unix, well, you might 
need mv, cp, rm, mkdir, cd, ls/dir, and maybe chmod, chown, chgrp... a 
total of 10 commands, at least 5 of which are almost no-brainers (cd, 
mv=move, cp=copy, rm=del, mkdir ... ls has lots of switches, some are good 
:)

Anyway, you'd only need unix sometimes if you used that kind of server, and 
you don't even want a server... if you can get your computer to run CGI's, 
and check out 11.2 (or so) of the library reference manual for the CGI 
commands, you should be most of the way there.

Sorry for the longness :)


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From krissepu at vip.fi  Wed Feb 20 15:55:31 2002
From: krissepu at vip.fi (Pekka Niiranen)
Date: Wed, 20 Feb 2002 22:55:31 +0200
Subject: WXWINDOW: CANCEL -button and validation
Message-ID: <3C740D43.E925C8F@vip.fi>

Hi,

In code below:

-  when user enters empty values and presses OK -button, validator
    warns about the empty values.

-  when user enters empty values and presses CANCEL -button, validator
    is not run and I must add "useless checks for empty values" in code
+ default values

- I can kill the wxDialog with empty values on boxes.

Questions:

1) How can I run Validate(self, parent) -method when CANCEL -button is
pressed ?
2) How can disable the close button (X -button on top right corner of
window) from wxDialog ?

 -------------- Code starts here --------------------------------

class EditConversionParameters(wxDialog):
    def __init__(self, parent, conv_fileprefix, conv_fileprefix_default,

                 conv_prefix, conv_prefix_default,
                 conv_suffix, conv_suffix_default, style):
        wxDialog.__init__(self, parent, -1, "  Conversion parameters",
                          wxDefaultPosition, wxSize(280, 150), style)

        # Defaults
        self.conv_fileprefix_default = conv_fileprefix_default
        self.conv_prefix_default = conv_prefix_default
        self.conv_suffix_default = conv_suffix_default

XXXXXXX Useless checks starts here XXXXXXXXXXXXXXXX

        # Make sure user does not leave with empty boxvalues ever !
        if conv_fileprefix == "":
            self.conv_fileprefix = conv_fileprefix_default
        else:
            self.conv_fileprefix = conv_fileprefix

        if conv_prefix == "":
            self.conv_prefix = conv_prefix_default
        else:
            self.conv_prefix = conv_prefix

        if conv_suffix== "":
            self.conv_suffix = conv_suffix_default
        else:
            self.conv_suffix = conv_suffix

XXXXXXX Useless checks stops here XXXXXXXXXXXXXXXX

        # Create validators
        self.conv_fileprefix_validator=InputValidator(ID_FPREF)
        self.conv_prefix_validator=InputValidator(ID_SPREF)
        self.conv_suffix_validator=InputValidator(ID_SSUFF)

        # Add static texts
        self.text1 = wxStaticText(self, -1, "Filename prefix:",
wxPoint(10,24))
        self.text2 = wxStaticText(self, -1, "Conversion prefix:",
wxPoint(10,54))
        self.text3 = wxStaticText(self, -1, "Conversion suffix:",
wxPoint(10,84))

        # Add dialog boxes
        self.box1  = wxTextCtrl(self, -1, self.conv_fileprefix,
wxPoint(100,20),
                                size=(50, -1),
validator=self.conv_fileprefix_validator)
        self.box2  = wxTextCtrl(self, -1, self.conv_prefix,
wxPoint(100,50),
                                size=(50, -1),
validator=self.conv_prefix_validator)
        self.box3  = wxTextCtrl(self, -1, self.conv_suffix,
wxPoint(100,80),
                                size=(50, -1),
validator=self.conv_suffix_validator)

        # Add some buttons
        self.button1 = wxButton(self, ID_CONV_DEFAULT, " Defaults ",
wxPoint(180,20), wxDefaultSize)
        self.button2 = wxButton(self, wxID_OK, " OK ",  wxPoint(180,50),
wxDefaultSize).SetDefault()
        self.button3 = wxButton(self, wxID_CANCEL, " Cancel ",
wxPoint(180,80), wxDefaultSize)

        # Add button events
        EVT_BUTTON(self, ID_CONV_DEFAULT, self.set_defaults)

    def set_defaults(self, event):
        self.box1.SetValue(self.conv_fileprefix_default)
        self.box2.SetValue(self.conv_prefix_default)
        self.box3.SetValue(self.conv_suffix_default)


class InputValidator(wxPyValidator):
    def __init__(self, flag):
        wxPyValidator.__init__(self)
        self.flag = flag
        EVT_CHAR(self, self.OnChar)

    def Clone(self):
        """ All validator classes must implement the Clone function,
        which returns an identical copy of itself."""
        return InputValidator(self.flag)

    def Validate(self, parent):
         """ Validate the contents of the given text control.
         Validator gets executed only when OK -button is pressed"""
         textCtrl = self.GetWindow()
         text = textCtrl.GetValue()
         if len(text) == 0:
             dlg = wxMessageDialog(parent, "All boxes must contain some
text!", " Error", wxOK )
             dlg.ShowModal()
             dlg.Destroy()
             textCtrl.SetFocus()
             return false
         else:
             return true

    def TransferToWindow(self):
         return true # Prevent wxDialog from complaining.

    def TransferFromWindow(self):
         return true # Prevent wxDialog from complaining.

    def OnChar(self, event):
       # Control which characters can be entered to each box
        key = event.KeyCode()
        if key == WXK_BACK or key == WXK_DELETE:
            event.Skip()
            return
        if self.flag == ID_FPREF and (chr(key) in string.ascii_letters
or
                                      chr(key) in string.digits or
                                      chr(key) == "_"):
            event.Skip()
            return
        if self.flag == ID_SPREF and (chr(key) in string.punctuation):
            event.Skip()
            return
        if self.flag == ID_SSUFF and (chr(key) in string.punctuation):
            event.Skip()
            return

        # Returning without calling even.Skip eats the event before it
        # gets to the text control : no text is entered into box.
        return

-pekka-



From tim.one at home.com  Sat Feb  2 14:54:00 2002
From: tim.one at home.com (Tim Peters)
Date: Sat, 2 Feb 2002 14:54:00 -0500
Subject: "Private" attributes - a possible idea (maybe PEP-worthy).
In-Reply-To: <VUM68.1204$3E5.108514@newsread2.prod.itd.earthlink.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEHFNJAA.tim.one@home.com>

[Chris Tavares]
> How about changing the name mangling rule so that the current
> number of base classes is added to the secret string?

I'm not sure what "current number of base classes" means.  In

class X(Y):
    __pvt = 1

what's the current number of base classes?  Keep in mind that name-mangling
is done at compile-time, and has only the static text of the module being
compiled to examine, so the only feasible answer I can come up with is "1".
But that wouldn't help (if Y derived from another class named X which
derived from Z, the count would be 1 on both incarnations of X).




From cce at clarkevans.com  Tue Feb 12 17:42:35 2002
From: cce at clarkevans.com (Clark C . Evans)
Date: Tue, 12 Feb 2002 17:42:35 -0500
Subject: Re-throwing an exception without loosing the stack trace
Message-ID: <20020212174235.A17928@doublegemini.com>

How do I catch and then re-throw an exception without
loosing the information about the stack trace?  I've
looked around alot and don't see the "obvious" answer.

Thanks!

Clark

P.S. I need to catch exceptions to clean up some
     resources (database locked table), but
     then I want to re-throw the error... in
     it's original stack trace so that I can
     find it... loosing the stack trace makes
     for brutal debugging.




From emile at fenx.com  Wed Feb  6 09:29:04 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 06:29:04 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <st888.7971$oI4.33917985@newsserver.ip.pt>
Message-ID: <a3retd$19cgik$1@ID-11957.news.dfncis.de>

"Hugo Martires" <hmartires at cultalg.pt> wrote in message
news:st888.7971$oI4.33917985 at newsserver.ip.pt...
> but i don?t need all the numbers, i only need 0.6

See 4.98 in the FAQ at http://python.org/doc/FAQ.html

--

Emile van Sebille
emile at fenx.com

---------



From tim.one at comcast.net  Tue Feb 26 01:43:07 2002
From: tim.one at comcast.net (Tim Peters)
Date: Tue, 26 Feb 2002 01:43:07 -0500
Subject: size of socket.recv buffer
In-Reply-To: <a5elgo$5bu$1@news.cis.ohio-state.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGNNPAA.tim.one@comcast.net>

[Locke]
> The argument to recv is supposed to be the buffer, acording to the
> documentation.

Certainly not according to Python's docs.  It's the maximum length of the
string you want Python's recv to return.

> But the doc doesnt say: buffer measured in what?

Python's recv() returns a string; it's an upper bound on

    len(sock.recv())

> Is it bytes?  Octets?  (by the way, what is the diff between a byte
> and an octet?)

Nothing in real life <wink>.

> See, I am writing a pop3 client type thing, after requesting a
> message from the server, my first s.recv tells me how long the
> message will be in octets.  So then I call s.recv again passing the
> number of octets as the argument to s.recv. Shouldn't this download
> the entire message? What actually happens is that it cuts off
> right in the middle of the message.
> ...

This is what you need:

    http://py-howto.sourceforge.net/sockets/sockets.html

Read that and it will be clearer.  Note that you can't even count on your
*first* recv to tell you the number of octets in one gulp, although that's
quite likely to work most of the time.  Sockets are a low-level facility, as
the title of the Python doc section hints:

    7.2 socket -- Low-level networking interface




From dsavitsk at e-coli.net  Mon Feb 11 20:08:30 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 12 Feb 2002 01:08:30 GMT
Subject: just a moron trying to get python working on win98 and personal web  server
References: <3C685DDE.AA8AFD10@kc.rr.com>
Message-ID: <iWZ98.3434$pC1.723529@newssrv26.news.prodigy.com>

not clear what you want to do.

python .asp should work so long as vbscript .asp works.  for python .cgi see
http://www.pythonpros.com/arw/pws.html.

also see www.modpython.org, modsnake.sourceforge.net/,
www.bel-epa.com/pyapache/ for using python w/ apache.


"james" <jdaily at kc.rr.com> wrote in message
news:3C685DDE.AA8AFD10 at kc.rr.com...
>
> RE: just a moron trying to get python working on win98 and personal web
> server
>
> i am trying to get python to work on my MS personal web server ( which
> is
> supposed to support python in a web url ).
>
> what do i need to do?
>
> i have tried setting the registry.
>
> i have set the PATH in autoexec.bat.
>
> is there an apache alternative.
>
>




From graeme.matthew at unite.com.au  Wed Feb 13 06:19:32 2002
From: graeme.matthew at unite.com.au (Graeme Matthew)
Date: Wed, 13 Feb 2002 22:19:32 +1100
Subject: Having a mare with Classes
Message-ID: <3C6A4BC4.4060409@unite.com.au>

Please can someone help, im having an absolute mare, I have Wesley Chuns 
book on core python programming and I cannot get casses to work, im just 
using a simple test from the examples in the book.

OK I HAVE A CLASS PERSON

class Person:

     def printName(self):
         print 'You printed it'

I HAVE A SCRIPT:

#!/usr/local/bin/Python2.2


import sys

sys.path.append('/www/cgi-bin')

import Person
person = Person()

person.printName()

I keep on getting this error:

#! /usr/local/bin/Python2.2


TypeError: call or non function (type module)

Please if you can help it would be much appreciated.

Thanks graeme



From mertz at gnosis.cx  Sat Feb  2 00:58:34 2002
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Sat, 02 Feb 2002 00:58:34 -0500
Subject: Why has 'dir()' changed in Python 2.2?
References: <mailman.1012624144.16504.python-list@python.org>
Message-ID: <KA4W8kKkXIda092yn@bellatlantic.net>                         

Me, previously:
|> I just noticed a change in Python 2.2... that turns out to break my
|> [xml_pickle] module.  It shouldn't be hard to fix (although I *do* have
|> a decision to ponder about the best fix)

"Tim Peters" <tim.one at home.com> wrote previously:
|I don't know what you were using it for exactly, but chances
|are good you should be looking at object.__dict__.keys() instead (which
|is all dir() often did before 2.2).

The timbot, as always, can peer through the surface words on the Usenet
into my deepest, dark thoughts.  The obj.__dict__.keys() is probably
what I'll change it to, which should restore the old behavior.  As the
name suggests, the [xml_pickle] module serializes Python objects to an
XML format... so dir() was a convenient shortcut for asking "what's this
object got?"

I probably should have known better, and -do- vaguely recall some
discussion about dir() on c.l.py around the time 2.2 alphas were coming
out.  The thing I want to think about is whether I actually *want* any
of the extra information that dir() gives me; for example, maybe the
docstring.

Do I understand the new behavior correctly in thinking that the
following function is equivalent to dir(), at least for old-style class
instances:

    def dir2(o):
        lst = o.__dict__.keys()
        lst.extend(o.__class__.__dict__.keys())
        return lst

Or is there something else thrown in there that this overlooks?

Yours, David...

--
mertz@  | The specter of free information is haunting the `Net!  All the
gnosis  | powers of IP- and crypto-tyranny have entered into an unholy
.cx     | alliance...ideas have nothing to lose but their chains.  Unite
        | against "intellectual property" and anti-privacy regimes!
-------------------------------------------------------------------------




From paul at prescod.net  Mon Feb 11 11:54:53 2002
From: paul at prescod.net (Paul Prescod)
Date: Mon, 11 Feb 2002 08:54:53 -0800
Subject: RFC: Option Parsing Libraries
Message-ID: <3C67F75D.D1EF37DA@prescod.net>

Greg Ward has proposed to add his Optik module to the Python library. 

 * http://mail.python.org/pipermail/python-dev/2002-February/019934.html

Optik is a parser for command line options with the features described
here:

 * http://mail.python.org/pipermail/python-dev/2002-February/019937.html

If you have a competitive library, or suggestions for changes to Optik,
please forward your comments to python-dev mailing list
(python-dev at python.org). If a long-running conversation ensues, you may
need to join the list to participate. Discussions in comp.lang.python
will not be considered unless you at least forward a reference to
python-dev. If you propose a competitor to Optik, please describe how it
compares to the feature list described above.

 Paul Prescod



From spencer at efn.org  Thu Feb 21 17:52:13 2002
From: spencer at efn.org (Spencer Ernest Doidge)
Date: Thu, 21 Feb 2002 14:52:13 -0800
Subject: win32file.AllocateReadBuffer(..)
Message-ID: <Pine.GSU.4.21.0202211448560.17401-100000@garcia.efn.org>

If I call win32file.AllocateReadBuffer(..) in a method that gets called
repeatedly and terminates each time it is done, will I have a memory leak
if I don't somehow free the allocated memory before concluding the run of
the method? Or does the memory get de-allocated automagically when the
method is finished running?

Spencer Doidge
www.spencerdoidge.com



From nas at python.ca  Tue Feb 26 15:05:37 2002
From: nas at python.ca (Neil Schemenauer)
Date: Tue, 26 Feb 2002 12:05:37 -0800
Subject: Coredumps with Python 2.1.1-3
In-Reply-To: <20020226.185425.1884661237.5746@ic.ac.uk>; from p.mayers@ic.ac.uk on Tue, Feb 26, 2002 at 06:54:28PM +0000
References: <20020226.185425.1884661237.5746@ic.ac.uk>
Message-ID: <20020226120537.A27533@glacier.arctrix.com>

Phil Mayers wrote:
> Funnies - I'm using the psycopg database adaptor, *BUT* the crash is
> definitely happening in pure python code (isolated using sys.stderr.write
> and some sleeps).

You've probably got an extension module that is randomly corrupting
memory.

  Neil



From jkraska at san.rr.com  Wed Feb 20 00:08:56 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 20 Feb 2002 05:08:56 GMT
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org> <7xd6z07odl.fsf@ruckus.brouhaha.com>
Message-ID: <e6c67uonjofbhk711hqs58lhd1ocnqf669@4ax.com>

>They're spending quite a lot of $$ running those banner ads and I
>doubt if they have that many openings, so they must be looking for
>"perfect" candidates.  I'm sure they get quite a few responses.

Maybe, maybe not. There are a lot of highly imperfect ways
to get to a candidate's programming skills. Any kind of in-
house testing runs them afoul of federal labor laws, so
turning it upside down into a programming contest has appeal.
Likewise, you're pulling from a crowd of self-styled enthusiasts.
The approach has a big upside. They're likely awash in good
candidates. "Perfect"? One doesn't have to be perfect if one
is a motivated hacker. One is class-A material in this case,
a good deal of the time.

C//



From peter at engcorp.com  Thu Feb  7 20:05:51 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 07 Feb 2002 20:05:51 -0500
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <slrna650nr.ahr.philh@comuno.freeserve.co.uk> <a3v1fc$jmc$1@newshost.accu.uu.nl>
Message-ID: <3C63246F.135516E@engcorp.com>

Martijn Faassen wrote:
> 
> phil hunt <philh at comuno.freeserve.co.uk> wrote:
> > One thing that puzzles me is why Tim wants help: as he says, we're
> > all too stupid to understand his grand design (as well as being
> > arrogant).
> 
> > Since he is so much cleverer and far-seeing than everyone else, he'll
> > have to code it himself, as no-one else is clever enough to understand it.
> 
> No no no..while *some* of us are stupid enough to understand it, some of
> us would understand it if we only tried, and in fact are ignoring
> auto-coding as it's a threat. It's a psychological thing; we're not stupid,
> but we don't like to see auto-coding so we just ignore it. 

My god, it's infectious!  Martijn, did the PSU put you up to


From mal at lemburg.com  Tue Feb 26 05:06:45 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 26 Feb 2002 11:06:45 +0100
Subject: PEP 263 comments
References: <200202250520.g1P5KKD01484@mira.informatik.hu-berlin.de>
Message-ID: <3C7B5E35.129E5501@lemburg.com>

"Martin v. Loewis" wrote:
> 
> To make some progress on PEP 263, I suggest that some of the open issues
> are resolved as follows:

Thanks for the comments. I've update the PEP at SourceForge...
 
> - Comment syntax: I suggest to use the form
>   -*- coding: <coding name> -*-
>   Emacs already recognizes this syntax, as does patch #508973
>   on IDLEfork. The other proposed syntaxes should be removed from the
>   PEP.

+1
 
> - In addition, to simplify usage on Windows, Python recognizes the
>   UTF-8 file signature (e.g. as generated by notepad). Any file
>   starting with \xef\xbb\xbf is treated as being UTF-8; a coding
>   comment different from "utf-8" in such a file is an error.

+1
 
> - identifiers remain restricted to ASCII

+1
 
> - Implementation strategy: I believe the proposed strategy (change the
>   tokenizer) is overly complicated, and likely inefficient. Instead, I
>   suggest that the encoding directive applies only to Unicode literals.
>   It will still be formally an error if comments or string literals do
>   not follow the declared encoding, but the Python parser won't detect
>   this error.
> 
>   For use in Unicode literals, the parser will continue to work as it
>   does now, except that it applies the declared coding in compile.c.
>   To do so, PyUnicode_DecodeRawUnicodeEscape and
>   PyUnicode_DecodeUnicodeEscape will expect an additional flag
>   indicating whether they operate on a char* or a Py_UNICODE*.
> 
>   The only problem with this approach is that encodings where " or '
>   could be the second byte of a multi-byte character cannot be
>   supported as a source encoding. Python supports no such encoding
>   in the standard library at the moment, anyway, so this should not
>   be a problem.

I've added a two phase approach to the PEP: first we only
handle Unicode literals, then we do the whole file in a later
step.
 
> - Backwards compatibility: I'm in favour of leaving mostly everything
>   as-is, i.e. if there is no declared encoding, it should be possible
>   to put arbitrary bytes in string literals and comments; the proposed
>   implementation strategy supports that. However, I think that Unicode
>   literals which use the Latin-1 fallback should be deprecated, and that
>   the implementation should raise a DeprecationWarning: Anybody relying
>   on that feature should declare that the encoding is Latin-1.

Python will have to use Latin-1 as fallback encoding anyway,
so I don't think it's worth the trouble...
 
> - Changes to IDLE: When IDLE opens a file, it shall look for the UTF-8
>   signature. If no UTF-8 signature is found, it shall look for the
>   coding comment. If none is found, it shall apply the locale's
>   coding, which is determined as follows:
>   - on windows, it is "mbcs"
>   - on Unix, it is the one returned by nl_langinfo(CODESET)
>   Otherwise, it is the system default encoding.
> 
>   When saving a file, IDLE shall preserve the UTF-8 signature if there
>   was one. If not, and if there is a coding comment, that should be
>   used to encode the file. If there is none, the locale's encoding
>   should be used. If encoding fails (whether the coding was found in
>   the comment or in the locale), the file shall be UTF-8 encoded, and
>   an UTF-8 signature added.

I did not add the IDLE changes to the PEP. Please upload them
as feature request to SF.

Thanks,
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From me at mikerobin.com  Thu Feb  7 16:08:14 2002
From: me at mikerobin.com (Michael Robin)
Date: 7 Feb 2002 13:08:14 -0800
Subject: wxPython book
References: <88bc63c6.0202040505.2a9c3016@posting.google.com> <cb4ba455.0202060441.54e1d529@posting.google.com> <T%u88.94053$KM2.3390206@atlpnn01.usenetserver.com>
Message-ID: <52e5ab5f.0202071308.34ec94cc@posting.google.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message news:<T%u88.94053> 
> The thing I find most frustrating (and have just aded a FAQ question about
> on the Wiki) is the lack of a connection between my intuition and the way
> sizers work. I have licensed a copy of wxDesigner just to try and get over
> this hurdle, but dammit I *still* can't put interfaces together that work
> the way I expect them to under window resizing.

I have the same problem. It doesn't help that wxDesigner does sizers
but doesn't do anchors via layout constraints - and Boa does anchors
but doesn't do sizers, so you can't really experiment/maintain at the
designer level with one tool. (Is it true that anchors are going away?
- I remember seeing in the wx pages that they were being deprecated.
They seem to be simple and give predictable results vis-a-vis heavily
nested sizers.)

On a related note, can a wxPython/wxDesigner user help me with the
below?
(Will crosssend to wxpython-users at lists.wxwindows.org.)

thanks,
mike

---------------------------------------------------
I can't seem to find the wxDesigner tool 
to turn on parameter 2 ('option') for wxAddwindow to 
make the item stretchable along the sizer axis.

For example, change:
    item0.AddWindow( item4, 0, ...)
to
    item0.AddWindow( item4, 1, ...)

Also, if anyone has any hints/tips on using wxdesigner in practice,
I'd appreciate it. me at mikerobin.com


From mark0ta at gmx.de  Tue Feb 12 17:36:39 2002
From: mark0ta at gmx.de (marko)
Date: 12 Feb 2002 14:36:39 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org>
Message-ID: <7c696d2d.0202121436.4950e979@posting.google.com>

Chris,

I use this 'COM Up!' tool when playing with COM/ActiveX.
Hope it helps you
http://www.bitwheel-technologies.com/products/comup/
http://www.bitwheel-technologies.com/products/comup/comup!_eval.zip



"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1013487810.369.python-list at python.org>...
> I am developing a headache messing with an OCX I picked up from the
> Internet.  It seems that my forays into ActiveX never pan out; I should
> stick with Linux.
> 
> The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> have to be such a pain to find?)  I can instantiate it:
> 
> >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> >>>
> 
> (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
> 
> >>> o.AboutBox()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py", line
> 29, in AboutBox
>     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
> com_error: (-2147418113, 'Unexpected failure', None, None)
> >>>
> 
> Any method or property, same answer.
> 
> The URL for the control is:
> 
>     http://www.qwerks.com/download/3693/report.zip
> 
> It is supposed to be a general printing control, but nothing works.
> 
> Chris Gonnerman -- chris.gonnerman at newcenturycomputers.net
> http://newcenturycomputers.net


From donn at u.washington.edu  Wed Feb 27 17:09:32 2002
From: donn at u.washington.edu (Donn Cave)
Date: 27 Feb 2002 22:09:32 GMT
Subject: kill childs
References: <a5j1p9$ftb$1@news.rz.uni-karlsruhe.de> <m3sn7misch.fsf@mira.informatik.hu-berlin.de>
Message-ID: <a5jles$317u$1@nntp6.u.washington.edu>

Quoth martin at v.loewis.de (Martin v. Loewis):
| "Christian Schmidt" <christian.schmidt at REMOVEaifb.uni-karlsruhe.de> writes:

|> I use Python to start other commands via os.system("cmdline")
|> When I kill the script with 'kill <scriptpid>' the command is not killed,
|> only the script.
|> How can I kill the child, too?
|> Can I use something like os.kill(-1) in an except-statement?

| If you use os.system, you cannot find out what the pid of the child
| is. So as a prerequisite, you need to replace the call to system()
| with a fork/exec sequence. Then, you can install a signal handler
| (using signal.signal) to kill the subprocesses if any.

He's aiming at the process group, which is a good idea.  os.kill(-1, sig)
will work, though of course the signal will be delivered to the calling
process as well.  It might be a good idea to set a flag indicating that
the process group kill has already happened, to avoid an infinite loop.

If for some reason a PID is needed, os.spawnv(os.P_NOWAIT, ...) returns it.

	Donn Cave, donn at u.washington.edu


From ark at research.att.com  Mon Feb 11 12:25:44 2002
From: ark at research.att.com (Andrew Koenig)
Date: Mon, 11 Feb 2002 17:25:44 GMT
Subject: Question on if, while, for....  Figured it out..
References: <e959e7b.0202110901.485da0b0@posting.google.com>
Message-ID: <yu99y9i03p7r.fsf@europa.research.att.com>

Fred> OK, I figured it out (one other way at least).

Does it detect that 121/143 is really 11/13?

-- 
Andrew Koenig, ark at research.att.com, http://www.research.att.com/info/ark


From pinard at iro.umontreal.ca  Mon Feb 18 22:00:55 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 18 Feb 2002 22:00:55 -0500
Subject: Guido wins FSF Award
In-Reply-To: <3c72aa10.15827358@127.0.0.1>
References: <Pine.LNX.4.32.0202171452390.1394-100000@aesaeion.com>
	<mailman.1014004828.14630.python-list@python.org>
	<3c72aa10.15827358@127.0.0.1>
Message-ID: <oqn0y65g60.fsf@titan.progiciels-bpi.ca>

[(Five Fresh) Fish]
> [chris.gonnerman at newcenturycomputers.net]

> > [...] and as far as I am concerned, Vorlons are welcome here.

> >"All Vorlons are Kosh"...

> Maybe we should all be Kosh!  :-)

This has not been over-stressed in the dialogs, but there are many Python
programmers living on Babylon 5.  Some of you might remember that episode
where Kosh got out of his suit once in public, to save the life of the
station commander who fell out of the transportation device (my English
vocabulary is severely lacking).  Everybody saw something different, but
surely for the Python guys, Kosh was like a phosphorescent, glowing Guido!

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



From roccomoretti at netscape.net  Wed Feb 13 23:07:14 2002
From: roccomoretti at netscape.net (Rocco Moretti)
Date: 13 Feb 2002 20:07:14 -0800
Subject: Python License Issues
References: <mailman.1013627905.10961.python-list@python.org>
Message-ID: <ff133276.0202132007.61937647@posting.google.com>

christoforou at gaaros.com wrote in message news:<mailman.1013627905.10961.python-list at python.org>...

> Some clients demand a certain level of 'protection' from situations of gross 
> misconduct, say python has stolen major chunks of code from other apps and 
> thus exposing them to huge legal risks. We can consider accepting liability 
> on all third party code we use, for such gross misconduct situations but is 
> it technically possible to do so given the PSF license? Ideally we would 
> like to negotiate a single license agreement with these clients (our license 
> agreement) which provides additional liability over and above to that which 
> PSF provides. However given clause 4 it seems that the client *must* 
> directly be bound by the PSF license. 
> 
> Any comments? 

Yes: What are you trying to get out of asking
comp.lang.python/python-list? If you have legal questions, it is best
to refer them to a lawyer who is certified to practice law in your
jurisdiction. This is doubly important as you are in London, England,
UK, and the PSF is based out of (Delaware? Virginia?) USA. The
international implications of the licensing are not trivial and
require the expertise of a professional. (i.e. If you're really
concerned about this, don't hesitate on shelling out a retainer for a
good lawyer.)

Regarding Python stealing major chunks of code: Is this a hypothetical
argument, or has this charge been leveled at Python by your clients?

# Addressing everyone now

On a somewhat tangential note, I realize that most of the PythonLabs
team at this point says "Done is good!" with the Python license and
wish to wash their hands of it, but the ~70 line contract with the
four different copyright owners and different licensing clauses for
each makes me a little queasy, even with assurances that it is "GPL
compatible."  (Legalese makes me nervous.)

>From what I understand, the copyright owners can change the licensing
terms at will, so if the PSF has the copyright to *all* of the code in
Python, they could replace the four section license with a single
section license like a BSD style license or even with (just) the
current PSF section and a listing of the Copyright: Year-Year Holder
list at the top. (Aside: Why *does* the PSF use an 8 subpoint license
instead of a BSD style? Is it strictly to maintain compatibility with
the BeOpen and CNRI sections?)

So who actually holds the copyright for all the bits of Python?

>From what I understand, BeOpen is no longer in business. Has the PSF
considered obtaining the copyright for the portions of code written at
BeOpen from whoever got it from the fire sale?
Although Stichting Mathematisch Centrum still exists, would the PSF
consider approaching them about a transfer of the Python copyrights?
CNRI probably is the toughest sell, as they still exist and have
concerns about liability. Would it be possible to negotiate some sort
of contract with them where the PSF assumes all liability for Python
versions other than 1.6.z, and CRNI keeps perpetual rights to the
1.6.z codebase?

I'm not really asking about if it is legally possible, but more so to
see if the PSF would be willing to take on such a task.

"Simple is better than complex"-applies-to-licenses-too-ly'rs 

-Rocco


From gcordova at hebmex.com  Thu Feb 21 12:54:22 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Thu, 21 Feb 2002 11:54:22 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <19650FF08A99D511B19E00B0D0F06323060ACB@SRM201002>

> 
> Anyway, =/== is a dead horse--didn't we agree that using := for assigments
> in expressions totally fixes it?  So there's no reason to keep coming
> back to it.
> 

Er, no, we didn't :-)

_YOU_ and _I_ agreed that it's more readable and less
error prone than simply using "=" in an expression.
That's it. Don't include me in your revolution.

> > Also, putting multiple conceptual things on one line isn't good for code
> > clarity.
> 
> That depends on your style and on the context.
> 

I agree with the earlier statement.

> 
> If you want Python to catch more typos for you,
> you're better off asking for Python to check for
> variable declarations (I don't mean static types,
> I just mean "local x") and flag undeclared variables
> like you can with perl's "use strict" option.
> That will prevent a lot more bugs than := is likely to cause.
> 

Now this is just plain nasty.

> 
> And it fills more space on the screen, crowding
> out other code, so you can't see as much of the
> program flow.  I like the short version better.
>

So, get a larger monitor, or use a smaller font,
or make the editor window larger, or change
your text mode (if using text mode), or ...

>
> > But in the end, neither of our subjective opinions
> > matter - Guido's does. That's probably why this
> > language is so great.
> 
> Guido doesn't write our code; we do.

That's all nice and useful. I'm not defending
Guido, but hey, let's keep this civil, ok?

*If* it weren't for Guido, we/you wouldn't *have*
Python to hack on.

Second, Guido's a pretty smart person. So maybe
he did give quite a bit of thought about the
expression/asignment thing.

Now, I have an idea which is at the back of my
mind. I don't know if it's worth anything, it's
my own hypothesis, it may be totally bogus, on
the other hand, it may be real and we don't know
about it. So here it is, for your perusal and
entertainment.

"Language" is the expression of a person's thoughts,
and thoughts and ideas expressed in a Language tend
to reflect the structure of that persons thoughts.

On the other hand, the inverse process also happens:
"Language" tends to impose, with time, it's structure
and semantics upon it's users mind. Like a kind of
feedback. The Object modifying the Observer.

One thing which is quite delightful about Python is
it's high-quality, simple-to-use, and quite powerful,
included library. Even if lots of modules have "amateur"
origins, it's of very high quality, and usability.

So. If the Language (Python) changes, in a way that
it doesn't impose a Neat, Orderly, Clean, programming
style, then I really believe that the Language will
--passively-- encourage sloppy code.

There.

That's the stone in my beans (er, mexican saying :-)

I love Python because it steadily forces upon me
some rules about writing neat, readable, correct
code.

I liked C because it let me get away with some
sloppy and hideous stuff.

Hmmm... no need to mention assembler. :-)

On the other hand, I may be totally, utterly
wrong; but it won't be the first or last time :-)

Thursday!!

-gustavo



From emile at fenx.com  Tue Feb 12 19:14:56 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 12 Feb 2002 16:14:56 -0800
Subject: Why no documentation of basic data structure performance?
References: <eppstein-0310EF.13300812022002@news.service.uci.edu>
Message-ID: <a4cbfq$1e2751$1@ID-11957.news.dfncis.de>

"David Eppstein"
> Does anyone else notice a serious lack of timing information in the
Python
> documentation?

[snippity-snippity]

For you it's timing information... Me, I notice a serious lack of run
time memory usage introspection/profiling...  ;-)
Why it's so slow I can test, why it's so big is a guess...


Time-to-take-out-the-thrash-ly y'rs,

--

Emile van Sebille
emile at fenx.com

---------



From amuys at shortech.com.au  Sun Feb 10 20:18:13 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 10 Feb 2002 17:18:13 -0800
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com>
Message-ID: <7934d084.0202101718.404404d1@posting.google.com>

rony.steelandt at bucodi.com (Rony) wrote in message news:<b34d3835.0202080025.1d69d198 at posting.google.com>...
> YOU SHOULD BE ASHAME !!!!!!!!!!!!!!!!!!!!!
Ashamed?  No, indeed he should be applauded for his stand for signal
over noise!
 
> Who on earth you think you are, to just post a message on usenet
> saying don't talk to that guy
>
> 1. Nobody or nothing gives you that right.
> 2. It's a attack to peoples intelegence here, i assume they very well
> can decide by themself if they want to dicusse or not
> 3. It's against the spirit of usenet
> 4. You are not God to decide this
> 5 ....

(referencing <4egfvd$2pl2 at news-s01.ny.us.ibm.net>#1/1 from
compl.lang.rexx in 1996)

A quick grep through the deja archives indicates you have been on
usenet long enough to know that what you're spouting here is absolute
nonsense.  Everyone, by virtue of their involvement in a newsgroup
community, has the right to defend that group's signal-noise
ratio---you have been on usenet too long to claim ignorance of this. 
It is not an attack on anyone's intellegence, rather an appeal to the
same, a request that they step back and reconsider the value of the
(now rediculously extended) cascade they are contributing to.  That
they pause, before automatically replying to the troll in our midst,
and consider the effect of this threads continued existence on c.l.py
itself.

It appalls me that someone who has been involved in usenet for so
long, would so willfully and deliberately, disregard the nature and
culture of usenet communities in this way.  To appeal to the 'spirit
of usenet' in defense of such a position maybe understandable, if
mistaken, in a newbie making their first post, but for someone who
appears to have been active here for over 5 years?

The spontaneous, self-organising, emergent communities of people
formed around this asyncronous text medium is the spirit of usenet you
so glibly refer to.  The reaction of this community (by flame, by
appeal, and eventually by ostracising) to defend itself against the
parasitic troll, spammer, or pathological cross-poster, is a
legitimate manifestation of this spirit.  An appeal by active
participants of a newsgroup for silence, for signal over noise, is
likely the purest expresion of the 'spirit of usenet' I can imagine.

Andrae Muys


From m.andrzejewski at dmz.com.pl  Mon Feb  4 04:26:21 2002
From: m.andrzejewski at dmz.com.pl (marcin andrzejewski)
Date: Mon, 4 Feb 2002 10:26:21 +0100
Subject: module from BLOB
Message-ID: <a3lk3g$fcs$1@news.tpi.pl>

is it possible to import module directly from database BLOB/CLOB field ?

--
marcin andrzejewski
m.andrzejewski at dmz.com.pl
keson at post.pl




From jeff at ccvcorp.com  Mon Feb  4 14:57:54 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 11:57:54 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com><Wls68.14178$om6.357057@news1.tin.it> <mailman.1012590723.7301.python-list@python.org> <a3kb3p$se44@news1.gtech.com>
Message-ID: <3C5EE7C2.4DF4A885@ccvcorp.com>


Peter Milliken wrote:

> G'Day Cliff,
>
> "Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
> news:mailman.1012590723.7301.python-list at python.org...
> > On Fri, 01 Feb 2002 08:29:42 GMT
>
> > I much prefer the "continuous ongoing code review" idea better than a
> > formalized procedure (if I understand what you mean by it correctly).
> > Working in pairs provides this as a side-effect rather than having a
> formal
> > procedure for code-review and seems less time-consuming overall.
> >
>
> Not sure what you mean by "continuous ongoing code review". I used to be a
> "team leader",  .....

I hesitate to speak for others, but...   I'm going to stomp right on in anyhow.
:)

I believe that the "continuous ongoing code review" being mentioned is in
reference to the Extreme Programming practice of pair programming -- all code is
written jointly by two people, sitting at the same computer.  Thus, everything
is, in effect, being reviewed by the other person as it's being written.  This
should (in principle ;) ) greatly reduce the need for formal group reviews...
though I expect they'd probably still be desirable regardless.


> ..... So I assume that "continuous ongoing code review" is similar to
> my idea of "early and often"? :-)

That's more or less my interpretation of it, too.  :)

Jeff Shannon
Technician/Programmer
Credit International




From joekm at earthlink.net  Wed Feb 27 21:34:06 2002
From: joekm at earthlink.net (Joe)
Date: Thu, 28 Feb 2002 02:34:06 GMT
Subject: Stumped by PhotoImage issue -- Think I've got it now
References: <3c7bcefb.501573@news.earthlink.net>
Message-ID: <3c7d92d4.351007@news.earthlink.net>

Managed to solve the problem described below by tying the "OK" button
to a separate function in the "About" class 
[ i.e. def buttonAction(self, event=NONE)]

Anyway, once I did that, the PhotoImage generated labels would
display.  I'm guessing that when Tk() finishes up, PhotoImage releases
its gifs.  What I was lead to believe on the net was that they needed
to be defined in a persistant way.  So I was trying to make them class
variables (self.keepMeAround], declaring them global, defining them
outside of the class then passing them to it and then redefining them
in the __init__ method, and other similar approaches.

The only clue other than that was discovering that if I broke the
button on purpose, the gifs would display (but the button would not).
So, not really thinking it would work I made this change:

instead of:

self.OK = Button(yada yada, command=AboutDialog.destroy)

I did this:

self.OK = Button[yada yada, command=self.buttonAction]

and then wrote a short function to destroy the About dialog window.


I'm guessing that this keeps a pending operation on the Tk() widget
which, in turn, keeps PhotoImage from loosing track of the gifs.  As
it happens, all the other times I've used PhotoImage in this program
the buttons all had functions defined separately.


......Anyway, that worked but if any of you programming experts out
there can add further clarification I'd appreciate it.  

On Tue, 26 Feb 2002 18:22:11 GMT, joekm at earthlink.net (Joe) wrote:

>I've actually seen that there are others with a similar problem that
>have posted on the web.  I've read G.R's FAQ and looked at the
>solutions others have used but I am still confused.....
>
>In a program that I am writing, I have an "about" dialog box that is
>written as a class that is in turned called from the "help : about"
>pulldown menu item in my "menubar" class.  This "about" dialog has
>three .gif images that are handled by PhotoImage.  The problem is that
>they appear to be displaying transparant.
>
>Now, if I pull the class out of my main program and wrap a minimalist
>mainloop around it, everything works fine.  Also, if I intentionally
>put an error in the part of the class that draws the button, the gif's
>are drawn without a problem (but I get an error message and the button
>does not draw).  
>
>I've tried setting instance variables, global variables, etc. and I'm
>running out of ideas.  Any suggestions would be appreciated.  Just so
>you know, this is my first original python program and my first real
>intro into object oriented programming.  OTOH, I have read everything
>I could download and purchased a couple of books on the subject
>including "Python and Tkinter Programming".
>
>..and no, I do not wish to use PMW at this time, I wish to gain
>experience with Tkinter first.
>
>The class follows (needs to be un-word-wrapped):
>
>class About:
>
>    def __init__(self, parent, GraphicsPath):
>
>        self.parent = parent
>
>        self.deltaLogo = PhotoImage(file=GraphicsPath+'DeltaLogo.gif')
>        self.pythonLogo =
>PhotoImage(file=GraphicsPath+'PythonPowered.gif')
>        self.tkLogo = PhotoImage(file=GraphicsPath+'TCL-TK_logo.gif')
>
>        AboutDialog = Toplevel(self.parent, relief=RAISED,
>borderwidth=3)
>        AboutDialog.geometry('+300+150')
>        AboutDialog.overrideredirect(1)
>        AboutDialog.tkraise()
>        self.AboutBorder = Frame(AboutDialog, relief=GROOVE,
>borderwidth=2)
>
>        title = Label(self.AboutBorder, text='Simple Airframe Fracture
>Evaluation Tool [SAFE-T]',
>                             font=('Ariel', 11, 'bold'))
>        title.pack(side=TOP, padx=2, pady=5)
>        
>        global Revision_Type, Revision
>        version = Label(self.AboutBorder, text=Revision_Type+'
>'+Revision)
>        version.pack(pady=5)
>        
>        copyright1 = Label(self.AboutBorder, text='Copyright 200X by
>Joseph K. Mooney and')
>        copyright1.pack(pady=5)
>
>        self.delta = Label(self.AboutBorder, image=self.deltaLogo)
>        self.delta.pack() 
>
>        copyright2 = Label(self.AboutBorder, text='ALL RIGHTS
>RESERVED\nSee License Agreement for Further Information\n')
>        copyright2.pack()
>        
>        description = Label(self.AboutBorder, text="Programed in
>PYTHON - GUI  built with Tkinter")
>        description.pack(pady=8)
>
>        self.python = Label(self.AboutBorder, image=self.pythonLogo)
>        self.python.pack(side=LEFT) 
>
>        self.tclTk = Label(self.AboutBorder, image=self.tkLogo)
>        self.tclTk.pack(side=RIGHT) 
>
>        self.AboutBorder.pack(fill=X, expand=YES)
>
>        self.OK = Button(self.AboutBorder, text='OK', font=('Arial',
>9, 'bold'), command=AboutDialog.destroy)
>        self.OK.place(relx=.48, rely=.8, anchor=NW)
>#



From weeks at vitus.scs.agilent.com  Mon Feb  4 14:42:21 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Mon, 4 Feb 2002 19:42:21 +0000 (UTC)
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <slrna5s3d6.h88.grante@isis.visi.com>
Message-ID: <1012851741.725670@cswreg.cos.agilent.com>

Grant Edwards (grante at visi.com) wrote:
: Everything in Python is passed by reference.  Always.

True.  But it is equally true that:

    Nothing in Python is passed by reference.  Never.

"passed by reference" is sufficiently ambiguous to cover both statements.

Once upon a time, "passed by reference" meant that a procedure call

    f(x)

passed the address of the variable x.  Nowadays, it sometimes means merely
that the value of x is an address.

Being old, I don't like the new vocabulary.  I like to say this, for
example:

    In Python, all data are represented internally as addresses, and these
    are passed by value.

Others might say:

    In Python, all data are represented internally by blocks of memory, and
    it is the addresses of these that are passed (by value):

Or:

    In Python, all data are represented internally by blocks of memory, and
    these are passed by reference.

They all mean the same thing.


Greg


From phr-n2002a at nightsong.com  Tue Feb  5 16:42:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 13:42:04 -0800
Subject: Random 32Bit integer was: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de> <mailman.1012902609.27332.python-list@python.org> <3C5FB52A.20902@thomas-guettler.de>
Message-ID: <7x7kprwqqb.fsf@ruckus.brouhaha.com>

Thomas Guettler <st-newsgroups at thomas-guettler.de> writes:
> This would give me a long. But I need a 32Bit Integer.
> Unfortunately there is no sys.minint.
> 
> OK, it is not that important for me. But it would be nice to know
> the answer (How to get a radom 32 Bit integer)

The largest 32-bit integer (sys.maxint on a 32 bit system) is 2**31-1.
The smallest integer is -(2**31).


From weeks at vitus.scs.agilent.com  Thu Feb  7 13:19:30 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Thu, 7 Feb 2002 18:19:30 +0000 (UTC)
Subject: threads killing other threads
Message-ID: <1013105970.135322@cswreg.cos.agilent.com>

I've occasionally wanted thread objects to have an exit() method, so that
one thread could interrrupt and kill another.  This doesn't seem to be 
available, perhaps because POSIX threads don't support it.

Is that correct?

Is there some reason why threads killing threads would be undesirable or
hard to implement?


Greg


From jason at jorendorff.com  Fri Feb 15 12:34:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 15 Feb 2002 11:34:02 -0600
Subject: help with silly algorhythm
In-Reply-To: <5e8bd451.0202142155.143ca4d9@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAEMODCAA.jason@jorendorff.com>

Kevin Parks writes:
> I realize that this is very odd and convoluted. If anyone can make
> heads or tails of this explanation and would like to help me I
> would be grateful.

This might be more readable than the previous poster's answers.
It should be faster, too, especially for large 'partners' lists.
You could do even better if you could guarantee that 'values'
is sorted too.

## Jason Orendorff    http://www.jorendorff.com/



# sillyalg.py - Silly algorithm

import bisect

values = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
partners = [1, 3, 7, 10]


# Mode 1
def upper(list, item):
    """ Return the least element greater than item in list.
    The list must be sorted!

    If item is greater than all items in 'list', then wrap around and
    return the least item.
    """
    index = bisect.bisect_right(list, item)
    if index == len(list):
        index = 0
    return list[index]

print "Mode 1"
print [[i, upper(partners, i)] for i in values]
print [upper(partners, i) for i in values]


# Mode 2
def lower(list, item):
    """ Analogous to upper(). """
    index = bisect.bisect_left(list, item)
    if index == 0:
        index = len(list)
    return list[index - 1]

print "Mode 2"
print [[i, lower(partners, i)] for i in values]
print [lower(partners, i) for i in values]


# Mode 3
def even_upper(list, item, n):
    functions = (upper, lower)
    fn = functions[n%2]  # choose which function
    return fn(list, item)

print "Mode 3"
print [[values[n], even_upper(partners, values[n], n)]
       for n in range(len(values))]
print [even_upper(partners, values[n], n) for n in range(len(values))]





From jeff at ccvcorp.com  Thu Feb 21 17:09:47 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 14:09:47 -0800
Subject: Hosting
References: <mailman.1014315035.1046.python-list@python.org>
Message-ID: <3C75702B.DD007357@ccvcorp.com>


Steven Majewski wrote:

> On 21 Feb 2002, Kevin Dahlhausen wrote:
>
> > That's actually not such a bad idea.  I'm only paying $10/month for
> > dialup, and looking to pay about the same for hosting.   Does DSL
> > require you to continue phone service once it is installed?  That
> > would make up the difference and it would be affordable then.
> >
> > > With the money you pay to ISP, why don't you get DSL with static IP, and
> > > run your own server?
>
> Prices will vary, but where I am you'll pay a LOT more for DSL
> service with a static IP than $10/month. However, sometimes you
> can get a package plan ( w/ phone service and a bunch of extras:
> caller ID, etc. ) so it might make sense -- you'll probably still
> pay more but the connection is worth it.

My DSL costs me approx. $50 / month, plus bandwidth charges if I exceed a
certain amount, compared to ~$20 / month for typical dialup packages in this
area, though some dialup goes as low as $10, and some DSL around $40.  Part of
my extra cost is because I required an external DSL modem (to be able to connect
multiple computers without fiddling with MS's internet sharing mess) and a
static IP (actually I get two, though I only use one for my router).  In theory
I could host a low-traffic server, but if I was doing anything more than a small
personal page, I'd probably end up paying more for the extra bandwidth than just
dropping the $10-20 for a hosting service (and then I'd have to deal with system
maintenance for the server, security updates, etc...)

(Server issues aside, after a few months of DSL the thought of going back to
dialup is actively painful. ;)  And really, I pay less than many people pay for
cable TV, and I enjoy the DSL a lot more than I would the TV...)


> The other thing is that you still might end up paying an ISP :
>  For DSL service, it was a package provided by Sprint (my local
> phone Co.) and Earthlink, so you HAD to get an account with
> Earthlink. ( Somebody told me that since Sprint doesn't know
> what they're doing with DSL, that it's really a marketing arrangement
> with Earthlink, who really provide most of the "service" )
> With my ISDN, I'm paying Sprint for the ISDN lines and I'm paying
> our local ISP (cstone.net) for an ISDN to internet connection on the
> other end.

There's always two parts of DSL service, the physical link and the ISP.  Some
phone companies offer both;  some places, you can get the physical link through
someone other than your phone company (but it's usually more expensive to do
so).  I deliberately chose to have an ISP (northwest link) *other* than my phone
company (which is Qwest) because I don't trust the phone company to know how to
run an ISP properly.  :)


>  For small to moderate usage, I think an ISP is always going to
> be cheaper than paying for a thick pipe and doing it yourself --
> but if you want the thick pipe anyway, then it might be worth
> setting up your own server. ( But if you do anything "commercial"
> with the server, they won't want to let you use any of the
> end-user oriented packages and pricing! )

Agreed.  In general, if you're wanting anything more than a "toy" server, you're
going to be better off paying for a commercial host than messing with it
yourself.  But if you've got the pipe, and won't have that much traffic, you
*can* do it yourself if you want to.  (If you've got the tools and the time, you
can fix your own car, too, but isn't it usually easier to take it to a
professional mechanic?)

Jeff Shannon
Technician/Programmer
Credit International




From 3dsphere at home.com  Fri Feb 15 19:33:26 2002
From: 3dsphere at home.com (Kris Botha)
Date: Fri, 15 Feb 2002 16:33:26 -0800
Subject: impossible to reboot OS in python?
Message-ID: <a4k9as$ak8$1@nntp.itservices.ubc.ca>

I heard it was impossible to induce a reboot in python as the os module does
not cater for that. Is this true?

I know in DOS you could just set a hex value at the correct address to
induce a cold or warm reboot, but this won't work for windows machines.

Does it depend on the os installed?

cheers,
kris




From kalath at lycos.com  Thu Feb 21 09:27:27 2002
From: kalath at lycos.com (Mark Brady)
Date: 21 Feb 2002 06:27:27 -0800
Subject: Interesting for Trekkers and Python Developers ... ST: Bridge Commander
References: <6c92a8bc.0202210257.702ee9d4@posting.google.com>
Message-ID: <e840346c.0202210627.6cc55458@posting.google.com>

tiberius at ms28.hinet.net (Tiberius Teng) wrote:
> http://www.bridgecommander.com/
>
> It's very interesting that looks like this game uses extensive Python
> scripts in its control - AI, ships, weapons, control panels, even
> mission progress (story ?!). Check the 'scripts' folder after
> installation and you'll find out lots of familiar filenames and folder
> structures ;)
> 

Great stuff, thanks for the info. As I'm sure most other Python gamers
know, Severance is another game that uses Python and of course there's
EVE (http://www.eve-online.com/) which uses Stackless. Anybody know of
any other commercial games which use Python?

Marcus.


From mhuening at zedat.fu-berlin.de  Mon Feb 11 04:58:07 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 11 Feb 2002 09:58:07 GMT
Subject: No explanation for weird behavior in re module!
References: <mailman.1013386937.27383.python-list@python.org>
Message-ID: <Xns91B26FC9941CEmhueningzedatfuberli@130.133.1.4>

"Tim Peters" <tim.one at home.com> wrote in
news:mailman.1013386937.27383.python-list at python.org: 

> 
>>      Other than the fact that 'T?r' has the '?' unicode charcater, I
>>      fail 
>> to see any difference! 
> 
> Heh.  Leaving this joy to someone else <wink>.
> 

Okay, I'll try...
If your string comes in 'Latin-1' you will have to tell Python to treat it 
as Unicode. And when you want to print it afterwards, you'll have to 
encode it as 'Latin-1'.

---
import re
txt = 'die T?r, T?ren'
txt = unicode(txt, 'latin-1')
pattern = re.compile(ur'(der|die|das)\s+(\w+)', re.UNICODE)

match = pattern.match(txt)
article = match.group(1)
noun = match.group(2)

print article.encode('latin-1')
print noun.encode('latin-1')
---

Hope this helps.
Matthias


From shredwheat at attbi.com  Fri Feb 15 11:32:33 2002
From: shredwheat at attbi.com (Pete Shinners)
Date: Fri, 15 Feb 2002 16:32:33 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131> <3C6CAFF8.4D3B5092@dotcom.ee>
Message-ID: <3C6D3963.7060102@attbi.com>

>I've been learning JAVA for a while, but now I've heard avout Python.
>Any coments concerning these two languages?

from my experiences..

java's strengths are academia and marketing
python's strength is real world coding

for me, one gets the job done. one looks good on glossy brochures.




From amuys at shortech.com.au  Wed Feb 27 01:58:25 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 26 Feb 2002 22:58:25 -0800
Subject: Oddity question
References: <mailman.1014745204.15257.python-list@python.org>
Message-ID: <7934d084.0202262258.53fbe36@posting.google.com>

Mariano Corsunsky <core.lists.python at core-sdi.com> wrote in message news:<mailman.1014745204.15257.python-list at python.org>...
> Hi there!
> 
> I was fooling around with sockets last night, using a thread to deal 
> with the read part and another method to do the print.
> 
> like this:
> 
> def thread read (self):
>   while 1:
>     self.buf=self.buf+self.my socket.recv (8)
> 
> def print buf(self):
>   if buf:
>     print "bufeer-->",buf
>     buf='""
> 
> But, what happened to me was that the -buf = ""- line seemed to be not 
> working because I got outputs like this
> 
> buffer->a
> buffer->ab
> buffer->abpleasework
> buffer->abpleaseworkpleaseflush!
> 
> So, I dont know why I changed thread read function to:
> 
>   while 1:
>     aux=self.my socket.recv (8)
>     self.buf=self.buf+aux
> 
> And this version worked properly..... but I still do not know why. To 
> me, I've only split the line in two, using an auiliar variable... but 
> THAT made the trick! 

<aside>
Heh, I seem to recall discussing the dangers of newbies messing with
threads without an understanding of concurrency just last week :)
</aside>

As I aluded above, the problem is that you haven't considered how your
two threads are interacting.  As far as I can tell you have two
threads, a producer (reading from network->buffer) and a consumer
(reading from buffer->stdout).

You must remember that the threads run *concurrently* this means that
the atomic operations that make up the statements in each thread are
interleaved in a non-deterministic fashion (or rather, any reliance on
a given order is a bug).

The code you're having problems with is appears to be effectively the
following...

thread1:
    while 1:
        buf = buf + socket.recv(8)

thread2:
    while 1:
        print buf
        buf = ""

Now when discussing threading bugs, I'm perfectly entitled to order
these operations anyway I like.  In this case I will order them
understanding that the socket.recv() call is likely to trigger a
schedule(); assignment, concatination is not; and print may or maynot
trigger depending on how agressive stdout is buffered.

However do remember that *ANY* interleving is valid, so your code
*MUST* behave correctly for *EVERY* possible ordering of instructions
between threads!

On the otherhand the instructions inside each thread execute in a
known order (well most of the time ;).  So the challange is to use the
known ordering of instructions inside each thread to impose an
ordering (where required) on the instructions *between* each thread.

But first a look at your code (so you can understand what you have
seen).

Consider

buf = buf + socket.recv(8)

This is three independent python operations, use dis to uncover how
many CPython instructions, and gdb can show you how many assembly
instructions, not all of which are atomic themselves.  For now we will
only concern ourselves with python operations/instructions (however
for correct operation, your code *MUST* handle the problem of
non-atomic assembly instruction execution)

Let's consider the following code:

>>> def read():
	return "World"

>>> def concat(a, b):
	return a + b

>>> def test():
	buf = "Hello "
	print concat(buf, read())

	
>>> test()
Hello World
>>> import dis
>>> dis.dis(test)
          0 SET_LINENO               1

          3 SET_LINENO               2
          6 LOAD_CONST               1 ('Hello ')
          9 STORE_FAST               0 (buf)

         12 SET_LINENO               3
         15 LOAD_GLOBAL              1 (concat)
         18 LOAD_FAST                0 (buf)
         21 LOAD_GLOBAL              2 (read)
         24 CALL_FUNCTION            0
         27 CALL_FUNCTION            2
         30 PRINT_ITEM          
         31 PRINT_NEWLINE       
         32 LOAD_CONST               0 (None)
         35 RETURN_VALUE        
>>> 

Note the order of the instructions @18->27.  Specifically that buf is
loaded @18, then read() is called @24, then concat is called @27.

So your analogous code is correctly ordered:

tmp1 = buf             # @18
tmp2 = socket.recv(8)  # @24
tmp3 = tmp1 + tmp2     # @27
buf = tmp3             # @35 (return vs. assign, but analogous [see @6
& @9])

(note tmp variables are references to the stack, so require no
operation to access)

For now we will treat these as atomic (I believe this is the case atm,
although the language spec provides no guarantee to that effect).

In a similar manner thread2's code:

>>> def blank():
	buf = "Hello World"
	print buf
	buf = ""

	
>>> blank()
Hello World
>>> dis.dis(blank)
          0 SET_LINENO               1

          3 SET_LINENO               2
          6 LOAD_CONST               1 ('Hello World')
          9 STORE_FAST               0 (buf)

         12 SET_LINENO               3
         15 LOAD_FAST                0 (buf)
         18 PRINT_ITEM          
         19 PRINT_NEWLINE       

         20 SET_LINENO               4
         23 LOAD_CONST               2 ('')
         26 STORE_FAST               0 (buf)
         29 LOAD_CONST               0 (None)
         32 RETURN_VALUE        
>>> 

So ordered:

tmp4 = buf  # @15
print tmp4  # @18
tmp5 = ""   # @23
buf = tmp5  # @26

So we have to interleave the following operations:

(A1) tmp1 = buf              (B1) tmp4 = buf
(A2) tmp2 = socket.recv(8)   (B2) print tmp4
(A3) tmp3 = tmp1 + tmp2      (B3) tmp5 = ""
(A4) buf = tmp3              (B4) buf = tmp5

remembering that A2 will almost certianly trigger a schedule().

Consider the ordering:
    A1,A2,B1,B2,B3,B4,A3,A4,repeat
Assuming buf starts equal to "", and do three loops:
first-loop:
tmp1 = buf              # buf == ""; tmp1 == ""; tmp2 == "";
tmp2 = socket.recv(8)   # buf == ""; tmp1 == ""; tmp2 == "foo";
tmp4 = buf              # buf == ""; tmp1 == ""; tmp2 == "foo";
print tmp4              # buf == ""; tmp1 == ""; tmp2 == "foo"; 
    output>>> ""
tmp 5 = ""              # buf == ""; tmp1 == ""; tmp2 == "foo"; 
buf = tmp5              # buf == ""; tmp1 == ""; tmp2 == "foo";
tmp3 = tmp1 + tmp2      # buf == ""; tmp1 == ""; tmp2 == "foo";
buf = tmp3              # buf == "foo"; tmp1 == ""; tmp2 == "foo";

second-loop:

tmp1 = buf              # buf == "foo"; tmp1 == "foo"; tmp2 == "foo";
tmp2 = socket.recv(8)   # buf == "foo"; tmp1 == "foo"; tmp2 == "bar";
tmp4 = buf              # buf == "foo"; tmp1 == "foo"; tmp2 == "bar";
print tmp4              # buf == "foo"; tmp1 == "foo"; tmp2 == "bar";
    output>> "foo"
tmp 5 = ""              # buf == "foo"; tmp1 == "foo"; tmp2 == "bar"; 
buf = tmp5              # buf == "foo"; tmp1 == "foo"; tmp2 == "bar";
tmp3 = tmp1 + tmp2      # buf == "foo"; tmp1 == "foo"; tmp2 == "bar";
buf = tmp3              # buf == "foobar"; tmp1 == "foo"; tmp2 ==
"bar";

third-loop:

tmp1 = buf              # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"foo";
tmp2 = socket.recv(8)   # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
tmp4 = buf              # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
print tmp4              # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
    output>> "foobar"
tmp 5 = ""              # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
buf = tmp5              # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
tmp3 = tmp1 + tmp2      # buf == "foobar"; tmp1 == "foobar"; tmp2 ==
"baz";
buf = tmp3              # buf == "foobarbaz"; tmp1 == "foo"; tmp2 ==
"baz";

and of course next time the print statement comes along you'll get
"foobarbaz".

So how do you deal with this problem?

Well the first, and most successful, approach is "Don't use threads"
:).  And that's not really a joke.  Most of the time threading is
overkill, there are often other non-threaded designes that are both
faster *and* simpler.

If however the nature of your program is such that threads are the
preferred approach, then you have use the features of the threading
module to impose your desired ordering on the instructions.

The problem you are currently facing is called a data-race.  You have
two threads both accessing the same variable (when talking threads,
variables, files, db-connections, et al are called resources).  The
traditional example used is a bank account:

balance = 1000

def get_balance():
    return balance
def put_balance(amount):
    balance = amount

Thread 1                             Thread 2
def deposit(amount):                 def withdraw(amount):
    log_deposit_start()                  log_withdraw_start()
    balance = get_balance()              balance = get_balance()
    balance = balance + amount           balance = balance - amount
    put_balance(balance)                 put_balance(balance)
    log_deposit_finish()                 log_withdraw_finish()

deposit(100)                         withdraw(100)

Which should naturally leave you with a final balance of 1000.

Now you should be able to use the same sort of thinking I used above
on your code to demonstrate that as the code stands possible answers
include 900, 1000, and 1100 (be aware these are not the only possible
answers, just the most likely ones).

The solution of course is to prevent deposit from reading or writing
to balance while withdraw is using it (ie between withdraw's
get_balance and put_balance calls), and vis-versa.  It is also
important to notice that deposit/withdraw can be executing any of the
four log_* functions while the other is modifying balance without
causing any problems.  In concurrency parlance, the balance modifying
code is a "critical section", while the log_* code is a "non-critical
section".

The important thing then when dealing with data-races is to ensure
"mutual exclusion" inside critical section, also refered to as
"serialising access" to the critical section.  The most common
approach to achiving this is via locking.

In general this works like so:

def deposit(amount):                    def withdraw(amount):
    non_critical_code()                     non_critical_code()
    balance_lock = get_balance_lock()       balance_lock =
get_balance_lock()
    balance_lock.lock()                     balance_lock.lock()
    critical_code()                         critical_code()
    balance_lock.unlock()                   balance_lock.unlock()
    non_critical_code()                     non_critical_code()

Now I haven't ever written a threaded python application before ---
I've never had the need ;).  However from the documentation for the
threading module it appears that balance_lock would be a Lock() or
RLock() [preferably the latter] and lock() and unlock() map to
aquire() and release() respectively.

So returning to your code, a minimal change, corrected version would
look like:

def __init__(self):
  self.buf = ""
  self.buflock = RLock()

def thread read (self):
  while 1:
    self.buflock.aquire()
    self.buf=self.buf+self.my_socket.recv (8)
    self.buflock.release()
 
def print buf(self):
  self.buflock.aquire()
  if self.buf:
    print "bufeer-->",self.buf
    self.buf=""
  self.buflock.release()

you should then see the behaviour you expect.

However:

a) holding a lock while performing a blocking operation is a very
*bad* thing, so at the very least read() should be rewritten:

def thread read (self):
  while 1:
    tmp = self.my_socket.recv(8)
    self.buflock.aquire()
    self.buf=self.buf + tmp
    self.buflock.release()

b) If this is a server please consider why you need threads to handle
your network io, especially when nfsd, sendmail, apache, wuftpd, lpd,
and tcpd don't.

c) If this is a gui client, understand that threads have a tendency to
interact poorly with many widget libraries, which generally provide
select/poll based event-loops with support for socket-io already.

d) If you still decide you want to use threads, consider using the
battery's included Queue module, instead of rolling your own with
string concatenation.

Anyway, regardless good-luck.

Andrae Muys


From tim-dated-1013892102.1a1a7a at catseye.net  Fri Feb  8 15:41:40 2002
From: tim-dated-1013892102.1a1a7a at catseye.net (Tim Legant)
Date: 08 Feb 2002 14:41:40 -0600
Subject: What is the most efficient way to do this?
In-Reply-To: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
References: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
Message-ID: <86vgd7d7uj.fsf@skitty.catseye.net>

"MDK" <mdk at mdk.com> writes:

> What is the most efficient way to remove empty items from this list?
> 
> ['', '1', '2', '3', '', '456', '789', 'wow']

>>> lst = ['', '1', '2', '3', '', '456', '789', 'wow']
>>> filter(None, lst)
['1', '2', '3', '456', '789', 'wow']


Tim



From magnus at thinkware.se  Wed Feb 20 14:48:55 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 11:48:55 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014136412.9849.python-list@python.org> <m2y9ho5hwi.fsf@desk.crynwr.com>
Message-ID: <258fd9b8.0202201148.51cbbd8b@posting.google.com>

Russell Nelson <nelson at crynwr.com> wrote in message news:<m2y9ho5hwi.fsf at desk.crynwr.com>...
 
> print "Frankly, I don't know why comp.lang.python discussion is not in Python"

Because that would be to close to abominations like "Perl Poetry"?


From jkraska at san.rr.com  Wed Feb 13 22:35:07 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 14 Feb 2002 03:35:07 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org>
Message-ID: <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com>

>to go when coding in Python.  Further, since Emacs (and I've heard, vim)
>support treating groups of spaces as a single tab (tab inserts n number of
>spaces, backspace takes you back n spaces), there really isn't much
>functional difference between the two when editing.

While I'm sure you know what you mean, other readers might not realize
that the genuinely useful capability of vim (and I suspect Emacs, if you
mean what I think you mean) is to not insert some constant 'n' spaces,
but rather to:

"insert just enough spaces to get you to the next tabstop".

And while vim indeed does just what I said, when you backspace over
these runs of spaces while in insert mode, it knows how those spaces
where generated and so:

"deletes any such spaces inserted through a tab, just as if
it were deleting a single tab."

This is a very useful capability. As for how to get it, one might
insert the following into their .vimrc:

set tabstop=8
set softtabstop=4 
set shiftwidth=4 
set et
"ascii-183==?
"ascii-187==?
set lcs=tab:?\ 
set list

The above configuration has the following properties:

1. "set tabstop=8". This leaves *hard* tabs at 8, so that whenever
you open a file with "real" tabs in it, they will have the same
indentation level as more, cat, lpr, and so forth are likely to.

2. "set softtabstop=4". This governs the behavior of the normal
tab key in vim.

3. "set shiftwidth=4". This sets autoindent to use a indentation
level of 4 after code blocks are begun.

4. "set et". This tells vim to never insert tabs. Use runs of
spaces instead.

5. "set lcs=tab:?\ ". This tells vim to use ascii-187 (the single
'>>' char) as a visible marker of all *real* tabs in your file.
This makes physical tabs as plain as day in your file: helpful
for seeing what's wrong with makefiles. Similar to the "Show-P"
feature in Word.

6. "set list". Turns #5 on.

Two notes. You can't type ascii 187 with an ordinary character.
In vim, you get it buy typing '<ctrl>V187'. The two lines before
the "set lcs" line are just vim comments that I have so that I
don't forget the numeric codes.

One last note. You're going to forget this eventually.

What if you want to type a *real* tab?

<ctrl>V<tab>

Or you can "set noet" to temporarily turn of tab expansion.

C//





From andersaa at stud.cs.uit.no  Wed Feb 27 05:03:16 2002
From: andersaa at stud.cs.uit.no (Anders Åberg)
Date: Wed, 27 Feb 2002 11:03:16 +0100
Subject: Thread support in NetBSD
References: <a5i40k$1k59$1@news.uit.no> <m3lmdf1dv8.fsf@mira.informatik.hu-berlin.de>
Message-ID: <a5iarm$1nvp$1@news.uit.no>

Thanks for your reply..

> > In the pthread directory, there are three subdirectories: bin, include
and
> > lib. Should I have passed one of these as parameter to configure,
instead
> > of the parent directory?
>
> I think this kind of setup is not supported. The directory it asks for
> is the library directory (it goes into the compiler line as a -L
> option), but there is no provision for locating header files in a
> different place.


So i should use
./configure --with-threads=/usr/pkg/pthread/lib
instead of
./configure --with-threads=/usr/pkg/pthread ?

> I'm not sure what MIT threads are, but unless they closely resemble
> Posix threads, you are out of luck. In that case, you'd first need to
> port Python to MIT threads.

The mit-pthread library is an implementation of posix threads. So i think
it should work.

> If the pthreads module is suited for your thread library, you can try
> setting CPPFLAGS and LDFLAGS before invoking configure. If that fails,
> you need to adjust configure.in.

What should i set these flags to? Sorry but i don't know much about these
things..


Anders.




From m.faassen at vet.uu.nl  Mon Feb 11 19:53:56 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Feb 2002 00:53:56 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net>
Message-ID: <a49p34$jlc$1@newshost.accu.uu.nl>

Fredrik Lundh <fredrik at pythonware.com> wrote:
> of course, Rue is going to argue that this doesn't apply to him, but
> to everyone else.  after all, he KNOWS he's smarter than everyone
> else, and we're just too stupid to realize how stupid we are ;-)

No, no, you wouldn't be classified under the 'stupid' category of people
who reject autocoding. You'd be either in the 'fear' category or more likely,
the 'evil' category. :)

I have no clue what category he'd put me in. Now I'll go and look through
the article you pointed me at..

Regards,

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


From DeepBleu at DeepBleu.org  Thu Feb 14 12:59:40 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Thu, 14 Feb 2002 11:59:40 -0600
Subject: zlib.so and libz.so and Python
References: <u6niisr909gtd6@corp.supernews.com> <mailman.1013703624.30857.python-list@python.org>
Message-ID: <u6nuqr108os0ae@corp.supernews.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.1013703624.30857.python-list at python.org...
>
>     DeepBleu> Python comes with zlib.so (for Sun OS).  But I noticed that
on
>     DeepBleu> the same machine, there is also libz.so.  Can someone please
>     DeepBleu> tell me what is the difference between these two and what
>     DeepBleu> libz.so is used for?
>
> "zlib" is the name of a Python module.  "zlib.so" is the runtime
incarnation
> of that module.  My guess is that libz.so is a shared library version of
the
> C "zlib" library.  The Python module is essentially just a wrapper around
> the zlib library.  On my linux machine ldd shows that zlib.so depends on
> libz.so:
>
>     % ldd zlib.so
>             libz.so.1 => /lib/libz.so.1 (0x40015000)
>             libgcc_s.so.1 => /usr/local/lib/libgcc_s.so.1 (0x40024000)
>             libc.so.6 => /lib/libc.so.6 (0x4002c000)
>             /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
>
> My guess is that you will see something similar on your Solaris system.
>
I used
% ldd zlib.so
         libz.so =>       /usr/openwin/lib/libz.so

So it means that in order to use compression/decompression in Python using
zlib module, one needs libz.so too.
The problem I am running into is that on one Solaris System (out of three),
I can not find libz.so.
and running
% ldd zlib.so
         libz.so =>       (file not found)

So on this partuclar machine, when I try using
from zipfile import ZipFile
new_zip_file = ZipFile("test.zip","w",8)

I am getting an error message that the zlib module is missing.

Where do I get the libz.so from?  Apparently it is included on newer Solaris
systems (I am guessing, since the machine it is not on happens to be a bit
older).

>     DeepBleu> Also, I want to find out if zlib.so (the one that comes with
>     DeepBleu> the Python distribution) is all we need to work with ZIP
files
>     DeepBleu> in Python.
>
> Technically, I suppose you could, but I think you'll find it a heck of a
lot
> easier to use the zipfile module (which uses zlib underneath the covers)
to
> work with ZIP files.
>
I know, I just meant if we also need libz.so in addition to zlib.so on the
system for zipfile etc... to work fine.
I am new to working on Solaris that is why I am a bit confused for now over
this.
DeepBleu
> --
> Skip Montanaro (skip at pobox.com - http://www.mojam.com/)
>




From store_li at sina.com  Mon Feb 25 22:42:23 2002
From: store_li at sina.com (Ricky)
Date: Tue, 26 Feb 2002 11:42:23 +0800
Subject: images
References: <iGhe8.20623$vP.68360@rwcrnsc51.ops.asp.att.net>
Message-ID: <a5f093$6mqk7$1@ID-12869.news.dfncis.de>

If you want to display a pic, you must use one GUI toolkit. If your program
is not very explicit, you may use PIL+Tkinter.

"Ali K" <akakakhel at attbi.com> wrote in message
news:iGhe8.20623$vP.68360 at rwcrnsc51.ops.asp.att.net...
> I would like to know how to show a picture in Python so that it shows up
on
> my window.
>
> Please Reply.
>
> Thank You.
>
> Ali K.
>
>




From phr-n2002a at nightsong.com  Mon Feb 11 06:12:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 03:12:34 -0800
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk>
Message-ID: <7xeljsp90d.fsf@ruckus.brouhaha.com>

Tim Howarth <tim at worthy.demon.co.uk> writes:
> (Or worse, I can convert with eval
> 
> a = str(hex(a))[2:-1]
> a = eval('0x'+a)

  a = eval('0x%x' % a)

might be a little more palatable.  


From hsauer at marnie.teuto.de  Sun Feb 24 12:45:50 2002
From: hsauer at marnie.teuto.de (Henning Sauer)
Date: 24 Feb 2002 18:45:50 +0100
Subject: Extracting data from telnet session ?
Message-ID: <m3d6yuai41.fsf@marnie.henning-sauer.de>

Hello,

I want to extract status information from a router to which I can
connect via telnet. 
Currently I try the following:

import getpass
import sys
import telnetlib

HOST = "192.168.0.1"
password = getpass.getpass()

tn = telnetlib.Telnet(HOST)
tn.set_debuglevel(1)
tn.read_until("Password:")
tn.write(password+"\n")

tn.write("24\n")
tn.write("1\n")
print tn.read_all()

this gets me output like this:
 
Telnet(192.168.0.1,23): recv '7;1H'
Telnet(192.168.0.1,23): recv '\033[8;1H'
Telnet(192.168.0.1,23): recv '\033[8;26H8\033[8;38H76\033[8;53H24542\033[8;65H696\033[8;79H2\033[8'
Telnet(192.168.0.1,23): recv ';1H'
Telnet(192.168.0.1,23): recv '\033[11;1H'
Telnet(192.168.0.1,23): recv '\033[12;1H'
Telnet(192.168.0.1,23): recv '\033[16;12H'
Telnet(192.168.0.1,23): recv '\033[18;23H'
Telnet(192.168.0.1,23): recv '\033[3;72H'
Telnet(192.168.0.1,23): recv '\033[4;62H'
Telnet(192.168.0.1,23): recv '\033[21;49H'
Telnet(192.168.0.1,23): recv '\033[14;22H'
Telnet(192.168.0.1,23): recv '\033[14;32H9\033[14;22H'
Telnet(192.168.0.1,23): recv '\033[7;1H'
Telnet(192.168.0.1,23): recv '\033[7;25H123\033[7;37H813\033[7;55H244\033[7;64H7121\033[7;79H6\033'
Telnet(192.168.0.1,23): recv '[7;1H'
Telnet(192.168.0.1,23): recv '\033[8;1H'
Telnet(192.168.0.1,23): recv '\033[8;25H832\033[8;37H217\033[8;54H8205\033[8;65H302\033[8;79H3\033'
Telnet(192.168.0.1,23): recv '[8;1H'
Telnet(192.168.0.1,23): recv '\033[11;1H'
Telnet(192.168.0.1,23): recv '\033[12;1H'
Telnet(192.168.0.1,23): recv '\033[16;12H'
Telnet(192.168.0.1,23): recv '\033[18;23H'
Telnet(192.168.0.1,23): recv '\033[3;72H'
Telnet(192.168.0.1,23): recv '\033[4;62H'
Telnet(192.168.0.1,23): recv '\033[21;49H'

What I would like to know is how can I extract the information 
from this data stream ? 
Is this really the data from the router or might it be debug 
information ? When I'm not setting the set_debuglevel I'm 
getting no output.
 
Bye,
Henning


From nospam at bigfoot.com  Mon Feb 18 09:07:10 2002
From: nospam at bigfoot.com (Gillou)
Date: Mon, 18 Feb 2002 15:07:10 +0100
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>
Message-ID: <a4r1ch$ghd$1@norfair.nerim.net>

mxODBC is the only (AFAIK) known python wrapper over unix ODBC managers (any
of iODBC or UnixODBC)
Alternate solution, use a direct connection to your DBMS client (no ODBC)
Search from http://www.python.org/sigs/db-sig/

HTH

--Gilles

"Ville Vainio" <vvainio at karhu.tp.spt.fi> a ?crit dans le message news:
yoxadu7xhk1.fsf at karhu.tp.spt.fi...
> Briefly put: can it be done? I'm mainly concerned with the price tag
> on mxODBC...
>
> --
> Ville Vainio - http://www.tp.spt.fi/~vvainio - ICQ #115524762
>  Wild geese have no intention to cast a reflection
>  Water has no mind to assume their form




From sholden at holdenweb.com  Wed Feb 13 11:21:09 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 11:21:09 -0500
Subject: Proceedings CD-ROM of IPC10?
References: <dinu-781536.17015113022002@news.t-online.com>
Message-ID: <Lnwa8.208046$KM2.8599052@atlpnn01.usenetserver.com>

"Dinu C. Gherman" <dinu at reportlab.com> wrote in message
news:dinu-781536.17015113022002 at news.t-online.com...
> Hi,
>
> I'm not sure if Foretec plans to do this or not, but
> is there already a CD-ROM with Proceedings for IPC10?
> I vagualy remember having seen something similar for
> IPC9, or so...
>
It was on sale for $50 at IPC10.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From henrik.motakef at web.de  Mon Feb 18 14:51:16 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 18 Feb 2002 20:51:16 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <mailman.1014064953.24595.python-list@python.org>
Message-ID: <86vgcud0wb.fsf@apo.hmotakef.int>

Gerhard H?ring <gh_pythonlist at gmx.de> writes:

> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> > Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> > czignat?  Classifero en bragna!
> Woas da Deifi wos des hoasn mog.

Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...

mfg
Henrik


From mdk at mdk.com  Tue Feb 12 08:24:41 2002
From: mdk at mdk.com (MDK)
Date: Tue, 12 Feb 2002 08:24:41 -0500
Subject: Why prefix with an underscore?
Message-ID: <a4b52q$1dq8gi$1@ID-98166.news.dfncis.de>

I am wondering why I see some people's code with function definitions
prefixed with an underscore.

For example:

def _myfunct(foo):

Does this mean something to the code or is it for some other reason?

Thanks.




From guy at wyrdrune.com  Sun Feb  3 13:15:26 2002
From: guy at wyrdrune.com (Guy Gascoigne - Piggford)
Date: Sun, 3 Feb 2002 10:15:26 -0800
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <mailman.1012745045.21128.python-list@python.org>
Message-ID: <3c5d7e28_2@news.nwlink.com>

Actually I did look at them, one of them seemed somewhat limited for what I
needed and the other uses an external expect library for functionality.
What I want need has to execute under jython - so I can extend it in java
but not C (at least easily).

So that leaves me with my original syntax question.

Guy

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.1012745045.21128.python-list at python.org...
>
>     Guy> I'm trying to find a way to write something like expect in a
python
>     Guy> module, ...
>
> Any particular reason you're not using one of the other Expect-like Python
> thingies:
>
>     http://www.vex.net/parnassus/apyllo.py?find=expect
>
> ?
> --
> Skip Montanaro (skip at pobox.com - http://www.mojam.com/)
>




From Simon.Budig at unix-ag.org  Fri Feb  1 19:00:31 2002
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 2 Feb 2002 01:00:31 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net>
Message-ID: <3c5b2c1f@si-nic.hrz.uni-siegen.de>

Michael Hudson <mwh at python.net> wrote:
> Simon Budig <Simon.Budig at unix-ag.org> writes:
>> Something is broken with the completion of readline:
> 
> The more I learn about readline, the less I like about it.  Not that
> I'd want to live without it, of course.

Heh - add ncurses to that list...   :-)

>> It seems that readline has a fallback to filename completion when
>> no matches are available. Even if I use my own completion function:
>> 
>> >>> def nullcompleter (text, state):
>> ...    print "\nBuh!"
> 
> <aside>
> That's probably a bad idea; you have no idea what state the terminal
> will be in when this function gets called...
> </aside>

Yeah -  I think so too, but it seems to work Ok and it helped me
to understand when the custom function gets called...

[...]
>> Is this a known Problem? 
> 
> Well, I knew about it.  I don't regard it as a huge problem.

Hmm. Imagine you are writing an interactive interpreter of a
custom language to manipulate something non-file'ishes and
suddenly the commandline starts completing *filenames*. I
think this is a good showstopper...

>> Is there an evil hack to avoid this?
> 
> Dunno.  Doubt it, somehow.  You'll need to look at the documentation
> for readline, though.

Well - I looked at the documentation for the readline-module, but that
doesn't say anything about it. Maybe I should read the sources.

Oh well. I am currently a little bit clueless...
        Simon
-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From morgan at hahaha.org  Sun Feb 24 13:06:27 2002
From: morgan at hahaha.org (Morgan Fletcher)
Date: Sun, 24 Feb 2002 18:06:27 GMT
Subject: Vehicle registry solution?
Message-ID: <87r8naoiu5.fsf@fletcher.sonic.net>

I am going to be building a set of registries for a certain brand/marque
automobile. (We've got a club, on-line registry is a project of mine.)

Do you know of any python cgi or zope-based solutions I could leverage for
something like this?

We'd want pictures and maybe ten or twenty database fields, as well as an
associated mailing list. We're already running Mailman 2.0.8 on the site.

Server is composed of netbsd, postgresql, python, apache.

Thanks,

Morgan



From martin at v.loewis.de  Wed Feb 13 02:41:04 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 08:41:04 +0100
Subject: UTC time conversion to internal time
References: <49ja8.17872$N31.902613@ozemail.com.au>
Message-ID: <m3n0yd3k33.fsf@mira.informatik.hu-berlin.de>

"Colin Brown" <cbrown at met.co.nz> writes:

> I wish to convert UTC times to internal time representation (the inverse of
> time.gmtime). There does not seem to be any function to do this. Does anyone
> have a clean solution please.

Surprisingly, this is not supported in the standard C library, and
hence also not supported in Python.

Regards,
Martin



From joost_jacob at hotmail.com  Sun Feb  3 18:57:09 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 3 Feb 2002 15:57:09 -0800
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <13285ea2.0202030404.5dfe9b66@posting.google.com> <a3j9du$181gf1$1@ID-7529.news.dfncis.de>
Message-ID: <13285ea2.0202031557.244bb70d@posting.google.com>

[Karl M. Syring]>
> This is utter nonsense. Pdcurses simply works on Windows.

Maybe -you- can get it to work?
Suppose you have the Python binaries Windows distribution 2.1 like we
have, how do you get it to work?  Is it in the Python 2.2
distribution?  Did you read the previous post?


From tim.one at comcast.net  Sat Feb 23 09:44:09 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sat, 23 Feb 2002 09:44:09 -0500
Subject: strings and sort()
In-Reply-To: <m2vgco1mun.fsf@mycroft.actrix.gen.nz>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEMHNOAA.tim.one@comcast.net>

[Paul Foley]
> ...
> There *is* a built in SORT function in CL.  It's destructive.

How odd that one could call a function that brings order to its unruly
argument "destructive".

just-more-proof-that-lisp-warps-the-mind<wink>-ly y'rs  - tim




From mlh at vier.idi.ntnu.no  Tue Feb  5 21:27:58 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 6 Feb 2002 02:27:58 +0000 (UTC)
Subject: ANN: Anygui 0.1 (final) released
Message-ID: <slrna6155d.h9.mlh@vier.idi.ntnu.no>

Greetings!

Anygui 0.1 has been released, and may be dowloaded from the anygui
project page (http://sourceforge.net/projects/anygui).

Anygui is a generic GUI package which allows you to write programs
that will work with several backends, such as Tkinter, wxPython,
PythonWin and Java Swing. The name is inspired by the standard library
package anydbm, because the behaviour is similar: You write your
program using the Anygui API, and Anygui figures out which back-end to
use.

For more information, please visit the Anygui web site at

                        http://www.anygui.org


			

Magnus Lie Hetland,

Anygui admin


--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From logiplexsoftware at earthlink.net  Thu Feb 28 17:11:09 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 28 Feb 2002 14:11:09 -0800
Subject: listing threads?
In-Reply-To: <a5m66j+lg39@eGroups.com>
References: <a5m66j+lg39@eGroups.com>
Message-ID: <20020228141109.1fe13f7c.logiplexsoftware@earthlink.net>

On Thu, 28 Feb 2002 21:07:31 -0000
maximilianscherr wrote:

> in the docs theres the information, that i can list all active 
> threads with threading.enumerate(), but i always just get the 
> Mainthread displayed in the interpreter console.
> 
> im doing somethin like:
> 
> class test(threading.Thread):
>   def t(self):
>        print "self"
> 
> te = test()
> te.start()
> 
> then if i do threading.enumerate() i just get the mainthread int he 
> list.
> 
> so how can i now list all (active) threads?

If test really looks like that, then I'm guessing it exits before you call
enumerate, so all that is left is the main thread.  Also, if you derive
from Thread, you need to tell it what it's running:

import threading
import time

class test(threading.Thread):
    def __init__(self):
        threading.Thread.__init__(self, target = self.t)
    def t(self):
        while 1:
            time.sleep(1)

te = test()
te.start()

threading.enumerate()


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From mhammond at skippinet.com.au  Mon Feb  4 02:44:08 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 07:44:08 GMT
Subject: Clipboard viewer
References: <mailman.1012681806.23290.python-list@python.org>
Message-ID: <3C5E3C19.1080809@skippinet.com.au>

Subramaniam Karthik wrote:
> Hi,
>    I am trying to implement a clipboard viewer in python. and i need to know
> which application posted the data to the clipboard. any ideas as to how this
> can be done? the GetClipboardOwner() just returns the window handle and not
> the application name.
> 
> Thanks
> karthik


All apps I have seen simply use the title of the window - eg, that 
returned by GetWindowText().

Mark.



From bh at intevation.de  Thu Feb 28 05:56:15 2002
From: bh at intevation.de (Bernhard Herzog)
Date: 28 Feb 2002 11:56:15 +0100
Subject: regular expression help
References: <1014853112.982.15.camel@beast.l3.com> <mailman.1014854866.5558.python-list@python.org>
Message-ID: <6qy9hd98og.fsf@abnoba.intevation.de>

Skip Montanaro <skip at pobox.com> writes:

>     Drew> I'm having some issues doing a small but irritating regular
>     Drew> expression match.
> 
> As Jamie Zawinski said probably more than once on comp.lang.emacs (and
> Fredrik Lundh has reminded us here):

There is no comp.lang.emacs :) At least not on google. The posting was
on comp.emacs.xemacs and alt.religion.emacs

>     Some people, when confronted with a problem, think "I know, I'll use
>     regular expressions".  Now they have two problems.

The quote is correct, though:
http://groups.google.de/groups?selm=33F0C496.370D7C45%40netscape.com

The two pragraphs preceeding the quoted one in JZWs posting are also
interesting:

  The notion that everything is a stream of bytes is utterly braindead.
  The notion that regexps are the solution to all problems is equally
  braindead.

  Just like Perl.

:-)


   Bernhard

-- 
Intevation GmbH                                 http://intevation.de/
Sketch                                 http://sketch.sourceforge.net/
MapIt!                                               http://mapit.de/


From codetube at yahoo.com  Sun Feb  3 08:24:49 2002
From: codetube at yahoo.com (codetube)
Date: Sun, 03 Feb 2002 13:24:49 -0000
Subject: Implementing PythonWin console from a Visual Basic application!
Message-ID: <a3jdn1+pftc@eGroups.com>

Hello Everyone,

I wonder if any one knows some thing about implementing the PythonWin 
console and other capabilities from with in a Visual Basic 
application.
I am writing an application in which I propose to have a console 
similar to PythonWin and also be able to run python files etc.
This might be a "big" question, so I would appreciate any clues or 
leads in that direction.
I had looked into some examples given by Mark Hammond in his 
book "Programming Python" but then I am looking for a more tight 
integration into the VB application.

Any help would be appreciated...

Best Regards,
Code Tube





From office at eu-domain.biz  Fri Feb  8 16:28:56 2002
From: office at eu-domain.biz (office at eu-domain.biz)
Date: 08 Feb 2002 22:28:56 +0100
Subject: Preregister your .eu domain now!
Message-ID: <UTC200202082129.WAA02406@hera.cwi.nl>

Pre-register your .eu domain now!

>From now on we accept pre-registrations 
for the Top Level Domain .eu.

Pre-register at:

http://www.eu-domain.biz/english

All pre-registrations are accepted on the 
basis of "first come, first serve". Each domain 
can only be pre-registered once.

office at eu-domain.biz

P.S. pre-register now, tomorrow may be to late!





From aries at bitstream.net  Mon Feb 18 19:28:39 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Mon, 18 Feb 2002 18:28:39 -0600
Subject: help - inserting values into database tables
References: <mailman.1013981787.10266.python-list@python.org><mailman.1014010317.7276.python-list@python.org> <mailman.1014053070.9613.python-list@python.org>
Message-ID: <3c719a9d$0$22477$6536d21b@news.bitstream.net>

Hi again,
Just wondering if you can help?

I installed the python ver 2.0(from CD)for windowsX on a windows98
system.  Then I installed the windows32all stuff.
So are the dbi and the odbc modules should now be installed
on my system now, or do I need some other package ??????

It seems I can import the dbi then the odbc with no errors...
But my simple script below won't run; has an error that I
don't understand ???????

I am running from the command line, heres my environ stuff
******* Do I need to put in another path to Lib ?????

COMSPEC=C:\WINDOWS\COMMAND.COM
windir=C:\WINDOWS
PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\JDK11~1.8\BIN;C:\PYTHON20;

******** Heres the error I get upon running from prompt

C:\codepy>python conntest.py

Traceback (most recent call last):
  File "conntest.py", line 8, in ?
    conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
dbi.operation-error: [Microsoft][ODBC Driver Manager] Data source name not
found and no default driver specified in LOGIN

The Manufacturers.mdb is the ms-access database of ver office XP
2002.  It is just a small database with a small table in it.
Very simple kind.  The file is in that codepy directory also.
So I don't understand ?????????????????
What is DSN, does it allow for a ms-access file??????????

Well anyways heres my test database connection script, I am
just trying to connect and print some output from the database.
I think I am doing this correct but not sure ??????????

The script:  conntest.py listing
#! /usr/bin/env python    #not needed for windowspython

#test script to connect to a ms-access database

import dbi
import odbc

conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
rs = conn.cursor()
rs.execute("select * from ManufName")
print rs.description

rs.close
conn.close

Can you shine any light my way ???????????

Thank you (working on a windows system today(UGH...)),
Joe F.
############

chajadan <python at chajadan.net> wrote in message
news:mailman.1014053070.9613.python-list at python.org...
> I always use PythonWin, part of the win32all distribution. It's a
> color-coded ide, with a basic syntax check, and has one nifty advantage
> over any other input editor I've tried: when you call a function, it takes
> the document strings and actually gives you a little note about what kind
> of input it's expecting!
>
> So all in all it should be the same thing as it ever was, but now in the
> environment your particular task seems to ask for. The only "new" things
> will be the other possible windows-only features you might want to learn
> and use in your time on windows.
>
> Happy computing!
>
> --chajadan
>
> At 08:15 AM 2/18/02 -0600, mylinuxboxroot wrote:
> >Hey thanks,
> >This is great, and I have the project now that it will be developed on a
> >windows
> >server so I guess I don't need to worry about going from linux to the
> >MS-Access
> >database.  I will develop it right on the server using the win32all
> >distribution that
snip





From morton at dennisinter.com  Fri Feb 22 15:42:50 2002
From: morton at dennisinter.com (damien  morton)
Date: 22 Feb 2002 12:42:50 -0800
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56>
Message-ID: <4abd9ce7.0202221242.2cc3410@posting.google.com>

Lutz Schroeer <Lutz.Schroeer at kybernetik-manufaktur.de> wrote in message news:<Xns91BAE535689F7Latzikatz at 139.174.2.56>...
> I'm creating a website which makes intensive use of cgi scripts and there 
> is the possibility for the visitor to register as a user for having the 
> ability to customize the site appearance, contents and other stuff.
> 
> Jumping from one script to the other I need to remember at least the user's 
> name. The W3C recommends using cookies (RFC 2109). Unfortunately many 
> people don't like cookies and filter them using WebWasher or a personal 
> firewall. 
> 
> Is there any other technique for keeping the user's data while he is using 
> my site without installing any large suite as Zope or the like? I've read 
> an article about utilizing soft links but I didn't quite get the point.
> 
> Lutz

Ive seen http state management done with wildcard DNS.

<sessiod-id>.foo.com/bar.html

simpler and easier to use cookies though.


From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 23:21:58 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 22:21:58 -0600
Subject: ActiveX Questions
Message-ID: <002e01c1b37c$d1471de0$0101010a@local>

I am developing a headache messing with an OCX I picked up from the
Internet.  It seems that my forays into ActiveX never pan out; I should
stick with Linux.

The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
have to be such a pain to find?)  I can instantiate it:

>>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
>>>

(I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:

>>> o.AboutBox()
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py", line
29, in AboutBox
    return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
com_error: (-2147418113, 'Unexpected failure', None, None)
>>>

Any method or property, same answer.

The URL for the control is:

    http://www.qwerks.com/download/3693/report.zip

It is supposed to be a general printing control, but nothing works.

Chris Gonnerman -- chris.gonnerman at newcenturycomputers.net
http://newcenturycomputers.net





From brian at sweetapp.com  Sat Feb  2 16:20:38 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sat, 2 Feb 2002 13:20:38 -0800
Subject: ANN: SilverCity 0.1.0
Message-ID: <000c01c1ac2f$76669800$445d4540@Dell2>

SilverCity is a Python extension that allows you to retrieve lexical
information about different languages (e.g. C/C++, HTML, Python).
Amongst other things, you can use it to style source code into styled
HTML. It is based on Neil Hodgson's excellent Scintilla project.

I'm pretty far from release (at least another week :-) so I'd really
like a sanity check for the API. Any feedback would be greatly
appreciated.

Windows installers are available for Python 2.0, Python 2.1 and Python
2.2 (I haven't gotten around to a Linux build yet). There is also an
example script that can create HTML output for Python and C/C++ files.
Other than that, you will really have to read the code to figure out the
project but the code is pretty simple.

See:
http://www.sweetapp.com/silvercity

Cheers,
Brian




From gerson.kurz at t-online.de  Mon Feb  4 04:55:58 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 04 Feb 2002 09:55:58 GMT
Subject: lambda expressions
References: <mailman.1012756205.9196.python-list@python.org>
Message-ID: <3c5e5962.591375@news.t-online.de>

On Sun, 3 Feb 2002 11:04:57 -0600, "Jason Orendorff"
<jason at jorendorff.com> wrote:

>Gerson Kurz writes:
>> So, I see few limitations in the current implementation of lambda for
>> the DEDICATED lambda-user.
>
>Lack of tail recursion could be a problem.

I'm not really sure - I left university six years ago -, but isn't

WHILE = lambda x,y: BOOL(x()) and (FALSE(y()) or WHILE(x,y))

tail recursive? Its not optimized - that is, the recursion needs
stack, but it seems to fit the general definition:

"""Tail Recursion
==============
A call is tail-recursive if nothing has to be done after the the call
returns, i.e. when the call returns, the returned value is immediately
returned from the calling function.  In this example, the recursive
call to `myfun' is tail-recursive:
     
     (defun myfun (x)
       (if (oddp (random x))
           (isqrt x)
           (myfun (1- x))))
"""

(Quoted from
http://www.cis.ksu.edu/VirtualHelp/Info/develop/cmu-user.info.Tail_Recursion.html)

Bye, Gerson
http://p-nand-q.com


From dsh8290 at rit.edu  Sat Feb  2 22:36:32 2002
From: dsh8290 at rit.edu (dman)
Date: Sat, 2 Feb 2002 22:36:32 -0500
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
In-Reply-To: <slrna5p5ba.g91.tim@vegeta.ath.cx>
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx>
Message-ID: <20020203033632.GA8468@dman.ddts.net>

On Sun, Feb 03, 2002 at 01:40:12AM +0000, Tim Hammerquist wrote:
 
| BTW, less wasted v-space is one very attractive feature of Python.

Yes, but it still doesn't mean you should hide the braces (in the
inferior languages) on other lines.

| My eyes just can't parse:
| 
| if ( some_condition() ) {
|   {
|     for (i = first_i();i <= 10; i++)
|       {
|         initialize_some_stuff();
|         while ( some_other_condition() )
|           {
|             do_some_more_stuff()
|           }
|       }
|   }

I have trouble with that too.  My preference (FWIW, probably nothing) :

if ( some_condition() )
{
    for (i = first_i();i <= 10; i++)
    {
        initialize_some_stuff();
        while ( some_other_condition() )
        {
             do_some_more_stuff()
        }
    }
}


Notice how the matching braces are horizontally aligned and at the
same level as the line that initiates the bock?  If the braces are
removed, and colons added it will look like python (with some
extraneous parens).

| ...yet I have no trouble parsing
| 
| if ( &some_condition ) {
|     for $i (&first_i .. 10) {
|         &initialize_some_stuff;
|         &do_some_more_stuff while &some_other_condition;
|     }
| }

This is hard for me because I can't see what the '}' goes with.
Combine this with leaving out the braces on one-line compound
statements, and I'll have real problems.

-D

-- 

A perverse man stirs up dissension,
and a gossip separates close friends.
        Proverbs 16:28




From andre at vandervlies.xs4all.nl  Fri Feb 22 14:38:26 2002
From: andre at vandervlies.xs4all.nl (Andre van der Vlies)
Date: Fri, 22 Feb 2002 20:38:26 +0100
Subject: BloodPressure.py
Message-ID: <3C769E32.D6D76ECC@vandervlies.xs4all.nl>

#!/usr/local/bin/python
#
# Tue Feb 19 08:59:38 CET 2002 -- A.G. van der Vlies
<andre at vandrevlies.xs4all.nl>
# Orignal coding. Made for my brother and cardiologist (??). One
regulary asked
# what my blood pressure was, the other was really happy with a
printout....
# Obviously you'll need gnuplot.
# I guess (hope) there are more people who have an 'interest' in this.

import time
import Gnuplot

HTML_PATH = '......../public_html/BloedDruk/'
FILE = 'bloeddruk-'
EXT = time.strftime("%b-%Y", time.localtime(time.time()))
Gfx = HTML_PATH+FILE+EXT+".gif"
Stat = HTML_PATH+FILE+EXT+".stat"
OutPut = open(Stat, "a+", 0)
#
## This is dutch. I'll try an english translation...
#
# bovendruk = raw_input('Geef de bovendruk: ')
# onderdruk = raw_input('Geef de onderdruk: ')
# pulse = raw_input('Geef de hartslag: ')
bovendruk = raw_input('Upper pressure: ')
onderdruk = raw_input('Lower pressure: ')
pulse = raw_input('Pulse: ')
NOW = time.strftime("%d-%b-%Y/%H:%M", time.localtime(time.time()))
item = "%s %s %s %s\n" % (NOW, bovendruk, onderdruk, pulse)
OutPut.write(item)
OutPut.close()
print bovendruk, onderdruk, pulse, NOW
g = Gnuplot.Gnuplot()
g('set xdata time')
g('set timefmt "%d-%b-%Y/%H:%M"')
g('set format x "%d-%m\\n%H:%M"')
g('set grid')
g('set terminal gif')
g('set output "%s"' % Gfx)
g.plot(Gnuplot.File(Stat, using='1:2', title='Sys', with='linespoints'),
        Gnuplot.File(Stat, using='1:3', title='Dia',
with='linespoints'),
        Gnuplot.File(Stat, using='1:4', title='Pulse', with='impulses'),
)

--
			   Andre van der Vlies <andre at vandervlies.xs4all.nl>
			   Homepage: http://vandervlies.xs4all.nl/~andre
Key fingerprint =  99 DC D6 01 96 C2 48 80  DC 9C D5 D2 34 F6 A6 5D
--
	Open Source is a philosophy not a price tag!
--
-------------- next part --------------
A non-text attachment was scrubbed...
Name: andre.vcf
Type: text/x-vcard
Size: 237 bytes
Desc: Card for Andre van der Vlies
URL: <http://mail.python.org/pipermail/python-list/attachments/20020222/33ec7f70/attachment.vcf>

From logiplexsoftware at earthlink.net  Fri Feb  1 14:41:45 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 11:41:45 -0800
Subject: "Private" attributes - a possible idea (maybe PEP-worthy).
In-Reply-To: <PTB68.107937$8b.4076617@atlpnn01.usenetserver.com>
References: <B43D149A9AB2D411971300B0D03D7E8BF0A154@natasha.auslabs.avaya.com>
	<mailman.1012522682.17576.python-list@python.org>
	<PTB68.107937$8b.4076617@atlpnn01.usenetserver.com>
Message-ID: <20020201114145.65a8ed94.logiplexsoftware@earthlink.net>

On Fri, 1 Feb 2002 14:17:39 -0500
Steve Holden wrote:

> "Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
> news:mailman.1012522682.17576.python-list at python.org...
> > On Fri, 1 Feb 2002 10:40:45 +1100
> > Delaney, Timothy wrote:
> >
> > > I'd like to close this thread here (which I created because I thought
> I'd
> > > come up with a way to get around the wart) unless someone *can* come
up
> > with
> > > a way to deal with it.
> >
> > With Python you can start a thread, but you can't stop it.  Sorry. 
You'll
> > have to wait until reaches the end of execution.
> 
> So, just the same as c.l.py, then?

Tim _was_ referring to a c.l.py thread, then I equated what he said to
Python threads, now you're equating what I said back to c.l.py, that's
likeTraceback (most recent call last): RuntimeError: maximum recursion
depth exceeded


-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From tjreedy at home.com  Tue Feb 26 10:06:29 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 26 Feb 2002 15:06:29 GMT
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com>
Message-ID: <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7x3czobs0x.fsf at ruckus.brouhaha.com...
> Local variable declarations are for code cleanliness.  Currently,
using
> some name for a temp variable leaves detritus around after you're
done
> with it.  To figure out what the program is doing, you have to check
> whether further references to 'temp' really wanted the value left
from
> someplace far away, or were holding just a different temporary
value.
> When the usefulness of a variable ends, its scope should also end.

An explicit 'del temp' ends the scope of the variable.  Its not much
more to type that any sort of brackets and allows for overlapping
scope of temporaries instead of artificially forcing nesting.  If you
want to extend the scope of 'temp', it is probably easier to get rid
of the explicit deletion than a pair of brackets.

Terry J. Reedy






From logstx at bellatlantic.net  Wed Feb 20 13:29:22 2002
From: logstx at bellatlantic.net (logistix)
Date: Wed, 20 Feb 2002 18:29:22 GMT
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <6WRc8.2681$ss5.759@nwrddc02.gnilink.net>

Your ideoms are only clearer if you've been poisoned by C.  Don't worry,
we've all been guilty of that at one point or another.  Too many languages
have been screwed up by trying to maintain some sort of C ideom status quo.

Someone who's never programmed before won't find your proposed syntax clear
at all.

--
-

"Russell Nelson" <nelson at crynwr.com> wrote in message
news:m2vgcs5hgf.fsf at desk.crynwr.com...
> There are two Python ideoms that bother me:
>
>     while 1:
>         line = file.readline()
>         if not line: break
>
> and:
>
>     matched = re.match(r'egexp(.*)', line)
>     if matched:
>         print matched.group(1)
>
> I'd really, really like it if you could do one(1) assignment within
> the boolean expression in a while and an if.  Like this:
>
>
>     while line = file.readline():
>
> and:
>
>     if matched = re.match(r'egexp(.*)', line):
>         print matched.group(1)
>
> I just think it would be easier to read programs if they could be
> written like this.  Should I write up a PEP or has this already been
> discussed and dismissed?
>
> --
> -russ nelson              http://russnelson.com | Crypto without a threat
> Crynwr sells support for free software  | PGPok | model is like cookies
> 521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
> Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   |




From meowbot at meowing.net  Sat Feb 16 03:52:56 2002
From: meowbot at meowing.net (A Meowbot)
Date: Sat, 16 Feb 2002 03:52:56 -0500
Subject: Wow.
References: <mailman.1013536760.2921.python-list@python.org>
Message-ID: <MPG.16d7e7e7ccbc325d989684@news.supernews.com>

James_Althoff at i2.com wrote:

> "wackyparse"  -- I like that.  Is it original?

Nah, I just hang out with the wrong crowd.

> (Passing on the "Meowbot" handle, though. <wink>)

Good idea; same explanation as above.


From joshm at taconic.net  Sun Feb 10 23:39:35 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sun, 10 Feb 2002 23:39:35 -0500
Subject: A newbie question about class
References: <mailman.1013380096.28178.python-list@python.org>
Message-ID: <3c6748b8_4@corp.newsgroups.com>

Heh, I knew I'd get someone's hackles up about this with my comment.  :-)

I was simply playing the strawman here.  I see nothing wrong with comparing
to None by using "is [not]".  I was merely responding to the previous
poster's comments which said basically "do it this way because unexpected
things can occur if you don't".  While I agree with this practice
(wholeheartedly!), I have a problem with the justification.  I would have
preferred "if you don't do it this way, and you get this sort of result,
look for this type of cause".  The distinction is that in the former, the
*reason* for doing it will not propagate.  This means that nobody will learn
from this lesson.

[ObTangent]  A friend was being interviewed for a programming job and was
asked this question (about code in C):

Given:

static int x;
void foo()
{
    printf("%i", x);
}

Will the call to foo produce 0, 1, -1, or Unknown?  (Choose one.)

The *right* answer should be "Always initialize your damn variables, and
don't count on the compiler being compliant, even if the language spec
explicitly defines the behavior.  Oh, and fire those on your staff who
don't."  But that wasn't one of the choices.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('If you clean your gun without unloading it, will you be (a) ok, (b)
dead, or (c) Unknown?')




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


From fredrik at pythonware.com  Sun Feb  3 15:11:35 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 03 Feb 2002 20:11:35 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org>
Message-ID: <XPg78.22996$l93.4173649@newsb.telia.net>

Tim Peters wrote:
> Oddly enough, none of the Python developers argue about brace placement,
> except for the occasional radical who sneaks in
>
> } else {
>       }

that should of course be:

> } else {
> }

(how come Guido ignores K&R and Indian Hill, but still insists
on Python programmers using the GvR style guide? ;-)

> What the Python developers do get testy about is living with Guido's
> seemingly inconsistent demand that Python code use 4-space indents
> with no hard tabs, C code use 8-space hard-tab indents.

really?  I thought the Python C code used 4-space hard-tab
indents...

</F>




From miracle at paradise.net.nz  Mon Feb 25 15:17:32 2002
From: miracle at paradise.net.nz (Matthew Sherborne)
Date: Tue, 26 Feb 2002 09:17:32 +1300
Subject: Is it possible in Python too?
References: <3C7A6BA1.26999.19EEC27@localhost>
Message-ID: <005b01c1be39$75c23640$08020196@Jonah>

OK, Sorry for the smart remark.

How would it be done normally; by COM automation or by editing files?

Python does both. Tell how and I'll send you the python script that does it.

Matthew Sherborne




From magnus at thinkware.se  Mon Feb 25 05:31:42 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Mon, 25 Feb 2002 11:31:42 +0100
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com> <slrna784si.m30.akuchlin@dust.mems-exchange.org> <de3fc1ef.0202201835.7f49f4be@posting.google.com> <3C793FEB.CD45FF95@engcorp.com>
Message-ID: <3C7A128E.6090906@thinkware.se>


Peter Hansen wrote:


> Mere seconds with Google could give you the answer.  Try "download"
> and "turtle.py" for example.


Mere seconds with Python 2.1.1 gave me the answer. ;-)

 >>> import turtle
 >>> turtle.demo()

I'm afraid there is more in Python than in the documentation that
comes with it. (Like everything that has to do with Tkinter...)
Actually, it IS mentined in Appendix A: Undocumented modules.



From martin at v.loewis.de  Sat Feb 16 08:13:11 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 14:13:11 +0100
Subject: XmlprocDriver instance has no attribute 'feed'
References: <7c4fa2db.0202160235.5efdccac@posting.google.com>
Message-ID: <m3pu35k1so.fsf@mira.informatik.hu-berlin.de>

jesusdb at gmx.at (Dominik) writes:

> When using the xml.dom.minidom module from the standard python2.2
> installation, the an error msg '..XmlprocDriver instance has no
> attribute 'feed' appeared.

That is quite unlikely. XmlprocDriver does not ship with Python 2.2.

> To solve this problem a new version of PyXML has to be installed. (In
> my case version 7)

Most likely, you had an older version of PyXML installed. This
specific problem was indeed fixed in PyXML 0.7.

Regards,
Martin



From anthony_barker at hotmail.com  Tue Feb 12 15:13:30 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 12 Feb 2002 12:13:30 -0800
Subject: problems with a proxy & authentication
Message-ID: <899f842.0202121213.603e6617@posting.google.com>

I am having problems with python and our company's proxy server, which
requires  authentication.

here is the code I am using:

import urllib2 
proxy = urllib2.ProxyHandler({"http":"http://myname:mypass at proxyserver.company.com:8080/"})
opener = urllib2.build_opener(proxy) 
f = urllib2.urlopen('http://www.python.org') 

results
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
    return _opener.open(url, data)
  File "/usr/lib/python2.2/urllib2.py", line 322, in open
    '_open', req)
  File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.2/urllib2.py", line 785, in http_open
    return self.do_open(httplib.HTTP, req)
  File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
    raise URLError(err)
urllib2.URLError: <urlopen error (101, 'Network is unreachable')>

Under windows error is the same except for the last line (and line
numbers)
URLError: <urlopen error (10051, 'Network is unreachable')>

Both perl and wget work. wget only supports Basic Authentication, so I
assume it is the same. Thinking it might be a network type issue I
changed the domain names to IP addresses. (results follow)

Any help would be appreciated................

File "<stdin>", line 1, in ?
  File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
    return _opener.open(url, data)
  File "/usr/lib/python2.2/urllib2.py", line 322, in open
    '_open', req)
  File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.2/urllib2.py", line 785, in http_open
    return self.do_open(httplib.HTTP, req)
  File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
    raise URLError(err)
urllib2.URLError: <urlopen error (111, 'Connection refused')>


From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 23:00:03 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 22:00:03 -0600
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <mailman.1012787645.17462.python-list@python.org> <a3kv4l$jki$1@news.efn.org>
Message-ID: <00e401c1ad30$6e62f8e0$0101010a@local>

----- Original Message -----
From: "Spencer Doidge" <spencer at spencerdoidge.com>


> Thanks for all your patient help, Chris. Now I have run into an apparent
> contradiction:
>
> This is a method within SerialPort_win.py from Isaac Barona's
site-package,
> modified by myself because his 2nd ReadFile(..) parameter was an integer,
> and I want the buffer instead. I need to see what was read...

[[ code block omitted ]]

> After I have run the script that defines read(...), the other methods
> defined by the script work fine, but with read(..) I get this result:
>
> >>> tty = SerialPort("COM1",None,None,None)
> >>> b = [0] * 256
> >>> tty.read(1,b)
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "C:\Python22\Lib\site-packages\MyScripts\MySerialPort_win.py", line
> 238, in read
>     hr, bfr = ReadFile(self.__handle, bfr, overlapped)
> TypeError: Second param must be an integer or a buffer object
>
> If bfr isn't a "buffer object," then what is?

According to the help file, you need a PyOVERLAPPEDReadBuffer object here,
whatever that is... Mark Hammond, are you out there?

I've lost track; why are you passing in a buffer?  Don't you get the bytes
read as the second return value from ReadFile?  Have you read the help
file yet?






From thomas at weholt.org  Sun Feb 10 14:11:41 2002
From: thomas at weholt.org (Thomas Weholt)
Date: Sun, 10 Feb 2002 19:11:41 GMT
Subject: Using cgi.FieldStorage with enctype=multipart/form-data in SimpleHTTPServer = nothing
Message-ID: <NBz98.1108$%m1.23743@news4.ulv.nextra.no>

I'm using a simple webserver based BaseHTTPServer.HTTPServer and I want to
be able to post files to a page served by this server, using
ENCTYPE="multipart/form-data". I've got this code in my handler, subclassed
from  BaseHTTPServer.BaseHTTPRequestHandler :

def do_POST(self):
    # read in posted parameteres and data
    if self.headers.dict.has_key('content-length') and
string.lower(self.headers.dict['content-type']) ==
'application/x-www-form-urlencoded':
        print "Parsing data, content-type: %s" %
'application/x-www-form-urlencoded'
        content_length = string.atoi(self.headers.dict['content-length'])
        parameters = cgi.parse_qs(self.rfile.read(content_length))
        # do something
    else:
        print "Parsing multipart/form-data"
        import cgi
        fs = cgi.FieldStorage() # also tried : fs =
cgi.FieldStorage(self.rfile)
        # do something with the data in fs

When I use plain post, no ENCTYPE specified it works fine ( the first
if-sentence), but parsing input from multipart/form-data shows nothing in
the fs-object. I've looked into the data sent to the server and it's there,
but nothing seem to appear using the cgi-module. What is the easiest way of
doing this ( ie. uploading files to a webpage using these modules) ? How can
I parse the posted data manually, using mimetools etc. ?

Best regards,
Thomas Weholt






From goodger at users.sourceforge.net  Wed Feb  6 19:11:30 2002
From: goodger at users.sourceforge.net (David Goodger)
Date: Thu, 07 Feb 2002 00:11:30 GMT
Subject: ANN: HappyDoc 2.0.1
References: <mailman.1012913477.7851.python-list@python.org> <a3pqbh$nru$2@peabody.colorado.edu> <Xns91ADF0977CADRASXnewsDFE1@130.133.1.4> <a3s8di$5r4$1@peabody.colorado.edu>
Message-ID: <B88730C4.1E468%goodger@users.sourceforge.net>

> Philip Swartzleonard wrote:
>> ... I don't see any moral reason why happydoc's html output system
>> and better introspection (and whatever else it does good) couldn't be
>> merged into pydoc. :\

Fernando P?rez wrote:
> My point exactly. We have two open projects with vast amounts of overlap and
> certain individual strengths each. *Exactly* the problem calling for a
> judicious integration of the better parts of each, all for Python's eternal
> glory :)

More than two! For a grand-unification plan and work in progress, please see
PEPs 256 & 258 and the first two projects listed in my sig.

    http://www.python.org/peps/pep-0256.html
    http://www.python.org/peps/pep-0258.html

-- 
David Goodger    goodger at users.sourceforge.net    Open-source projects:
 - Python Docstring Processing System: http://docstring.sourceforge.net
 - reStructuredText: http://structuredtext.sourceforge.net
 - The Go Tools Project: http://gotools.sourceforge.net



From jeff at ccvcorp.com  Mon Feb 25 16:03:12 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 25 Feb 2002 13:03:12 -0800
Subject: PIL jpeg problems
References: <mailman.1014236017.29342.python-list@python.org> <x7vgcohwny.fsf@linux.local>
Message-ID: <3C7AA690.E8DF2A05@ccvcorp.com>


Dieter Maurer wrote:

> Jonathan Miller <notjonathan at yahoo.com> writes on Wed, 20 Feb 2002 12:12:41 -0800 (PST):
> > I'm having trouble using jpegs with PIL on my ISP's server--FreeBSD
> > 4.1.1-STABLE, Python 2.0 (#4, Nov 22 2000), GCC 2.95.2 19991024 (release), at
> > pair.com--a good box.
> > ...
> > IOError: decoder jpeg not available
> I once read (either here or in the Zope mailing lists) that
> JPEG support must be enabled explicitly.
>
> Please search Google to find out the details.
>
> Dieter

IIRC, PIL needs to have libjpeg available at compile-time in order to build with jpeg
support.  This shouldn't be an issue if using prebuilt binaries (as I have been), but I
don't know whether any BSD prebuilt binaries are available...

Jeff Shannon
Technician/Programmer
Credit International









From aahz at panix.com  Fri Feb  8 12:58:33 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 09:58:33 -0800
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <a413k9$s8h$1@panix2.panix.com>

In article <mailman.1013181915.301.python-list at python.org>,
Skip Montanaro  <skip at pobox.com> wrote:
>
>Guido posted a bunch of talk slides to the Python website in PowerPoint
>format.  I understand his reasons for not creating HTML versions (mostly
>because the HTML generated by PPT sucks).  Still, that leaves people without
>access to PPT sort of out in the cold.  What non-MS tools are out there
>(open source or otherwise) for reading, displaying or converting PPT slides
>to other, more accessible, formats?  If we can get a reasonable (but fairly
>short) list of tools, I can probably see they get referenced on the relevant
>page of the Python website.

Just figured I'd make a plug for ReportLab, which avoids this problem by
creating PDF instead of PowerPoint.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From replogle2 at cox.net  Sat Feb  2 13:01:12 2002
From: replogle2 at cox.net (Andrew Replogle)
Date: Sat, 02 Feb 2002 18:01:12 GMT
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
Message-ID: <IPV68.8359$Xk4.443398@news1.east.cox.net>

"Andrew Replogle" <replogle2 at cox.net> wrote in message
news:LwK68.7172$Xk4.381742 at news1.east.cox.net...
>    Anyone know of any books out there for python system administration or
> should I just stick to perl for these tasks?
>
>
Well, for example, I'd like to write a python script to tail a ftp log file
& when a connection from a customer is made and the file transmits
succesfully to email them a confirmation. I am usually very efficient if I
have just a small example or concept of how to do this and just wanted to
know if anyone had any suggestions on such a book. Thats all. :)

Andrew




From rony.steelandt at bucodi.com  Tue Feb 19 03:34:41 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 19 Feb 2002 00:34:41 -0800
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com> <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com>
Message-ID: <b34d3835.0202190034.24337a3e@posting.google.com>

Michael Kelly <mkelly2002NOSPAM at earthlink.net> wrote in message news:<nlh27ukfu8u2ga4g2n876b211idb024rkt at 4ax.com>...
> Out of curiosity, did they give you a type library
> with the control?  Also you might try running
> it through Makepy with -o option to get a module?
> Perhaps if you posted the CoClass part of the
> ActiveX module produced by Makepy somebody 
> could see if something is weird.

Still no succes with that one :(
It don't seems to me that it needs a hosting form, it looks like it
creates it own modal form to me

Here is the file created by makepy :
#------------------------------------
# Created by makepy.py version 0.3.3
# From type library 'REPORTOCX.OCX'
# On Sun Feb 17 17:18:27 2002
"""ReportOcx ActiveX Control module"""
makepy_version = '0.3.3'

import win32com.client.CLSIDToClass, pythoncom

# The following 3 lines may need tweaking for the particular server
# Candidates are pythoncom.Missing and pythoncom.Empty
defaultNamedOptArg=pythoncom.Missing
defaultNamedNotOptArg=pythoncom.Missing
defaultUnnamedArg=pythoncom.Missing

CLSID = pythoncom.MakeIID('{E80AB8D2-EB1E-49EE-8C2A-57BB2E70302A}')
MajorVersion = 1
MinorVersion = 0
LibraryFlags = 10
LCID = 0x0

from win32com.client import DispatchBaseClass
class _DReportOcx(DispatchBaseClass):
	"""Dispatch interface for ReportOcx Control"""
	CLSID = pythoncom.MakeIID('{BE1E9308-9F3C-4D83-B7BC-D0F6F6DCE778}')

	def AboutBox(self):
		return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)

	def AddRecord(self, szRecordString=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2f, LCID, 1, (24, 0), ((8,
0),),szRecordString)

	def Clear(self):
		return self._oleobj_.InvokeTypes(0x3c, LCID, 1, (24, 0), (),)

	def DoPrint(self):
		return self._oleobj_.InvokeTypes(0x33, LCID, 1, (11, 0), (),)

	def DoPrintPreview(self):
		return self._oleobj_.InvokeTypes(0x34, LCID, 1, (11, 0), (),)

	def DoPrinterSetup(self):
		return self._oleobj_.InvokeTypes(0x35, LCID, 1, (11, 0), (),)

	def GetPageHeight(self):
		return self._oleobj_.InvokeTypes(0x31, LCID, 1, (3, 0), (),)

	def GetPageWidth(self):
		return self._oleobj_.InvokeTypes(0x32, LCID, 1, (3, 0), (),)

	def GetPages(self):
		return self._oleobj_.InvokeTypes(0x30, LCID, 1, (2, 0), (),)

	def SetBottomString(self, szBottomString=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2c, LCID, 1, (24, 0), ((8,
0),),szBottomString)

	def SetChangePageLine(self, cPageLine=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2e, LCID, 1, (24, 0), ((2,
0),),cPageLine)

	def SetColAlign(self, szFmt=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x39, LCID, 1, (24, 0), ((8,
0),),szFmt)

	def SetColNums(self, nColNums=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0xf, LCID, 1, (24, 0), ((2,
0),),nColNums)

	def SetFixedLeftMargin(self, nFixedLeftMargin=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x11, LCID, 1, (24, 0), ((2,
0),),nFixedLeftMargin)

	def SetFixedTopMargin(self, nFixedTopMargin=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x10, LCID, 1, (24, 0), ((2,
0),),nFixedTopMargin)

	def SetGridHeadLineColor(self,
nGridHeadLineColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1b, LCID, 1, (24, 0), ((3,
0),),nGridHeadLineColor)

	def SetGridLineColor(self, nGridLineColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1c, LCID, 1, (24, 0), ((3,
0),),nGridLineColor)

	def SetPageFoot(self, szPageFoot=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x37, LCID, 1, (24, 0), ((8,
0),),szPageFoot)

	def SetPageFootAlign(self, cAlign=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1d, LCID, 1, (24, 0), ((2,
0),),cAlign)

	def SetPageFootFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x25, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetPageFootLineColor(self,
nPageFootLineColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x19, LCID, 1, (24, 0), ((3,
0),),nPageFootLineColor)

	def SetPageHead(self, szPageHead=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x36, LCID, 1, (24, 0), ((8,
0),),szPageHead)

	def SetPageHeadAlign(self, cAlign=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x21, LCID, 1, (24, 0), ((2,
0),),cAlign)

	def SetPageHeadFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x26, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetPageHeadLineColor(self,
nPageHeadLineColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1a, LCID, 1, (24, 0), ((3,
0),),nPageHeadLineColor)

	def SetRecordAlign(self, cAlign=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1e, LCID, 1, (24, 0), ((2,
0),),cAlign)

	def SetRecordBackColor(self, nRecordBackColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x3b, LCID, 1, (24, 0), ((3,
0),),nRecordBackColor)

	def SetRecordBottomMargin(self,
nRecordBottomMargin=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x16, LCID, 1, (24, 0), ((2,
0),),nRecordBottomMargin)

	def SetRecordFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x23, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetRecordHeight(self, nRecordHeight=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x15, LCID, 1, (24, 0), ((2,
0),),nRecordHeight)

	def SetRecordTopMargin(self, nRecordTopMargin=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x17, LCID, 1, (24, 0), ((2,
0),),nRecordTopMargin)

	def SetReportHeadBackColor(self,
nReportHeadBackColor=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x3a, LCID, 1, (24, 0), ((3,
0),),nReportHeadBackColor)

	def SetReportHeadFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x24, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetReportHeadHeight(self,
nReportHeadHeight=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x14, LCID, 1, (24, 0), ((2,
0),),nReportHeadHeight)

	def SetReportSubTitle(self, szFormat=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2a, LCID, 1, (24, 0), ((8,
0),),szFormat)

	def SetReportSubTitleHeight(self,
nReportSubTitleHeight=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x13, LCID, 1, (24, 0), ((2,
0),),nReportSubTitleHeight)

	def SetReportSubTitleRowsCols(self,
nReportSubTitleRows=defaultNamedNotOptArg,
nRecordSunTitleCols=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x18, LCID, 1, (24, 0), ((2, 0),
(2, 0)),nReportSubTitleRows, nRecordSunTitleCols)

	def SetReportTitle(self, szReportTitle=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x38, LCID, 1, (24, 0), ((8,
0),),szReportTitle)

	def SetReportTitleHeight(self,
nReportTitleHeight=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x12, LCID, 1, (24, 0), ((2,
0),),nReportTitleHeight)

	def SetReportTotal(self, szFormat=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x29, LCID, 1, (24, 0), ((8,
0),),szFormat)

	def SetSubTitleAlign(self, cAlign=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x1f, LCID, 1, (24, 0), ((2,
0),),cAlign)

	def SetSubTitleFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x27, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetTitleAlign(self, cAlign=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x20, LCID, 1, (24, 0), ((2,
0),),cAlign)

	def SetTitleFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x28, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetTopString(self, szTopString=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2b, LCID, 1, (24, 0), ((8,
0),),szTopString)

	def SetTotalFormat(self, szFontName=defaultNamedNotOptArg,
cxSize=defaultNamedNotOptArg, cySize=defaultNamedNotOptArg,
nTextColor=defaultNamedNotOptArg, bBold=defaultNamedNotOptArg,
bItalic=defaultNamedNotOptArg, bUnderline=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x22, LCID, 1, (24, 0), ((8, 0),
(2, 0), (2, 0), (3, 0), (11, 0), (11, 0), (11, 0)),szFontName, cxSize,
cySize, nTextColor, bBold, bItalic, bUnderline)

	def SetWidthString(self, szWidthString=defaultNamedNotOptArg):
		return self._oleobj_.InvokeTypes(0x2d, LCID, 1, (24, 0), ((8,
0),),szWidthString)

	_prop_map_get_ = {
		"Appearance": (-520, 2, (2, 0), (), "Appearance", None),
		"BackColor": (-501, 2, (19, 0), (), "BackColor", None),
		"BorderStyle": (-504, 2, (2, 0), (), "BorderStyle", None),
		"Caption": (-518, 2, (8, 0), (), "Caption", None),
		"Enabled": (-514, 2, (11, 0), (), "Enabled", None),
		# Property 'Font' is an object of type 'Font'
		"Font": (-512, 2, (9, 0), (), "Font",
'{BEF6E003-A874-101A-8BBA-00AA00300CAB}'),
		"FontBold": (6, 2, (11, 0), (), "FontBold", None),
		"FontItalic": (7, 2, (11, 0), (), "FontItalic", None),
		"FontName": (4, 2, (8, 0), (), "FontName", None),
		"FontSize": (5, 2, (2, 0), (), "FontSize", None),
		"FontUnderline": (8, 2, (11, 0), (), "FontUnderline", None),
		"FootLine": (13, 2, (11, 0), (), "FootLine", None),
		"ForeColor": (-513, 2, (19, 0), (), "ForeColor", None),
		"HeadLine": (14, 2, (11, 0), (), "HeadLine", None),
		"Height": (10, 2, (2, 0), (), "Height", None),
		"ReadyState": (-525, 2, (3, 0), (), "ReadyState", None),
		"Visible": (11, 2, (11, 0), (), "Visible", None),
		"Width": (9, 2, (2, 0), (), "Width", None),
		"bHeadLine": (12, 2, (11, 0), (), "bHeadLine", None),
		"hWnd": (-515, 2, (3, 0), (), "hWnd", None),
	}
	_prop_map_put_ = {
		"Appearance" : ((-520, LCID, 4, 0),()),
		"BackColor" : ((-501, LCID, 4, 0),()),
		"BorderStyle" : ((-504, LCID, 4, 0),()),
		"Caption" : ((-518, LCID, 4, 0),()),
		"Enabled" : ((-514, LCID, 4, 0),()),
		"Font" : ((-512, LCID, 4, 0),()),
		"FontBold" : ((6, LCID, 4, 0),()),
		"FontItalic" : ((7, LCID, 4, 0),()),
		"FontName" : ((4, LCID, 4, 0),()),
		"FontSize" : ((5, LCID, 4, 0),()),
		"FontUnderline" : ((8, LCID, 4, 0),()),
		"FootLine" : ((13, LCID, 4, 0),()),
		"ForeColor" : ((-513, LCID, 4, 0),()),
		"HeadLine" : ((14, LCID, 4, 0),()),
		"Height" : ((10, LCID, 4, 0),()),
		"ReadyState" : ((-525, LCID, 4, 0),()),
		"Visible" : ((11, LCID, 4, 0),()),
		"Width" : ((9, LCID, 4, 0),()),
		"bHeadLine" : ((12, LCID, 4, 0),()),
		"hWnd" : ((-515, LCID, 4, 0),()),
	}

# Little helper to transform event arguments to more usable objects.
from pywintypes import UnicodeType
dispatchType = pythoncom.TypeIIDs[pythoncom.IID_IDispatch]
def transformarg(arg):
	if type(arg)==dispatchType:
		return Dispatch(arg)
	elif type(arg)==UnicodeType:
		return str(arg)
	return arg
arg_transformer = lambda object, args: map(transformarg, args)

class _DReportOcxEvents:
	"Event interface for ReportOcx Control"
	CLSID = CLSID_Sink =
pythoncom.MakeIID('{AF07ADBE-9F30-4E5E-B377-B9ABD83B0E28}')
	_public_methods_ = [] # For COM Server support
	_arg_transformer_ = arg_transformer
	_dispid_to_func_ = {
		        1 : "OnUserError",
		}

	def __init__(self, oobj = None):
		if oobj is None:
			self._olecp = None
		else:
			import win32com.server.util
			cpc=oobj._oleobj_.QueryInterface(pythoncom.IID_IConnectionPointContainer)
			cp=cpc.FindConnectionPoint(self.CLSID_Sink)
			cookie=cp.Advise(win32com.server.util.wrap(self))
			self._olecp,self._olecp_cookie = cp,cookie
	def __del__(self):
		try:
			self.close()
		except pythoncom.com_error:
			pass
	def close(self):
		if self._olecp is not None:
			cp,cookie,self._olecp,self._olecp_cookie =
self._olecp,self._olecp_cookie,None,None
			cp.Unadvise(cookie)
	def _query_interface_(self, iid):
		import win32com.server.util
		if iid==self.CLSID_Sink: return win32com.server.util.wrap(self)

	# Handlers for the control
	# If you create handlers, they should have the following prototypes:
#	def OnUserError(self, nErrCode=defaultNamedNotOptArg):

class CoClassBaseClass:
	def __init__(self, oobj=None):
		if oobj is None: oobj = pythoncom.new(self.CLSID)
		self.__dict__["_dispobj_"] = self.default_interface(oobj)
	def __repr__(self):
		return "<win32com.gen_py.%s.%s>" % (__doc__,
self.__class__.__name__)

	def __getattr__(self, attr):
		d=self.__dict__["_dispobj_"]
		if d is not None: return getattr(d, attr)
		raise AttributeError, attr
	def __setattr__(self, attr, value):
		if self.__dict__.has_key(attr): self.__dict__[attr] = value; return
		try:
			d=self.__dict__["_dispobj_"]
			if d is not None:
				d.__setattr__(attr, value)
				return
		except AttributeError:
			pass
		self.__dict__[attr] = value

# This CoClass is known by the name 'REPORTOCX.ReportOcxCtrl.1'
class ReportOcx(CoClassBaseClass): # A CoClass
	# ReportOcx Control
	CLSID = pythoncom.MakeIID("{856933B7-BD84-4308-AA83-486418BC9EC3}")
	coclass_sources = [
		_DReportOcxEvents,
	]
	default_source = _DReportOcxEvents
	coclass_interfaces = [
		_DReportOcx,
	]
	default_interface = _DReportOcx

RecordMap = {
}

CLSIDToClassMap = {
	'{AF07ADBE-9F30-4E5E-B377-B9ABD83B0E28}' : _DReportOcxEvents,
	'{856933B7-BD84-4308-AA83-486418BC9EC3}' : ReportOcx,
	'{BE1E9308-9F3C-4D83-B7BC-D0F6F6DCE778}' : _DReportOcx,
}
CLSIDToPackageMap = {}

win32com.client.CLSIDToClass.RegisterCLSIDsFromDict( CLSIDToClassMap )


From skip at pobox.com  Wed Feb 13 18:58:15 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 13 Feb 2002 17:58:15 -0600
Subject: Simple Question
In-Reply-To: <6QCa8.7212$zj7.1668243@news1.telusplanet.net>
References: <6QCa8.7212$zj7.1668243@news1.telusplanet.net>
Message-ID: <15466.64919.829009.596007@12-248-41-177.client.attbi.com>

    Tristan> Is there a way to pass references or pointers as parameters to
    Tristan> functions in python?

In a manner similar to using & in C?  No.  As a hackish way to do things?
Sure.  However, there's probably a better (that is, "more Pythonic") way
that doesn't involve needing pass-by-reference semantics.  You'll have to
tell us more about what you're trying to do for us to suggest something
plausible.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From mwh at python.net  Wed Feb 13 08:25:25 2002
From: mwh at python.net (Michael Hudson)
Date: Wed, 13 Feb 2002 13:25:25 GMT
Subject: Re-throwing an exception without loosing the stack trace
References: <mailman.1013552683.30067.python-list@python.org> <Xns91B4F52BC58cliechtigmxnet@62.2.16.82> <3C69BB96.369FC648@stroeder.com>
Message-ID: <uheolh5p1.fsf@python.net>

Michael Str?der <michael at stroeder.com> writes:

> Chris Liechti wrote:
> > 
> > "Clark C . Evans" <cce at clarkevans.com> wrote in
> > news:mailman.1013552683.30067.python-list at python.org:
> > 
> > > How do I catch and then re-throw an exception without
> > > loosing the information about the stack trace?  I've
> > > looked around alot and don't see the "obvious" answer.
> > 
> > try a simple, naked "raise" ;-)
> 
> Is that thread-safe?

Yes (the exception data gets stored in the threadstate).

Cheers,
M.

-- 
  languages shape the way we think, or don't.
                                        -- Erik Naggum, comp.lang.lisp


From klaatu at evertek.net  Tue Feb 26 19:21:51 2002
From: klaatu at evertek.net (Mike Dean)
Date: Tue, 26 Feb 2002 18:21:51 -0600
Subject: Find an Item in a Sorted List
In-Reply-To: <3c7c1b5b.34558051@news.la.sbcglobal.net>; from hankc@nospam.com on Tue, Feb 26, 2002 at 11:50:58PM +0000
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <20020226182151.A2940@evertek.net>

Greetings, and welcome to Python!

Python does have a (very fast) way to do a search such as you are
looking at with its Dictionaries (hash tables in other languages).  For
example:

# The data - these can be lines in a file rather than a list of strings;
# in that case, use the file reading and looping code in a previous
# recommendation
data = ['Arty,1000',
'Bobby,2000',
'Charlie,3000']

d2={}    # Initialize an empty dictionary

for line in data:
    d2[line.split(',')[0]] = line    # Store the line in the dictionary,
                                     # using the first comma-delimited
                                     # item as the key

# Later on...
print d2['Bobby']

Dictionary searches are implemented in C in the Python core, and execute
very quickly (IIRC, they use b-tree indexes so that the search time is
relatively flat, even with very large dictionaries).

HTH,
Michael


P.S.  I remembered to test my code this time :-)

* HankC <hankc at nospam.com> [2002-26-02 23:50]:
> Greetings:
> 
> I'm 2 days new to Python and hope I can get some pointers here...
> 
> I have a sorted, delimited text file such as:
> 
> Arty,1000
> Bobby,2000
> Charlie,3000
> 
> Knowing a name, I want to return the entire line from the file.  My
> approach/problems (any comments appreciated):
> 
> # use the filter method
> # filter fxn
> def Match(line): return line[:toklen] == tok
> 
> # find match fxn
> def FindMatch:
>       fp = open(fname)
>       flist = fp.readlines()
>       dataline = filter(Match, eodflist)
> 
> My problem is that I don't understand how, if it's even possible, to
> set tok and toklen values in the FindMatch fxn and pass them to Match.
> Is it possible to pass multiple values to the function used by
> filter()?
> 
> More general...  does filter() have any performance advantage over
> just using a for loop?  The datafile list has 10,000+ items.  My
> background is in object pascal where I would read the items into a
> list then use a binary search to find the target.  I'm hoping there's
> an efficient python method to perform a search easily on a sorted
> list.
> 
> Suppose I want to make tok and toklen global to the module.  Is this
> possible?
> 
> TIA!
> -- 
> http://mail.python.org/mailman/listinfo/python-list



From quinn at spew.ugcs.caltech.edu  Fri Feb  8 01:43:59 2002
From: quinn at spew.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Feb 2002 06:43:59 GMT
Subject: Standing Objects ... possible?
References: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>
Message-ID: <slrna66std.qu3.quinn@spew.ugcs.caltech.edu>

On 2 Feb 2002 20:25:52 GMT, Philip Swartzleonard <starx at pacbell.net> wrote:
>    	I have this vague idea that this would be useful in the near 
>future, and I was wondering if it was at all possible without recoding 
>the interperter. What i want is a construct like:
>
>object Foo( ancestor ):
>    	pass
>
>    	or more likely:
>
>class Foo( ancestor, standing_mixin ):
>    	pass
>
>    	to be more or less equivelent to:
>
>class _Foo( ancestor ):
>    	pass
>Foo = _Foo()
>
>    	The second option looked easy at one point, but I don't think that 
>I can make it self-initiate... In any case, is there a prefered way of 
>writing hidden stuff like the _Foo for doing it the more possible way?

Well, the obvious is:

class foo(ancestor):
    pass
foo = foo()

That'll get the name out of the way at least.  You could also del it, if it's
really causing you aesthetic pain.

I've no clue what you mean by "self-initiate".  Of course, 'foo's __init__
will be called, as always.  I've never heard the term "standing object" before.
If it's something that's obscure, or which you just made up, it could stand a
little more elaboration.


From kfarmer at thuban.org  Mon Feb 25 03:14:39 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 25 Feb 2002 08:14:39 GMT
Subject: PyAppBar/wxAppBar opened at SourceForge
References: <adec2554.0202172056.62b0f735@posting.google.com> <Xns91B92597C587BRASXnewsDFE1@130.133.1.4>
Message-ID: <Pnme8.42894$nf3.1803412783@newssvr14.news.prodigy.com>

"Philip Swartzleonard" <starx at pacbell.net> wrote in message
news:Xns91B92597C587BRASXnewsDFE1 at 130.133.1.4...
> tell me what to do with this... maybe i'm looking in the wrong places

Install.  Run the demo.  This isn't anything more than a proof of concept,
and it's not all that complex.

> but i couldn't even find definate information on something both tiny and
> important: which OS(s) it is designed for. I can only guess that it
> works on windows due to the DLLs available for download.

See what you quoted...

>      This project aims to create Application Toolbars (a class of
>      dockable window under Win32) in Python. Examples of AppBars

However, I'd be just as happy if an API was developed to deal with these
things in a generic way.  I'm not certain, though, that it's particularly
feasible on anything less than an OS-by-OS basis.  Certainly, *I* think this
sort of thing is dreadfully missing from the Python GUIs I've dealt with
(PythonWin, wxPython, Tkinter).

> It is a nice idea though, if it does what i think it does =).

Well, it's been done before, certainly, but as far as I can tell, I'm the
only one to have bothered posting code.  If the resize/move bugs could be
squashed by someone more skilled in win32 than I, it'd be basically done for
what I'm wanting.  Even with the bugs (which I never encounter since I never
bother undocking the thing) it's proven useful for me.

Thinking of tackling DeskBands soon.  Might be cleaner.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org





From DeepBleu at DeepBleu.org  Fri Feb  1 14:51:24 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Fri, 1 Feb 2002 13:51:24 -0600
Subject: telnetlib read_until()( question
References: <u5lqt76m0ogfd8@corp.supernews.com> <a3er56$17h5lp$1@ID-11957.news.dfncis.de>
Message-ID: <u5lsfu2m34gpa9@corp.supernews.com>

Thanks Emile
I am trying to find out why a telnet connection (using a Python script)
dropped out.
DeepBleu

"Emile van Sebille" <emile at fenx.com> wrote in message
news:a3er56$17h5lp$1 at ID-11957.news.dfncis.de...
>
> "DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
> news:u5lqt76m0ogfd8 at corp.supernews.com...
> > For telnetlib's read_until(expected [,timeout])
> >
> > a.  Does anyone know the default value of the timeout when no timeout
> is
> > given, and
> > b.  When a timeout is given:  Is the numeric value in seconds or
> mintues?
>
> telnetlib uses select.select, passing the timeout value.  From the
> select docs:
>
> FUNCTIONS
>     select(...)
>         select(rlist, wlist, xlist[, timeout]) -> (rlist, wlist, xlist)
> [snip]
>         The optional 4th argument specifies a timeout in seconds; it may
> be
>         a floating point number to specify fractions of seconds.  If it
> is absent
>         or None, the call will never time out.
>
>
> HTH,
>
> --
>
> Emile van Sebille
> emile at fenx.com
>
> ---------
>




From huaiyu at gauss.almadan.ibm.com  Tue Feb 19 14:17:35 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Tue, 19 Feb 2002 19:17:35 +0000 (UTC)
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <slrna74ssf.21s.QnickQm@localhost.localdomain>
Message-ID: <slrna7596e.3v7.huaiyu@gauss.almadan.ibm.com>

On Tue, 19 Feb 2002 15:44:09 GMT, Nick Mathewson <QnickQm at alum.mit.edu>
wrote: 
>Making a real static type system is harder than you'd think.
>Consider this function:
>
>def fn(lst,item):
>   for i in lst:
>     if i == item:
>        return repr(i)
>
>What is the type of fn?  You could call it:
>
>     (list[int], int) -> str
>
>But then you'd lose the polymorphism of the function.  To a first
>approximation, you might say:

[snip excellent analysis of need of polymorphism]

>IMNSHO, some kind type inference is the only tractable answer.

This is an excellent point.  A static typechecking system added to Python
will be overly restrictive unless it has the power of polymorphic type
systems found in some functional programming languages (ML, Haskell).

To further expand on this point, a full typechecking system in this example
must be able to handle the following type semantics:

fn :: I(A), B -> str where:

    # These can be inferred automatically by type checker
    I in Iter
    (A, B) in Eq
    A in Repr

    # These can be imported as standard Python design interfaces
    typeclass Iter(I): I(A) => (def I(A).next() :: A) and (raises StopIteration)
    typeclass Eq(A, B): def A.__eq__(B) :: bool
    typeclass Repr(A): def A.__repr__() :: str

It is not necessary to write all these on the spot --- predefined
typeclasses can be imported.  And it is not necessary to make all of them
explicit --- much can be derived automatically using type inference as in ML
or Haskell.  But the point is that a whole new set of mechanisms must be
developed to represent and handle such type semantics.

Such a system would certainly be marvelous to have.  It is certainly doable.
It is likely to be difficult.  Sounds like a good candidate for a
challenging project.

Huaiyu


From l0819m0v0smfm001 at sneakemail.com  Mon Feb 25 23:50:59 2002
From: l0819m0v0smfm001 at sneakemail.com (Joshua Macy)
Date: Tue, 26 Feb 2002 04:50:59 GMT
Subject: What's wrong with this subroutine?
References: <Xns91C17CC1E3820AusYourStandingInIt@130.133.1.4>
Message-ID: <3C7ADA4C.8090705@sneakemail.com>

A.Newby wrote:

> First, here's what I'm trying to do ......
> 
> I'm working on a html, cgi based chat script. It's working pretty well so 
> far. 
> 
> I want to get this subroutine so that the user just has to post a url into 
> the chat box to get a post a picture or a link in chat. For e.g, to post a 
> pic of a walrus, you would simply type ....
> 
> http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg
> 
> ... and the script would take that, and before writing it to the chat log 
> it would convert it to ...
> 
> <img src = http://www.pbs.org/kratts/world/oceans/walrus/images/walrus.jpg>
> 
> ... or to post a link, the user would simply type ... 
> 
> http://www.goodvibes.com/
> 
> ... and the script would convert it to ....
> 
> <a href = "http://www.goodvibes.com/>Link!</a>
> 
> 
> Here's the subroutine I'm using to try and do this.... it is called from a 
> function named "infilter" which filters the posts before writing them to 
> the chat log ...
> 
> def search4URL(string):
>     #the following bits are used to alter the post when it finds the URL
>     beginpic = " <img src =" 
>     endpic = ">"
>     beginlink = " <a href ="
>     endlink = " target = blank>Link</a>"
>     
>     URLStarts = string.find(' http:') #space is so when the function     
> 	    	    	    	    	    	  #repeats, it won't convert the same 
>     	    	    	    	    	    	  #block of text again
> 



   What happens when there are no more ' http:' patterns in your string 
and string.find returns -1?  I would try this either with a regular 
expression (r'(http://\S+)' would probably do the trick), or in a while 
loop that makes sure that there is something to replace before trying to 
replace it:

while string.find(filterstring, ' http:') > -1:
     filterstring = search4URL(filterstring)

or even by chewing up the string so the search doesn't have to keep 
starting from the beginning:

newstring = ''
while filterstring:
    nextHttp = string.find(filterstring, ' http:')
    if nextHttp > -1:
         newstring = newstring + filterstring[:nextHttp]
         linkstring, filterstring = search4URL(filterstring[nextHttp:]) 

         newstring  = newstring + linkstring
    else:
       newstring = newstring + filterstring

where search4URL returns the link string that it constructs and the 
remainder of the filterstring seperately.

   BTW, just as a stylistic matter, it's probably better not to shadow 
the string module with a local variable named 'string' inside your 
function.  This works in your current function because 'string' contains 
a string, and strings now all have methods such as find that have the 
same name as the module-level function (i.e. 'this is a 
string'.find('this') does the same thing as string.find('this is a 
string', 'this')), but that's not generally the case--if you 
accidentally did this with a module like math or re it would be a bug. 
Also, the intent would be clearer if you called the function something 
like replaceURLwithLink instead of search4URL, since it doesn't just 
search for the url.

   I hope this helps,

   Joshua



From grante at visi.com  Fri Feb 22 11:54:24 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 22 Feb 2002 16:54:24 GMT
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4> <Nktd8.1458$N7.276578@ruti.visi.com>
Message-ID: <4Jud8.1483$N7.287257@ruti.visi.com>

In article <Nktd8.1458$N7.276578 at ruti.visi.com>, Grant Edwards wrote:
> In article <Xns91B6DFD6A36CDRASXnewsDFE1 at 130.133.1.4>, Philip Swartzleonard wrote:
> 
>>> Personally, I would like to see octal number entry migrated from 0---
>>> to something like 0o---.  This would bring things more in line with
>>> the entry of hex numbers (0x---)
>>> 
>>> I know it's only a minor thing, but it'd be nice if my numbers behaved
>>> as I expected...
>> 
>> Steve has mentioned that UNIX people are used to it... 
> 
> I'm a Unix person, I'm used to it, and I still think it sucks.
> 
>> i think it's a little deeper than that, b'cause as far as i
>> know, _every_ programming language that has ever supported
>> octal does it this way...
> 
> And I still think it sucks.  :)

I just got bit by the leading-zero thing again about 10 minutes
ago when I changed the minor version number of something from 07
to 08.  Had to skip from 1.07 to 1.10...

-- 
Grant Edwards                   grante             Yow!  With this weapon
                                  at               I can expose fictional
                               visi.com            characters and bring about
                                                   sweeping reforms!!


From jeff at ccvcorp.com  Thu Feb 21 14:45:20 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 11:45:20 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com> <3C751FEA.24CEFBBC@ford.com>
Message-ID: <3C754E4F.271F00C1@ccvcorp.com>


Jeff Henry wrote:

> Roy Smith wrote:
> >
> > Roy Smith <roy at panix.com> wrote:
> > > PS -- In the lists of false things in python that have been mentioned on
> > > this thread, I don't think I've seen anybody mention {}, the empty
> > > dictionary.
> >
> > Along those same lines, I half-expected the empty function to be false,
> > i.e.:
> >
> > def foo():
> >    pass
> >
> > if (foo):
>
> Oops! I think you meant to say -
>
> if (foo()):

No, he meant the first way.  The intent is to test whether an "empty" code block
is considered false or not.  It's not -- even "empty", there's something there,
so it's considered true.

(This may not be so surprising, when you consider that this function is *not*
equivalent to a NOP -- it creates a frame object, passes control to it, returns
None, and destroys the frame object.)

Jeff Shannon
Technician/Programmer
Credit International




From tdelaney at avaya.com  Tue Feb  5 17:52:48 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 6 Feb 2002 09:52:48 +1100 
Subject: when does python free memory?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A168@natasha.auslabs.avaya.com>

> From: reanes at banta-im.com [mailto:reanes at banta-im.com]

> The first number that comes out of the "print memsize" statement is
> the pid, the second is the amount of memory used.  It doesn't seem to
> be actually freeing the memory that was allocated to variable z.  Now
> I played around with it a little more and if I reallocate another list
> it does re-use that memory:
> 
> So apparently python knows that the memory formerly allocated to z is
> available, but it doesn't tell the OS.  Is this the case, and if so,
> is there any way around it?  I have an application that needs to run
> 80-150 python processes that can stay around for 15min to an hour, so
> I'd like them not to hold onto the maximum amount of memory that they
> use throughout the whole life of the process.

This is normal with most applications - they grow, but rarely shrink their
total used memory. It depends primarily on the C library memory allocator
(malloc) implementation for the platform.

If you want 80-150 python processes running, each with their own copy of the
python VM, you are going to end up with 80-150 copies of the VM using up
memory. This in itself may or may not be significant, however it leads to
the same problem - each instance of the VM will grow its memory usage, but
usually not shrink.

Normally a better way would be to either have each "process" running as a
different Python thread, or build Python as a shared library - then the
"processes" would then be able to reuse memory freed by a different
"process" (at least, I think this would work with a shared library - it will
definitely work with threads) leading to a smaller amount of memory being
used overall.

Tim Delaney



From jkraska at san.rr.com  Tue Feb 19 10:19:44 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 19 Feb 2002 15:19:44 GMT
Subject: Adding static typing to Python
References: <ur047u8lrm308jbivi5h8g7ld4dugfb5i7@4ax.com> <a4tb3n$pc0$1@thorium.cix.co.uk>
Message-ID: <pmr47uovm87hkng4ma22h231fdrbq1rmnv@4ax.com>

>I haven't worked out 2.2 either.  But Jython does present Java's getters 
>and setters as attributes.

Oh, yeah. That's pretty cool. What they've done here is put a
Python getattr/setattr adapter around the BeanInfo interface
for Java. 

C//



From joshm at taconic.net  Tue Feb 26 18:52:23 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 26 Feb 2002 18:52:23 -0500
Subject: Oddity question
References: <mailman.1014759240.15400.python-list@python.org>
Message-ID: <3c7c1d54_2@corp.newsgroups.com>

FAILS>         self.buf=self.buf+self.my_socket.recv (8)

OK>         aux=self.my_socket.recv (8)
OK>         self.buf=self.buf+aux

This is just a guess, but here goes.  The call to recv() takes some amount
of time.  Perhaps Python is evaluating the "self.buf+" part *before* calling
recv().  It caches the old buffer contents.  Meanwhile, your other thread
prints and clears the value of self.buf.  When recv() returns, it appends
its results to the *already evaluated* self.buf and resets it.  When you use
aux, the caching of self.buf doesn't happen.

you could test this by trying something like this:

cache = self.buf
aux = self.my_socket.recv(8)
self.buf = cache + aux

This should also fail.

But the larger issue here is that what you are experiencing is a "race
condition".  Two threads are each trying to modify the same object
(self.buf), and they are interfering with each other.  A better way to deal
with this is to write separate functions (perhaps named "AppendBuffer" and
"ClearBuffer"?) which use a common mutex.  A mutex is a magic lock where
only one thread can own it at a time.  This makes it impossible for one
thread to muck around with the buffer while the other thread is modifying
it.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Race conditions suck to debug')




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


From aries at bitstream.net  Sun Feb 17 13:00:12 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Sun, 17 Feb 2002 18:00:12 +0000
Subject: help - inserting values into database tables
Message-ID: <3C6FEFAB.DE9907CC@bitstream.net>

Hey all,

Ok I am wondering something:
I have a python list which has values I would like to insert into a
database, the
database is a microsoft (ugh)  Access type of database.
-- I don't know much about the python for windows side of things like
ODBC
----Is this type of thing possible ????????????

What are some basic things I need to know about this problem space???
Where shall I look (on the web) for guideance for this type of thing ???

Basically the MS - Access database tables do exist and I want to insert
   info from my python list into it.  But I have no clue if this can be
done.
   Can I connect to the MS - Access database with a python script to
    insert my list into the database ??????????

Thanks,
Joe F.
-----



From emile at fenx.com  Tue Feb 12 07:57:49 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 12 Feb 2002 04:57:49 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <3C688CED.4583E7DA@cosc.canterbury.ac.nz>
Message-ID: <a4b3r4$1dgjaa$1@ID-11957.news.dfncis.de>

"Greg Ewing"

> Consider the fact that most right-handed
> guitar players use the left hand for fingering, and the
> right hand for the relatively dextrously undemanding
> function of strumming. Why is that?
>

When you move past folk music to classical guitar, the situation
reverses.  Then you use the left hand for the relatively dextrously
undemanding function of fingering, while the right hand produces the
sound.  ;-)


--

Emile van Sebille
emile at fenx.com

---------



From phr-n2002a at nightsong.com  Thu Feb 21 03:12:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 00:12:34 -0800
Subject: strings and sort()
References: <mailman.1014274835.21287.python-list@python.org>
Message-ID: <7xu1sbdzil.fsf@ruckus.brouhaha.com>

"Sean 'Shaleh' Perry" <shalehperry at attbi.com> writes:
> > I don't understand this reply.  The sort method would be no less efficient
> > if it returned the sequence:
> > 
> >   def nsort(a):
> >      a.sort()
> >      return a
> > 
> of course it would.  returning has a cost.  Most uses of sort() just
> need to in place update the list.  If there was actually a copy made
> as some people expect that would be even more costly.

Huh?  The current version of sort returns None, which has exactly
the same cost as returning the sorted list.  If you look at the
definition of nsort above, it doesn't copy the list.  It sorts it
in place and then returns the sorted list instead of returning None.

> When programming in a language, you learn the idioms of that language.

When an "idiom" is really a "workaround", it's sometimes preferable
to fix it rather than to get used to it.


From Scirious at Scirious.com  Thu Feb 14 17:00:48 2002
From: Scirious at Scirious.com (Scirious)
Date: Thu, 14 Feb 2002 22:00:48 GMT
Subject: Java vs Python ... ???
Message-ID: <Xns91B5CBEF0A103Scirious@140.99.99.131>

Dear friends;

I've been learning JAVA for a while, but now I've heard avout Python.
Any coments concerning these two languages?

TIA;
Scirious.


From sdm7g at Virginia.EDU  Tue Feb 19 16:43:03 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 16:43:03 -0500 (EST)
Subject: execfile question
In-Reply-To: <18682.020219@storm.ru>
Message-ID: <Pine.OSX.4.43.0202191640240.1674-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, Kerim Borchaev wrote:
> And can you explain who holds a reference on C instance in my example
> when 'execfile' completes?

Python 2.2 (#36, Jan 30 2002, 17:50:25)
[GCC 2.95.2 19991024 (release)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> print execfile.__doc__
execfile(filename[, globals[, locals]])

Read and execute a Python script from a file.
The globals and locals are dictionaries, defaulting to the current
globals and locals.  If only globals is given, locals defaults to it.

>>> execfile( 'executable.py' )
it's me
init
>>> dir()
['C', '__builtins__', '__doc__', '__name__', 'c']
>>> del c
deleted
>>>

-- Steve Majewski


> ----------------------------------
> My original question:
>
> > Can someone explain why while running this script I don't see C.__del__
> > eventually called?
> >
> > For me the output of it's execution is:
> >
> > >main.py
> > it's me
> > init
> > executed
> > >
> >






From winelight at spamcop.net  Wed Feb 20 16:13:58 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Wed, 20 Feb 2002 21:13:58 -0000
Subject: threads and urllib
References: <b03e80d.0202201228.76ac8525@posting.google.com>
Message-ID: <nkUc8.19043$kX6.133584@NewsReader>

>         self.webpage = urllib.urlopen(self.url).read()

Yes, this can cause Python to hang (thread passim).

It *seems* that using httplib() instead *might* avoid the hang.

Andrew ward





From bc90021 at bc90021redundantdot.net  Mon Feb 18 16:43:02 2002
From: bc90021 at bc90021redundantdot.net (bc90021)
Date: Mon, 18 Feb 2002 21:43:02 GMT
Subject: Newbie question...
References: <3C716B27.3090408@bc90021redundantdot.net>
Message-ID: <3C71750E.9020305@bc90021redundantdot.net>

X-No-Archive: Yes

> error: failed dependencies:
>     rpmlib(PartialHardlinkSets) <= 4.0.4-1 is needed by python-2.2-7


Yes, I really like answering my own posts.  ;)

I found the below regarding the above, but it doesn't really tell how to 
fix it... I'm upgrading my RPM rpm to 4.04, to see what happens.

http://groups.google.com/groups?hl=en&threadm=20020202063217.Z12158%40eastham.
redhat.com&rnum=4&prev=/groups%3Fhl%3Den%26q%3D%2522
rpmlib(PartialHardlinkSets)%2522%26btnG%3DGoogle%2BSearch

Sorry for the link formatting...

bc90021



From mhammond at skippinet.com.au  Thu Feb 21 23:21:31 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 22 Feb 2002 04:21:31 GMT
Subject: win32file.AllocateReadBuffer(..)
References: <Pine.GSU.4.21.0202211448560.17401-100000@garcia.efn.org>
Message-ID: <3C75C7C4.6020808@skippinet.com.au>

Spencer Ernest Doidge wrote:
> If I call win32file.AllocateReadBuffer(..) in a method that gets called
> repeatedly and terminates each time it is done, will I have a memory leak
> if I don't somehow free the allocated memory before concluding the run of
> the method? Or does the memory get de-allocated automagically when the
> method is finished running?

Normal Python reference counting semantics apply - ie, when there are no 
references to the buffer it will be deleted.  Thus, there should be no 
leaks.

Mark.



From aleax at aleax.it  Mon Feb 18 11:45:57 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 16:45:57 GMT
Subject: Python License Issues
References: <mailman.1014039130.7808.python-list@python.org>
Message-ID: <9dac8.83655$3J1.2675331@news2.tin.it>

Michael Chermside wrote:
        ...
> Of couse I should point out that this is ALL theoretical. There is no
> way that you could provide such a service unless you are a good-sized
> insurance firm, and there is no corporate lawyer in the world who is
> going to allow such a policy to be written (despite the fact that (in my
> opinion) it would never have to pay out anything).

If your opinion that you'd never have to pay out on such policies
is firm enough, why would you need a good-sized insurance firm or
any corporate lawyers?  Set up in business for yourself (maybe in
Guernesey, Virgin Islands, wherever) and start selling these "you
pay me a premium and I'll assume a liability that will never come
to pass" policies, for whatever the market will bear.  If your
opinion is correct, you should make enough money to live out the
rest of your life in ease (or at least to get the kids an ice
cream now and then, if the market basically shares your own
opinion and thus 'will bear' little:-).  If your opinion turns
out to be wrong, you'll forfeit whatever sureties you put up to
convince your customers that your policies are worth buying, or,
if you have unlimited liability (as Lloyds used to), then you'll
forfeit basically everything.

My personal opinion on this issue isn't firm enough to prompt me
to do any such thing, but surely the profit potential for those
whose opinion _is_ firm indeed should be alluring...?


Alex



From gimbo at ftech.net  Tue Feb 26 05:13:55 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 10:13:55 +0000
Subject: Website to post code for peer review
In-Reply-To: <a5dl24.3vs4847.1@kserver.org>
References: <mailman.1014676907.27964.python-list@python.org> <a5dl24.3vs4847.1@kserver.org>
Message-ID: <20020226101355.GB18498@andy.tynant.ftech.net>

On Mon, Feb 25, 2002 at 03:25:55PM -0800, Sheila King wrote:

> There is no "community" site in the sense of the perlmonks.org site (AFAIK)
> Most Python discussion takes place on mailing lists or in newsgroups rather
> than on web sites.
> 
> The closest you may be able to come to what you're looking for at this
> time, is a combination of
> Useless Python: http://www.lowerstandard.com/python/whatsthepoint.html
> and a subscription to the Python-Tutor mailing list:
> http://mail.python.org/mailman/listinfo/tutor
> 
> I don't know about any "voting" or "experience points", but you can get
> your code reviewed and discussed.

There's also the ActiveState Python Cookbook - you can submit code and
people can comment on it and rate it.  I've never seen perlmonks.org
though so have no idea how they compare.  There's not an _enormous_
amount of stuff there, mind, ie it doesn't seem incredibly active.
But somewhat active.  :-)

http://aspn.activestate.com/ASPN/Cookbook/Python

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From kdahlhaus at yahoo.com  Wed Feb  6 08:03:35 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 6 Feb 2002 05:03:35 -0800
Subject: Web Hosting - one report
Message-ID: <283adf56.0202060503.e5bd15f@posting.google.com>

Time for the weekly where-can-I-cheaply-host-pythonic-web-app posting.

I thought I had found a place at 'Hostdepartment.com'  $5/month for
100Mb and a full shell. They have 1.6 installed.  You can compile your
own python, but many modules do not build.  It turns out that they
have disabled dynamic loading of modules from user accounts.  So if
you have a swigged extension, python can load any shadow files, but
the .so is never found.  That's a shame, as I was hoping to actually
post an *answer* to this reoccuring question.  Personally, I need to
be able to use extension modules - at least PIL, GD charts, etc.

(I was quite annoyed when after 3 days of notices that my trouble
ticket was being looked at, a message came that said 'your ticket is
completely resolved.  you cannot load dynamic modules from user
accounts'.  That's explained, not resolved.)

Anyway, so that's one report.  Has anyone found one of these
inexpensive providers that allows you to use your own native-language
extension modules?


From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 16:32:47 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 15:32:47 -0600
Subject: Compiling from Pythonwin?
References: <a3k0cf$dko$1@news.efn.org>
Message-ID: <000e01c1acfa$55118ea0$0101010a@local>

----- Original Message ----- 
From: "Spencer Doidge" <spencer at spencerdoidge.com>


> Is there a way to compile .py scripts from the Pythonwin GUI screen other
> than using compileFile(...)?

Any time you import a Python script from a drive where you have write 
access, it will be compiled automatically.  Your "top-level" script
ordinarily won't be, and generally doesn't need to be.

So,
    import mymodule

will compile mymodule.py to mymodule.pyc automatically.





From michele at diee.unica.it  Sun Feb 10 08:29:08 2002
From: michele at diee.unica.it (Michele Marchesi)
Date: Sun, 10 Feb 2002 14:29:08 +0100
Subject: XP2002 Conference - Call for Practitioner's Reports
Message-ID: <3C6675A4.2080203@diee.unica.it>

  ***** XP2002 - CALL FOR PRACTITIONER'S REPORTS *****

     3rd International Conference on eXtreme Programming
     and Agile Processes in Software Engineering - XP2002

         26-29 May, 2002, Alghero, Sardinia, Italy
                  http://www.xp2002.org
General Chair:  Kent Beck, Three Rivers Institute, USA
Program Chair:  Michele Marchesi, University of Cagliari, Italy
Program Co-chair:  Giancarlo Succi, University of Bolzano, Italy

GOAL and SCOPE:
Practitioners' Reports provide a lightweight format for presenting
late-breaking work that is not developed enough  for a full paper.

PhD candidates are particularly encouraged to submit reports describing
their work.

Submissions are solicited in all relevant areas of XP and Agile
Methodologies.

Accepted Practitioners' Reports will be presented as posters during
interactive sessions and published in conference  proceedings.

HOW TO SUBMIT:
The Practitioner's Report abstract should include a title, author names and
affiliations and an optional URL for  further reading or reference.

Submitted reports should follow the same format guidelines of regular
papers, but are limited to two A4 pages.

DATES:
Practitioner's Reports submission: February 28, 2002
Practitioner's Reports acceptance notification: March, 15, 2002
Final manuscript submission: April 1, 2002

For more information, including conference topics and instructions to
authors, please refer to XP2002 Web site: http://www.xp2002.org or send 
an e-mail to:  michele at diee.unica.it

----------------------------------------------------------------
Please forward this message to others who might be interested!
Apologize if you receive this announcement more than once.



From jason at jorendorff.com  Fri Feb  1 21:43:35 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 20:43:35 -0600
Subject: typetesting, adaptation, typeclasses, ... (was Re: isinstance() considered harmful)
In-Reply-To: <a3eks9$tb$1@panix2.panix.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKENGCPAA.jason@jorendorff.com>

Aahz wrote:
> Jason Orendorff wrote:
> > Alex Martelli wrote:
> > > 
> > > Refusing to guess in presence of ambiguity is one of Python's
> > > tenets, though.
> >
> > It's the language that doesn't guess.  The language designer is
> > notorious for picking one thing, seemingly arbitrarily, from a number
> > of reasonable alternatives.
>
> Mind expanding that?  (I could guess, but I don't want to flame you
> until I know for sure what you're saying. ;-)

Oh, I don't remember the context exactly.  It had to do with how
useful the recursive quality of typeclasses is.

My point was that when the BDFL has to choose between a number
of reasonable alternatives, he often picks one without seriously
trying to justify the choice.  (Hence "BDFL pronouncements"
in many PEPs: 201, 202, 223, 255...)  In PEP 255:  "No argument
on either side is totally convincing, so I have consulted my
language designer's intuition."

Or if you prefer: in the role of designing Python's behavior in
the presence of ambiguous input, the BDFL prefers to have Python
throw exceptions.  But when the BDFL himself is presented with
ambiguous input (or internal ambivalence), he (often enough)
makes a decision and moves on.

Flame away...  :)

## Jason Orendorff    http://www.jorendorff.com/



From dsh8290 at rit.edu  Thu Feb  7 16:01:27 2002
From: dsh8290 at rit.edu (dman)
Date: Thu, 7 Feb 2002 16:01:27 -0500
Subject: threads killing other threads
In-Reply-To: <3C62E0D5.99C586DF@ccvcorp.com>
References: <1013105970.135322@cswreg.cos.agilent.com> <3C62E0D5.99C586DF@ccvcorp.com>
Message-ID: <20020207210127.GB29888@dman.ddts.net>

On Thu, Feb 07, 2002 at 12:17:26PM -0800, Jeff Shannon wrote:
| Greg Weeks wrote:
| 
| > I've occasionally wanted thread objects to have an exit() method, so that
| > one thread could interrrupt and kill another.  This doesn't seem to be
| > available, perhaps because POSIX threads don't support it.
| >
| > Is that correct?
| >
| > Is there some reason why threads killing threads would be undesirable or
| > hard to implement?
| 
| I don't know the reasoning behind threads being unkillable (from outside),

Here's an analogy :

If someone is sitting on the toilet, then you come and kill him, how
is he supposed to flush?

Processes can be killed by other processes since their resources are
separate and the kernel can "flush" for the now-dead process.  With
threads, there is one process and all resources are shared.  No one is
left around to "flush".

-D

-- 

He who spares the rod hates his son,
but he who loves him is careful to discipline him.
        Proverbs 13:24




From gerhard at bigfoot.de  Mon Feb 25 14:54:31 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 25 Feb 2002 19:54:31 GMT
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com> <slrna7l380.16u.gerhard@lilith.my-fqdn.de>
Message-ID: <slrna7l5jn.1l8.gerhard@lilith.my-fqdn.de>

Gerhard H?ring wrote in comp.lang.python:
> Bob Lancaster wrote in comp.lang.python:
>> I am in a bit of a quandry. 
>> I am writing a python module which converts some data from various
>> python variables, lists, etc. into SQL, which in turn is inserted to
>> or selected from a database.
>> 
>> I need to support three combinations.
>> (1) PostgreSQL and Linux, for academic customers.  No serious problems
>> here.  In fact, I have learned why folks love PostgreSQL so much.
>> 
>> (2) Oracle 9i and Linux.  I haven't tried this one yet, but it seems
>> to be at least as easy as the third combination.
>> 
>> (3) Oracle 9i and Windows 2000, for industrial customers.  Trouble!
>> I have tried two approaches.  
>>   (a) DCOracle2:  builds, connects, but then does not work properly. 
>> I see two possible sources of the problem.
> 
> Or:
> 
> (4) Use a commercially supported DB module like mxODBC.

Oops. I meant it could be a possible solution for your Oracle
troubles, and it does run on Linux and Windows.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From martin at v.loewis.de  Sun Feb 17 08:16:34 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 17 Feb 2002 14:16:34 +0100
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com>
Message-ID: <m37kpcuu31.fsf@mira.informatik.hu-berlin.de>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> I'm wondering how difficult it would be to add an editable (mutable)
> string class to Python 2.2 implemented in C under the new type/class
> unification scheme.

There is already an efficient mutable vector type, the array type,
which can be used to implement both mutable byte strings and mutable
character strings.

> Java has a StringBuffer class which Java programmers use idiomatically
> whenever they want to build a big string from smaller ones.  Python
> surprisingly lacks a counterpart to StringBuffer.

The Python counterpart of the Java StringBuffer is the StringIO module.

Regards,
Martin


From ragu at daylight.com  Thu Feb  7 13:21:57 2002
From: ragu at daylight.com (Ragu Bharadwaj)
Date: Thu, 07 Feb 2002 11:21:57 -0700
Subject: encapsulation, typing & other questions
Message-ID: <3C62C5C5.7FEF0B10@daylight.com>

Hi Guys,
	I'm considering Python for enterprise level development and being new
to Python but having used ObjC & Java a lot, have the following
questions. Any answers would be appreciated.

1. Are there plans to include type support for method/function calls in
Python? Seems like it would lead to much faster error detection &
correction.

2. How easy is it to use design patterns in Python? A lack of typing
suggests it wouldn't be too easy, but I'd like to hear back from folks
that have tried this.

3. Are there plans to include method & variable encapsulation in the
object models in python? The answer I've heard is that since it's
possible to overcome encapsulation by long or dirty ways, it isn't worth
it. But my comeback is that if one sticks to the rules while coding,
this won't happen and the resulting gains are worth it.

4. How advanced is db access in Python? Does it use connection pooling
and other optimizations like the JDBC drivers?

cheers
-Ragu


From kp87 at lycos.com  Mon Feb 11 08:29:04 2002
From: kp87 at lycos.com (kevin parks)
Date: 11 Feb 2002 05:29:04 -0800
Subject: filling lists with different types of noise [brown, 1/f...]
Message-ID: <5e8bd451.0202110529.2a917602@posting.google.com>

Hi.

Anyone here have a module that creates different types of noise? I
know that Numpy has some, but what i am looking for is a way to fill a
list with 1/f noise and brownian noise.

Anyone know how to generate brown and 1/f noise in Python?

cheers,

kevin


From emile at fenx.com  Sun Feb 17 14:52:55 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 17 Feb 2002 11:52:55 -0800
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1> <8867.813T2365T9245607threeseas@earthlink.net>
Message-ID: <a4p219$1o0ne$1@ID-11957.news.dfncis.de>

Complete re-posts smell like spam.

I have no doubt that your nine steps can be super-imposed on everything.
However, as long as the deep magic involved in converting these steps
into things people want done is kept as magic, let Aahz's quote reign:

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py

Emile van Sebille
emile at fenx.com







From phr-n2002a at nightsong.com  Sun Feb 10 02:23:33 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 09 Feb 2002 23:23:33 -0800
Subject: Request regarding announcements
Message-ID: <7xn0yhg5qi.fsf@ruckus.brouhaha.com>

For those of you posting announcements (like "ANN: PyFrob version 3.2
released"), could you please include a few sentences in the
announcement saying what the program does?  Assume that there are a
lot of people on the newsgroup who have never heard of the program
but might like to use it if they knew about it. 

Thanks.


From tdelaney at avaya.com  Sun Feb 17 23:49:15 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 18 Feb 2002 15:49:15 +1100
Subject: Guido wins FSF Award
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A190@natasha.auslabs.avaya.com>

> From: mlh at vier.idi.ntnu.no [mailto:mlh at vier.idi.ntnu.no]
> 
> In article <3C706035.6AC11606 at earthlink.net>, Hans Nowak wrote:
> >Magnus Lie Hetland wrote:
> [...]
> >> No, no! This isn't simple enough. The *real* truth (and 
> it's physics,
> >> so you can't contradict me) is that there is only one true action:
> >> 
> >> DS - Do Stuff
> >> 
> >> Now deal with it and go somewhere else with your false theory.
> >
> >Do you also have a model with zero actions?
> 
> How could that possibly cover everything? Sorry, one action is
> minimum.

Ah - but the effect of *not acting* can cover everything if you time it
correctly. So what we obviously need is a timetable of when not to act at
all and so long as we follow it we won't need to do anything at all!

Tim Delaney



From sanjay at jarna.com  Fri Feb 15 13:38:12 2002
From: sanjay at jarna.com (Sanjay)
Date: 15 Feb 2002 10:38:12 -0800
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com> <3C6D08C1.2000307@skippinet.com.au>
Message-ID: <9d4c7023.0202151038.1d3307f@posting.google.com>

Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6D08C1.2000307 at skippinet.com.au>...
> Sanjay wrote:
> > I saw a thread last september stating that this limitation may soon be
> > lifted.  Does the latest release of PythonCOM support this feature? 
> 
> It supports implementing such interfaces, but not calling them.
> 
> > If not, are there any workarounds?  I also saw the suggestion that
> > creating your own pyd to wrap the custom interface would be a good
> > alternative.  Is this still the best way to go?
> 
> For calling interfaces, using SWIG is still the best option - though I 
> would much prefer to bite the bullet and see the universal gateway stuff 
> extended so that we can call those interfaces. :-)
> 
> Mark.

What I am trying to do is something that was somewhat discussed on a
previous thread, but I'm not sure the thought was fully completed. 
Specifically, I would like to write an exchange server event sink by
implementing two com interfaces, IExStoreAsyncEvents and
IExStoreSyncEvents.  The problem I'm having is that when the python
event listener is loaded as a com+ application, and I register this
application for events with exchange, I get the error TypeError ...
IID not supported.  From looking at the previous threads, it seems as
this problem is caused by not being able to handle arbitrary vtable
interfaces, such as the interface methods of IExStoreAsyncEvents and
IExStoreSyncEvents.  If, however,  the two com classes
IExStoreAsyncEvents and IExStoreSyncEvents were defined in a pyd, then
pythoncom would have knowledge of these interfaces, and properly call
my application.

I hope I'm explaining the problem clearly. I am a little confused as
to how to get around this with pythoncom.

Sanjay

By the way, the code sample looks like:

# Methods implemented by the interfaces.
IExStoreAsyncEvents_methods = ["OnSave", "OnDelete"]
IExStoreSyncEvents_methods  = ["OnSyncSave", "OnSyncDelete"]

try:
    # import event sink type library
    universal.RegisterInterfaces('{1F28FCC2-8B0D-11D2-9C95-00C04F79F1DB}',
0, 1, 0, ["IExStoreSyncEvents", "IExStoreAsyncEvents"])
except:
    raise RuntimeError, "could not import event sink type library"

class MyEventListener:
	_public_methods_ = IExStoreAsyncEvents_methods +
IExStoreSyncEvents_methods
	_com_interfaces_ = ['IExStoreAsyncEvents', 'IExStoreSyncEvents']

	_reg_clsid_      = "{D7CDBFB7-7B49-4EA4-9F8E-47152ED86991}"
	_reg_progid_     = "MyEventListener"
	_reg_desc_       = "My Event Listener"


        ...
 
        # IExStoreSyncEvents interfaces	
	def OnSyncSave(self, pEventInfo, strUrl, lFlags):
		# Called by a store when an item is saved.
		pass
	
	def OnSyncDelete(self, pEventInfo, strUrl, lFlags):
		# Called by a store when an item is deleted.
		pass

        ...


From just at xs4all.nl  Wed Feb 20 15:55:10 2002
From: just at xs4all.nl (Just van Rossum)
Date: Wed, 20 Feb 2002 21:55:10 +0100
Subject: Warning about "for line in file:"
References: <mailman.1014235657.26957.python-list@python.org>
Message-ID: <just-AB9A6B.21550920022002@news1.xs4all.nl>

In article <mailman.1014235657.26957.python-list at python.org>,
 "Jason Orendorff" <jason at jorendorff.com> wrote:

> > This is highly unintuitive to me, indeed so much so that I would like to 
> > call it a bug!
> 
> I agree! 

Heh, I was starting to think I'm the only one...

> This could be improved by having file.__iter__()
> always return the same object.  But that doesn't solve the problem
> in full.

Still, it would be 100% better than the current situation.

> One possibility is to change file.__iter__() to return
> iter(file.readline, ''), and have users call file.xreadlines()
> explicitly, and accept the semantic oddness, if they need the
> extra speed.  :-\

No, "for line in file" should become *the* idiom, and therefore it 
should also be the fastest idiom.

> Another possible hack would be to have xreadlines.next() call
> file.seek() on the underlying file object, to make sure the two
> are in sync.  That sounds like a pain to me.

I have no idea how expensive seek() is in general, but it would indeed 
clear up the semantics. But what about the other way around? Should a 
seek() also update the iterator? It's just really messy to have these 
two states exist at the same time :-(.

Just


From huaiyu at gauss.almadan.ibm.com  Fri Feb  1 14:51:51 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Fri, 1 Feb 2002 19:51:51 +0000 (UTC)
Subject: Timing and aspect oriented programming (Was Re: Equivalent of Perl chomp?)
References: <mailman.1012509053.25355.python-list@python.org> <slrna5lpiu.o0.huaiyu@gauss.almadan.ibm.com>
Message-ID: <slrna5lse2.o0.huaiyu@gauss.almadan.ibm.com>

The results I gave earlier was not quite right: the first part of the numbers
was pasted from an earlier run which loops over ten times as much.

Following are comparible numbers in a more readable format.

Using [-1:]
----plain old test ---

slice endswith
 0.0196  0.0268
 0.0196  0.0267
 0.0195  0.0345
 0.0196  0.0269
 0.0195  0.0269

endswith slice
 0.0270  0.0195
 0.0270  0.0194
 0.0269  0.0194
 0.0268  0.0197
 0.0275  0.0198
---- aspect test ---

slice endswith
 0.0778  0.1017
 0.0772  0.0874
 0.0773  0.0874
 0.0773  0.0934
 0.0773  0.0875

endswith slice
 0.0875  0.0767
 0.0871  0.0860
 0.0875  0.0775
 0.0877  0.1090
 0.0954  0.0780


Using [-len(x):]
----plain old test ---

slice endswith
 0.0573  0.0271
 0.0566  0.0268
 0.0565  0.0270
 0.0586  0.0270
 0.0632  0.0269

endswith slice
 0.0271  0.0568
 0.0271  0.0567
 0.0270  0.0563
 0.0270  0.0566
 0.0271  0.0567
---- aspect test ---

slice endswith
 0.1244  0.0871
 0.1240  0.0891
 0.1312  0.0873
 0.1241  0.0875
 0.1312  0.0877

endswith slice
 0.0875  0.1241
 0.0891  0.1305
 0.0876  0.1241
 0.0874  0.1241
 0.0938  0.1243


From tismer at tismer.com  Tue Feb 26 12:31:01 2002
From: tismer at tismer.com (Christian Tismer)
Date: Tue, 26 Feb 2002 18:31:01 +0100
Subject: Time to rename Stackless?
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz> <mailman.1014633824.31429.python-list@python.org> <915a998f.0202260438.246be39c@posting.google.com> <mailman.1014738044.20018.python-list@python.org> <a5gevj$77e6d$1@ID-11957.news.dfncis.de>
Message-ID: <3C7BC655.3070605@tismer.com>

Emile van Sebille wrote:

...


> I hope you've found funding to continue.  Do we send contributions to
> Kaunstr.?


Yes, there is some funding already, at the moment just something
to fill the wholes partially that Stackless poked into my wallet.

I have set up a PayPal account which I will add to the web pages.

It also works via email:
https://www.paypal.com/xclick/business=tismer%40tismer.com&item_name=Stackless

Thanks to all supporters of Stackless - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From jeff at ccvcorp.com  Thu Feb 28 13:48:28 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 10:48:28 -0800
Subject: ActiveState version of 2.2???
References: <1001ff04.0202170702.6ebaaf55@posting.google.com> <3C704157.B872722F@bellatlantic.net> <1001ff04.0202261911.22131410@posting.google.com> <3C7D2879.F54DF385@ccvcorp.com> <3C7D7D0B.7050007@skippinet.com.au>
Message-ID: <3C7E7B7C.435BEB1@ccvcorp.com>


Mark Hammond wrote:

> Jeff Shannon wrote:
> > For me, the big advantage of ActiveState's is that it
>  > automatically registers all of the COM and Active Scripting
>  > stuff, which I've been using fairly extensively, but it's not
>  > *that* difficult to do this manually.
>
> The win32all installers also do this, and have for quite some time now
> (years).  Do you get any error messages at install time?

Hm, the first time I installed Python (BeOpen 2.0, about a year ago), I
seem to remember having to run a couple of registration scripts in order
to get active scripting to work.  (I don't recall any error messages, but
it's been a while.)  Since then, I've just been using ActiveState's, since
it was easier to get a single download.  Haven't tried 2.2 + win32all
myself, yet, so...  maybe I should.  :)

Jeff Shannon
Technician/Programmer
Credit International




From wtanksley at bigfoot.com  Wed Feb 20 15:33:23 2002
From: wtanksley at bigfoot.com (William Tanksley Google)
Date: 20 Feb 2002 12:33:23 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <de3fc1ef.0202201233.2228e5b1@posting.google.com>

rmunn at pobox.com (Robin Munn) wrote:

> * Batteries included. This is the big one, which makes Python more than

My friend was griping about one battery that Basic comes with but
Python doesn't (at least not that I've seen): graphics.  Yes, I know
there's no single way to draw in Basic, but every version has a
platform-specific graphics library.

Does Python?  If so, I've never seen it.  I'd like to.

Personally, I need pixel graphics.  Others will want vector graphics.

-Billy


From jamesk at ifm-services.com  Mon Feb 25 03:50:58 2002
From: jamesk at ifm-services.com (James Knowles)
Date: Mon, 25 Feb 2002 01:50:58 -0700
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <a50hd2$6vt$1@panix2.panix.com> <a50im3$bf1$1@news.island.liu.se> <slrna7dipm.5et.quinn@regurgitate.ugcs.caltech.edu> <BEEd8.66441$dG.27583658@news1.rdc1.sdca.home.com>
Message-ID: <3C79FAF2.78B8BDB1@ifm-services.com>

> > My s?ster was bitten by a m??se once.
... 
> python-related conversation is expected to take place here; 

Obviously you've not paid sufficient attention to the beginning. That's one
of the best parts, IMHO. 

http://www.smouse.force9.co.uk/monty.htm


Mynd you, m??se bites Kan be pretty nasti...


From philh at comuno.freeserve.co.uk  Thu Feb  7 18:45:43 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:45:43 +0000
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de>
Message-ID: <slrna664d7.d87.philh@comuno.freeserve.co.uk>

On Thu, 7 Feb 2002 10:26:02 -0500, MDK <mdk at mdk.com> wrote:
>I have this:
>
># myprog.py
>
>from mymod import *
>
>def dosomething(x):
>    dosomething(1000)
>
># End code
>
>I want the dosomething(1000) run from the function in mymod.  However,
>instead it is running dosomething(x) in myprog.py.

Do:

# myprog.py
import mymod
def dosomething(x):
   mymod.dosomething(1000)

>import mymod

Thisi s the prefered way of doing things.



-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jim at publishingresources.com  Tue Feb 26 16:44:24 2002
From: jim at publishingresources.com (Jim Abrams)
Date: Tue, 26 Feb 2002 21:44:24 -0000
Subject: fun with ADO
References: <2ec1bc1c.0202221202.1e0ea6aa@posting.google.com> <f20ea932.0202230533.6a01ac4b@posting.google.com>
Message-ID: <Xns91C1AA4A0ABC6jimpublishingresourc@209.249.90.101>

tom at peresys.co.za (Tomasz Stochmal) wrote in 
news:f20ea932.0202230533.6a01ac4b at posting.google.com:

>> 
>>     rs=win32com.client.Dispatch(r'ADODB.Recordset')
>>     rs.Open(sql,connectString)
>>     while not rs.EOF:
>>         name=str(rs.Fields("clientname").Value)  #oink!
>>         gid=str(rs.Fields("gid").Value)
>>         (- do something fun with fields -)
>>         rs.MoveNext()
>>     rs.Close()
>> 
> 
> Use rs.GetRows() to retrieve all rows & fields with one call only
> on python side you can extract the field values that you need
> 

I've found GetRows() to be the fastest method of all. 
Here's my gig:

def getdicts(self, rows=0, _lower=1):	
    	fieldnames = map(str, self.get_field_names())
	if _lower: fieldnames = map(lower, fieldnames)
	ret = []
	if not self.EOF: ret = [dict(zip(fieldnames, l)) for l in zip
(*self.getrows(rows))]
	return ret


Watch the wrapping.


From fredrik at pythonware.com  Mon Feb 11 05:09:18 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 10:09:18 GMT
Subject: box on soap
References: <mailman.1013283562.647.python-list@python.org> <Lli98.24849$l93.4538901@newsb.telia.net> <F_z98.32573$3E5.2317813@newsread2.prod.itd.earthlink.net>
Message-ID: <iLM98.25034$l93.4623038@newsb.telia.net>

Chris Tavares wrote:

> Being a professional training celebrity will do that to you

Especially when you're an expert on a technology that's quickly
becoming obsolete <0.5 wink>

> Actually, there *is* one advantage to having WSDL that isn't about enforcing
> contracts - having WSDL around will let you build better tools.

Oh, sure.  Programmers will generate WSDL (or whatever) if
they find that it saves them some time, think that it's fun, or
if their customers want it.

I will, too.

And I'm sure that some programmers will write WSDL files just
to make Don Box happy.  Heck, some people will probably stay
away from dynamic languages just because he doesn't use them.

I won't.

(the real problem here is that RPC-over-the-WEB is a dead
end in itself.  XML-RPC works because it's soo damn simple:
it's simplicity often overcomes the limitations. But that's not
true for SOAP+WSDL.  Just lots of complexity, and no real
advantages over better designs.  But that's another story)

</F>

    "I have less and less need to convince anyone of any-
    thing. This feels like evolution, rather than apathy"
        - evan williams




From piedra at filosofal.com  Wed Feb 13 12:54:59 2002
From: piedra at filosofal.com (NI UN EURO DE MAS)
Date: Wed, 13 Feb 2002 18:54:59 +0100
Subject: WEBCAMS
Message-ID: <a4e8rh$oeb$1@news.wanadoo.es>

http://msgsl.4d2.net/
Alucinaras por un tubo, aplaudiras con las orejas, te sonaras en estereo,
etc..
M?s de 300 c?maras funcionando, en empresas, domicilios particulares,
iglesias, establecimientos publicos, en ciudades, etc..

Los mejores programas de la red para video conferencia, envio de v?deo
e-mail y para poder montarte tu WebCams. Y lo mejor de todo solo pagaras en
las conexiones NADA ni un EURO solo tu linea ADSL y conectate los 365 d?as
del a?o con las antipodas y las 24 h. del d?a. Pueden conectarse al mismo
tiempo 5 WebCams en el programa mas peque?o de hogar.






From jbarham at octopuswireless.com  Mon Feb  4 14:26:27 2002
From: jbarham at octopuswireless.com (John E. Barham)
Date: 4 Feb 2002 11:26:27 -0800
Subject: CPython vs. Jython for Web applications?
Message-ID: <c775c8db.0202041126.5177e606@posting.google.com>

Good day,

I'm considering writing a fairly sizable Web app in Python and would
appreciate advice on whether to use CPython or Jython.  My main
motivation for considering Jython is "native" access to Java's more
mature XML libraries for XSLT (e.g., Saxon) and SVG (e.g., Apache's
Batik).

The constraints are:
 - data to be stored in a relational database, preferably open-source
 - app must run on both Win32 and some free Unix, preferably FreeBSD
 - interface will be generated using XSLT
 - app must run in Apache using some form of embedding (i.e.,
out-of-process overhead of classic CGI is unacceptable)

I'm more interested in ease of development than optimal performance
but performance still matters...

Some questions:
  Has anyone written a sizable Web app in Jython?
  Are there real-world benchmarks comparing, say, Jython & Saxon vs.
CPython & Pyana/Xalan?
  How do Jython servlets compare to using something like mod_python in
terms of speed and ease of use and scalability?
  How painful is it to do Jython database access vs. the standard
Python DBI?

FWIW, I have a lot of experience with C++, CPython and XSLT but just
an undergrad course's worth in Java.

I appreciate that Jython's performance is dependent on the JVM, so
advice would be appreciated there as well, especially regarding the
performance of JVMs for the free Unices.

Thanks in advance.

  John


From rhymes at myself.com  Thu Feb 14 13:43:34 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Thu, 14 Feb 2002 19:43:34 +0100
Subject: Highlights & Cursors
Message-ID: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com>

I'm just learning Tkinter but I have a little problem.
Can anyone of you explain me how to use the highlight
for the focus in a widget and the cursor option?

Thanks in advance.

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From usenet at thinkspot.net  Sat Feb 23 13:16:13 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 10:16:13 -0800
Subject: Newbie and exceptions.
References: <f0gf7u8h6tn05bf4a7orbnsm0e0lsn3oa6@4ax.com> <a58ffq$uak$1@norfair.nerim.net> <o8hf7u8mpbmb3d4fi84cliqscb4tbgvqbu@4ax.com>
Message-ID: <a57q5e.3vsin0h.1@kserver.org>

On Sat, 23 Feb 2002 16:35:15 GMT, Jan Johansson <spamfilter2 at mupp.net>
wrote in comp.lang.python in article
<o8hf7u8mpbmb3d4fi84cliqscb4tbgvqbu at 4ax.com>:

> ># Near the start of the script
> >from ftplib import error_perm
> 
>    except ftplib.error_perm:
> NameError: name 'ftplib' is not defined

ftplib is not defined because you did a 
from ... import

Instead do

import ftplib

Then
except ftplib.error_perm should work.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From roy at panix.com  Sat Feb  2 14:59:03 2002
From: roy at panix.com (Roy Smith)
Date: Sat, 02 Feb 2002 14:59:03 -0500
Subject: Problem building gdbm module (python-2.1.1)
Message-ID: <roy-AE0761.14585102022002@news1.panix.com>

I'm trying to build python-2.1.1 with gdbm support on a Solaris (SunOS-5.7) 
box.  It looks to me like there's a bug in the configure script.

I've got /usr/local/include/gdbm.h and /usr/local/lib/libgdbm.a on the 
system.  Modules/gdbmmodule.c has '#include "gdbm.h"' in it, but the 
configure script is looking for gdbm/ndbm.h, which it doesn't find.

If I change the configure script to look for gdbm.h instead of gdbm/ndbm.h, 
it finds the library, builds the module, and seems to work just fine.

Is this a bug, or am I just doing something wierd?


From joshm at taconic.net  Fri Feb 22 16:18:14 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Fri, 22 Feb 2002 16:18:14 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3c7455a8_4@corp.newsgroups.com> <jwbaxter-CE1C5D.00553921022002@corp.supernews.com> <3C74BD08.FD8CAACE@alcyone.com> <jwbaxter-E8520F.18591421022002@corp.supernews.com>
Message-ID: <3c76b327_3@corp.newsgroups.com>

> Yes...it was knowing that which led to the "test" and the remark.  Which
> should have spoken of a string of length 1 rather than a one-character
> string.

And also the reason for my use of the word "philosophically" in my original
reply.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From jmackd1 at yahoo.com  Thu Feb 28 09:22:10 2002
From: jmackd1 at yahoo.com (Jeff McComas)
Date: 28 Feb 2002 09:22:10 -0500
Subject: sorting
Message-ID: <3c7e3d12_7@news.teranews.com>

As a newbie i find the python scripting language very interesting and useful.
I am trying to write a script that will search a .txt file and have the script
tell me how many of the same numbers are in the .txt file. I know you can do
it with the f.count command but my problem is how can you do an input that
makes the numbers strings? Or puts double quotes and a comma around a number?
This is what a raw input from a .txt file looks like:

34.6        25        78        89.2       45.7
23          92        47        20         85
56          78        23        49         92

out of that how can i pull out the numbers that are the same, say 78 = 2

Any help would be appreciated!

Thanks,
Jeff McComas
Austin, TX



From sholden at holdenweb.com  Sun Feb 17 17:21:55 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 17 Feb 2002 17:21:55 -0500
Subject: __class__ for 2.2 types
References: <2ec1bc1c.0202071240.44185531@posting.google.com> <a3upaq$1arn1e$1@ID-59885.news.dfncis.de> <a4otl1$44d$1@news.idiom.com> <a4p0ep$1ptki$1@ID-11957.news.dfncis.de>
Message-ID: <G7fc8.189$Zp2.58090@atlpnn01.usenetserver.com>

"Emile van Sebille" <emile at fenx.com> wrote...
> "Jim Dennis"
> >  That's an interesting "gotchya"  I understand why 2.__class__ is a
> >  syntax error (ambiguous with a float: 2.0 etc).  But I'm surprised
> >  that 2 . __class__ works; the spaces seem unnatural here.
> >
>
> I was surprised by this recently as well.  It turns out the '.' is
> simply a delimiter, and whitespace is allowed either side of a
> delimiter.
>
> >>> sys . path . pop ( )
> 'F:\\Python22\\lib\\site-packages'
> >>>
Well, syntactically the period acts as a delimiter when there are no spaces,
but here the spaces themselves are the delimiters. Technically isn't the
period acting as an operator here, the operation being "find in namespace"
or some such?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From phr-n2002a at nightsong.com  Wed Feb 20 18:35:40 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 15:35:40 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org> <7xr8ng83k0.fsf@ruckus.brouhaha.com> <3C741EAF.B8E6E14A@stroeder.com>
Message-ID: <7xeljfwwtv.fsf@ruckus.brouhaha.com>

Michael Str?der <michael at stroeder.com> writes:
> > <base href=http://foo.bar/cgi-bin/_SID_=36987348732/>
> 
> Are you sure all browsers handle this properly?

I can't be certain about very old browsers, but it's part of standard
HTML these days.  I'm not sure when it became part of the standard.



From shochet at hotmail.com  Mon Feb 11 19:43:56 2002
From: shochet at hotmail.com (Joe Shochet)
Date: 11 Feb 2002 16:43:56 -0800
Subject: Compiling Python 2.2 in Microsoft VC7 (.net)
Message-ID: <a85e8a65.0202111643.5f5fda89@posting.google.com>

I'm trying to get Python 2.2 to compile in VC7 (part of Visual Studio
.net), but I'm running into all sorts of troubles. Before digging into
the problems, has anybody else already done this?

The main problems seem be related to missing files (largeint.h,
traceback.c were the first two I saw). Many more followed these
errors.


From othello  Wed Feb 27 01:36:22 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 01:36:22 -0500
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <a5hur4$3qg$1@bob.news.rcn.net>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xpu2slwtl.fsf_-_ at ruckus.brouhaha.com...
<snipped>
>    Example 2:
>      local x(int)
>      x = "banana"    # implementation-dependent behavior
>
>    Some ways Python can handle example 2:
>
>    1a) Compiler does enough static analysis to flag the "banana"
assignment
>       as a compile time error
>
>    1b) Compiler doesn't flag; interpreter raises runtime TypeError
exception.
>
>    2) Compiler doesn't flag; interpreter assigns "banana", i.e. the
>       advice is completely ignored and does nothing, the program works
>       exactly the same way it does now without the advice
>
>    3) Compiler doesn't flag, interpreter has totally undefined behavior.
>       It can work normally, or crash, or delete all your files, etc.
>       The idea is that an optimizing compiler can believe the advice and
>       generate code based on x always being an int (it can put it in a
>       register, perform native machine code arithmetic on it etc).
>

How about another option:

4) Code is generated along the lines of:
if type('banana')==int:
   <run optimized code that assumes 'banana' to be an int>
elseif USESTRICT:
   raise TypeError
else:
   <run normal, slow code which binds 'banana' to int regardless of type>

In other words, generate optimized code in case the type advice is correct;
otherwise, run normally unless strict type checking is in effect.

This allows optimization based on the advice and error checking only when
wanted.

In general, I like your proposals (after Greg Ewing's correction).  It
reminds
of importing packages in that it is a simple tool for building large, fast,
reliable
programs while staying out of the way (and out of mind) for more casual
programming.


Raymond Hettinger






From mhammond at skippinet.com.au  Wed Feb 27 19:49:49 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 28 Feb 2002 00:49:49 GMT
Subject: Problems compiling python Debug under windows
References: <757c9986.0202270735.446b8a0f@posting.google.com>
Message-ID: <3C7D7EB5.8010609@skippinet.com.au>

Marco Gillies wrote:
> I'm having a bit of a nightmare trying to write a python extension in
> C++ under windows. I traced the original problem to the fact that I
> should be using the debug version of python with a debug extension. So
> I'm trying to compile python as debug. Which is where the problems
> start.
> 
> I've got as far as compiling the pythoncore module (the python dll and
> executables). However, I also need to compile the external modules,
> _sre, _socket, parser etc. I've compiled these and the .pyd files are
> there and I've tried to import them into python the give an error:
> 
> Import Error:dynamic modules does not define init function

You should just open the workspace in MSVC, select the Debug 
configuration, and build it.

I assume this works, and you are left with _sre_d.pyd, _socket_d.pyd etc 
in your PCBuild directory?

Then, start python_d, and attempt to import your module.

Assuming that all works, then we are left with a configuration issue in 
your specific project.  (If it doesn't work, post as many details about 
this specific problem - eg, a directory listing of PCBuild\*_d.pyd, and 
a Python prompt that prints sys.path then attempts to import _sre.)

For your projects, don't touch USE_DL_EXPORT - just set DEBUG and 
_DEBUG, then #include "Python.h".  It should then "just work" ;)

Mark.



From martin at v.loewis.de  Sat Feb 16 16:45:49 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 22:45:49 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122> <xyuk7tdku0w.fsf@sex.ifi.uio.no>
Message-ID: <m3y9hthzhu.fsf@mira.informatik.hu-berlin.de>

Jan-Fredrik Braseth <janfbr at sex.ifi.uio.no> writes:

> > # I would change opt to c in the following line
> >                     self.msg('IAC %d not recognized' % ord(opt))
> 
> But if there really is something wrong with that, shouldn't someone
> send a bug report or something to the creators of Python?

Please do: send to to sf.net/projects/python. A patch might help even
more; please use a context or unified diff.

Regards,
Martin



From trentm at ActiveState.com  Wed Feb 20 17:08:06 2002
From: trentm at ActiveState.com (Trent Mick)
Date: Wed, 20 Feb 2002 14:08:06 -0800
Subject: How do you conver output of "popen('ls')" to a list?
In-Reply-To: <3C737245.BC5D7164@macrozon.net>; from husam@macrozon.net on Wed, Feb 20, 2002 at 10:54:13AM +0100
References: <3C737245.BC5D7164@macrozon.net>
Message-ID: <20020220140806.A30148@ActiveState.com>

$ ls
bye.txt  hello.txt
$ python
>>> import os
>>> o = os.popen('ls')
>>> lines = o.readlines()
>>> o.close()
>>> lines
['bye.txt\012', 'hello.txt\012']
>>> files = [line.strip() for line in lines]   # strip off EOL characters
>>> files
['bye.txt', 'hello.txt']
>>>


Cheers,
Trent

-- 
Trent Mick
TrentM at ActiveState.com



From SBrunning at trisystems.co.uk  Fri Feb  8 03:55:08 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 8 Feb 2002 08:55:08 -0000 
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <31575A892FF6D1118F5800600846864DCBCBFF@intrepid>

> From:	aahz at panix.com [SMTP:aahz at panix.com]
> Heh.  My preferred working style is to use two computers, one with each
> hand when I'm not doing anything that requires actual typing.  And, of
> course, I use one mouse in each hand.
 
Only two? Haven't you got feet?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From jason at jorendorff.com  Tue Feb 26 17:56:00 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 16:56:00 -0600
Subject: Class functions
In-Reply-To: <NVSe8.151$en5.1429@typhoon.sonic.net>
Message-ID: <HFEKILOLEFEFMKAECNDLMEIEDFAA.jason@jorendorff.com>

Chat Maine wrote:
> I'd like to access the data updated in real time via a class' function.

Chad,

Your source code doesn't provide quite enough information for
us to give you good advice.  (It actually sounds like you want
Python 2.2's generators, but I'm not sure I believe that.)

Here is one easy way to do this...

# theclass.py
class MyClass:
    ...
    ...
    def func(self):
        return UPDATE

# mainprogram.py
x = theclass.MyClass()
data = x.func()
while data != '':
    print data
    data = x.func()

I may be missing the point, though...

## Jason Orendorff    http://www.jorendorff.com/



From marklists at mceahern.com  Thu Feb 14 12:00:30 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 14 Feb 2002 09:00:30 -0800
Subject: Beginner question please help!
In-Reply-To: <28eed654.0202140843.510948d5@posting.google.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEPOKCAA.marklists@mceahern.com>

Well, for starters, what are x and y?  There's a built-in cmp function:

>>> cmp(4, 5)
-1

If you define your own classes and you want them to allow comparison by
value, you can simply define the special method __cmp__ for them.

#! /usr/bin/env python

class Foo:

	def __init__(self, value=None):
		self.value = value

	def __cmp__(self, other):
		return cmp(self.value, other.value)

f = Foo(4)
g = Foo(5)

print "cmp(f, g) == %d" % (cmp(f, g))
print "f < g == %d" % (f < g)
print "f == g == %d" % (f == g)
print "f > g == %d" % (f > g)

// mark

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Michael Tu
> Sent: Thursday, February 14, 2002 8:43 AM
> To: python-list at python.org
> Subject: Beginner question please help!
>
>
> Hello,
>
> I am trying to do the following exercise but after spending couple of
> hours without any luck, I think it is time to get some helps. Thank
> you in advance.
>
> Write a compare function that return 1 if x > y, 0 if x == y, and -1
> if x < y.
>
> Mike
> --
> http://mail.python.org/mailman/listinfo/python-list
>
>




From smulloni at bracknell.smullyan.org  Fri Feb 22 18:49:53 2002
From: smulloni at bracknell.smullyan.org (Jacob Smullyan)
Date: Fri, 22 Feb 2002 23:49:53 -0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <3C6D8B63.5070909@geocities.com> <mailman.1013864067.16756.python-list@python.org> <69587ukeu1m51mf46kpumjnt9qfujho7ju@4ax.com> <a52h06$43hjh$1@ID-59885.news.dfncis.de> <e7fd7u0co9g72l5p3qi5ohgmtqlq48cglf@4ax.com>
Message-ID: <slrna7dm2f.c5k.smulloni@bracknell.smullyan.org>

Along similar lines, I'm currently developing a product archive format
for SkunkWeb which uses Gordon McMillan's iu.py.  I first attempted to
use imputil.py and ran into serious problems -- iu.py really saved the
day.  I recommend it; either it should become part of the core, or
PythonLabs should bite the bullet and rewrite the import mechanism
with a much higher degree of generality, so that weird import hooks
won't be necessary any longer.


-- 
Jacob Smullyan | smulloni at smullyan.org
http://www.smullyan.org/smulloni/


From xtian at toysinabag.com  Mon Feb 18 16:05:48 2002
From: xtian at toysinabag.com (xtian)
Date: 18 Feb 2002 13:05:48 -0800
Subject: Object copying itself
References: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com>
Message-ID: <d111eafa.0202181305.1c671297@posting.google.com>

Dale Strickland-Clark <dale at riverhall.NOTHANKS.co.uk> wrote in message news:<27mq6ug68esgsh7in71scb9m30phojjf3g at 4ax.com>...

> I want object references copied (shallow copy) but the lists need to
> be copied entirely (deep copy).

It sounds like you may have your terminology confused. Copying an
object reference is not a shallow copy - it's not a copy at all
(except in a degenerate sense), because the references, while copies
of each other, still point to the same object.

If you've called copying a list a deep copy to contrast with the
non-copy of the objects, then you probably mean shallow copy. This is
a copy, because a new list is created, but then all of the list's
items are *non-copied* into it.

This seems to make more sense to me. If it's actually what you want,
then this should implement it:

def copyObj(obj):
	newObj = obj.__class__()
	for name, val in obj.__dict__.iteritems():
		if type(val) is list:
			val = copy.copy(val)
		setattr(newObj, name, val)
	return newObj

This uses some 2.2isms, and doesn't take care of __init__ processing
(that'll depend on how it should be handled in your app), and doesn't
do __slots__ (Alex has discussed how they could be handled).

And I may have misunderstood what you wanted, so it could also be
completely wrong! That's enough disclaimers for today.

Hope it helps...
xtian

(The snipers are passed out in the bushes again.)


From amuys at shortech.com.au  Sun Feb 17 23:09:43 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 17 Feb 2002 20:09:43 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <mailman.1013816158.26552.python-list@python.org>
Message-ID: <7934d084.0202172009.75324c2a@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1013816158.26552.python-list at python.org>...
> On Sat, 16 Feb 2002 05:57:09 +0700
> Tripp Scott wrote:
> 
> > which ones do you recommend? examples in any language is fine, 
> > though i prefer python, perl, or ruby. what i would like is 
> > guidance on what is thread safety, how to achieve it, what are 
> > the issues in multithreading programming, the global interpreter 
> > lock, and other basics.
> 
> It's been many years since I've read a book on multi-threading, so I can't
> suggest a good title, but it's a lot simpler than you might think (given
> the fact you're looking for a book ;)  This is especially true with Python,
> which has a remarkably easy-to-use threading library (use the threading
> module, not the thread module).
> 
> Thread-safety boils down to this: lock any /shared/ resource that might be
> /modified/ by a thread.  End of story.  You can get into all sorts of
> debates about when these locks are necessary, but 99% of the time they are,
> and the other 1% isn't worth the effort to decide whether they really are
> or not.
> 

"End of Story" is a rather interesting way to summarise the issues of
Deadlock, Lock-ordering, Livelock, Starvation, Convoying, and
Lock-Contention to name a few of the problems you may face moving from
serial to parallel algorthims (although the last two only affect
perfomance, not correctness).

I don't have the time to discuss these further atm, and I am not
personally aware of a readily accessable text on concurrent
programming.  Ben-Ari "Principles of Concurrent Programming" PH '82
was the text used when I studied this back in university, however it
is a little dry.  OTOH, I am very reluctant to recommend any of the
language/library specific texts to someone without an existing
knowledge of concurrency issues.  Even Kleiman, Shah, and Smaalders
"Programming with Threads" (an excelent book, covers libpthread based
programming in C/C++) gives rather preemptory coverage of the problems
involved.

Unfortunately without an understanding of the problems, it is hard to
understand/apply the solutions provided by the various
languages/libraries described.

Cliff: Sorry I can't answer your question straight away, maybe check
your local library (possibly uni-library) for a copy of Ben-Avi.  If
you have any problems mapping the concepts he discusses to specific
Python code I (and I imagine the rest of c.l.py) will do what I can to
help.  Other then that, WHY do you feel you need threads?  Or is this
just a "Something new to learn" question?

Tripp: If I sound a bit harsh, I'm sorry.  Understand that as someone
with their name attached to a unexpectedly widely distributed tutorial
on threading, I get *alot* of help requests from new programmers who
think that threading is "just another library to learn", and approach
it as such (with the inevitable trouble).  Your "End-of-story"
statement was just more then I would reasonably let pass without
comment.

Everyone: Does anyone know of an accessable concurrency text?  One
that moves from discussing the basics (data-races, critical sections,
mutural exclusion, lock-order based deadlock, resource based deadlock,
etc) to the various solutions (mutex/cond-var, semaphore, monitor,
rendezvous) to the various design approaches (master-slave, pipeline,
work-pile, etc).  Possibly even a discussion of various performance
considerations (convoy's, wait-free synchronisation, cache-aliasing,
spin vs. blocking locks, etc).

Andrae Muys


From sdm7g at Virginia.EDU  Tue Feb 19 14:35:40 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 14:35:40 -0500 (EST)
Subject: Finding nonprintable characters?
In-Reply-To: <3C72A41F.20908@verio.net>
Message-ID: <Pine.OSX.4.43.0202191420260.1674-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, VanL wrote:
>
> I have a function
>
> isBinary(filehandle)
>
> that I'm not sure how to implement.  I've decided to define binary as
> containing characters above \x80.  But  what is the best way to do this?
>
> 1. iterate through xreadline, so the whole thing doesn't get loaded into
> memory?

I would use file.read( bytes ) -- if it's binary, then you probably
don't need to read the whole file in. Most programs I've seen that
try to determine 'binaryness' only check the first N bytes anyway.
( I've seen some that want a certain percentage of non-printing chars
  per block -- not just a single out of range char. )

> 2. String searching?  If so, for what string?  Searching for anything
> greater  than \x7f?
>
> 3. Re searching?  for what class?
>

How about something like:

  filter( lambda c: ord(c) > value, file.read( blocksize ) )

or, as you note, save the ord() call and use an octal or hex string
literal. If you want to use list comprehensions it would be something
like:

  [ c for c in file.read( blocksize ) if c > '\x7f' ]

but list comprehensions give you a list while filter on a string
yields a string. You can divide the (float) length of the filtered value
by the length of the original ( blocksize ) for a ratio if you
want to use that instead of a single out of range char.


-- Steve Majewski





From mdk at mdk.com  Thu Feb 14 16:32:04 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 16:32:04 -0500
Subject: Convert String to Dictionary question
References: <a4h851$a8g7$1@ID-98166.news.dfncis.de> <Pine.SOL.4.30.0202142127220.24038-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <a4hacl$argt$1@ID-98166.news.dfncis.de>

"John J. Lee" <jjl at pobox.com> wrote in message
news:Pine.SOL.4.30.0202142127220.24038-100000 at mimosa.csv.warwick.ac.uk...
> On Thu, 14 Feb 2002, MDK wrote:
>
> > I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> >
> > When I do a readline the I get back a string.
> >
> > How do I convert this string to a dictionary?
>
> You probably don't want to.  Read the documentation for the pickle module.
>
>
> John
>

Pickling seems too slow.




From seebs at plethora.net  Wed Feb  6 00:34:20 2002
From: seebs at plethora.net (Peter Seebach)
Date: 06 Feb 2002 05:34:20 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net>
Message-ID: <3c60c05c$0$79564$3c090ad1@news.plethora.net>

In article <1372.801T1070T13994028threeseas at earthlink.net>,
Timothy Rue <threeseas at earthlink.net> wrote:
>Why would  be interested in a programming language by a company that has
>been found guilty of Anti-trust? Not to mention that language is not
>available anywhere near as much as something like Python.

Because the language fits your project a lot better.  The *community* is
perfect for you.  These people share your basic mindset.  They're heading
where you seem to be already.

>It'd really be a good thing if the MS cronies state so in their sig.

I am not a MS crony.

>This way I don't have to try and figure out why you have a problem knowing
>the difference between what is idea and what is natural law physics. For
>MS seems to think it own reality. It doesn't!

This isn't about ownership, it's about finding the community most naturally
suited to your project.  The C# people are the community least contaminated
by existing notions of software engineering.

-s
-- 
   Copyright 2001, all wrongs reversed.  Peter Seebach / seebs at plethora.net
   $ chmod a+x /bin/laden      Please do not feed or harbor the terrorists.
     C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/


From kragen at pobox.com  Fri Feb  1 19:51:01 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 01 Feb 2002 19:51:01 -0500
Subject: Enhanced Generators - the difference between sources and iterators
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org>
Message-ID: <83g04kzoy2.fsf@panacea.canonical.org>

Oren Tirosh <oren-py-l at hishome.net> writes:
> My implementation of xmap, xfilter and xzip returns a source, not an
> iterator.  It can be iterated multiple times, just like the output of map,
> filter or zip.

That won't work when passed iterators as arguments, right?

> The logical opposite of a source is a destination.  Destinations should
> be able to produce a temporary object mirroring an iterator.  I call such 
> an object a sink (PLUG: see http://tothink.com/python/dataflow)

Well, if you want to have two pieces of code with independent local
control state connected by a stream of values, in which the sink runs
until it needs a new value, and then the source runs until it produces
a new value, after which the sink runs again etc., generators give you
that already.  Having "sinks" just gives you another way to do the
same thing.  Are there some situations where structuring a piece of
code with some hypothetical "accept" expression would yield (pardon
the pun!) something simpler than just doing the same thing with
generators?

Suppose you have this program:

def uniq(iterable):
    state = {}
    for x in iterable:
        if not state.has_key('prev') or x != state['prev']:
            print x
        state['prev'] = x

Now, if we have this hypothetical "accept" operation that is the other
side of 'yield', we can make this into a lazy sink:

def uniq():
    state = {}
    while 1:
        x = accept
        if not state.has_key('prev') or x != state['prev']:
            print x
        state['prev'] = x

That's not any simpler.  The code that represents uniq's source also
doesn't get simpler.  It changes from this:

def uniqsource():
    mystate = initializemystate()
    while mystate.continuing():
        mystate.update()
        yield mystate.nextvalue()

to this:

def uniqsource(outputthing):
    mystate = initializemystate()
    while mystate.continuing():
        mystate.update()
        outputthing.feed(mystate.nextvalue())

That's also not any simpler.  Are there examples where the second pair
is noticeably simpler than the first?  I doubt it.



From mcherm at destiny.com  Wed Feb 13 14:35:55 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Wed, 13 Feb 2002 14:35:55 -0500
Subject: How to generate XML
Message-ID: <3C6AC01B.7070404@destiny.com>

>>> I'm amazed that people always start proposing difficult-to-use
>>> libraries when this question comes up. I'd prefer to use
>>>
>>> print "<foo/>"
> 
> For the special case where you have a tree structure whose nodes
> are instances of Python classes, implement an "export" method in
> each class something like the following:

 >
 >    def export(self, outfile, level):
 >             [...]


I tend to agree with the original poster: simply writing your own print 
routines isn't difficult at all. However, if you're going to go around 
putting in standard "export()" methods, perhaps you'd prefer something 
which does it automatically.

The following code will create an xmlPrint() method. Pass it an object, 
and it prints it out in a pseudoreadable fashion. If I were really going 
to be using this, I'd modify _xmlPrintObj() so it checks for a .toXML() 
method and invokes that instead if such a thing exists. But there's no 
reason to write boilerplate in Python, where introspection is so easy.

-- Michael Chermside


------------------------ <code> -------------------------
def xmlPrint(obj, writer=sys.stdout, indentText='   '):
     """Print an object in XML format."""
     _xmlPrintObj(obj, _IndentingWriter(writer, indentText))

class _IndentingWriter:
     def __init__(self,
                  writer=sys.stdout,
                  indentText='   ',
                  indentLevel=0):
         self.writer = writer
         self.indentText = indentText
         self.indentLevel = indentLevel
         self.indentStr = indentText * indentLevel
     def incr(self):
         return IndentingWriter(self.writer, self.indentText,
                                self.indentLevel + 1)
     def writeline(self, text):
         self.writer.write(self.indentStr)
         self.writer.write(text)
         self.writer.write('\n')


def _xmlPrintObj(obj, writer):
     """Print an object in XML format (internal interface)."""
     # Print tags for the object itself based on class name
     tagname = obj.__class__.__name__
     writer.writeline('<%s>' % tagname)

     # for all fields not beginning with an underscore...
     for fieldName in filter(lambda x: x[0]!='_', dir(obj)):
         _xmlPrintField(getattr(obj, fieldName), fieldName,
                        writer.incr())

     # close the tags for the object itself
     writer.writeline('</%s>' % tagname)

def _xmlPrintField(value, fieldName, writer):
     """Print a field of some object, labled with its fieldName."""
     if type(value) == types.InstanceType:
         # for sub-objects we print field name then recurse
         writer.writeline('<%s>' % fieldName)
         _xmlPrintObj(value, writer.incr())
         writer.writeline('</%s>' % fieldName)
     elif type(value) in (types.ListType, types.TupleType):
         # for sequences we print all items
         writer.writeline('<%s>' % fieldName)
         for v in value:
             _xmlPrintField(v, 'value', writer.incr())
         writer.writeline('</%s>' % fieldName)
     else:
         # for all other fields just use str()
         writer.writeline('</%s>%s</%s>' % (fieldName, value, fieldName))
------------------------ </code> -------------------------




From BPettersen at NAREX.com  Thu Feb 28 10:50:59 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Thu, 28 Feb 2002 08:50:59 -0700
Subject: Status of PEP's?
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192215147@admin56.narex.com>

> From: Remco Gerlich [mailto:scarblac at pino.selwerd.nl] 
> 
> Fernando P?rez <fperez528 at yahoo.com> wrote in comp.lang.python:
> > Python is a general purpose language, not one to be used only by 
> > number
> > theorists. Putting at a level as basic as the integers a 
> convention which 
> > allows things like 'for i in 3' to be valid iteration code 
> really sounds 
> > bizarre and counter intuitive to me.
> > 
> > But maybe it's just me.
> 
> It's not just you.
> 
> I believe syntax like 'for i in 5:' is confusing (at first 
> sight I'd think that i would get one value, 5), and has no 
> merit except saving a few key strokes compared to 'for i in 
> range(5):', which is clearer as well.
> 
> Compare:
> 
> for i in 6: print i
> for i in 6,7: print i
> for i in 6,7,8: print i
> 
> You wouldn't think the first line does something different, would you?

Personally, I find the proposal quite easy to understand, and I have no
problem grokking that 

  for i in len(mylist):

Creates an iterator over the indices in mylist. After over four years
programming in Python, the above is still an error I make frequently. I
also see no conceptual problem in understanding what:

  if i in len(mylist):

does. In fact I think it is more intuitive than the current:

  if i in range(len(mylist)):

Finally, I would almost expect the proposal to work, when 

  for line in file:

works.

Oh, and as for the proposal to have range accept a sequence, I believe
that died by BDFL pronouncement (although my memory is a bit fuzzy here
;-)

-- bjorn



From pbienst at MIT.EDU  Tue Feb 26 13:20:51 2002
From: pbienst at MIT.EDU (Peter Bienstman)
Date: 26 Feb 2002 13:20:51 -0500
Subject: Precompiled regular expressions slower?
In-Reply-To: <XFMail.20020226094959.shalehperry@attbi.com>
References: <XFMail.20020226094959.shalehperry@attbi.com>
Message-ID: <1014747651.11444.14.camel@hunsmac.mit.edu>

> Perhaps Peter should post the relevant code so we can see what is actually
> happening for ourselves?

Here is the relevant portion of the code:

###########################################################################
#
# Get all the information from input file.
#
###########################################################################

ymin, ymax = 0.0, 0.0

bands, x = 0, 0
sym_points, sym_labels = [], []
xtics, labels = [], []
data, kpts = {}, {}
bands_min, bands_max = {}, {}

recip = zeros([3,3], Float)
last_k = zeros([3], Float)

last_label, gap_label = "", ""

title = replace(strip(os.popen("pwd").readline()), "_", "-")

print "---- reading data from", input_fname, "..."

line = inputfile.readline()
while line:

  s = search(r"^gap (.+) with", line)
  if s: gap_label = "Gap: " + s.group(1)

  s = search(r"^Computing (\S+) bands", line)
  if s: bands = int(s.group(1))

  if search(r"^Reciprocal lattice vectors \(\/ 2 pi\):", line):
    for i in range(3):
      s = search(r"\((\S+), (\S+), (\S+)\)", inputfile.readline())
      recip[0,i] = float(s.group(1))
      recip[1,i] = float(s.group(2))
      recip[2,i] = float(s.group(3))

  s = search(r"Gnuplot Title: +(.+)$", line)
  if s: title = s.group(1)

  s = search(r"^Gap from band (\d+) \((\S+)\) to band (\d+) \((\S+)\)",
line)
  if s:
    bands_max[int(s.group(1))] = float(s.group(2))
    bands_min[int(s.group(3))] = float(s.group(4))

  s = search(r"\#\(\S+ +\S+ +\S+\) (\".+\") k-point$", line)
  if s:
    sym_labels.append(s.group(1))
    s = search(r"\#\((\S+) +(\S+) +(\S+)\)", line)
    sp = array([float(s.group(1)),float(s.group(2)),float(s.group(3))])
    sym_points.append(sp)

  s = search(r"^t?e?m?freqs:, (\d+), (\S+), (\S+), (\S+),", line)
  if s:
    ik = int(s.group(1)) - 1 # We always count from zero.
    k = array([float(s.group(2)),float(s.group(3)),float(s.group(4))])

    label = ""
    for i in range(len(sym_points)):
      if sum((sym_points[i] - k)**2) < 1e-6:
        label = sym_labels[i]
    
    if kpts.has_key(ik):
      x = kpts[ik]
    elif last_label and label == last_label:
      kpts[ik] = x # Do nothing, as this is the same point as the last
one.
    else:
      x += sqrt(sum(matrixmultiply(recip, k - last_k)**2))
      kpts[ik] = x
      if label:
        xtics.append(x)
        labels.append(label)
      
    data[ik] = zeros([bands], Float)
    f = string.split(line, ",")
    for n in range(6,len(f)):
      data[ik][n-6] = float(f[n])
    
    ymin = min(ymin, min(data[ik]))
    ymax = max(ymax, max(data[ik]))
    
    last_k     = k
    last_label = label
    
  line = inputfile.readline()






From scarblac at pino.selwerd.nl  Thu Feb 21 03:03:31 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Feb 2002 08:03:31 GMT
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com>
Message-ID: <slrna79a8r.6ot.scarblac@pino.selwerd.nl>

Jason <caljason76 at yahoo.com> wrote in comp.lang.python:
> In my experimentation 0, None, "", [], and () all seem to be
> considered false.  Just to confound me none of them are == to each
> other.  Coming from a Scheme world this is confusing.  Are there any
> other values that are false?

{}, at least. Any classes that return the right results on certain special
functions (__nonzero__(), if I remember correctly).

Why is it normal to you that loads of values are true when they're not equal
to each other, but only one can be false? Python isn't Scheme.

> Why aren't () and "" the same thing?

Because one is an empty tuple and one is an empty string.

> Don't both describe an immutable sequence of values (basically, why
> aren't strings either lists or tuples)?

That would be purity. In Python, practicality beats purity. Tuples are
pretty heavy structures, every element has a reference to an object. A
string is just a row of bytes. Python has to interface with C code and files
a lot, and both in C and in typical files, strings are just rows of bytes.

There are other reasons - to me, they're not at all the same thing
conceptually. Tuples nest, strings don't. For a string a="x", a[0] == a.
You can't do that well and still have normal nested tuples, I think.

To pre-empt your next question: watch out, lists aren't like Lisp list.
They're not cons pairs, they're just an array of references in memory.

-- 
Remco Gerlich


From logstx at bellatlantic.net  Thu Feb 14 18:45:10 2002
From: logstx at bellatlantic.net (logistix)
Date: Thu, 14 Feb 2002 23:45:10 GMT
Subject: help!: optimization of search problem (example code)
References: <3c6bcf70.16509609@news.easynews.net>
Message-ID: <a_Xa8.46$LG6.16@nwrddc04.gnilink.net>

Sorry I can't just spit out some python code, but you might want to
investigate some space partitioning trees.  The three major types are
bsp-trees quadtrees and octrees.  Inserting points will take longer than
building a list, but finding closest points should be significantly faster.

It's definately the way you want to go if your non-test data is dynamic.
--


"Marcus Stojek" <stojek at part-gmbh.de> wrote in message
news:3c6bcf70.16509609 at news.easynews.net...
> Hi,
>
> after some days of playing around I am not getting
> any further and have to ask for some more help.
>
> Here is my problem. I have two large lists ("la" and "lb") of tuples.
> Each tuple represents a point in space and has four entries:
> x, y, z, pointID
>
> Now I have to find out for each point of list "la" which point of
> "lb" is closest. The result should be for all points of "la":
>
> Point 123 of "lb" is the closest one to Point 5932 of "la"
> where 123 and 5932 is the pointID.
>
> So here is what I have so far.:
>
> ----------------------------------------------------------------
> from random import *
> from math import *
> import bisect
> import time
>
>
> # timer for benchmarking
> def startclock():
>     d=time.clock()
>     return d
>
> def stopwatch(start):
>     return time.clock()-start
>
> #generating two arbtitrary lists
> seed(0)
> a={}
> b={}
>
> for i in range(20000):
>     a[i]=(100*random(),100*random(),100*random())
>     b[i]=(100*random(),100*random(),100*random())
>
> la=[]
> lb=[]
>
> for i in range (20000):
>     dummy=(a[i][0],a[i][1],a[i][2],i)
>     la.append(dummy)
>     dummy=(b[i][0],b[i][1],b[i][2],i)
>     lb.append(dummy)
>
> lb.sort() #source list
>
> start=startclock()
> maxdist=5.0 # I know that there are no two points with a distance > 5
>
> for i in range(500): #should be range(len(la))
> #slicing a part out of lb that is closer than maxdist for the first
> #coordinate
>     lp=(la[i][0]-maxdist,la[i][1],la[i][2],la[i][3])
>     rp=(la[i][0]+maxdist,la[i][1],la[i][2],la[i][3])
>     left=bisect.bisect_right(lb,lp)
>     right=bisect.bisect_left(lb,rp)
>     if right==len(lb): right-=1
>     lsearch=lb[left:right] # only compare la[i] to this list
>
>     mindist=10000
>     minhit=-1 #pointID of closest point
>
>     for j in lsearch:
> # i can exclude points with coordinate 2 or 3 further away than
> #maxdist
>         if abs(j[1]-la[i][1])< maxdist and
>            abs(j[2]-la[i][2]) < maxdist:
> #calculate distance:
>
>
d=sqrt(pow((j[0]-la[i][0]),2)+pow((j[1]-la[i][1]),2)+pow((j[2]-la[i][2]),2))
>             if d < mindist:
>                 mindist=d
>                 minhit=j
>
>     if i <= 10 :
>    print la[i],minhit,mindist
>
> bench=stopwatch(start)
> print bench, " sec"
> --------------------------------------------------------------------------
------------------------------
>
> This takes 5.3 sec on my machine for 500 points. That is much too
> long. Any idea how I could significanty improve this code?
>
> Tanks,
> Marcus




From eppstein at ics.uci.edu  Thu Feb 28 18:43:29 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 15:43:29 -0800
Subject: Status of PEP's?
References: <mailman.1014933174.3556.python-list@python.org>
Message-ID: <eppstein-9F0A0A.15432928022002@news.service.uci.edu>

In article <mailman.1014933174.3556.python-list at python.org>,
 "Bjorn Pettersen" <BPettersen at NAREX.com> wrote:

> Since many of the brightest minds on c.l.py have tried hard to come up
> with a general range syntax without much success

I would argue with the "without success" opinion you express. I still like
    for lower_bound <= var < upper_bound
(or similar three-way comparisons, with middle argument of comparison 
required to be a variable), in parallelism with the current "for var in 
seq" with first argument of "in" required to be a variable.  I guess we 
haven't made it into a PEP for the BDFL to rule on, if that's your measure 
of success, but the only arguments I can remember seeing against it are (1) 
some people view "in" as a keyword integrally attached to "for" rather than 
part of an expression, and don't like for loops without "in", and (2) this 
isn't sufficiently terse for those situations where you want an integer 
range outside of a for loop.  Am I misinterpreting the previous discussion?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From cmueller at onebox.com  Fri Feb 22 11:50:28 2002
From: cmueller at onebox.com (rockmuelle)
Date: 22 Feb 2002 08:50:28 -0800
Subject: wxPython MDI question
References: <aaa059f1.0202211632.1f9e9d98@posting.google.com>
Message-ID: <aaa059f1.0202220850.dc7f3c2@posting.google.com>

Ok, found some code for this hidden in the demo directory. 

wxPython/demo/MDISashDemo.py shows how to do it.

-Chris

cmueller at onebox.com (rockmuelle) wrote in message news:<aaa059f1.0202211632.1f9e9d98 at posting.google.com>...
> I am trying to create an MDI application that has a list control on
> the left side of the main MDI frame and the MDI client area on the
> right side, similar to the way Visual Studio has other windows
> surrounding its client area.
> 
> I've tried creating the list control and then adding it and the client
> area to a sizer, but the client area always takes up the whole main
> frame. The client area and the list control are situated at (0, 0), as
> if there was no sizer.  If I add more controls to the sizer, the
> non-client area windows get layed out correctly, but the sizer and
> client area still overlap.
> 
> Does anyone know how to fix this?
> 
> I'm using wxPython 2.3.1 with Python 2.1.1 on Windows 2K.  Code
> demonstrating what I've tried is appended to this message.
> 
> Thanks.
> 
> -Chris
> 
> 
> from wxPython.wx import *
> 
> class MyMDIFrame(wxMDIParentFrame):
>   """
>   Main frame that has list control and MDI client area.
>   """
> 
>   def __init__(self, parent, id=-1, title='MDI Example'):
>     wxMDIParentFrame.__init__(self, parent, id, title=title)
> 
>     sizer = wxBoxSizer(wxHORIZONTAL)
>     sizer.Add(wxListCtrl(self, -1))
>     sizer.Add(wxListCtrl(self, -1))  # add another list
>     sizer.Add(self.GetClientWindow())
>     sizer.Layout()
> 
>     self.SetSizer(sizer)
>     self.Fit()
>     
>     return
> 
> 
> if __name__=="__main__":
> 
>   class App(wxApp):
> 
>     def OnInit(self):
>       frame = MyMDIFrame(NULL)
>       frame.Show(true)
>       self.SetTopWindow(frame)
> 
>       return true
> 
>   app = App(0)
>   app.MainLoop()


From dieter at handshake.de  Sat Feb 23 13:46:39 2002
From: dieter at handshake.de (Dieter Maurer)
Date: 23 Feb 2002 19:46:39 +0100
Subject: Problems with xreadlines and SuSE 7.3
References: <mailman.1014388376.11740.python-list@python.org>
Message-ID: <x7n0y0hw8g.fsf@linux.local>

My VDI Freemail <wagner.nils at vdi.de> writes on Fri, 22 Feb 2002 15:30:57 +0100:
> Python 2.1.1 (#1, Sep 24 2001, 05:28:47)
> [GCC 2.95.3 20010315 (SuSE)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>> import xreadlines
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named xreadlines
> >>>
I can -- but I use a self compiled Python 2.1.2.

I think, you should probably not use the Python distributed with SuSE 7.3.
I found out, that a self compiled one was more than 50 percent
faster than the distributed one.


Dieter


From starx at pacbell.net  Mon Feb 11 22:08:47 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 12 Feb 2002 03:08:47 GMT
Subject: Global Variables in Modules
References: <20020211111025.63f4c956.jsc-news@bluesine.com>
Message-ID: <Xns91B2C2C16D511RASXnewsDFE1@130.133.1.4>

Jeff Clement || Mon 11 Feb 2002 10:10:25a:

> Good Afternoon All,
> 
> I have a project where it would be extremely handy (albeit maybe not so
> clean codeing), to have a global variable in one module be available to
> another module (without a lot of work.  I know I can do module.varname
> = varname from the importing module but I would prefer not to). 
> 
> Below is my sample program.  I would like to launch test.py and declare
> some global variable A, then import test_mod and hopefully get access
> to A from within test_mod.  
> 
> What I'm actually doing is running a code block in an exec block and
> declaring a bunch of globals for that exec.  The code block imports
> some modules and I need those to be able to see the global variables. 
> I can make this work by adding them to __builtins__ but that doesn't
> seem right somehow :) 
> 
> Anyone have any ideas?
> 
> Thanks,
> Jeff
> 
> ---- test program -----------------------------------------
> 
> $ python test.py
> 
> test.py
> -------
> A=100
> import test_mod
> 
> test_mod.py
> -----------
> print A

Well, i'm doing something like this:

gloabls.py:
texture_enabled = 1;

filea:
if button_pushed = something_or_other:
  toggle( globals.texture_enabled ) # i.e x = not x, a helper

fileb:
if globals.texture_enabled:
    	do_long_texturing_stuff()
else:
    	do_something_simpler()


Which is basically the same thing that jason said, except i populate the 
file IN the file. Also, you should probably name it something other than 
'globals', seeing how that is a builtin function name (didn't know at the 
time, got real confused when i went to use it :).

The benefit in either case, is that you put all of your global data into a 
named structure, which makes things less messy. (You could even import it 
as gb or G or something easy to type, just be consistant :)
-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From tjreedy at home.com  Tue Feb 19 11:51:09 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 19 Feb 2002 16:51:09 GMT
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org> <a4ss56$akj6@news1.gtech.com>
Message-ID: <1ovc8.54286$X64.19378143@news1.rdc2.pa.home.com>

"Peter Milliken" <peter.milliken at gtech.com> wrote in message
news:a4ss56$akj6 at news1.gtech.com...
> Strongly typed languages encourage you to override the base types so
that
> you aren't forced one day (as an example) to go through all of the
source
> looking for places where floats are used and change them to double
floats -
> happened on one rather large project I worked :-) They changed the
compiler
> and it defined float as being 32-bit, the previous version defined
float as
> 64-bit - caused no end of heartache as you can imagine :-)

To me, this says that the 'strong typing' of C is a bit of a fraud!
To paraphrase:
Strongly typed languages encourage you to override the base types
because the base types are only loosely defined, so that you cannot
know what actual types they will translate into unless you specify the
machine, compiler, and version.

Python's int and float types are such overridings as you recommend.
They are usually the same as the int32 and float64 typedefs that many
C programmers use (and are guaranteed to be no smaller in any standard
implementation).  If your large project had used Python, it would not
have had that particular heartache.

The ongoing int/long unification will eliminate some current problems
even further by autoswitching base type as needed.  What
strongly-typed-language compiler will do that for you?  I believe such
behaviour violates the usual notion of 'strongly-typed'.  Does that
make it bad?

> So type'ing encourages things like defining data types that
encompass the
> range of values your variable is *expected* to have at the time you
> design/write your code i.e. you might be talking the temperature of
water -
> 0 - 100 Degrees Celsuis for instance.

Strictly speaking, these exact limits only apply to pure water in a
clean vessel bounded above by clean air at exactly standard pressure
and measure by an exactly accurate thermometer.  Even so, I believe
one can, with care, bring the water into a (temporary, unstable)
supercooled state below 0.0C.  Hard-coding ideal expectations into
types can make one's code unnecessarily brittle.

> Languages that are strongly typed like
> this then allow static checking at compile time i.e. the compiler
will tell
> you if you attempt to add a variable declared as Litres to a
Variable
> declared to hold Degrees Celsius.

The same compiler will probably

wrongly reject   2 * 3Liters (= 6 Liters),
wrongly accept  2Liters * 3Liters (= 6 Liters**2, not 6 Liters),
wrongly reject   2LitersperHour * 3Hours (= 6 Liters),

and similarly for division, so such a compiler is worse than useless
unless you restrict yourself to +- operations.  Do you know of any
compiler for any general-purpose language that correctly handles
measurement unit arithmetic out of the box?

One can write a class (or base class and set of classes) in Python
that will do so *correctly*, and I believe there is at least one such
module available.  Such can probably be done more efficiently that
before in 2.2 by deriving from float.

> Similarly, the compiler will complain if you attempt to assign a
temperature of
> 101 Degrees to a variable declared  with the above limits.

"Var = contstant"  is trivial to check, but *most* assignments are of
values not known until runtime.  How about

ta, tb = input("Enter two temperatures: ") # suppose user enters 70,80
tc = ta + tb # should this be rejected as illegal?
tc = tc/2.0 # maybe not
td = (ta + tb) /2.0 # should give 75 even though intermediate 150 is
'illegal'

Will your unnamed compiler generate the *proper* runtime checks?
Should they be at the point of computation, at the point of
assignment, or perhaps the end of a series of computations and
assignments?  Will the compiler generatethe same as what the
programmer considers proper?  Does the programmer get to catch an
'error' or does the program just stop?

> Ada allows run-time checking as well, so if you
> somehow manage to write code that passes the static checking

'somehow manage'?  as if writing 'tc = ta + tb' is difficult?

>it will blow up at run-time with a nice exception and traceback
> that tells you exactly what  and where it all happened :-)

As will raising an exception within a method of a Python class derived
from int or float.

Summary of my comments above:  Static typing only solves a small part
of the programming problem (and possibly less than its advocates
think).  It correctly rejects some incorrect code and enables more
efficient object code, but at the cost of writing more code,
incorrectly rejecting some correct code (hence coercion as a partial
solution), and giving a patina of correctness to some incorrect code.

Another point:  Static typing works against generic programming, which
to me is part of the essence of Python.  The unions and untyped (!)
preprocessor function definitions [as in "#define max(x,y) (x<y?y:x)"]
of C and templates of C++ are attempts to work around this clash.

Terry J. Reedy





From logiplexsoftware at earthlink.net  Mon Feb 11 13:56:44 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 11 Feb 2002 10:56:44 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com>
References: <3C62EE4A.5090305@destiny.com>
	<20020207140755.364d5d77.logiplexsoftware@earthlink.net>
	<200202081311.g18DBQGo020260@ratthing-b246.strakt.com>
	<20020208103714.11a2ae01.logiplexsoftware@earthlink.net>
	<200202110940.g1B9eTGo023834@ratthing-b246.strakt.com>
Message-ID: <20020211105644.4cacaa31.logiplexsoftware@earthlink.net>

On Mon, 11 Feb 2002 10:40:28 +0100
Laura Creighton wrote:

> Grin.  'Cause I know better! <sticks out tongue>

<looks aghast>  Well... um... this sounds like dangerous territory, so I'll
just concur.

> Don't formalise the process too much.  You are not trying to bring order
> to your establishment by building rules that work top-down.  You are
> trying to foster order that works from the bottom up.  How any two
> people best collaborate can vary with the people -- and if you are
> on a roll you may want, need, and deserve a few days with each other
> and only with each other -- otherwise 10 minutes, or an hour to
> kill this one nasty bug may be enough.

I guess I wasn't expressing myself well, but this sounds much like what I
was saying.  Will you please stop using my own arguments against me!  I
have a natural tendency to argue so you'll end up making me argue against
my own position and then I can't tell if I won or not ;)

> Have you read 'Engineering and The Minds Eye'?  I just finished it.
> Wonderful.  

Not yet, but I will now.

Regards,
Cliff

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From tim-dated-1015476021.3d1cbb at catseye.net  Wed Feb 27 23:40:21 2002
From: tim-dated-1015476021.3d1cbb at catseye.net (Tim Legant)
Date: 27 Feb 2002 22:40:21 -0600
Subject: Regular expression problem
In-Reply-To: <nxhf8.149057$dh.38723170@typhoon.nyroc.rr.com>
References: <Npdf8.3552$V15.578531@typhoon.nyroc.rr.com>
	<3C7D85BC.71550E5F@gmx.de>
	<nxhf8.149057$dh.38723170@typhoon.nyroc.rr.com>
Message-ID: <864rk2rzgq.fsf@skitty.catseye.net>

Asheesh Laroia <pan-news at asheeshenterprises.com> writes:

> This is great, thanks!
> 
> Only one problem.  I'm having trouble (I did give it a try) making the
> following work:
> 
> 	<@Trap Body text:>Useful Text
> 
> I need to still be able to extract "Useful Text", not delete it.
> 
> Thanks again!

Try this:

>>> rc = re.compile(r'<@Trap\s+\w+\s+\w+=?(?:<.+?>)*>', re.MULTILINE|re.DOTALL)

>>> text = """<@Trap Body text=<FONT "Times"><CCOLOR\n   "Black"><
11><HORIZONTAL 100><LETTERSPACE 0><CTRACK 127><CSSIZE 70><C+SIZE\n
58.3><C-POSITION 33.3><C+POSITION 33.3><P><CBASELINE 0><CNOBREAK
0><CLEADING -0.05\n  ><GGRID 0><GLEFT 0><GRIGHT 0><GFIRST 19.2><G+BEFORE
0><G+AFTER 0><GALIGNMENT \n  "justify\n  "><GMETHOD "proportional"><G&
"ENGLISH"><GPAIRS 4><G% 120><GKNEXT 0><GKWIDOW \n  1><GKORPHAN\n
1><GTABS $><GHYPHENATION 2 36 0><GWORDSPACE 75 100 150><GSPACE -5 0
25>>Useful Text"""

>>> rc.sub('', text)
'Useful Text'


Tim



From huaiyu at gauss.almadan.ibm.com  Thu Feb 28 20:22:47 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Fri, 1 Mar 2002 01:22:47 +0000 (UTC)
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>        <mailman.1014861466.23611.python-list@python.org>        <slrna7suon.5gr.huaiyu@gauss.almadan.ibm.com> <mailman.1014929989.16553.python-list@python.org> <a5md0a$an8$1@animus.fel.iae.nl>
Message-ID: <slrna7tlv7.68p.huaiyu@gauss.almadan.ibm.com>

On 1 Mar 2002 00:03:38 +0100, Carel Fellinger <cfelling at iae.nl> wrote:
>Skip Montanaro <skip at pobox.com> wrote:
>...
>> Don't assume for loop iteration is your only culprit... ;-) List creation
>> overhead can be as big a problem.  I added another test function called iter
>
>yep, see my other posting:)

Thank you both for 
1) convincing me to install 2.2 (even the same algorithm speeds up 6-7%)
2) altogether speeding it up seven times over my initial version

>Ah a new contestent, double fun, better algoritme better timing (linux
>too), but the real iter still wins, although by a very slight margin:)

>iter 10.95
>iteriteritems 10.43
>iteritems 52.15
>mapvalue 12.88

They are very similar, to say the least.  In any case, the remainder time
seems mostly function call overhead.  Here's my result (also linux).

def emptyloop(f, d):
	for i in xrange(len(d)):		pass

def emptyloopcall(f, d):
	for i in xrange(len(d)):		f(None)

emptyloop 0.67
emptyloopcall 2.39
iter 2.87
iteriteritems 2.94
iteritems 18.63
mapvalue 3.94


Huaiyu


From tripps81 at yahoo.com  Tue Feb 26 14:43:23 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Wed, 27 Feb 2002 02:43:23 +0700
Subject: "pass by reference?"
In-Reply-To: <3C7822E1.508@thinkware.se>
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <5.1.0.14.2.20020227024016.09964cb8@202.95.131.154>

At 24/02/2002 06:16, Magnus Lyck? wrote:
>>Suppose I want to make a function that modifies its argument:
>>  a = -1
>>  make_abs(a)
>>  print a # = 1
>>What is the Pythonic way of doing this? Wrap it in a list?
>>  make_abs([a])
>
>You could make 'a' point to a mutable variable, such as
>a list.
>
>def make_abs(l):
>     return map(abs,l)
>
>a = [-1]
>make_abs(a)
>print a[0] # 1

Ah, yes. Thanks for the correction.

>a = abs(a) is the Pythonic way to do it as others have said.

Got it. Guess I was still thinking in terms of PHP's 
func(&$foo). Also, thanks for the juicy and lengthy explanation!

t




From ron.l.johnson at cox.net  Mon Feb 18 19:06:25 2002
From: ron.l.johnson at cox.net (Ron Johnson)
Date: Tue, 19 Feb 2002 00:06:25 GMT
Subject: confusion regarding os.path.walk()
Message-ID: <20020218180623.086f6c58.ron.l.johnson@cox.net>

Hello,

In the visit callout function, it looks as though after the top level
is processed, os.path.isfile() and os.path.isdir() both return 0 when 
passed elements from the names parameter.

Am I doing something wrong?

Here is an examle:
me at rebel:~/test$ python
Python 2.2 (#1, Jan  8 2002, 01:13:32) 
[GCC 2.95.4 20011006 (Debian prerelease)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> 
me at rebel:~/test$ cat test.py
import os.path, string
def test(arg, dirname, names):
    print arg
    print dirname
    names.sort()
    for n in names:
        print n,
        if os.path.isdir(n): 
            print ' is a DIR'
        elif os.path.isfile(n):
            print ' is a FILE'
        else:
            print ' is UNKNOWN'
    print '================================'
os.path.walk('.', test, None)
me at rebel:~/test$ python test.py
None
.
bar  is a DIR
foo  is a DIR
snafu  is a DIR
test.py  is a FILE
================================
None
./bar
================================
None
./foo
snaggle  is UNKNOWN
swizzle  is UNKNOWN
================================
None
./foo/swizzle
================================
None
./snafu
================================

TIA,
Ron
-- 
+------------------------------------------------------------+
| Ron Johnson, Jr.        Home: ron.l.johnson at cox.net        |
| Jefferson, LA  USA      http://ronandheather.dhs.org:81    |
|                                                            |
| 484,246 sq mi are needed for 6 billion people to live, 4   !
! persons per lot, in lots that are 60'x150'.                |
! That is ~ California, Texas and Missouri.                  !
! Alternatively, France, Spain and The United Kingdom.       |
+------------------------------------------------------------+


From vincent_a_primavera at netzero.net  Mon Feb  4 10:57:10 2002
From: vincent_a_primavera at netzero.net (Vincent A. Primavera)
Date: 04 Feb 2002 15:57:10 +0000
Subject: Python and Postfix filters...
Message-ID: <1012838230.3041.10.camel@del-vap-bos.ralphpill.com>

Hello,
	Has anyone written Postfix content filters in Python?  If so would you
be willing to share an example?
~
	Thank you,
	
	Vincent A. Primavera.





From marklists at mceahern.com  Sun Feb 10 18:02:30 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 15:02:30 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
In-Reply-To: <20020210145042.A25774@glacier.arctrix.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEAGKCAA.marklists@mceahern.com>

[Neil Schemenauer]
> > Wouldn't it make sense to return list('abc') ?
> 
> It would also make sense to return ['a', 'b', 'c'].  Since it's not
> obvious what you want Python raises an error.

That's what that is:

	>>> list('abc')
	['a', 'b', 'c']

// mark



From michael at rcp.co.uk  Thu Feb  7 08:15:14 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Thu, 7 Feb 2002 13:15:14 +0000 (UTC)
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <Xns91AE6BE28FCD8michaelrcpcouk@194.238.50.13> <uk7tpa574.fsf@python.net>
Message-ID: <Xns91AE86CF4427Cmichaelrcpcouk@194.238.50.13>

Michael Hudson <mwh at python.net> wrote in news:uk7tpa574.fsf at python.net:

> Michael Abbott <michael at rcp.co.uk> writes:
> 
>> I reckon to pay a cost of perhaps a factor of 100 in raw CPU for the 
>> privilege of using Python. 
> 
> But not over Java, surely?  I guess a JIT or native compilation will
> help, but...

I think so.  Java byte codes are pretty low level, and don't have to do 
very much, and of course JIT or native compilation can be *very* effective.  
Python byte codes are, on the other hand, much more powerful, and have to 
do quite a lot of run-time work.

If you're doubtful, take a quick peek at what the Python byte code 
interpreter does to implement a method call (I'm thinking of call_object in 
ceval.c).  There's nothing wrong with this (that's not my point), but 
there's quite a lot of work to be done (including memory allocation, I'm 
afraid).

This is one of the most widely used operations in the language (excepting, 
perhaps, method lookup), and can be reduced in a natively compiled language 
to a small handful of instructions or even eliminated (when we're lucky).

My factor of a 100 is a guess; am I wrong?  I believe it's fair.


From magnus at thinkware.se  Tue Feb 19 19:10:48 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 19 Feb 2002 16:10:48 -0800
Subject: Adding static typing to Python
References: <mailman.1014084270.25240.python-list@python.org>
Message-ID: <258fd9b8.0202191610.3ecfaaf1@posting.google.com>

"Milliken, Peter" <Peter.Milliken at GTECH.COM> wrote in message news:<mailman.1014084270.25240.python-list at python.org>...
> I haven't tried any other
> strongly typed language than Ada, so my statements are necessarily limited
> by that background.

Then I can understand that you come to another
conclusion than those who come from a C/C++
background.

I've only used ADA in school, and that was in the
late 80's, but I can certainly say that the static
typing in ADA is much better than the C/C++ variant.

I've just toyed with Java, so I don't know that well
enough to judge.

Of course, you can always define classes to restrict
value sets and limit confusion, but classes are a bit
of overkill for simple values that don't represent any
unique objects. (I mean, from a design point of view,
a car is an object, its maximum speed is not, it's an
attribute of the car, whether speed is implemented as
a class or not.)

> The original post states that the typing of Java was
> perceived as a distinct advantage, so presumably the level of type checking
> there is enough to request the feature in Python. 

But still he used Python instead of Java, since he
had given up on having Java do this text processing.

Since there is no such thing as a Python with Java typing
we don't know if such a thing would be even better. It
is easy to imagine a percieved gain from such a change,
but it's not so easy to predict all the consequences
such a change would cause.

Even if Python is really simple to learn, it does require
a slightly different mindset than C++ or Java, and that
takes a while. It seems that people worry less about static
typing the more they use Python. Sure, humans are very
adaptable, and can get used to almost anything, but I don't
think people would get more and more relaxed about this if
it really caused a lot of problems.

I mean, I've used Windows for more than a decade, I only 
got more and more frustrated with it for each month that 
passed. It's better now, but that's just partly because I
"got used to it", and due to improvements from Microsoft.
The 2 main reasons I can live with Windows are Python and
Cygwin's unix tools.

With C++ new stumble blocks seemed to crop up as I got over
the initial ones. There's a reason there are books like Scott
Meyer's "Effective C++" series that spends 100's of pages on
handling quirks in the pure core of the languare. Python books
can spend less on such things, and instead describe the rich
library.

There were never any major frustrations with learning Python,
and having programmed Perl for less than a year I was more
productive in Python than in Perl in days, but of course we
tend to aquire certain mindsets. Last summer I tried to do
some Perl programming, and it was a rather bad experience. I
was thinking in Python and trying to coax that to Perl. It 
just didn't work...


From tim at vegeta.ath.cx  Wed Feb 13 06:46:34 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Wed, 13 Feb 2002 11:46:34 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013534600.28550.python-list@python.org> <slrna6ivjb.4ol.philh@comuno.freeserve.co.uk>
Message-ID: <slrna6kkf8.is0.tim@vegeta.ath.cx>

phil hunt <philh at comuno.freeserve.co.uk> graced us by uttering:
> Gustavo Cordova <gcordova at hebmex.com> wrote:
>>> >"-ist" - an individual professionally or avocationally 
>>> occupied with the
>>> >idea or activity defined by the root [1].
>>> 
>>> >Of course, in either Spanish or Esperanto it would more commonly be
>>> >"Pythonisto" anyway, so who knows.
>>
>>Kinda wrong for Spanish though.
>>
>>A follower of Zapata ("Emiliano Zapata", perhaps?) would be "Zapatista",
>>be it a male or female. "Zapatisto" is a gross error, and it's funny
>>also :-)
>>
>>It's common for little children to mistakenly use "-isto" as well as
>>"-ista" when talking about boys or girls, that's why it's funny, because
>>it's an infantile mistake.
> 
> Is -ista always feminine? E.g. is it always _la dentista_ not _el dentista_?

Nope. -ista nouns end in -a regardless of gender. If you're
referring to a female dentist, you say "la dentista".  For a male
dentist, "el dentista".  Likewise, a male follower of Zapata: "un
Zapatista". Female: "una Zapatista".

This is one of the more standard exceptions in Spanish.  A very
confusing one regards words coming from Greek. For example:

    "una problema"      # WRONG
    "un problemo"       # WRONG
    "un problema"       # Correct
    "un tema"           # Correct
    "un programa"       # Correct

Tim Hammerquist
-- 
Anyone who thinks that the act of programming is worth
less than $50/hr should leave the industry.
    -- Randal L. Schwartz (clpm, 27 Jul 2000)


From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 10:19:45 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 09:19:45 -0600
Subject: Win32 comport extensions
References: <a3im40$rjt$1@news.efn.org>
Message-ID: <007b01c1acc6$37b81ec0$0101010a@local>

----- Original Message ----- 
From: "Spencer Doidge" <spencer at spencerdoidge.com>


> In the win32 extensions script win32comport_demo.py, line 102 reads:
> rc, data = ReadFile(self.handle, comstat.cbInQue, overlapped)
> 
> I don't see the five arguments I'm used to associating with Win32 API's
> ReadFile(). The nBytesToRead and lpNumBytesRead args are not there.

You REALLY need to read the docs provided in the Win32\Help folder; look
for the file named PythonWin32Extensions.hlp.  Use the Index, Find feature 
to locate the ReadFile function.

I considered explaining the details here, but there are two very different
cases, so you should just RTM yourself on this one.

BTW Many people don't even realize that help file is there...





From logiplexsoftware at earthlink.net  Tue Feb  5 20:21:50 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 5 Feb 2002 17:21:50 -0800
Subject: ASPI from Python
In-Reply-To: <a3putt$9sp$1@news.wplus.spb.ru>
References: <a3putt$9sp$1@news.wplus.spb.ru>
Message-ID: <20020205172150.22ffdafd.logiplexsoftware@earthlink.net>

On Wed, 6 Feb 2002 03:55:25 +0300
Dmitry Rozmanov wrote:

> Hi.
> 
> How can I interface to ASPI from python in Windows2000? Is it possible?
> 
> How can I talk with a scanner from python in Linux?

print "hello, scanner"

There is supposed to be a module that supports SANE under Linux:

About: http://www.linuxjournal.com/article.php?sid=2395
SANE: 	http://www.mostang.com/sane/
Python API: http://starship.skyport.net/crew/amk/projects/

I haven't checked this out so YMMV

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


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From altis at semi-retired.com  Sat Feb  9 21:14:44 2002
From: altis at semi-retired.com (Kevin Altis)
Date: Sat, 9 Feb 2002 18:14:44 -0800
Subject: Animated graphs / wxPython
References: <a43rur$31f$1@newsg1.svr.pol.co.uk>
Message-ID: <IGk98.194$YT2.114543@news.uswest.net>

PythonCard http://pythoncard.sourceforge.net/ includes a BitmapCanvas class
(component) that provides automatic double-buffered drawing and you can
control when updates are made to the on-screen buffer (what the user sees).
The doodle, hopalong, and turtle sample apps included with PythonCard use
the BitmapCanvas class, so there is example code to learn from. You can see
screen shots of the samples in action at
http://pythoncard.sourceforge.net/samples.html The BitmapCanvas component
supports many of the wxDC drawing primitives, but I haven't implemented all
of them yet. If there are particular ones you need such as ellipses
(mentioned below) that would be easy to add.

Since this was done for PythonCard it won't work without modification in a
plain wxPython application, even though PythonCard uses wxPython as its GUI
toolkit. Depending on how far along you are on your wxPython app you might
consider just using PythonCard instead. You could also adapt the
BitmapCanvas class to your own needs or use it as an example of
double-buffered drawing and roll your own.

ka

"Duncan Smith" <buzzard at urubu.freeserve.co.uk> wrote in message
news:a43rur$31f$1 at newsg1.svr.pol.co.uk...
> I need to frequently redraw a graph (ie. nodes and edges) to a wxPython
> panel.  ie. I remove and/or add edges to the graph and then redraw it.  I
am
> currently using wxPython ellipses and lines for drawing the nodes and
edges.
> The problems are the speed of redrawing and the flickering as the window
is
> redrawn.  SetClippingRegion() doesn't help a great deal as the clipping
> region is usually nearly as large as the full graph.  I am currently using
a
> workaround whereby I draw edges in the background colour in order to
remove
> them.  But after a number of redraws the graph starts to get noticeably
> whited out by drawing over existing nodes and edges.
>
> Does anyone know of a better approach?  Ideally I would like to be able to
> simply remove and add edges and send them to the back of the nodes rather
> than redrawing the relevant nodes each time I add an edge.  I have looked
at
> a number of packages, but none seem to be able to do what I want.
> Essentially I want an animated graph that will show a user what the graph
> algorithms are up to.  (Gato won't do the job.)  Thanks in advance.
>
> Duncan
>
>




From christophertavares at earthlink.net  Sun Feb 10 14:38:13 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Sun, 10 Feb 2002 19:38:13 GMT
Subject: box on soap
References: <mailman.1013283562.647.python-list@python.org> <Lli98.24849$l93.4538901@newsb.telia.net>
Message-ID: <F_z98.32573$3E5.2317813@newsread2.prod.itd.earthlink.net>

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:Lli98.24849$l93.4538901 at newsb.telia.net...
> Mark McEahern wrote:
> > Don Box on the Importance of Being WSDL
>
> David McCusker on the Importance of Ignoring Don Box:
>
> http://www.treedragon.com/ged/map/ti/newFeb02.htm#06feb02-box
>
> (I'll just give you a summary. I won't argue.  David's right
> and Don's wrong ;-)
>
> </F>
>

Don hasn't shipped an actual product in years, so his perspective is a
little odd. Being a professional training celebrity will do that to you. :-)

Actually, there *is* one advantage to having WSDL that isn't about enforcing
contracts - having WSDL around will let you build better tools. Codegen of
proxies & stubs. Better IDE integration. Web services "browsers". These are
all things worth having, I would think. It doesn't need to be WSDL
specifically, of course, but *some* sort of interface definition needs to be
around. WSDL is a baroque mess right now, but it will get better.

-Chris





From aldo at nullcube.com  Sun Feb 17 23:13:20 2002
From: aldo at nullcube.com (Aldo Cortesi)
Date: Mon, 18 Feb 2002 15:13:20 +1100
Subject: Warning about "for line in file:"
In-Reply-To: <20020216093228.A15411@glacier.arctrix.com>
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu> <20020216134444.GA12676@nullcube.com> <20020216093228.A15411@glacier.arctrix.com>
Message-ID: <20020218041320.GA20517@nullcube.com>

Thus spake Neil Schemenauer (nas at python.ca):

> > iter(file) creates a line iterator that does the same
> > thing as file.readline() every time .next() is called,
> > until it reaches the end of the file.
> 
> It does not.  iter(file) calls file.xreadlines().
> xreadlines() internally calls readlines(CHUNKSIZE).

Sure... I had misread the question by the OP - I thought he
was baffled by the fact that the file position didn't get
"rewound" when a new iterator was created...

> > But file.readline(), just like any other file read,
> > starts reading at the _current seek position_ of the
> > file. 
> 
> True but irrelevant to the discussion at hand.

Not so. The nub if the matter is still that when the second
iterator is created, it "picks up" at the current file
position. The difficulty is just that in the case of
xreadlines, reads happen in chunks of 8192 bytes. 

---

class my_xreadlines:
    def __init__(self, file):
        self.file, self.xr = file, file.xreadlines()
        print "Creating iterator at:", file.tell()

    def __iter__(self):
        return self

    def next(self):
        self.pos = self.file.tell()       
        line = self.xr.next()
        if self.pos < self.file.tell():
            print "Read up to:", self.file.tell()
            self.pos = self.file.tell()
        return line

def main():
    # "test" is a file larger than 8192 bytes
    f = open("test")
    for i in my_xreadlines(f):
        pass

    for i in my_xreadlines(f):
        break

main()

---


-- 
Aldo Cortesi
aldo at nullcube.com
www.nullcube.com




From tanzer at swing.co.at  Thu Feb 21 01:49:16 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Thu, 21 Feb 2002 07:49:16 +0100
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: Your message of "20 Feb 2002 09:57:36 EST."
             <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <E16dn2a-0003aq-00@tswings.swing.cluster>

Russell Nelson <nelson at crynwr.com> wrote:

> There are two Python ideoms that bother me:
>
>     while 1:
>         line = file.readline()
>         if not line: break

No need for this one in 2.2.

> and:
>
>     matched = re.match(r'egexp(.*)', line)
>     if matched:
>         print matched.group(1)

It's trivial to write a class Regexp which allows this style of code:

    pat = Regexp(r'egexp(.*)')
    ...
    if pat.match (line) :
        print pat.last_match.group (1)

> Should I write up a PEP or has this already been discussed and
> dismissed?

It's been discussed time and again. Writing a PEP and having it
dismissed might make future discussions a bit shorter <wink>. And
then, it might not.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From Stefaan.Eeckels at ecc.lu  Thu Feb  7 15:44:27 2002
From: Stefaan.Eeckels at ecc.lu (Stefaan A Eeckels)
Date: Thu, 7 Feb 2002 21:44:27 +0100
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net>
 <uelk6750s.fsf@synopsys.com>
 <27327.798T2437T205791threeseas@earthlink.net>
 <uadup5dsf.fsf@synopsys.com>
 <4898.802T2975T395701threeseas@earthlink.net>
 <ug04ey1on.fsf@synopsys.com>
 <1731.803T2524T145894threeseas@earthlink.net>
 <uy9i5b5e6.fsf@synopsys.com>
Message-ID: <20020207214427.0a85ecd6.Stefaan.Eeckels@ecc.lu>

On 07 Feb 2002 08:53:05 -0800
David Masterson <dmaster at synopsys.com> wrote:

> >>>>> Timothy Rue writes:
> 
> > I've identified and defined the tool and it's natural components.
> 
> > The problem is in getting an industry that has been programmed to
> > depend on some level of manual control over what the users can and
> > can't do, to let go of such control and make available such a
> > natural automation tool as identified, so that the user can do
> > things for themselves, rather than being dependant on the
> > programmers and whether or not they have the resources to get a
> > programmer to do what they need done.
> 
> Thus far, I see nothing "natural" in your approach.
> 
> > The problem is perhaps to prevasive for you to see?
> 
> Possibly.  I may have been in the "forest" for so long that all I can
> see are "trees".  On the reverse side, perhaps you've been looking at
> the "forest" for so long that you don't see how each "tree" has an
> impact on it.

IMHO, Timothy's problem is that he overestimates the importance
of the tool. If the computer has to supply the information to
disambiguate the user's instructions, it needs to be provided
with that information.  He's right in his perception that the
tool itself can be rather simple, and concerned mostly with the
dynamic creation of links between bits of information; after all,
that's the way the brain goes about it.

The problem is how to get the human context in the computer. 
So much of what we take for granted depends wholly on us being
humans, with human I/O systems, that trying to simulate it with
a keyboard and a (graphical) screen as frontends to a hierarchical
filing system is all but impossible. Growing up as humans in a
human society is probably the only way to equip a "thinking machine"
with the context to exhibit human intelligence. Surely, the machine's
capability to handle the complexity needs to be present, but our
capability to formalize our own thought processes is so deficient
we cannot hope to feed it into something that is not a human
brain. 

Take care,

-- 
Stefaan (GPG Fingerprint 25D8 551B 4C0F BF73 3283 21F1 5978 D158 7539 76E4)
-- 
"Object-oriented programming is an exceptionally bad idea which
 could only have originated in California." --Edsger Dijkstra


From rvr at idec.es  Sat Feb 23 11:54:38 2002
From: rvr at idec.es (=?ISO-8859-15?Q?V=EDctor_R=2E_Ruiz?=)
Date: Sat, 23 Feb 2002 16:54:38 +0000
Subject: Python XML-RPC server as CGI
Message-ID: <a58hff$5ep4i$1@ID-76296.news.dfncis.de>

  Hello:

  I'm using Python 2.1 I want to run an xmlrpc server as a python CGI 
script, executed via Apache: it is possible with the Python 2.1's standard 
xmlrpc library?

  Thanks,

-- 
V?ctor R. Ruiz
http://www.infoastro.com


From bldrake1 at yahoo.com  Wed Feb  6 10:46:05 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 6 Feb 2002 07:46:05 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <st888.7971$oI4.33917985@newsserver.ip.pt>
Message-ID: <fb072181.0202060746.434ae0e4@posting.google.com>

Should have added one more line to my previous post.  Here is the
entire story:

Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import math
>>> n = 10
>>> nf = 10.0
>>> a = 1.0
>>> b = 0.5
>>> a/n
0.10000000000000001
>>> b + a/n
0.59999999999999998
>>> a/nf
0.10000000000000001
>>> b + a/nf
0.59999999999999998
>>> b
0.5
>>> 1.0 + (0.6 - (b + a/n))
1.0
>>> 

This last line is the crucial one.  It's says that within the
numerical precision of the machine b + a/n "looks and behaves as close
as I can make it" like 0.6.  Remember 64 bit doubles are finite.

Barry Drake

"Hugo Martires" <hmartires at cultalg.pt> wrote in message news:<st888.7971$oI4.33917985 at newsserver.ip.pt>...
> but i don?t need all the numbers, i only need 0.6
> 
> 
> "Jason Orendorff" <jason at jorendorff.com> wrote in message
> news:mailman.1012956250.23108.python-list at python.org...
> > Hugo Martires writes:
> > > suposing that n=10 we get: myArray[0] = 0.600000023 ("something
> > > like this")
> > >
> > > but if i do: myArray[i] + (1.0/n) it give's me 0.6
> > >
> > > what's the problem of the array, and how can i solve this ?
> >
> > It's only half of the normal precision.  From the documentation:
> >
> >     The following type codes are defined:
> >         Type code   C Type             Minimum size in bytes
> >         'f'         floating point     4
> >         'd'         floating point     8
> >
> > Try using 'd' instead of 'f'.
> >
> > ## Jason Orendorff    http://www.jorendorff.com/
> >
> >


From skip at pobox.com  Wed Feb 27 17:15:17 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 27 Feb 2002 16:15:17 -0600
Subject: distutils source dist question
Message-ID: <15485.23157.157705.938801@beluga.mojam.com>

I have a small library consisting of three C source files:

    alf.c
    alf.h
    alfmodule.c

The first two files are a standalone library, while the third is an
extension module wrapping the library.

The naive "python setup.py install" command creates alfmodule.so and
installs it in site-packages.  I'd like libalf.so or libalf.a to also get
created and installed in ${prefix}/lib and alf.h to be installed in
${prefix}/include.  Can this be done with distutils or do I need to fall
back to Makefile for this?

Thanks,

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From starx at pacbell.net  Sat Feb 16 00:17:30 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 05:17:30 GMT
Subject: newbie asking for info about Python
References: <xkka8.2932$VX6.245554@weber.videotron.net> <mailman.1013620406.29808.python-list@python.org>
Message-ID: <Xns91B6D890E6813RASXnewsDFE1@130.133.1.4>

Laura Creighton || Wed 13 Feb 2002 09:09:59a:
> Can you please not send HTML mail out to this list.  It is very hard
> on those of us that read this not as a newsgroup but as a digestified
> mailing list.  

It's not particularly nice to those of us who use a newsgroup program 
that knows it's job and only displays raw text, either =).


Joseph Youssef || Tue 12 Feb 2002 06:21:25p:
> Hello, I know nothing about Python but before I start learning I would
> like to know what Python is usually "used" for, I mean I don't think
> it's made to write some kinda application or something, so what
> exaclty is it used for? 

Well, it certanly is used for applications, although i can't think of 
any off the top of my head. Games too, see pygame for that. I'm actually 
working on a 3d roguelike completly in python, and it's only a little 
bit slow (and thats only because in my current setup every frame drawn 
involves either 1200 or 1600 function calls, i can do better =). (There 
are some screenshots at rubydragon.com/image/sproj -- not very good 
looking yet though :).

The other thing i find it useful for is 'complex file operations'. 
Simple: i have a script that recursivly calls 'unzip xxx.zip -d xxx' for 
each file in a tree. Complex: I have a suite of scripts that turns a 
tree full of .msxhtml, .sxhtml, and .info.sx files into useable .html 
files (composing the gallery is particularly involved), copies all 
relevent files to a 'dist' directory, and (using Stefan Schwarzer's 
upload.py, slightly tweaked), uploads everything to my site. Nothing is 
generated if its source file hasn't changed.

There's lots of other good stuff too =).

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From marknenadov at winningteam.com  Mon Feb 25 15:05:37 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Mon, 25 Feb 2002 15:05:37 -0500
Subject: mcrypt core dumping
Message-ID: <20020225.150537.1605908235.23755@d57-33-82.home.cgocable.net>

Is anyone out there using the Python mcrypt module?

I am using mcrypt and every third or fourth time I run my program I get a
segmentation fault (core dump).  I have narrowed it down to the line that
calls mcrypt_generic(). The following is the function that I
made:

from mcrypt import *

def mcrypt_encode( text, passcode ):
	cipher = "tripledes"
	key_size = mcrypt_module_get_algo_key_size( cipher, genNULL() )
	key = gen_key( key_size, passcode, len(passcode))
	
	module = mcrypt_module_open( cipher, genNULL(), "cfb", genNULL() )
	mcrypt_generic_init( module, key, key_size, gen_iv( module ) )
	
	buffer = mkbuffer( text )
	
        #this is the line that appears to be causing the core dump
        mcrypt_generic( module, buffer, len( buffer ) )
	ctext = getbuffer( buffer, len( buffer ) )

	mcrypt_free( buffer )	
	mcrypt_generic_end( module )
	mcrypt_module_close( module )
	mcrypt_free( key )
		
	return ctext	


Is there anything wrong with that? Anybody have any clues as to what may
be causing this?

~Mark Nenadov


From donn at drizzle.com  Fri Feb  1 00:15:02 2002
From: donn at drizzle.com (Donn Cave)
Date: Fri, 01 Feb 2002 05:15:02 -0000
Subject: Java vs Python Benchmarks: Java is faster
References: <Ldc68.67$8u4.28600@wagner.videotron.net>
Message-ID: <1012540501.71692@yasure>

Quoth Bryan <bryan at eevolved.com>:
|
| For what it is worth to people, I have taken the liberty of benchmarking 
| the new Java SDK ( 1.4.0 ) against Python 2.1 and 2.2. The tests originate 
| from:
|
| http://www.twistedmatrix.com/users/glyph/rant/python-vs-java.html
|
| Point form Summary:
...
| Interesting differences between Python 2.1 and 2.2:
|
| - I/O is slower in 2.2
| - List and Hash are slightly improved in 2.2
| - Object allocation shows slight improvement
| - Interpreter intialization time has doubled in 2.2

Doubled!!?  Could that be some kind of site glitch?

	Donn Cave, donn at drizzle.com


From bosahv at netscapenospam.net  Sat Feb 16 03:31:49 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Sat, 16 Feb 2002 08:31:49 GMT
Subject: Has anyone used Zorp????
Message-ID: <VNob8.383622$pa1.90623055@news3.rdc1.on.home.com>

I'm looking to replace a firewall and I thought that a python configurable
wall would be better than not.

I haven't heard anything about Zorp, which could be a good thing, so I was
wondering if anyone here has played with it???

Bo




From brian at sweetapp.com  Mon Feb  4 11:48:42 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Mon, 4 Feb 2002 08:48:42 -0800
Subject: SilverCity 0.1.0
In-Reply-To: <d3m78.34171$Ni2.194073@news-server.bigpond.net.au>
Message-ID: <003301c1ad9b$ce1344f0$445d4540@Dell2>

Neil Hodgson wrote:
> Assuming you just mean moving it from private to public, OK, I'll
> change that after 1.44 is released.

Thanks.
  
> Not my preferred way of working. I generally prefer writing
> navigational code rather than  visitors or event handlers. 
> Other people really like these styles of interface and they 
> are certainly popular for XML.

Fair enough.

Thanks again!

Cheers,
Brian




From rdsteph at earthlink.net  Sun Feb  3 00:51:40 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 03 Feb 2002 05:51:40 GMT
Subject: Automatic daily updates on Pythonic web articles works-finally!
References: <3C5CCE13.42EB95BF@earthlink.net>
Message-ID: <3C5CD0C4.5E7B7329@earthlink.net>

By the way, any Mac fans out there should definitely check out Steven
Figgin's O'Reilly Network article at
http://www.onlamp.com/pub/a/python/2002/1/31/pythonnews.html about
"Cocoa for your Python". I think its a great article and I don't want it
to get buried and lost in the pile of IBM Developer Works articles at
the top of the list right now. The "Cocoa for your Python" article is
just the sort of gem you like to find in the list, so if there's a
string of "bad" articles all from one place at any time, just scroll
down and look lower in the short list; at least the list of 200
downloads really quickly and you can scan it for articles of interest
quickly. 

Try to remember its a work in progress and I hope we can improve its
output over time...


Ron Stephens wrote:
> 
> http://www.awaretek.com/plf.html now has a link that gives a new list
> each day at 7:30 A.M. of the last 200 web articles about Python.
> 
> The full archive of all of the thousands of such articles since the dawn
> of time is also still available, but it takes a much longer time to
> download ;-)))
> 
> It's not perfect; it is still a work in progress, but at least now (if
> it all continues to work !) we can expect a new list each morning of
> interesting articles. Perhaps we are dealing with a case of nascent
> artificial intelligence, since the resulting list is just a wee bit
> indeterminate; so at least you know you'll see new articles each day,
> although a few might be actually 'old" articles that didn't show up
> before for some reason (right now IBM's Developer Works has a string of
> such articles at the top of the list, but hey they're great articles and
> we should all read Dr. David Mertz' stuff again anyway;-))). Also, a few
> non-Python but nevertheless interesting programming related articles
> show up from time to time.
> 
> I can't guarantee it will continue to work each day (I am out of town
> again all week and the computer could crash) but I think it will and it
> is interesting to see fresh Pythonic articles each day, thanks to Hans
> Nowak for the mygale web spider program!
> 
> Also, you might want to check out the link to the "Tech News Web
> Robot-updated each 15 minutes". This is a more general tech news web
> spider that updates more or less continuously all the time and grabs
> links to breaking news stories about technical subjects (mostly
> computer-related) and scrolls them across the screen, with older stories
> dropping off the bottom, each story might be up for 2-3 hours on a week
> day before it disappears into the ether.
> 
> Also yet again ;-))), I have entered the first installments of my
> experiences with Linux under the Linux Chronicles. I switched cold
> turkey three weeks ago, and although I have been gone from home most of
> the time since then, I am finding Linux to be mostly great, and
> certainly a great environment for Python, but of course I have
> encountered some difficulties.
> 
> And more new stuff is there also...
> 
> http://www.awaretek.com/plf.html
> 
> Ron Stephens
> 
> P.S. Thanks again to Hans Nowak...


From jeff at ccvcorp.com  Wed Feb 27 14:35:24 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 11:35:24 -0800
Subject: [Python-Help] advocacy help ( Python vs VB )
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01> <mailman.1014706963.7898.python-list@python.org> <9cem7uc521ni47v2mo0qiv1dfougdciqp6@4ax.com> <9588af6.0202262221.41b4ae9d@posting.google.com>
Message-ID: <3C7D34FB.DBFDD0F2@ccvcorp.com>

mike wrote:

> Courageous <jkraska at san.rr.com> wrote in message news:<9cem7uc521ni47v2mo0qiv1dfougdciqp6 at 4ax.com>...
> > While this is academic, things like this are going to come up again
> > in the future. What would really sell them is a Python which compiled
> > to CLR and gave them full access to the suite of COM/.NET objects.
> > Many "VB shops" require this. He should take a long gander at win32all
> > and make sure that it fits their need.
>
> Python.NET does exactly this. This is available from Activestate.

Python.NET *would* have done exactly this, had it ever come to fruition.  It was an experimental project,
however, that has apparently been discontinued.  While the experimental code might or might not be
available from ActiveState, I sincerely doubt that it's useful in a production environment.

Note that Python.NET is distinct from VisualPython.Net -- the former is a Python implementation that will
run in MS's .NET CLR; the latter is a Python-editing plugin for Visual Studio.Net.  The plugin is
available, and works (or so I presume, haven't tested it myself though ;) ).  The Python-for-.NET-CLR is
still vaporware.

Jeff Shannon
Technician/Programmer
Credit International




From jason at jorendorff.com  Tue Feb 26 12:52:58 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 26 Feb 2002 11:52:58 -0600
Subject: socket in python 2.2 vs 2.1
In-Reply-To: <442684269.20020226084449@kupol.ru>
Message-ID: <HFEKILOLEFEFMKAECNDLOEFKDFAA.jason@jorendorff.com>

Ivan Bykov wrote:
> IMHO, '255.255.255.255' is a valid IP number. This is
> number mask, use for broadcast packet.
> MvL> Just remove the connect call.
> Then program to work not will.

You are sending UDP broadcast packets?
I think the right way to do this is:

  import socket
  s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
  s.sendto(text, ('255.255.255.255', port))

That is, I think it is a mistake to call s.connect().
I believe most sockets libraries allow you to call connect()
on a DGRAM socket, and they interpret that as the programmer
setting a default destination for datagrams.  However, it
seems more natural (and more faithful to the nature of UDP)
to specify the destination with sendto().

(Martin:  255.255.255.255 is the limited broadcast address.)

## Jason Orendorff    http://www.jorendorff.com/




From fredrik at pythonware.com  Tue Feb 12 10:42:15 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 12 Feb 2002 15:42:15 GMT
Subject: Drawing in "Slow motion".
References: <3C692698.C9DAD820@inenco.no> <a4ba6i$14bg$1@norfair.nerim.net>
Message-ID: <rJaa8.17932$n4.3113013@newsc.telia.net>

Eric Brunel wrote:
> - keep your time.sleep calls, but call the update_idletasks method on the
> canvas after each sleep. The call should redraw the screen.
> - a bit trickier: use the delayed callback facility in Tk via the "after"
> method to draw while your application is running. This is a bit more
> complicated and you would certainly have to rewrite your code differently.

if you're using 2.2, you can replace the call to time.sleep() with a
yield, and add a small wrapper like the "func" function below:

from __future__ import generators

import random
from Tkinter import *

w, h = 500, 500

root = Tk()
cnv = Canvas(root, width=w, height=h)
cnv.pack(side=TOP)
Button(root, text='Enough', command=root.quit).pack(side=TOP)

def draw():
    x, y = random.randrange(w), random.randrange(h)
    while 1:
        oldx, oldy = x, y
        x, y = random.randrange(w), random.randrange(h)
        cnv.create_line(oldx, oldy, x, y)
        yield 500 # sleep

def func(iter):
    root.after(iter.next(), func, iter)

func(draw()) # get things going

mainloop()

</F>

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






From sag at hydrosphere.com  Wed Feb 20 16:17:29 2002
From: sag at hydrosphere.com (Sue Giller)
Date: Wed, 20 Feb 2002 14:17:29 -0700
Subject: No space left on device
Message-ID: <20020220211326403.AAA230@mail.climatedata.com@SUEW2000>

I am hoping someone here can give me some insight.

I have a program that generates a LOT of small files in a 
Windows2000 subdirectory.  It is my understanding (often faulty) 
that there is no limit to the number of files such a subdirectory can 
contain.  However, I am getting an error when I hit about 21,843 
files.

  File "C:\Python21\Scripts\Tests\ManyFiles.py", line 14, in ?
    fp = open(fileName, 'wb')
IOError: [Errno 28] No space left on device: 'c:\\legacystoret\\~916-
22850tst.pck'

I have over 5 G of space on the drive, and the test files are 0 len 
anyway.  (It still happens if the files are not 0 len)

I can then create additional files in that directory from within a DOS 
window, but I cannot rerun my file generator to make even one more 
file.   So I am wondering if the limit comes from Windows, from 
Python 'open...' or some other innards of Python.

I'm hoping someone can shed some light as to where this error is 
coming from.  I am hoping I don't have to totally change my file-
generating system to lessen the number of files, since 22,000 is just 
a start on the count I may need.

TIA

I have reduced the code to the following simple case. 

# ----- CODE -----
# manyFiles - generate many files with temp names to see if we 
reach a limit

import tempfile

maxFileCount = 100000           # fails at about 21,843 files
seed = 'tst'                              # base for generated filenames
base = r'c:\\storetlegacy'		# dir that gets the files

for ii in xrange(maxFileCount):
    tempfile.tempdir = base
    temp = tempfile.mktemp(seed)
    fileName = "%s.pck" % (temp)    
    fp = open(fileName, 'wb')	# error comes here
    fp.close()
    if ii % 50 == 0:
        print "filecount", ii			# kekep track of how far we are 
along the count



From MaximilianScherr at T-Online.de  Tue Feb 12 17:23:30 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Tue, 12 Feb 2002 22:23:30 -0000
Subject: hex to int
Message-ID: <a4c4l2+hmnu@eGroups.com>

how can i convert hex to int?




From seebs at plethora.net  Sat Feb  9 12:57:16 2002
From: seebs at plethora.net (Peter Seebach)
Date: 09 Feb 2002 17:57:16 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <2106.802T789T14243611threeseas@earthlink.net> <3c647f02$0$22477$3c090ad1@news.plethora.net> <3665.805T682T7313596threeseas@earthlink.net>
Message-ID: <3c6562fb$0$79557$3c090ad1@news.plethora.net>

In article <3665.805T682T7313596threeseas at earthlink.net>,
Timothy Rue <threeseas at earthlink.net> wrote:
>On 08-Feb-02 20:44:34 Peter Seebach <seebs at plethora.net> wrote:
>>In article <2106.802T789T14243611threeseas at earthlink.net>,
>>Timothy Rue <threeseas at earthlink.net> wrote:
>>>C# is a subtopic of .NET / Mono, as is many other programing languages,
>>>including Python.

>>Unless, of course, you use that to build *your* framework.  They need
>>options.  Give them some.

>Who needs options?

The C# community.

>Or the community most likely to mimic it and try and shoot it down. MS has
>all the bucks to bullshit the public into believing what ever it is they
>want the public to believe. Embrase and extend right?

The programmers aren't the same as MS.  Anyway, if they *do* manage to mimic
it, then you win - we all win.  Autocoding will become a practical reality.

>>How would you know?  You never really know; you simply observe that your
>>current model covers everything you've seen.

>???? You assume what I've presented is a model that I've come up with and
>based on only what I've seen? You are greatly mistaken and in such, you
>also expose a bit of C# mindset. If you are to be an example of the
>Acceptance of what I present, then you are only accepting what you want
>to believe, not what is.

I'm not a C# person.

My assumption is that all human beliefs are based entirely on the available
data, and are subject to change when new data becomes available.

>There are over three thousand programming language. The VIC is not
>programming language dependant and as such I don't see any reason for me
>to learn another language, because there is always going to be someone
>that says what you do about some other language.

Ahh, but what you need isn't *language*, it's *community*, and the C#
community is the best match for you.

>I'm sure the VIC could be
>written in many different languages, so? What is important is having it
>written in a language that is widely (cross platform) available and freely
>available. C# doesn't fit that spec. but it doesn't mean the VIC can't be
>used to generate C# code.

Miguel de Icaza thinks that the .NET framework is the salvation of software
development on a lot of platforms.  I think you and he have a lot in common,
and if you can convince people like him of your plan, you have a better chance
of making it work.

>I'm in/on usenet where there is an unbiased date time stamped archiving
>machinery attachability (important in legal matters).

Good point.

>BUT YOU HAVEN'T NAMED A USENET C# NEWSGROUP, HAVE YOU?

No.  I assume there is one, although I don't know where it is.

>I have no fear of going into such a newsgroup. But then that is not what
>you are suggesting, is it? You really want me to just leave here and maybe
>go behind some damned closed door mailing list that has no inherent
>assurence that it can provided any unbiased date time stamped public
>accessible archive.

Actually, that had never occurred to me.  That does look like a problem,
but I'm sure you can find one.

Hmm.

microsoft.public.dotnet.languages.csharp

Well, that's got its up side and its down side.  The up side is, Google
caches it, so it's got the public archive.  The down side is, the microsoft
hierarchy isn't universally accessible, although I know all the news servers
I use carry it.

>Name your favorite C# usenet newsgroup. If you pick a moderated one, then
>you pick a censored one, and you will not know the truth from it.

I object to the allegation that all moderated newsgroups are "censored" in
any meaningful sense.  The ones I'm familiar with don't post porn spam,
but that's about it.

Anyway, the above-named group is a public Usenet group,
moderately-well-propagated, and, despite the name, it's Usenet, and Microsoft
has no control over it.

>I have no reason to believe what I present will have any different effect
>on such a group of people as it apparently has here. You should think
>about that.

The people here (myself included) are hopelessly tied to conventional notions
of software engineering practice.  You need minds unfettered by such concerns.

-s
-- 
   Copyright 2001, all wrongs reversed.  Peter Seebach / seebs at plethora.net
   $ chmod a+x /bin/laden      Please do not feed or harbor the terrorists.
     C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/


From jkraska at san.rr.com  Fri Feb  8 18:49:00 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 23:49:00 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu> <3c641e04.3543807919@news>
Message-ID: <tcp86uggrfsvupadn6jt4h322m8nbcsdjn@4ax.com>

>People take turns. Everyone agrees there's no point to replying,
>but at any given time there are plenty of people with nothing
>better to do (or too pissed off at the guy to be rational about
>it, or both.)

The basic reality of it is that sometimes arguing with an oddball
is a tad more interesting than nursing one's ennui while watching
the latest poorly inspired television show. Either which way, it
also offers more entertainment than watching paint dry. :-)

C//



From jkraska at san.rr.com  Tue Feb  5 21:41:14 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 06 Feb 2002 02:41:14 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c6011e3$1_5@news.bluewin.ch> <tor06u0j1ce99ibgd98uflckcms89apfci@4ax.com> <3c607be8$1_1@news.bluewin.ch> <9k116u4r58thf9lp69cso89ubtvvhhsm7f@4ax.com> <7xr8nze4w4.fsf@ruckus.brouhaha.com>
Message-ID: <2d516ukpf7l5p1tp0v19a3ekadkmehuefv@4ax.com>

>> At the moment, no. I have my own lexer, parser, validator
>> and emitter. The first two are fairly close to done, missing
>> a few notables, like comprehensions. The last are very much
>> a work in progress. Lotto work there; still in design on those.
>
>Is this written in Python?

No. They are written in C++. It _interprets_ (well, right now,
just parses) Python. It's pretty fun learning the Python team's
own lessons the hard way. I bump into these problems and then
say to myself, "God, this is really hard, how the Devil did the
Python guys get past this," only to discover they didn't and now
I've learned why. Most educational.

Let's just say that a language which uses implicit indent and
dedent as a grammatical delimiter is limited in the ways it can
be practically instantiated in the real world by virtue of the
constraints themselves. Given the nature of that particular
set of constraints, it turns out that Python is a very good
realization of the solution -- something very similar to what
I'd call a "natural mapping".

If I were to do my own indent-generated scoped language, I
can't say that it would really _look_ all that different from
Python.

C//



From piet at cs.uu.nl  Sat Feb 16 09:48:45 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 16 Feb 2002 15:48:45 +0100
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <wzadu9scs2.fsf@nono.cs.uu.nl>

>>>>> rmunn at pobox.com (Robin Munn) (RM) writes:

RM> Reading the first few chapters of _How to Think Like a Computer
RM> Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/ for those who haven't
RM> heard of this book), I was struck by how Python resembles BASIC in some
RM> ways that make it a very good language for teaching:

BASIC a very good language for teaching????
Well, it depends on what you want to teach, I guess.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From martin at v.loewis.de  Wed Feb 27 15:26:38 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 21:26:38 +0100
Subject: kill childs
References: <a5j1p9$ftb$1@news.rz.uni-karlsruhe.de>
Message-ID: <m3sn7misch.fsf@mira.informatik.hu-berlin.de>

"Christian Schmidt" <christian.schmidt at REMOVEaifb.uni-karlsruhe.de> writes:

> I use Python to start other commands via os.system("cmdline")
> When I kill the script with 'kill <scriptpid>' the command is not killed,
> only the script.
> How can I kill the child, too?
> Can I use something like os.kill(-1) in an except-statement?

If you use os.system, you cannot find out what the pid of the child
is. So as a prerequisite, you need to replace the call to system()
with a fork/exec sequence. Then, you can install a signal handler
(using signal.signal) to kill the subprocesses if any.

Regards,
Martin



From skip at pobox.com  Thu Feb 14 18:00:25 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 14 Feb 2002 17:00:25 -0600
Subject: hex to int
In-Reply-To: <HFEKILOLEFEFMKAECNDLEEKFDCAA.jason@jorendorff.com>
References: <3C6C1E76.8AE06332@ix.netcom.com>
        <HFEKILOLEFEFMKAECNDLEEKFDCAA.jason@jorendorff.com>
Message-ID: <15468.16777.137805.376338@12-248-41-177.client.attbi.com>

    >>> int('0xff', 16)
    255

    >> In what version of Python do they introduce that?  

    Jason> Waaaaay back in Python 1.6.

Just for the record, if you're still stuck with 1.5.2, string.atoi supports
the same interface:

    >>> string.atoi("0xff", 16)
    255

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From markabarclay at mediaone.net  Mon Feb 25 14:16:02 2002
From: markabarclay at mediaone.net (Mark Barclay)
Date: Mon, 25 Feb 2002 14:16:02 -0500
Subject: running a COM server in Python
Message-ID: <3C7A8D72.9089BC7A@mediaone.net>

I have succeded in getting a COM server implemented in Python, so that
it can be instantiated by a COM client., thanks to ActivePython's
excellent groundwork.

I'm wondering if it is also possible to create an application in Python
that acts as a COM server but can be already running before teh COM
client attempts to attach to it. This is how Excel behaves. If Excel is
already running when a COM client attempts to attach to it. This is the
bahaviour I want.

Can anyone give me a pointer towasr how to get similar functionality?




From mscheid at iam.unibe.ch  Fri Feb 15 07:27:24 2002
From: mscheid at iam.unibe.ch (Matthias Scheidegger)
Date: Fri, 15 Feb 2002 13:27:24 +0100
Subject: module loading/unloading mechanism
Message-ID: <3C6CFEAC.15F9C17F@iam.unibe.ch>

Hi,

I'm need some way to do module loading/unloading similar to the way it's
done in the Linux kernel.
A module should be loadable from a string or marshalled object (ie. have
its own namespace and execute some initialization code) and be
unloadable (ie. remove all its namespace and execute cleanup code).
An important requirement is that the interpreter can't be restarted.
Module loading/unloading should be the only way to add/remove
"programs". The interpreter is embedded in a C program.
Afaik frozen modules could do the loading pretty smoothly. Any ideas
about unloading?

cheers

Matthias



From sorr at rightnow.com  Fri Feb 15 14:07:06 2002
From: sorr at rightnow.com (Orr, Steve)
Date: Fri, 15 Feb 2002 12:07:06 -0700
Subject: Java vs Python ... ???
Message-ID: <D408B6AC060DB24790622E1A54204DB00823AAA2@xchange.rightnowtech.com>

Thanks but could you make your answers a little more brief. :-)

Eschew obfuscation when your loquacity is exceeded only by your verbosity.
Steve 

-----Original Message-----
From: James_Althoff at i2.com [mailto:James_Althoff at i2.com]
Sent: Friday, February 15, 2002 11:49 AM
To: python-list at python.org
Subject: RE: Java vs Python ... ???

<Steve Orr>
For instance, with the Oracle Application Server (not the database) you
supposedly can have all your Java apps load balanced "for free" without
having to write load balancing code. So I was wondering the other day,
since
Jython compiles to bytecode, will it run under applications servers (such
as
Oracle) and enjoy the same Java features those servers provide?
</Steve Orr>

Yes.

Jim


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



From emile at fenx.com  Fri Feb  1 14:58:34 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 1 Feb 2002 11:58:34 -0800
Subject: telnetlib read_until()( question
References: <u5lqt76m0ogfd8@corp.supernews.com> <a3er56$17h5lp$1@ID-11957.news.dfncis.de> <u5lsfu2m34gpa9@corp.supernews.com>
Message-ID: <a3esav$17p2n5$1@ID-11957.news.dfncis.de>

"DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
news:u5lsfu2m34gpa9 at corp.supernews.com...
> Thanks Emile
> I am trying to find out why a telnet connection (using a Python
script)
> dropped out.


Would you by chance be trying to connect out to a system behind a
firewall?  Perhaps a non-activity timer is dropping you while you wait
for processing to complete on the remote host.

--

Emile van Sebille
emile at fenx.com

---------



From erno-news at erno.iki.fi  Mon Feb  4 05:43:55 2002
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 04 Feb 2002 12:43:55 +0200
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org>
Message-ID: <ku665d4jdg.fsf@lasipalatsi.fi>

In article <a3knt6$gr3$1 at news.efn.org>, "Spencer Doidge"
<spencer at spencerdoidge.com> writes:

| I really tried to RTM first this time, but I failed.
| Can someone give me a hint where in the doc html's it explains how to pass
| by reference?
| Specifically, I want to do the equivalent of this:

| char s[10];

| void foo(char *bfr)
| {
|     bfr[0] = 'a';
|     bfr[1] = 'b';
|     return;
| }

| void main(void)
| {
|     foo(s);
|     return;
| }

s = [''] * 10

def foo(bfr):
    bfr[0] = 'a'
    bfr[1] = 'b'

def main():
    foo(s)
    return

  -- erno




From roy at panix.com  Sun Feb  3 20:32:30 2002
From: roy at panix.com (Roy Smith)
Date: Sun, 03 Feb 2002 20:32:30 -0500
Subject: Finding all of a classes methods?
Message-ID: <roy-CA3285.20323003022002@news1.panix.com>

Is there any way to find all of a class's methods?  I know I can use dir(), 
but that gives you the names of the methods (as strings), not the methods 
themselves (as callable MethodType objects).

I'm building a production system.  Basicly, I want to run every one of a 
class's methods on an object.  The number of methods is large, and will 
grow in the future.  Right now, I'm building a list of the methods as I 
define them.  Something like:

methods = []

def foo (self, x):
   pass
methods.append (foo)

def bar (self, x):
   pass
methods.append (bar)

and so on down the list (yes, the real ones do something more useful than 
pass).  Obviously, the class is keeping track of its methods already.  If I 
could get at that list, I could avoid having to build my own.

Also, I'm slightly mystified about what, exactly, I'm adding to my list.  
When I do:

for method in self.methods:
   method (x)

I get an error saying that the method expected 2 arguments, but only got 1.  
Obviously, it's not getting self because I'm not calling it as a method of 
self.  If I change it to:

for method in self.methods:
   method (self, x)

I get what I want, but I'm not 100% sure that's kosher.


From cfelling at iae.nl  Sun Feb 24 12:00:49 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 24 Feb 2002 18:00:49 +0100
Subject: is their any limit on the size of the data passed to a progra   m via popen2?
References: <mailman.1014401619.20291.python-list@python.org> <3C76CAB9.50509@student.kun.nl> <a56o39$7dg$1@animus.fel.iae.nl> <3C7761D4.7050100@student.kun.nl> <a58k1m$8t1$1@animus.fel.iae.nl> <a58kjb$927$1@animus.fel.iae.nl> <3C78B231.1010805@student.kun.nl>
Message-ID: <a5b681$clq$1@animus.fel.iae.nl>

husam <h.jehadalwan at student.kun.nl> wrote:
...
> hee, NEDERLANDER, leuk je te ontmoeten hier. Jammer dat er geen 
> Nederlandse newsgroup is voor python.

Ah, maar dit *is* de Nederlandse newsgroep voor Python.
Zo ongeveer iedereen hier beheerst Nederlands alsof het zijn
moers taal is, per slot komt Python ook uit Nederland:)


-- 
groetjes, carel


From jason at jorendorff.com  Sat Feb 16 09:40:24 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 08:40:24 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <E16bzNd-0006sf-00@tswings.swing.cluster>
Message-ID: <HFEKILOLEFEFMKAECNDLOEAFDDAA.jason@jorendorff.com>

"Jason Orendorff" <jason at jorendorff.com> wrote:
> >   2.  eval()
> >       Well-known gaping security hole.
> >
> > It seems to me that security is a compelling reason to choose
> > pickle() over eval().  Am I wrong?
> 
> With regard to eval, yes. Check out the two optional arguments to
> eval, before spreading FUD.

Well, the bottom line is:  If I'm looking for a vulnerability in
an app, I'm searching for calls to eval(), not pickle.

For example, the idiom you suggested elsewhere in this thread:
  >>> eval ("""os.system ("echo 'kilroy'")""", {}, {})

is actually exploitable:
  >>> s = "__import__('os').system('echo kilroy')"
  >>> eval (s, {}, {})
  kilroy
  0

If you want to know how to do it correctly, consult the pickle
source code <wink>.  But even then, there's stuff like
"100L**100**100**100".

Whereas there are no known security holes in pickle.

## Jason Orendorff    http://www.jorendorff.com/


P.S. Note that you *can* make pickle throw a SyntaxError:

  >>> y = "S'foo'\001\np1\n."
  >>> cPickle.loads(y)
  SyntaxError: invalid syntax
  >>> pickle.loads(y)
  SyntaxError: unexpected EOF while parsing

This is a bug.

P.P.S.  Separate idea: the use of eval() in pickle could
be replaced by a call to compile(), arguably safer but more
brittle in terms of maintenance:

  >>> s = 'foo\nbar'
  >>> co = compile(repr(s), '<pickle>', 'eval')
  >>> print co.co_consts[0]
  foo
  bar

Probably not worth it.




From emile at fenx.com  Tue Feb 26 02:11:17 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 23:11:17 -0800
Subject: ask help about raise, thank u
References: <mailman.1014686626.31452.python-list@python.org>
Message-ID: <a5fcom$70pjd$1@ID-11957.news.dfncis.de>

"dongkui" dongkui at huawei.com
> pls give me one example about raise which have three para, and the >
third para is traceback obj
>
> like
> raise "a", "b", c #c as traceback object
>

There appears to occur six times in the 2.2 standard library.  Four
times in urllib.py, and once each in mhlib.py and urllib2.py

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From cfelling at iae.nl  Wed Feb 13 20:01:34 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 14 Feb 2002 02:01:34 +0100
Subject: How to capture output of os.spawnv()?
References: <6JAa8.34711$Wf1.6786156@ruti.visi.com> <1013641215.358571@yasure>
Message-ID: <a4f29e$bhm$1@animus.fel.iae.nl>

Donn Cave <donn at drizzle.com> wrote:
...
> Check it out.  Not popen, but as they say, "popen2 et al."
> It will exec the command on its own, no shell, if you give it
> an argument list.  If you give it a string, it will use the
> shell.

How is one supposed to know this?  (excluding reading the source or
remembering from the last time you brought this up:)

I've reread the fine manuals, but can't find a clue hinting at this
interesting behaviour.
-- 
groetjes, carel


From timr at probo.com  Sun Feb 17 01:37:59 2002
From: timr at probo.com (Tim Roberts)
Date: Sat, 16 Feb 2002 22:37:59 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu>
Message-ID: <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com>

Fernando P?rez <fperez528 at yahoo.com> wrote:

>I.J. wrote:
>
>> Is there a module for python that grants ussage of decimal numbers with more
>> than 14 decimal places, including all scientific functions to do with those
>> numbers?
>
>No. Python uses C doubles underneath, and those give you only 14 places. 
>...
>What you're asking for isn't particularly easy to find outside of systems 
>like Mathematica which implement their own arithmetic subsystem, ...

And the original poster should understand that this is true because such
systems are of little practical use.  There are very, very few physical
processes where floating point is useful AND precision of more than 6
significant digits is really required.  Large integers can be useful (and
Python supports them), but a requirement for high precision floats is
usually a sign that the requirer does not understand his problem space.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From sdm7g at virginia.edu  Mon Feb 25 16:19:21 2002
From: sdm7g at virginia.edu (Steven Majewski)
Date: Mon, 25 Feb 2002 16:19:21 -0500 (EST)
Subject: subclassing float (or not)
In-Reply-To: <XFMail.20020222123018.shalehperry@attbi.com>
Message-ID: <Pine.OSX.4.40.0202251510050.509-100000@d-128-54-224.bootp.virginia.edu>


On Fri, 22 Feb 2002, Sean 'Shaleh' Perry wrote:

>
> On 22-Feb-2002 Thomas Heller wrote:
> > I'm currently porting my own rewrite of Konrad Hinsen's PhysicalValue
> > module to Python 2.2.
> > Basically a PhysicalValue instance is a float conbined with a unit.
> > Should I subclass float or should I not? What would I gain by doing
> > this (except maybe some memory savings when I use __slots__)?
> >
>
> Generally when you subclass you are saying "I am a special type of foo" where
> foo is your parent.  I think in this instance it would make more sense to make
> a new class which contains a float and a unit.  Give it methods so it acts how
> you expect.
>

Except that how he expects it to act is exactly like a float except for
carrying and propagating a little extra bit of state at each operation.

Sounds like a good application for subclassing floats to me!

You can do it (and could have always done it) with a container class,
but there are always holes where you need the operation to be
coerced on the classes value and not the class itself. In some
cases that still may be what you really want (because you also want
to coerce the result to my_float rather than float), but at least
if you miss something, a subclassed float will do something closer
to the right thing.

An example of what I mean is 'sqrt(MyFloat)' -- if MyFloat is a
class that contains a float and has methods for __add__, __mul__,etc.
sqrt(MyFloat) will raise an exception if passed a container class --
you need to call 'sqrt(MyFloat.value)' , while:

>>> class F(float):
...     pass
...
>>> from math import *
>>> sqrt( F(1.0))
1.0

Will return a normal float where you may really want a custom F instance.
( Which is the better way to go wrong here may vary. )

Two other float-like classes that I've had use for that are easier to
do now with float subclasses:

Values that carry and propagate standard errors or standard deviations
with them.

Values that carry an "audit trail" ( for taking chained ratios of
experimental measures and trying to track down the inconsistent
data. )


-- Steve Majewski






From jason at jorendorff.com  Mon Feb 25 02:04:51 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 01:04:51 -0600
Subject: Newbie Q: Import code at run-time?
In-Reply-To: <3C79DCFB.D216144A@ifm-services.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEODDEAA.jason@jorendorff.com>

James Knowles wrote:
> Now I want to generalize the program so that I can pass in
> the name of the Python module with the dictionary definitions
> along with the database type, e.g.
> 
> $ python builddb dbA mysql >dbA-mysql.sql
> $ python builddb dbB ib >dbB-ib.sql

There are a few ways.  Easiest:

import sys
module_name = sys.argv[2]
__import__(module_name)   # Note *double* underscores here
my_module = sys.modules[module_name]

If you don't like that one, check out execfile().

## Jason Orendorff    http://www.jorendorff.com/



From martin at v.loewis.de  Tue Feb 12 19:56:12 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 01:56:12 +0100
Subject: Extracting bzip2 & gzip files
References: <mailman.1013534839.29210.python-list@python.org>
Message-ID: <m3d6zaw66r.fsf@mira.informatik.hu-berlin.de>

"Jaco Greeff" <jaco at puxedo.org> writes:

> Any ideas? Help on this?

It works fine for me, just the way you posted it (invoking the bzcat
version directly, for some compressed file). Perhaps the file you are
trying to uncompress is really corrupted?

Regards,
Martin



From markabarclay at attbi.com  Tue Feb 19 07:51:10 2002
From: markabarclay at attbi.com (Mark Barclay)
Date: Tue, 19 Feb 2002 12:51:10 GMT
Subject: execfile question
References: <mailman.1014117330.24204.python-list@python.org>
Message-ID: <3C724B05.E9167812@attbi.com>

"del c" will cause your __del__() to be called.
So will any other action that causes your instance to be garbage-collected.

For example, try:

>>> class D:
...     def __init__(self):
...         self.myC = C()
...
>>> d = D()
init
>>> del d
deleted
>>>


Kerim Borchaev wrote:

> Can someone explain why while running this script I don't see C.__del__
> eventually called?
>
> For me the output of it's execution is:
>
> >main.py
> it's me
> init
> executed
> >
>
> ##main.py######################
> execfile('executable.py', {})
> print 'executed'
> ########################
>
> ##executable.py######################
> print "it's me"
> class C:
>     def __init__(self):
>         print 'init'
>     def __del__(self):
>         print 'deleted'
>
> c = C()
> ########################
>
> Thanks in advance.

--
Mark Barclay (markabarclay at attbi.com)




From martin.franklin at westerngeco.com  Wed Feb 20 03:59:58 2002
From: martin.franklin at westerngeco.com (Martin Franklin)
Date: Wed, 20 Feb 2002 08:59:58 +0000
Subject: MIDI
References: <mailman.1014155972.28486.python-list@python.org>
Message-ID: <a4vogp$vf2$1@mail1.wg.waii.com>

Mariano Corsunsky wrote:

> Does someone knows where can I find a module to play midi files from a
> python script?
> 
> 
> thanks!
> 
> 
http://www.vex.net/parnassus/apyllo.py?find=MIDI




From gerson.kurz at t-online.de  Mon Feb 25 12:36:11 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 17:36:11 GMT
Subject: Trouble with getting remote Computer Name
References: <d7d5ebdf.0202251024.236b2c1b@posting.google.com>
Message-ID: <3c7a75ea.16613828@news.t-online.de>

On 25 Feb 2002 10:24:02 -0800, matthias_j at lycos.com (Matthias Janes)
wrote:

>My Problem:
>
>I have the IP address of all the Servers in a VPN for instance
>(10.251.23.6,10.251.24.6, 10.251.25.6 and so on)
>
>How can I get from each Server the Name of the Server in python.

Example:

>>> socket.gethostbyname("www.google.com")
'216.239.37.101'
>>> socket.gethostbyaddr('216.239.37.101')[0]
'www.google.com'



From mhammond at skippinet.com.au  Sat Feb 23 20:27:21 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 24 Feb 2002 01:27:21 GMT
Subject: problems with pythoncom and VB-generated OLE/ActiveX DLL
References: <e34e8ce6.0202231659.609787bd@posting.google.com>
Message-ID: <3C7841F6.8040206@skippinet.com.au>

Domenico wrote:


>>>>import display
>>>>d = display.Class1()
>>>>
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "C:\compilers\python\display.py", line 40, in __init__
>     if oobj is None: oobj = pythoncom.new(self.CLSID)
> pywintypes.com_error: (-2147221164, 'Interfaccia non registrata.',
> None, None)
> 
> 'Interfaccia non registrata' (in Italian) means something like
> 'Unregistered interface'. So I did:

Assuming self.CLSID is the CLSID of the *object*, not the *DLL*, this 
should work.

Have you run regsvr32.exe on the DLL?  Can you create the object from 
another VB program, or using WSH?

> 
> 
>>>>import pythoncom
>>>>d = pythoncom.new("{***CLSID of the DLL***}")
>>>>

The CLSID of the DLL will never work - it must be the object's CLSID.

I assume you have a good reason for using pythoncom.new() over 
win32com.client.Dispatch()?

Mark.



From eric.brunel at pragmadev.com  Mon Feb 18 05:07:32 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 18 Feb 2002 10:07:32 +0000
Subject: Can Tk be connect with Copyboard of Linux
References: <a4jlhf$6uv$1@pegasus.csx.cam.ac.uk>
Message-ID: <a4qg1f$2pq3$1@norfair.nerim.net>

<posted & mailed>

> Hi, everybody,

Hi,

> Under windows, the copyboard is shared by all the applications. So I can
> copy some messages from different applications, and paste them into a Text
> or Entry widgit based on Tkinter/Tk. But I found the same operation cannot
> work under linux. Under KDE, I can not paste messages from other
> applications into a widget based on Tkinter/Tk. Is it possible to solve
> this problem?

I tried to solve the same problem and I soon realized that talking about 
_the_ clipboard with X11 was quite meaningless: there may be as many 
clipboards as there are applications. And that seems to be the problem: 
several applications use the selection named "CLIPBOARD" to store their 
copied data, so the Windows way of doing things should also work on Linux. 
But many applications do not use this selection at all, and I couldn't 
figure out a way of finding what was the name of the selection they use. So 
if you copy things from these, you won't be able to paste them in your 
application, but because of them, and not because of Tk or Tkinter...

For example, I've got two editors installed on my Linux box, namely GEdit 
and NEdit. GEdit uses the "regular" CLIPBOARD selection, so copy/paste 
between GEdit and Tk works. But NEdit doesn't use this selection, so 
copy/paste doesn't work...

There is also a way to know if an application uses the CLIPBOARD selection 
or not: with every X11 server, a small tool named xclipboard should be 
installed. This tool shows the contents of the CLIPBOARD selection. So you 
should just run xclipboard, open the application, then select something and 
copy it. If it appears in xclipboard, you will be able to paste it back in 
your Tk application. If it doesn't, the application doesn't use the regular 
CLIPBOARD selection, and you won't be able to get it.

Note however that you can always use the PRIMARY selection that Cameron 
talked about, and that there's probably no bindings to do: just copy by 
selecting with the left button, and paste by using the middle button. That 
should work on any Unix/X11 box. And that seems to be the standard way of 
doing things, more than the CLIPBOARD / Ctrl-C / Ctrl-V stuff, which is 
more Windows-like.

HTH
 - eric -



From bkd at graphnet.com  Fri Feb  8 10:05:03 2002
From: bkd at graphnet.com (Bruce Dykes)
Date: Fri, 8 Feb 2002 10:05:03 -0500
Subject: [NEWBIE] ftplib again...
References: <mailman.1013094492.31504.python-list@python.org> <sJC88.96852$KM2.3618183@atlpnn01.usenetserver.com>
Message-ID: <020301c1b0b1$fff74660$5102020a@graphnet.com>

----- Original Message -----
From: "Steve Holden" <sholden at holdenweb.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, February 07, 2002 16:55
Subject: Re: [NEWBIE] ftplib again...


> > All I need to do is get the file contents into a list. That's it. Now it
> > looks as if retrlines("RETR filename") will do that, with some
additional
> > manipulations.
> >
> > But it doesn't. Every single variation I've tried,
> > dataset=[host.retrlines("RETR filename") ],  host.retrlines("RETR
> filename",
> > dataset.append) , dataset=list(host.retrlines("RETR filename") ), etc.
> > always results in the file contents getting dumped to the
> > console/stdout/interactive window, and the completion code, '226 Command
> > Completed Successfully' getting assigned to the variable.
> >
> > What am I missing? Do you need to see more of my code?
> >
> """
> retrlines (command[, callback])
> Retrieve a file or directory listing in ASCII transfer mode. command
should
> be an appropriate "RETR" command (see retrbinary() or a "LIST" command
> (usually just the string 'LIST'). The callback function is called for each
> line, with the trailing CRLF stripped. The default callback prints the
line
> to sys.stdout.
> """
> This doesn't really make it that obvious, but what you have to do is
provide
> a callback, which is called, with the line as an argument, each time a
line
> is retrieved from the server. The code you need is something like this:
>
> linelist = []
> host.retrlines("RETR filename", linelist.append)

Thanks loads Steve...that was perfect. I coulda sworn I tried that
particular construct after seeing it in some sample code (ftpmirror.py, I
think) and it didn't work then...but hey...it works...8-)

Thanks
bkd




From cfelling at iae.nl  Tue Feb 26 20:25:05 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 27 Feb 2002 02:25:05 +0100
Subject: Python 2.2 Socket Test Failure
References: <Q%Ve8.8688$Dv1.29388@rwcrnsc53>
Message-ID: <a5hchh$jod$1@animus.fel.iae.nl>

Michael Strayer <mstrayer1 at attbi.com> wrote:
...
> I've built Python 2.2 on a Red Hat 7.2 system and when
> I run make test I get the following error:

> test test_socket crashed -- socket.gaierror: (-2, 'Name or service not 
> known')
> test_socket_ssl
> test test_socket_ssl skipped -- Use of the `network' resource not enabled
> test_socketserver
> test test_socketserver skipped -- Use of the `network' resource not enabled

This looks familiar:)  I had to build me a more modern ssl suite:

    libssl0.9.6_0.9.6b-4_i386.deb
    libssl-dev_0.9.6b-4_i386.deb
    openssl_0.9.6b-4_i386.deb

on Debian potato.  You are using Red Hat, so try to find rpm's that work
on your version of Red Hat.
-- 
groetjes, carel


From andtherewaswater at yahoo.com  Thu Feb 14 01:32:52 2002
From: andtherewaswater at yahoo.com (A.A)
Date: 13 Feb 2002 22:32:52 -0800
Subject: Bug ? xml.dom.minidom
References: <3C63D2F3.79EB997F@yahoo.com> <m31yfsf7rb.fsf@mira.informatik.hu-berlin.de>
Message-ID: <45595e8e.0202132232.395085c4@posting.google.com>

martin at v.loewis.de (Martin v. Loewis) wrote in message news:<m31yfsf7rb.fsf at mira.informatik.hu-berlin.de>...
> "A.A" <andtherewaswater at yahoo.com> writes:
> 
> > What is happening ???
> 
> I can't reproduce this. If you can provide more details, please submit
> a bug report to SourceForge. Any chance you have PyXML installed? What
> version?
> 
> Regards,
> Martin

Sigh.

Apologies. It was a false alarm. It seems that in my home directory
there was xml.py file, and unwittingly, I was importing THAT file.

Regards,
A.A.


From niemeyer at conectiva.com  Mon Feb 18 13:58:34 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Mon, 18 Feb 2002 15:58:34 -0300
Subject: super(obj)?
In-Reply-To: <h_9c8.69760$om6.1918534@news1.tin.it>
References: <mailman.1014045511.10659.python-list@python.org> <h_9c8.69760$om6.1918534@news1.tin.it>
Message-ID: <20020218155834.A4044@ibook.distro.conectiva>

Hi Alex!

[...]
> I guess I must be missing something, but really can't
> see what.  Explain please?  Thanks!

I'm the one who was missing something. What I explained works
perfectly for one level hierarchies, but will fail miserably
once you create a class inheriting the bogus class.

Thanks!

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From buzzard at urubu.freeserve.co.uk  Mon Feb 18 21:03:34 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Tue, 19 Feb 2002 02:03:34 -0000
Subject: Refresh()
Message-ID: <a4sbud$n8t$1@news7.svr.pol.co.uk>

I am having problems with the following wxPython code.  I simply want to
change the background colour of 'panel1' and then (re)draw to 'panel1'.  For
some reason it does not redraw (but I can redraw by calling the
OnGraphviewMove(event) function again, eg. by moving the window).  Maybe I
just need a decent explanation of exactly what Redraw() does?  Can anyone
please help?  Thanks in advance.  (Win2000, Python 2.1, wxPython 2.2.?.)

Duncan Smith


    def OnToolbar1tools4Tool(self, event):
        data = wxColourData()
        data.SetChooseFull(true)
        dlg = wxColourDialog(self, data)
        try:
            if dlg.ShowModal() == wxID_OK:
                data = dlg.GetColourData()
                self.panel1.SetBackgroundColour(data.GetColour())
                self.panel1.Refresh()                   #change the
background
                self.OnGraphviewMove(event)    #(re)draw the graphic
        finally:
            dlg.Destroy()





From bkelley at wi.mit.edu  Fri Feb 15 13:25:32 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 15 Feb 2002 13:25:32 -0500
Subject: Problems with Tkinter based scripts under emacs...
References: <wk4rkj583y.fsf@dof.se>
Message-ID: <3C6D529C.4070206@wi.mit.edu>

Rod Stephenson wrote:

> I'm having problems with emacs on win95 when running a python script
> that invokes Tkinter. I'm using the bash shell under emacs, and
> python-mode, so that I can start up an interpreter which appears in a
> comint buffer and issue commands. Theres no problem with regular
> python scripts, and its a nice environment for edit-run-debugging.
> 
> I've just started using Tkinter and I find the following: The script
> runs fine, but when I quit the application, the shell just hangs and I
> don't get the >>> prompt back. I end up having to kill the process and
> restart the interpreter.
> 
> Actually I'm not sure whether this is specifically an emacs issue. 


I think this is more likely a Windows issue.  Emacs also doesn't seem to 
like to run wxPython in shell mode either.  My guess is that on windows, 
the standard gui programs fiddle with the stdin/stdout and stderr pipes 
which gets emacs fairly confused.

This problem doesn't seem to occur on unix using the same toolkits and 
emacs.

It's not free but the WING IDE seems to fix some of these problems under 
windows, take a look at the feature list here 
http://archaeopteryx.com/wingide/features

> Everything works just fine in the DOS command line shell, but this is
> so annoying to use! Any hints about setting up emacs or idle to cope
> with Tkinter would be much appreciated.


The solution I use for debugging with Tkinter type ctrl-C ! in the 
source window to open up a shell and then type ctrl-C ctrl-C to run the 
source.  This allows print statements and exceptions to be viewable at 
run time.  The caveat is that the shell hangs when I leave Tk so the 
only benefit is the real time print statements.


Brian Kelley




From zvonkoka at yahoo.com  Sat Feb 23 12:38:36 2002
From: zvonkoka at yahoo.com (zvonko)
Date: Sat, 23 Feb 2002 18:38:36 +0100
Subject: cannot start pythonwin
Message-ID: <a58k0o$31ur$1@as201.hinet.hr>

i've just installed python 2.2 and the pythonwin extensions.
however, when i try to start pythonwin, i get the following error:


The application cannot locate win32ui.pyd (1157)

but the file is in the same directory where the pythonwin is.

HELP!





From justin at iago.org  Wed Feb 13 12:00:06 2002
From: justin at iago.org (Justin Sheehy)
Date: Wed, 13 Feb 2002 12:00:06 -0500
Subject: Problem with socket
In-Reply-To: <3c6a7e08$0$12221$4d4efb8e@news.be.uu.net> ("DelPiccolo
 Ivano"'s message of "Wed, 13 Feb 2002 15:54:25 +0100")
References: <3c6a7e08$0$12221$4d4efb8e@news.be.uu.net>
Message-ID: <x7zo2ds4fd.fsf@ra.iago.org>

"DelPiccolo Ivano" <ivano.delpiccolo at degroof.be> writes:

> I tried to impement a very simple client-server application.
> Hre's the output from the server part :
>
> Traceback (most recent call last):
>   File "C:\Python22\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "D:\server.py", line 10, in ?
>     data = conn.recv(1024)
>   File "<string>", line 1, in recv
>   File "C:\Python22\lib\socket.py", line 159, in __getattr__
>     raise error(9, 'Bad file descriptor')
> error: (9, 'Bad file descriptor')
>
>
> Here's my server.py :
> import socket
> host = ''
> port = 50007
> s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> s.bind ((host,port))
> s.listen(1)
> conn,addr=s.accept()
> print 'Connected by : ', addr
> while 1:
>     data = conn.recv(10)
>     if not data : break
>     conn.send(data)
>     conn.close()

The contents of the while loop is almost certainly not what you want.

The first recv will generally return some data.  You then proceed to
send data back, and then _close the socket_. After that, the while
loop starts over and you try to read again... from the socket you just
closed.

If you just change the loop from:

while 1: 
    data = conn.recv(10) 
    if not data : break 
    conn.send(data) 
    conn.close() 

to this:

while 1:    
    data = conn.recv(10)
    if not data : break
    conn.send(data)
conn.close()

you should be all set with regard to this example.

-Justin

 




From hfoffani at yahoo.com  Sat Feb  2 13:46:38 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Sat, 2 Feb 2002 19:46:38 +0100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com>
Message-ID: <20020202134649.468$Gg@news.newsreader.com>

> > I fully agree that when a manager evaluates techs alternatives the
> > decision may be affected by non-tech factors. I don't think many
> > readers here would deny such claim and c.l.py is a great forum to
> > look for answers on TECH's issues of Python.
>
> Wait just a moment here ... what are you saying here? c.l.py is a
> great place to tooks for tech's issues But we don't consider
managment
> issues ...
> Well in the ***real*** world where i live all descission are taken
by
> managers !

I meant to say managment issues that doesn't have to do or relate to
Python itself but, say, to business or customer policies.
For example, if your CEO states: "We are an Oracle partner, so
everything in PL/SQL" how much is left to say?

Of course, I accept that turning those cases in Python succesful
stories
won't be offtopic.

What I do saw in c.l.py from time to time and annoys me a lot is
confusing techs issues with business ones and perhaps I was being a
unfair with you.

Worst, may be is just that *I* don't want to read hundreds of posts
on proj-mgmnt here... Heh... So that's MY problem, aint so? :-(

Regards,
-Hernan







From fperez528 at yahoo.com  Sun Feb 10 11:11:35 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 10 Feb 2002 16:11:35 +0000
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <eppstein-A87A6C.14430710022002@news.service.uci.edu>
Message-ID: <a46uqm$gc7$1@peabody.colorado.edu>

David Eppstein wrote:

>> Pythonista - Pie thun ee sta
>> I took it as a unintended play on Sandanista, which is a group of  "freedom
>> fighting" rebels in Nicaragua.
> 
> I think it's a normal Spanish suffix, but it seems to be moving into
> English -- I've also seen "Gapatista" used by some Calif. activists
> <http://www.ics.uci.edu/~eppstein/pix/sv/j4p/Gapatista.html>

Indeed, it's just the Spanish form of the 'ist' English suffix (pianist, 
violinist, etc.) It simply means 'who performs an activity'.

Its use is more common in Spanish than in English, though (pianista, 
violinista, but also futbolista, tenista, equilibrista, ...). Interestingly, 
even though in Spanish the -a ending typically denotes a feminine noun, the 
-ista ending applies to both male and female practitioners.

And yes, I'm a native Spanish speaker, so I should know.

Cheers,

f.



From wolfson at uchicago.edu  Sun Feb 17 21:37:04 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Sun, 17 Feb 2002 20:37:04 -0600
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1> <8867.813T2365T9245607threeseas@earthlink.net> <slrna70n1f.som.mlh@vier.idi.ntnu.no> <3C706035.6AC11606@earthlink.net>
Message-ID: <pan.2002.02.17.20.37.03.908190.9995@uchicago.edu>

On Sun, 17 Feb 2002 19:59:34 -0600, Hans Nowak wrote:

> Do you also have a model with zero actions? This one seems kind of
> difficult to me.

I do.  Here it is:





Refute THAT.

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From loewis at informatik.hu-berlin.de  Sun Feb  3 09:35:37 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Feb 2002 15:35:37 +0100
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org>
Message-ID: <j47kpuhbuu.fsf@informatik.hu-berlin.de>

Bas van Gils <bas.vangils at home.nl> writes:

> I've been wondering, lately, about lambda-expressions. Most programming
> languages support and I wondered why (mostly because I find them hard to
> read, valueing _readable_ code). Collegues at work couldn't give me a
> satisfying answer so ... can anyone explain to me why lambda-expressions
> are so useful?

I think they got introduced because some people found them convenient,
especially those coming from a functional background. To them, it is a
religious matter: If you have map and filter, you also ought to have
lamba.

Later, they found that, even though they got what they asked for, they
did not get what they wanted: inside a lambda expression, you can only
use expressions, but they also wanted anonymous code blocks.

It is unlikely that they will ever get that, since nested functions
can now refer to local variables conveniently.

Regards,
Martin


From Moo at houston.rr.com  Wed Feb 27 01:25:30 2002
From: Moo at houston.rr.com (Avi Homes)
Date: Wed, 27 Feb 2002 06:25:30 GMT
Subject: Python GUI
Message-ID: <uZ_e8.24022$dj3.835926@typhoon.austin.rr.com>

New to python (still.. I made a post last week with my ridiculous error
about using () for indexing in a dictinoary instead of [], ah well, its the
easy mistakes that get us. anyway) I was curious if there were some type of
gui builder or somesuch for python, alla Borland Builder ? This is a
delightful language,everything seems so easy for it. as a friend pointed out
"yeah ,but you can do that in c++ as well.." and i replied "yeah, but its
not this easy! =)"
thanks guys

~Avi




From jgardn at alumni.washington.edu  Mon Feb  4 11:51:56 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Tue, 5 Feb 2002 01:51:56 +0900
Subject: Stop feeding the trolls! Kill the autocoding thread!
In-Reply-To: <200202041602.g14G2Oi28655@my.knctv.co.kr>
References: <83bsfkk31e.fsf@panacea.canonical.org> <3c5e9cb9.17829734@news.t-online.de> <200202041602.g14G2Oi28655@my.knctv.co.kr>
Message-ID: <200202041657.g14GvNi29821@my.knctv.co.kr>

On Tuesday 05 February 2002 12:56 am, Jonathan Gardner wrote:
> A quote of your post is below. Go ahead and reply, but don't post a copy to
> the list!
>
Damn I feel stupid...

=(



From amuys at shortech.com.au  Sun Feb  3 21:06:46 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 3 Feb 2002 18:06:46 -0800
Subject: Please comment  on Draft PEP for Enhanced Generators
References: <a38c1u$945$1@bob.news.rcn.net> <a3976j$qkt$1@serv1.iunet.it> <a39cpr$ic1$1@bob.news.rcn.net>
Message-ID: <7934d084.0202031806.4e7dff9a@posting.google.com>

"Raymond Hettinger" <othello at javanet.com> wrote in message news:<a39cpr$ic1$1 at bob.news.rcn.net>...
> "Alex Martelli" <aleax at aleax.it> wrote in message
> news:a3976j$qkt$1 at serv1.iunet.it...
> > def xmap( fun, *sequences ):
> >     ...
> >  for args in xzip( *sequences ):
> >      yield fun(*args)
> >
> > has the wrong behavior when the sequences have different lengths.
> > http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/66448 has
> > an implementation which I believe satisfies the specs (complicated,
> > alas, but those specs ARE tricky;-).
<snip>
> I have one misgiving -- padding shorter sequences with None makes sense
> when all of the sequences are expected to be finite, but may not be the
> desired
> behavior when one of the sequences is potentially infinite. I would like to
> xmap( None, xrange(sys.maxint), iter(f.readline,'') ) to return a finite
> sequence
> and respond to list(mygen) without crashing in an infinite loop.

As unpadding is impossible, while padding is trivial, this suggests
the existence of an additional generator to pad where required.

def xpad(iterator, default=None):
  try:
    yield iterator.next()
  except StopIteration:
    yield default

which would then allow you to pad shorter sequences with None in those
cases where it is desired.

Andrae Muys


From sabren at manifestation.com  Tue Feb 19 13:06:18 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 19 Feb 2002 13:06:18 -0500 (EST)
Subject: Adding static typing to Python
In-Reply-To: <HFEKILOLEFEFMKAECNDLOELPDDAA.jason@jorendorff.com>
Message-ID: <Pine.LNX.4.44.0202191250110.13883-100000@mercury.sabren.com>

On Tue, 19 Feb 2002, Jason Orendorff wrote:

> class Frob(object):
>     def get_name(self):
>         return self.__name
> 
>     def set_name(self, name):
>         if re.match('^\w+$', name) is None:
>             raise ValueError("Frob.name may only contain name characters")
>         self.__name = name
> 
>     name = property(get_name, set_name, None,
>                     """ The Frob's canonical given name """)
> 
> Guido hints there may be some syntactic sugar for this
> in an upcoming version.


He created a metaclass called autoprop that does the
name=property(...)  assignment automatically..

   http://www.python.org/2.2/descrintro.html#metaclasses


I incorporated this into my strongbox package... It also
lets you define something like "slots" automatically:


   class Person(Strongbox):
       # static typing of slots:
       name = attr(str)
       
       # default values, lambda typing:
       age = attr(int, default=18, okay=lambda v: v > 0)

       # regexp typing:
       tel = attr(str, okay="[0123456890()-]+") 

       # enumerations:
       favorite_color = attr(str, okay=["red","green","blue"])
       
       # autoprop style access:
       def get_canDrink(self):
           return self.age > 21


Then:

>>> person = Person(name="fred")
>>> person.name
fred
>>> person.age = 5
>>> person.canDrink
0
>>> person.age = "55"  # automatic type casting!
>>> person.age
55
>>> person.age = "apple"
# (raises TypeError)
>>> person.age = -5
# (raises ValueError)
>>> person.newAttribute = 5  
# (raises attribute error)
>>> person.private.newAttribute = 5 # works fine.


The point of doing all this isn't to speed up the compiler
or anything. It's just to save me from having to write the
same kinds of validation code over and over again when
dealing with records that have to be saved to MySQL. The
automatic type casting is great when you're pulling data
from an HTML form...

I don't have the code online anywhere, but I can post it if
anyone's interested.

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From threeseas at earthlink.net  Fri Feb  1 23:36:25 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 02 Feb 2002 04:36:25 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <690.793T1095T13354397threeseas@earthlink.net> <un0yxkupb.fsf@synopsys.com> <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com>
Message-ID: <27327.798T2437T205791threeseas@earthlink.net>

On 31-Jan-02 13:24:19 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> Language is abstraction, created in order to communicate, but it can
>> never completely communicate reality.  A pictures says a thousand
>> words..... And that ain't language.

>Yes it is.  Language exists in order to communicate an abstraction of
>reality from one entity (usually people, but it could also be animals)
>to another.  That abstraction can be as simple as "hunger" to as
>complex as the theory of relativity.  The same is true with pictures
>since they are also an abstraction of reality and are used to
>communicate what something (might have) looked like at some time.
>There are many pictographic languages around.

>> This project would move forward faster if people would chose to
>> help.  And the sooner it get more complete the sooner more can be
>> explained by way of example.

>We *CANNOT* help if we cannot understand the abstraction of reality
>that this project represents.  You must find the proper language by
>which to communicate that abstraction.

>> #! Python

>I don't know Python (yet), but I think I get the idea.

>> # PK file array - VIC Dictionary
>> # Setting the standard variables to defaults

>For brevity, I'm going to cut out your code and simply comment on it.

>> vic = {

>* Your case statement doesn't provide any info on what the VIC is
>  supposed to do.  It appears to be merely creating an array.

>> # Test!!!

>* Your test is merely a printout of various array items -- again, no
>  info on what the VIC is supposed to do.

>> /*Reading in the PK file.

>* Okay, I assume you get some information from the PK file, but I see
>  nothing to tell me what it's supposed to contain.

>>     current_line = READLN('PK_file')
>>     PARSE VAR current_line vic.1 ':' vic.1.1 ';' vic.1.2 ';' vic.1.3

>* Whoops, there is some reading going on here and a basic format
>  (lines of "something" with 4 fields separated by ":" and ";").

>> /* Writing the PK file */

>* Writing?  We haven't done anything with what we read in yet?

>> CLOSE('PK-file-out')

>* What was achieved in this program?  Did you merely rearrange the
>  data in the PK file?  If so, why?



Ok, What you saw wasn't the all that I had, the following is, (though not
as much of it as the last post, has been converted to python)..

There is code that has been converted to python, code that is still in
arexx form needing to be converted to python and more that hasn't been put
into either arexx or python.

What is between /* and */ are arexx type comments.

Along the way there is code for testing such as printing and writing and
reading a file, the PK file. The PK file keeps track of information
regarding the other commands, the information they make use of.

the "SAY" command is Arexx and like the Prints statement. For now there is
alot of say/print that can later be easily changed to a function call that
does something more. But for now the say/print allows us to know we are
getting the to the point of a place for a function call.


The sections like:

    WHEN input_command = AI THEN SELECT

        WHEN user_instruction = r1 THEN SAY vic.1.1
        WHEN user_instruction = r2 THEN SAY vic.1.2
        WHEN user_instruction = r3 THEN SAY vic.1.3

        WHEN user_instruction = c1 THEN vic.1.1=change2make
        WHEN user_instruction = c2 THEN vic.1.2=change2make
        WHEN user_instruction = c3 THEN vic.1.3=change2make

Are short direct accesses to the array.


maybe read the later half of http://www.mindspring.com/~timrue/KC.html
web page where you might recognize the below as being taken from it.

The PK file and it's manipulation will alter the direction the VIC takes
in doing things.



#! Python

# PK file array - VIC Dictionary
# Setting the standard variables to defaults

vic = {

    'AI': {
        'AI_1': AI_name_number,
        'AI_2': PK_file_directory,
        'AI_3': Current_Directory },

    'PK': {
        'PK_1': PK_filename,
        'PK_2': last_alt_PK_filename,
        'PK_3': default_PK_filenane },

    'OI': {
        'OI_1': OI_filename,
        'OI_2': last_alt_OI_filename,
        'OI_3': opt_default_OI_filename },

    'IP': {
        'IP_1': device,
        'IP_2': preprocess_or_Class,
        'IP_3': BOI_EOI,
        'IP_4': opt_last_alt_set },

    'OP': {
        'OP_1': device,
        'OP_2': postprocess_or_Class,
        'OP_3': BOO_EOO,
        'OP_4': opt_last_alt_set },

    'SF': {
        'SF_1': SF_LPC_flags,
        'SF_2': last_alt_flags,
        'SF_3': SF_fname_at_line_no,
        'SF_4': opt_last_alt_set,
        'SF_5': SF_filename_at_line_no_running_stack },

    'IQ': {
        'IQ_1': IQ_flags,
        'IQ_2': last_alt_flags,
        'IQ_3': IQ_fname_at_line_no,
        'IQ_4': opt_last_alt_set,
        'IQ_5': IQ_filename_at_line_no_running_stack },

    'ID': {
        'ID_1': ID_flags,
        'ID_2': last_alt_flags,
        'ID_3': ID_fname_at_line_no,
        'ID_4': opt_last_alt_set,
        'ID_5': ID_filename_at_line_no_running_stack },

    'KE': {
        'KE_1': Master_teeth,
        'KE_2': last_alt_Master_teeth,
        'KE_3': KE_fname,
        'KE_4': opt_last_alt_set }

}



/* Test!!!
SAY vic.1":" vic.1.1";"vic.1.2";"vic.1.3
SAY vic.2":" vic.2.1";"vic.2.2";"vic.2.3
SAY vic.3":" vic.3.1";"vic.3.2";"vic.3.3
SAY vic.4":" vic.4.1";"vic.4.2";"vic.4.3
SAY vic.5":" vic.5.1";"vic.5.2";"vic.5.3
SAY vic.6":" vic.6.1";"vic.6.2";"vic.6.3
SAY "  "vic.6.5
SAY vic.7":" vic.7.1";"vic.7.2";"vic.7.3
SAY "  "vic.7.5
SAY vic.8":" vic.8.1";"vic.8.2";"vic.8.3
SAY "  "vic.8.5
SAY vic.9":" vic.9.1";"vic.9.2";"vic.9.3
*/

/*==================================*/
/*Reading in the PK file.
  If no filename is present then read default.
  if default does not exist, warn and create*/

PKfilename = 0

/*check internal variables*/

IF vic.2.1 > '' THEN PKfilename = vic.2.1

    ELSE DO

    IF VIC.2.2 > '' THEN PKfilename = vic.2.2

        END


/* If above fails try "default.pk" */

IF PKfilename = 0 THEN DO

    IF EXISTS(vic.2.3) = 0 THEN DO

    SAY 'WARNING: default PK file ('vic.2.3') not available. Creating It'

    /* create/write PK file from internal default*/

        END

    ELSE PKfilename = vic.2.3

    END


IF PKfilename ~= 0 THEN DO

    IF OPEN('PK_file',PKfilename,'R') ~= 0 THEN DO

    /* read the result file if available */

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.1 ':' vic.1.1 ';' vic.1.2 ';' vic.1.3

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.2 ':' vic.2.1 ';' vic.2.2 ';' vic.2.3

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.3 ':' vic.3.1 ';' vic.3.2 ';' vic.3.3

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.4 ':' vic.4.1 ';' vic.4.2 ';' vic.4.3

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.5 ':' vic.5.1 ';' vic.5.2 ';' vic.5.3

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.6 ':' vic.6.1 ';' vic.6.2 ';' vic.6.3
    vic.6.5 = READLN('PK_file')

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.7 ':' vic.7.1 ';' vic.7.2 ';' vic.7.3
    vic.7.5= READLN('PK_file')

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.8 ':' vic.8.1 ';' vic.8.2 ';' vic.8.3
    vic.8.5 = READLN('PK_file')

    current_line = READLN('PK_file')
    PARSE VAR current_line vic.9 ':' vic.9.1 ';' vic.9.2 ';' vic.9.3

/* this is how we will keep track of sf,iq,id line numbers ???*/

    SAY SEEK('PK_file',0,'C')


        END


    ELSE DO

        SAY  "variable_pkfile not available"

        END

    END


/*=====================*/
/* Writing the PK file */

SAY OPEN('PK-file-out','*','W')

WRITELN('PK-file-out',vic.1":"vic.1.1";"vic.1.2";"vic.1.3)
WRITELN('PK-file-out',vic.2":"vic.2.1";"vic.2.2";"vic.2.3)
WRITELN('PK-file-out',vic.3":"vic.3.1";"vic.3.2";"vic.3.3)
WRITELN('PK-file-out',vic.4":"vic.4.1";"vic.4.2";"vic.4.3)
WRITELN('PK-file-out',vic.5":"vic.5.1";"vic.5.2";"vic.5.3)
WRITELN('PK-file-out',vic.6":"vic.6.1";"vic.6.2";"vic.6.3)
WRITELN('PK-file-out',vic.6.5)
WRITELN('PK-file-out',vic.7":"vic.7.1";"vic.7.2";"vic.7.3)
WRITELN('PK-file-out',vic.7.5)
WRITELN('PK-file-out',vic.8":"vic.8.1";"vic.8.2";"vic.8.3)
WRITELN('PK-file-out',vic.8.5)
WRITELN('PK-file-out',vic.9":"vic.9.1";"vic.9.2";"vic.9.3)

CLOSE('PK-file-out')

/*====================================*/
/* passing lines not used by the vic */

/*vic.5.1 = 'rexx_csh'

line_to_pass =   dir /*'echo "line to_pass via OP setting"'*/

ADDRESS VALUE vic.5.1 ; line_to_pass*/

echo vic.

echo "line to_pass via OP setting  vic"

/*============================================*/
DO FOREVER
OPTIONS PROMPT "INPUT PLEASE:"
PULL input_command user_instruction change2make

SELECT

    WHEN input_command = AI THEN SELECT

        WHEN user_instruction = r1 THEN SAY vic.1.1
        WHEN user_instruction = r2 THEN SAY vic.1.2
        WHEN user_instruction = r3 THEN SAY vic.1.3

        WHEN user_instruction = c1 THEN vic.1.1=change2make
        WHEN user_instruction = c2 THEN vic.1.2=change2make
        WHEN user_instruction = c3 THEN vic.1.3=change2make


/* AI Sets up the PK file and makes available the internal commands.
   May be used with any of the following options*/

     /*AI or -n name*/
        WHEN user_instruction = '' THEN SAY "startup of AI w/defaults"

        WHEN user_instruction = '-'n THEN SAY "startup w/AI file"

     /*AI -pk filename or filename.pk*/
        WHEN user_instruction = '-'pk THEN vic.2.1=change2make

        WHEN user_instruction = .pk THEN SAY "startup w/.pk file"

     /*AI -oi filename or filename.oi */
        WHEN user_instruction = '-'oi THEN vic.3.1=change2make

        WHEN user_instruction = .oi THEN SAY "startup w/.OI file"

     /*AI -ip (set options) or -i (set options) */

        WHEN user_instruction = '-'ip THEN vic.4.1=changes2make

        WHEN user_instruction = '-'i THEN SAY "startup w/i (ip) settings"

     /*AI -op (set options) or -o (set options) */
        WHEN user_instruction = '-'op THEN SAY "startup w/op settings"

        WHEN user_instruction = '-'o THEN SAY "startup w/i (op) settings"

     /*AI -sf filename at line# or filename.sf at line# */
        WHEN user_instruction = '-'sf THEN SAY "startup w/sf file at line#"

        WHEN user_instruction = .sf THEN SAY "startup w/.sf file at line#"

     /*AI -iq filename or filename.iq */
        WHEN user_instruction = '-'iq THEN SAY "startup w/iq file"

        WHEN user_instruction = .iq THEN SAY "startup w/.iq file"

     /*AI -id filename or filename.id */
        WHEN user_instruction = '-'id THEN SAY "startup w/id file"

        WHEN user_instruction = .id THEN SAY "startup w/.id file"

     /*AI -ke filename or filename.ke */
        WHEN user_instruction = '-'ke THEN SAY "startup w/ke file"

        WHEN user_instruction = .ke THEN SAY "startup w/.ke file"

     /* AI -? or AI -h */
        WHEN user_instruction = '-'? THEN SAY "AI help"

        WHEN user_instruction = '-'h THEN SAY "AI HELP"

     /*AI -d  or -w debug/watch toggle */
        WHEN user_instruction = '-'d THEN SAY "startup w/debug"

        WHEN user_instruction = '-'w THEN SAY "startup w/watch"

     /*AI exit or -x AI configuration */
        WHEN user_instruction = exit THEN SAY "exit AI configuration"

        WHEN user_instruction = '-'e THEN SAY "Exit AI configuration"


/* Example: AI -pk DF1:Testfile script.sf -ke Holder */

        OTHERWISE SAY "AI instruction not valid!!"
        END



    WHEN input_command = PK THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.2.1
        WHEN user_instruction = r2 THEN SAY vic.2.2
        WHEN user_instruction = r3 THEN SAY vic.2.3

        WHEN user_instruction = c1 THEN vic.2.1=change2make
        WHEN user_instruction = c2 THEN vic.2.2=change2make
        WHEN user_instruction = c3 THEN vic.2.3=change2make


/*PK*/

    /* PK returns PK pk line*/
        WHEN user_instruction = '' THEN SAY vic.2.1 vic.2.2 vic.2.3


    /* PK -? help on PK*/

        WHEN user_instruction = '-'? THEN SAY PK help

     /*--------------------------------------------------*/

     /*PK pk toggle positions 1 & 2*/

        WHEN user_instruction = pk THEN SAY vic.2.2 vic.2.1 vic.2.3

     /*PK pk(filename)             make pk file current*/

        WHEN user_instruction = pk.file THEN SAY pk.file vic.2.1 vic.2.3

     /*PK -c pk                    cancel current pk file*/

        WHEN user_instruction = '-'c.pk THEN SAY vic.2.2 ',' ',' vic.2.3

     /*PK -c pk(filename)          replace current with*/

        WHEN user_instruction = '-'c.pk.file THEN SAY c.pk.file vic.2.2 vic.2.3

/*---------------------------------------------------------------------*/

     /*PK oi                       toggle positions 1 & 2*/

     /*PK oi(filename)             make oi file current*/

     /*PK -c oi                    cancel current io file*/

     /*PK -c oi(filename)          replace current with*/

/*---------------------------------------------------------------------*/

     /*PK ip                       toggle positions 1 & 2*/

     /*PK ip(set-options)          make ip set current*/

     /*PK -c ip                    cancel current ip set*/

     /*PK -c ip(set-options)       replace current with*/

/*---------------------------------------------------------------------*/

     /*PK op                       toggle position 1 & 2*/

     /*PK op(set-options)          make op set current*/

     /*PK -c op                    cancel current op set*/

     /*PK -c op(set-options)       replace current with*/

/*---------------------------------------------------------------------*/

/*     PK sf                       swaps current sf filename at line# with
                                 previous sf filename at line#

     PK sf filename at line#...     add filename at line# to top of stack

     PK -c sf                    cancels PK SF line

     PK -c sf filename at line#...  changes PK SF line to
                                 *note this can be rather destructive
                                 to script execution and OI var lists

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

     PK iq                       swaps current iq filename at line# with
                                 previous iq filename at line#

     PK iq filename at line#...     make IQ file current or put on top
                                 of stack

     PK -c iq                    cancels PK IQ line

     PK -c iq filename at line#...  changes PK IQ line to

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

     PK id                       swaps current id filename at line# with
                                 previous id filename at line#

     PK id filename at line#...     make ID file current or put on top
                                 of stack

     PK -c id                    cancels PK ID line

     PK -c id filename at line#...  changes PK ID line to

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

     PK ke                       toggles positions 1 & 2

     PK ke filename              make ke file current keychain

     PK -c ke                    cancel current keychain file

     PK -c ke filename           replace current keychain file with
*/
        OTHERWISE SAY PK help
        END


/*OI*/


    WHEN input_command = OI THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.3.1
        WHEN user_instruction = r2 THEN SAY vic.3.2
        WHEN user_instruction = r3 THEN SAY vic.3.3

        WHEN user_instruction = c1 THEN vic.3.1=change2make
        WHEN user_instruction = c2 THEN vic.3.2=change2make
        WHEN user_instruction = c3 THEN vic.3.3=change2make

        OTHERWISE SAY OI help
        END

/*
OI

     OI                          returns PK OI line

     OI -?                       help on OI

     OI -oi                      returns current list of vars & values

     OI variable                 gets input from user into variable
                                 value using IP set. if input is
                                 greater than 255 a file is created
                                 and input is to file and value is
                                 set to filename O#I.var where # is
                                 unique (perhaps filedate/time/AI#)

     OI -c var                   cancels variable

     OI -c var value             creates/changes var value

*/

    WHEN input_command = IP THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.4.1
        WHEN user_instruction = r2 THEN SAY vic.4.2
        WHEN user_instruction = r3 THEN SAY vic.4.3

        WHEN user_instruction = c1 THEN vic.4.1=change2make
        WHEN user_instruction = c2 THEN vic.4.2=change2make
        WHEN user_instruction = c3 THEN vic.4.3=change2make

        OTHERWISE SAY IP help
        END

/*
IP            only affects OI

     IP                          returns PK IP line

     IP -?                       help on IP

*/

    WHEN input_command = OP THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.5.1
        WHEN user_instruction = r2 THEN SAY vic.5.2
        WHEN user_instruction = r3 THEN SAY vic.5.3

        WHEN user_instruction = c1 THEN vic.5.1=change2make
        WHEN user_instruction = c2 THEN vic.5.2=change2make
        WHEN user_instruction = c3 THEN vic.5.3=change2make

        OTHERWISE SAY OP help
        END
/*
OP            only affects ID, IQ

     OP                          returns PK OP line

     OP -?                       help on OP

*/
    WHEN input_command = SF THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.6.1
        WHEN user_instruction = r2 THEN SAY vic.6.2
        WHEN user_instruction = r3 THEN SAY vic.6.3

        WHEN user_instruction = c1 THEN vic.6.1=change2make
        WHEN user_instruction = c2 THEN vic.6.2=change2make
        WHEN user_instruction = c3 THEN vic.6.3=change2make

        OTHERWISE SAY SF help
        END

/*
SF

     SF                          returns PK SF line

     SF -?                       help on SF

     SF -sf                      returns SF line contents

     SF filename at line#...        returns given filename at line# contents

*/

    WHEN input_command = IQ THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.7.1
        WHEN user_instruction = r2 THEN SAY vic.7.2
        WHEN user_instruction = r3 THEN SAY vic.7.3

        WHEN user_instruction = c1 THEN vic.7.1=change2make
        WHEN user_instruction = c2 THEN vic.7.2=change2make
        WHEN user_instruction = c3 THEN vic.7.3=change2make

        OTHERWISE SAY IQ help
        END

/*
IQ

     IQ                          returns PK IQ line.

     IQ -?                       help on IQ

     IQ -iq                      returns IQ search path based on
                                 KEychain.

     IQ -iq word                 returns IQ search path based on
                                 KEychain and word

     IQ word                     Searches for word using KEychain
                                 access and executes line_contents
                                 commands based on KEychain teeth
                                 and OP set.

     IQ -c word                  Cancels word/line_contents from
                                 file(s) using KEychain access.


     IQ -c word line_contents    Creates or Changes word/line_contents
                                 using KE keychain access for search.
                                 If Creating it will add to the end of
                                 path - end of path is end if first
                                 file.

*/

    WHEN input_command = ID THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.8.1
        WHEN user_instruction = r2 THEN SAY vic.8.2
        WHEN user_instruction = r3 THEN SAY vic.8.3

        WHEN user_instruction = c1 THEN vic.8.1=change2make
        WHEN user_instruction = c2 THEN vic.8.2=change2make
        WHEN user_instruction = c3 THEN vic.8.3=change2make

        OTHERWISE SAY ID help
        END

/*
ID
          Test word for being a var, var-contents, alias, alias
          contents, IQ word, directory, file. If file will try to
          determine type of file. executes commands based upon
          findings (wordf).
          Tests what the word is in relationship to the related AI
          system.

     ID                          returns PK ID line.

     ID -?                       help on ID

     ID -id                      returns ID search path based on
                                 KEychain.

     ID -id word                 returns ID search path based on
                                 KEychain and wordf

     ID word                     Searches for wordf using KEychain
                                 access and executes line_contents
                                 commands based on KEychain teeth
                                 and OP set.

     ID -c wordf                 Cancels word/line_contents from
                                 file(s) using KEychain access.


     ID -c wordf line_contents   Creates or Changes word/line_contents
                                 using KE keychain access for search.
                                 If Creating it will add to the end of
                                 path - end of path is end if first
                                 file.
*/



    WHEN input_command = KE THEN SELECT
        WHEN user_instruction = r1 THEN SAY vic.9.1
        WHEN user_instruction = r2 THEN SAY vic.9.2
        WHEN user_instruction = r3 THEN SAY vic.9.3

        WHEN user_instruction = c1 THEN vic.9.1=change2make
        WHEN user_instruction = c2 THEN vic.9.2=change2make
        WHEN user_instruction = c3 THEN vic.9.3=change2make

        OTHERWISE SAY KE help
        END

/*
KE
               To be used by IQ and ID mainly.

     KE                          returns PK KE line.

     KE -?                       help on KE.

     KE -ke                      returns current keychain list w/values.

     KE keyname                  returns value of keyname if exist.

     KE -c keyname               cancels keyname.

     KE -c keyname key-teeth     creates or changes keyname.

*/


    OTHERWISE SAY Not Valid command':>' input_command user_instruction change2make
    LEAVE
    END

/* Writing the PK file */

SAY OPEN('PK-file-out','*','W')

WRITELN('PK-file-out',vic.1":"vic.1.1";"vic.1.2";"vic.1.3)
WRITELN('PK-file-out',vic.2":"vic.2.1";"vic.2.2";"vic.2.3)
WRITELN('PK-file-out',vic.3":"vic.3.1";"vic.3.2";"vic.3.3)
WRITELN('PK-file-out',vic.4":"vic.4.1";"vic.4.2";"vic.4.3)
WRITELN('PK-file-out',vic.5":"vic.5.1";"vic.5.2";"vic.5.3)
WRITELN('PK-file-out',vic.6":"vic.6.1";"vic.6.2";"vic.6.3)
WRITELN('PK-file-out',vic.6.5)
WRITELN('PK-file-out',vic.7":"vic.7.1";"vic.7.2";"vic.7.3)
WRITELN('PK-file-out',vic.7.5)
WRITELN('PK-file-out',vic.8":"vic.8.1";"vic.8.2";"vic.8.3)
WRITELN('PK-file-out',vic.8.5)
WRITELN('PK-file-out',vic.9":"vic.9.1";"vic.9.2";"vic.9.3)

CLOSE('PK-file-out')

END

/*=====================*/
/* Writing the PK file */

SAY OPEN('PK-file-out','test.pk','W')

WRITELN('PK-file-out',vic.1":"vic.1.1";"vic.1.2";"vic.1.3)
WRITELN('PK-file-out',vic.2":"vic.2.1";"vic.2.2";"vic.2.3)
WRITELN('PK-file-out',vic.3":"vic.3.1";"vic.3.2";"vic.3.3)
WRITELN('PK-file-out',vic.4":"vic.4.1";"vic.4.2";"vic.4.3)
WRITELN('PK-file-out',vic.5":"vic.5.1";"vic.5.2";"vic.5.3)
WRITELN('PK-file-out',vic.6":"vic.6.1";"vic.6.2";"vic.6.3)
WRITELN('PK-file-out',vic.6.5)
WRITELN('PK-file-out',vic.7":"vic.7.1";"vic.7.2";"vic.7.3)
WRITELN('PK-file-out',vic.7.5)
WRITELN('PK-file-out',vic.8":"vic.8.1";"vic.8.2";"vic.8.3)
WRITELN('PK-file-out',vic.8.5)
WRITELN('PK-file-out',vic.9":"vic.9.1";"vic.9.2";"vic.9.3)

CLOSE('PK-file-out')

/**********************************************************************


File/Line Formats

OI - "variable : value : opt comment"

IP - "input type/format : input device : end of input char"

OP - "output type/format : output device : selection flags"

    selection flags:

                   -f = First match
                   -a = All matches
                   -b = Best match
                   -e = Exact match(s)
                   -l = Last match

                   -w = wait until search is complete before applying
                        found. A temp file is used to hold findings.
    IQ only?       -d = dual/two way pattern matching
    IQ only?       -s = soundex (a future application maybe)

                   -p at name = position name of match(s) contents to use.
                            These variables are related to key-teeth.
                            (may specify more than one)
                   *note: standards on -p at name use should be developed.

ID - "wordf to test : command-line to do testing : command/contents
                                                   -p at name posititions"

     If file - lines beginning with FILE>var will be tested:

     "FILE>type : command-line to do testing : command/contents
                                               -p at name positions"

     "keychain-pattern : keyname : file to check"


IQ - "word-pattern : opt command-line if match : command/contents
                                                 -p at name posititions"

     "keychain-pattern : keyname : file to check"

---------------
KE - "keyname:key-teeth/-p at name (to compare w/file-comment):level-check"

            key-teeth/-p at name:
     user definable but defaults/examples are:

                            -t = text
                            -e = executable calls
                            -s = shell script code
                            -ed = education
                            -ref = referance
                            -appl = application

When the keychain is called:
                          :: key-name is looked for in current KE file
                            and if found... does level-check...
***file-comment/OP-p at name 1: key-teeth/-p at name are compared with file
                            comment and if there is a fit...
***ID/IQ LISTS            2: ID/IQ list is searched for a
                            path:filename(date) match and if not found
                            then adds filename to top of stack and list.
                            If found then...
                          3: PK's ID/IQ line is searched for filename
                            and if found moves it to top of stack and
                            picks up at line#. If not found then skips
                            filename (it's already been looked at)...

NEED TO PUT MORE THOUGHT INTO THIS!!!
***********************************************************************

                    ----- END OF DOCUMENT -----
*/


/*--------------------------------------------------------------------------
Search on my name at http://www.dejanews.com for other posts/peices.

   Timothy Rue                             Email: timrue at mindspring.com
                Virtual Interaction Configuration Web Page:
                    http://www.mindspring.com/~timrue/
   A Contribution to the Paradigm Shift of software development and use.
--------------------------------------------------------------------------*/



From sdm7g at Virginia.EDU  Fri Feb 15 15:51:01 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Fri, 15 Feb 2002 15:51:01 -0500 (EST)
Subject: Warning about "for line in file:"
In-Reply-To: <20020215110159.B12454@glacier.arctrix.com>
Message-ID: <Pine.OSX.4.43.0202151547390.318-100000@d-128-61-180.bootp.virginia.edu>


On Fri, 15 Feb 2002, Neil Schemenauer wrote:

> Russell E. Owen wrote:
> > The readline or xreadline file methods work fine, of course.
>
> Why "of course"?  iter(file) does the same thing as file.xreadlines().
> Have you tested xreadlines?
>

I also get the same behavior on any example I can contrive on
file iterators that I get with with either readlines or xreadlines.

Russell: could you post an example that shows the behaviour you're
seeing ?

-- Steve Majewski





From store_li at sina.com  Thu Feb  7 12:46:35 2002
From: store_li at sina.com (Ricky)
Date: Fri, 8 Feb 2002 01:46:35 +0800
Subject: Questions  about pyscript
References: <a3tegq$1ahish$1@ID-12869.news.dfncis.de> <3C62727A.1090306@skippinet.com.au>
Message-ID: <a3uekm$1avlmo$1@ID-12869.news.dfncis.de>

Thank you! So, I need to mixed DHTML and python mothed to code client side
program, right?

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C62727A.1090306 at skippinet.com.au...
> Ricky wrote:
> > I make a try a pyscript. But there is one thin confuse me. When I use
print
> > statement intent to send something to browse. It doesn't work, while
> > document.write can make it. What make me confused is what part is
controlled
> > under python(such as flow control, functions, package etc) and which
part is
> > controlled by DHTML(output?).
> >
> > Could some one give me an over view option of what pyscript is and the
> > object tree map of pyscript is a plus.
>
> ActiveScripting (ie, pyscript) is a technology that allows an
> application to expose an object model, and allow many languages to use
> scripting.  Python works with IE, ASP, WSH and a number of other
> applications - thus there is no "pyscript" documentation that can help
you.
>
> IE is just one such application (and the one you seem to be using).  MS
> documents this object model with IE - however, you are only likely to
> find examples for VBScript or jscript, so will need to convert them to
> Python yourself.
>
> Since pyscript doesn't really know about IE in any special way, it
> doesn't know how to redirect 'print' statements to document.write()
> calls.  That has been my exuse anyway - hard-coding for IE wouldn't
> necessarily be too evil :)
>
> Mark.
>




From tim.hochberg at ieee.org  Wed Feb 20 13:51:21 2002
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 20 Feb 2002 18:51:21 GMT
Subject: add-a-gram
References: <mailman.1014175232.10747.python-list@python.org> <slrna76tq6.915.mlh@vier.idi.ntnu.no>
Message-ID: <JeSc8.2069$XT1.78163@news1.east.cox.net>

"Magnus Lie Hetland" <mlh at vier.idi.ntnu.no> wrote in message

> That's pretty quick... Anyway, if you ignore the setup-time (reading
> in the file and storing it in a useful way; this took 6 seconds),
> the search for a solution took 0.9 seconds on my PC -- implemented in
> Python, mind you.

Interesting, I must have done something different. My setup time is longer
(~15 s), but once it's setup I can find an anagram in something like 2/100
of a second, most of which is printing.

> (Of course ignoring the setup isn't very realisting
> in this case, where there is only one solution -- but if you were to
> be able to find an addagram of a user-supplied length, for instance,
> that would be realistic.)

I'm coming up with two words of length X for possible answers, did I mess
that up, or did you just mean that all we care about it X? X left
unspecified in case someone is still working on this.

-tim





From SBrunning at trisystems.co.uk  Wed Feb 20 10:58:47 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 20 Feb 2002 15:58:47 -0000
Subject: Guido wins FSF Award
Message-ID: <31575A892FF6D1118F5800600846864DCBCCC4@intrepid>

> From:	grante at visi.com [SMTP:grante at visi.com]
> > what are revealing the Nine Pythonic Actions underlying all that we do.
> > can you explain what that means ??????
> 
> Bhuddism allusion?
 
Autocoding allusion, I think.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From jeff at ccvcorp.com  Mon Feb  4 21:54:10 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 18:54:10 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com> <7xheow3hl2.fsf@ruckus.brouhaha.com> <3C5F3A07.45126DAF@ccvcorp.com> <7xg04gisvz.fsf@ruckus.brouhaha.com>
Message-ID: <3C5F4951.226DF029@ccvcorp.com>


Paul Rubin wrote:

> ...So I wonder now, whether more people are using 2.1.* than
> 1.5.2, and whether there will still be a significant amount of 2.1
> installations when 3.0 is released.  If it sounds like there won't be,
> then I'll probably just put
>
>    if 3/2==1: <print upgrade-to-new-script message and exit>
>
> at the top of my scripts and then code for 2.1.  When 2.1 gets more
> displaced by 2.2 and later, I can switch over to import future division.

My suspicion is that it's not that 2.0 never caught on, but that once 2.1
came out, there was no reason *not* to use it.... except for those
circumstances that had people stuck with 1.5.2.  Similarly, I think that
once 2.2 has been out for a bit longer (and more binary packages are built
for it--notably, I'm still waiting for ActivePython to have a 2.2final
release...), there won't be much reason to hang on to 2.1 (presuming, of
course, that RH8.0 uses 2.2 and not 2.1).  At least, that's what I'm
*hoping* will happen.  :)

Jeff Shannon
Technician/Programmer
Credit International




From justin at iago.org  Fri Feb 15 09:27:42 2002
From: justin at iago.org (Justin Sheehy)
Date: Fri, 15 Feb 2002 09:27:42 -0500
Subject: sorting list of tuples by second (third...) tuple item
In-Reply-To: <5e8bd451.0202142238.580f15c1@posting.google.com> (kp87@lycos.com's
 message of "14 Feb 2002 22:38:38 -0800")
References: <3c6bbe96.12195968@news.easynews.net>
	<mailman.1013718563.21631.python-list@python.org>
	<5e8bd451.0202142238.580f15c1@posting.google.com>
Message-ID: <x7it8yrfa9.fsf@ra.iago.org>

kp87 at lycos.com (kevin parks) writes:

> How would you sort something based on 2 criteria. Say you have a list
> of lists (i hate tuples). first thing you want is everything sorted by
> the first value and then everything that has the same value for [0]
> then sorted by [1]

This is the default behavior of Python's listobject's sort method.

It will sort by the first element, then the second, then the third, etc...

If you want some other order, you have two choices:

 - write a custom cmp function to pass to sort

 - decorate your list so that the default sort does what you want

-Justin

 




From marklists at mceahern.com  Tue Feb 12 16:24:29 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 12 Feb 2002 13:24:29 -0800
Subject: searching backwards in a string
In-Reply-To: <7x7kpi76p0.fsf@ruckus.brouhaha.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEIFKCAA.marklists@mceahern.com>

[Paul Rubin]
> Thanks but I don't think that solution works if the pattern being
> searched for is a complicated regexp.  Even if the pattern is a simple
> string, it has to be reversed too.

That's probably true.  Would you be so kind as to post an example of the
input (string, complicated regexp) and output (do you want it to tell you
the starting position from left to right of the first match found backwards
in the string?) you'd expect?

Thanks,

// mark




From threeseas at earthlink.net  Tue Feb  5 23:54:36 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:36 GMT
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <Xns91AB3616C9C9Bahmebahhotmailcom@198.99.146.10>
Message-ID: <811.802T1774T34536threeseas@earthlink.net>

On 04-Feb-02 08:17:44 ameoba <ahmebah at hotmail.com> wrote:
>"Timothy Rue" <threeseas at earthlink.net> wrote in
>news:972.799T919T11694325threeseas at earthlink.net:

>> Autocoding will be a nice thing to have, it'll get rid of all the
>> arrogant jerks.


>Unfortunately, it will do nothing to remove people spouting off meaningless
>drivel and claiming it's revolutionary.

Hmmmm, try evolution a bit.

Was GNU revolutionary or evolutionary?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From nas at python.ca  Sat Feb 16 12:32:28 2002
From: nas at python.ca (Neil Schemenauer)
Date: Sat, 16 Feb 2002 09:32:28 -0800
Subject: Warning about "for line in file:"
In-Reply-To: <20020216134444.GA12676@nullcube.com>; from aldo@nullcube.com on Sun, Feb 17, 2002 at 12:44:44AM +1100
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu> <20020216134444.GA12676@nullcube.com>
Message-ID: <20020216093228.A15411@glacier.arctrix.com>

Aldo Cortesi wrote:
> Actually this has nothing to do with iterators, or a "read
> cache".

It does.

> iter(file) creates a line iterator that does the
> same thing as file.readline() every time .next() is called,
> until it reaches the end of the file.

It does not.  iter(file) calls file.xreadlines().  xreadlines()
internally calls readlines(CHUNKSIZE).

> But file.readline(), just like any other file read,
> starts reading at the _current seek position_ of the file. 

True but irrelevant to the discussion at hand.

  Neil



From NOSpawnPAM at lightspawn.org  Fri Feb  1 04:37:00 2002
From: NOSpawnPAM at lightspawn.org (pawn)
Date: 1 Feb 2002 01:37:00 -0800
Subject: Running CGIs under my uid - going slowly insane.
Message-ID: <4b4c5d99.0202010136.8e6ff51@posting.google.com>

Here's the situation:

I need my Python CGIs to run scripts under my own uid, since they read
and update .html files which are core to the site.

The company hosting my site, addr.com, seems unable or unwilling to
help.

Several people have pointed me in the direction of 'cgi wrappers', but
I don't seem to get anything to work; perhaps I'm using things not
meant for this. Ideally, I'd just change the interpreter from
#!/usr/bin/python to the wrapper.

Solutions I've thought about:

* Switch back to perl which respects 04755 permissions.
* Kick addr.com and find a reasonable hosting provider (time+money
expense)
* Find out if anybody here can help
* Inflict physical harm upon self until cgi problems are either
forgotten or seem unimportant by comparison.
* Swear off computers forever, move to the woods.

What do you think? is the hosting company at fault here? Am I trying
to do something ridiculous? All I know is with perl it was trivial, in
python it's a pain (just like using the CGI package, when I get
something that seems to be a dictionary except it's read only and some
method can't be applied to it - why? why?)


From sholden at holdenweb.com  Thu Feb  7 08:09:20 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 08:09:20 -0500
Subject: wxPython book
References: <88bc63c6.0202040505.2a9c3016@posting.google.com> <cb4ba455.0202060441.54e1d529@posting.google.com>
Message-ID: <T%u88.94053$KM2.3390206@atlpnn01.usenetserver.com>

"Bill Tate" <tatebll at aol.com> wrote in message
news:cb4ba455.0202060441.54e1d529 at posting.google.com...
> writeson at earthlink.net (Doug Farrell) wrote in message
news:<88bc63c6.0202040505.2a9c3016 at posting.google.com>...
> > Hi all,
> >
> > I'm busy evangilizing my office co-workers on the wonders of Python
> > and wxPython, to which they are responding pretty well. We are
> > currently working on a test project that uses both, and the most
> > frequent complaint I get is "there's no book on wxPython..... wah wah
> > wah". I have seen mention of a book being written, presumably by Robin
> > Dunn. Is this true and if so, any idea when it might be coming out?
> >
> > Thanks,
> > Doug
>
> Doug,
> While I would say it would be nice to have a book, suggest you check
> out wxDesigner (it still has a way to go in terms of ALL the bells &
> whistles but what's there already is pretty damn nice).  I like that I
> can use one visual designer for the interface and build the skeleton
> of the C++ and Python code from within the same IDE.  It made using
> wxWindows/wxPython much more intuitive.  I must say, however, it
> really is a bummer when you get those quick responses to questions
> from the wxwindows/wxpython discussion list - I much prefer that
> "timely and high quality" support you get from commercial vendors :>).

Well, I too would be interested in a wxPython book, but I'd prefer Robin
Dunn didn't write it do he can carry on developing his excellent software
products <wink> (I'm also a great Xitami promoter).

The thing I find most frustrating (and have just aded a FAQ question about
on the Wiki) is the lack of a connection between my intuition and the way
sizers work. I have licensed a copy of wxDesigner just to try and get over
this hurdle, but dammit I *still* can't put interfaces together that work
the way I expect them to under window resizing.

Can anyone explain to this bear of little brain?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From fredrik at pythonware.com  Mon Feb 11 07:41:54 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 12:41:54 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com> <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu> <8852d33c.0202110117.374f062@posting.google.com> <dAO98.148833$KM2.6668696@atlpnn01.usenetserver.com>
Message-ID: <m_O98.17698$n4.3035458@newsc.telia.net>

Steve Holden wrote:
> > So + has standard behaviour but will  adopt whatever is specified in
> > the __add__ function if one is present for the class.
> >
> > Yes?
>
> Correct

if standard behaviour means "raise a TypeError exception", yes.

if standard behaviour means "add the things in a standard way", no.

</F>




From janfbr at sex.ifi.uio.no  Sat Feb 16 16:15:43 2002
From: janfbr at sex.ifi.uio.no (Jan-Fredrik Braseth)
Date: 16 Feb 2002 22:15:43 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122>
Message-ID: <xyuk7tdku0w.fsf@sex.ifi.uio.no>

bokr at oz.net (Bengt Richter) writes:

[snip]

> # I would change opt to c in the following line
>                     self.msg('IAC %d not recognized' % ord(opt))

But if there really is something wrong with that, shouldn't someone
send a bug report or something to the creators of Python?

-- 
Jan-Fredrik Braseth		www.ifi.uio.no/~janfbr/
2B|!2B


From James_Althoff at i2.com  Fri Feb  8 20:23:27 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 8 Feb 2002 17:23:27 -0800
Subject: OT: (ambi)dexterity of Pythonistas
Message-ID: <OF222ED1CC.94E657B8-ON88256B5B.000736FD@i2.com>

[Aahz]
>Yes, I'm aware that prices have come down a lot, but the equivalent
>flat-panel display is still twice as expensive.

I picked up the 18" flatpanels for $849.  The equivalent monitors -- 20"
and totally flat in both dimensions (Sony, Mitsubishi, etc.) -- were
similarly priced.

Still, it would be nice to see them at $100 a pop.  Maybe next year.  :-)

Jim




From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 20:17:37 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 19:17:37 -0600
Subject: just a moron trying to get python working on win98 and personal web  server
References: <3C685DDE.AA8AFD10@kc.rr.com>
Message-ID: <010101c1b363$1256d7e0$0101010a@local>

----- Original Message ----- 
From: "james" <jdaily at kc.rr.com>
> 
> RE: just a moron trying to get python working on win98 and personal web
> server

You aren't a moron if you asked here...

> i am trying to get python to work on my MS personal web server ( which
> is supposed to support python in a web url ).

Never worked for me.
 
> what do i need to do?
> 
> i have tried setting the registry.
> 
> i have set the PATH in autoexec.bat.
> 
> is there an apache alternative.

Try Xitami (we've been discussing it here).  Put the following at the top
of your CGI scripts:

    #! c:\python22\python.exe -u

Change the 22 to whatever version you are using.  Note the space after the
exclamation point.  The # should be in the first column; indenting the line
here is for emphasis.

Get it from http://www.xitami.com

Apache is great, but until the new version for Windows comes out, I can't
recommend it on Win98.  I run Apache on several Linux servers.





From thomas.heller at ion-tof.com  Thu Feb  7 12:54:34 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 7 Feb 2002 18:54:34 +0100
Subject: Getting correct file names under Windows
References: <a3ud3i$17g3$1@norfair.nerim.net>
Message-ID: <a3uf1h$1arllf$1@ID-59885.news.dfncis.de>

"Eric Brunel" <eric.brunel at pragmadev.com> wrote in message news:a3ud3i$17g3$1 at norfair.nerim.net...
> Hi all,
>
> Windows case insensitivity striked again: I've got a Python/Tkinter program
> that uses the new tk function named tk_chooseDirectory (new in Tk 8.3),
> displaying a directory selection dialog. For some reason, this function
> insists in returning the selected directory all uppercase. Unfortunately,
> even if Windows does not care about case, we definetely don't want that: the
> disk we work on is mounted from a Samba Unix - therefore case sensitive -
> server, and we want to use the same file names on both environments.
> So I tried to get the "real" file name, i.e. the one that is defined on
> Unix, and the one I actually see in Windows Explorer or DOS windows. I
> didn't succeed...
>
> Does anybody know any way to get the "real" name of a file or directory
> under Windows, i.e. the one that Explorer displays?
You know that the Explorer doesn't always display the correct case?
There's an option somewhere (Enable Uppercase Filenames or so).

To your original question: win32api.FindFiles() and os.listdir()
returns filenames in the correct case.

Thomas




From core.lists.python at core-sdi.com  Tue Feb 26 12:31:45 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Tue, 26 Feb 2002 14:31:45 -0300
Subject: Oddity question
Message-ID: <00a601c1beeb$7685d980$6e42a8c0@corelabs.coresdi.com>

Hi there!

I was fooling around with sockets last night, using a thread to deal with the read part and another method to do the print.

like this:

def thread_read (self):
  while 1:
    self.buf=self.buf+self.my_socket.recv (8)

def print_buf(self):
  if buf:
    print "bufeer-->",buf
    buf='""

But, what happened to me was that the -buf = ""- line seemed to be not working because I got outputs like this

buffer->a
buffer->ab
buffer->abpleasework
buffer->abpleaseworkpleaseflush!

So, I dont know why I changed thread_read function to:

  while 1:
    aux=self.my_socket.recv (8)
    self.buf=self.buf+aux

And this version worked properly..... but I still do not know why. To me, I've only split the line in two, using an auiliar variable... but THAT made the trick! 
why? 
 WHY?
   W H Y?

(thanks for answers)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020226/a622d95a/attachment.html>

From debl2NonoSpammyWhammy at bellatlantic.net  Sun Feb 17 18:48:30 2002
From: debl2NonoSpammyWhammy at bellatlantic.net (David Lees)
Date: Sun, 17 Feb 2002 23:48:30 GMT
Subject: ActiveState version of 2.2???
References: <1001ff04.0202170702.6ebaaf55@posting.google.com>
Message-ID: <3C704157.B872722F@bellatlantic.net>

Mark Hammond who created PythonWin is no longer employed by ActiveState,
but he has been kind enough to continue to update to 2.2 in the form of
Win32all-146 that can be downloaded from his home page at:
http://starship.python.net/crew/mhammond/win32/Downloads.html

David Lees


Bob wrote:
> 
> When?


From amuys at shortech.com.au  Wed Feb 27 21:07:40 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 27 Feb 2002 18:07:40 -0800
Subject: Problems with select
References: <pan.2002.02.26.04.17.29.538545.11917@onesite.org> <slrna7nfkh.2bd.rmunn@rmunn.dyndns.org> <pan.2002.02.27.01.03.24.922015.11917@onesite.org>
Message-ID: <7934d084.0202271807.639ea900@posting.google.com>

Alain Tesio <alain at onesite.org> wrote in message news:<pan.2002.02.27.01.03.24.922015.11917 at onesite.org>...
> Hi, thanks for the explanation, but select still returns files where
> there is nothing to read, after the end of the stream.
> 
<SNIP to output>
> 4 select returns :  ([<open file '(fdopen)', mode 'r' at 0x8060450>, <open file '(fdopen)', mode 'r' at 0x808d510>], [], [])
> something on out : ''
> something on err : ''

Quoting from the Python Library Reference "2.1.8.9 File Objects"

readline([size]) 
Read one entire line from the file. A trailing newline character is
kept in the string2.7 (but may be absent when a file ends with an
incomplete line). If the size argument is present and non-negative, it
is a maximum byte count (including the trailing newline) and an
incomplete line may be returned. An empty string is returned when EOF
is hit immediately. Note: Unlike stdio's fgets(), the returned string
contains null characters ('\0') if they occurred in the input.

So that's why you're getting an empty string, ie. EOF.  Quoting
again...

read([size]) 
Read at most size bytes from the file (less if the read hits EOF
before obtaining size bytes). If the size argument is negative or
omitted, read all data until EOF is reached. The bytes are returned as
a string object. An empty string is returned when EOF is encountered
immediately. (For certain files, like ttys, it makes sense to continue
reading after an EOF is hit.) Note that this method may call the
underlying C function fread() more than once in an effort to acquire
as close to size bytes as possible.

Notice the comment regarding tty's.  This is the reason select keeps
returning.  You can always read an EOF, so select returns to let you
read it.
You should probably close() the file once you it EOF, and remove it
from the select call.

Andrae Muys


From djmitchell at optushome.com.au  Mon Feb 11 15:23:12 2002
From: djmitchell at optushome.com.au (djmitchell)
Date: Tue, 12 Feb 2002 07:23:12 +1100
Subject: Zope cookbook?
References: <3c6786c7$0$17962$afc38c87@news.optusnet.com.au> <mailman.1013435799.31256.python-list@python.org>
Message-ID: <3c682874$0$21300$afc38c87@news.optusnet.com.au>

Paul Everitt wrote:

> 
> Hello Dave.  I think you are looking for the recipes at ZopeLabs:
> 
>    http://www.zopelabs.com/
> 
> --Paul
> 

Thanks Paul,

That's exactly what I was looking for.

Dave Mitchell


From chris.gonnerman at newcenturycomputers.net  Tue Feb 19 09:03:14 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 19 Feb 2002 08:03:14 -0600
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com> <3C71B291.B8574221@ccvcorp.com> <mailman.1014085471.28317.python-list@python.org> <slrna73h0o.18i.grante@tuxtop.visi.com> <20020219095832.GE2611@lilith.hqd-internal>
Message-ID: <009d01c1b94e$2d6b9f20$0101010a@local>

I've tried to think of a way to articulate this... but I am not
getting very far.

When I wrote mostly C, I spent a lot of time engineering library-
like functions, usually in several layers, carefully hiding data
as much as possible.  Often I had to resort to globals to cut
development time; then I regretted it later.  Simple things like
string manipulation required a lot of extra variable declarations,
checks for overruns, etc. which naturally made the code difficult
to read.

Sure, I commented, chose obvious variable names, and I tended to
segment my code into "operation"-sized units, but still, getting
an overall view of a function (or module, for that matter) was 
far from trivial.  Bugs hid in there.  I have recently been running
one of my old C programs from about ten years back, which I had
believed mostly bug-free; the many bugs I am finding are surprising.

Now most of my work is in Python.  It's true, most of my programs
are small anymore; I rarely *need* to write large programs.  One 
line in Python often equals three or more in C, and more often than
that, two or three lines replace ten to twenty lines of C.  I get
my work done, without static typing, typechecks, or assertions
(once in a while I do use an assertion...)  They *just work*, and
when they don't it's REAL obvious how I've been stupid.

Python is *readable*.  I have yet to see any proposal for static
typing that improves on that; and as I have said before, with
static typing it's not Python anymore.  Design by contract 
implemented for Python would have to be pragmatic (i.e. by
"suggestion" rather than "mandate").  

Python isn't about tying the programmer's hands, it's about freeing 
his or her mind.






From kosh at aesaeion.com  Sat Feb 16 21:35:23 2002
From: kosh at aesaeion.com (kosh at aesaeion.com)
Date: Sat, 16 Feb 2002 19:35:23 -0700 (MST)
Subject: Guido wins FSF Award
In-Reply-To: <3s4u6ugk038esv6drk5go253u95fa5o5fg@4ax.com>
Message-ID: <Pine.LNX.4.32.0202161933280.1394-100000@aesaeion.com>

On Sun, 17 Feb 2002, Courageous wrote:

>
> Really, Geedo didn't deserve this award. Since plainly Python is
> a system designed to obfuscate code, perpetuate complexity, and other-
> wise fails to recognize the simplicity of the nine command division
> of the entire physical universe, I redirect the committee to Sir
> Timothy Rue. My vehement protest to the comittee will be publically
> filed by 1100 GMT tomorrow, I swear. And well, if none of you see or
> understand why it is I have to do this, that's understandable. You are
> perhaps simply incapable of comprehending my rationale.
>

ROFLMAO!!!

Very well done comment and I agree completely with the point at which it
is aimed. :) However I think you need to state your response in terms of
the 9 basic commands that underly all things so as to make your point
better. Be sure to include about 3000 pages of explaining documentation so
that the simplicity of your answer can be seen.




From tismer at tismer.com  Sat Feb 23 12:08:31 2002
From: tismer at tismer.com (Christian Tismer)
Date: Sat, 23 Feb 2002 18:08:31 +0100
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com> <20020223001425.A8624@ibook.distro.conectiva>
Message-ID: <3C77CC8F.9040608@tismer.com>

Gustavo Niemeyer wrote:

>>>- Platform independence is gone.
>>>
>>I hope not...  I would hope to hear that "platform indepedence has
>>_ARRIVED_"
>>
> 
> That's the human nature. Tismer writes half a dozen mails, some with
> unbelivable results, and some asking for help. One of the first answers
> he gets is complaining about something.


Oh, that's fine with me. I'm happy to donate my old,
independent implementation over to the first one who
can prove to understand it. Abandoning this code
has been kind of lifesaver for me. I will not go
that path again anymore. Instead, if Guido decides
to re-write half of Python in order to make it
stackless by nature, then I'll be happy to help.

> Thanks Tismer. I'll let you know if I do any work about it or find
> someone to sponsor you. Personally, I'd like to see it running in PowerPC
> Linux. Maybe it's a good reason to learn ppc assembly.


Thank you very much, I'd appreciate any help!

Have you had a look into "switch_MS_WIN32.h"?
The switching code is below 10 assembly insns.
It should not be a problem to implement, at
least for all machines which don't keep the
stack in registers or such. Spark is such an animal.

It is funny: The switching code is much smaller
than the code that makes Stackless stackless.
The latter has more housekeeping, since I wanted
to do the wrapped interpreter call in C and not
in assembly, in order to make porting easier.

Well, back to work - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From DavidA at ActiveState.com  Tue Feb 12 20:30:46 2002
From: DavidA at ActiveState.com (David Ascher)
Date: Tue, 12 Feb 2002 17:30:46 -0800
Subject: 2001 Turing Award
References: <OF5A92159D.F5FA1C05-ON88256B59.0069ED9D@i2.com>
Message-ID: <3C69C1C6.3FE74E40@activestate.com>

OT: The press release from the ACM at
http://www.acm.org/announcements/turing_2001.html is subtitled:

  "Norwegian Team Developed Concepts for Software Now in Home
Entertainment Devices"

So the fact that the x-box uses OO is the most important impact of OO in
the world.

Gotta love the press release world. =)



From starx at pacbell.net  Sat Feb 16 02:13:41 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 07:13:41 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com> <3C6C680D.B35E7719@cosc.canterbury.ac.nz> <jash4a.sm1.ln@127.0.0.1> <8m2p6u40evl9k04e8jq14tfvpvsov96df0@4ax.com> <5nai4a.0a.ln@127.0.0.1>
Message-ID: <Xns91B6EC446DAB3RASXnewsDFE1@130.133.1.4>

Carl Banks || Thu 14 Feb 2002 10:43:49p:

> Courageous wrote:
>> 
>>>> This sounds ingenious, but what happens when you save the
>>>> file and load it back? Does it remember this information
>>>> somehow, or does it get lost?
>>>
>>>The editor remembers this information by saving it in the text file
>>>itself.
>> 
>> I believe that you're answering a different question than what
>> he asked.
> 
> Yeah.  Although the question was about vim, my mind was still thinking
> Emacs from a few posts ago.  Emacs scans the line(s) above to
> determine how to unindent, i.e., all of the unindent information is
> saved in the file.

Perhaps more clearly: The formatting of the code itself is the unindent 
information; emacs examines the code struture to figure out how to do this 
(and other things).

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From tjreedy at home.com  Sat Feb  2 10:47:28 2002
From: tjreedy at home.com (Terry Reedy)
Date: Sat, 02 Feb 2002 15:47:28 GMT
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <kST68.20653$X64.6553605@news1.rdc2.pa.home.com>

"Fred" <c at tech.usd253.org> wrote in message
news:e959e7b.0202020710.417801c8 at posting.google.com...
> Hello, I am just starting out with Python. I have a very basic book
> that is not helping me much. I have done some simple C programming
in
> the past and I think if I could see the below C code done in Python
it
> would get me over the beginners hump. > //Simple C code to create a
times table.
>
> #include <stdio.h>
> int main (void)
>
> {
>
> int i;
> int j;
>
> for (i = 1; i < 13; i++)
> {
> for (j = 1; j < 13; j++)
>   printf("%4d", i*j);
>         printf("\n");
>    }
> return 0;
> }

"Oo, oo, I got it, I got it"

for i in range(1,13):
  for j in range(1,13):
    print '%4d'%(i*j,),
  print

#outputs following (but with columns lined up in fixed-pitch font)
   1    2    3    4    5    6    7    8    9   10   11   12
   2    4    6    8   10   12   14   16   18   20   22   24
   3    6    9   12   15   18   21   24   27   30   33   36
   4    8   12   16   20   24   28   32   36   40   44   48
   5   10   15   20   25   30   35   40   45   50   55   60
   6   12   18   24   30   36   42   48   54   60   66   72
   7   14   21   28   35   42   49   56   63   70   77   84
   8   16   24   32   40   48   56   64   72   80   88   96
   9   18   27   36   45   54   63   72   81   90   99  108
  10   20   30   40   50   60   70   80   90  100  110  120
  11   22   33   44   55   66   77   88   99  110  121  132
  12   24   36   48   60   72   84   96  108  120  132  144

So, delete declarations and braces and adjust syntax of rest ...

Terry J. Reedy






From xander at yifan.net  Mon Feb  4 20:50:12 2002
From: xander at yifan.net (Alex Holkner)
Date: Tue, 5 Feb 2002 12:50:12 +1100
Subject: Floating-point multiplication
Message-ID: <OJEJKEOFJEGJNCEHNGHKGEFOCBAA.xander@yifan.net>

I'm experiencing discrepencies in floating point multiplication between the
Python interpreter and embedded python.  Running Python 2.2 under Win2k,
compiling with VC++.  The interpreter gives:

  >>> print sys.version
  '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
  >>> print 0.785398163397 * 4.0
  3.14159265359
  >>> print math.pi
  3.14159265359

Which is expected.  Under my embeddeded python, however:

  print sys.version
    '2.2 (#28, Jan  4 2002, 19:01:19) [MSC 32 bit (Intel)]'
  print 0.785398163397 * 4.0
    3.14159274101
  print math.pi
    3.14159274101

Which is quite wrong, and is upsetting 'import random', which requires
math.pi to be within a certain tolerance.

I expect that this problem is related to the runtime C libraries in use,
however both the interpreter and embedded python22.lib/python22.dll were
compiled from the source.  Building Debug vs. Release does not make a
difference.

Thanks,
Alex.





From db3l at fitlinxx.com  Thu Feb 28 20:42:18 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Feb 2002 20:42:18 -0500
Subject: win32+popen4
References: <$4LupWAmqsf8Ew0o@jessikat.fsnet.co.uk>
Message-ID: <ud6ypoyh1.fsf@ctwd0143.fitlinxx.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> What's the correct way to give a command to popen4 under win32 when the
> command/arguments may have spaces in it. EG I want to use
>
> "c:\program files\dimbo\bongo.exe" "c:\program files\dumbo\my file.dat"

Are you including the quotes in your call to popen4?  E.g.:

   os.popen4(r'"c:\program files\dimbo\bongo.exe"
               "c:\program files\dumbo\my file.dat"')

And if, so which version of Windows are you using?  I thought there
might have been an issue under Win9x platforms with losing quoted
information on the way to using the helper module w9xpopen, but a
quick Google search didn't show up anything, so I may be
mis-remembering.

However, if you are on a 9x platform, and quoting the command just as
you want the command interpreter to see it on the command line doesn't
work, you might just want to revert to using the short (non-space)
version of the name.  So in your example you would replace "program
files" with "progra~1" (or whatever it may be in the unlikely event
it's different than that).

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


From kp87 at lycos.com  Wed Feb 13 11:20:04 2002
From: kp87 at lycos.com (kevin parks)
Date: 13 Feb 2002 08:20:04 -0800
Subject: Python tutor or study groups in KOREA
Message-ID: <5e8bd451.0202130820.714d28ea@posting.google.com>

Are there any folks who use Python and are willing to help a python
newbee in Korea? I am looking to learn  a few tricks with python and
would love to have someone's brain to pick. Or if there was a group
that was not too advanced here in Seoul that meets from time to
time...

My Korean is servicable (I am from New York and am not
Korean-American, it is just a coincidence that my name is Parks (not
Park)), but maybe English is better for me as things might be too
complex for my Korean ability.

cheers,

kevin

(as this may not be of interest to others here, perhaps it is best to
e-mail me off of the comp.lang.python wire)


From dgallion1 at yahoo.com  Sun Feb 10 23:54:55 2002
From: dgallion1 at yahoo.com (Darrell)
Date: Mon, 11 Feb 2002 04:54:55 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com>
Message-ID: <z8I98.61299$QG.13109718@typhoon.nyroc.rr.com>

>>> from Numeric import *
>>> ar1=array(map(ord,list('0A')))
>>> ar2=array(map(ord,list('AB')))
>>> map(chr,maximum(ar1,ar2))
['A', 'B']
>>> maximum(ar1,ar2)
array([65, 66])
>>>

Ok, not a builtin and maybe not all that fast.
But I just read the Numeric tutorial.
--Darrell


dsavitsk wrote:

> maybe this isn't xor, but i would like something that will smush strings
> ... that is, something that would compare two strings position by position
> and return the higher (or lower i suppose) value piece in each position,
> i.e.
> 
>>>> fun('00', 'AA')
> 'AA'
> 
>>>> fun('00', 'A')
> 'A0'
> 
>>>> fun('00', '')
> '00'
> 
> i understand that this is easy to write, but i was looking for a builtin.
> does it exist?
> 
> -d




From jason at jorendorff.com  Fri Feb 15 19:48:27 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 15 Feb 2002 18:48:27 -0600
Subject: Object copying itself
In-Reply-To: <27mq6ug68esgsh7in71scb9m30phojjf3g@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEONDCAA.jason@jorendorff.com>

Dale Strickland-Clark writes:
> I want object references copied (shallow copy) but the lists need to
> be copied entirely (deep copy).
>
> Do I have to do this by hand or is there a neat way of doing it?

Well, depending on what you need, one rather unwholesome way might be:

class MyClass:
    def __init__(self, other):
        self.__dict__ = copy.deepcopy(other.__dict__)

    def __copy__(self):
        return self

    def __deepcopy__(self):
        return self

## Jason Orendorff    http://www.jorendorff.com/




From sjmachin at lexicon.net  Sun Feb 17 16:21:52 2002
From: sjmachin at lexicon.net (John Machin)
Date: 17 Feb 2002 13:21:52 -0800
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com> <m37kpcuu31.fsf@mira.informatik.hu-berlin.de> <7x4rkgw79y.fsf@ruckus.brouhaha.com>
Message-ID: <c76ff6fc.0202171321.4bd2e69c@posting.google.com>

Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7x4rkgw79y.fsf at ruckus.brouhaha.com>...
> martin at v.loewis.de (Martin v. Loewis) writes:
> > > I'm wondering how difficult it would be to add an editable (mutable)
> > > string class to Python 2.2 implemented in C under the new type/class
> > > unification scheme.
> > 
> > There is already an efficient mutable vector type, the array type,
> > which can be used to implement both mutable byte strings and mutable
> > character strings.

The array module is one of the included batteries that needs a
publicity agent. It is good value for:

(a) vectors of all-same-type numerics where the power of Numpy is not
required. Existing code using lists can often be simply tweaked to use
arrays when the vector is created -- myvector = array.array('l', [0,
1, 42]) instead of myvector = [0, 1, 42] -- and everything else just
works. Like how's this for hot-pluggable poly-whatsit:

>>> import array, bisect
>>> v = array.array('l', [10, 20, 30])
>>> v
array('l', [10, 20, 30])
>>> bisect.insort_left(v, 15)
>>> v
array('l', [10, 15, 20, 30])
>>>

(b) mutable strings are a godsend if you have to create a file of the
type that seems to be favoured by government departments:
fixed-column-size records (often several hundreds of bytes wide).

> 
> Very interesting.  I was about to answer that:
> 
>   1) There are functions to insert and delete single values into the array
>      but not multi-element slices.  
> 
>   2) I don't think the existing string functions work on byte arrays.
>      In particular, I don't think you can search for a regular expression in one.
> 
> But it turns out that both operations are supported!  I discovered
> this by looking in the arraymodule.c source to see how the buffer
> interface was implemented, and then experimenting.  The docs for the
> array module gave no clue at all that these operations were supported.
> 

Second sentence of the documentation gives a bit of a clue on your (1)
above:
"""
Arrays are sequence types and behave very much like lists, except that
the type of objects stored in them is constrained.
"""

Functions in string module may have worked on 'c' arrays in 1.5.2 but
now string.find(strg, query) is implemented as a string method --
strg.find(query)

It's a pleasant surprise that the re module supports searching etc on
'c' arrays.

> 
> Anyway, it looks like the array module may do what I need.

You'll find one nuisance when working with 'c' arrays; it doesn't
support coercing an ordinary string to an array -- you have to do it
yourself. For example:

x = array.array('c', 'abc')
x = x + 'def'

doesn't work. You have to do 

x = x  + array.array('c', 'def')

>  So I guess I'll submit a doc bug saying the above items should be described.

Certainly the re docs should be a tad more explicit about what a
"string" can be, or what behaviour(s) it must provide.


From klaatu at evertek.net  Sun Feb 24 23:22:28 2002
From: klaatu at evertek.net (Mike Dean)
Date: Sun, 24 Feb 2002 22:22:28 -0600
Subject: Correction: Re: string substitutions
In-Reply-To: <c76ff6fc.0202242012.48b17a85@posting.google.com>; from sjmachin@lexicon.net on Sun, Feb 24, 2002 at 08:12:18PM -0800
References: <mailman.1014504639.21135.python-list@python.org> <c76ff6fc.0202232228.201f829e@posting.google.com> <mailman.1014583431.3100.python-list@python.org> <c76ff6fc.0202242012.48b17a85@posting.google.com>
Message-ID: <20020224222228.B16970@evertek.net>

* John Machin <sjmachin at lexicon.net> [2002-24-02 20:12]:
> <snipped text about RE engine handling \n natively>
> This important fact is often not documented well or at all.

Is there an appropriate place to request that this be added to the next
incarnation of the documentation?  It seems too useful to be ommitted...
And does it work for any other special characters like \t?

Thanks,
Mike



From ivano.delpiccolo at degroof.be  Wed Feb 13 09:54:25 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Wed, 13 Feb 2002 15:54:25 +0100
Subject: Problem with socket
Message-ID: <3c6a7e08$0$12221$4d4efb8e@news.be.uu.net>

I tried to impement a very simple client-server application.
Hre's the output from the server part :

Traceback (most recent call last):
  File "C:\Python22\Pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "D:\server.py", line 10, in ?
    data = conn.recv(1024)
  File "<string>", line 1, in recv
  File "C:\Python22\lib\socket.py", line 159, in __getattr__
    raise error(9, 'Bad file descriptor')
error: (9, 'Bad file descriptor')


Here's my server.py :
import socket
host = ''
port = 50007
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.bind ((host,port))
s.listen(1)
conn,addr=s.accept()
print 'Connected by : ', addr
while 1:
    data = conn.recv(10)
    if not data : break
    conn.send(data)
    conn.close()

Here's my client.py :

import socket
host = '172.16.4.8'
port = 50007
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

s.connect((host,port))
s.send("123456789")
data = s.recv(10)
s.close()
print `data`


WHERE IS THE ERROR ?????????????
I don't have any error on the client side, and the client recive correctly
the data.

Thanks in advance




From niemeyer at conectiva.com  Mon Feb 18 12:40:49 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Mon, 18 Feb 2002 14:40:49 -0300
Subject: super(obj)?
In-Reply-To: <j4sn7yybzf.fsf@informatik.hu-berlin.de>
References: <mailman.1014045511.10659.python-list@python.org> <j4sn7yybzf.fsf@informatik.hu-berlin.de>
Message-ID: <20020218144049.A3464@ibook.distro.conectiva>

> It would usually be wrong. Consider
[...]
> This could does the expected: Derived2.doit call2 Derived1.doit calls
> Base.doit. If you change Derived1 to
[...]
> then self.__class__ would be Derived2, so the Derived1 would call
> itself recursively, instead of performing a super call.

Sure... you're 100% right. I forgot about the cascade effects. To make
this work, some way to know the local class would be necessary, and
this is probably not worth.

Thanks.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From peter at engcorp.com  Mon Feb 25 22:19:15 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 25 Feb 2002 22:19:15 -0500
Subject: How do I get stdout directly into a Python variable
References: <3C7ADA7E.1FFB5A4F@bellatlantic.net>
Message-ID: <3C7AFEB3.11A8723A@engcorp.com>

David Lees wrote:
> 
> I am playing with a Python script that issues system commands on linux
> and I want the stdout to go directly into a Python variable.  Right now
> I am writing to a temp file and then opening it up.  This seems like a
> hack.
> 
> I am doing something like this:
> 
> os.system('diff file1 file2 | wc > temp')
> fh = open('temp','r')
> line = fh.read()
> fh.close()

You're looking for os.popen() and it's ilk, including the
popen2 module.


From clpy at snakefarm.org  Fri Feb 22 11:06:21 2002
From: clpy at snakefarm.org (Carsten Gaebler)
Date: Fri, 22 Feb 2002 17:06:21 +0100
Subject: running system command in cgi problem
References: <a55a6d$4orlk$1@ID-79019.news.dfncis.de>
Message-ID: <3C766C7D.EE4B9A40@snakefarm.org>

Felix Seeger wrote:

> I've writen a small cgi/python script that starts a command on the server.
> This command may take about 30 minutes to finish. So my browser becomes a
> timeout.
> 
> Is there any way to run the command, and while it works send the browser
> every 30 secs a please wait string + refresh.

You could fork a child process which executes the command and let the
parent process wait for it to finish and print keepalive messages in
the meantime. This could look like the following. Note the '-u'
argument to python; it ensures that output is not buffered but sent to
the browser immediately.


#! /usr/bin/python -u
import os, time

print "Content-Type: text/plain"
print

print "Forking child"
if os.fork() == 0:
    # child process
    time.sleep(20)  # your command here
    os._exit(0)
else:
    # controlling parent process
    while 1:
        # check if child has exited, but don't block
        (pid, status) = os.waitpid(-1, os.WNOHANG)
        if pid > 0:
            print "Child finished!"
            break
        else:
            print "Please wait!"
            time.sleep(4)


Hope this helps.
cg.
-- 
Save the earth - it's our only source of chocolate.


From gerson.kurz at t-online.de  Wed Feb 27 12:41:53 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Wed, 27 Feb 2002 17:41:53 GMT
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
Message-ID: <3c7d199c.17109656@news.t-online.de>

On Wed, 27 Feb 2002 15:05:01 +0000, Chris <nospam@[127.0.0.1]> wrote:

>In the olden days I played around with BASIC
>just for fun, writing trivial games and making patterns move on the 
>screen.
>
>Can I do that sort of thing with Python?
Yup! 

>Will I be able to make a ball bounce around inside a box etc.?
Yup! I recommend you take a look at http://www.vpython.org/, get
amazed, and motivated to learn python really quickly. 

VPython is a set of add-on libraries called "3D Programming for
Ordinary Mortals" to python (so you need to have basic python
understanding first) that will allow you to create rotating balls
using opengl, and quite easily so.

There is a very intuitive first-time sample for that here:

http://www.vpython.org/vpythonprog.htm

That looks feasible, right?


From mats at laplaza.org  Mon Feb 25 14:37:01 2002
From: mats at laplaza.org (Mats Wichmann)
Date: Mon, 25 Feb 2002 19:37:01 GMT
Subject: [Tutor] unrelated bug issue
References: <mailman.1014530559.26764.python-list@python.org>
Message-ID: <3c7a9170.22360632@news.laplaza.org>

On Sun, 24 Feb 2002 01:00:59 -0500, Tim Peters <tim.one at comcast.net>
wrote:

:[Lloyd Hugh Allen]
:> So I tried Ruby for a bit, got frustrated by their ide and figured that
:> I'd try it later in a proper linux environment since the windows version
:> seems to be (I think, anyway) built on CygWin. When I uninstalled Ruby,
:> IDLE stopped working (no message--just didn't start. Command line python
:> worked fine). Turned out that this was because Ruby had put a few lines
:> in the autoexec.bat, one of which said "please use the Tcl library in
:> the Ruby directory", which no longer existed. Cleared out the
:> autoexec.bat, now Python works again. Is this a Python bug (ie, should
:> Python tell the user that the Tcl library doesn't exist or at least give
:> some sort of feedback), a Ruby bug (that the uninstaller didn't clean up
:> after itself), or both?
:
:I got a better one:  suppose you shoved a banana down your throat, then a
:carrot in each nostril.  Would it then be the banana's or the carrots' fault
:that you couldn't breathe anymore?  That's right:  it would be the carrots',
:because you didn't have any trouble breathing before they wound up in your
:nose.  Or maybe it was the banana's fault after all, since if it weren't
:there, the carrots' presence wouldn't matter.  It's a puzzler.  Just so long
:as it's clear that it's never ever your fault <wink>.

Maybe it's the fault of <insert-appropriate-entity-here> who designed
your body for not providing a robust protocol for fruits and
vegetables to register themselves when they're being inserted in an
orifice, and which other fruits and vegetables could check before they
insert into other orifices to make sure the coast (or windpipe) is
clear, as it were.

Applications that muck with autoexec.bat are likely products of
satan.org anyway.



Mats Wichmann



From shaun at laughey.com  Sun Feb 17 16:40:28 2002
From: shaun at laughey.com (Shaun Laughey)
Date: Sun, 17 Feb 2002 21:40:28 -0000
Subject: Newbie: Problems with File IO
In-Reply-To: <Xns91B8833A74178AusYourStandingInIt@130.133.1.4>
Message-ID: <000601c1b7fb$b82b3c00$0100a8c0@HOME.LAUGHEY.COM>

If you want to remember them forever (sic) then use cookies.
http://www.python.org/doc/current/lib/module-Cookie.html

If you want to remember them from post to post during the same "session"
then use a session object (a temporary cookie in effect) or encode a
unique session into the url (not as easy).

pyweblib.session contains all the functions you need.

Then you would simply

User connects
Do they have a session?
No - generate one.
Do they have a username?
No - ask for it and store in session username
and
Call the clean up function periodically to clear out old sessions.
On your page display the session username.

Remember to use try: or has_key or else you will get errors if there is
no username key in the session object.

For pyweblib see http://www.stroeder.com/pylib/PyWebLib/

Documentation isn't a strong point but the library should give you a
help.


Hope this helps
Shaun Laughey,

-----Original Message-----
From: python-list-admin at python.org [mailto:python-list-admin at python.org]
On Behalf Of Flavian Hardcastle
Sent: 17 February 2002 02:25
To: python-list at python.org
Subject: Re: Newbie: Problems with File IO

Chris Liechti <cliechti at gmx.net> wrote in 
news:Xns91B81F8828230cliechtigmxnet at 62.2.16.82:

> Flavian Hardcastle <deathtospam43423 at altavista.com> wrote in 
> news:Xns91B873A49D077AusYourStandingInIt at 210.49.20.254:
> 
>> out_file = open("log.txt", "w")
> mode 'a' does append (thats the one you want), 'w' starts a new file 
> for writing, adding a plus 'a+' or 'w+' opens for read/write
> 

Thanx! That's got me a bit further. What part of the standard Python
manual 
(the html one you get at Python.org) would that be under?

Also, if you could give me a hint as to how I might get the script to 
remember people's usernames everytime they post, that'd be good.

-- 
"You can tell whether a man is clever by his answers. You can tell
whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au
-- 
http://mail.python.org/mailman/listinfo/python-list




From starx at pacbell.net  Sat Feb  9 18:06:29 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 23:06:29 GMT
Subject: Animated graphs / wxPython
References: <a43rur$31f$1@newsg1.svr.pol.co.uk>
Message-ID: <Xns91B099B307C26RASXnewsDFE1@130.133.1.4>

Duncan Smith || Sat 09 Feb 2002 11:04:01a:

> I need to frequently redraw a graph (ie. nodes and edges) to a wxPython
> panel.  ie. I remove and/or add edges to the graph and then redraw it. 
> I am currently using wxPython ellipses and lines for drawing the nodes
> and edges. The problems are the speed of redrawing and the flickering
> as the window is redrawn.  SetClippingRegion() doesn't help a great
> deal as the clipping region is usually nearly as large as the full
> graph.  I am currently using a workaround whereby I draw edges in the
> background colour in order to remove them.  But after a number of
> redraws the graph starts to get noticeably whited out by drawing over
> existing nodes and edges. 
> 
> Does anyone know of a better approach?  Ideally I would like to be able
> to simply remove and add edges and send them to the back of the nodes
> rather than redrawing the relevant nodes each time I add an edge.  I
> have looked at a number of packages, but none seem to be able to do
> what I want. Essentially I want an animated graph that will show a user
> what the graph algorithms are up to.  (Gato won't do the job.)  Thanks
> in advance. 

Well, if you want to solve the flickering you either need to use double 
buffering or draw really, really, really fast. I'd suggest the former. I 
don't really know any other options, but you could use basic 2d opengl with 
pyopengl and pyopenglcontext's wxcontext... or something like that.

Actually, i felt like thinking a little bit, so i worked this out. This is 
a rough draft of a class that would do what you want -- just alter it's 
points and hit it's draw() for every update. You'd just have to set a 
background color (optionally) and point the 'camera' where you want it 
once, and that's about it-- just call this with an initilized GL setup 
available. Note that i haven't tested this at all, but if you want to go 
this way it should get you started :).

(Also, it's probably overkill, but it's the only graphics system i'm very 
familiar with, and it can do this job :)

class graph_rep(object):
    def __init__( self, initial_points, nodesize=.25 ):
        self.points = initial_points
        self.nodelist = self.gennode( nodesize )

    def __del__( self ):
        glDeleteLists( self.nodelist, 1 )


    def gennode( self, nodesize ):
        # Generate a simple hexagon
        glNewList( listnum, GL_COMPILE )
        glBegin( GL_POLYGON )
        
        for deg in range( 0, 360, 60 ):
            rads = math.pi * (deg/180.0)
            glVertex2d( nodesize * math.cos(rads),
                        nodesize * math.sin(rads) )

        glEnd
        glEndList()

        return listnum

    def draw( self, mode ):

        # So we don't do this for every mouseover, etc
        if mode == 1:
            glBegin( GL_LINESTRIP )
            glColor3d( 0,0,0 ) #black
            for point in self.points:
                glVertex2d( point.x, point.y )
            glEnd()

            glColor3d( 1,0,0 ) #red
            for point in self.points:
                glPushMatrix()
                glTranslated( point.x, point.y, 0 )
                glCallList( self.nodelist )
                glPopMatrix()
                
        myapp.3display.triggerRedraw()
        # or a swap-buffer command and GLUTpostRedisplay, or whatever
        # it is that you need to do to update the screen. This could
        # easily be done in whatever calls the draw function, too.
    	


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From tismer at tismer.com  Fri Feb 22 14:11:30 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 22 Feb 2002 20:11:30 +0100
Subject: Stackless: 1,000,000 Switches in 0.6 seconds
References: <mailman.1014390936.27277.python-list@python.org> <3C768F95.5050107@mxm.dk>
Message-ID: <3C7697E2.1070004@tismer.com>

Max M wrote:

> Christian Tismer wrote:
>  > Sorry, I can't hold it back.
>  >
>  > Not only that tasklets are smaller than expected, they
>  > are also really fast.
> 
> No worry ... just keep it comming ...
> 
> I'll bet that no one will mind if they become even smaller and faster.


No, I'm pretty sure that this will become the break-through
for Stackless.
I'm applying for world-dominance :-)

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From joonas at olen.to  Thu Feb 21 07:50:39 2002
From: joonas at olen.to (Joonas Paalasmaa)
Date: Thu, 21 Feb 2002 12:50:39 GMT
Subject: Parsing date strings into tuples
Message-ID: <3C74EC73.6030204@olen.to>

Question 1:
How can I parse date strings into tuples returned by localtime().
Date strings are in in some of the formats listed below.

'Wed, 20 Feb 2002 09:25:29 EET'
'Wed, Feb 20 2002 00:44:38 -0800'
'Wed, 20 Feb 2002 10:09:27 -0500 (EST)'

Question 2:
Why doesn't this work?

 >>> time.strptime('Wed, 20 Feb 2002 09:25:29 EET',
     "%a, %d %b %Y %H:%M:%S %Z")
Traceback (most recent call last):
   File "<pyshell#5>", line 1, in ?
     time.strptime('Wed, 20 Feb 2002 09:25:29 EET', "%a, %d %b %Y 
%H:%M:%S %Z")
ValueError: unconverted data remains: 'EET'



From idot at vt.edu  Fri Feb 15 01:43:49 2002
From: idot at vt.edu (Carl Banks)
Date: Fri, 15 Feb 2002 01:43:49 -0500
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com> <3C6C680D.B35E7719@cosc.canterbury.ac.nz> <jash4a.sm1.ln@127.0.0.1> <8m2p6u40evl9k04e8jq14tfvpvsov96df0@4ax.com>
Message-ID: <5nai4a.0a.ln@127.0.0.1>

Courageous wrote:
> 
>>> This sounds ingenious, but what happens when you save the
>>> file and load it back? Does it remember this information
>>> somehow, or does it get lost?
>>
>>The editor remembers this information by saving it in the text file
>>itself.
> 
> I believe that you're answering a different question than what
> he asked.

Yeah.  Although the question was about vim, my mind was still thinking
Emacs from a few posts ago.  Emacs scans the line(s) above to
determine how to unindent, i.e., all of the unindent information is
saved in the file.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From aries at bitstream.net  Mon Feb 18 09:15:57 2002
From: aries at bitstream.net (mylinuxboxroot)
Date: Mon, 18 Feb 2002 08:15:57 -0600
Subject: help - inserting values into database tables
References: <mailman.1013981787.10266.python-list@python.org> <mailman.1014010317.7276.python-list@python.org>
Message-ID: <3c710b01$0$79561$6536d21b@news.bitstream.net>

Hey thanks,
This is great, and I have the project now that it will be developed on a
windows
server so I guess I don't need to worry about going from linux to the
MS-Access
database.  I will develop it right on the server using the win32all
distribution that
you have suggested.  It will be somewhat easier now I think. I should be
able to use
the modules dbi and odbc windows stuff.
    I have not tried python on windows yet though, do you get an output
screen
    terminal like using it on linux ???

Well I will play with it too.
thanks much,
Joe F.


chajadan <python at chajadan.net> wrote in message
news:mailman.1014010317.7276.python-list at python.org...
> My apologies, odbc is part of the win32all distribution, and is therefore
> available only under Windows.
>
> However, the module you found, or other similar modules, are bound to be a
> viable solution for you, database interaction being so widespread as it
is.
>
> I know of another possible solution, by the creators of mxTime and
whatnot,
> available in this general location:
> http://www.lemburg.com/files/python/eGenix-mx-Extensions.html
>
> --chajadan
>
> At 04:15 PM 2/17/02 -0800, Paul Rubin wrote:
> >chajadan <python at chajadan.net> writes:
> >
> > > Yes, this is altogether possible.
> > >
> > > There is a module include in the main distribution called odbc, and
> > > one called dbi.
> >
> >Whaaa?  That's the first I've heard of this.  In Python 2.2 under Linux,
> >"import odbc" and "import dbi" both throw ImportError.
> >
> >There's an ODBC module in Vaults of Parnassus that looks pretty good, but
> >I haven't
> >tried it yet.  Is that what you meant?
> >--
> >http://mail.python.org/mailman/listinfo/python-list
>
>




From jkraska at san.rr.com  Sun Feb 10 13:15:45 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 10 Feb 2002 18:15:45 GMT
Subject: RFC -- Hygienic Macros in Python
References: <mailman.1013231420.29837.python-list@python.org> <slrna6bfrc.o7n.philh@comuno.freeserve.co.uk>
Message-ID: <iudd6uccfj973u1vdb8ta28bfudgaqs820@4ax.com>

>>Well, I don't think that's particularly realistic code. 
>
>Nor do I, particularly. I can't offhand think of anything where I'd
>particularly want to use macros in Python, except perhaps to define my
>own control structures.

A toy example or referring to control structures are fine examples.

One doesn't have to have a useful example per se; first-class macro
systems are usually used by those who wish to craft a domain-specific
dialect of a language because in so doing they will be able to more
easily and succinctly represent their problems in the language they
are working in. This is a somewhat esoteric application; it's one
of the reasons that many AI researchers continue to use Lisp, Scheme,
et al.

> local lock
> Where the "local lock" declaration means something like: in the
> current namespace create a local variable name that doesn't clash with
> anything else.

That's a good idea that I'll think about. Surely it's easy enough
syntactically, in particular because the dialect of Python that I'm
creating has optional declarations (to support strict functions and
strict classes).

If a macro weren't hygienic by default in my system, I'd probably do
something like you suggest. Another alternative would be to require
a first-class closure form and use that when you wanted hygienic
behavior, ala:

defmacro incf3(j):
	closure(j):
		i=3
		j=j+i

In this syntax, i is created in the protected closure namespace, but
j (as indicated by the parameter to the block closure) is promulgated
_through_ into the closure, and moved by reference. This continues to
behave cleanly.

I could, I suppose, also invert it, and allow blocks of unhygienic space
with a block keyword of some kind.

C//



From philh at comuno.freeserve.co.uk  Thu Feb  7 07:03:16 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 12:03:16 +0000
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1731.803T2524T145894threeseas@earthlink.net>
Message-ID: <slrna64r84.ahr.philh@comuno.freeserve.co.uk>

On Thu, 07 Feb 2002 04:28:54 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:
>
>
>>That is putting the cart in front of the horse.  We cannot create a
>>tool to solve a problem until we know what the problem is.
>
>That's not true. Not only has the problem been identified but it has been
>pointed out that it is not a simple one time problem/solution matter.
>
>The symbols of mathmatics weren't designed to solve one problem, they
>were designed to solve all problems in that class of problems.
>
>I have pointed out that programming is the act of automating complexity
>that is made up of simpler things, in order to make use of complexity
>simple for the users of that complexity.
>
>The problems is that there is no general automation tool oriented to all
>users. Yet the essence of programming boils down to two deminsional bit
>flipping.
>
>I've identified and defined the tool and it's natural components.

So, when can we expect the finished version, Tim?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jeff at ccvcorp.com  Thu Feb 28 17:07:20 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 14:07:20 -0800
Subject: Pointers for solving windows GUI problem.
References: <f72dac0d.0202280747.2360c608@posting.google.com> <Pine.GSU.4.21.0202281241260.23516-100000@garcia.efn.org>
Message-ID: <3C7EAA17.D1CDCAEF@ccvcorp.com>


Spencer Ernest Doidge wrote:

> Like you, I am looking for a pure Python/WinGUI solution. AS I just
> learned from this group, http://www.activestate.com has Visual Python, but
> you have to have Visual Studio .NET for that. That means many $$$ :-(

You do not need Visual Python in order to use the Win32 gui stuff from
ActiveState.  The PythonWin IDE is, in fact, a Python-wrapped MFC
application.  Everything you need to do the same comes with it, as part of the
ActivePython installation (or, I believe, in the win32all extensions).  I've
never *done* this myself, because I find that using wxPython is a higher level
of abstraction than what I remember of C++ Win/MFC programming, and is thus
easier to do -- and has the added benefit of being cross-platform, should I
ever have reason to care about that.  But you can certainly do wrapped-MFC
programming from PythonWin.

Jeff Shannon
Technician/Programmer
Credit International




From jolsen at mailme.dk  Tue Feb 26 13:10:53 2002
From: jolsen at mailme.dk (Jesper Olsen)
Date: 26 Feb 2002 10:10:53 -0800
Subject: Numeric arrays - initialising with binary data
References: <cf0ad9fb.0202250446.3df07a08@posting.google.com> <a5dc4f$6dr9c$1@ID-11957.news.dfncis.de>
Message-ID: <cf0ad9fb.0202261010.4c778108@posting.google.com>

"Emile van Sebille" <emile at fenx.com> wrote in message news:<a5dc4f$6dr9c$1 at ID-11957.news.dfncis.de>...
> "Jesper Olsen" <jolsen at mailme.dk> wrote in message
> news:cf0ad9fb.0202250446.3df07a08 at posting.google.com...
> > Hi,
> >
> > The build in array type has the methods
> >
> > array.tostring()
> > array.fromstring()
> >
> > which make it easy to initialise arrays with binary data
> > read from a file.
> >
> > The Numeric array, however, apperantly only has the tostring method.
> >
> 
> Numeric 20.1.0 has fromstring.

You are right, thanks.

Numeric 21 is out now by the way.

I missed fromstring(), because it is not a method on the array object,
as in the build in type.

Which documentation are you using?

The most recent I can find is this

http://www.pfdubois.com/numpy/html2/numpy-9.html#pgfId-36538

I hate to complain so much :-), but why do they not distribute it with
the module itself. Also, I think it should be updated.

Cheers
Jesper


From phr-n2002a at nightsong.com  Wed Feb 27 23:07:55 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 20:07:55 -0800
Subject: Rotor in Jython?
References: <mailman.1014818684.32712.python-list@python.org>
Message-ID: <7x8z9e45b8.fsf@ruckus.brouhaha.com>

"Scherer, Bill" <Bill.Scherer at VerizonWireless.com> writes:
> I've a need to perform rotor decryption in Jython.  A quick 
> search didn't reveal anything to help.  I've begun coding it in 
> Python from rotormodule.c. My C knowledge is not strong, so there 
> are spots that are surely not right.

I hope you're not the rotor module for computer security.  That cipher
is nowhere near as safe to use as a modern one like AES.

If you just have some old rotor-encrypted files you have to decrypt,
I'd say get to a CPython environment and use the existing rotor module
to decrypt them and be done with it.

Don't use the rotor module for new traffic that needs good security.


From fumingw at yahoo.com  Wed Feb 20 14:04:47 2002
From: fumingw at yahoo.com (Fuming Wang)
Date: 20 Feb 2002 11:04:47 -0800
Subject: Python 2.2 on windows 98
Message-ID: <b0fd9226.0202201104.2c6a220@posting.google.com>

Hi,

I am having trouble using the arrow key to recall commands in console
interactive mode.  I even installed Cygwin and start Python from there
but still got the same effect.  Is there anything I can do to get
arrow key working?

Thanks!

Fuming Wang


From chrishbarker at attbi.com  Fri Feb 15 16:59:47 2002
From: chrishbarker at attbi.com (Chris Barker)
Date: Fri, 15 Feb 2002 13:59:47 -0800
Subject: MemoryError using NumPy
References: <mailman.1013804009.30807.python-list@python.org>
Message-ID: <3C6D84D3.48293AD5@attbi.com>

Gustavo Cordova wrote:
> 
> > > I suspect that memory has fragmented such that a sufficiently
> > > large contiguous block is not available to set up the array.
> > > Is this the likely cause, and if so is there a way to un-fragment
> > > memory? Or might there be an alternative to Numeric.take that
> > > gets to the same place?
> 
> Er... This isn't an Amiga, you know. :-)
> 
> Really. Under UNIX, or anything "Unixy", where there's
> a memory management unit on the CPU and a virtual memory
> subsystem in the OS, memory fragmentation really doesn't
> mean much, because the OS remaps the memory layout
> so that each and every process receives (virtually) the
> same memory space, in a contiguous manner.
> 
> I remember having to reboot my Ami because I didn't have
> enought ram left, when in fact it was all fragmented.
> Virtual memory would have been good then.  :-)
> 
> (regardless, they rocked)
> 
> > I think you have two functions in one line that create a copy (take()
> > and astype()). I have never gotten an answer as to when CPython frees
> > the memory for temporary variables in one statement, but if you can
> > delete myDataArray before calling astype(), you might save something
> > there.
> 
> The problem is the C library, seems like it doesn't ever return
> allocated-and-then-freed memory to the OS, so, CPython never
> returns unused memory to the OS neither.

OK, so the Python process keeps the memory. Does the memory that the
Python process holds get fragmented? If so, same problem. If not, then
what is going on?

Also, a few years back there was a discussion on the MATLAB newsgroup
about a similar problem, and The Mathworks claimed that it was not
possible to return memory to the os once a process has used it. This was
the case on Linux and I'm pretty sure, all the other *nixs that they
supported. Perhaps it's a library issue, but is there a different one(s)
that we all should be using to compile Python?

-Chris




-- 
Christopher Barker,
Ph.D.                                                           
ChrisHBarker at attbi.net                ---           ---           ---
                                     ---@@       -----@@       -----@@
                                   ------@@@     ------@@@     ------@@@
Oil Spill Modeling                ------   @    ------   @   ------   @
Water Resources Engineering       -------      ---------     --------    
Coastal and Fluvial Hydrodynamics --------------------------------------
------------------------------------------------------------------------


From phr-n2002a at nightsong.com  Thu Feb  7 06:44:40 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 07 Feb 2002 03:44:40 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <e840346c.0202070321.452d405d@posting.google.com>
Message-ID: <7xr8nxy0rb.fsf@ruckus.brouhaha.com>

kalath at lycos.com (Mark Brady) writes:
> As python lover trapped developing in Java I see this a lot. Sun's
> marketing has convinced a lot middle/upper management that Java/J2EE
> is the only viable option to C/C++ for large server side systems.
> Actually the endorsement and support from IBM/Oracle/HP/etc. for Java
> is what seems to have given Java the 'respectability' it currently
> enjoys as a C/C++ alternative. On a personal note Java/J2EE while
> keeping me employed is driving me insane. Thank god for Pygame
> (www.pygame.org). Java feeds the body but Python feeds the soul!

As much as I like Python, I don't see it working so well in the kind
of large, real-world projects that Java works pretty well in:

  - big codebase (>200K LOC) being developed/maintained by lots (say
    20 or more) of people

  - Program stays around for years but developers come and go.
    Average turnover is maybe 1 programmer per month.  So there
    are always people hacking on the codebase who aren't that familiar
    with it.

  - The programmers are what you get through normal industry-style
    recruiting processes, i.e. they are mostly pretty good, but not great.

In this environment, C is an unholy nightmare, C++ is somewhat better,
but Java seems to do a much better job than either of keeping
complexity under control and avoiding crazy application bugs.

I don't know of any Python projects on such a scale.  The biggest ones
I know of (e.g. Zope) are smaller and have extremely good programmers
rather than just ordinary programmers.  So they don't face the problems
that typical Java projects do.

I don't know what would happen if someone tried a big project in
Python (Zope is medium sized).  Maybe it would work ok, maybe not.

I do believe a native-code compiler would make Python a lot more
interesting for many purposes.


From logiplexsoftware at earthlink.net  Mon Feb 25 18:26:53 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 25 Feb 2002 15:26:53 -0800
Subject: dlg.Destroy() behaviour
In-Reply-To: <a5e561$q8v$1@news5.svr.pol.co.uk>
References: <a5e561$q8v$1@news5.svr.pol.co.uk>
Message-ID: <20020225152653.03935d89.logiplexsoftware@earthlink.net>

On Mon, 25 Feb 2002 19:58:42 -0000
Duncan Smith wrote:

> Just a quick question about whether or not I need to destroy dialogues
> (wxPython) with dlg.Destroy().  When creating a number of
wxTextEntryDialogs
> in succession only the first dialogue gets the focus.  I need a mouse
click
> to set the focus on subsequent dialogues.  However, if I remove the usual
> 'dlg.Destroy()', then subsequent dialogues do get the focus.  Is this
> advisable?  If not, what is the neatest way of setting the focus on the
> subsequent dialogues?  Thanks in advance.

I don't know if this works (never tried it on a dialog), but dlg.SetFocus()
?

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From stauffer at swarthmore.edu  Tue Feb  5 11:06:32 2002
From: stauffer at swarthmore.edu (Glenn Stauffer)
Date: Tue, 5 Feb 2002 11:06:32 -0500
Subject: Python 2.2 dynamic library problem on Compaq Tru64
Message-ID: <20020205160719.2D8CF17E8D@juniper.its.swarthmore.edu>

This is a shot in the dark, but maybe someone here has a clue to this error 
(my Unix admin doesn't).

I compiled and installed Python 2.2 on a Compaq server running Tru64 v5.0.   
I am using cx_Oracle to access an Oracle 8.1.6.2 database.  Running my 
scripts at the unix prompt, all is well.  When I run them through cron, I get 
this error:

> Traceback (most recent call last):
> ? File "/usr/users/stauffer/python/dbamon/dbchk.py", line 3, in ?
> ? ? import cx_Oracle, os, sys, smtplib
> ImportError: dlopen: cannot load
> /usr/local/lib/python2.2/site-packages/cx_Oracle.so

I've copied the cx_Oracle.so into a variety of locations - the local script 
directory, site-packages, lib-dynload - with the same results.  

Any clues about how to get this working would be greatly appreciated.

Thanks, Glenn Stauffer



From janez.jere at void.si  Sat Feb  2 18:23:29 2002
From: janez.jere at void.si (Janez Jere)
Date: Sun, 03 Feb 2002 00:23:29 +0100
Subject: win32 AddFontResource RemoveFontResource
Message-ID: <3C5C74F1.2010106@void.si>

Hi!

How to install a font with python for windows? I read about 
AddFontResource and RemoveFontResource win32 API functions which should 
do it. Is there another way to do it, becase win3api is without direct 
wrapper for them.

Janez



From nospam at [127.0.0.1]  Thu Feb 28 03:56:21 2002
From: nospam at [127.0.0.1] (Chris)
Date: Thu, 28 Feb 2002 08:56:21 +0000
Subject: Extremely Newbie wants to program and pleads to be pointed in the right direction
References: <UrIWUMDdWPf8EwAj@[127.0.0.1]>
 <a0up7usn1v48q14i1eopb5n8mfdthfpqkm@4ax.com>
 <54abe19d.0202271315.4e699ed0@posting.google.com>
Message-ID: <b7yVwiC1Cff8EwBv@[127.0.0.1]>

In article <54abe19d.0202271315.4e699ed0 at posting.google.com>, Anders 
Schneiderman <schneida at seiu.org> writes
>I'd choose Python over Java; it's easier to learn, and it's a much less 
>wordy language.  Every once in a while I'll see something new in the 
>Java world I want to check out, and every time I do, I'm reminded of 
>how much I like Python.  Java might be a good second language.
>
>One more thing.  One of the reason it's easy to learn Python is that 
>when you run into a really tricky problem you can't solve, if you post 
>it to the newsgroup, someone will give you an answer.  I just posted 
>something an hour ago, and already a really nice guy sent me the answer!

Thank you.  These comments are very useful.
I had already had a stab at Java and found it too much of a stretch to 
start with.
Can you do ordinary 2D graphics stuff without VPython - simply using the 
ordinary Python?
-- 
Chris


From iddwb at moroni.pp.asu.edu  Fri Feb  1 15:57:14 2002
From: iddwb at moroni.pp.asu.edu (David Bear)
Date: 01 Feb 2002 13:57:14 -0700
Subject: reading sendmail alaises
References: <m34rl2jh7m.fsf@moroni.pp.asu.edu> <864rl2p09z.fsf@k0dez.scriptkiddie.net>
Message-ID: <m31yg4diol.fsf@moroni.pp.asu.edu>

>Or, if that's still too fiddly for you, you could spawn out to the
>praliases program.

what is the 'praliases' program? 
-- 
David Bear
College of Public Programs/ASU
480-965-8257
...the way is like water, going where nobody wants it to go


From bkelley at wi.mit.edu  Fri Feb 15 16:40:02 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 15 Feb 2002 16:40:02 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org> <a4j1h7$kker$1@ID-98166.news.dfncis.de> <a4jp10$r5pc$1@ID-98166.news.dfncis.de>
Message-ID: <3C6D8032.2050306@wi.mit.edu>

MDK wrote:


> Maybe cPickle won't work for me:
> 
>   File "C:\Python22\!MyTest\pipe.py", line 88, in put
>     pic = cPickle.dumps(object,1)
> cPickle.UnpickleableError: Cannot pickle <type 'array'> objects


Array's aren't inherently picklable.  You might have to do something 
like this.

text = pickle.dumps((object.typecode, object.tolist()))

and load with

object = apply(array.array, pickle.loads(text))

Of course this completely removes the reason to use pickle though since 
you need to know the datatype in the first place...

It also looks like your SOL when trying to subclass from array.array or 
array.ArrayType for including pickle cababilities.  The array module is 
really confusing in this respect, array.array is a function that returns 
an array.ArrayType which thinks of itself as <type 'array.array'> when 
it really is <type 'array.ArrayType'>.

Numeric arrays in contrast are picklable.

import Numeric
object = Numeric.array([1,2,3], Numeric.Float)
text = pickle.dumps(object)


Brian Kelley





From gerson.kurz at t-online.de  Thu Feb 21 23:20:00 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Fri, 22 Feb 2002 04:20:00 GMT
Subject: Unintelligible Lambda Tricks
References: <mailman.1014338076.13501.python-list@python.org>
Message-ID: <3c75c52d.86345109@news.isar.de>

On Thu, 21 Feb 2002 19:44:34 -0800, chajadan <python at chajadan.net>
wrote:

>data.__setitem__(4,42)
>
>def swap(a,x,y):
>a[x] = (a[x], a[y])
>a[y] = a[x][0]
>a[x] = a[x][1]
>You can write this as a lambda expression like this:
>
>>swap = lambda a,x,y:(lambda
>>f=a.__setitem__:(f(x,(a[x],a[y])),f(y,a[x][0]),f(x,a[x][1])))()
>
>I don't understand why __setitem__ takes 2 arguments, but in the lambda () 
>there are 3...

Well, it doesn't? Its a matter of bracket counting. Here is what the
expression does:

swap = lambda a,x,y:(lambda
f=a.__setitem__:(f(x,(a[x],a[y])),f(y,a[x][0]),f(x,a[x][1])))()

Take out these parts:

part_1 = f(x,(a[x],a[y]))
part_2 = f(y,a[x][0])
part_3 = f(x,a[x][1])

Those are our three swap statements from above. Note that part_1 is a
function that takes a tuple of *two* elements: x, and a tuple. Its not
three elements! This is just like the code in the "regular" swap()
function above. 

So you can think of the lambda as

swap = lambda a,x,y:(lambda f=a.__setitem__:(part_1,part_2,part_3))()

(this is not valid python, its just to show the structure of the
code). Next, the inner lambda function is just a helper. By defining

lambda f=a.__setitem__:(expression that uses f)

I define a function that can use "f" in any place it would have to
write a.__setitem__. Writing three times "f" is shorter than writing
three times "a.__setitem__". To invoke that expression, include it in
brackets and add a (), as in 

(function-declaration)()

so, mystery solved, I hope ;)






From nhodgson at bigpond.net.au  Sun Feb  3 06:58:05 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Sun, 03 Feb 2002 11:58:05 GMT
Subject: SilverCity 0.1.0
References: <mailman.1012684745.32090.python-list@python.org>
Message-ID: <hB978.33300$Ni2.188457@news-server.bigpond.net.au>

Brian Quinlan:
> SilverCity is a Python extension that allows you to retrieve lexical
> information about different languages (e.g. C/C++, HTML, Python).
> Amongst other things, you can use it to style source code into styled
> HTML. It is based on Neil Hodgson's excellent Scintilla project.
>
> I'm pretty far from release (at least another week :-) so I'd really
> like a sanity check for the API. Any feedback would be greatly
> appreciated.

   Not sure which level you are interested in. The _SilverCity interface
looks straightforward. There could be a script for generating language
modules or outlines of language modules from the constants: generate("CPP",
"SCE_C_*"). BTW, the verbatim lexical class that got the "XXX what is this?"
comment is for C#'s multiline verbatim strings: @"Just like
""Python"" uses
triple quotes for long texts."

   Neil





From jcromwell at ciena.com  Sat Feb  2 18:21:36 2002
From: jcromwell at ciena.com (Cromwell, Jeremy)
Date: Sat, 2 Feb 2002 15:21:36 -0800 
Subject: hex to normal chars?
Message-ID: <FB77A8A317F5D31199F1009027DE4A8701B001E4@wntcsdexg01.csd.ciena.com>

I think you want this:
>>> chr(int("41", 16))
'A'

int() can take a string and a base, and returns an int
chr() takes an int and returns the ASCII character
-Jeremy Cromwell


-----Original Message-----
From: maximilianscherr [mailto:MaximilianScherr at T-Online.de]
Sent: Saturday, February 02, 2002 2:47 PM
To: python-list at cwi.nl
Subject: hex to normal chars?


how can i transfer hex numbers to normal chars, like "41" to "A" etc.?


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



From gh_pythonlist at gmx.de  Fri Feb  8 13:39:09 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Fri, 8 Feb 2002 19:39:09 +0100
Subject: [OT] PPT reader?
In-Reply-To: <15459.60914.405905.301501@12-248-41-177.client.attbi.com>
References: <15459.60914.405905.301501@12-248-41-177.client.attbi.com>
Message-ID: <20020208183908.GC1206@lilith.hqd-internal>

Le 08/02/02 ? 09:25, Skip Montanaro ?crivit:
> 
> Guido posted a bunch of talk slides to the Python website in PowerPoint
> format.  I understand his reasons for not creating HTML versions (mostly
> because the HTML generated by PPT sucks).  Still, that leaves people without
> access to PPT sort of out in the cold.  What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT slides
> to other, more accessible, formats?  If we can get a reasonable (but fairly
> short) list of tools, I can probably see they get referenced on the relevant
> page of the Python website.

Well, reading Guido's post, I tried to open one of his powerpoint slides
with kpresenter (of the KOffice suite). It was a nice way to test the
Linux OOM (out of memory) killer. At 1500 MB process size, Linux killed
kpresenter and an unrelated daemon :-) But then, I wasn't that surprised
- KOffice never was stable for me.

OpenOffice worked just great, though. It can also save PowerPoint, which
few Open Source tools can.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.3 ?C      Wind: 2.5 m/s



From chris at cmb-enterprises.com  Mon Feb 25 14:53:01 2002
From: chris at cmb-enterprises.com (Chris Dutton)
Date: Mon, 25 Feb 2002 19:53:01 -0000
Subject: Simple regexp problem
References: <3C75289B.8050504@nyu.edu> <ueljelpvz.fsf@python.net>
Message-ID: <20020225145225834-0500@news.fltg.net>

In <ueljelpvz.fsf at python.net> Michael Hudson <mwh at python.net> wrote:
>Shane Hoversten <srh232 at nyu.edu> writes:
>
>> Hi all,
>> 
>> 	This question has gotta be real simple.  I want to remove
>> preceeding whitespace from a string.  Seems like I should be able to
>> say (using another example to make it easier to see):
>> 
>> re.sub("^1", "", "11111junk")
>>re.sub("^1*", "", "11111junk")

Or better yet:
re.sub("^1+", "", "11111junk")


From deathtospam43423 at altavista.com  Mon Feb 11 16:03:09 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 11 Feb 2002 21:03:09 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com>
Message-ID: <Xns91B34DA3C4E54AusYourStandingInIt@130.133.1.4>

only at example.com wrote in 
news:oNz98.2509$q95.1624800457 at newssvr21.news.prodigy.com:

> Okay... you're a little confused about where the data from your form is
> going. The data will not be STORED in script.cgi ...  script.cgi is
> supposed to be a program (in this case, a python program) that processes
> all of the VARIABLES (e.g. name, etc) from your form. Take a look at the
> tutorial Devshed has and you will begin to understand how cgi works. 
> 
> http://www.devshed.com/Server_Side/Python/CGI/page1.html
> 
> Hope this helps.
> 

Thanx! You hit the nail right on the head. I was under the false impression 
that the script was merely some sort of blank file that the browser 
deposits raw data into.

I will work through that tutorial. If I have any more trouble I'll post em 
here if that's ok.

> 
> In article <mailman.1013350037.30507.python-list at python.org>, "Chris
> Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote:
> 
>> ----- Original Message -----
>> From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>
>> 
>> 
>>> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
>>> news:mailman.1013316023.10782.python-list at python.org: 
>>> 
>>> > "PERL type CGI's" should be read "scripts".  You can't run Python,
>>> > Perl, Ruby,
>>> > Lua, etc. from this particular server.
>>> > 
>>> > Try Xitami.  It is quite compatible with Python.  http://xitami.com
>>> > 
>>> Thanx. I'm downloading it as we speak.
>> 
>> I have used it a little; Steve Holden wrote about it in his book which I
>> just finished reading.
>>  
>>> So with this server, will I be able to just send info direct from my
>>> browser form to a file? If I can do that, then I'm up and running. 
>> 
>> I'm not sure what you are saying here.  You don't send info "to a file,"
>> you send it to a CGI script.  Your script must handle the data from
>> there. 
>> 
>>> [snip]
>>> > 
>>> > The form looks good, but where is script.cgi?
>>> > 
>>> In the same directory as the html form, cgi-bin.
>> 
>> Ah... what I meant was, that you should *post* the source to script.cgi.
>>  Anyway, if your form is HTML it is bad form to have it in the cgi-bin
>>  folder. 
>> 
>> GENERALLY (though it's not real safe to say this) CGI scripts can be
>> stored in the cgi-bin folder, or in the htdocs/html/webroot folder; this
>> depends on webserver settings.  The Xitami default allows CGI scripts
>> only in the cgi-bin folder, but you can define that cgi-bin is actually
>> the same as webroot.  Xitami recognizes executable files by extension on
>> Windows, so your script should likely be called script.py (assuming it
>> is Python).
>> 
>> Gah.  So.  Put script.py in the Xitami cgi-bin folder, and your form in
>> the Xitami webpages folder, and if you are using the default
>> configuration, you should be in good shape.  Start Xitami and then surf
>> to http://localhost/yourform.html (whatever you called it) and when you
>> click the Submit button, the form data is sent to
>> http://localhost/cgi-bin/script.py (which should be in the action=
>> attribute of the form tag, perhaps abbreviated as /cgi-bin/script.py).
>> The script receives the data, does whatever, and then must send a new
>> page to the web browser... this is why I want to see your script.
>> 
>> 
>> 
>>



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From lac at strakt.com  Wed Feb 27 08:16:06 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 27 Feb 2002 14:16:06 +0100
Subject: Tkinter list boxes, selection emphasis, and the keyboard focus 
In-Reply-To: Message from Tokomar <no@spam.com> 
   of "Tue, 26 Feb 2002 19:21:48 GMT." <gfRe8.6453$c15.3356020763@newssvr21.news.prodigy.com> 
References: <K0Qe8.48006$BA5.2207030344@newssvr14.news.prodigy.com>  <gfRe8.6453$c15.3356020763@newssvr21.news.prodigy.com> 
Message-ID: <200202271316.g1RDG6tB026202@ratthing-b246.strakt.com>

Make your listboxes like this:
lb = Listbox(exportselection=0)

http://www.pythonware.com/library/tkinter/introduction/x5453-patterns.htm

I don't understand all of your problem so I don't know if this will fix
all of it or only some of it.

Laura Creighton



From jim.vickroy at noaa.gov  Mon Feb  4 15:09:32 2002
From: jim.vickroy at noaa.gov (jim.vickroy)
Date: Mon, 04 Feb 2002 13:09:32 -0700
Subject: heres a dumb question....
References: <a3mp6v$u04$1@wiscnews.wiscnet.net>
Message-ID: <3C5EEA7C.F6380391@noaa.gov>

How about the raw_input() function?

> x = raw_input('Enter an integer: ') # Enter 123
> x
123

The prompt string (e.g., 'Enter an integer: ') is optional.


Jeremy Bush wrote:

> Hello, ive just started fooling around with python, I know C++ pretty well,
> and it looks familiar.  The only problem im having is a cant seem to figure
> out how to get keyboard input...ive read the whole readme that came with it,
> the tutorial, everything.  I just want to know how to do a simple "cin >>
> X;"  Can anyone help?



From jason at jorendorff.com  Mon Feb 25 06:11:41 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 05:11:41 -0600
Subject: Editable strings anyone?
In-Reply-To: <3C742016.9030507@wi.mit.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLOEOODEAA.jason@jorendorff.com>

Brian Kelley wrote:
> Jason Orendorff wrote:
> > Brian Kelley wrote:
> > > What would it take to make array.array or array.ArrayType an
acceptable
> > > base type for subclassing?
> > >
> >
> > Wow, you're right, this is a mess.  I will put together a patch
> > for it.
>
> Cool.  I'll buy you a beer at the next python conference :)

I'll make a note of that.  The patch is up, but it has a flaw or two
that I need to fix.

http://sourceforge.net/tracker/index.php?func=detail&aid=520694&group_id=547
0&atid=305470

Paul's "gap" feature, I fear, will have to wait for someone a bit more
motivated than I.

## Jason Orendorff    http://www.jorendorff.com/




From mgash at trisystems.co.uk  Wed Feb 20 11:14:04 2002
From: mgash at trisystems.co.uk (Mark Gash)
Date: 20 Feb 2002 08:14:04 -0800
Subject: Adding GUI to application
Message-ID: <9120a667.0202200814.1bd679cb@posting.google.com>

Hi,

I have written an application that will check the percentage free on
the local harddisks and compare against a predetermined level.
The application is written purely for Win32 and uses the performance
monitor modules to extract the disk information.
The level is set by using a parameter within an ini file. 
If the level is exceeded, a mail is sent to the administrator. 
Once the comparison is made (and mail sent as required), the process
then goes to 'sleep' for 2 hours.
The process then wakes up again, and the comparison repeats.

The application is confirmed as working as required.

I wish to amend the application so that (i) it runs as before, (ii)
there is an icon in the taskbar, via which the application can be
stopped.

As a newbie to programming, I am finding this task difficult, to say
the least. The problem that I am having, is that by using the
win32gui_taksbar.py from the win32/demos module, I can get the icon
down in the taskbar - but as the diskchecking application 'goes to
sleep' rather than completes/exits, the control is never returned to
the win32gui_taskbar to await the next interupt.

I feel that this change will have to be multi-threaded, but as I have
said, I am a newbie to programming and am 'drowning' whilst trying to
complete this.

Full code listing can be sent/attached if required.

Many Thanks for your help in advance.

Mark Gash
mgash at trisystems.co.uk


From dvass at felis.uni-freiburg.de  Thu Feb 14 06:05:45 2002
From: dvass at felis.uni-freiburg.de (=?iso-8859-1?Q?J=F6rg=20W=F6lke?=)
Date: Thu, 14 Feb 2002 12:05:45 +0100
Subject: spawn and standard output
References: <ce0b6758.0202131543.1707e80c@posting.google.com> <3c6b7581$0$12226$4d4efb8e@news.be.uu.net>
Message-ID: <3C6B9A09.9CDCDAF3@felis.uni-freiburg.de>


DelPiccolo Ivano schrieb:
 
> You can also redirect your command to a specific file and read it.
> example :
> os.system("ls -la>foo.txt")
> after you open the file foo.txt and you read it.
> 
> The problem of using that method, stdout and stdin are in the same file.
                                   
s/stdin/stderr

> When you read your file you have to do treatment to detect if the
> information is coming from stdout or stderr.

So why not:

os.system("ls -la > foo.txt 2>foo.error.txt")

This won't work with a cshell.

HTH,HAND
and Greetings J"o!

-- 
$ make LOVE
Make: Don't know how to make LOVE.  Stop.
$


From aahz at panix.com  Fri Feb  1 10:41:47 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 07:41:47 -0800
Subject: An improvement to Queue
References: <mailman.1012515655.8263.python-list@python.org> <Xns91A85CDCAE3FBmichaelrcpcouk@194.238.50.13>
Message-ID: <a3ecvr$si7$1@panix2.panix.com>

In article <Xns91A85CDCAE3FBmichaelrcpcouk at 194.238.50.13>,
Michael Abbott  <michael at rcp.co.uk> wrote:
>"Tim Peters" <tim.one at home.com> wrote in
>news:mailman.1012515655.8263.python-list at python.org: 
>> 
>> It *sounds* like you're using Queue to manipulate items at too low a
>> level of granularity.  If you created a class to encapsulate the notion
>> of "one round of work", then the consumer could pull off "one complete
>> round of work" per Queue.get(), and you wouldn't need other synch
>> gimmicks (or other complications) to communicate where "one round of
>> work's" boundaries lie. 
>
>The problem is that when the consumer gets around to performing its round 
>of work, rather than just performing the "jobs" in sequence and in 
>isolation, in fact the the consumer is in a position to merge jobs together 
>into a single action.
>
>Think of a cursor addressable terminal the wrong end of a serial line (this 
>is close to my application): sending characters down the serial line is the 
>main limiting resource, and the "consumer" in this case is in the business 
>of formatting up an incremental update and sending it on.
>
>However, changes to the display are coming in all the time, too fast to 
>display, on a separate thread.  I therefore want to accumulate these 
>changes and when there is room on the serial line send the entire batch.  
>The consumer should therefore be spending most of its time waiting for the 
>serial line to become free, and then formatting up a batch of updates.
>
>Arguably this accumulation should be done in the generating thread, but I 
>don't really wish to do it that way in this particular case.  In 
>particular, the model I propose makes synchronisation almost trivial, which 
>is I think quite an important consideration!

Well, you're adding complexity to the Python threading model, which is
already difficult enough for many people to understand.  Your getall()
idea *is* a good one, no argument there.  You should either do the
batching on the consumer side (if it's busy, don't do a get(); when it's
free, do a getall() and format) or create an intermediary pipeline to do
just the collapsing.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From cimarron+google at taylors.org  Thu Feb 28 14:47:58 2002
From: cimarron+google at taylors.org (Cimarron Taylor)
Date: 28 Feb 2002 11:47:58 -0800
Subject: Need advice for obfuscating Python code
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com> <29e28c51.0202270014.61dbd1d0@posting.google.com> <RUcf8.13561$TB2.72289@rwcrnsc52.ops.asp.att.net>
Message-ID: <29e28c51.0202281147.4b690b5b@posting.google.com>

"Alexander Gavrilov" <gavrilov at attbi.com> wrote in message news:<RUcf8.13561$TB2.72289 at rwcrnsc52.ops.asp.att.net>...
> Are you working in Microsoft, aren't you?
> 
> Alexander

Aren't we all?  :-)

Cim


From mcherm at destiny.com  Fri Feb  8 09:42:10 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Fri, 08 Feb 2002 09:42:10 -0500
Subject: Webizing Python
Message-ID: <3C63E3C2.6020807@destiny.com>

> 
> And seeing as I have very little information, I think it would be a good
> time to start criticizing.
> 


I agree... this could ONLY be undertaken if we were EXCEEDINGLY careful 
about security issues, versioning issues, and locality issues. Even Tim 
Berners-Lee pointed out that he'd rather download modules to his 
computer and run them locally -- and keep them stored locally, to be 
updated only with his permission.

Perhaps the right approach is to FIRST build the python version of CPAN 
-- a system for storing modules and libraries, which also provided some 
way of rating the modules, dealt with versioning issues gracefully, and 
also gave a practically-just-one-click-needed ease of downloading and 
installing. Once something like this was built and working, THEN it's 
time to add it to the language itself.

-- Michael Chermside







From jason at jorendorff.com  Mon Feb 11 00:44:19 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 23:44:19 -0600
Subject: xor on strings
In-Reply-To: <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIEKJDBAA.jason@jorendorff.com>

dsavitsk wrote:
> >>> red = hex(56)[2:4]
> >>> green = hex(156)[2:4]
> >>> blue = hex(12)[2:4]
> >>> thestring = '&H' + blue + green + red
> 
> but here, since the blue string is only 1 character it throws the 
> rest off.

>>> colors = (56, 156, 12)
>>> print '&H%02x%02x%02x' % colors
&H389c0c

If you would prefer to have a function instead,
you can use string.zfill().

## Jason Orendorff    http://www.jorendorff.com/



From jnix at cafeconleche.de  Mon Feb  4 18:06:19 2002
From: jnix at cafeconleche.de (Johannes Nix)
Date: 05 Feb 2002 00:06:19 +0100
Subject: strange error with dynamic loading of sound extension
Message-ID: <67heown8yc.fsf@zosma.lokal>

I am trying to build a python extension type for real-time audio
processing with Python. The extension module is a wrapper around the
ALSA library (ALSA is the advanced Linux sound architecture); more
specific, the digital audio or PCM device. ALSA has the advantage that
it supports a number of 24-bit sound cards and it is possible to work
with very small latencies, even running Python.

I've suceeded with SWIG-generated wrappers for ALSA-0.5.9 (you can
find them as "rtaudio" in the Vaults), but with ALSA-0.9 I run in a
very strange problem - and I need this version urgently, so I'm
begging for help.


Basically, when I try to initialize a PCM handle (which is wrapped
into the extension type), I get the following error:

-----------------------------------------------------------------------
jnix at sirrah:~/zosma/rtaudio > ./test.py default
ALSA lib dlmisc.c:97:(snd_dlsym_verify) unable to verify version for symbol snd_config_hook_load
ALSA lib conf.c:2452:(snd_config_hooks_call) symbol snd_config_hook_load is not defined inside (null)
ALSA lib conf.c:2859:(snd_config_update_r) hooks failed, removing configuration
errcode is: -6Traceback (most recent call last):
  File "./test.py", line 2, in ?
    from rtaudio import pcm
  File "/home/jnix/lib/python/rtaudio/__init__.py", line 5, in ?
    import alsa_pcm
  File "/home/jnix/lib/python/rtaudio/alsa_pcm.py", line 13, in ?
    from pcm import *
pcm.error: device open error: No such device or address
------------------------------------------------------------------

The location of the error is in the following extension type method,
it is triggered by snd_pcm_open() :

pcmmodule.c ------------------------------------------------------------------

static PyObject*
pcm_open(pcm_device *self, PyObject *args)
{             
  int err;
  snd_pcm_stream_t stream = SND_PCM_STREAM_PLAYBACK;
  char *devicename;
  char *mode;
  static snd_pcm_t *handle;


  if (!PyArg_ParseTuple(args,"ss",&devicename,&mode)){  
    return NULL;
  }

  

  if (strcmp(mode,"w") == 0)
    stream = SND_PCM_STREAM_PLAYBACK;
  else if (strcmp(mode,"r") == 0)
    stream = SND_PCM_STREAM_CAPTURE;
  else {
    sprintf(errstr,"open failed: illegal mode '%s'\n",mode);
    raise_Error(errstr);
  }
  
  fprintf(stderr,"trying to open \"%s\"\n",devicename);

  err = snd_pcm_open(&(self->pcm_handle), devicename, stream, 0);
  fprintf(stderr,"tried to open \"%s\"\n",devicename);
  fprintf(stderr,"errcode is: %i",err);
  if (err < 0) {
    sprintf(errstr,"device open error: %s\n", snd_strerror(err));
    raise_Error(errstr);
  }


  self->closed = 0;
  strncpy(self->name, devicename, MAX_NAME_LEN);
  if (strlen(devicename) > MAX_NAME_LEN )
     self->name[MAX_NAME_LEN-1] = '\0';
  return (PyObject*) self;
  
}
----------------------------------------------------------


Now, I've glazed about four hours at the signature of snd_pcm_open -
which is right. It seems also true that snd_pcm_open() is the first
function that has to be called, it initializes the PCM device.  I
concluded that perhaps something is going wrong because of dynamic
loading, which seems to be added in ALSA-0.9.

Next, I took an example program, which plays a sound file in various formats, and stripped off everything beyound snd_pcm_open(). The results looks like this:

ap.c -------------------------------------------------------------

#include <stdio.h>
#include <alsa/asoundlib.h>

static snd_pcm_t *handle;
int main(int argc, char *argv[])
{
        int err;

        err = snd_pcm_open(&handle, "default", SND_PCM_STREAM_PLAYBACK, 0);
        if (err < 0) {
                error("audio open error: %s", snd_strerror(err));
                return 1;
        }


}
--------------------------------------------------------------------

This program runs smoothly, when compiled with this line:

gcc ap.c -lasound -o ap

, it does not produce any error message.

However, when I change the same program into this tiny extension module:

ad.c ----------------------------------------------------------


#include <stdio.h>
#include <alsa/asoundlib.h>

static snd_pcm_t *handle;                                     

int initad(int argc, char *argv[])
{
        int err;

        err = snd_pcm_open(&handle, "default", SND_PCM_STREAM_PLAYBACK, 0);
        if (err < 0) {
                error("audio open error: %s", snd_strerror(err));
                return 1;
        }


}
----------------------------------------------------------------

and compile it with

jnix at sirrah:~/tmp > gcc   -c ad.c -o ad.o
jnix at sirrah:~/tmp >  gcc -shared ad.o -lasound -o ad.so


and try to import it dynamically into python, I get:

jnix at sirrah:~/tmp > python
Python 2.0 (#1, May 16 2001, 00:02:45) 
[GCC 2.95.3 20010315 (SuSE)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import ad
ALSA lib dlmisc.c:97:(snd_dlsym_verify) unable to verify version for symbol snd_config_hook_load
ALSA lib conf.c:2452:(snd_config_hooks_call) symbol snd_config_hook_load is not defined inside (null)
ALSA lib conf.c:2859:(snd_config_update_r) hooks failed, removing configuration
python: : Unknown error 1075756497
jnix at sirrah:~/tmp > 
-------------------------------------------------------

which is obviously the same error as with the full extension module.

So I conclude that the difficulty is related to dynamic loading.
ALSA-0.9 seems to load dynamic stuff too, it has a lot of flexibility
built in, e.g. defining abstract device names in configuration
files. It seems that both systems bite each other.

How can I reconcile them ? I would rather prefer to maintain a dynamic
extension.

I appreciate your help very much.

dynamic-ally but a little bit frustrated, Johannes Nix




From quinn at regurgitate.ugcs.caltech.edu  Fri Feb 22 17:50:30 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Feb 2002 22:50:30 GMT
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <a50hd2$6vt$1@panix2.panix.com> <a50im3$bf1$1@news.island.liu.se>
Message-ID: <slrna7dipm.5et.quinn@regurgitate.ugcs.caltech.edu>

On Wed, 20 Feb 2002 17:23:22 +0100, Anders Dahlberg <andda715 at student.liu.se>
wrote:
>
>"Aahz Maruch" <aahz at panix.com> skrev i meddelandet
>news:a50hd2$6vt$1 at panix2.panix.com...
>> In article <m2wux85hud.fsf at desk.crynwr.com>,
>> Russell Nelson  <nelson at crynwr.com> wrote:
>> >Henrik Motakef <henrik.motakef at web.de> writes:
>> >> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
>> >>> Le 18/02/02 ? 15:28, Tim Peters ?crivit:
>> >>>>
>> >>>> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
>> >>>> czignat?  Classifero en bragna!
>> >>> Woas da Deifi wos des hoasn mog.
>> >> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
>> >
>> >Even better: we could switch languages within threads, to *really*
>> >keep the monolinguists confused.
>>
>> B'vakasha
>
>Ja, varf?r inte?

My s?ster was bitten by a m??se once.


From benji_york at cal-na.com  Fri Feb  8 17:18:02 2002
From: benji_york at cal-na.com (Benji York)
Date: 8 Feb 2002 14:18:02 -0800
Subject: What is the most efficient way to do this?
References: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
Message-ID: <59667b81.0202081418.26709747@posting.google.com>

How about:

>>> l = ['', '1', '2', '3', '', '456', '789', 'wow']
>>> filter(None, l)
['1', '2', '3', '456', '789', 'wow']

--
Benji York
york6 at mindspring.com


From starx at pacbell.net  Sat Feb  2 23:38:36 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 3 Feb 2002 04:38:36 GMT
Subject: The object named 'type', in 2.2
Message-ID: <Xns91A9D1FE73881RASXnewsDFE1@130.133.1.4>

I would like to know just where i can figure out the logic behind this, 
because of the name i can't seem to do a meaningful search on this. Here 
is what i know:

1. In Python 2.2, there is an builtin object in python named 'type'. It 
is of type 'type'.

>>> type
<type 'type'>


2. Every object that the dot operator can be used on, except for classic 
classes, has 'type' as it's ancestor that can be accessed through 
__class__ at some level.

>>> type
<type 'type'>
>>> [5].__class__ is type
0
>>> [5].__class__.__class__ is type
1
>>> class x(object): pass
>>> x.__class__ is type
1
>>> a = x()
>>> a.__class__ is x
1
>>> a.__class__.__class__ is type
1
>>> class y: pass
>>> y.__class__
Traceback (most recent call last):
  File "<pyshell#26>", line 1, in ?
    y.__class__
AttributeError: class y has no attribute '__class__'


3. Type is it's own parent:

>>> type.__class__ is type
1


It is this last one particularly that I do not understand. I ran into it 
today trying to fix a recrusion-death problem in PyCrust under 2.2 -- it 
recursivly accesses thing's __class__es in it's attribute-finding 
routines. I've tried to find information on this, but it has eluded me so 
far...

Thanks

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From jimmy at retzlaff.com  Mon Feb 18 16:46:49 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Mon, 18 Feb 2002 13:46:49 -0800
Subject: Performances of Pyhton programs
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F55A8@MAYOR.retzlaff.com>

Christian Tanzer wrote:
> In my experience, a Python solution might be anywhere between 100
> times slower to 1000 times faster than a C/C++ solution.

I used to work at a company that sent between 25K and 200K distinct
emails per day through Microsoft Exchange Server (it wasn't spam, people
had to explicitly give us a lot of personal information just to get the
emails). The application that composed the emails from information in a
database and handed them off to Exchange (essentially a mail-merge
engine) was originally written in C++. It took about 2 months to
develop, and we spent a lot of time optimizing it over the years.

Finally, maintaining it and adding new features was becoming too
burdensome so I translated it to Python. It took all of a weekend to do
the translation. Of course, I was only testing with one mail note at a
time while I translated it. When I got back to the office and tried it
on 1000 notes, it finished in about 1/4 the time the C++ version would
normally take. I naturally assumed it was broken and didn't send the
mails. But sure enough, 1000 emails showed up in my inbox. I was very
confused and asked my buddy to help me review/compare the code.

It turned out that I'd implemented an optimization in the C++ code a few
years earlier and made a mistake in the implementation such that the
optimization wasn't as effective as it should've been. In translating to
Python, I "accidentally" fixed it and enjoyed the full benefit. It took
a great deal of comparison between the C++ and Python code to even
realize that the C++ code was broken (I never realized it during the
translation). This C++ code had been through several code reviews over
the years that never spotted the problem.

The moral is that, in Python, it's often easy to do the "right thing" by
accident, whereas in C++ accidents aren't often so fruitful. Within a
week we had implemented two new features in the mailer that we'd been
putting off for months because of the complexity of integrating them
into the mailer. And we still had significantly better performance than
before. And we had <400 lines of Python code to maintain in place of the
3,000+ lines of C++ code we'd had before. And it became a lot more fun
to work with. And I was pissed at myself for not making the change 3
years earlier when I first started using Python at work!

It can't be said often enough that artificial benchmarks measure
"theoretical speed" which often has almost nothing to do with "practical
speed." An overarching Python theme seems to be that the practical is
much more important than the theoretical when they don't coincide
nicely.

Jimmy



From gh_pythonlist at gmx.de  Mon Feb 18 16:22:48 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 22:22:48 +0100
Subject: Newbie question...
In-Reply-To: <3C716B27.3090408@bc90021redundantdot.net>
References: <3C716B27.3090408@bc90021redundantdot.net>
Message-ID: <20020218212247.GA9600@lilith.hqd-internal>

Le 18/02/02 ? 21:00, bc90021 ?crivit:
> [installing Ptyhon 2.2 RPMs on Redhat]

... but of course hacking the RPMs can be fun. When I was using SuSE, I
always changed the .SPEC file and recompiled the source RPMs. Then
changed all the other Python related .SPEC files, if necessary and
rebuilt everything Python related. Most of the time, this is pretty
straightforward and a simple replace of "1.5" with "2.2" in the .SPEC
file is enough.

I don't want to discourage you :-)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.2 ?C      Wind: 3.2 m/s



From othello  Wed Feb 27 00:52:52 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 00:52:52 -0500
Subject: Precompiled regular expressions slower?
References: <mailman.1014753464.19952.python-list@python.org>
Message-ID: <a5hs7q$p8s$1@bob.news.rcn.net>

"Sean 'Shaleh' Perry" <shalehperry at attbi.com> wrote in message
news:mailman.1014753464.19952.python-list at python.org...
> >
> > The times of the following three tests are:
> > 1.94322395325   # search(string, string)
> > 2.62431204319   # search(re_obj, string)
> > 0.667925000191  # re_obj.search(string)
> >
> > Someone care to explain why the second ends up slower than the first?
> >
>
>  1 def _compile(*key):
>  2    # internal: compile pattern
>  3    p = _cache.get(key)
>  4    if p is not None:
>  5        return p
>  6    pattern, flags = key
>  7    if type(pattern) not in sre_compile.STRING_TYPES:
>  8        return pattern
>  9    try:
> 10        p = sre_compile.compile(pattern, flags)
> 11    except error, v:
> 12        raise error, v # invalid expression
> 13    if len(_cache) >= _MAXCACHE:
> 14        _cache.clear()
> 15    _cache[key] = p
> 16    return p
>
> My guess is #2 is slower because of the if type() check on line 7.
>

I think the culprit is in line #3.  Looking-up a string in the cache
dictionary
is very fast -- hashing in C and comparing in C or even better matching an
interned string with 'is'.  Looking up a compiled regular expression object
involves checking for a Python class definition of  __hash__  and __cmp__,
then not finding the definitions, revert to default hash and followed by an
id() based compare.

The moral is to give strings to functions that expect strings and make
instance.method calls when a pre-compiled instance is available.


Raymond Hettinger





From kennedy.bill at freeuk.com  Tue Feb 26 12:31:47 2002
From: kennedy.bill at freeuk.com (William Kennedy)
Date: 26 Feb 2002 17:31:47 GMT
Subject: ANN: RNG Lotto / winLotto
Message-ID: <Xns91C1B31005C37kennedybillfreeukcom@195.92.193.157>

I have just completed my first two Python programs, not that they do a lot.

RNG_Lotto is a Lottery generator using basic python and winLotto is a 
version that uses wxPython.

They are both FREEWARE and you can download them at

http://www.kennedy.bill.freeuk.com/lotto.html and
http://www.kennedy.bill.freeuk.com/winlotto.html

I am planning to put a 'how I wrote the code...' section but would first 
like to find out what platforms they work on.  I only have Win98 to test 
them on.

Any help would be appriciated.

Many thanks


From gbreed at cix.compulink.co.uk  Thu Feb 21 08:05:48 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Thu, 21 Feb 2002 13:05:48 +0000 (UTC)
Subject: Adding static typing to Python
References: <3C73E7EE.D0F6DE21@alcyone.com>
Message-ID: <a52rbc$p1j$1@thorium.cix.co.uk>

Erik Max Francis wrote:

> Strong typing simply means that objects have an intrinsic notion of type
> and that operations on them have different meanings depending on their
> types.  Weak typing is the reverse; objects (or variables) do not have
> any intrinsic type and the operations that are performed determine the
> type.  Obviously there are varying degrees of strong typing; C++ is more
> strongly typed than C, for instance, but both are strongly typed. 
> Python is strongly typed; Perl would be a good example of a weakly-typed
> language (e.g., 2 + 2 and 2 . 2 mean different things because "2"
> doesn't have any intrinsic type).

At the point you describe C as strongly typed, you've lost me.  With this 
code

int main() {
        printf("%i", *"\100\0\0\0");
}

I get

$ make typetest
gcc     typetest.c   -o typetest

$ ./typetest
64

Every object in C is a "manipulable region of storage", and can have 
whatever type you want.  How could they have any weaker intrinsic types 
than this?  

As for Perl, from the top of perldata we have

"""
Perl has three built-in data types: scalars, arrays of scalars, and 
associative arrays of scalars, known as ``hashes''. 
"""

So it *does* have strong typing, but describes different things as "types" 
to what you assumed.  Further down, it even says

"""
Although strings and numbers are considered pretty much the same thing for 
nearly all purposes, references are strongly-typed, uncastable pointers 
with builtin reference-counting and destructor invocation.
"""

Okay, so we have a continuum and Perl isn't quite purely weak typing.  But 
I still don't see a trace of strong typing, by this definition, in C.  I 
can also find plenty of contradictory definitions ...


                    Graham

        <http://www.microtonal.co.uk/>


From jason at jorendorff.com  Tue Feb 19 22:06:23 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 19 Feb 2002 21:06:23 -0600
Subject: print and unicode strings
Message-ID: <HFEKILOLEFEFMKAECNDLIENNDDAA.jason@jorendorff.com>

The following 3 lines are now in my PYTHONSTARTUP file
on Windows.

# Enable friendly Unicode output
import codecs
sys.stdout = codecs.getwriter('cp437')(sys.stdout, 'replace')

Check it out.

  >>> print u'\xbfQu\xe9 pas\xf3?'
  ?Qu? pas??

What, no UnicodeError?  What happened?

It's even better in IDLE, with 'utf-8' as the encoding.
Try it.  This is guaranteed to make you grin.

  # in IDLE
  >>> sys.stdout = codecs.getwriter('utf-8')(sys.stdout)
  >>> print u'\N{COMET}'

I think IDLE should do this by default.  It's fine for
Python to "refuse to guess" encodings, but in the case of Tk
it's UTF-8 on all platforms... right?

## Jason Orendorff    http://www.jorendorff.com/




From ajeru at vknn.org  Wed Feb 20 02:54:16 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 23:54:16 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com> <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com>
Message-ID: <24c39b2c.0202192354.765da777@posting.google.com>

That's obviously not the opinion of the members of the "The Python
type system SIG" including Guido van Rossum, otherwise they wouldn't
consider adding static types as an optional feature.

Alexander


Daniel Klein <danielk at aracnet.com> wrote in message news:<55367u4imqamlmv8ou707t40dk1jolrkmc at 4ax.com>...
> The point is that it ceases to be the same language with the same attractions
> (or flaws) depending on how you look at it.
> 
> Dan
> 
> On 19 Feb 2002 16:46:00 -0800, ajeru at vknn.org (Alexander Jerusalem) wrote:
> 
> >Daniel Klein <danielk at aracnet.com> wrote in message
>  news:<ghp47u02vceavgohdod9c8vbs9u5po91bt at 4ax.com>...
> >> Let me reverse the question: What do you think the response would be if
>  Sun
> >> was asked to drop static type checking (as well as checked exceptions).
> >> 
> >> When you can answer this, then you'll have _your_ answer.
> >> 
> >> Daniel Klein
> >
> >Well, my hope is that open source projects are more flexible than
> >bureaucracies like Sun.
> >
> >
> >Alexander


From quinn at hurl.ugcs.caltech.edu  Thu Feb 14 17:23:12 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Feb 2002 22:23:12 GMT
Subject: Convert String to Dictionary question
References: <a4h851$a8g7$1@ID-98166.news.dfncis.de> <Pine.SOL.4.30.0202142127220.24038-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <slrna6oe6g.d0b.quinn@hurl.ugcs.caltech.edu>

On Thu, 14 Feb 2002 21:27:48 +0000, John J. Lee <jjl at pobox.com> wrote:
>On Thu, 14 Feb 2002, MDK wrote:
>
>> I saved my dictionary {'hello1': [1, 0, 0]} to a file.
>>
>> When I do a readline the I get back a string.
>>
>> How do I convert this string to a dictionary?
>
>You probably don't want to.  Read the documentation for the pickle module.

Or if you do, say because 'repr' format is more human readable than pickle,
do 'd = eval(fp.readline())'.  As always with eval(), if bad guys can edit your
file, you're in trouble.  Of course, if they can edit your pickle, you're in
trouble too.


From phd at phd.pp.ru  Mon Feb  4 04:49:06 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 4 Feb 2002 12:49:06 +0300
Subject: Possible bug in Email  and Multifile Modules ?
In-Reply-To: <a3kmhn.3vvfim9.1@kserver.org>; from usenet@thinkspot.net on Mon, Feb 04, 2002 at 01:01:43AM -0800
References: <a3kmhn.3vvfim9.1@kserver.org>
Message-ID: <20020204124906.A729@phd.pp.ru>

Hi!

On Mon, Feb 04, 2002 at 01:01:43AM -0800, Sheila King wrote:
> >>> msg = email.message_from_file(open('virus.txt'))
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
>   File "E:\PYTHON\PYTHON22\lib\email\__init__.py", line 35, in
> message_from_file
>     return _Parser(_class).parse(fp)
>   File "E:\PYTHON\PYTHON22\lib\email\Parser.py", line 40, in parse
>     self._parsebody(root, fp)
>   File "E:\PYTHON\PYTHON22\lib\email\Parser.py", line 116, in _parsebody
>     raise Errors.BoundaryError(
> BoundaryError: Couldn't find terminating boundary:
> ====_ABC1234567890DEF_====
> =============(end interpreter session)================
> 
> Now, it turns out that this message is not well formed. Instead of the
> correct closing boundary at the end of the message, it has a boundary which
> indicates that another part should follow.
> 
> >  No parts were removed, it's 100% intact. And you're right, it's not
> >  well-formed, but that's the whole idea behind it. It tricks Outlook
> >  into executing the attachment.
> > But not being well-formed shouldn't break anything...

   GIF viewer should not display EXE-files. PDF viewer should not print
broken PDFs. And email scanner should not scan files that are not emails.
Just drop it on the flor and forget.
   That M$ Outofluck id dangerously broken, but I don't see a reason to
repeat its bugs and put them into Python libraries.
   If you want to automatically scan emails - scan those that are real
well-formed emails, and put broken files into separate mail folder to deal
with them manually.

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



From jbauer at rubic.com  Wed Feb 20 15:33:14 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 20 Feb 2002 14:33:14 -0600
Subject: MySQLdb build on Solaris 7
Message-ID: <3C74080A.2D415438@rubic.com>

Asim Beg wrote:
> Frankly, I am surprised (shocked...) to find that an 
> important package such as MySQLdb for python would 
> not compile on Solaris...

Let's get some perspective here.  MySQLdb python has compiled
and is running on thousands of Solaris machines ... just not
yours <0.5 wink> -- yet.

Further questions:

1. What is in your LD_LIBRARY path?

2. Which version of the MySQL server is running
on your Solaris machine?

3. Did you install MySQL database server on the
machine or did the admin?  If the latter, did they
compile it on the machine or install a binary
distribution (which might result in a version
mismatch)?

And one suggestion:  If all else fails you might
consider compiling a local MySQL server from the
sources -- producing the libs relative to your home 
directory -- and then building the MySQLdb python 
interface against those libs --- then testing your 
setup your local MySQL server.

Jeff Bauer
Rubicon Research



From st-newsgroups at thomas-guettler.de  Fri Feb  8 04:55:51 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Fri, 08 Feb 2002 10:55:51 +0100
Subject: High-level languages, large projects, GNOME and the .NET controversy
References: <23891c90.0202070501.78df2371@posting.google.com>
Message-ID: <3C63A0A7.5060507@thomas-guettler.de>

Paul Boddie wrote:

> Slightly off-topic, but the recent controversy stirred up by Miguel de
> Icaza and Richard Stallman on the future of GNOME and its relation to
> .NET technologies produced an interesting response:
> 
>   http://www.newsforge.com/article.pl?sid=02/02/06/1536255&mode=nocomment
> 
> Here's the relevant section:


[cut]


Would be nice to use perl libraries in python. But I think that can
be done easier without CIL (Common Intermediate Language. AFAIK somthing 
like bytecode in java or .pyc).


Would be nice to have the intellisence (dropdown-list if you type 
object.) of the Visual Studio. Is there already something like this?


Would be nice to have a component architecture which allows 
plug-and-play usage of things like e.g. html-editor. On the shell level
integration is easy. Just use popen(), but with GUIs it is more difficult.

thomas







From franke at ableton.com  Sun Feb 10 13:36:35 2002
From: franke at ableton.com (Stefan Franke)
Date: Sun, 10 Feb 2002 19:36:35 +0100
Subject: Are there any Python books or sites with good exercises?
Message-ID: <g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r@4ax.com>

(WRT your personal definition of 'good', of course). I'm going to 
prepare a Python introductory course for experienced programmers
and I'm looking for handy exercises which hopefully provide them
numerous Aha! effects.

Thanks,
Stefan




From sheershion at mailexpire.com  Fri Feb 22 08:34:49 2002
From: sheershion at mailexpire.com (Robert Amesz)
Date: Fri, 22 Feb 2002 13:34:49 -0000
Subject: wxPython MDI question
References: <aaa059f1.0202211632.1f9e9d98@posting.google.com>
Message-ID: <Xns91BD914C1D0D1rcamesz@amesz.demon.nl>

rockmuelle wrote:

> I am trying to create an MDI application that has a list control
> on the left side of the main MDI frame and the MDI client area on
> the right side, similar to the way Visual Studio has other windows
> surrounding its client area.
> 
> I've tried creating the list control and then adding it and the
> client area to a sizer, but the client area always takes up the
> whole main frame. The client area and the list control are
> situated at (0, 0), as if there was no sizer.  If I add more
> controls to the sizer, the non-client area windows get layed out
> correctly, but the sizer and client area still overlap.
> 
> [snip]
>
>     sizer = wxBoxSizer(wxHORIZONTAL)
>     sizer.Add(wxListCtrl(self, -1))
>     sizer.Add(wxListCtrl(self, -1))  # add another list
>     sizer.Add(self.GetClientWindow())

I can see what you're trying to do here, but it's obvious the 
wxMDIParentFrame doesn't cooperate, even after correcting the code (see 
caveats below).

This would seem a task for manual resizing, as sizers obviously don't 
work properly here. Modify your MyMDIFrame thus (tested with wxPython 
2.3.2, Python 2.1.1 and Win98):


class MyMDIFrame(wxMDIParentFrame):
    """
    Main frame that has list control and MDI client area.
    """
    
    def __init__(self, parent, id=-1, title='MDI Example'):
        wxMDIParentFrame.__init__(self, parent, id, title=title)
        self.list = wxListCtrl(self, -1)
        EVT_SIZE(self, self.onSize) 
        return

    def onSize(self, evt):
        width, height = self.GetClientSizeTuple()
        listwidth = self.list.GetSizeTuple()[0]
        self.GetClientWindow().SetDimensions(listwidth, 0,
                                width - listwidth, height)
        self.list.SetDimensions(0, 0, listwidth, height)


Note that I've removed the second wxListCtrl to make the calculation 
simpler. As such, it is simply a matter of dividing the available space 
into two parts, using SetDimensions() to resize and position both 
windows.


Robert Amesz
-- 
A few caveats concerning sizers: don't call Layout() before SetSizer(), 
for obvious reasons. And after SetSizer(), there's no need to call 
Layout(), that should happen automatically, but *do* put a 
self.SetAutoLayout(true) in your frame or panel, otherwise nothing will 
happen at all.


From phr-n2002a at nightsong.com  Wed Feb 20 22:41:31 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 19:41:31 -0800
Subject: strings and sort()
References: <mailman.1014257380.19140.python-list@python.org>
Message-ID: <7xsn7vbixg.fsf@ruckus.brouhaha.com>

"Sean 'Shaleh' Perry" <shalehperry at attbi.com> writes:
> > Why doesn't sort() return the sorted list.  I would like to chain it
> > to other operations:
> >         b=[x for x in a].sort()
> > 
> efficiency.  Why return the sequence when we usually just need to sort the
> list.  Sort is also a meothod of an object, why should it return a new object
> instead of just affecting the one it belongs to?

I don't understand this reply.  The sort method would be no less efficient
if it returned the sequence:

  def nsort(a):
     a.sort()
     return a

The reason it doesn't work that way is so people don't do

   a = b.sort()

expecting to get a sorted copy of b, ending up with a rude surprise
when b gets clobbered.

In Scheme, there's a convention that function names ending with "!"
operate are destructive:

   (reverse x)     ; makes  a copy of list x, in reverse order
   (reverse! x)    ; reverses list x, in place

In Common Lisp, there's a similar convention that the destructive
version starts with "n":

   (reverse x)    ; make reverse copy of x
   (nreverse x)   ; reverse in place

   (concat a b)   ; copy and concatenate two lists, like "a+b" in Python
   (nconcat a b)  ; concatenate in place, like "a.extend(b)" returning a

If there was a built-in sort function, CL would use "nsort" like the
above, and "sort" as the equivalent of "a[:].nsort()" (make a
copy of a, sort the copy in place, and return the copy).

I think Python's definition of "sort" is a minor wart.  It's confusing
that sort doesn't make a copy, and to ameliorate the confusion, sort
is made to not return sorted list, causing inconvenience.
I think the CL approach would have been better, to have separate
sort() and nsort() operations for sorting in place and making a copy.
Or just have nsort, whose weird-looking name alerts the user that it
has special behavior.  


From jason at jorendorff.com  Mon Feb 11 17:08:53 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 11 Feb 2002 16:08:53 -0600
Subject: subclass list
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBKEEEKCAA.mark@mceahern.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIENPDBAA.jason@jorendorff.com>

Mark McEahern writes:
> Q:  How do I override append() so that I can determine whether to add the
> value or not?
>
> I tried:
>
>   def append(self, value):
>       if not self.unique or (self.unique and not
self.keys.has_key(value)):
>           list.append(value)
>           if self.unique:
>               self.keys[value] = None

list.append(self, value)
  or
super(Bucket, self).append(value)

More on super:
  http://www.python.org/2.2/descrintro.html#cooperation

## Jason Orendorff    http://www.jorendorff.com/




From henrik.motakef at web.de  Sun Feb 10 15:52:19 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 10 Feb 2002 21:52:19 +0100
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org> <3C66F2CA.802A00BF@earthlink.net>
Message-ID: <86lme1xdoc.fsf@apo.hmotakef.int>

Ron Stephens <rdsteph at earthlink.net> writes:

> (I still can't paste the whole crontab file because I only have a
> two button mouse and K-Edit doesn't copy and paste into Netscape
> messenger ;-)))

This might be slightly off-topic, but usually pressing both buttons at
once emulates pressing the middle one...

hth
Henrik


From debl2NoNoSpammyKwhammy at bellatlantic.net  Sun Feb 17 17:46:52 2002
From: debl2NoNoSpammyKwhammy at bellatlantic.net (David Lees)
Date: Sun, 17 Feb 2002 22:46:52 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <3C7032E3.D975D089@bellatlantic.net>

I have read through this thread and have a few historical comments on
BASIC.
First, the original BASIC circa 1964, which ran on a GE-235 based time
sharing system
was compiled, not interpreted.  So was its successor which ran on a
GE-635 about 2 years
later.  It was a very fast compiler, so that it had the feel of an
interpreter,
but the execution time was better than the interpreters of the day.  It
was much later,
70's I guess, that DEC and later Mr. Gates created interpreted versions
of BASIC.

Second, BASIC was intended from the very start (by Kurtz and Kemeney, it
inventor)
to be used as a teaching language for everyone; College students and
high schools students
In fact if my failing memory serves :) they had an NSF grant which put
those old ASR-33
and KSR-35 teletypes online at high schools using the old 110 baud
acoustic modems.
I have noticed (purely subjective) that people trained in computer
science
tend to look down at BASIC.  If the quote from Edsger Dijkstra is real,
it just
confirms some of the snottyness by CS types. Dartmouth had a time-shared
version of
Algol-60 also available, which was not used particularily heavily.

Third, as pointed out in this thread, BASIC has been wildly successful
if judged by
usage.  It replaced FORTRAN (or was it FORTRAN II, I forget) for
teaching computer
literacy to Dartmouth undergrads.

Fourth (not a pun) little piece of historical trivia.  The original
BASIC evolved as a language and splintered
all over the place.  One of the excellent things I see in Python is the
tight control
of the language evolution.  In the year that I have been intermittently
using Python
I have seen nice improvements in the language, but it is great to see
public discussion
and thought about what will break when changes are made.

Enough rambling.

David Lees


Robin Munn wrote:
> 
> Reading the first few chapters of _How to Think Like a Computer
> Scientist_ (http://www.ibiblio.org/obp/thinkCSpy/ for those who haven't
> heard of this book), I was struck by how Python resembles BASIC in some
> ways that make it a very good language for teaching:
> 
> * Interpreted, so there's immediate feedback. Students can just type
>   'python' at the command prompt and start coding, receiving *immediate*
>   feedback on the effects of what they type.
> 
> * Non-scary assignment and output statements. In fact, Python and
>   BASIC's output statments look identical. "Print" is a natural choice
>   for ease of understanding. Using sys.stdout.write can come later, once
>   students begin to realize that they don't always *want* a newline or
>   space appended to their output.
> 
> * No need to declare variables. Use them and they are immediately
>   created for you, ex nihilo.
> 
> All three of these things make for a very low barrier-to-entry for
> people who've never done any programming before. I'm sure there are
> other similarities as well, that maybe someone else will point out. Of
> course, Python goes beyond BASIC in many ways:
> 
> * NO LINE NUMBERS! Enough said.
> 
> * Rich control-flow statements. Of course, this is available in more
>   modern versions of BASIC, but back when I was first learning to
>   program (at age six, using BASIC), all I had was conditional GOTO. You
>   know, IF condition GOTO line-number. Or, for the advanced students,
>   IF condition GOSUB line-number (I never did understand the necessity
>   for that kind of statement until well after I'd stopped using BASIC
>   entirely and moved to more "modern" languages like... Pascal).
> 
> * Batteries included. This is the big one, which makes Python more than
>   just a toy language suitable for learning. The richness of the
>   standard library makes Python incredibly useful, and I for one would
>   like to see it continue to grow. I understand the concern that people
>   developing Python-based programs for distribution would like their
>   users to be able to run a 100k script without downloading a 10M
>   distribution, but I don't want to see that cause a reduction in the
>   standard library. Instead, we could make make two Python
>   distributions, a "full" install containing everything and a "lite"
>   install containing only the interpreter and maybe -- *maybe* -- a
>   couple of essential modules like os and sys. Then people developing
>   Python programs like, say, a pygame-based game, could distribute a
>   single package containing the "Python lite" distribution, their own
>   code, and any other modules their code depended on -- all in one
>   easy-for-the-end-used-to-install package.
> 
> Anyhow, I hope this helps develop some ideas.
> 
> --
> Robin Munn
> rmunn at pobox.com


From pje at telecommunity.com  Mon Feb 25 09:49:59 2002
From: pje at telecommunity.com (Phillip J. Eby)
Date: Mon, 25 Feb 2002 09:49:59 -0500
Subject: "Data" vs. "Non-data" Descriptors in PEP 252, and "once" attributes
References: <MPG.16e373104557a63b989685@news.supernews.com> <j48z9hn8tl.fsf@informatik.hu-berlin.de>
Message-ID: <MPG.16e4191e85b18f71989686@news.supernews.com>

In article <j48z9hn8tl.fsf at informatik.hu-berlin.de>, 
loewis at informatik.hu-berlin.de says...
> Phillip J. Eby <pje at telecommunity.com> writes:
> 
> > Interestingly, this does *not* work if you try to do something similar 
> > with the built-in 'property' type.  
> 
> I don't find this surprising. The builtin property type has a fixed
> meaning for its constructor argument: they are callables, and they are
> called on each access. How did you mean to implement your "once"
> properties with it?

By supplying only an 'fget' function, so that the 'property' object would 
be a "non-data" descriptor.  Which doesn't work.  Even if you use a 
function which caches its result in the instance dictionary, the function 
is still called every time, because 'property' instances are always 
"data" descriptors.

In retrospect, I realize that it wouldn't make sense for 'property' 
to create a new type on each call, or use a different type based on which 
functions you supply.  And since the data/non-data determination is made 
via type slots, it's understandable in retrospect that it would not 
change based on the functions passed.

This wasn't really obvious to me at first, however.  On the surface, 
creating a 'property' seemed isomorphic to writing a property class of my 
own with '__get__' and '__set__' methods; I was therefore initially 
puzzled by the behavior difference between writing my own property type, 
and using the built-in one.  Although I was supplying a only a "get" 
method in both cases, the behavior was different, and that was what 
caused me to go PEP-hunting and source-digging to understand the 
difference.


> > Operationally, it seems it would be hard to change any of this in a way 
> > that would break my "once" examples without also breaking lots of other 
> > things.  But the PEP doesn't really give an official way to create either 
> > a data or non-data descriptor; it just says what the current 
> > implementation does.  Hence my concern.
> 
> Nobody can tell the future with certainty. If you think that certain
> wording would tighten the PEP to make you feel more safe, just submit
> a patch for the PEP to SF.

I wasn't aware that one could do that.  Thanks for the tip; I'll look 
into it.  I don't imagine it'll even be a big patch; just an explanation 
that a data descriptor is one which includes a "tp_descr_set" function.  
Or perhaps it would be better to drop the "data/non-data" terms and refer 
to "writable" vs. "read-only" descriptors.


From ajeru at vknn.org  Wed Feb 20 16:00:29 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 20 Feb 2002 13:00:29 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com> <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com> <24c39b2c.0202192354.765da777@posting.google.com> <mailman.1014210154.9297.python-list@python.org>
Message-ID: <24c39b2c.0202201300.3dfe7dbc@posting.google.com>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message news:<mailman.1014210154.9297.python-list at python.org>...
> ----- Original Message -----
> From: "Alexander Jerusalem" <ajeru at vknn.org>
> 
> 
> > That's obviously not the opinion of the members of the "The Python
> > type system SIG" including Guido van Rossum, otherwise they wouldn't
> > consider adding static types as an optional feature.
> >
> > Alexander
> >
> >
> > Daniel Klein <danielk at aracnet.com> wrote in message
>  news:<55367u4imqamlmv8ou707t40dk1jolrkmc at 4ax.com>...
> > > The point is that it ceases to be the same language with the same
>  attractions
> > > (or flaws) depending on how you look at it.
> > >
> > > Dan
> 
> Static types as an optional feature for function/method interfaces is
> not a bad thing, and does not really change the language; but the key
> word here is "optional." 

My suggestion for adding static typing included this word "optional"
from the very beginning. I'm not seriously suggesting that everyone
reengineer their code to include static typing and I see many benefits
in dynamic typing as well.

> If static typing of function arguments is
> optional, I for one am not likely to use it, so the language hasn't
> changed.

Exactly, that's what I said from the beginning.

> 
> Static typing is used by compilers to improve performance, I'm told;
> but to do it with an interpreter would require implicit typechecking
> by the interpreter core.  This would probably be faster than explicit
> typechecking (as in, write it yourself) but I can't help but see it as
> a performance loss vs. the usual method (no typechecks at all,
> exceptions thrown for interface mismatches).

The proposals of the types-SIG say that static typing is used on a per
module basis. That means that weather there is or is not a performance
improvement it's only going to affect you if you choose to use the
static types.

Alexander


From kkeller at speakeasy.net  Thu Feb  7 15:20:50 2002
From: kkeller at speakeasy.net (Keith Keller)
Date: Thu, 7 Feb 2002 12:20:50 -0800
Subject: 2001 Turing Award
References: <mailman.1013109679.5469.python-list@python.org>
Message-ID: <2jnu3a.4sq.ln@wombat.san-francisco.ca.us>

In article <mailman.1013109679.5469.python-list at python.org>,
	James_Althoff at i2.com writes:

> ----- Message from Gary Chapman <gary.chapman at mail.utexas.edu> on Thu, 7
> Feb 2002 08:41:00 -0600 -----
> 
> I'm happy to pass on the news that the 2001 Turing Award, the highest
> award in computer science

Damn--I thought the award was for the person who best impersonated
a Turing machine.  :)

--keith

-- 
kkeller at speakeasy.net
public key:  http://wombat.san-francisco.ca.us/kkeller/public_key
alt.os.linux.slackware FAQ:  http://wombat.san-francisco.ca.us/perl/fom



From no at spam.com  Tue Feb 26 13:45:22 2002
From: no at spam.com (Tokomar)
Date: Tue, 26 Feb 2002 18:45:22 GMT
Subject: Tkinter list boxes, selection emphasis, and the keyboard focus
References: <K0Qe8.48006$BA5.2207030344@newssvr14.news.prodigy.com>
Message-ID: <6JQe8.6350$V23.3347875968@newssvr21.news.prodigy.com>

Tokomar wrote:

> But has anyone addressed this problem directly? Is there a configuration
> option I overlooked?

The ready-made workaround is called "activate".

But the outer problem with Tkinter's style remains.

-- 
Tokomar


From nordwick at scam.xcf.berkeley.edu  Thu Feb 21 15:08:58 2002
From: nordwick at scam.xcf.berkeley.edu (Jason Nordwick)
Date: 21 Feb 2002 12:08:58 -0800
Subject: test
Message-ID: <kpg03ubns5.fsf@scam.xcf.berkeley.edu>

this is only a test of the emergency broadcast system, in the event
of a real emergency...




From jo at johnochiltree.uklinux.net  Sun Feb 10 08:49:33 2002
From: jo at johnochiltree.uklinux.net (John Ochiltree)
Date: Sun, 10 Feb 2002 13:49:33 GMT
Subject: Python IDEs
References: <zfC08.15504$Hg7.1623775@news11-gui.server.ntli.net> <a1uv39$o79$1@rex.ip-plus.net> <3c664c2b@nntp.server.uni-frankfurt.de> <867kplznbt.fsf@apo.hmotakef.int>
Message-ID: <NTu98.19162$H9.2428975@news1.cableinet.net>

On Sunday 10 Feb 2002 9:40 am, Henrik Motakef wrote:

> Michael 'Mickey' Lauer <mickey at tm.informatik.uni-frankfurt.de> writes:
> 
>> From my point of view, I'd like to have the time to add Python
>> capabilities to kdevelop, which seems the most promising open source
>> IDE around.
> 
> AFAIK, there is Python (and other non-C++ languages) support in the
> CVS version of Gideon (the successor of KDevelop). So no need to take
> the time yoursef.
> 
> Regards
> Henrik

Is there a URL for Gideon, the Gideon on SourceForge isn't it.

Ta
John


From threeseas at earthlink.net  Sun Feb 17 17:06:20 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 17 Feb 2002 22:06:20 GMT
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1> <8867.813T2365T9245607threeseas@earthlink.net> <a4p219$1o0ne$1@ID-11957.news.dfncis.de>
Message-ID: <5098.813T2333T10665859threeseas@earthlink.net>

On 17-Feb-02 14:52:55 Emile van Sebille <emile at fenx.com> wrote:
>Complete re-posts smell like spam.

Show me that you know what spam is rather than using the word to insinuate
a negative. Hell, for that matter, show me that you know what a "complete
re-post" is. So far, I have no evidence that you know either.

>I have no doubt that your nine steps can be super-imposed on everything.

Good, that's a big step in the right direction!!!!

>However, as long as the deep magic involved in converting these steps
>into things people want done is kept as magic, let Aahz's quote reign:

>"This is Python.  We don't care much about theory, except where it
>intersects with useful practice."  Aahz Maruch on c.l.py

I'd have to say you took two steps backward in "converting  these steps
into things people want done" as this suggest that you have no idea what
you said above that caused you to take a big step forward.

As these steps don't covert into what people want done, but rather are
used to automate conversion of abstract programming languages into the
programs the user then uses to do, or are used by the user to automate
what they have done, so that they can do it again and again in a simpler
manner.

As you apparently subconsciously recognized, to be able to super-impose
these steps onto what you do, specifically thru computers, means you can
identify the automation control points, and define them in a manner to
automate what you super-imposed them upon.


I'm not stopping you or anyone from converting verified physical reality
into the environment of computing where they can be used by anyone and
everyone that wants to make use of them. And I'm not going to be so
arrogant as to assume I know what people want done, but I do know that
what ever it is they want done, it can be mapped using the nine action
constants in a manner that allows it to be automated, should they so chose
to do it more than once. This is really no different than knowing the
elements of mathmatics, regardless of what people want to calculate, can
be used to calculate whatever people want to mathmatically calculate.

There is no theory or even as you say "magic". But if you think there is
some magic then I suppose you really are not much of a programmer but more
likely an exdotcomer. What there is is just plain logic.

As a matter of fact I've done nothing but put forth effort to help by
identifying and defining these nine action constants in terms of computer
functionality descriptions.

I suspect the difficulity you have in comprehending this is as a result
of your simply not wanting to see it. And that is your choice. Or perhaps
as indicated above, you are simply trying to preceive these nine actions
as a replacement for programming languages (as some here have recently
insinuated - are you so weak minded to follow such error?). These nine
actions don't replace programming languages but rather allow you to
automate the use of them. Consider super-imposing the nine steps (actually
not steps but actions that can be done in any sequence order and even
repeated each as needed) on some common and often repeated coding you do
manually, even if there is some variation in it. So long as you can
super-impose the action set then you can automate it's use or application,
even dynamicly. And once you do it, you can if you want, make the
automation available for other to use/benefit from, as they can do the
same for you. Certainly automating such certainly wantingly repeatable
things as do's, don't's and standard of a given language use is something
everyone using the language would benefit from. And once you automate its
use, no more bug typos in what is done right once, and then automated in
applying it.

There is no magic to any of this and there need not be any magic or
complexity beyond the typical user/consumer available resources to use an
autocoding tool set to do program creation for themselves.

In fact it's the software industries defined job to make things easier for
the consumer. But I suppose there is some magical reason why they don't
ever quite do it, huh?

Now why would I be doing all of this? Which answer would you want, world,
group or self? Lets' try self!

So I can Do the things I need to get done without having to reinvent what
so many have already invented and reinvented. Where in return I can return
the favor of making automations I create available for other to build
upon.

Is that a bad thing? Does that make me the nut or fool others want to make
me out to be? Or are their actions and words against me only exposure of
their self limiting greed?

Do you have a problem with automating the use of Python and doing so by
using Python to create the automation tool?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From steve at lurking.demon.co.uk  Mon Feb  4 08:03:16 2002
From: steve at lurking.demon.co.uk (Steve Horne)
Date: Mon, 04 Feb 2002 13:03:16 +0000
Subject: lambda expressions
References: <mailman.1012737185.19345.python-list@python.org>
Message-ID: <8p0t5u4a5504t5sjl2s6h9sjseegse0qjj@4ax.com>

On Sun, 3 Feb 2002 12:52:48 +0100, Bas van Gils <bas.vangils at home.nl>
wrote:

>Pythonians,
>
>I've been wondering, lately, about lambda-expressions. Most programming
>languages support and I wondered why (mostly because I find them hard to
>read, valueing _readable_ code). Collegues at work couldn't give me a
>satisfying answer so ... can anyone explain to me why lambda-expressions
>are so useful?
>
>thankx in advance
>
>    Bas

First off, I don't think most languages support lambda expressions.
It's really just languages which have a functional influence.

In functional languages, the reason is simply that they are based
heavily on lambda calculus, and therefore need to be able to express
all the core concepts from lambda calculus.

In Python, it's more to do with convenience and readability when
passing functions as parameters. If the function to be passed is very
simple, it is easier to define it as an anonymous lambda because it
saves having to write a named function elsewhere, and it saves future
readers from having to look up a separate function elsewhere in the
code.

Consider trying to read the following code...

  print wibble (a, b)

You'd be pretty annoyed to find that the definition, 1000 lines of
code away, was...

  def wibble (a, b) :
    return a + b

You'd much rather see...

  print a + b


Similarly, this...

  print map (lambda x : x*2, [1,2,3])

...is clearer than...

  def wibble (x) :
    return x*2

  <thousands of lines of code>

  print map(wibble, [1, 2, 3])


Though only if you understand the 'lambda', of course.

-- 
Steve Horne
steve at lurking.demon.co.uk


From robin at jessikat.fsnet.co.uk  Mon Feb  4 09:45:23 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Feb 2002 14:45:23 +0000
Subject: Pythonservice stuck
Message-ID: <ot678KAD6pX8Ewmt@jessikat.demon.co.uk>

Win2k SP2

I tried out the latest Zope 2.5 and after playing a bit I stopped the
service and uninstalled the software.

However, in my services I still see a service for Zope_2_5 (the name I
gave it instead of WebSite). A reboot doesn't clear this.

I see several entries relating to Legacy zope_2_5 in the registry, but
cannot delete them even with Administrator luser privs.

Any way to remove these pesky entries?
-- 
Robin Becker


From grimmtoothtoo at yahoo.com  Sat Feb  9 16:55:04 2002
From: grimmtoothtoo at yahoo.com (Jeff Grimmett)
Date: 9 Feb 2002 13:55:04 -0800
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>	<3c62a989.3448437947@news>	<a3ucsk$12h8$1@nntp6.u.washington.edu>	<a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org>
Message-ID: <2df3d89d.0202091355.222422d1@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1013203955.3594.python-list at python.org>...

> Arguing with irrational oddballs on usenet is today's equivalent of the
> Rubik's Cube.

Only thing is, there is an actual solution to Rubik's Cube.

I remember Timbo from when I used to hang out in c.s.amiga.*, years
ago, so consider that if he's still at it, there's a pretty good
chance that nothing anyone here can say will put an end to the
nonsense. Killfile him and move on.

I guess his Amiga broke, so he's looking to Python instead of ARexx to
implement his ... whatever it is.

Timbo, it won't work. You need to use Perl. Perl, or C#.


From ecastro at cicei.ulpgc.es  Sat Feb 23 07:46:48 2002
From: ecastro at cicei.ulpgc.es (Enrique)
Date: 23 Feb 2002 12:46:48 GMT
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <3C778FB0.7AB1A2E4@cicei.ulpgc.es>

Hi Tripp

Tripp Scott wrote:
> 
> Suppose I want to make a function that modifies its argument:
> 
>   a = -1
>   make_abs(a)
>   print a # = 1
> 
> What is the Pythonic way of doing this? Wrap it in a list?
> 
>   make_abs([a])
> 
> t

Simply make the function return the modified argument.

a=function(a)

This follows the rule "explicit better than implicit".


From peter.milliken at gtech.com  Tue Feb 19 01:41:44 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Tue, 19 Feb 2002 17:41:44 +1100
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org>
Message-ID: <a4ss56$akj6@news1.gtech.com>

"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1014098489.15429.python-list at python.org...
> Milliken, Peter wrote:
> > Strong typing requires that you design before you code (Extreme
> > Programmers beware! :-)) - therefore by the time a clean compile
> > is achieved 95% of the bugs *are* out :-) That is one of the
> > primary advantages to strong typing, it *forces* you to do the
> > design up front otherwise you will never get a clean compile.
>
> Hmm.  Perhaps you can explain this in more detail.
>

Statement is true when a number of conditions are satisfied :-) One of which
is that we are talking a non-trivial program!

Strongly typed languages encourage you to override the base types so that
you aren't forced one day (as an example) to go through all of the source
looking for places where floats are used and change them to double floats -
happened on one rather large project I worked :-) They changed the compiler
and it defined float as being 32-bit, the previous version defined float as
64-bit - caused no end of heartache as you can imagine :-)

So type'ing encourages things like defining data types that encompass the
range of values your variable is *expected* to have at the time you
design/write your code i.e. you might be talking the temperature of water -
0 - 100 Degrees Celsuis for instance. Languages that are strongly typed like
this then allow static checking at compile time i.e. the compiler will tell
you if you attempt to add a variable declared as Litres to a Variable
declared to hold Degrees Celsius. Similarly, the compiler will complain if
you attempt to assign a temperature of 101 Degrees to a variable declared
with the above limits. Ada allows run-time checking as well, so if you
somehow manage to write code that passes the static checking it will blow up
at run-time with a nice exception and traceback that tells you exactly what
and where it all happened :-)

So, if you follow this form of design approach i.e. taking the real world
objects and modelling them in your code - something I believe to be
encouraged in OO? :-) then if you make a mistake and attempt to add the
contents of two variables of differing types then the compiler will tell you
about it - doesn't matter wether they are both integers at the end of the
day, if you have told the compiler that variable 'A' is a temperature of a
certain range and variable 'B' is a volume of another range then you will
not be allowed to add them together - rather simplistic example, but conveys
the concept I hope.

Of course, a strongly typed checked language won't guarantee that such
features are used! This is always up to the programmer (I once saw a
programmer mangle his Forth code so badly it looked like his previous
language - Fortran! :-)). But if they are used then you are forced to make
sure that you stick with the "rules" as you define them via your type
declarations. So our argument here is presumptive of the features being
used - you could argue the same thing about any language feature i.e. I
could argue that the OO features of Python offer no advantage - based upon
never using them! :-)

> Are you arguing that compilers for strongly typed languages tend to
> reject ill-designed or thrown-together code?  I hadn't noticed.
>

Assuming the above, strongly type checked compilers/languages will indeed
reject ill-designed and thrown together code - it might take a while, but
eventually the programmer will back himself into a hole :-). You usually
have some form of "override" even so, so you could litter your code with
coercions (I have seen this too! :-)) but this is relatively easy to
discover, so the culprit is soon exposed.

> Surely you're not arguing that extra-productive programming tools
> are bad because they encourage the programmer to waste his or her
> own time by skipping the design step.

Sorry too many "negatives" here for me to work out what you are driving at -
I just came back from a "big" lunch, so the old brain isn't functioning on
all cylinders too well :-)

>
> Maybe this is true in some sense - since I enjoy coding Python so
> much, perhaps I'm overeager to get started?  But if so, then the
> solution is sane development practices and a touch of
> self-discipline.  Which will likely come in handy on any ten-year
> project.
>
>

Yes, sane development practices are essential on such large projects -
doesn't matter what language is used. Part of my argument is that language
features are designed and provided as further "encouragement" down the path
of "correct" practices. Typing is recognised as one of these features......

> In any case, it's not nice to pick on XP and then say:
>
> > [...] I have probably got the idea of Extreme
> > Programming wrong - I really don't care, from what I have seen
> > described the style doesn't suite me :-)
>
> Blah.  :(

True, very wrong of me - my apologies :-) That was a cheap shot.

>
> ## Jason Orendorff    http://www.jorendorff.com/
>




From jw9 at cse.buffalo.edu  Wed Feb  6 16:31:24 2002
From: jw9 at cse.buffalo.edu (J. Wang)
Date: Wed, 6 Feb 2002 16:31:24 -0500
Subject: Is there anyway to obtain client's IP address using cgi module?
Message-ID: <Pine.SOL.4.30.0202061627250.17425-100000@hadar.cse.Buffalo.EDU>

I am trying to create a python CGI script. While I'd like to log the
user's IP address, i don't know how to do it. Any advice will be
appreciated. Do I need to use some other module besides cgi?

Thanks!





From nospam at bigfoot.com  Wed Feb  6 11:12:26 2002
From: nospam at bigfoot.com (Gillou)
Date: Wed, 6 Feb 2002 17:12:26 +0100
Subject: re difference between python 1.5 and python 2.1.2
Message-ID: <a3rk7v$vlb$1@norfair.nerim.net>

Hi,

I got a re that searches for FTP or HTTP URLs in a text.
This worked perfectly with python 1.5.2 but it raises an exception with
Python 2.1.2.
Other regular expressions I use work perfectly when running with Python
2.1.2.

Has anybody some hint ?

Thanks in advance.

--Gilles

>>> import re
>>> pat = re.compile(r'((ftp|http)://[\w-]+(?:\.[\w-]+)*(?:/[\w-\.?=]*)*)')
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "C:\Python21\lib\sre.py", line 90, in compile
    return _compile(pattern, flags)
  File "C:\Python21\lib\sre.py", line 136, in _compile
    raise error, v # invalid expression
error: bad character range





From mdk at mdk.com  Tue Feb  5 14:28:01 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 14:28:01 -0500
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de> <a3p2fg$49g$1@mail1.wg.waii.com> <a3p9b6$1a7qtb$1@ID-98166.news.dfncis.de> <a3p9uh$19q6me$1@ID-11957.news.dfncis.de>
Message-ID: <a3pbo5$19rrgt$1@ID-98166.news.dfncis.de>

"Emile van Sebille" <emile at fenx.com> wrote in message
news:a3p9uh$19q6me$1 at ID-11957.news.dfncis.de...
>
> "MDK" <mdk at mdk.com> wrote in message
> news:a3p9b6$1a7qtb$1 at ID-98166.news.dfncis.de...
> > >>> class MyList:
> > ...     def __init__(self):
> > ...             self.list=[]
> > ...     def append(self, item):
> > ...             self.list.append(item)
> > ...
> > > >
> > >>> import mylist
> > >>> x = mylist()
> > Traceback (most recent call last):
> >   File "<interactive input>", line 1, in ?
> > TypeError: object of type 'module' is not call
> >
> > Why is this happening?  My list object must be a class in its own,
> > importable file.
>
> s/b
>
> x = mylist.MyList()
>
> --
>
> Emile van Sebille
> emile at fenx.com

Ah, thanks.




From richmond at ev1.net  Fri Feb 15 17:15:06 2002
From: richmond at ev1.net (Charles Richmond)
Date: Fri, 15 Feb 2002 22:15:06 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net> <a4geh2$hir$12@bob.news.rcn.net> <140220020703593280%jwbaxter@spamcop.net> <a4j24g$jao$9@bob.news.rcn.net>
Message-ID: <3C6DA497.9AB8D624@ev1.net>

jmfbahciv at aol.com wrote:
> 
>      [snip...]     [snip...]     [snip...]
> 
> I do have two memoes that I found in my office boxes that were
> addressed to PDP-1 users.  I don't know the people who wrote
> them (Roland Silver, Edward Fredkin) nor where they were.
> 
>      [snip...]     [snip...]     [snip...]
> 
> Well, if you can give me a clue about these memoes that I have, I'd
> appreciate it.
> 
Ed Fredkin is semi-famous. He was mentioned prominantly in David
Levy's book _Hackers_. He also invented some data structures that
are used today, notably the "trie". 

-- 
+-------------------------------------------------------------+
|     Charles and Francis Richmond     <richmond at plano.net>   |
+-------------------------------------------------------------+


From cfelling at iae.nl  Wed Feb 27 22:45:34 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 04:45:34 +0100
Subject: map-like function on dict values?
References: <slrna7r0hi.bkl.huaiyu@gauss.almadan.ibm.com>
Message-ID: <a5k94u$dhr$1@animus.fel.iae.nl>

Huaiyu Zhu <huaiyu at gauss.almadan.ibm.com> wrote:
> Is there a way to do this

> for k, v in dict.items():
>     dict[k] = f(v)

> without involving a Python for loop?  I'd imagine that a function
> implemented in C could speed things up quite a bit.  Something like

what about using an iterator, like

  for k, v in dict.iteritems():
      dict[k] = f(v)

would be interesting to know whether saving to build a (huge) list would
save you the time you need.

As for the C function, for each item you'll have to get to the python
interpreter to evaluate the `f(v)' call. My gut feeling would be that
that would be costly. But as you know, gut feelings are worthless,
time it instead.

-- 
groetjes, carel


From sdm7g at Virginia.EDU  Mon Feb  4 08:32:34 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Mon, 4 Feb 2002 08:32:34 -0500 (EST)
Subject: WANTED: IBM 1401 AUTOCODING WAREZ [was: Autocoding - the resistance
 to expect from psuedo programmers]
In-Reply-To: <slrna5sudd.uef.mill@pedgr571.sn.umu.se>
Message-ID: <Pine.OSX.4.43.0202040829420.4495-100000@d-128-61-180.bootp.virginia.edu>


HEY DUDEZ !
 I WANT TO GET INTO THIS AUTOCODING THING.
 ANYONE KNOW WHERE I CAN GET SOME IBM 1401 WAREZ ?
-- MULTICS-MAN





From st-newsgroups at thomas-guettler.de  Tue Feb  5 04:16:36 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Tue, 05 Feb 2002 10:16:36 +0100
Subject: Size of integers (2**32-1)
Message-ID: <3C5FA2F4.8020905@thomas-guettler.de>

I have two python version installed:
with the cygwin version "print 2**32-1" works
but with the zope version I get OverFlowError.

Background:
I want to get a random 32-Bit integer like this randrange(0, 2*32-1)

thomas



From jeff at ccvcorp.com  Thu Feb 14 19:35:56 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 16:35:56 -0800
Subject: Finding path to __init__.py in package
References: <mailman.1013726177.3628.python-list@python.org>
Message-ID: <3C6C57EB.129E14D4@ccvcorp.com>

Lulu of the Lotus-Eaters wrote:

> What I would like is a function that asks "what directory is this
> current __init__.py script running from?"

Every module has a __file__ attribute that is the (fully qualified) filename
that the module was loaded from -- so in your example __init__, it should be
"$PYTHONPATH/gnosis/xml/pickle/doc/__init__.py"  with, of course, $PYTHONPATH
properly expanded.  (Shouldn't that be sys.prefix, though, and not
$PYTHONPATH??)  I'm not positive whether the __file__ attribute is accessible
from *inside* the module, but it should be.  In which case, all you need to
do is os.path.dirname(__file__).

Jeff Shannon
Technician/Programmer
Credit International




From emile at fenx.com  Fri Feb  8 23:31:04 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 8 Feb 2002 20:31:04 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org><a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org>
Message-ID: <a4290e$1b4hcp$1@ID-11957.news.dfncis.de>

"Cliff Wells" >
> What's wrong with Fry's?  As long as you don't listen to the
salespeople,
> that is.
>
>

... and check there's no recycle tag before buying.

--

Emile van Sebille
emile at fenx.com



From core.lists.python at core-sdi.com  Tue Feb 19 16:53:42 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Tue, 19 Feb 2002 18:53:42 -0300
Subject: MIDI
Message-ID: <03b101c1b98f$e5f7d1d0$6e42a8c0@corelabs.coresdi.com>

Does someone knows where can I find a module to play midi files from a python script?


thanks! 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020219/704d6c71/attachment.html>

From kondor at wanadoo.fr  Tue Feb  5 04:07:31 2002
From: kondor at wanadoo.fr (kondor)
Date: Tue, 05 Feb 2002 10:07:31 +0100
Subject: wxPython and Boa-Constructor
References: <88bc63c6.0202040509.74258657@posting.google.com>
Message-ID: <3C5FA0D3.2000806@wanadoo.fr>

I have python 2.2 and the last version of wxpython (http://www.wxpython.org) and 
it work fine.


Doug Farrell wrote:

> Hi all,
> 
> I have Python 2.2 and wxPython 2.3 installed on my machine. I've also
> been curious about using Boa-Constructor to generate some code and
> look at what it generates as a tutorial to me on what to do. I've
> downloaded the latest Boa-Constructor, but it doesn't like the version
> of Python/wxPython that I've got installed. Does anyone know what I
> might have to do to get it running or when it might be upgraded again?
> 
> Thanks
> Doug Farrell
> 



From paul at prescod.net  Thu Feb 14 14:14:30 2002
From: paul at prescod.net (Paul Prescod)
Date: Thu, 14 Feb 2002 11:14:30 -0800
Subject: PullDOM docs?
References: <slrna6l7i2.8hs.mlh@vier.idi.ntnu.no>  <m33d05jhzx.fsf@mira.informatik.hu-berlin.de> <slrna6o26h.fbh.mlh@vier.idi.ntnu.no>
Message-ID: <3C6C0C96.79C715DE@prescod.net>

Magnus Lie Hetland wrote:
> 
>...
> 
> But is pulldom really no more efficient thatn minidom? E.g. if I'm
> looking for a record in a Really Large XML file? I would have thought
> that not having to build a Really Large DOM tree would help
> significantly... 

Yes, it would help significantly for a Really Large XML file. That's
precisely why pulldom exists. It isn't going to be a speed demon but it
will use a tiny bit of RAM which is what you're mostly trying to save
with a Really Large XML File.

 Paul Prescod



From jwbaxter at spamcop.net  Tue Feb 19 11:05:27 2002
From: jwbaxter at spamcop.net (John Baxter)
Date: Tue, 19 Feb 2002 08:05:27 -0800
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl> <xyuvgcujsyl.fsf@sex.ifi.uio.no> <a4shr2$49s$0@216.39.172.122> <a4t9fr$500$1@odysseus.uci.kun.nl>
Message-ID: <jwbaxter-020177.08052619022002@corp.supernews.com>

In article <a4t9fr$500$1 at odysseus.uci.kun.nl>,
 "Patrick Vrijlandt" <p.vrijlandt at aig.azn.nl> wrote:

> Although telnetlib seems rather buggy, there is not much request for
> improvement on c.l.py, so I think it is barely used, or
> it works fine with most servers.

Well, telnetlib works for the limited purposes we use it for here.  Or 
does it?  I found the cause (in my code) of fairly frequent false alarms 
in a monitor I wrote using it, but we still have infrequent false alarms 
which I haven't yet understood.

  --John


From sjmachin at lexicon.net  Mon Feb 18 04:51:12 2002
From: sjmachin at lexicon.net (John Machin)
Date: 18 Feb 2002 01:51:12 -0800
Subject: python10 digest + articles at pythonjournal.cognizor.com  comprise PythonJournal 2(2)
References: <mailman.1013981368.9153.python-list@python.org> <c76ff6fc.0202171926.36131461@posting.google.com>
Message-ID: <c76ff6fc.0202180151.5e5a06e0@posting.google.com>

sjmachin at lexicon.net (John Machin) wrote in message news:<c76ff6fc.0202171926.36131461 at posting.google.com>...
> editor at pythonjournal.com wrote in message news:<mailman.1013981368.9153.python-list at python.org>...
> > Python Journal new issue 2(2) is out and about at pythonjournal.cognizor.com
> > with articles and a fairly detailed digest of the content of the 4 very full 
> > days of the 2002 Python10 conference.   Focus in the articles of this issue 
> > is on deepening understanding of how python works -- recursion, PySpy.
> > 
> > First -- The pythonista thinks about repeating functions: Alternative
> > recursion methods with iteration splitoff auxiliaries, having different oopses
> > and efficiencies. Then -- Inspect your own pet python's looping and leaping.
> > In realtime! Tutorial on Malcolmson's PySpy shows what documentation ;)
> > looks like inside out, how basic program structures including recursion
> > vs. iteration really slither along. Screenshots, congeners  (and cf.
> > inspect.py). 
> > 
> 
> The journal quotes an URL http://www.malcolmson.com/pyspy ... this cracks a 404.
> 
> http://www.malcolmson.com/ relates to a photographer named Matt Malcolmson.
> 

Correction: the journal (incorrectly) has http://www.malcomson.com/pyspy

The correct URL is http://www.malcolmson.com/pyspy


From sjmachin at lexicon.net  Thu Feb 21 15:56:55 2002
From: sjmachin at lexicon.net (John Machin)
Date: 21 Feb 2002 12:56:55 -0800
Subject: mxDateTime in Python2.2 on Tru64
References: <695dd93e.0202210758.3b64f79a@posting.google.com>
Message-ID: <c76ff6fc.0202211256.4f6953b3@posting.google.com>

richard.papworth at bbsrc.ac.uk (Richard Papworth) wrote in message news:<695dd93e.0202210758.3b64f79a at posting.google.com>...
> Hi
> 
> After upgrading from Python 1.52 I'm having trouble getting mx
> Extensions to work with Python 2.2.
> 
> Compilation went OK (although config.h was missing from the Python
> includes directory). However, when I try to load up the DateTime
> module I run into problems:

[snip]
 
>   File "/nfs3/users/postgres/usr/local/lib/python2.2/site-packages/mx/DateTime/DateTime.py",
> line 441, in ?
>     MaxDateTime = DateTime(sys.maxint / 366 - 1,12,31)
> OverflowError: signed integer is greater than maximum
> >>> 
> 
> The other mx extensions seem to be OK. Does anyone know what's going
> wrong?

Here's a *temporary* patch: change the offending line to be:

MaxDateTime = DateTime(5867440,12,31)
# magic number is result of sys.maxint / 366 - 1 on a 32-bit machine

This should keep you going for a little while ...

Hope I don't go down in history as the cause of the Y5867.44K bug.

Cheers,
John


From skip at pobox.com  Mon Feb 25 15:30:33 2002
From: skip at pobox.com (Skip Montanaro)
Date: Mon, 25 Feb 2002 14:30:33 -0600
Subject: Any Pythonistas in Mexico?
In-Reply-To: <200202220112.g1M1C0k28307@mail.aldeasystems.com>
References: <200202220112.g1M1C0k28307@mail.aldeasystems.com>
Message-ID: <15482.40681.274503.725628@12-248-41-177.client.attbi.com>

    Carlos> my company is located in Mexico and we use Python for our
    Carlos> development projects, I am trying to find out how many of us
    Carlos> Python enthusiasts live in Mexico and see if maybe we can form a
    Carlos> user group or something.

I lived in Mexico last week. ;-)

While there it occurred to me that having a Python conference in some
Mexican coastal resort (I was in Puerto Vallarta and would vote to hold it
there or someplace similar) would be fun.  I suspect it wouldn't be all that
much more expensive than holding it in the US, especially if you time it to
miss the winter/spring break crunches.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From jason at jorendorff.com  Fri Feb  8 21:04:42 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 8 Feb 2002 20:04:42 -0600
Subject: encapsulation, typing & other questions
In-Reply-To: <3C62C5C5.7FEF0B10@daylight.com>
Message-ID: <HFEKILOLEFEFMKAECNDLGEFHDBAA.jason@jorendorff.com>

> 1. Are there plans to include type support for method/function calls in
> Python?

No.

> Seems like it would lead to much faster error detection &
> correction.

It seems that way to me, too.  Many disagree.

In any case, this doesn't bother me as much in practice
as I expected it to, coming from C++ as I did.  Many things
work to offset this penalty:  Python's high-level-ness,
its wonderful standard library, its simple syntax, its
brevity, its readability, etc.

Sometimes error detection is quick but error correction is
a matter of "First, write an Adapter class..."  (Today I had
an ASP script fail because TextStream, a standard object
distributed by Microsoft, doesn't implement IStream, a
standard interface invented by Microsoft.  This is the
worst of both worlds:  the error isn't detected until
runtime *and* it's a static typing error that is
prohibitively difficult to fix.)

> 2. How easy is it to use design patterns in Python?

Extremely easy!  Here's a Singleton:

  class Logger:
      def __init__(self): ...
      def log(self, description, priority=3): ...
      def set_threshhold(self, threshold): ...

  logger = Logger()
  del Logger

Here's an implementation of Observer in pure Python:

  # An event object...
  class LineEvent:
      def __init__(self, target, line):
          self.target = target
          self.line = line

  # And an observable object that fires an event...
  class FileTailer:
      def __init__(self, file):
          self.__observers = []
          self.file = file
      def add_observer(self, obs):
          self.__observers.append(obs)
      ...
      def process(self):
          while self.has_more_lines():
              line = self.file.readline()
              for obs in self.__observers:
                  obs(LineEvent(self, line))

  # And an example of an object that observes a FileTailer.
  class FileWindow(JTextArea):
      def __init__(self, tailer):
          JTextArea.__init__(self)
          self.tailer = tailer
          tailer.add_observer(self.handle_line)
      def handle_line(self, event):
          self.text += event.line

(Jython also has special support for Java's event model,
which is not to be confused with what I'm illustrating above.
They're called "listeners" in Java, of course, and there are
other differences.)

The Distutils setup function is a Facade.

The standard library's SocketServer is rather like a
Template Method, in that it defines the common bits of
the algorithm and lets the subclass implement other bits.

Python 2.2 has new language support for iterators, so
the Iterator pattern is essentially built-in.

Python's __getattr__() provides truly remarkable support for
Proxy objects.  There are a few examples in the standard library:
Bastion (which is rather old) and weakref.proxy.  Mark Hammond's
win32com support creates Python proxies that look just like
COM objects, to the programmer.

If that's not enough to convince you, Bruce Eckel is writing
a book on the subject:  "Thinking in Python".  (It's on his
web site somewhere, but in very rough form.)

> A lack of typing suggests it wouldn't be too easy [...]

Maybe.  Or maybe static typing doesn't have anything to do with
design patterns.  Hmmmm.


> 3. Are there plans to include method & variable encapsulation in the
> object models in python?

No.

> The answer I've heard is that since it's possible to overcome
> encapsulation by long or dirty ways, it isn't worth it. But my
> comeback is that if one sticks to the rules while coding,
> this won't happen and the resulting gains are worth it.

You are right about the second part.

But the ordinary Python answer is *not* that encapsulation isn't
worth it, but rather that encapsulation by convention
is just as effective as encapsulation that's enforced by
the compiler.

Just "stick to the rules", as you say -- namely, don't call methods
named with leading underscores -- and you too can achieve
encapsulation in Python.

Your mileage may vary.


> 4. How advanced is db access in Python?

Not as good as JDBC, but then, what is?

It's a toss-up between Python's db support and Microsoft's ADO.
Python's API is probably a little better than ADO, but driver
quality seems to vary more.


> Does it use connection pooling
> and other optimizations like the JDBC drivers?

Are JDBC drivers doing connection pooling these days?

If I recall correctly, this is ordinarily done at a higher
level -- as it should be! -- by e.g. application servers.
Similarly, a Python application server would be expected
to provide connection pooling *on top of* the ordinary
database API.  Zope does this.

## Jason Orendorff    http://www.jorendorff.com/



From sjones16 at rochester.rr.com  Wed Feb 27 23:07:29 2002
From: sjones16 at rochester.rr.com (Seth Jones)
Date: Thu, 28 Feb 2002 04:07:29 GMT
Subject: cursor control
References: <3c731781.5352364@nntp.surry.net>
Message-ID: <52if8.149161$dh.38774212@typhoon.nyroc.rr.com>

Have you looked into curses?

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

This seems to be what you're looking for.


"Charles W. Strickland" <cstrickland at surry.net> wrote in message
news:3c731781.5352364 at nntp.surry.net...
> Is it possible to control cursor positioning in python without using
> TKinter?  I wanted to write a program that required it, but could not
> find anything anywhere to explain how to do it.  There were a couple
> references awhile back, but I never did see a definitive answer.  I
> finally wrote the program in C - what a mess!




From jason at jorendorff.com  Wed Feb 13 15:58:52 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 13 Feb 2002 14:58:52 -0600
Subject: Coming of age...
In-Reply-To: <iHya8.25576$l93.4762435@newsb.telia.net>
Message-ID: <HFEKILOLEFEFMKAECNDLKEGJDCAA.jason@jorendorff.com>

U.S. Patent #5,206,951 wrote:
>     An object based data processing system including an extensible
>     set of object types and a corresponding set of "object managers"
>     wherein each object manager is a program for operating with the
>     data stored in a corresponding type of object. The object managers
>     in general support at least a standard set of operations. Any
>     program can effect performance of these standard operations
>     on objects of any type by making an "invocation" request. In
>     response to an invocation request, object management services
>     (which are available to all object managers) identifies and invokes
>     an object manager that is suitable for performing the requested
>     operation on the specified type of data. A mechanism is provided
>     for linking data from one object into another object. An object
>     catalog includes both information about objects and about links
>     between objects.

This quote comes from the part at the top of the patent document,
under the heading "Abstract."

Which I thought was rather apt.

## Jason Orendorff    http://www.jorendorff.com/



From bh at intevation.de  Tue Feb 12 05:32:39 2002
From: bh at intevation.de (Bernhard Herzog)
Date: 12 Feb 2002 11:32:39 +0100
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4> <eSz98.4759$zb.260473@news1.east.cox.net> <3C682A16.1D6C800D@ccvcorp.com> <Xns91B28C2B6EBE0RASXnewsDFE1@130.133.1.4>
Message-ID: <6qk7tjm1mg.fsf@abnoba.intevation.de>

Philip Swartzleonard <starx at pacbell.net> writes:

> Then it becomes 'Guidoista' perhaps, for following a specific leader? ;)

Or, taking a different cultural context: a Guidoka. 

Gui-do: the way of the python

   Bernhard

-- 
Intevation GmbH                                 http://intevation.de/
Sketch                                 http://sketch.sourceforge.net/
MapIt!                                               http://mapit.de/


From PoulsenL at capecon.com  Tue Feb 26 11:10:09 2002
From: PoulsenL at capecon.com (PoulsenL at capecon.com)
Date: Tue, 26 Feb 2002 11:10:09 -0500
Subject: Time to rename Stackless?
Message-ID: <72F73B808A78D511A73600034771D9FF717C46@DC_EXCHANGE2>

How about High Performance Python?

Absolutely no information is conveyed whatsoever, but it is a cool name :)

Loren Poulsen

-----Original Message-----
From: Christian Tismer [mailto:tismer at tismer.com]
Sent: Tuesday, February 26, 2002 10:40 AM
To: Hamish Lawson
Cc: python-list at python.org
Subject: Re: Time to rename Stackless?


Hamish Lawson wrote:

> Christian Tismer wrote:
> 
> 
>>Well, it doesn't switch between multiple stacks, it
>>repeatedly replaces parts of the same stack.
>>(But this would have the same effect, of course).
>>
> 
> If it essentiually has that effect, how about Switching Python then as a
name?


Not bad!
Should I start a contest?

ciao - chis

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/



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



From m.faassen at vet.uu.nl  Thu Feb  7 18:09:32 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Feb 2002 23:09:32 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <slrna650nr.ahr.philh@comuno.freeserve.co.uk>
Message-ID: <a3v1fc$jmc$1@newshost.accu.uu.nl>

phil hunt <philh at comuno.freeserve.co.uk> wrote:
> One thing that puzzles me is why Tim wants help: as he says, we're
> all too stupid to understand his grand design (as well as being 
> arrogant). 

> Since he is so much cleverer and far-seeing than everyone else, he'll
> have to code it himself, as no-one else is clever enough to understand it.

No no no..while *some* of us are stupid enough to understand it, some of
us would understand it if we only tried, and in fact are ignoring 
auto-coding as it's a threat. It's a psychological thing; we're not stupid,
but we don't like to see auto-coding so we just ignore it. This is 
arrogant and pretentious, perhaps.

Anyway, besides those two categories (too dumb, unconsciously scared) there
are those of us who are smart enough to see it and can look it straight in
the eye, but deliberately suppress the knowledge because we'd be out of
a job. We'll call this the 'evil' category.

As Tim has some faith in humanity, he assumes there's a rare fourth
category that is smart enough and brave enough and is in fact *good*,
and that this category will help him create the auto-coding system.

At least that's my theory. I may be off completely.

Regards,

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


From jason at jorendorff.com  Fri Feb  1 00:04:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 31 Jan 2002 23:04:02 -0600
Subject: Java vs Python Benchmarks: Java is faster
In-Reply-To: <20020201015223.GC4394@dman.ddts.net>
Message-ID: <HFEKILOLEFEFMKAECNDLIEJICPAA.jason@jorendorff.com>

> Interestingly enough, too, there doesn't seem to be a way to increase
> the max heap size of javac and I've seen it bomb with
> java.lang.OutOfMemoryError [...]

java -cp $JAVA_HOME/lib/tools.jar -Xmx... sun.tools.javac.Main ...

> A little while back I wrote a script in python to obfuscate package
> and class names in a java source tree.  (for in-house use at my
> current employer)  I used CPython 2.2 and a bunch of regexes for
> identifying the occurences of the class/package names in the source
> files.  It can process the whole 15MB source tree in about 30 seconds
> and using no more than 7MB RAM at peak with 2-3MB most of the time.
> I'll likely have to recode it in java (no one else there speaks
> python), but I'll bet it won't be nearly as good (both in
> terms of read/write-ability of the code and performance).
> 
> Any comments?

1.  You can buy a real obfuscator for a few hundred bucks.

2.  Whatever you buy will be written in Java.  (And it will
    probably work at the class-file level rather than the
    source level, which is a good thing.)

3.  If you must do this yourself:  ANTLR (for example) comes
    with everything you need to tokenize Java source code (or
    fully parse it, if you prefer).  It should be quite speedy,
    and more foolproof than slinging regular expressions.

## Jason Orendorff    http://www.jorendorff.com/



From lac at strakt.com  Wed Feb 27 11:12:34 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 27 Feb 2002 17:12:34 +0100
Subject: SSL Server Sockets? 
In-Reply-To: Message from "Thomas Weholt" <thomas@gatsoft.no> 
   of "Wed, 27 Feb 2002 13:52:15 GMT." <jw5f8.129$5vh.170404352@news.telia.no> 
References: <a5ijos$1565$1@news.idiom.com>  <jw5f8.129$5vh.170404352@news.telia.no> 
Message-ID: <200202271612.g1RGCYtB026632@ratthing-b246.strakt.com>

Martin Sj?gren (martin at strakt.com) made pyOpenSSL work for servers.  
See http://pyopenssl.sourceforge.net There is a debian package.

Laura Creighton




From quinn at regurgitate.ugcs.caltech.edu  Tue Feb 19 16:51:41 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: Tue, 19 Feb 2002 13:51:41 -0800
Subject: Adding static typing to Python 
In-Reply-To: Your message of "Tue, 19 Feb 2002 11:43:17 PST."
             <Pine.LNX.4.33.0202191140530.5828-100000@ns0.tbye.com> 
Message-ID: <20020219215146.CF6548046A@regurgitate.ugcs.caltech.edu>

> On 19 Feb 2002, Quinn Dunkan wrote:
> > Or you can just look and see how the 'person' is used in the function.  If 
> the
> > first statement is a 'o = db.by_id[person]' then you know it's an id.  If i
> t
> 
> An even better approach is to ask yourself: if it's a person's ID, then
> why isn't the variable named personID?

I didn't say that because the OP could have objected that "then you're just
encoding the type information in the variable name".

And I would agree with him to an extent.  Naming variables based on their type
leads to the same problem you get in static systems, when the type becomes
obsolete.  The variable doesn't actually have to be an id, it just has to be
something that you can look up in the db.  So the actual type requirements are
in whoever is calling the function, and the db object.  They could both change,
and now 'personID' is misleadingly named.  (of course, 'db.by_id' would then
also be misnamed, but lets assume it was a more generic 'db[...]')

Taken to its logical extreme, this seems to imply that all variable names
should be quite vague if the function is just passing it along, which could get
silly.  But taking things to their logical extremes often does get silly :)



From paul at boddie.net  Tue Feb 19 07:34:16 2002
From: paul at boddie.net (Paul Boddie)
Date: 19 Feb 2002 04:34:16 -0800
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi>
Message-ID: <23891c90.0202190434.147ee6d1@posting.google.com>

Ville Vainio <vvainio at karhu.tp.spt.fi> wrote in message news:<yoxr8niyiyf.fsf at karhu.tp.spt.fi>...
> 
> Direct connection doesn't work in my case (my DBMS only offers ODBC
> and JDBC). However, I overreacted a little bit w/ mxODBC - I thought
> it was $500 (as it used to be), but I was emailed that it is in fact
> $75.

You could always make a SWIG wrapper for iODBC - I started one to see
how hard it might be - but in the end it's probably cheaper (time is
money, so they say) to license mxODBC.

Paul


From erno-news at erno.iki.fi  Fri Feb  1 08:41:20 2002
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 01 Feb 2002 15:41:20 +0200
Subject: What is a "bus error"?
References: <qj658.1752$HL4.244689@newssrv26.news.prodigy.com> <a32tpp02vve@enews3.newsguy.com> <3C56448B.58833301@engcorp.com> <Pine.LNX.4.43.0202011202560.4970-100000@zeon.well.ox.ac.uk>
Message-ID: <ku3d0l2ubj.fsf@lasipalatsi.fi>

In article
<Pine.LNX.4.43.0202011202560.4970-100000 at zeon.well.ox.ac.uk>, Kenny
<kccw at well.ox.ac.uk> writes:

| 	I recently installed Python 2.2-5 from Fink (Unix Software for
| MacOSX) and tried running a few sample scripts/programs.
| 	However, some of these programs now give a "bus error" message -
| what does this mean?

it means python got a SIGBUS signal from the kernel, ie it crashed.
SIGBUS commonly results from attempting to access unaligned memory.

so it seems that there is a bug in the fink packaging of python, or
python itself. you could try asking the fink people and/or compiling
python yourself and seeing if that works. (if it doesn't, file a
bug in the python bug tracker.)

  -- erno


From jimd at vega.starshine.org  Sat Feb  2 17:54:58 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 2 Feb 2002 22:54:58 GMT
Subject: addition to MLab: randperm
References: <a2972632.0201171005.15e660e3@posting.google.com>
Message-ID: <a3hqo2$nsr$1@news.idiom.com>

In article <a2972632.0201171005.15e660e3 at posting.google.com>, les ander wrote:

> I needed the matlabs randperm function so i wrote one and 
> wanted to see if there is a better way. I think this function would
> be a good addition to MLab:

> # given a number n, return an array that is a shuffling of 
> # integers from range 0 to n 

 I must be missing something in your requirements.
 What's wrong with:

	 def randperm(n):
		import random
		t = range(n)
		random.shuffle(t)
		return t


> def randperm(n):
>  import MLab,Numeric
>  r=MLab.rand(n)
>  dict={}
>  for i in range(n):
>     dict[ r[i] ]=i
>  r=Numeric.sort(r)
>  out=Numeric.zeros(n)
>  for i in range(n):
>     out[i]=dict[r[i]]
>  return out


> i encourage other people to post the code if they write
> more matlab type functions using the Numeric package



From syring at email.com  Sun Feb  3 07:26:38 2002
From: syring at email.com (Karl M. Syring)
Date: Sun, 3 Feb 2002 13:26:38 +0100
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net>
Message-ID: <a3jbna$18pllq$1@ID-7529.news.dfncis.de>

"wrbt" <whoRunBartertown at masterBlaster.net> schrieb
> Hi and hello all, I'm the latest Python convert... 2 months and counting.
> 
> I LOVE IT.
> 
> You'll see me around.

You should know about the Python that eats anonymous posters for breakfast.

Karl M. Syring




From alain at onesite.org  Mon Feb 25 22:17:30 2002
From: alain at onesite.org (Alain Tesio)
Date: Tue, 26 Feb 2002 04:17:30 +0100
Subject: Problems with select
Message-ID: <pan.2002.02.26.04.17.29.538545.11917@onesite.org>

Hi, I can't manage to make select work, in both 1.52 and 2.1.2

I just want to read stderr and stdout when executing a command
with popen3, and I must use select, read() blocks when there is
more than a fixed amount of data (4 KB I think)

The problems is that select still gives something after the end
of the file, or sometimes never returns anything at all.

I create the pipes with :

(pipe_out,pipe_in,pipe_err)=popen2.popen3(command)

and use select like this:

pollstatus=select.select([],[pipe_out],[pipe_err],1)


(the doc says : select(iwtd, owtd, ewtd[, timeout])
I didn't put pipe_in as it's closed)


Here is my test program:

========================
#!/usr/bin/python

import sys,select,popen2,string

def execute(command,maxiter):
	(pipe_out,pipe_in,pipe_err)=popen2.popen3(command)
	pipe_in.close()
	pipes=[pipe_out,pipe_err]
	gotsomething=1
	i=0
	while gotsomething:
		gotsomething=0
		i=i+1
		if i>maxiter:
			print "maxiter reached"
			sys.exit(0)
		pollstatus=select.select([],[pipe_out],[pipe_err],1)
		print "select returns : ",pollstatus
		for (pollitem,label,pipe) in ((1,"out",pipe_out),
									  (2,"err",pipe_err)):
			if pollstatus[pollitem]:
				print "%i on %s : '%s'" % (i,label,string.strip(pipe.readline()))
				gotsomething=1

execute("cat 3lines",10)
========================

And the ouput continues after the file is exhausted (readline doesn't
fail), same problem with no timeout.

04:09:32 alain ~/tmp $wc -l 3lines
      3 3lines
04:09:32 alain ~/tmp $./test_select.py
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
1 on out : 'first'
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
2 on out : 'second'
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
3 on out : 'third'
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
4 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
5 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
6 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
7 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
8 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
9 on out : ''
select returns :  ([], [<open file '(fdopen)', mode 'r' at 0x8060480>], [])
10 on out : ''
maxiter reached


And sometimes (really randomly, about 1 time out of 10) it just gives
nothing, with no timeout it hangs forever.

04:11:28 alain ~/tmp $./test_select.py
select returns :  ([], [], [])


What did I miss ?
I've succeeded with select.poll but I must use select.select to have my
program usable with python 1.52 (still in Debian stable)


Thanks
Alain


From sjmachin at lexicon.net  Tue Feb 12 21:10:44 2002
From: sjmachin at lexicon.net (John Machin)
Date: 12 Feb 2002 18:10:44 -0800
Subject: Reading and writing C structs from Python
References: <3C696FC7.1090600@erols.com>
Message-ID: <c76ff6fc.0202121810.296bb1b8@posting.google.com>

"Edward C. Jones" <edcjones at erols.com> wrote in message news:<3C696FC7.1090600 at erols.com>...
> Suppose I have a C struct for example

Initial reaction is to say "Don't do that". There is no answer that is
both helpful and easy. Trying to establish how many degrees of freedom
you have:
Are you wrapping *existing* C code as a Python extension? 
Are you planning to write some C code whose sole purpose is to be
called from Python as an extension?
Are you embedding Python in an existing application?
Who will use it? (You / colleagues / universe)
Multi-platform ? E.g. is the interface meant to work for all
reasonable values of sizeof(int)?
One-off throw-away or critical 24x7 app?

> 
> typedef struct {
> float x;

float or double?

> int n[4];
> } test;
> 
> and a pointer to an instance of the struct
> 
> test* t = (test*) malloc(sizeof(test));

If possible, use PyMem_Malloc() and friends.

> 
> I would like to wrap this in a python object "py_t" so that (in Python) 
> "x = py_t.x" reads "t.x" and "py_t.x = 13.0" writes to "t.x".

Do you mean not "t.x" but "t->x"?

> Ditto for "py_t.n[i]".

So n is to be interpreted by the Python side as a list, not a tuple,
one gathers. What would you like to happen if the Python side does:

py_t.n[0] = x where x is not an int

py_t.n.append(any_object)

py_t.n.extend(some_list)

etc etc

Is it possible that you can regard your struct as logically an
instance of a class and in OO fashion provide methods to do whatever
the object should do, rather than letting the Python side peek and
poke at the attributes?

> Ideally I would like to do this using a C function 
> something like:
> 
> PyObject* PyObject_FromCStruct(void* t, char* format)
> void* PyObject_AsCStruct(PyObject* pyo, char* format)
> 
> The format strings used in module "struct" would be fine.

Yes, but if you really need to handle arrays, you would need to
introduce brackets, so that you can describe your example struct as
"f[iiii]".

> 
> My guess is that this is a job for

I guess otherwise.

> the notoriously confusing "buffer interface".

Ha! Common ground at last :-)

> Does anyone have any sample code for problems like this?

If you really want to do the to/from struct thing, then perusal of the
source of (a) the struct module (b)PyArg_ParseTuple() and
Py_BuildValue() might be a good start.


From wookiz at hotmail.com  Mon Feb 11 04:17:40 2002
From: wookiz at hotmail.com (wooks)
Date: 11 Feb 2002 01:17:40 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com> <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu> <8852d33c.0202100022.70a05925@posting.google.com> <slrna6dqet.dm.quinn@hurl.ugcs.caltech.edu>
Message-ID: <8852d33c.0202110117.374f062@posting.google.com>

quinn at hurl.ugcs.caltech.edu (Quinn Dunkan) wrote in message news:<slrna6dqet.dm.quinn at hurl.ugcs.caltech.edu>...
> 
> Polymorphism has traditionally been a big deal for static languages, because
> their compile time typechecking machinery needs to be powerful enough to
> generate polymorphic signatures.  Dynamic languages don't have compile time
> typechecking machinery, so it's not an issue.
> 
> For that matter, '+' works on both floats and ints in C, so it's polymorphic
> there too.  But if you wanted to *write* a function like '+' in C, it would
> want you to declare the types of the arguments, at which point you'd be stuck
> between saying "add() takes two ints" and "add() takes two floats" (in the C
> case, ints have built in magic that let them automatically turn into floats,
> but that's just for numbers (and I assume to make C's lack of polymorphism more
> bearable) and not all static languages do automatic conversion like that (e.g.
> in haskell "let {x=1::Int; y=2.0} in x+y" is an error)) so you can see why
> static languages make a ruckus about supporting (or not) polymorphism, while
> dynamic programmers might not even know what it is, but use it all the time
> anyway.

If I am getting you right the rest of the post translates what you
said in the first sentence.

Further if I get magic methods right, they are special functions which
if present enable the object to apply a unique interpretation of a
python operator.
So + has standard behaviour but will  adopt whatever is specified in
the __add__ function if one is present for the class.

Yes?


From sdm7g at Virginia.EDU  Mon Feb  4 21:32:21 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Mon, 4 Feb 2002 21:32:21 -0500 (EST)
Subject: [osx] milter problem
In-Reply-To: <644f6688.0202041801.7b0a5968@posting.google.com>
Message-ID: <Pine.OSX.4.43.0202042131160.317-100000@d-128-61-180.bootp.virginia.edu>


On 4 Feb 2002, rdack wrote:

> osx 10.1.2, python2.1, milter-0.3.9
> i have installed the milter ok, but when i do:
> python sample.py
> i get:
> # python sample.py
> dyld: python Undefined symbols:
> _pthread_sigmask
> _sigwait
>
> any ideas?

Maybe you need:  "-lpthread"  added to the cc switches. (?)

-- Steve Majewski





From tripps81 at yahoo.com  Tue Feb 19 11:58:46 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Tue, 19 Feb 2002 23:58:46 +0700
Subject: python threads and perl/ruby threads? (Re: basic books/guides
  on multithreading programming?)
In-Reply-To: <7934d084.0202172009.75324c2a@posting.google.com>
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154>
 <mailman.1013816158.26552.python-list@python.org>
Message-ID: <5.1.0.14.2.20020219235359.07ab8e50@202.95.131.154>

hi andrea,

At 18/02/2002 11:09, Andrae Muys wrote:
>Cliff:

yes, i'm afraid you confused me with cliff.

>Sorry I can't answer your question straight away, maybe check
>your local library (possibly uni-library) for a copy of Ben-Avi.  If
>you have any problems mapping the concepts he discusses to specific
>Python code I (and I imagine the rest of c.l.py) will do what I can to
>help.  Other then that, WHY do you feel you need threads?  Or is this
>just a "Something new to learn" question?

both reasons, i think. :-) you don't know what you want or need 
(or don't) unless you get to know more about the thing, right?

another question (or, thread): what differentiates python 
threads with perl and ruby threads, in the current 
implementations of the languages? i believe python use the 
native threads whenever possible, is that correct? i know that 
ruby always provides threads, for example, so that means it 
implement its own threads.

t




From mwh at python.net  Thu Feb 21 06:19:32 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 11:19:32 GMT
Subject: Idiots guide to code objects, frame objects and settrace?
References: <VoAc8.7320$k91.367319@wards> <slrna77r4h.dg2.quinn@schilling.ugcs.caltech.edu>
Message-ID: <ueljfm64s.fsf@python.net>

quinn at schilling.ugcs.caltech.edu (Quinn Dunkan) writes:

> On Tue, 19 Feb 2002 22:28:50 -0000, Tim Rowe <tim at eltig.freeserve.co.uk> wrote:
> >Is there an idiots guide to code objects, frame objects and settrace
> >anywhere? I want one program to monitor the execution of another. I've
> >looked at coverage.py from the Vaults of Parnassus, but can't make enough
> >sense of it to adapt it to do what I want.
> 
> A combination of the langref and interactive mucking around worked
> just fine for me.  They're not very complicated.  The only possibly
> confusing bit is that tb_next points downstream and f_back points
> upstream... or is it the other way around?

tb_next points "cross-stream" more than anything doesn't it?  It's
almost always None, anyway (if I'm thinking of the right field).

Cheers,
M.

-- 
  If Unicode is a horde of zombies with flaming dung sticks, 
  the hideous intricacies of JIS, Chinese Big-5, Chinese 
  Traditional, KOI-8, et cetera are at least an army of ogres 
  with salt and flensing knives.        -- Eric S. Raymond, python-dev


From dalke at dalkescientific.com  Thu Feb 21 14:27:48 2002
From: dalke at dalkescientific.com (Andrew Dalke)
Date: Thu, 21 Feb 2002 12:27:48 -0700
Subject: What values are considered false?
References: <mailman.1014312336.21048.python-list@python.org>
Message-ID: <a53i07$njv$1@slb2.atl.mindspring.net>

Harvey Thomas:
> The empty function is not only false, it is also None
>
> In the interpreter,
>
> def foo():pass
>
> foo() is None
>
> 1

Not quite.  You called the fuction 'foo', which doesn't
define a return value so the value None is used instead.
To compare the function to None, do

  foo is None

and you'll get 0.

                    Andrew
                    dalke at dalkescientific.com





From kosatab at vscht.cz  Wed Feb 27 09:45:02 2002
From: kosatab at vscht.cz (Beda Kosata)
Date: Wed, 27 Feb 2002 15:45:02 +0100
Subject: list vs. dict
Message-ID: <3C7CF0EE.3010009@vscht.cz>

Hi,
for a piece of code I need to store relatively large amount of records 
(100-1000). Each record will contain few (10-20) pieces of information 
(mostly strings or integers).
For convenience it would be better for me to make this records as dicts, 
so I don't have to remember what data has what position in the list. 
However the most important for me is speed and I suppose that working 
with lists is faster than with dicts. So my question is - how much do 
lists and dicts differ in speed? The most important is the speed of 
comparison ( (dict == dict) vs. (list == list)) and also the speed of 
setting and retrieving data (dict[key] = a vs. list[key] = a) and (a = 
dict[key] vs. a = list[key]).
Does anyone have some experience? How big the difference between the two 
is? I have done some small tests and it seems its not a big difference, 
but I don't have any real-world data.

Thanks in advance

			BEDA



From BPettersen at NAREX.com  Mon Feb  4 13:32:38 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Mon, 4 Feb 2002 11:32:38 -0700
Subject: module from BLOB
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158CF8@admin56.narex.com>

> From: Fredrik Lundh [mailto:fredrik at pythonware.com] 
> 
> marcin andrzejewski wrote:
> > is it possible to import module directly from database 
> BLOB/CLOB field 
> > ?
> 
> something like this could work:
> 
> import imp, marshal
> from my_database_api import fetch_blob_from_database
> 
> def import_from_blob(name):
>     # fetch code object
>     code = marshal.loads(fetch_blob_from_database(name))
>     # create a new module object
>     module = imp.new_module(name)
>     # execute the code in the module's namespace
>     exec code in vars(module)
>     return module

Don't you also have to say:

  sys.modules[name] = module

Or am I missing something?

-- bjorn



From justin at iago.org  Tue Feb 19 13:32:44 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 19 Feb 2002 13:32:44 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <a4k2si$1skr$1@norfair.nerim.net> ("Gillou"'s message of "Fri,
 15 Feb 2002 23:49:44 +0100")
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
	<a4k2si$1skr$1@norfair.nerim.net>
Message-ID: <x7r8nh725v.fsf@ra.iago.org>

"Gillou" <nospam at bigfoot.com> writes:

>>   Instead, we could make make two Python
>>   distributions, a "full" install containing everything and a "lite"
>>   install containing only the interpreter and maybe -- *maybe* -- a
>>   couple of essential modules like os and sys. Then people developing
>>   Python programs like, say, a pygame-based game, could distribute a
>>   single package containing the "Python lite" distribution, their own
>>   code, and any other modules their code depended on -- all in one
>>   easy-for-the-end-used-to-install package.
>
> And when downloading another one he gets the python interpreter again, and a
> new copy of some packages...
> No, Python deserves a dedicated RPMfind  like utility that manages
> dependencies (OS, python version, required modules...).
> This would be a really great extension for distutils.

A standard or easily-available "lite" distribution would be far better
for use in applications.  Having the application use its own copy of
the runtime system is a big win for reliability, as the application
programmer can be sure that the application is running on exactly the
same version/build of Python that they tested with.

-Justin

 




From shalehperry at attbi.com  Tue Feb 19 17:28:13 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Tue, 19 Feb 2002 14:28:13 -0800 (PST)
Subject: newbieish question
In-Reply-To: <c2Ac8.26231$zJ3.1876517@typhoon.maine.rr.com>
Message-ID: <XFMail.20020219142813.shalehperry@attbi.com>

On 19-Feb-2002 Mike Carifio wrote:
> Is there a way to turn off the interpretation of "\n" on sys.platform ==
> "win32"
> when writing to a file?
> 
> Here's an example:
> 
> fh = open("a_file.text", "w") # say
> fw.write("\n")  # writes \r\n on win32
> fw.write(chr(10))  # also appears to write \r\n on win32; can I get just
> line feed?
> 

open the file in binary mode.  The newline interpolation is only done in ascii
mode.



From gh_pythonlist at gmx.de  Thu Feb 21 22:48:35 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Fri, 22 Feb 2002 04:48:35 +0100
Subject: Embedding Python in C#
In-Reply-To: <3c75b88c$1_5@news1>
References: <3c75b88c$1_5@news1>
Message-ID: <20020222034834.GA6456@lilith.hqd-internal>

Le 21/02/02 ? 22:19, Adonis Vargas ?crivit:
> has this been done?

No idea.

> is it even possible?

I think so. The question is, how hard is it really?

> if it has where can i obtain information on doing so?

AFAIK you can still use or link to native "unsafe code" in MSIL compiled
programs. So might be able to link to the normal Python DLL.

No, there is no Python.NET except a proof-of-concept and this is not
likely to change in the forseeable future.

Another approach could be to reuse the good old Windows Scripting Host
with Python/COM. IIRC the .NET framework has a COM bridge.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: -3.3 ?C      Wind: 2.3 m/s



From vdkamp at inter.nl.net  Sat Feb  9 13:30:37 2002
From: vdkamp at inter.nl.net (Peter HJ van der Kamp)
Date: Sat, 09 Feb 2002 19:30:37 +0100
Subject: Python MySQL and MAC OS X
Message-ID: <3C656ACD.1020907@inter.nl.net>

Hello,

I'm trying to install the Python MySQL module, without success.
According to the MAC OS X documentation of Python 2.0 I have
to drop the setup.py file onto the interpreter and pass the
build argument by setting a Unix style command line (which can
be set by holding the option key during interpreter startup). But
this procedure leads to a -43 error, file not found. Anyone any
idea how to build and install the module?

Thanks.

Peter van der Kamp



From nikolai.kirsebom at siemens.no  Thu Feb  7 11:12:31 2002
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Thu, 07 Feb 2002 16:12:31 GMT
Subject: String Resource Viewer (DLL)
Message-ID: <3c62a715.190928891@news.mch.sni.de>

I've made a small tool which I'm using to inspect string resources in
DLL files.  The tool uses some win32all featrues and the GUI is based
on wxPython.

Source and prebuilt versions are available at
www.micon.no/PythonPage.asp

Use/modify as you please.

Any comments are welcome.
Nikolai Kirsebom



From cmaloney at physics.ucsb.edu  Tue Feb 26 12:20:21 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: 26 Feb 2002 09:20:21 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014311147.15619.python-list@python.org> <3C7A751D.1080703@physics.ucsb.edu> <a5ekpq$n83$1@bob.news.rcn.net>
Message-ID: <6bf54244.0202260920.5920efc6@posting.google.com>

Hi David.

Since I'm not a "computer guy" per se, I was wondering if you could 
clear up some more of my confusion.  I don't know much about how 
compilers work, but I would presume that any compiler (eg g++) would
create some internal representation, similar to the .cil file, after
parsing but before compiling.  

Presumably the EDG front end is geared toward producing this information
in a format more suited to being used by the "wrapper generators" in contrast 
with a compiler which produces this information only for itself and with 
the sole goal of compiling an object file?  

What I'm getting at is that I'm confused as to why people wouldn't have
used information from the compiler *itself* to automate the process of
wrapper generation.  Obviously the compiler writer people would have to
collaborate with the wrapper generator people... but there must be technical
obstacles?

Cheers,
Craig 

"David Abrahams" <david.abrahams at rcn.com> wrote in message news:<a5ekpq$n83$1 at bob.news.rcn.net>...
> I've been interested in systems for automatically wrapping C++ and
> generating documentation for a while, so I downloaded the PDT (the parser
> toolkit used by SILOON) and tried it on one of my source files. After lots
> of hacking, I was able to get the EDG front-end component to parse my source
> and generate an intermediate language (.cil) file, but the PDT executable
> which is supposed to interpret the .cil and produce a high-level description
> of the code crashed with an assertion:
> 
> Assertion failed: ptr->assoc_template, file c:\documents and
> settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-tau\il_display_tau.
> c, line 3047
> 
> Since the PDT doesn't come with source, there was no way for me to debug/fix
> the problem.
> Furthermore, the PDT developers don't seem to want to be reached. Their
> contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/) says:
> 
>     To contact the developers of PDT, please use the "Comments" section in
> the download page.
> 
> But of course, there is no "Comments" section on the download page. :=(
> 
> -Dave
> 
> "Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
> news:3C7A751D.1080703 at physics.ucsb.edu...
> > Hi Craig.
> >
> > Since your site specifically lists g++ 2.95 and NOT 3.x, am I to infer
> > that there will be problems with the new ABI?  -- Or any problems that
> > will, in practice or principle, cause conflicts with g++ 3.0?
> >
> > *Holds breath... and crosses fingers...*
> >
> > --Craig
> >
> > rasmussn at lanl.gov wrote:
> >
> > > On Wednesday, February 20, 2002, at 04:48 PM, David Abrahams wrote:
> > >
> > >>> As I understand it, the SILOON project uses a commercial parser known
>  as
> > >>> PDT from uoregon.  Have you, or has anyone else had any experience
>  with
> > >>> the SILOON/PDT project?  On the face of it, my guess would be that you
> > >>> would be somewhat skeptical of their claims?
> > >>
> > >>
> > >> I haven't had any experience with it. I just found that site. My
> > >> skepticism
> > >> just dropped by a factor of 8. If they are really using the EDG
> > >> front-end as
> > >> they claim to, then I am sure the parser is excellent. I do have to
> > >> wonder
> > >> who is paying for the FE. The issues of dealing with the semantics of
> > >> what
> > >> they've parsed remain, but these guys at least seem to be doing the
> > >> parsing
> > >> part right.
> > >
> > >
> > > It is my understanding that an educational or research license to the
>  EDG
> > > front end is fairly easy to obtain.  The University of Oregon has
>  obtained
> > > the right to distribute the EDG front-end as binaries and does so for
> > > several
> > > platforms in the Program Database Toolkit (PDT).  So for anyone wanting
> > > interface level information from C++ or Fortran 90 source files, the PDT
> > > is an excellent tool.  We use it in several projects.
> > >
> > > See http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/ and
> > > http://www.acl.lanl.gov/siloon/ for more information.
> > >
> > > Craig
> > >
> > >
> >
> >


From pobrien at orbtech.com  Thu Feb 14 11:58:14 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Thu, 14 Feb 2002 10:58:14 -0600
Subject: The joys of weakrefs... references to instance methods (callbacks)
In-Reply-To: <3C6B614C.7040003@geocities.com>
Message-ID: <NBBBIOJPGKJEKIECEMCBOEIKMCAA.pobrien@orbtech.com>

Cool. Since you mentioned ZODB, I'm collecting samples to get a better
understanding of how people are using ZODB outside of Zope. If you (or
anyone else for that matter) have any ZODB code you'd be willing to share,
please send it my way. Thanks.

---
Patrick K. O'Brien
Orbtech

> -----Original Message-----
> From: Mike C. Fletcher [mailto:mcfletch at geocities.com]
> Sent: Thursday, February 14, 2002 1:04 AM
> To: pobrien at orbtech.com; Python List
> Subject: Re: The joys of weakrefs... references to instance methods
> (callbacks)
>
>
> Actually, with the work I just finished, I replaced all watcher
> instances with property objects and call dispatcher directly, so I no
> longer need the watchers at all :) .  The version up on the site is the
> old one, the new one is rather trivial (which is why I got rid of it in
> my code):
>
[snipped code]
>
> Anywho, need to get back to work. This whole Python
> 2.2-on-ZODB-with-automatic-event-generation excursion has sucked up
> almost 4 full days of working time.
>
> Enjoy,
> Mike




From logiplexsoftware at earthlink.net  Wed Feb 13 20:34:02 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 13 Feb 2002 17:34:02 -0800
Subject: Python License Issues
In-Reply-To: <20020213191640.13956.qmail@mail.yifansoft.com>
References: <20020213191640.13956.qmail@mail.yifansoft.com>
Message-ID: <20020213173402.5cc46966.logiplexsoftware@earthlink.net>

On Wed, 13 Feb 2002 19:16:40 GMT
christoforou at gaaros.com wrote:

> Hello from rainy London
> 
> We have an issue with the python license which basically boils down to
this: 
> 
> If we ship an application that uses python, is the user of our
application 
> bound by the python license agreement (PSF)? It seems so given clause 8
of 
> the PSF: 
> 
>    8. By copying, installing or otherwise using Python 2.1.2, Licensee
>       agrees to be bound by the terms and conditions of this License
>       Agreement. 
> 
> If such is the case we will encounter problems with clients that do not
like 
> clause 4 of the PSF which states: 
> 
>  4. PSF is making Python 2.1.2 available to Licensee on an "AS IS"
>     basis.  PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
>     MPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND
>     DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR
FITNESS
>     FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON 2.1.2 WILL NOT
>     INFRINGE ANY THIRD PARTY RIGHTS. 
> 
> Some clients demand a certain level of 'protection' from situations of
gross 
> misconduct, say python has stolen major chunks of code from other apps
and 
> thus exposing them to huge legal risks. We can consider accepting
liability 
> on all third party code we use, for such gross misconduct situations but
is 
> it technically possible to do so given the PSF license? Ideally we would 
> like to negotiate a single license agreement with these clients (our
license 
> agreement) which provides additional liability over and above to that
which 
> PSF provides. However given clause 4 it seems that the client *must* 
> directly be bound by the PSF license.


Not a lawyer, but it seems clear that the license is suggesting that the
_PSF_ is not liable for these things.  I don't see where it stops anyone
else, Christ-like, from accepting responsibility for them. 
It seems true that the client will also be held to this license: they can't
hold the PSF responsible either.  But if your client has an agreement
holding _you_ responsible, I don't see where the Python license has any
relevance.

Further, I doubt that disclaimers regarding gross misconduct amount to
much, but if they do, I hereby disclaim any responsibility or knowledge of
anything bad I have ever done or might do in the future ;)

Besides, if there were any future legal issues regarding Python, I imagine
Guido would already know about it ;)


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From editor at pythonjournal.com  Sun Feb 17 18:28:18 2002
From: editor at pythonjournal.com (editor at pythonjournal.com)
Date: Sun, 17 Feb 2002 16:28:18 -0700
Subject: python10 digest + articles at pythonjournal.cognizor.com
  comprise PythonJournal 2(2) 
Message-ID: <E16cYrA-00033e-00@harrier.prod.itd.earthlink.net>

Python Journal new issue 2(2) is out and about at pythonjournal.cognizor.com
with articles and a fairly detailed digest of the content of the 4 very full 
days of the 2002 Python10 conference.   Focus in the articles of this issue 
is on deepening understanding of how python works -- recursion, PySpy.

First -- The pythonista thinks about repeating functions: Alternative
recursion methods with iteration splitoff auxiliaries, having different oopses
and efficiencies. Then -- Inspect your own pet python's looping and leaping.
In realtime! Tutorial on Malcolmson's PySpy shows what documentation ;)
looks like inside out, how basic program structures including recursion
vs. iteration really slither along. Screenshots, congeners  (and cf.
inspect.py). 

PySpy's 5th generation architecture mixes several programming languages like
many of the python10 works.  Expand to the level you want, then watch the
changing of your 
variables and function calls as python proceeds along its train of
thought.... With 
its introspection, PySpy teases us with the possibility of python 
consciousness.  Great study tool.  

NOTE:  Usually at <A HREF="http://pythonjournal.com">
http://pythonjournal.com</a>, look for it presently at <A
HREF="pythonjournal.cognizor.com">***pythonjournal.cognizor.com***.  NB.

Dr. S. Candelaria de Ram
        Editor-in-Chief, PythonJournal.com
        editor at pythonjournal.com





From skip at pobox.com  Thu Feb 14 14:17:53 2002
From: skip at pobox.com (Skip Montanaro)
Date: Thu, 14 Feb 2002 13:17:53 -0600
Subject: zlib.so and libz.so and Python
In-Reply-To: <u6nuqr108os0ae@corp.supernews.com>
References: <u6niisr909gtd6@corp.supernews.com>
        <mailman.1013703624.30857.python-list@python.org>
        <u6nuqr108os0ae@corp.supernews.com>
Message-ID: <15468.3425.126059.104317@beluga.mojam.com>

    DeepBleu> Where do I get the libz.so from?

http://www.gzip.org/zlib/

Cheers,

Skip



From tim.one at home.com  Mon Feb 18 15:28:12 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 18 Feb 2002 15:28:12 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: <oqofim3dei.fsf@titan.progiciels-bpi.ca>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com>

[Aahz]
> I think the occasional non-English post simply reflects the
> international nature of Python.

[Fran?ois Pinard]
> Exactement.  Bravo!  Bien dit! :-)

Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
czignat?  Classifero en bragna!

tims-speak-their-own-language-ly y'rs  - just another tim




From shredwheat at attbi.com  Fri Feb 15 00:35:10 2002
From: shredwheat at attbi.com (Pete Shinners)
Date: Fri, 15 Feb 2002 05:35:10 GMT
Subject: Recommendations for cross-platform GUI Bindings ?
References: <xBMa8.217$B92.52511@news.xtra.co.nz> <ictn6uomd63papjtut186c67lickq0cjj9@4ax.com>
Message-ID: <3C6C9F4E.3050905@attbi.com>

Courageous wrote:
> For low level graphics, pygame is the best of the lot. It's on
> top of SDL.

pygame's great for graphics. but i'll warn you now there's no standard 
GUI for it. there's a couple options out there, but odds are you may 
need to 'roll your own' little ui stuff. not hard for little buttons and 
checkboxes, but gets to be a lot of work for text input and list boxes :]



From gcordova at hebmex.com  Fri Feb 15 10:07:40 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 09:07:40 -0600
Subject: Thread and CGI
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A80@SRM201002>

> 
> Hi All,
> 
> The CGI I request run on a long time displaying regulary a 
> point for the advance. If I stop the page with the "stop
> button in IE", I stop the job !!
> So I would like to print "Your job is in background now. 
> Thanks for your request."
> 
> The Thread approch is may be an alternative ??
> 
> Have you examples of thread approach if it is good one or 
> others tips ?
> 
> FR
> 

What I did once, when I had such a problem (maybe a bit
worse, because in my case the browser timed-out) was to
send the message to the browser, "thank you, an acknowledge
will be sent to you shortly", re-open all standard
filehandles (stdin, stdout, stderr) to "/dev/null",
and then fork.

The parent process then exits, and the child process lowers
it's priority, and does the work. Upon finishing, it
dispatches an email to the used with the acknowledgment.

It worked pretty nicely :-)

Remember:
	REOPEN STDIN, STDOUT, STDERR TO /dev/null !!!!

I know this is a python group, but I originally did this
in perl, so I'm posting the bit I used:

Because, if not, the child won't "daemonize", and the
parent process will hang until the child finishes, or
the browser times-out (upon which the child will be
summarily terminated).

Good luck!

-gustavo



From PoulsenL at capecon.com  Tue Feb 26 13:02:25 2002
From: PoulsenL at capecon.com (PoulsenL at capecon.com)
Date: Tue, 26 Feb 2002 13:02:25 -0500
Subject: Time to rename Stackless?
Message-ID: <72F73B808A78D511A73600034771D9FF717C47@DC_EXCHANGE2>

Christian Tismer wrote:

...

>Yes, there is some funding already, at the moment just something
>to fill the wholes partially that Stackless poked into my wallet.

>I have set up a PayPal account which I will add to the web pages.

>It also works via email:
>https://www.paypal.com/xclick/business=tismer%40tismer.com&item_name=Stackl
ess

If you have a pay pal set up then I propose you allow groups to donate money
in favor a particular name.  In the end the name with the highest cash
amount wins and you use all the proceeds towards its development.

Loren



From sholden at holdenweb.com  Thu Feb  7 08:48:45 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 08:48:45 -0500
Subject: OT again... is there an IMAP expert in the house?
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org> <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org> <mailman.1012224450.11377.python-list@python.org> <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu> <mailman.1012892049.28494.python-list@python.org> <Pine.OSF.4.44.0202042332410.184841-100000@goedel2.math.washington.edu> <mailman.1012915928.22449.python-list@python.org>
Message-ID: <SAv88.94356$KM2.3410687@atlpnn01.usenetserver.com>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in ...
[ ... ]
>
> It would be easier for me if I had time to grok IMAP like I do SMTP and
> POP3.
> This is my first painful exposure.
>
Grokking IMAP would take a very long time indeed. Life's too short if there
are other ways to achieve your ends ;-)

I just had a reviewer complain that my book mentions that IMAP is superior
to POP3 but then doesn't cover IMAP. While the reviewer complained that I
lamented the lack of space to cover it, it really *would* have taken too
long to thoroughly explain IMAP, which uses a complex on-line mail reading
model understood by relatively few.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From tim.one at comcast.net  Fri Feb  8 23:31:55 2002
From: tim.one at comcast.net (Tim Peters)
Date: Fri, 08 Feb 2002 23:31:55 -0500
Subject: threads killing other threads
In-Reply-To: <a41fsq$bvf$1@newsreader.mailgate.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEMENKAA.tim.one@comcast.net>

[Les Smithson]
> But doesn't pthread_testcancel allow cooperative threads to be
> canceled?

Yes.

> This is the portable equivalent to the thread signaling scheme
> described earlier.

It's only portable among platforms with pthreads implementations recent
enough to include it.  The thread facilities Python exposes are much more
portable than that.

> It also gives the canceled thread a chance to clean itself up before
> exiting.

Yes, provided you expose more and more of pthread-specific functionality to
get at all this stuff.

> Not ideal, but better than nothing.

I'm afraid the truly portable part of it remains nothing.  Someone may care
to write a pthreads-specific extension to Python's threading facilities,
though.  Note that what people actually ask for is a way to *force* another
thread to terminate (even if it's, e.g., stuck in a blocking I/O call), so
the audience for this kind of gimmick gets slashed repeatedly each time
reality objects <wink>.

don't-use-threads-when-you-want-processes-ly y'rs  - tim




From danb_83 at yahoo.com  Mon Feb 18 20:07:03 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 18 Feb 2002 17:07:03 -0800
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <ad052e5c.0202181707.69c22978@posting.google.com>

"jimh" <jim at jim.jim> wrote in message news:<a4rjq4$4n8$1 at web1.cup.hp.com>...
> ...
> I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
> a one-line 'print "hi"' script, and this is, of course, not acceptable for a
> web server.

Because you expect that running a ten-line script will take a minute?


From jkraska at san.rr.com  Wed Feb 27 21:09:28 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 28 Feb 2002 02:09:28 GMT
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com> <a5jbra$7ok00$1@ID-11957.news.dfncis.de> <m3ofiair04.fsf@mira.informatik.hu-berlin.de>
Message-ID: <4q4r7uklijltcd3i5v455hh44gi3bmtqfc@4ax.com>

On
>As Zenin points out, Python uses objects to emulate closures, whereas
>Lisp uses closures to emulate objects.

Ah. I like that turn of phrase. Very astute.

In one brief sentence, it explains how and why I already think
about Python but could never put to words.

C//



From wookiz at hotmail.com  Fri Feb 15 06:09:09 2002
From: wookiz at hotmail.com (wooks)
Date: 15 Feb 2002 03:09:09 -0800
Subject: Windows Message Hooks
References: <8852d33c.0202140224.13840b45@posting.google.com> <3C6C4920.4000301@skippinet.com.au>
Message-ID: <8852d33c.0202150309.187b311f@posting.google.com>

Mark Hammond <mhammond at skippinet.com.au> wrote in message news:<3C6C4920.4000301 at skippinet.com.au>...
> wooks wrote:
> > Can anyone direct me to a resource where I can find info Python
> > support for the above.
> > 
> > Thanks.
> 
> I don't think you will find direct support.  Adding such support to the 
> win32gui module should not be difficult.
> 
> Mark.

As yet I have am still very much a newbie. I bought your book but
haven't got far enough to learn about the win32gui
. I came across this book.

http://www.oreilly.com/catalog/subhookvb/desc.html

I am wondering whether it would be possible to apply any techniques I
learn from it to Python via the win32gui module that you mentioned. I
presume the book tells you what API calls to make and you just make
them from Python rather than VB. Yes?


From gustav at morpheus.demon.co.uk  Fri Feb 22 16:50:02 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Fri, 22 Feb 2002 21:50:02 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <3C6D8B63.5070909@geocities.com> <mailman.1013864067.16756.python-list@python.org> <69587ukeu1m51mf46kpumjnt9qfujho7ju@4ax.com> <a52h06$43hjh$1@ID-59885.news.dfncis.de>
Message-ID: <e7fd7u0co9g72l5p3qi5ohgmtqlq48cglf@4ax.com>

On Thu, 21 Feb 2002 11:08:39 +0100, "Thomas Heller" <theller at python.net>
wrote:

>Another interesting technology are Jean-Claude Wippler's
>'Scripted Documents'. The main link seems to be
>
>http://www.equi4.com/scripdoc/
>
>but you should also google for others.

That's very interesting. Thanks for the pointer.

Paul.


From aahz at panix.com  Fri Feb  1 12:56:25 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 09:56:25 -0800
Subject: typetesting, adaptation, typeclasses, ... (was Re: isinstance() considered harmful)
References: <mailman.1011997845.31162.python-list@python.org>
Message-ID: <a3eks9$tb$1@panix2.panix.com>

In article <mailman.1011997845.31162.python-list at python.org>,
Jason Orendorff <jason at jorendorff.com> wrote:
>Alex Martelli wrote:
>> 
>> Refusing to guess in presence of ambiguity is one of Python's
>> tenets, though.
>
>It's the language that doesn't guess.  The language designer is
>notorious for picking one thing, seemingly arbitrarily, from a number
>of reasonable alternatives.

Mind expanding that?  (I could guess, but I don't want to flame you
until I know for sure what you're saying. ;-)
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From wayne at mishre.com  Mon Feb 11 18:34:08 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Mon, 11 Feb 2002 15:34:08 -0800
Subject: Help: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254>
Message-ID: <3C6854F0.9040003@mishre.com>

I haven't used Xitami in a while, but IIRC I had some problems with it 
when I had the first line on a Windows based system.  Try remving the 
first line and testing it.  I don't think it should matter, technically, 
since it's a comment...but it used to.  Mind you it's been about 2 years 
since I've worked with Xitami...

Do you have .py and .pyc files associated within Xitami to launch the 
Python interpreter?  Just having .py files won't necessarily work, since 
you will be using .pyc files after the first run.

Flavian Hardcastle wrote:
> The example at http://www.devshed.com/Server_Side/Python/CGI/page3.html is 
> supposed to produce the text in my browser "Hello, Python", but it doesn't 
> do a thing.
> 
> Instead, using the Xitami personal server, I get the error 
> 
> 
>>Internal error: your request was unsuccessful
>>Cannot create CGI process - program not found
>>
> 
> Here is the text of the example...
> 
> #!/usr/bin/python
> 
> # Tell the browser how to render the text
> print "Content-Type: text/plain\n\n"
> 
> print "Hello, Python!"  # print a test string
> 
> 
> I had a theory that the first line was the problem. The tut says that it's 
> used by Unix and Linux servers to tell the OS to initiate Python. I'm using 
> Win98, btw. I tried changing it to the following......
> 
> 
>>#path = c:/Python22/Python.exe
>>
> 
> and then directing to the file location on my HD ...
> 
> 
>>#path = d:/Docs/Web/cgi-bin/index.py
>>
> 
> ... but in both cases the browser merely gave me the option of downloading 
> the file, or opening it using Python.
> 
> So what am I missing? 
> 


-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From rhymes at myself.com  Sat Feb 16 08:42:42 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sat, 16 Feb 2002 14:42:42 +0100
Subject: Highlights & Cursors
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com> <3daa1aa8.0202141551.1dae3651@posting.google.com> <2apq6ucuh9n7s446rrcsdi2fu7eis0qaqu@4ax.com> <13285ea2.0202151501.47080e91@posting.google.com>
Message-ID: <bcos6ugt62vln6puk2tu8fgcnm29qr67e8@4ax.com>

On 15 Feb 2002 15:01:12 -0800, joost_jacob at hotmail.com (J.Jacob)
wrote:

>from Tkinter import *
>widget = Button(text='new cursor', padx=10, pady=4)
>widget.pack()
>widget.config(cursor='cross')    # others: 'gumby' 'hand2' 'pencil'
>'watch'
>mainloop()

Thanks

----
Lawrence "Rhymes" Oluyede
rhymes at myself.com


From max at alcyone.com  Sun Feb 10 18:01:52 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 10 Feb 2002 15:01:52 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com> <7xlme1hv0y.fsf@ruckus.brouhaha.com> <3C66EA99.B22478E3@alcyone.com> <roy-F1119D.17441210022002@news1.panix.com>
Message-ID: <3C66FBE0.54733C3D@alcyone.com>

Roy Smith wrote:

> I think it makes plenty of sense.

Sure, but your repeated opinion isn't going to change anything.  Since
that's how things work, the onus on you is to convince people that it
should be changed.  That you think it would be nice isn't very
convincing.

> If the delimter was written as a constant '', I might be inclined to
> agree,
> but it could be computed on the fly, and having a delimter of '' mean
> "split the thing up with nothing between the parts" is a very logical
> extension of the general idea.

That doesn't change the fact that it is likely a mistake and there's no
reason to do it, given that there are easier, clearer, and more direct
ways to get the effect that you want.

> Should we outlaw raising things to the zero power because somebody
> thinks
> it "doesn't make much sense at all"?  I'll be a lot of people think it
> doesn't make much sense, but the mathematicians would howl if you
> tried to
> take it away.

You can keep repeating this analogy but it doesn't apply.  x^0 is well
defined (with a questionable area if x = 0, but this isn't sci.math). 
Splitting a string by delimiter with no delimiter doesn't make much
sense, because in that case it's not clear what you mean, and you're
better off doing it otherwise.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From mwh at python.net  Mon Feb 18 07:12:34 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 18 Feb 2002 12:12:34 GMT
Subject: Searching program for sorting cvs log output of several files
References: <3C6D78F0.F5A9A4E1@ndh.net>
Message-ID: <uit8vc7f7.fsf@python.net>

Stefan Schwarzer <s.schwarzer at ndh.net> writes:

> Hello Pythonistas :)
> 
> I'm looking for a program who will generate the output of "cvs log"
> for several files but make a single merged log of it with _all_
> accumulated entries sorted by date and time.
> 
> Does anyone know such a program/module, preferably in Python? Thanks
> in advance for all help.

Tools/scripts/logmerge.py?

Cheers,
M.

-- 
  SCSI is not magic. There are fundamental technical reasons why it
  is necessary to sacrifice a young goat to your SCSI chain now and
  then.                                                  -- John Woods


From mpoulin at verinet.com  Fri Feb  1 10:35:15 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Fri, 01 Feb 2002 10:35:15 -0500
Subject: Help with time
References: <a3ecqt$16bk9d$1@ID-98166.news.dfncis.de>
Message-ID: <3c5ad0bf$0$2200$7586b60c@news.frii.com>

In article <a3ecqt$16bk9d$1 at ID-98166.news.dfncis.de>, "MDK" <mdk at mdk.com>
wrote:

> I am passing to a function how long in milliseconds I want to wait for

> import time
> 
> def getdata(ms):
>     starttime = time.time()
>     while len(data) < 20:
>         now = time.time()
>         print '%.3f seconds %.3f' % (now - starttime, timeout / 1000) if
>         now - starttime > timeout/1000:
>             break
>     return data
> 
> What am I doing wrong?

Maybe a truncation problem (mixing ints and floats)?

My Python interpreter says
>>> print 100/1000
0
>>> print 100/1000.0
0.1
>>>


From hamish_lawson at yahoo.co.uk  Thu Feb 21 14:02:59 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 21 Feb 2002 11:02:59 -0800
Subject: Using JPE and Jython together?
Message-ID: <915a998f.0202211102.6a8396b9@posting.google.com>

Jython provides Python implemented for the Java Virtual Machine, and
as such can't directly make use of Python extensions written in C. JPE
(Java-Python Extension) on the other hand provides a bridging
mechanism between Java and the native Python interpreter (including
Python C extensions). Though I haven't used JPE at all and Jython only
a little, I'm wondering whether it might be possible to use JPE from
within Jython to give access to Python C extensions (I'm presuming
that whatever mechanism JPE provides for Java to call across to the
native Python interpreter could be used by any language that compiles
to Java bytecode - including Jython). If so, then the next step could
be to hide the explicit invocation of the JPE mechanism from within
Jython and instead hook into the import mechanism to allow direct
importing of Python C extensions. Might such integration be a feasible
development route for Jython?

Hamish Lawson


From emile at fenx.com  Thu Feb 28 10:35:56 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 07:35:56 -0800
Subject: sorting
References: <3c7e3d12_7@news.teranews.com>
Message-ID: <a5lj2v$8afvv$1@ID-11957.news.dfncis.de>

"Jeff McComas" <jmackd1 at yahoo.com> wrote in message
news:3c7e3d12_7 at news.teranews.com...
> As a newbie i find the python scripting language very interesting and
useful.
> I am trying to write a script that will search a .txt file and have
the script
> tell me how many of the same numbers are in the .txt file. I know you
can do
> it with the f.count command but my problem is how can you do an input
that
> makes the numbers strings? Or puts double quotes and a comma around a
number?


When you read the .txt file you get strings.  Use split to make a list,
use the list elements as keys in a dict where the accumulated count is
the value.  You may want to start with:

vals = """ 34.6        25        78        89.2       45.7
 23          92        47        20         85
 56          78        23        49         92 """

and add the file reading part later.


Post your efforts next time so we can point out specific problems and
answers.

--

Emile van Sebille
emile at fenx.com

---------



From greg at cosc.canterbury.ac.nz  Thu Feb 14 00:02:20 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 14 Feb 2002 18:02:20 +1300
Subject: Roses are blue, violets are red (Re: xor on strings)
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com> <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com> <vqI98.3303$pC1.669077@newssrv26.news.prodigy.com> <3c6763c6_3@corp.newsgroups.com> <9UJ98.3347$pC1.674754@newssrv26.news.prodigy.com>
Message-ID: <3C6B44DC.850A48E9@cosc.canterbury.ac.nz>

dsavitsk wrote:
> 
> VB does BBGGRR for no reason that i understand.

It's because the Microsoft programmers that wrote VB
are actually immigrants from the planet Roloc, who
have the red, green and blue sensors in their retinas
wired to their brains in a different order.

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


From deathtospam43423 at altavista.com  Thu Feb 21 09:50:59 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 21 Feb 2002 14:50:59 GMT
Subject: newbie: Some file object tips?
References: <mailman.1014133831.1330.python-list@python.org>
Message-ID: <Xns91BDDAD21D41AusYourStandingInIt@130.133.1.4>

Thanx! This little code has been immensely useful to me. I've learned a lot 
from it.

I'm currently thinking of using parts of it. Instead of actually scrolling 
back from the end of the log, and counting markers indicating the spaces 
between posts ... I'm thinking of using a dictionary array to keep a 
database of the precise byte position of each chat post. I would determine 
this using seek(0,2) and tell() operands as you describe below. Does that 
sound like a better way of doing it?


Gustavo Cordova <gcordova at hebmex.com> wrote in 
news:mailman.1014133831.1330.python-list at python.org:
>> 
>> So does this command ....
>> 
>> > lines = open("log.txt").readlines()
>> 
>> ... actually load the whole log into RAM.
>> 
> 
> Yes it does.
> 
> But then, what's actually flawed is the algorithm,
> not the program, or the tools.
> 
> You need a new way to store the message which will
> let you read only part of the log into memory.
> Right?
> 
> Well, sorry. Maybe "flawed" is a too harsh qualification,
> but overly simple. Ok.
> 
> Hmmm... What would I do?
> 
> I wouldn't read the "last 'n' lines", as you put it,
> but something easier to calculate, "the last 'n' K of text",
> which is more "file-like".
> 
> So, to read the lines in the last "n" Kb of text:
> 
> ====== SOME CODE ======
> def GetLog(filename, maxSize=None):
>      log = open(filename)
>      if maxSize:
>           # Seek to end of file.
>           log.seek(0,2)
> 
>           # Obtain current position index.
>           fileLen = log.tell()
> 
>           # If file is larger than maxSize,
>           # seek to (EOF - maxSize).
>           if maxSize < fileLen:
>                log.seek(fileLen-maxSize)
> 
>                # Skip one line.
>                log.readline()
> 
>           # Or return to the file beginning.
>           else:
>                log.seek(0)
> 
>      # Return all lines.
>      return log.readlines()
> ====== SOME CODE ======
> 
> So now you only get the last maxSize Kb lines-worth of
> lines from the log file.
> 
> Maybe it's a little clumsy, I bet someone can clean it
> up a bit.
> 
> Good luck :-)
> 
> -gustavo
> 
> 



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From fperez528 at yahoo.com  Tue Feb 19 06:23:22 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 19 Feb 2002 11:23:22 +0000
Subject: Adding static typing to Python
References: <mailman.1014142051.2177.python-list@python.org>
Message-ID: <a4u5ad$h0e$1@peabody.colorado.edu>

Michal Wallace wrote:

> 
> I don't have the code online anywhere, but I can post it if
> anyone's interested.
> 

That would be great, it sounds nice and useful.

Cheers,

f.


From newgene at bigfoot.com  Sat Feb  9 19:35:19 2002
From: newgene at bigfoot.com (newgene)
Date: 9 Feb 2002 16:35:19 -0800
Subject: A newbie question about class
Message-ID: <a50631f6.0202091635.1dc42d4b@posting.google.com>

Hi, group,
     I have a newbie question about building a class:

eg.

class A:
   def __init__(self,x=None):
       self.x=x
       if self.x != None:
            self.y=x**2
       else:
            self.y=x**2

If I initialize A like "a=A(2)", then I can get "a.y=4" automatically;
but if I initialize A like "a=A()", then when I give "a.x=2" later, I
can not get the value of "a.y" like the former. Of course I can use a
method to set the value of "a.y", but is there a machanism to get the
"a.y" automatically whenever I change the value of "a.x"?


Thank you in advance.

Newgene


From eric.brunel at pragmadev.com  Tue Feb  5 10:05:49 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 5 Feb 2002 15:05:49 +0000
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com> <60018f02.0202050547.4836c5ff@posting.google.com>
Message-ID: <a3oolc$8g9$1@norfair.nerim.net>

Hi Marc,

Could it be the return of the son of the utterly useless CR character at 
the end of lines? If your file is in DOS format, the shell tries to execute 
"/usr/bin/python^M", and obviously doesn't find it... The result is what 
you saw: either "Command not found" or "No such file or directory", or 
something similar...

HTH
 - eric -


From bougle at gluce.com  Tue Feb 19 14:20:23 2002
From: bougle at gluce.com (bougle gluce)
Date: Tue, 19 Feb 2002 11:20:23 -0800
Subject: newbie: Some file object tips?
References: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4> <mailman.1014120512.7783.python-list@python.org> <Xns91BAF4112B7BEAusYourStandingInIt@130.133.1.4>
Message-ID: <pxxc8.43$K_1.81711@news.uswest.net>

"Flavian Hardcastle" <deathtospam43423 at altavista.com> wrote in message
news:Xns91BAF4112B7BEAusYourStandingInIt at 130.133.1.4...
> Gerhard =?unknown-8bit?Q?H=E4ring?= <gh_pythonlist at gmx.de> wrote in
> news:mailman.1014120512.7783.python-list at python.org:
>
> > Le 19/02/02 ? 11:00, Flavian Hardcastle ?crivit:
> >>
> >> I'm writing a crude cgi chat script. I've already got it up and
> >> working with Xitami server, and have chatted with a few net buddies.
> >>
> >> How it works is very rough and simple... it basically just takes data
> >> from people's browsers (using the FieldStorage function) and stores it
> >> in a file called log.txt, and then displays the contents of that file
> >> back to them.
> >>
> >> My problem is ...
> >>
> >> The longer you chat, the bigger the log.txt file gets, and
> >> consequently the web page the chatters are viewing gets bigger and
> >> bigger, and becomes hard to download.
> >>
> >> So I want to include a "scroll back" feature ... which will enable the
> >> user to control how many previous posts (s)he views. It will probably
> >> take the form a simple input box on the form ... if you want to see
> >> the last 8 posts, you put 8 in the box, if you want to see the last 50
> >> posts, you put 50 in the box and so on.
> >>
> >> So, I need the script to be able to search the log.txt file. Any tips
> >> as to how I might go about this?
> >
> > You could read the file into a list with
> >
> > lines = open("log.txt").readlines()
> >
> > but now, all the lines will include a newline character at their end.
> > To strip these, you can do the following instead:
> >
> > lines = map(string.rstrip, open("log.txt").readlines())
> >
> > Now you can use slicing on the list, say to return the last 8 lines,
> > you can do:
> >
> > lines[-8:]
> >
> > HTH,
> >
> > Gerhard
>
> Cool! Thanx very much. Never saw that in any of the tutorials!
>
> BTW, one thing I'm concerned about ... I want to avoid the script having
to
> load the entire log.txt file into RAM everytime someone makes a post. You
> see, I estimate that a log file with only 200 posts worth of text would
> weigh about 60k.
>
> If say, ten people, are all posting posting at once that could be as much
> as 600k in the memory. If it's 20 chatters, 1.2 meg. That could take a
> couple of seconds for the servers CPU to process, might noticeably slow
> things down ... especially on my 'puter which is only 500Mhz.
>
> So does this command ....
>
> > lines = open("log.txt").readlines()
>
> ... actually load the whole log into RAM.
>
> --
> cdewin at dingoblue.net.au

yes, readlines() will load the whole fileinto RAM. as an alternative you can
use readline(), which reads one line at a time, or read(n) where n is the
number of bytes you want to read. you can also use seek() to go directly to
a particular place in the file.




From brian_l at yahoo.com  Sun Feb 10 03:36:50 2002
From: brian_l at yahoo.com (Brian Lenihan)
Date: 10 Feb 2002 00:36:50 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org>
Message-ID: <afe3b146.0202100036.f221c9a@posting.google.com>

"Dr. David Mertz" <mertz at gnosis.cx> wrote in message news:<mailman.1012934533.15278.python-list at python.org>...
> |"Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
> |> I suppose it might be a brain thing. I use my left hand for writing and
> |> drawing while I use my right hand for more coarse-grained tasks, like
> |> throwing a ball and such.
> 
> "Steve Holden" <sholden at holdenweb.com> wrote
> |Strange. I don't come across many other people who do this. I've always
> |thought my brain must have been differently wired for some reason I'll never
> |know.
> 
> Same story for me.  I don't think it is that unusual though.  Culture is
> sort of geared to the right-handed majority, so us folks who are mostly
> left-handed wind up learning the things they can do "well enough" with
> the right hand in the common style.  Once you've learned something,
> propioceptic memory is more important than underlying "handedness."

I eat, write, mouse, and play golf right handed.  I kick equally well
with both feet, at least while playing soccer.  My surfing, skate
boarding, throwing, tennis, darts, archery and shooting are done
left-handed. My left eye is dominant. I agree with the cultural bias
against left-handers. However, nuns with rulers were probably
responsible for me writing with my right hand.  My mother is probably
responsible for me eating with my right hand. I play golf right-handed
because there were no left-handed clubs available to me when a learned
as a child.

I don't throw like a girl right-handed, so maybe I am actually
ambidextrous.

I have a brother who does everything right-handed except eating. He's
always been the stubborn one.

There-are-other-things-that-one-does-with-one-hand-or-the-other-but-we-don't-mention-ly-yours

Oh, and if throwing was really so "coarse-grained" I'd have been a
Major League pitcher.


From BPettersen at NAREX.com  Mon Feb 25 14:31:38 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Mon, 25 Feb 2002 12:31:38 -0700
Subject: On inner classes
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158D36@admin56.narex.com>

> From: Gon?alo Rodrigues [mailto:op73418 at mail.telepac.pt] 
> 
> Suppose we have something like
> 
> class <name>:
>     class <name>:
>         <whatever>
>     <whatever>
> 
> that is, a class defined inside another. How can the inner 
> class acess the methods/attributes of the outer one?

I'm assuming you're meaning something like:

  class Outer:
     def outerMethod(self):
        class Inner:
           def innerMethod(this):
              self.outerMethod() # call class Outer's outerMethod

-- bjorn



From marklists at mceahern.com  Thu Feb  7 11:22:48 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 08:22:48 -0800
Subject: FixedPoint
In-Reply-To: <Dhx88.95486$KM2.3465594@atlpnn01.usenetserver.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEGKKBAA.marklists@mceahern.com>

Steve Holden wrote:
> Since the *intent* for None is that there should only ever be one
> object of type None, the better test would be
>
>     if something is None

Thanks, that's how I generally test for None--I didn't know the bit about
"there should only ever be one."

However, you wrote:

> which can easily be used without requiring any changes to FixedPoint.

Well, sure, I could easily change my local copy of PyPgSql's PgSQL.py.  But
it seems like in this case it makes more sense for me to change my local
copy of FixedPoint.py.  Hmm, I'm having a hard time articulating *why*
though.  I guess because we use PyPgSql more, it's actively maintained,
etc.--I'm better off not maintaining our own branching of that code.

Of course, I could ask the PyPgSql folks not to test for None like that.
But then they're going to say, "how's the bug actually happening?"  And I'm
going to have to tell them I'm sending a FixedPoint instance as a param to
cursor.execute().  And *that* only works because I modified FixedPoint.py's
__repr__ to return str(self).  As you can tell, at the moment, I'm just
flying by the seat of my pants trying to make something work.

> Good general rule: if Tim made it that way, it's supposed to be
> that way :-)

I tend to agree.  However, do you think he purposefully avoided making the
code resilient to frippery like this?

	if FixedPoint(1.0) == None:
		...

Should it be resilient to code like that?  In the case of PyPgSql, one could
argue I'm using it incorrectly by passing it a FixedPoint to squirt into the
db.  But then that merely means I haven't framed my question correctly.  ;-)

Cheers,

// mark




From jim.vickroy at noaa.gov  Mon Feb  4 11:52:29 2002
From: jim.vickroy at noaa.gov (jim.vickroy)
Date: Mon, 04 Feb 2002 09:52:29 -0700
Subject: Pythonservice stuck
References: <ot678KAD6pX8Ewmt@jessikat.demon.co.uk>
Message-ID: <3C5EBC4D.DFF874E7@noaa.gov>

I'm not an experienced MS Windows programmer and I have no experience with
Zope, in particular, but I have always been able to remove services by
using the registry editor (regedit).  If memory serves, on some Windows
platforms, the entries remain visible until a reboot is performed.

Robin Becker wrote:

> Win2k SP2
>
> I tried out the latest Zope 2.5 and after playing a bit I stopped the
> service and uninstalled the software.
>
> However, in my services I still see a service for Zope_2_5 (the name I
> gave it instead of WebSite). A reboot doesn't clear this.
>
> I see several entries relating to Legacy zope_2_5 in the registry, but
> cannot delete them even with Administrator luser privs.
>
> Any way to remove these pesky entries?
> --
> Robin Becker



From jeff at ccvcorp.com  Thu Feb 14 19:29:30 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 16:29:30 -0800
Subject: More license stuff
References: <u6oe6n32m9pjfb@corp.supernews.com>
Message-ID: <3C6C566A.A51AD1E1@ccvcorp.com>


Virginia O wrote:

> I snagged this question from a thread below...
>
> >>OK, let's forget about the legal stuff for the moment.
>
>  More importantly, good manners. Let's say I'm wrapping up a Python program
> with py2exe, and I'm going to distribute it. Which files would I be expected
> to include as a matter of courtesy to the Python developers (and the
> developers of any additional modules that I use)? Just the licence files, or
> is there anything else?
>
> Cheers,
> Simon Brunning
> >>>>>>>>>
>
> I also, being a newbie, don't understand what all this licensing jargon
> means in terms of distribution. If I write a program, and sell it, then
> what? I do understand I need to speak with a intellectual properties lawyer,
> but I'd like to get a handle on things before I get started.'

The Python license explicitly states that this is fine.  First off, any code you
write (whether in Python or in C or whatever) is yours to do with as you
please.  The Python license explicitly gives you permission to use code modified
from the Python standard library, and to redistribute Python itself to anyone
you want to.  You could presumably even charge money to redistribute Python...
though, of course, since anyone can get it for free from the python.org site,
it'd be a tough sell.  ;)  The only real restriction that the Python license
has, is that if something bad happens, you can't sue the PSF or the Python dev
team over it.  So if that Python application dies and some client loses
thousands of dollars, well...  it's not the PSF's fault.  Whether or not that
client can sue *you* depends on the license that *you* put on the code that you
write, and whether that client agrees to it....

(Note, of course, standard disclaimers that I am not a lawyer nor an IP expert;
this is merely an anecdotal layman's understanding of the license.)

Jeff Shannon
Technician/Programmer
Credit International





From bokr at oz.net  Thu Feb 28 19:37:53 2002
From: bokr at oz.net (Bengt Richter)
Date: 1 Mar 2002 00:37:53 GMT
Subject: PEP 276 (was Re: Status of PEP's?)
References: <mailman.1014854027.3045.python-list@python.org> <a5lcvh$4tt$1@panix2.panix.com>
Message-ID: <a5mih1$hdh$0@216.39.172.122>

[just posted]
On 28 Feb 2002 05:57:05 -0800, aahz at panix.com (Aahz Maruch) wrote:
>[posted & e-mailed]
>
>In article <mailman.1014854027.3045.python-list at python.org>,
> <James_Althoff at i2.com> wrote:
>>
>>Rather than restart the entire debate on PEP 276, please send me
>>suggestions for specific changes to the PEP.
>
>My primary complaint about the PEP is that the following idiom is
>awkward, ugly, and prone to misunderstanding:
>
>    if i in 5:
>        ...
>
>Please add this to your list of objections.

how about

    if i in [:5]:  # (colon required to specify slice vs list)

The implication is an anonymous slice of xrange(sys.maxint+1)
(if that were legal).

<aside>
BTW, how do you specify a for-count ending in sys.maxint?

 >>> for i in xrange(sys.maxint-4,sys.maxint+1): print i
 ...
 Traceback (most recent call last):
   File "<stdin>", line 1, in ?
 OverflowError: long int too large to convert to int
 >>> for i in xrange(sys.maxint-4,sys.maxint): print i
 ...
 2147483643
 2147483644
 2147483645
 2147483646
 >>> print sys.maxint
 2147483647
</aside>

This could also allow [2:5] or [1:5:2] etc.

For an anonynous slice of integers, it might be reasonable
to interpret negative numbers in the slice as legitimate
indices backwards from zero. E.g., i in [-2:5:2] would be
mean "i in [-2,0,2,4]", not effectively "i in []".

Regards,
Bengt Richter



From nobody at nobody.com  Tue Feb  5 11:45:09 2002
From: nobody at nobody.com (John)
Date: Tue, 5 Feb 2002 10:45:09 -0600
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com> <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu> <M6M78.23304$l93.4247257@newsb.telia.net> <7xu1swe3bv.fsf@ruckus.brouhaha.com>
Message-ID: <hYT78.14612$tg4.168818@vixen.cso.uiuc.edu>

Is there any available code that does the below?

> but the OP asked for a generator where you could just specify what
> skewness you want.  The random module doesn't give a direct way to do
> that.  You have to pick a distribution, then crunch out some integrals
> and solve some equations to get the distribution parameters for the
> skewness you want.
>
> Further description:
> http://www.everything2.net/index.pl?node=skewness




From kondor at wanadoo.fr  Tue Feb  5 04:00:50 2002
From: kondor at wanadoo.fr (kondor)
Date: Tue, 05 Feb 2002 10:00:50 +0100
Subject: python and xmlrpc
Message-ID: <3C5F9F42.9040508@wanadoo.fr>

HI,
I'm looking for a document explained how to write xmlrpc servers and clients.
Can you help me ?
I've seen the articles on o'reilly.com (ONLAMP.org) and pythonware.org
Thanks.



From vodine at direct.ca  Thu Feb 14 17:25:32 2002
From: vodine at direct.ca (Virginia O)
Date: Thu, 14 Feb 2002 14:25:32 -0800
Subject: More license stuff
Message-ID: <u6oe6n32m9pjfb@corp.supernews.com>

I snagged this question from a thread below...

>>OK, let's forget about the legal stuff for the moment.

 More importantly, good manners. Let's say I'm wrapping up a Python program
with py2exe, and I'm going to distribute it. Which files would I be expected
to include as a matter of courtesy to the Python developers (and the
developers of any additional modules that I use)? Just the licence files, or
is there anything else?

Cheers,
Simon Brunning
>>>>>>>>>

I also, being a newbie, don't understand what all this licensing jargon
means in terms of distribution. If I write a program, and sell it, then
what? I do understand I need to speak with a intellectual properties lawyer,
but I'd like to get a handle on things before I get started.

Thanks,
Virginia







From jeff at ccvcorp.com  Thu Feb 28 17:13:19 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 28 Feb 2002 14:13:19 -0800
Subject: Can python handle operating a modem?
References: <a5ltoj$m59$1@newsg2.svr.pol.co.uk>
Message-ID: <3C7EAB7F.6CFF4E9F@ccvcorp.com>


"G. Willoughby" wrote:

> Can python handle operating a modem? i.e. can i say, lift the hook on the
> line to see if there is a dial tone? and test this? are there any modules
> available for this?

Yes, in the sense that all of this is done by sending commands to a serial
port, and listening for a reply -- modem commands tend to be reasonably
simple and straightforward, from what little I've seen.  There's a number of
serial port modules available;  I don't know if there's a higher-level
library specifically for sending modem commands, but you might try searching
the Vaults (http://www.vex.net/parnassus/).

Jeff Shannon
Technician/Programmer
Credit International




From gerhard at bigfoot.de  Mon Feb 18 19:50:19 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 19 Feb 2002 00:50:19 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s51k$8pd1@news1.gtech.com>
Message-ID: <slrna738ab.akv.gerhard@lilith.my-fqdn.de>

Peter Milliken wrote in comp.lang.python:
> There is no such effort to my knowledge. There is a tool called PyChecker
> that attempts to fill the gap, I have no experience with it though.
> 
> The loose typing of Python is perceived as one of its big features :-).
> There are regular discussions on the usefulness of Python for "larger"
> projects. Many claim that they use it without any problems - I suspect that
> it is from a background of ignorance and lack of metrics i.e. how much time
> is lost in their projects to misnamed variables etc? The only way in Python
> to find these is through (exhaustive) testing - such rigorous testing
> *requires* you to develop an exhaustive test suite that is capable of
> running periodically i.e. nightly etc to catch such errors. Whether the
> people who support Python for larger projects have the discipline within
> their project team to maintain such an effort is unclear - there has been no
> mention of it at all. So I doubt that the proponents of Python in these
> instances have any idea of the true cost of this language "feature" - they
> appear to be wound up in how quickly they can produce a working prototype
> (or whatever) so I suspect their day to day work involves very small,
> manageable projects where the lost efficiency due to loose typing isn't
> really that noticable for them i.e. they "live with it" :-).
> 
> Personally, I have used Python for several small, personal projects and the
> odd "script" that searches text files - even here, the only perceived
> advantage is the library support, I often wonder whether I wouldn't be
> better off using my real language of choice (Ada) because of it's strong
> typing -

Ada 95 is also one of my favourite languages, but I've never used it
for a real project (well, I once tried it use it for putting my
gateway computer online/offline, until I realized that Annex E wasn't
supported on the win32 version of GNAT :-( ).

> it too has interfaces to the standard GUI tools etc.

Yet the number of libraries is much smaller than for Python. There are
some nice data structure libraries and a nice COM binding, but I found
it much too exhausting to try to get the Ada libraries to work. For
example, there is AdaSockets for raw network programming, but the
Python standard library has implemented most of the interesting
protocols already, so Python is more usable out of the box for most of
my networking needs.

> My next personal project will not be in Python, it is a "cute"
> language and allows some nice facilities i.e. creating a script at
> run-time and then running it can be very powerful in the correct
> circumstances - so I will probably use an Ada-Python API for these
> situations and get the best of both worlds :-).

Perhaps you will find http://pyada.sf.net/ useful. It's a project I
started once for no good reason :-) It's now more or less
unmaintained.

> When I compile a Ada program cleanly I know that 95% of the bugs are
> out of it - before I even start testing! :-) Of course, it takes me
> longer to write up front but overall development effort (in my
> subjective opinion - since I have no metrics :-)) is that it is much
> lower.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From jeff at ccvcorp.com  Mon Feb  4 17:18:39 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 04 Feb 2002 14:18:39 -0800
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com>
Message-ID: <3C5F08BF.95D8ED5B@ccvcorp.com>


Guy Gascoigne - Piggford wrote:

> I'm trying to find a way to write something like expect in a python module,
> and I'm having trouble getting the syntax the way that I want.
>
>         while 1:
>             try:
>                 if self.expect( ( "ast login", 0 ),
>                                 ( "ogin", self.send_user ),
>                                 ( "sword", self.send_password ),
>                                 ( "\$", "done" ) ) == "done":
>                     break
>             except IndexError:
>                 skipcount += 1
>                 if skipcount == 2:
>                     raise IndexError

I'm not specifically familiar with expect, and I don't know if this will meet
with your requirements/preferences, but I'd be really tempted to replace your
self.expect() call with a dictionary, preloaded with functions.

expectdict = { "ast login": None,
                "ogin": self.send_user,
                "sword": self.send_password
                # ... etc,etc.
                }
if self.expect(expectdict):
    # ....


> Where send_user and send_password are separately defined functions.  What I
> really want to be able to do is to define the action beside the pattern,
> calling short functions that are defined separately makes it harder to read
> and trickier to follow exactly what is happening.  I've also played around
> with just returning an index, but that also gets ugly as the list of
> possible patterns gets larger.

Personally, I find that clearly named functions make more sense than short
anonymous ones;  if the function is well named, then it's easy to guess what it
does, and it's easy to find if you need more detail.  Having the function
definition separate from the list/dict then makes it easier to read the
list/dict, so you can see *what* patterns are being triggered.  Using a
dictionary like I suggested above, as (in essence) a jump table, makes
everything easy to find and keep track of.  Of course, that may just be me, and
YMMV.

Jeff Shannon
Technician/Programmer
Credit International




From aahz at panix.com  Wed Feb  6 20:18:53 2002
From: aahz at panix.com (Aahz Maruch)
Date: 6 Feb 2002 17:18:53 -0800
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <jAe88.87233$KM2.2811767@atlpnn01.usenetserver.com>
Message-ID: <a3sklt$7vr$1@panix2.panix.com>

In article <jAe88.87233$KM2.2811767 at atlpnn01.usenetserver.com>,
Steve Holden <sholden at holdenweb.com> wrote:
>"Spencer Doidge" <spencer at spencerdoidge.com> wrote ...
>>
>> I really tried to RTM first this time, but I failed.
>> Can someone give me a hint where in the doc html's it explains how to pass
>> by reference?
>
>The same place it explains how to use arguments :-) -- by reference is the
>only way to do it.
>
>References to immutable objects, however, (strings, integers, etc.)
>cannot be used to change the value, since the function argument is a
>*local copy* of the references passed in. Binding that local copy to a
>new value will have no effect on the argument that was passed in to the
>function.

Um, that's not IMO/IME a particularly useful way to look at it.  I don't
remember who first proposed it, but I find talking about "names" and
"bindings" to be much clearer.  When talking about parameter passing, one
would say that the function parameters create new names that get bound to
the passed in objects.  Any time one uses assignment on the function
parameter names (or any other name in Python, for that matter), the name
gets rebound to a different object.

If the object is mutable and mutated, all names that are bound to that
object see the changes.  Mutation, by definition, *never* involves a
rebinding of the object being mutated (that is, mutation never takes
place in an assignment -- though of course the assignment expression
might involve mutation as a side-effect).

(Yes, you're talking about binding, but you don't go whole hog, which
confuses the issue.)

>--
>Consulting, training, speaking: http://www.holdenweb.com/
>Author, Python Web Programming: http://pydish.holdenweb.com/pwp/
>
>"This is Python.  We don't care much about theory, except where it
>intersects with useful practice."  Aahz Maruch on c.l.py

While I'm pleased to be quoted, I do prefer being asked for permission.
Most people have that preference, IME.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From printers at sendme.cz  Fri Feb  1 08:47:58 2002
From: printers at sendme.cz (A)
Date: Fri, 1 Feb 2002 14:47:58 +0100
Subject: Why it does NOT work on Linux ?
Message-ID: <3C5AAA9E.18034.11AB3F0@localhost>

I have the following part of program that finds ItemID numbers.
Here, for example, are two
146759 and 146700 .
This program works well under windows but on Linux it does not 
find any number. Can you please help?
Thanks.
Ladislav

####################
import re
Text="""<tr BGCOLOR="#FFFFFF">
                      <td valign="top" align="left"><a 
href="lead.asp?ItemID=146759">[CN] Oak, Foiled &amp; Antique 
Furniture</a></td>
                      <td valign="top" align="center">18/12/2001</td>
                    </tr><tr BGCOLOR="#FFFFFF">
                    <td valign="top" align="left"><a 
href="lead.asp?ItemID=146700">[CN] Oak, Foiled &amp; Antique 
Furniture</a></td>
                      <td valign="top" align="center">18/12/2001</td>
                    </tr>"""

IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
Results=re.findall(IDs,Text)
print Results

##############
t



From howard at eegsoftware.com  Thu Feb 21 12:18:01 2002
From: howard at eegsoftware.com (Howard Lightstone)
Date: Thu, 21 Feb 2002 17:18:01 GMT
Subject: BLT Setup and Installation
References: <mailman.1014305575.22017.python-list@python.org>
Message-ID: <Xns91BC5EB5C235Dhowardeegsoftwarecom@66.75.162.196>

"Alves, Carlos Alberto - Coelce" <calves at coelce.com.br> wrote in
news:mailman.1014305575.22017.python-list at python.org: 

> Hi all,
> How can I install and setup the BLT extentions to tcl/tk to use in my
> python programs.
> I have downloaded the Pmw and BLT packages (binaries forms) and unzip
> them into C:\Python21 and C:\Python21\tcl. But when try the following
> code, it crashs:
> 
> from Tkinter import *
> import Pmw
> root=Tk()
> g=Pmw.Blt.Graph(root)
> Here, the system raise and error and say that "::blt::graph" commands
> not found, when callig Tk/tcl.
> 
> I'm running Python2.1 under Windows98, wich it's installed at
> C:\Python21 

I just had to do this myself for Python2.0.1 but it's the same----

BLT asks for the path where TCL is installed on the machine.  

I changed that to match the Python tcl path (c:\pythonxx\tcl) and 
completed the installation.

Then I copied the BLT 2.4 directory from ----tcl\lib 
                                    to   ----tcl\tcl8.3

Copy the blt24.dll from --tcl\bin to \windows\system.

This worked even when I installed python in a different place.


From joost_jacob at hotmail.com  Fri Feb 15 18:01:12 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 15 Feb 2002 15:01:12 -0800
Subject: Highlights & Cursors
References: <0jun6ukqqtq58t1dh7ustpu06patd71d2s@4ax.com> <3daa1aa8.0202141551.1dae3651@posting.google.com> <2apq6ucuh9n7s446rrcsdi2fu7eis0qaqu@4ax.com>
Message-ID: <13285ea2.0202151501.47080e91@posting.google.com>

[Lawrence Oluyede]
> My question is: how can i change the cursor in a frame or some others
> widgets?
>

You mean this ?

from Tkinter import *
widget = Button(text='new cursor', padx=10, pady=4)
widget.pack()
widget.config(cursor='cross')    # others: 'gumby' 'hand2' 'pencil'
'watch'
mainloop()

Don't ask me the best place to look up Tkinter things like this on the
internet i am still looking for that myself :)


From marklists at mceahern.com  Tue Feb 12 10:11:51 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 12 Feb 2002 07:11:51 -0800
Subject: searching backwards in a string
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBOEGGKCAA.marklists@mceahern.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEGHKCAA.marklists@mceahern.com>

Sorry, the sample I posted doesn't correctly adjust for the difference
between the length of the word and the last index of the word.  Replace
this:

>   # Express location relative to original orientation of word.
>   start = len(s) - m.start()

with:

    # Express location relative to original orientation of word.
    lastIndex = len(s) - 1
    start = lastIndex - m.start()
    print "%s found backwards in %s at %d." % (pattern, s, start)

// mark




From peter at engcorp.com  Sun Feb 24 14:35:26 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 14:35:26 -0500
Subject: win extension supports RS232 binary
References: <a4kc8e$ei5$1@news.efn.org>
Message-ID: <3C79407E.14F92E69@engcorp.com>

Spencer Doidge wrote:
> 
> Starting with Isaac Barona's SerialPort extension built on the win32
> extensions, I have a receiver working now on binary data. The original
> SerialPort handles ASCII, and you can use strlen() to determine the number
> of bytes received. My update on Barona's module returns a tuple giving the
> number of bytes received and a list containing those bytes.

I don't understand why the original would be limited to binary...
Python strings easily handle embedded ASCII NUL characters.


From mwh at python.net  Thu Feb 28 12:22:27 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 28 Feb 2002 17:22:27 GMT
Subject: string title() method does not work on Danish characters
References: <mailman.1014912949.13665.python-list@python.org>
Message-ID: <uzo1twmb4.fsf@python.net>

"Jason Orendorff" <jason at jorendorff.com> writes:

> Max M wrote:
> > Thanks for the suggestion, but it does not solve my problem. I am 
> > pulling the data from an older MS SQL server that returns standard ascii.
> 
> If it contains characters >127, it's not standard ASCII.
> It's probably ISO-8859-1, aka Latin-1, aka Western Europe.
> 
> If you're getting this data back as 8-bit strings, then you need
> to convert them to unicode first:
> 
> ustr = s.decode('latin-1')

This is 2.2 only.  In 2.1 you should use

ustr = unicode(s, "latin-1")

> ustr = ustr.title()

Cheers,
M.

-- 
  If I had wanted your website to make noise I would have licked
  my finger and rubbed it across the monitor.
                           -- signature of "istartedi" on slashdot.org


From ark at research.att.com  Sat Feb  2 13:28:08 2002
From: ark at research.att.com (Andrew Koenig)
Date: Sat, 2 Feb 2002 18:28:08 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012618084.26466.python-list@python.org> <a3gu2l$om7$1@panix2.panix.com> <m66o5uodgg4qikc2itemvsikjmmspb2a0i@4ax.com>
Message-ID: <yu99wuxv7n7r.fsf@europa.research.att.com>

> This is more or less true, yup. Just consider the vast hordes of
> programmers who, upon encountering Python, say "yuck! whitespace
> sensitive language!" who, only after trying it for a while, see the
> light. I was in that boat several years ago. It kept me from
> learning Python for a while. But then. But thennnnnnn. I _tried_ it.

It seems to be universal among programmers that they get hung up about
tiny syntax details until they get used to those details.  For example,
C and C++ programmers still argue about where to put the curly braces
in their programs.  At least Python doesn't have that particular problem...

-- 
Andrew Koenig, ark at research.att.com, http://www.research.att.com/info/ark


From ragnvald at captech.no  Thu Feb 21 06:00:39 2002
From: ragnvald at captech.no (Ragnvald Larsen)
Date: Thu, 21 Feb 2002 12:00:39 +0100
Subject: Mailing system for niceguys
Message-ID: <a52k0l$81g$1@tyfon.itea.ntnu.no>

We have made an OpenSource system for hiking information (1). It is
currently used by the Norwegian Hikers Association (2) with a total of
190.000 members. The system is currently in ASP (IIS) and the database is
MSSQL. As we during this summer/fall we will convert the system to PHP and
possibly PostgreSQL we are now interested in finding platform neutral
solutions for the further development.

In the further development of our intranet we would like to be able to send
the users individual updates on changes in the system. The system now has
4-500 users (local board members, volunteers, employees etc...) Today this
functionality is available online for the users. We do have some thoughts on
monthly newsletters etc. Perhaps even to interested members.

A script (in Python perhaps?) should get the appropriate information from
the database and que it up for mailing, and subsequent mailing (timed
event - every monday, month, etc...). I am aware of libraries for databases
and emails in Python. The system should be possible to run on a Windows 2000
server with Python, PHP and ASP. Transfer to Linux/PostgreSQL should be
possible.

Somebody probably already made this or a similar Python based system. Links
or other information would be greatly appreciated.


Regards,

Ragnvald Larsen
Chief developer Sherpa OpenSource

1) http://sherpa.sourceforge.net
2) http://www.turistforeningen.no




From roy at panix.com  Sun Feb 17 07:30:31 2002
From: roy at panix.com (Roy Smith)
Date: Sun, 17 Feb 2002 07:30:31 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <wzadu9scs2.fsf@nono.cs.uu.nl> <861yflwo4s.fsf@apo.hmotakef.int> <5d283efe.0202162142.53e3b402@posting.google.com>
Message-ID: <roy-05958C.07303117022002@reader2.panix.com>

daniels at mindspring.com (Alan Daniels) wrote:
> I *do* know several very capable programmers who started with GW-Basic
> back in the day, and they now work in C++/Java/Perl/My Favorite Language
> just fine.

Actually, basic was my 2nd language.  I started programming some big HP 
thing that was sort of a cross between an oversized desk calculator and a 
small computer.  9830 maybe?  Programs were written as a series of 
keystrokes.  My brain seems to have survived.  RPN, of course.  We had the 
optional pen plotter, which was pretty cool.

> They just need to unlearn the habits of using globals and
> variants everywhere.

Tell me about it.  After many years of resisting learning Perl, I was 
finally forced into it at a new job about a year ago.  The syntax is 
horrible (if you can have syntactic sugar, can you also have syntactic 
vinegar?) but the real thing that blew my mind was the pervasive use of 
globals.  I took over maintenance of a 1000+ line perl program that did all 
of its inter-routine communication with globals.  I've been hacking at it 
for a year and still don't understand what everything does.  Talk about a 
first language that can destroy your brain.

If you're looking for interesting programming languages to explore, you 
should give postscript a try.  I'm not sure I would advocate its use for 
anything but printing, but it's a facsinating departure from what most 
people are used to.  In some ways, it's an like an RPN version of Python, 
without the class machinery.  Or maybe it's closer to Ruby :-)


From tim.one at comcast.net  Wed Feb 13 01:41:23 2002
From: tim.one at comcast.net (Tim Peters)
Date: Wed, 13 Feb 2002 01:41:23 -0500
Subject: Re-throwing an exception without loosing the stack trace
In-Reply-To: <3C69BB96.369FC648@stroeder.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEENONLAA.tim.one@comcast.net>

[Chris Liechti]
> try a simple, naked "raise" ;-)

[Michael Str?der]
> Is that thread-safe?

Yes; exception info is maintained on a per-thread basis.





From sheershion at mailexpire.com  Sun Feb 17 15:47:47 2002
From: sheershion at mailexpire.com (Robert Amesz)
Date: Sun, 17 Feb 2002 20:47:47 -0000
Subject: Redefining "\n"
References: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>
Message-ID: <Xns91B8D86F45AA0rcamesz@amesz.demon.nl>

Dale Strickland-Clark wrote:

> When preparing strings for the Windows clipboard, I need \x0d\x0a for
> a newline which is a bit cumbersome to insert all over the place.
> 
> It would be neat if I could make "\n" insert some other sequence.

Uh, how about

   data = data.replace("\n", "\r\n")

before sending it to the clipboard?


Probably the best thing is to put everything in a function (caution - 
untested code):

def SetClipboardText(data):
    data = str(data)  # A precaution

    if wxPlatform == '__WXMSW__':
        data = data.replace("\n", "\r\n")

    dataobject = wxTextDataObject()
    dataobject.SetText(data)

    wxTheClipboard.Open()
    wxTheClipboard.SetData(dataobject)
    wxTheClipboard.Close()


That would neatly encapsulate it, and it's more portable that way, too.


Robert Amesz


From eric.brunel at pragmadev.com  Tue Feb  5 09:34:04 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 5 Feb 2002 14:34:04 +0000
Subject: Inherit Class Dynamicly
References: <a3ok79$bvr$1@pegasus.csx.cam.ac.uk>
Message-ID: <a3ompr$3cm$1@norfair.nerim.net>

Hi Xiao-Qin Xia,

> from Tkinter import *
> class NewWidget:
>         ...
>         def __init__(self, widget_name = "Button", **dict):
>                 # What is needed here to make the new born object(self)
>                 # behavior like both the class specified by widget_name
>                 # and NewWidget?

I'd do something like:

import Tkinter  # Simpler here than "from Tkinter import *"
class NewWidget:
        def __init__(self, widget_name = "Button", **dict):
                tkClass = getattr(Tkinter, widget_name)
                self.__actualWidget = tkClass( ... whatever ...)
        def __getattr__(self, attrName):
                return getattr(self.__actualWidget, attrName)

The attribute __actualWidget will contain an instance of the Tkinter class 
with the name passed to the constructor, and the __getattr__ method will 
ensure that all unknown attributes or methods in the NewWidget class are 
delegated to the Tkinter object. A kind of specilization by delegation 
instead of the usual one done by inheritance. You may also need a 
__setattr__ method in your NewWidget class.

But be warned: specializing Tkinter classes (like above or the usual way) 
may not have the result you expect since many things are "hardcoded" within 
Tk. Many methods available in Python are never actually called internally 
by Tk. For example, if you have a sub-class of StringVar overloading the 
"set" method, your method will never be called on user input.

Anyway, HTH
 - eric -


From db3l at fitlinxx.com  Thu Feb 28 22:34:36 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 28 Feb 2002 22:34:36 -0500
Subject: listing threads?
References: <mailman.1014930527.20065.python-list@python.org> <a5mmoj$2ji$1@panix2.panix.com>
Message-ID: <uadttot9v.fsf@ctwd0143.fitlinxx.com>

aahz at panix.com (Aahz Maruch) writes:

> DON'T DO THAT.  I'm serious.  You really do NOT want to use
> threading.enumerate(); you'll only confuse yourself.  I've pointed you
> at my web page on threading; is there some reason you haven't read it,
> or is there something you don't understand?

I wouldn't go all that far.  Clearly you have to understand the nature
of the snapshot that threading.enumerate() returns with respect to the
asychronous nature of threads (which of course is what was confusing
the OP, since the example had a thread that barely got off the ground
before exiting).

But I don't see that enumerate() is in and of itself a bad method,
when used properly - certainly not something to warrant such a
sweeping statement.  We certainly use it in some of our applications
as a passive way to get a snapshot of thread activity and it works
fine.

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


From anon at anon.com  Thu Feb  7 22:08:33 2002
From: anon at anon.com (Andrew)
Date: Fri, 08 Feb 2002 03:08:33 GMT
Subject: ActivePython and wxPython
Message-ID: <RiH88.37389$Jq.1973082@news2.calgary.shaw.ca>

Hello,

Sorry if this question is inappropriate for this newsgroup, but I was
wondering if there are known problems using wxPython with ActivePython?  I
try to run the first example (from www.wxpython.org) in the wxPython
tutorial but get a crash when I attempt to create a wxFrame object.  I
stepped into the code and it's failing on an 'apply' call.

Thanks,
Andrew




From pete at shinners.org  Fri Feb  1 01:42:06 2002
From: pete at shinners.org (Pete Shinners)
Date: Fri, 01 Feb 2002 06:42:06 GMT
Subject: ANN: Pygame 1.4 Released
Message-ID: <mailman.1012673766.20188.clpa-moderators@python.org>

Pygame, game development for python

Pygame 1.4 released on January 31, 2001
Pete Shinners (pete at shinners.org)
http://www.pygame.org


Pygame is a set of python modules written to help create games in
Python. It allows for the creation of high quality games, yet is also 
simple enough for the beginning user to get started immediately.

Version 1.4 adds support for Mac OSX. There are also many small
bugfixes and enhancements. For example pygame windows have default
icons, more flexible function arguments, and finer control over
cdrom audio playback.

Note that building pygame for yourself on OSX is still problematic
as the latest versions of SDL and Python have bugs on this platform.
These problems should be taken care of in following releases. In
the meantime, look for a complete binary download of pygame and
it's fixed dependencies.

For those new to the pygame library, it includes:
   * fullscreen or windowed framebuffer
   * multiple audio channel mixing, with streaming music
   * control of joystick, cdrom, mouse, and keyboard devices
   * antialiased truetype font rendering
   * variety of media formats: png, jpg, ogg, mp3, mod, and more

The pygame package also comes with complete documentation,
tutorials, and many playable examples. There is both an active
mailing list and irc channel available to help people get started.

Pygame is distributed under the LGPL licese, which is the same
license used by many its dependencies.









From nick.collier at verizon.net  Tue Feb 26 09:39:17 2002
From: nick.collier at verizon.net (Nick Collier)
Date: Tue, 26 Feb 2002 14:39:17 GMT
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu>
Message-ID: <3c7b9d52.5730880@news.verizon.net>

On Tue, 26 Feb 2002 08:25:28 -0500, Fernando Pereira
<pereira at cis.upenn.edu> wrote:

>On 2/26/02 4:32 AM, in article m3pu2sfv0m.fsf at mira.informatik.hu-berlin.de,
>"Martin v. Loewis" <martin at v.loewis.de> wrote:
>
>> Nick Collier <nick at src.uchicago.edu> writes:
>> 
>>> Anyway, I thought that the Python grammar was LL(1) compliant and
>>> yet when I translate the grammar into the Coco/R LL(1) parser
>>> generator I get lots of errors like:
>> 
>> [...]
>>> Where X and Y are productions such as arglist, etc.
>> 
>> My guess is that your translation into Coco/R is incorrect. I cannot
>> spot any ambiguity in arglist. Considering
>> 
>> arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**'
>> test)
>> 
>> it may be that, given "argument "," argument ","', it finds itself
>> incapable of determining whether this is the second argument belongs
>> to the first subexpression or the second one.
>
>This production cannot be disambiguated top-down with *any* lookahead, for
>the reason you suggest. Thus, it doesn't seem Python's grammar is LL(1). In
>general, languages with infix operators and complex nested structures like
>lists are not LL(k), because a parser can only decide the role of a
>subexpression after it sees the operator or terminator that follows it. But
>such languages are easily parsed bottom-up (shift-reduce) with bounded
>lookahead. That rule (in context) can be disambiguated bottom-up with two
>tokens of lookahead: once an 'argument" is parsed and pushed onto the
>shift-reduce stack and a "," is found, if the next token is ")" then the top
>of the stack is the second occurrence of "argument", otherwise it's the
>first.

I now have access to the actual error message and seems as if the
problem is with argument. So the error messages look like:

  LL1 warning in argument: NAME is  start & successor of deletable
structure
  LL1 warning in argument: NUMBER is  start & successor of deletable
structure
  LL1 warning in argument: STRING is  start & successor of deletable
structure
  LL1 warning in argument: LPAR is  start & successor of deletable
structure
  LL1 warning in argument: LSQB is  start & successor of deletable
structure
  LL1 warning in argument: PLUS is  start & successor of deletable
structure
  LL1 warning in argument: MINUS is  start & successor of deletable
structure
  LL1 warning in argument: BACKQUOTE is  start & successor of
deletable structure

The same errors are repeated in arglist, picking up on the errors from
argument.  Coco\R does create a parser here, so I guess I'm whether it
will work, after all the python one certainly works!

thanks,

Nick




From mcherm at destiny.com  Mon Feb 25 17:36:02 2002
From: mcherm at destiny.com (mcherm)
Date: Mon, 25 Feb 2002 17:36:02 -0500
Subject: On inner classes
Message-ID: <3C7ABC52.5080901@destiny.com>

bjorn writes:
> You can't. Think about it. Your inner_class is at class scope in
> outer_class. What you're after is a value of an *instance* of
> outer_class, however you have no handle to such an instance from within
> inner_class, ergo you can't get to it.

Maybe I can make this a little clearer. Your use of the term "inner 
class" suggests that you might be coming from a Java background. In 
java, there are things called inner classes, and the way that they work 
is that each instance of the inner class is "magically" connected to an
instance of the outer class.

That's not what happens when you declare one class inside of another in 
Python -- in Python the "inner" class is just another class, and its 
instances are not "magically" bound to instances of the outer class. 
It's as if (in Python language) they were declared to be STATIC inner 
classes.

Most of the time, this is the behavior that you want -- inner classes in 
Java were invented to solve problems that are really easy to handle in 
Python because of the existance of bound methods. (I can explain what 
that meant if you like.) But as you might expect, if you actually DO 
want the java-like behavior, Python allows it, you just have to be 
explicit and explain what you mean.

Here's some code illustrating what I mean:

----------------<sample>--------------------
Python 2.2c1 (#27, Dec 14 2001, 13:15:16) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
 >>> class Outer:
	def makeAnInner(self):

		class Inner:
			def __init__(self, myOuter):
				self.myOuter = myOuter
			def innerMethod(self):
				print "I'm an inner, my outer is %s" % self.myOuter
				self.myOuter.outerMethod()
		anInner = Inner(self)
		anInner.innerMethod()
	def outerMethod(self):
		print "This is an outer method."

		
 >>> anOuter = Outer()
 >>> anOuter.makeAnInner()
I'm an inner, my outer is <__main__.Outer instance at 0x00954E90>
This is an outer method.
----------------</sample>--------------------

There are only two things you have to do differently here than in 
Java... you have to declare the __init__() method of Inner to take a 
myOuter parameter and store it, and you have to explicitly pass the 
outer when you create it in the line "anInner = Inner(self)". By 
requiring you to explicitly create this outer-inner relationship when 
you want it, Python adds the ability to NOT have the relationship when 
you DON'T need it.

-- Michael Chermside




From gerson.kurz at t-online.de  Fri Feb 15 05:28:03 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Fri, 15 Feb 2002 10:28:03 GMT
Subject: help with silly algorhythm
References: <5e8bd451.0202142155.143ca4d9@posting.google.com>
Message-ID: <3c6ce21d.190762859@news.isar.de>

On 14 Feb 2002 21:55:27 -0800, kp87 at lycos.com (kevin parks) wrote:

>What I want to do is generate a new list of items based on
>an input list and a some criteria (list of acceptable values, for
>example). 

An ideal job for lambda! Lets start with the data.

> * list of input values =[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11] 
> * a list of acceptable new 'partner values' = [1, 3, 7, 10]

input_list = range(12)
partner_values = [1,3,7,10]

> so we might get (in mode 1: all closest higher value)
> [ [0, 1], [1, 3], [2, 3], [3, 7], [4, 7], [5, 7], [6, 7], [7, 11]....

GetUpper = lambda i,p:filter(lambda x:x[1],[(y,(filter(lambda
x:x>y,p)+[None])[0]) for y in i])

In typical Python fashion the code is obvious.

> -or (as a single new list) -
> [1, 3, 3, 7, 7, 7, 7, 11.....]

GetUpper = lambda i,p:filter(None,[(filter(lambda x:x>y,p)+[None])[0]
for y in i])

> mode 2 (all closest value down) might be:
> [[0,10], [1, 10], [2, 1], [3, 1], [4, 3], [5,3], [6,3], [7,3], [8, 7]....

Now this is a bit tricky, what actually would be "closest down" is 

 [[2, 1], [3, 1], [4, 3], [5,3], [6,3], [7,3], [8, 7]....

because the first two entries - [0,10] and [1, 10] - are "wrapped"
around. (Clearly, 10 is not "closest down" to 10). So, the actual
"closest down" list can be achived by using

GetLower = lambda i,p:filter(lambda x:x[1],[(y,([None]+filter(lambda
x:x<y,p))[-1]) for y in i])

while the version with the wrap-around you wrote looks like this:

GetLower = lambda i,p:[(y,([p[-1]]+filter(lambda x:x<y,p))[-1]) for y
in i]

> mode 3
> [ [0, 1], [1, 10], [2, 3], [3, 1].....

A generic mix function would read

Mix = lambda f1,f2,i,p:(lambda z=zip(f1(i,p),f2(i,p)):[z[i][i%2] for i
in xrange(len(z))])()

Note the fine use of a otherwise useless lambda call for allocating
local variables. You can use Mix() like this:

print Mix(GetUpper,GetLower,input_list,partner_values)
print Mix(GetLower,GetUpper,input_list,partner_values)




From phr-n2002a at nightsong.com  Mon Feb  4 17:46:01 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 14:46:01 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org> <7xu1sx41cl.fsf@ruckus.brouhaha.com> <3C5F04BF.3090806@yumpee.org>
Message-ID: <7x4rkwdfx2.fsf@ruckus.brouhaha.com>

Manoj Plakal <terabaap at yumpee.org> writes:
>           So if you're distributing Python code, it looks
>           like you have some choices:
>               - Require use of Python <= 2.1, no matter
>                 what the latest and greatest version of
>                 Python provides (perhaps bundle the interpreter
>                 with your software if people don't want to keep
>                 an extra copy of Python to run your stuff)

That bogus in most circumstances (though not always).  Python is
supposed to let me write small, self-contained programs that do
amazing things.  If I have to ship a huge interpreter along with my
scripts, they're no longer small.  If I make people find and download
some old Python version, they're no longer self contained.  So the
scripts should run on whatever Python versions most users have
installed.

>               - Require Python <= 2.1 now, but eventually
>                 migrate your users to 2.2 or 3.0 by releasing
>                 newer versions of your software

Won't work--I don't control what the users do.  I might issue new
releases but users will keep running the old ones, just like Python 2.2
has been released but most users are using older versions.

>               - Require Python >= 2.2 now, and hopefully
>                 use the from future mechanism to avoid
>                 the above headaches.

No good either.  I'd have to tell 90% of current Python users to go
upgrade their Python installation and I'd consider that incredibly
obnoxious.  Think of how angry you've gotten if you've looked at a web
page and seen "you must upgrade your browser to MSIE 6.9 (or whatever)
and reboot your system in order to view this page.  Click here to
start the 35 megabyte download".  Why couldn't they have just stuck
with standard HTML so the page would work your regular browser?

>           Also, I'm curious, why does your code depend
>           so critically on integer division? 

The program I'm hacking right now actually needs exact rational
division, but that's another story.  The issue is there's no way to
use integer division that works in both 2.1 and 3.0, without using
something like divmod that would shock the unwary.


From tim.one at comcast.net  Thu Feb 28 19:20:15 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 28 Feb 2002 19:20:15 -0500
Subject: PEP 276 (was Re: Status of PEP's?)
In-Reply-To: <OF85647636.477BD4AC-ON88256B6E.007DD753@i2.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGAOAAA.tim.one@comcast.net>

[Carel Fellinger]
> you being such a heavy weight that I fear
> Guido might be lured into thinking this to be a good idea
> that I felt compelled to react in public.

[James_Althoff]
> Well, thanks (I think <wink>), but I can assure you that I hold no
> special brownie points and have zero ability to "lure" Guido (or anyone
> on the Python team) into anything.  So, no need to lose any sleep over
> that one.  (On the other hand, if anyone knows *how* to "lure" any of
> them, be sure to let me know <wink>).

Well, I'm particularly swayed by posters who have a "<wink>" key and know
how to use it.  Still, my take remains what it was at the start:  it isn't
even half as disgusting as most people will say it is <wink>.

some-people-don't-even-like-listcomps-ly y'rs  - tim




From loewis at informatik.hu-berlin.de  Thu Feb 28 10:17:06 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 16:17:06 +0100
Subject: Trouble importing / using a custom text codec
References: <9ac02e81.0202280152.5bb1bf25@posting.google.com>
Message-ID: <j4r8n5fxfx.fsf@informatik.hu-berlin.de>

jhorneman at pobox.com (Jurie Horneman) writes:

> - With cp932.py in site-packages as described above, if I decode a
> text using encoding "MBCS_codecs.cp932", I get the error message
> "LookupError: unknown encoding". This happens even if I import cp932
> by hand before decoding the text.

I recommend a different strategy. Aim for allowing "cp932" as an
encoding name. To achieve this, put the following (or something like
this) into MBCS_codecs/__init__.py

def search_mbcs(encoding):
    if encoding == "cp932":
        import cp932
        return cp932.getregentry()
    return None

codecs.register(search_mbcs)

If you want this to happen at startup time of Python, just add a
MBCS_codecs.pth file in site-packages, which reads

import MBCS_codecs

HTH,
Martin



From jason at jorendorff.com  Mon Feb 11 01:57:27 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 11 Feb 2002 00:57:27 -0600
Subject: subclassing tuple
In-Reply-To: <4abd9ce7.0202102243.d078f9c@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEKMDBAA.jason@jorendorff.com>

damien morton wrote:
> class triple(tuple):
>     def __init__(self, x, y, z):
>         tuple.__init__(self, (x,y,z))
> 
> >>> t = triple(1,2,3)
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: tuple() takes at most 1 argument (3 given)
> 
> For some reason, subclasses of tuple dont call the proper __init__ code.

The magic __new__() method is called to create the object.
[The __init__() method is then called, but you might not need it.]

  class triple(tuple):
      def __new__(cls, x, y, z):
          return tuple.__new__(cls, (x, y, z))

More: http://www.python.org/2.2/descrintro.html

## Jason Orendorff    http://www.jorendorff.com/





From jkraska at san.rr.com  Tue Feb 19 10:40:49 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 19 Feb 2002 15:40:49 GMT
Subject: Adding static typing to Python
References: <mailman.1014098489.15429.python-list@python.org> <a4ss56$akj6@news1.gtech.com> <b7uc8.8411$Zp2.790511@atlpnn01.usenetserver.com>
Message-ID: <fjs47u0n8bscvgtme1vc7j6jajlt4og2n3@4ax.com>

>I don't see anything about strong typing that encourages you to define
>artificial types in this way. 

Speaking of typedef and the C/C++ languages, I've long thought
that in a statically-typed language, the types should never be
aliased, and should always be distinct. For example, the following
shouldn't compile:

typedef float F1;
typedef float F2;

void f (F1 f1);

int main (int argc, char** argv)
{
	F2 f2 = 3.0;

	f ( f2 );
}

That really doesn't sit well with my sensibilities, and frankly,
I believe this is antithetical to what a _statically_ typed
language is all about and intended to accomplish. It's also
internally inconsistent (two structs or classes with the same
fields can't be used as aliased without a hard type cast!), and
that's a bugaboo.

C//



From sdm7g at Virginia.EDU  Tue Feb 19 13:40:48 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 13:40:48 -0500 (EST)
Subject: confusion regarding os.path.walk()
In-Reply-To: <336594389.20020219181830@darwinwars.com>
Message-ID: <Pine.OSX.4.43.0202191327370.1674-100000@d-128-61-180.bootp.virginia.edu>


On Tue, 19 Feb 2002, Andrew Brown wrote:

>
> Thanks for your help. What you describe is what I thought should
> happen. But I couldn't figure out how to join the paths. It always
> looked right: I'd do debugging runs that simply printed out the
> command it was supposed to execute, and I could get results like
>
> 'ln -s ./wombats/habitat/burrows.html ./wombats/habitat/Burrows.htm'
>
> (which I think should work) but when I ran this through os.system(),
> it failed every time. So did every other variant I could think of
> except changing to the directory where the file was and linking
> without any paths involved.
>

Ok -- the problem there is that if you use a relative path in the
symbolic link, then when it's expanded, it's relative to the LINK,
and not to you current working directory. So it's trying to resolve
into something like:

  .../wombats/habitat/wombats/habitat/burrows.html

That's not really a Python problem: I'ld say that the man page for
'ln' is wrong because they list the arguments as both being files
( source_file, target_file ) when in fact, one of them is a file
name ( which it will create ) and the other is a path name of a
file that may or may not exist. Saying the args are files makes
you think that it's going to go out there and find the file and
do something with it (like normalize it's pathname) when all it
does it stick the pathname in the link.

os.path.abspath( link )  and os.readlink( link ) can both give
you the pathname of the file that is linked to, if you want to
check this stuff from Python.

>>> os.symlink( './mail/wwww', './mail/abcXYZ' )
>>> os.readlink( './mail/abcXYZ' )
'./mail/wwww'
>>> os.path.realpath( './mail/abcXYZ' )
'/Users/sdm7g/mail/mail/wwww'



-- Steve





From gbreed at cix.compulink.co.uk  Fri Feb 22 11:14:03 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Fri, 22 Feb 2002 16:14:03 +0000 (UTC)
Subject: Adding static typing to Python
References: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <a55qob$ebu$1@thorium.cix.co.uk>

Quinn Dunkan wrote:

> >But you can do that in Python:
> >
> >>>> def area(shape):
> >...   if isinstance(shape, Circle):
> >...     return math.pi*shape.radius**2
> >...   if isinstance(shape, Rectangle):
> >...     return shape.width*shape.height
> >...   if isinstance(shape, Square):
> >...     return shape.width**2
> >...   raise TypeError, "unknown shape"
> >...
> 
> This kind of giant typecase is exactly the sort of thing that OO is 
> supposed
> replace.

Yes, but then are overloaded functions "OO" either?

> >                                                     As for strict, 
> >strong, static, explicit typing (which may be a straw man) it looks 
> *less* >suitable for this example.

Note I did say "explicit" there.

> >You could make them all subclasses of Shape.  In which case, you have 
> >functions defined to use a polymorphic object that depend on function 
> >overloading by the real type.  Not all statically typed languages will 
> >allow that.
> 
> C++ requires special keywords for it, but all statically typed OO 
> languages I
> know of do dynamic dispatch.  I certainly don't know every static 
> language out
> there, but just because there's a limited implementation somewhere 
> doesn't
> mean the concept is unworkable :)

In Java, you need the same kind of "giant typecase" as in the Python 
example you objected to.  I think C++ works the same way, unless you use 
templates.  If I understand "dynamic dispatch" correctly, that's dynamic 
typing by the back door, so not much of an argument for static typing.

> >             If they do, you have the problem that you might have 
> > added >triangles for some other reason, and forgotten to cater for 
> them in >area().  You aren't getting static type checking at all.
> 
> Sure you are.  area will reject objects that don't have the proper 
> methods
> defined for them.  How it works in haskell is that each class has a 
> minimum
> interface that you must implement.  If you don't implement all the 
> methods
> of the Shape class, Shape-accepting functions won't accept your objects.
> This is less flexible than having to implement *only* the methods the
> function actually uses, as in python, but c++ templates are sorta like 
> that.

If the types have to be resolved at runtime, you aren't getting static 
type checking.  What you describe is called "genericity" right?  I wasn't 
thinking of that, and I did specify "explicit" above to exclude it.  AIUI, 
Haskell does static, implicit typing and so long as everything's 
statically compiled that gets you the same convenience as dynamic typing.

The proposed language changes for "optional static typing" all specify 
types explicitly.  A more productive approach may well be "how do we alter 
the language so that type inference can work?"  That may be why slots got 
added.

> >Wouldn't it be better to make area a method of the shapes?  And for 
> static >typing, make it abstract in the Shape class?
> 
> In your above example, that's what it is.  You're just spelling 
> 'circle.area()'
> as 'area(circle)'.

No, circle.area() and area(circle) mean very different things in Python, 
so none of your comments on circle.area() really apply to area(circle).

> I actually like the second better, because lets you do things like
> 'map(area, shapes)' instead of 'map(lambda o: o.area(), shapes)', as 
> well
> as multi-dispatch, and eliminates the 'len(o)' vs 'o.len()' thing.  It 
> also
> has other benefits like letting you define a method anywhere you want 
> which
> obeys the usual scoping rules (so you can add a 'frobbify' method to the
> string class, but not have to worry about it making a global change for
> everyone in all modules).

That's a matter of taste, but you can get area() to work much the same way 
Python implements len(), as

def area(shape):
  return shape.__area__()

which still uses polymorphism rather than operator overloading.  (Time for 
a definition of "polymorphism"?)

> I'm not even sure why the smalltalk style 'object followed by method' 
> notation
> caught on, especially since 'function followed by args' has always been
> firmly entrenched.

It happens to match the usual English word order: "the class's method" 
rather than "the method of the class".  If the field had been dominated by 
French speakers, perhaps it would have been the other way round.



                     Graham

         <http://www.microtonal.co.uk/>


From cliechti at gmx.net  Fri Feb 15 13:04:48 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 15 Feb 2002 19:04:48 +0100
Subject: cygwin and python
References: <u6oe6n32m9pjfb@corp.supernews.com> <3c6d2f5e@news.capgemini.se>
Message-ID: <Xns91B6C2A2B7870cliechtigmxnet@62.2.16.82>

"Stephane Tallard" <stallard at capgemini.fr> wrote in
news:3c6d2f5e at news.capgemini.se: 
> I use python under cygwin bash-rxvt . Trying to use omniorbPy,  I'm
> facing a strange behaviour :
> When executing the same command line under rxvt , nothing happens: 
> the script never ends. Moreover if I add print "Hello" at the 
> beginning of the script, "Hello" is never prompted ...

the rxvt port has some problems with output and not flushing it ... 
try:
python -u echo.py
or use the bash in the normal windows console window (cygwin.bat)

-- 
Chris <cliechti at gmx.net>



From carifio.nospam at nospam.usys.com  Tue Feb 19 19:35:14 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Wed, 20 Feb 2002 00:35:14 GMT
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
Message-ID: <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com>

# simple is good too

def traverse(list):
    if list:
        print list[0]  # or do something here
        traverse(list[1:])
    else:
        print "done"

# even simplier

def traverse1(list):
    if list:
        # do something
        traverse(list[1:])




From lucio at movilogic.com  Tue Feb  5 11:12:28 2002
From: lucio at movilogic.com (Lucio Torre)
Date: Tue, 05 Feb 2002 13:12:28 -0300
Subject: ANNOUNCE: autochino release 1
References: <3C5FFF8E.8090004@movilogic.com>
Message-ID: <3C60046C.4090709@movilogic.com>

Lucio Torre wrote:

> autochino release 1 is now available
>
> wee it working at:
> http://vampira.dyndns.org/autochino

As usual, the URL is wrong, and it downt work... why? my apache is 
really confused, so i can fix it, or fix you.

so find it at: http://vampira.dyndns.org/autochino/ (the slash does the 
magic)

>
> download it from:
> http://vampira.dyndns.org/autochino_rel1.zip


This is ok.

>






From skywalkerpackage at hotmail.com  Wed Feb  6 04:46:46 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Wed, 06 Feb 2002 09:46:46 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de>  <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org>
Message-ID: <3c60fb80.199334@news.online.de>

Hello Laura,

thanks again for your explanations.

Now I have another problem:

I want to use my own widgets in the same scheme. For example, let's
say I have created the widget

LabelEntry

which just combines a label and an entry side by side in a row. The
widget inherits from Frame:

class LabelEntry(Frame):
    """Valid options: labelText, padx, pady"""
    def __init__(self, master=None, labelText='', padx=4, pady=4):
        Frame.__init__(self, master)
        self.label = Label(self, text=labelText)
        self.entry = Entry(self)
        self.label.pack(side=TOP, anchor=W)
        self.entry.pack(side=TOP, anchor=W, padx=padx, pady=pady)

This doesn't work together with an optionDict={'labelText':'hello',
'padx':'6'}

and

wt=LabelEntry(...)
apply(wt.config, (), optionDict)

I get the following error message:

Traceback (most recent call last):
  File
"D:\python\activestate\python21\Pythonwin\pywin\framework\scriptutils.py",
line 301, in RunScript
    exec codeObject in __main__.__dict__
  File "D:\python\activestate\python21\myscripts\pane2c.py", line 42,
in ?
    apply(wt.config, (), dictOptions)
  File "D:\PYTHON\ACTIVESTATE\PYTHON21\lib\lib-tk\Tkinter.py", line
1085, in configure
    self.tk.call((self._w, 'configure')
TclError: unknown option "-labelText"

I guess the problem is that Python forwards the optionDict to Frame
and can't use them there.

I have also tried to understand how it is done in e.g. the Button
widget which inherits from Widget which inherits from BaseWidget
(a.a.), but I don't understand what's going on in Tkinter.py. Also, I
have tried to write the constructor of LabelEntry sort of

def __init__(self, master=None, **kw):
      Frame.__init__(self, master)
     ...
      self.label = Label(self, text=kw['labelText'])
...

But then, how do I handle default values, and how do I ensure that the
objects are created with the right parameters? Maybe I should use a
local dictionary

self.defaultOptionsLabel={'labelText':''}
self.defaultOptionsEntry={...}

then test if a key is in the optionDict parameter and, if yes, then
overwrite the dictionarys above, accordingly and use them to configure
the basic widgets.

But this seems rather cumbersome to me.


Is there an elegant solution to this problem?


ByeBye

Johannes





From db3l at fitlinxx.com  Fri Feb 15 19:00:07 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 15 Feb 2002 19:00:07 -0500
Subject: win32 DeleteObject ??
References: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk>
Message-ID: <uit8yuwhk.fsf@ctwd0143.fitlinxx.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> In Mark Hammond's excellent win32 extensions I can see how to obtain a
> handle to a bitmap, but how does one remove from memory? I can't seem to
> find DeleteObject.

Well, the extensions wrap the MFC objects with Python objects.  The
only one that seems to explicitly include the DeleteObject method is
the region object.  I would assume that the others automatically clean
up after themselves after the wrapping Python object is destroyed, but
am not completely certain.

> Similarly I can't seem to find things like CreateCompatibleDC, BitBlt
> etc which are part of win32. It almost seems as though I'm missing
> win32gdi.pyd :)

Those are all methods on a pyCDC object which you get from calls like
CreateDC().

Note that this is all in the win32ui.pyd module, which has its own
documentation and is located in the Pythonwin subdirectory rather than
the win32 subdirectory in an installation.

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


From joshm at taconic.net  Mon Feb 11 01:36:39 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 11 Feb 2002 01:36:39 -0500
Subject: xor on strings
References: <mailman.1013406617.24389.python-list@python.org>
Message-ID: <3c676428_2@corp.newsgroups.com>

> >>> colors = (56, 156, 12)
> >>> print '&H%02x%02x%02x' % colors
> &H389c0c

This tripped me up, too -- I was about to post the same thing, when I
realized that they were looking for a BGR string, not an RGB one.  They
wanted &H0c9c38.

-- j

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From mwh at python.net  Thu Feb 21 10:04:04 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 15:04:04 GMT
Subject: Warning about "for line in file:"
References: <just-8C3043.11330718022002@news1.xs4all.nl> <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com> <20020220143307.A2018@glacier.arctrix.com> <mailman.1014256478.16106.python-list@python.org>
Message-ID: <un0y2lvql.fsf@python.net>

Aldo Cortesi <aldo at nullcube.com> writes:

> Thus spake Neil Schemenauer (nas at python.ca):
> 
> > Jason Orendorff wrote:
> > > I agree!  This could be improved by having
> > > file.__iter__() always return the same object.  But that
> > > doesn't solve the problem in full.
> > 
> > Perhaps the buffer should be attached to the file object
> > itself instead of to the iterator.  read() and readline()
> > could then be changed to use the buffer if it exists.

Yes.

> ... but this would leave a discrepancy between the actual
> file position and the notional "read" position. Among other
> things, file.tell() would then give the wrong information,
> and file.write() would write to an unexpected location. 

I think if you interleave reads and writes without an intervening
seek(), you lose anyway.

Cheers,
M.

-- 
  > Emacs is a fashion statement.
  No, Gnus is a fashion statement.  Emacs is clothing.  Everyone 
  else is running around naked.
        -- Karl Kleinpaste & Jonadab the Unsightly One, gnu.emacs.gnus


From gcordova at hebmex.com  Wed Feb 20 18:29:25 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 17:29:25 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC7@SRM201002>

> 
> Next question: why is muddying the statement/expression distinction a
> detriment?  As opposed to the alternate view, that the 
> statement/expression
> distinction is itself a detriment?
> -- 

Or, if I may rephrase the question:

Why is the statement/expression distinction a detriment,
as opposed to the alternate view, that the statement/expression
distinction is a virtue?

I, at least, prefer the legibility and explicitness
of the current situation, where there can be no
misunderstanding, versus the common "Oops, I typed '='
instead of '==' mistakes".

-gustavo



From jason at jorendorff.com  Thu Feb 14 16:21:05 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 15:21:05 -0600
Subject: help!: optimization of search problem (example code)
In-Reply-To: <3c6bcf70.16509609@news.easynews.net>
Message-ID: <HFEKILOLEFEFMKAECNDLCEJPDCAA.jason@jorendorff.com>

Marcus Stojek wrote:
> This takes 5.3 sec on my machine for 500 points. That is much too
> long. Any idea how I could significanty improve this code?

On my machine I was able to reduce the time between startclock()
and stopwatch() by 25% just by wrapping all that stuff in a big
function.

In Python, access to local variables in functions
is faster than access to global variables.

More to come.

## Jason Orendorff    http://www.jorendorff.com/



From jkrushen at data-fortress.com  Sun Feb 24 23:42:06 2002
From: jkrushen at data-fortress.com (Jordan Krushen)
Date: Mon, 25 Feb 2002 04:42:06 GMT
Subject: Concurrent XML-RPC?
Message-ID: <ygje8.46554$Ym3.6525594@news0.telusplanet.net>

I know this has been asked before (or something close), but having explored
as much as I can, I still haven't found an existing, lightweight way to have
true concurrency when serving XML-RPC.  While Medusa is ok for IO-bound
stuff, we'll be doing some heavy db-driven processing, which when run under
Medusa is serialized.  Obviously, the SocketServer code behaves this way, as
well.

Is there a patch for the SocketServer code anywhere to allow it to run under
inetd?  I'd hate to have to rewrite all the xmlrpclib handling of the
connections, and this would allow me to run it under DJB's tcpserver,
obviously giving us the concurrency we need.

I've also heard that Zope's ZServer might do the trick, is this simple
enough?  (Note the above mentioning of 'lightweight' :)

What we're wanting is a framework where one can simply take the appropriate
library and use an automated installer, which would talk to a centralized
authority telling it which IP/port to bind to, set up all the
stunnel/tcpserver instances, generate certificates, etc.  I imagine
somewhere around 50 separate, independent XML-RPC services would exist at a
time, talking between themselves and the outside world.  Of course, many of
these instances may be running on the same host, so we'd like them to be as
lightweight as possible.

Any thoughts?

J.




From mpoulin at verinet.com  Wed Feb  6 10:58:01 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Wed, 06 Feb 2002 10:58:01 -0500
Subject: Is anyone capable of explaining ?
References: <mailman.1012947969.13013.python-list@python.org>
Message-ID: <3c616e7e$0$185$7586b60c@news.frii.com>

In article <mailman.1012947969.13013.python-list at python.org>, "A"
<printers at sendme.cz> wrote:

> Can anybody please answer the following question?

I think you can find the answer here:
http://www.cookiecentral.com/faq/


From phr-n2002a at nightsong.com  Tue Feb 12 20:50:03 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 17:50:03 -0800
Subject: Major league corruption - anyone seen anything like this?
References: <mailman.1013557302.9749.python-list@python.org>
Message-ID: <7xzo2ekv5g.fsf@ruckus.brouhaha.com>

Skip Montanaro <skip at pobox.com> writes:
> I'm developing a small library and accompanying wrapper.  I sometimes get a
> segmentation fault (environment is Mandrake 8.1, Python is from CVS).  Upon
> my return to the shell prompt, I discover that the /usr/local/bin/python
> link is gone as are a number of shared libraries (readline among them).
> 
> Obviously I have some serious problems in what was supposed to be a simple
> wrapper (perhaps the problems are with the underlying library which has only
> been lightly tested), but I've never seen a segfault take files with it like
> this.  Has anybody else encountered this before?

That should be impossible assuming you don't have write access to the
/usr/local/bin directory.  If you can reproduce it, it's a big Linux
security hole.



From compedu at nh.km.ua  Thu Feb 28 07:35:22 2002
From: compedu at nh.km.ua (Dmitry Savchuk)
Date: Thu, 28 Feb 2002 14:35:22 +0200
Subject: Suggest, please
Message-ID: <a5l84r$1c32$1@infocom.km.ua>

Hi,
Can I make applicatinion using Python that will administrate PC's via LAN
under Linux. I mean to blockade workstations after paying time is off (like
in the Internet Cafe). I did such program under Windows in VB 6.0 and now
I'd like to try to develop under Linux. I am interesting how can I send a
signal via LAN when time is off. Early I have used TCP sockets. As I
understand in this case I should do in such way, but it would be interesting
to look at such code first if it possible.

Thank you,
Dima.






From tim at vegeta.ath.cx  Fri Feb  8 23:12:13 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Sat, 09 Feb 2002 04:12:13 GMT
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <slrna6938e.8ot.tim@vegeta.ath.cx>

Aahz Maruch <aahz at panix.com> graced us by uttering:
> I know some of y'all still keep roots to the Perl community, so I was
> wondering if there's any kind of definitive nickname for Perl
> programmers, the way we call ourselves Pythonistas?

The JAPH tradition would tend to suggest "Perl hacker", though "Perl
monger" and "Perler" are not uncommon.

So, no, I guess there's not a "definitive" nickname for us. <wink>

Tim Hammerquist
-- 
$ du -sh /mnt/windows | awk 'END{printf "%s wastes %s\n",$2,$1}'


From cliechti at gmx.net  Tue Feb 19 17:13:36 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 19 Feb 2002 23:13:36 +0100
Subject: Alternative for \n when using telnetlib
References: <1103_1014153824@Johan>
Message-ID: <Xns91BAECD509164cliechtigmxnet@62.2.16.82>

Johan <aequitas at planet.nl> wrote in news:1103_1014153824 at Johan:
> When connecting to my adsl modem (alcatel speed touch pro) via
> telnetlib i can read the date write my username. But then i when i 
> give the \n to proceed to the password it doesn't work. In 
> interactive mode it has the same problem. It just writes a new line. 
> 
> Is there a alternate for the \n command?

not in that sense. its just that some protocols and especially windows 
stuff that require '\r\n' as line ending.

chris

-- 
Chris <cliechti at gmx.net>



From jeremy at alum.mit.edu  Tue Feb  5 22:09:42 2002
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: 5 Feb 2002 19:09:42 -0800
Subject: ZODB.FileStorage -- copy?  save as?
References: <3C606CCE.ACA1472D@bioreason.com>
Message-ID: <b0f083db.0202051909.2a8f0327@posting.google.com>

Mitch Chapman <Mitch.Chapman at bioreason.com> wrote in message news:<3C606CCE.ACA1472D at bioreason.com>...
> I'd like to be able to do something like "save as" with
> ZODB.FileStorage.FileStorage or a derivative thereof.  I'm having a 
> hard time even identifying alternatives, mostly because I don't yet
> understand the FileStorage implementation.

I hope you don't need to understand the implementation in order to use
it.  I'll grant that there's a lot of folklore about how to manage a
FileStorage that isn't documented anywhere.

> F'rinstance I don't know if/when it's safe to copy the associated 
> files ('foo', 'foo.tmp'), whether or not one needs to copy the tmp
> file, what the tmp file does, how one might attach an existing DB 
> to a new FileStorage instance, etc.

It's safe to copy the file "foo" -- assuming you mean what is by
convention called "Data.fs".  Data.fs.tmp is used to store temporary
data while a transaction is committing.  There's also Data.fs.index
which is good to copy, because having an index makes the initial open
of the storage faster.

I'm not sure that "attaching an existing DB to a new FileStorage" is a
sensible  thing to do; i.e. it's not a sensible question, although it
may not be obvious why.  A FileStorage is a log of all the
transactions committed on a DB.  You can't just copy (as in cp) one
file to another, because you don't know anything about the
transactions or objects in the log.

There's a storage level api for copyTransactionsFrom that copies
transactions from one storage to another, but that's still a feature
that should only be used if you know what you are doing.  It's more
likely that you'd want to copy objects at the ZODB level; load an
object graph out of one storage and create copies of all the objects
in a new storage.
 
> Can anyone provide clues?  So far all I have found is the source 
> code itself :) and 
> 	http://www.zope.org/Documentation/Developer/Models/ZODB/

Hope this was helpful.

Jeremy


From akhar at videotron.ca  Fri Feb 15 19:35:58 2002
From: akhar at videotron.ca (akhar)
Date: Fri, 15 Feb 2002 16:35:58 -0800
Subject: saving complex data structures?
Message-ID: <Ocfb8.9438$bS4.618144@weber.videotron.net>

How does one go about saving and restoring complex data structures? like if
I have made a daatabase in python how do I save the contents for fast
retrieval, in other ways than text files which do not always take into
account structure? ( I thought of XML but ... I am wondering if there is a
binary way?)

Akhar




From dgallion1 at yahoo.com  Mon Feb 11 19:24:01 2002
From: dgallion1 at yahoo.com (Darrell)
Date: 11 Feb 2002 16:24:01 -0800
Subject: socket.connect blocks main thread?
Message-ID: <b0c5da3a.0202111624.10b9210d@posting.google.com>

On Win2k s.connect blocks the printing of '.' while it times out.
Not what I expected, any ideas?
If sourceforge is wasn't soooo slow right now, I'd take a look.
--Darrell

import socket, time, thread

def getIpConnThread(self):
    while 1:
        time.sleep(3)
        host='localhostxxx'
        try:
            s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            s.connect((host,2000))
        except:
            print 'Failed to connect to:',host
        
thread.start_new_thread(getIpConnThread, (None,))
while 1:
    time.sleep(0.5)
    print '.',


From maxm at mxm.dk  Thu Feb 14 04:01:13 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 14 Feb 2002 09:01:13 +0000
Subject: py2exe & McMillan's installer (was Re: Is making Python executables a reality?)
References: <E16bC9r-0000zp-00@mail.python.org> <mailman.1013665162.16163.python-list@python.org>
Message-ID: <3C6B7CD9.6020405@mxm.dk>

ahimsa consulting wrote:


 > Could someone kindly post the URL for downloading the py2exe program and
 > also the "Gordon McMillan's installer" that David referenced (see below).


Here goes:

http://starship.python.net/crew/theller/py2exe/

http://www.mcmillan-inc.com/install1.html

I found them via Google and about 4 clicks.

regards Max M




From cmaloney at physics.ucsb.edu  Wed Feb 20 12:50:05 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: Wed, 20 Feb 2002 09:50:05 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu> <a4vaqi$j8e$1@bob.news.rcn.net>
Message-ID: <3C73E1CD.8070100@physics.ucsb.edu>

Thanks Dave.  You raise some issues that've been puzzling me.

  rasmussn at lanl.gov has been good enough to bring the lanl SILOON
project to my attention.

http://www.acl.lanl.gov/siloon/

David Abrahams wrote:

> "Craig Maloney" <cmaloney at physics.ucsb.edu> wrote in message
> news:3C7283E4.4050604 at physics.ucsb.edu...
> 
> 
> One of the design goals for Boost.Python was to avoid the language support
> issues that tend to plague systems which attempt to parse C++. Parsing C++
> is hard; AFAIK all of the wrapping systems which try to do it break down in
> some areas. Boost.Python uses the builtin type inference abilities of your
> C++ compiler instead of parsing the C++ code separately.

 > [...]

 

As I understand it, the SILOON project uses a commercial parser known as
PDT from uoregon.  Have you, or has anyone else had any experience with 
the SILOON/PDT project?  On the face of it, my guess would be that you 
would be somewhat skeptical of their claims?

As for the CERN Root interpereter (CINT) 
http://root.cern.ch/root/Cint.html -- which I had already heard about 
from my friends in the know in the particle physics community -- I 
would assume that many here also harbor the same skeptecism toward the 
C++ interpereter as they do toward the C++ parsing systems?


Cheers,
Craig



From sabren at manifestation.com  Wed Feb 20 01:13:19 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Wed, 20 Feb 2002 01:13:19 -0500 (EST)
Subject: calling a function indirectly
In-Reply-To: <a4v7s6$13tk@r02n01.cac.psu.edu>
Message-ID: <Pine.LNX.4.44.0202200112010.14394-100000@mercury.sabren.com>

On Tue, 19 Feb 2002, Rajarshi Guha wrote:

> Hi,
>   is it possible to have a variable contain a function name, and use the 
> variable to call the function? 
> An example:
> 
> def func:
>    some code
> 
> funcvar = 'func'
> 
> Now use funcvar (somehow!) to call func


I also think the easiest thing to use is funcvar = func,
but if for some reason it has to be a string, you can
use exec, or:

locals()[funcvar]("whatever")


Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From roy at panix.com  Wed Feb 20 23:14:00 2002
From: roy at panix.com (Roy Smith)
Date: Wed, 20 Feb 2002 23:14:00 -0500
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <slrna78p0v.1jg.grante@tuxtop.visi.com>
Message-ID: <roy-A9169A.23140020022002@reader1.panix.com>

grante at visi.com (Grant Edwards) wrote:
> for non-empty strings type(s) == type(s[0]).

Slightly more generally, type(s) == type(s[0:0]) for all strings, even 
empty ones.


From wurmy at earthlink.net  Wed Feb 20 23:06:18 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Thu, 21 Feb 2002 04:06:18 GMT
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com>
Message-ID: <3C747268.6C756B9D@earthlink.net>

Jason wrote:
> 
> How do I sort the characters in a string, so far I use this:
>         a="qwerasdfzxcv"
>         b=[x for x in a]
>         b.sort()
> 
> Why doesn't sort() return the sorted list.  I would like to chain it
> to other operations:
>         b=[x for x in a].sort()

The other replies told you why list.sort() doesn't return
the sorted list. You can easily roll your own function,
though:

>>> def sort2(lst):
	z = lst[:]
	z.sort()
	return z

>>> a = [1, 4, 6, 2, 9, -2]
>>> b = sort2(a)
>>> b
[-2, 1, 2, 4, 6, 9]
>>> a
[1, 4, 6, 2, 9, -2]

This sort2() function returns a new, sorted list
without affecting the original one. Don't use this
when performance is an issue, though... it's not
very efficient because it copies the original
list first.

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From brueckd at tbye.com  Thu Feb  7 17:55:43 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Thu, 7 Feb 2002 14:55:43 -0800 (PST)
Subject: encapsulation, typing & other questions
In-Reply-To: <3C62FD4D.2B908BA9@daylight.com>
Message-ID: <Pine.LNX.4.33.0202071444480.5364-100000@ns0.tbye.com>

On Thu, 7 Feb 2002, Ragu Bharadwaj wrote:

> brueckd at tbye.com wrote:
> >
> > On Thu, 7 Feb 2002, Ragu Bharadwaj wrote:
> >
> > I don't know of any plans to do this, and the benefits are probably less
> > than you'd imagine. This issue tends to come up as a theoretical problem
> > much more often (if ever) than it does in practice.
>
> Beg to disagree completely. I've seen too many folks encounter problems
> to believe this.

Beg to disagree completely. In your original post you said you were new to
Python, right? Well, in Python this problem does not happen much at all in
practice. And if it does happen, you will probably find out about it in
less time that it would have taken you to write out some type declaration.

Search comp.lang.python on Google groups - you'll find plenty of messages
about how type info in function decls might theoretically help and few or
no messages from people reporting actual errors that would have been
caught.

Flawed explicit type systems actually add complexity and bugs.

-Dave




From jkraska at san.rr.com  Fri Feb  8 20:30:02 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 09 Feb 2002 01:30:02 GMT
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <a40e2r.3vvegs9.1@kserver.org> <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net> <tbf14a.ep7.ln@wombat.san-francisco.ca.us> <a41ob2$9k3$1@panix2.panix.com> <Xns91AFAE30B3A22RASXnewsDFE1@130.133.1.4>
Message-ID: <5bv86ugamufirrio4k1pivtm4iktsakupv@4ax.com>

>Maybe shortened to Perlpro(s)?

Perlobites? Perlabees? Perl *weenies*?

:-)

C//



From starx at pacbell.net  Mon Feb  4 04:02:53 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 4 Feb 2002 09:02:53 GMT
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com> <7xbsf7dqj9.fsf@ruckus.brouhaha.com> <B883FC7B.4309%jonathan@onegoodidea.com>
Message-ID: <Xns91ABAA86A9D3RASXnewsDFE1@130.133.1.4>

Jonathan Hogg || Mon 04 Feb 2002 12:53:15a:
 
> Which *old* code of someone's has actually been broken so that it will
> not run correctly on the newer interpreters?

If you want a specific example: the introspect code in pyCrust. It recursevly 
dives through __class__ attributes, and now for everything except old classes  
this eventually leads to 'type'. type.__class__ is type, therefore runaway 
recursion. I think using dir() would have also made it Do Bad Things due to 
the 2.2 changes, but i fixed this quite a while before finding the real 
problem, so i don't know if it actually hurt.

If this is what you mean by '*old*' at least...


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From eriggs at kollmorgen.com  Fri Feb 15 12:00:27 2002
From: eriggs at kollmorgen.com (eriggs)
Date: 15 Feb 2002 09:00:27 -0800
Subject: How to disable a button control in wxPython?
Message-ID: <3bff5060.0202150900.77c5f5fa@posting.google.com>

I'm somewhat new to Python, and I'm trying to figure out how to
disable a button control.  I have included my test source code for
reference.  Thanks in advance for any help offered.  --  Earl

######################################################################
from wxPython.wx import *

class ButtonFrame(wxFrame):
    'Creates a frame with a single button in the center'
    def __init__(self):
        wxFrame.__init__(self, NULL, -1, 'wxPython',
                         wxDefaultPosition, (330, 210))

        panel = wxPanel(self, -1) 

        #wxTextCtrl(wxWindow* parent, wxWindowID id, 
        #           const wxString& value = "", 
        #           const wxPoint& pos,
        #           const wxSize& size = wxDefaultSize, 
        #           long style = 0, const wxValidator& validator, 
        #           const wxString& name = "text")
        tc = wxTextCtrl(panel, -1, "", wxPoint(10, 10), \
                    wxSize(300, 100), wxTE_MULTILINE|wxTE_READONLY)
        self.msgtext = tc

        button1 = wxButton(panel, 111, 'Start', wxPoint(10, 130), \
                           wxSize(100, 30))
        EVT_BUTTON(self, 111, self.onButtonStart)

        button1.SetDefault()
        self.StartButton = button1
 

        button2 = wxButton(panel, 112, 'Quit', wxPoint(210, 130), \
                           wxSize(100, 30))
        EVT_BUTTON(self, 112, self.onButtonQuit)
        EVT_CLOSE(self, self.OnCloseWindow)

        self.QuitButton = button2

        # Disable Quit button
        #self.QuitButton.Disable()
        # Enable Start button
        #self.StartButton.Enable()


    def onButtonStart(self, event):
        'Create a message dialog when the button is clicked'
        dlg = wxMessageDialog(self, 'Starting...',\
                              'Start', wxOK)
        dlg.ShowModal()
        dlg.Destroy()

        self.setText("Starting program...")
        # Enable Quit button
        #self.QuitButton.Enable()
        # Disable Start button
        #self.StartButton.Disable()
        # Start program loop


    def onButtonQuit(self, event):
        'Create a message dialog when the button is clicked'
        dlg = wxMessageDialog(self, 'Quitting...',\
                              'Stop', wxOK)
        dlg.ShowModal()
        dlg.Destroy()

        self.setText("Stopping program...")

        self.Close(true)


    def OnCloseWindow(self, event):
        self.Destroy()


    def setText(self, msg):
        'Create a message dialog when the button is clicked'
        self.msgtext.Clear()
        self.msgtext.AppendText(msg)


class App(wxApp):
    def OnInit(self):
        'Create the main window and insert the custom frame'
        frame = ButtonFrame()
        frame.Show(true)
        return true # Yes, continue processing

# Create the app and start processing messages
app = App(0)
app.MainLoop()


From alanmk at hotmail.com  Tue Feb  5 16:48:22 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 5 Feb 2002 13:48:22 -0800
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com> <f72dac0d.0202050751.68bc0a7@posting.google.com> <3j406uo6g577f6djre8jeunin6q11o8p6n@4ax.com>
Message-ID: <f72dac0d.0202051348.4c45f4ff@posting.google.com>

Steve Horne <steve at lurking.demon.co.uk> wrote:
> XML is bound to be easier to parse for the compiler than the existing
> Python language, because it avoids the following issues...
> 
> 1.  Resolution of ambiguities using associativity, precedence and
>     preference for shift over reduce.

But creates a whole new problem in that a DTD/Schema/RelaxNG has to 
be created that validates the structure of the xython program ....

It's been too long since I did compiler theory (1989 in fact, undergrad
course), so I don't even know if it would be possible to express the
necessary constraints for xython in an XML DTD.....

> 2.  Classifying words as identifiers or keywords, particularly where
>     this depends on 'from __future__'.

This is a good point, but to my mind, not enough of a benefit to justify 
a whole new syntax. And "from __future__" seems to do the job quite well.

> 3.  Awkward lexical handling such as recognition of indentation.

Let's not start on the pros and cons of indentation :-) Personally,
I like it, I think its one of python great features.

> Error reporting should also be simpler. In fact, syntax errors should
> be impossible (assuming the editor is configured properly) - only
> semantic errors should get as far as the compiler.

See above about creating a DTD. If you don't have a DTD, then all you're
guaranteed by your editor is a well-formed xython file, but it might
still contain meaningless *structures*

> The XML would certainly not be easy for people to edit, but it's not
> meant to be - the presentation is an issue for the editor.

Hand-editing would be a PITA. One of the xython proponents seems to be
focussed more on the ability to autogenerate python/xython from XSLT,
but for the life of me, I can't think of a use case for this that would
not be better served with a dedicated code manipulation/generation tool.

> But yes, I can see problems. For instance, the error reports would
> relate to the XML AST - not to the visible representation of the
> source code.

And don't forget that speedwise, an XML parser would be extemely unlikely
to be faster than a dedicated parser, even just checking for well-formedness,
let alone validity.....

Alan, the push-down automaton....


From mike.mellor at us.army.mil  Mon Feb  4 16:38:02 2002
From: mike.mellor at us.army.mil (Mike Mellor)
Date: Mon, 4 Feb 2002 16:38:02 -0500
Subject: directly executing a python script in Unix
In-Reply-To: <1012854277.1061.95147.m10@yahoogroups.com>
References: <1012854277.1061.95147.m10@yahoogroups.com>
Message-ID: <20020204163802.6f657228.mike.mellor@us.army.mil>

On Mon, 04 Feb 2002 20:24:37 +0000
python-list at yahoogroups.com wrote:

>    Date: 4 Feb 2002 12:08:43 -0800
>    From: laymon at crd.ge.com (Marc Laymon)
> Subject: directly executing a python script in Unix
> 
> I am just learning python.  I have written a few simple scripts which
> run fine when invoked using "python myscript.py".  The Python manual
> says you can execute a Python script directly In Unix by making the
> first line "#!/usr/bin/python", then giving the command
> "./myscript.py".  However, when I try this, I get an error "Command
> not found.".  I get the same error under both Linux and Solaris. 
> Using the same format for perl works fine.  Why doesn't the
> "#!/usr/bin/python" line work for Python ?  (perl and python are both
> in /usr/bin on my Linux system.)
> 
> Thanks.
> 
> Marc Laymon
> -- 

Marc - 

You need to make the script executable:

chmod +x myscript.py

That should do it.  Hope it helps.

Mike



From jkraska at san.rr.com  Mon Feb 11 21:24:42 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 12 Feb 2002 02:24:42 GMT
Subject: Fear of autocoding! >Re: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <3c681c43.13768618@127.0.0.1>
Message-ID: <abvg6ucfaeduk100pgdjh2oc12ifqqak7h@4ax.com>

>Perhaps one metric is to count the number of people who are telling you,
>bluntly, that you're a loon.  It just could be that they're right.  :-)

One of my two uncles is convinced that he's simply a creative genius
that other people can't be bothered to understand. He's been working
on the same perpetual motion device for the last 5 years. He believes
that he can use leverage to capture gravitic energy: essentially he
is attempting to reconfigure the device while in motion so that it
captures the "down" part of gravity more efficiently in parts of the
device such that the oppositional energy of the device is more than
equalized. It's the classic perpetual motion engine.

In a quiet moment quite unrelated to any description of his device to
me, I recommended for him SSRI medication. Serotonin deficiency is
associated with obsessive thinking patterns.


C//



From shoot at shootclub.com  Wed Feb 20 23:46:26 2002
From: shoot at shootclub.com (Shootclub Webmaster)
Date: 20 Feb 2002 20:46:26 -0800
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <6ef58307.0202202046.3e4f90cd@posting.google.com>

kdahlhaus at yahoo.com (Kevin Dahlhausen) wrote in message news:<283adf56.0202201551.7026d548 at posting.google.com>...
> Does anyone know of an inexpensive web host that allows compilation
> and use of Python extensions from  your own directory?

http://qwk.net

qwk.net has supplied me with tremendous service.

$30 a month on a shared machine with python (1.5.2)

I installed MySQLdb myself as well as DateTime in my home
directory (compiled using gcc).


Good Luck with your search,
Shootclub Webmaster
http://shootclub.com/cgi-bin/web.py


From tkeon0 at netscape.net  Tue Feb 19 23:29:07 2002
From: tkeon0 at netscape.net (Trevor Keon)
Date: 19 Feb 2002 20:29:07 -0800
Subject: importing dlls with python
Message-ID: <e6bacef5.0202192029.2d095cc8@posting.google.com>

Hi, I am new to Python, and was wondering if it was possible to import
dlls written in Delphi 5?  I have tried a number of methods to make it
work, and I either get:

An illegal operation in windows,

or:

dynamic module does not define init function (initxxx), 

where xxx is the dll name.  Now in the dll I have added a initxxx
function that returns an int, and have also tried returning a boolean
(also tried using a procedure, same result).  When I export the
initxxx function I get the illegal operation.  For those who are
familiar with Delphi:

function initxxx: Integer;  StdCall;
  export; 
...

function initxxx: Integer;
begin
...
end;  

And in the project file:

export
  initxxx;

  If I do not have the export in the project file then I get the
second error (no init function).  Can anyone help me here?  Or does
anyone know another script language that can call dlls easily?  I have
already looked at Tcl and Perl, but none of them seem feasible.

Thanks in advance,

Trevor


From pirx at mail.com  Wed Feb 13 17:05:17 2002
From: pirx at mail.com (pirx at mail.com)
Date: Thu, 14 Feb 2002 09:05:17 +1100
Subject: a short overview of Python's ecological niche
Message-ID: <a4eo0i$2s51$1@si05.rsvl.unisys.com>

http://www.ddj.com/documents/s=2287/ddj0202a/0202a.htm

fantastic read. about what languages should be, cultural differences and
plenty of pointers to useful ideas. I think that Python got _very_ unfairly
lumped with Perl, but I also think that Mondrian may be becaming interesting
to some people.





From brian at sweetapp.com  Tue Feb  5 13:14:36 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 5 Feb 2002 10:14:36 -0800
Subject: python and xmlrpc
In-Reply-To: <3C5F9F42.9040508@wanadoo.fr>
Message-ID: <000901c1ae70$f8a2e600$445d4540@Dell2>

Kondor wrote:
> HI,
> I'm looking for a document explained how to write xmlrpc servers and
> clients.
> Can you help me ?
> I've seen the articles on o'reilly.com (ONLAMP.org) and pythonware.org
> Thanks.

So what information are you looking for i.e. what do you need to know
that wasn't explained in those articles? The actual Python documentation
is at:

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

There are also examples in xmlrpclib.py and SimpleXMLRPCServer.py. 

Cheers,
Brian




From dyoo at hkn.eecs.berkeley.edu  Mon Feb 18 17:26:39 2002
From: dyoo at hkn.eecs.berkeley.edu (Danny Yoo)
Date: Mon, 18 Feb 2002 14:26:39 -0800 (PST)
Subject: My Spring project!  [A dictionary that allows substrings for key
 lookup]
Message-ID: <Pine.LNX.4.21.0202181410170.11029-100000@hkn.eecs.berkeley.edu>

Hi everyone,


I've heard so much good stuff about making extension libraries in SWIG,

    http://www.swig.org

that I finally put my foot down and tried to actually apply it toward a
real project.


I've written a wrapper around Dan Gusfield's 'strmat' Suffix Tree library,

    http://www.cs.ucdavis.edu/~gusfield/strmat.html


and I think it's all quite crazy.  *grin* If anyone's interested, I've
collected my notes and code here:

    http://hkn.eecs.berkeley.edu/~dyoo/programming/current.html


As a warning: there are definitely memory leaks in the wrapper.  Also,
parts of the 'strmat' library haven't been wrapped yet because they look a
little... well, fragile.  But it's still very interesting, and perhaps
even useful!


As an example, I've written a small SubstringDict class that allows one to
use substrings of dictionary keys:

###
[Adyoo at coffeetable:~/pystrmat-0.6$ python SubstringDict.py
Reading the dictionary.  Please wait.
45392 words now indexed.
Please enter a word; I'll look for all entries in the dictionary
that have the word as a substring.
word? py
59 words found: 
['espy', 'copy', 'swampy', 'skimpy', 'sleepy', 'photocopy',
'photocopying', 'Skippy', 'canopy', 'preoccupy', 'anisotropy', 'lumpy',
'puppy', 'occupy', 'occupying', 'drippy', 'pygmies', 'pygmy', 'pyramid',
'papyrus', 'floppy', 'physiotherapy', 'Harpy', 'pyramids', 'jumpy',
'snoopy', 'droopy', 'copying', 'copyright', 'copyrightable',
'copyrighted', 'copyrights', 'copywriter', 'unhappy', 'microscopy',
'pyre', 'peppy', 'creepy', 'spy', 'spyglass', 'spying', 'philanthropy',
'sloppy', 'Cappy', 'python', 'entropy', 'psychotherapy', 'shipyard',
'soapy', 'poppy', 'happy', 'hardcopy', 'crappy', 'choppy', 'capybara',
'snappy', 'syrupy', 'spectroscopy', 'therapy']
word? moon
14 words found: 
['mooned', 'mooning', 'moonlight', 'moonlighter', 'moonlighting',
'moonlit', 'moons', 'moonshine', 'honeymoon', 'honeymooned',
'honeymooner', 'honeymooners', 'honeymooning', 'honeymoons']
word? tutor
9 words found: 
['statutorily', 'statutoriness', 'statutory', 'tutor', 'tutored',
'tutorial', 'tutorials', 'tutoring', 'tutors']
###


This may look mundane, but what's neat about this is that when it looks up
a substring key, it isn't scanning through all of the dictionary.

I still have to do some more optimization and code cleanup, but I hope
that this is useful for people.  I did remember hearing a question about
this on Python-list a while back, so I'll forward this message to the main
list as well.




From kondor at wanadoo.fr  Tue Feb  5 11:28:00 2002
From: kondor at wanadoo.fr (kondor)
Date: Tue, 05 Feb 2002 17:28:00 +0100
Subject: SOAP and python
Message-ID: <3C600810.4080805@wanadoo.fr>

I've download the SOAP package for python and their is an exemple of a client 
and a server (echoServer.py and echoClient.py)
It function well on my machine (localhost)
but when I launch the server on my machine and try the client on an other 
machine (in my network), it dosen't work.
yet I've change localhost to the server ip adress in the clientEcho.py.
The client say :

PythonWin 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2001 Mark Hammond (MarkH at ActiveState.com) - see 
'Help/About PythonWin' for further copyright information.
Traceback (most recent call last):
   File "D:\Python21\Pythonwin\pywin\framework\scriptutils.py", line 301, in 
RunScript
     exec codeObject in __main__.__dict__
   File "C:\projet\echoClient.py", line 23, in ?
     print server.echo_ino("moo")
   File "D:\Python21\lib\SOAP.py", line 3603, in __r_call
     self.__hd, self.__ma)
   File "D:\Python21\lib\SOAP.py", line 3521, in __call
     config = self.config)
   File "D:\Python21\lib\SOAP.py", line 3397, in call
     r.putrequest("POST", real_path)
   File "D:\Python21\lib\httplib.py", line 437, in putrequest
     self.send(str)
   File "D:\Python21\lib\httplib.py", line 379, in send
     self.connect()
   File "D:\Python21\lib\httplib.py", line 363, in connect
     self.sock.connect((self.host, self.port))
   File "<string>", line 1, in connect
error: (10061, 'Connection refused')
 >>>


Can you help me ?



From phd at phd.pp.ru  Tue Feb 12 16:08:51 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 13 Feb 2002 00:08:51 +0300
Subject: PyFT and PIL
In-Reply-To: <oqofiu4j6o.fsf@titan.progiciels-bpi.ca>; from pinard@iro.umontreal.ca on Tue, Feb 12, 2002 at 02:02:55PM -0500
References: <mailman.987996754.9967.python-list@python.org> <9c11bn$sph@gap.cco.caltech.edu> <oqu1sm4vqr.fsf@titan.progiciels-bpi.ca> <20020212173402.G8503@phd.pp.ru> <oqofiu4j6o.fsf@titan.progiciels-bpi.ca>
Message-ID: <20020213000851.C13180@phd.pp.ru>

On Tue, Feb 12, 2002 at 02:02:55PM -0500, Fran?ois Pinard wrote:
> >    make -f Makefile.pre.in boot
> >    make
> 
> P.S. - There was no `Makefile.pre.in' in the distribution, but there were
> a few others here and there on my disks, it was enough to copy one of them.

   It (note single form) is standard file. Yo can take it from any Python
distribution up to (but not including) 2.2.
   Setup.in is specific for every project, though.

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



From gcordova at hebmex.com  Fri Feb  8 10:28:01 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 8 Feb 2002 09:28:01 -0600 
Subject: Saving Browser State without Cookies
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A0E@SRM201002>

I've use two ways to pass session-ident-cookies without
using browser cookies:

1. Append "/xxxxxxx" to all program URIs, and retrieve that
   information via the PATH_INFO environment variable.

2. Send all form parameters via POST, and append "?xxxxxx"
   to the form's action URI; retrieve all parameters using
   a POST processor, and the session identifier is in QUERY_STRING.

But, most cgi packages are quite lame, the mash together
all POST and QUERY_STRING parameters, so you end up having to
modify them.

Also, IIS has trouble with PATH_INFO, you can't give it
a URI like:

http://www.someserver.com/bleh/ackktthp.asp/uusjw2233

because it'll search for a file "uusjw2233" in a directory
called "ackktthp.asp", which is not what you want. I've
used this successfully in Apache though, which correctly
passes "/uusjw2233" in the PATH_INFO parameter, to the
"ackktthp.cgi" program (no ASPs in Apache for me).

Ah! I've seen another one, but not everywhere.

You *could* use Apache's capabilities to remap URIs,
and construct URIs like this:

http://www.someserver.com/AppName/xxxxxxx/blahblah.cgi

And you remap the ./xxxxxxxx/ directory to point to
the real application directory. When a user logs on,
you create the mapping, and when he logs off (or his
session expires) you destroy the mapping. Then, using
some judicious cutting (or regexing) you can extract
the session identifier (the "xxxxxx").

Another way is to use, if you have access to, a "wildcard"
domain, where anything of the form:

http://xxxxxx.appname.someserver.com/ ....

gets sent to:

http://appname.someserver.com/

you can parse the host name used, and obtain from there
the session identifier. I wouldn't know how to configure
the DNS for that though, but it's an idea.



So, choose your poison:

1. Client-side cookie.
2. Use POST params, and QUERY_STRING session identifier.
3. Use POST and GET params, and PATH_INFO session identifier.
4. Use a hidden form parameter.
5. Use a virtual directory mapping (apache?)
6. Use a "wildcard domain".

Others can give you more info.

Good luck!

-gustavo



From alan.gauld at bt.com  Thu Feb 28 09:59:11 2002
From: alan.gauld at bt.com (Alan Gauld)
Date: Thu, 28 Feb 2002 14:59:11 +0000
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net>
Message-ID: <3C7E45BF.1080804@bt.com>

Ron Stephens wrote:

> Well, let's see, I bought this 1292 page book from Amazon a week ago for
> $72. This is the most I have ever paid for a computer book. 


I've paid more but I'm glad that I got my complimentary copy
as a reviewer! Its a lot of money for paper!

> This book is one of the most enjoyable I have ever read. 


I'm glad you like it, books are very much a personal
preference kind of thing.

> the paper is the finest in any Python book yet; 


I assume you mean finest in the sense of thinest?
My copy feels like tissue paper...

> languages, and I sort of expected them to half-heartedly do a me-too
> book by just applying the same old formula to a Python version.


I think thats what they have done, looking at their C++ and
Java books they are very similar. But if those books work
then why not?

> Regular Expressions, and Unicode. There are even excellent and extensive
> appendixes 


One of my criticisms of the book is that the HTML and XHTML 
appendices really could have been rolled together - they are 
obviously cut n paste copies with minor differences where 
appropriate - there are whole pages with only a couple of
words different! But thats how you get 1300 pages in a book
I guess!

> very best college text books  I have encountered. Even the summaries
> after each chapter are useful; 


Here I agree, I did think te summaries were useful - for the 
first half of the book I just read the summary sections
without reading the main text,  then when I got to the
chapters I really wanted to read I read both.
That worked OK.

> I am excited.


In that case the book has succeeded, excited readers
are every authors dream! :-)

Alan Gauld
Author of the "Learn to program" website
http://www.freenetpages.co.uk/hp/alan.gauld



From SBrunning at trisystems.co.uk  Thu Feb 21 07:29:49 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Feb 2002 12:29:49 -0000
Subject: Why this behavior? (was: Stupid Lambda Tricks)
Message-ID: <31575A892FF6D1118F5800600846864DCBCCD6@intrepid>

> From:	Jonathan Gardner [SMTP:jgardn at alumni.washington.edu]
> I've always wondered:
> 
> >>> "abc".upper()
> "ABC"
> >>> [3,1,2].sort()
> < nothing >
> 
> Why is that? Is it just easier to implement it this way, or is there some 
> sound philosophical reason for this?
 
The latter.

Strings are immutable, so methods which 'change' them must in fact be
creating a new string. This has to be returned by the method - otherwise,
where would it go?

Lists are mutable. Methods actually *do* change them - 'sort' sorts the list
in place. To remind you of this fact, nothing is returned by the method. If
'sort' *did* return the sorted list, you could do:

spam = [3, 2, 1]
eggs = spam.sort()

Now, you might think that eggs is sorted and spam isn't, but if fact, spam
*is* sorted.

So instead, sort returns nothing, and you'll get a runtime error of some
sort in no time, and you'll find the problem quickly.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From quinn at hurl.ugcs.caltech.edu  Mon Feb  4 17:29:37 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 4 Feb 2002 22:29:37 GMT
Subject: obj2 = ojb1 = range(len(somelist))
References: <m31yg6jgta.fsf@moroni.pp.asu.edu> <mailman.1012524543.30091.python-list@python.org>
Message-ID: <slrna5u2qh.i0j.quinn@hurl.ugcs.caltech.edu>

On 31 Jan 2002 19:45:42 -0500, Tom Jenkins <tjenkins at devis.com> wrote:
>On Thu, 2002-01-31 at 17:28, David Bear wrote:
>> I wanted to create 2 distinct lists that were derived from range
>> function.  rather that doing
>> 
>> obj1 = range(len(somelist))
>> obj2 = range(len(somelist))
>> 
>> I wanted to do
>> 
>> obj2 = ojb1 = range(len(somelist))
>> 
>> Yet, both names point to the same object -- they're not two objects.
>> 
>> Is there some other syntax that I could use to avoid having to call
>> range(len(list)) twice and still create two objects?
>
>i didn't think this was going to work but hey the interpreter was there
><wink>
>>>> l1 = l2[:] = range(len(somelist))
>>>> print l1, l2
>[0, 1, 2, 3, 4] [0, 1, 2, 3, 4]
>>>> print id(l1), id(l2)
>135566380 135560684
>>>>

That should only work if 'l2' is already an object that supports slice
assignment.


From philh at comuno.freeserve.co.uk  Thu Feb  7 12:32:25 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 17:32:25 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com>
Message-ID: <slrna65eh9.ahr.philh@comuno.freeserve.co.uk>

On Wed, 06 Feb 2002 16:22:09 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>>>That's non hygienic behavior.
>
>>And what if I decide that for my particular application, that's just
>>the behaviour I want? Or do you think that you know better than all
>>the users of your macro system, even though you've not met them and don't
>>know anything about the software they are writing?
>
>Well, since one has to go through hoops to make a macro hygienic,
>it's not hard at all to make them unhygienic. The plan is for hygiene
>because this is generally regarded as a best practice by the community.
>Which is to say, this as a default is regarded as what is most likely
>to service _everyone_ best, allowing for exceptions of course.
>
>You say "that you think you know better" as a scathing remark

I apologize for the tone of my previous remark.

It's just that being forced -- "for my own good" -- not to use certain
features because someone has decreed them intrinsically bad, doesn't
appeal to me at all.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From rdsteph at earthlink.net  Thu Feb 28 17:37:55 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Thu, 28 Feb 2002 22:37:55 GMT
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net> <20020227163411.71b4b4d8.alex@netWindows.org> <Xns91C39C93F933Edenroearthlinknet@207.217.77.23>
Message-ID: <3C7EB1DC.58E7F1D5@earthlink.net>

Hmmm, very curious to me. I am thinking that , maybe, I enjoy the Deitel
and Deitel style partly because I have little formal academic background
in computer science. Maybe the Deitel books appeal more to those readers
who are relatively lacking in background.

I do think the Deitels use a lot of words to make their points, and
there is a lot of repetition, in that the book is full of highlighted
Common Programming Errors, Tips, and Observations, all of which repeat
points just made in the text. Some might find those Tips needlessly
redundant and time wasters. Likewise, some might find the relatively
verbose text to be overkill, while I find it helps me to understand. And
I tend to learn best by repetition. 

Anyway, its interesting to me. I don't think I can draw any conclusions,
but I tender the hypothesis that the Deitel Python book might be better
for relative neophytes rather than accomplished Python programmers.

I'll no doubt be thinking about this as I read the more advanced
chapters of the book.


From dmaster at synopsys.com  Mon Feb  4 12:59:12 2002
From: dmaster at synopsys.com (David Masterson)
Date: 04 Feb 2002 09:59:12 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <690.793T1095T13354397threeseas@earthlink.net> <un0yxkupb.fsf@synopsys.com> <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net>
Message-ID: <uadup5dsf.fsf@synopsys.com>

>>>>> Timothy Rue writes:

> NEED TO PUT MORE THOUGHT INTO THIS!!!
> *************************************

I couldn't have said it better myself!

Your (extended) program appears to be no better than the previous one
I looked at in that it doesn't appear to do anything useful.  My
memory of AREXX is ~10 years old, so some things in your program are
not obvious (like the PARSE command).  Overall, though, it appears to
merely do some convoluted printouts that I can't make heads or tails
of.

Since you can't seem to make us understand the value of V.I.C. from a
"whitebox" look at it, let's try a "blackbox" approach.  In a
"blackbox" approach, you give some examples of (expected) input and
(expected) output from V.I.C. and leave it to the reader to figure out
how V.I.C. will (or should) work.  Obviously, for this to work, your
examples must be well thought out and very concrete.  They must show
the usefulness of what V.I.C. does and they shouldn't need much in
explanation of how the input became the output.  And the more examples
you can provide, the better are the chances that someone will
understand what you're trying to do and develop the program to do it.

My recommendation, though, is that you put these examples on your
web-page and let us go look at them.  Posting them here would be bad
because they would probably get lost.

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From bkelley at wi.mit.edu  Wed Feb 20 17:13:28 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Wed, 20 Feb 2002 17:13:28 -0500
Subject: Python on Windows CE
References: <a4ev4g012jn@enews4.newsguy.com> <3C6D587B.5040203@wi.mit.edu> <Uvdc8.31$RP7.31260710@newssvr14.news.prodigy.com>
Message-ID: <3C741F88.4070304@wi.mit.edu>

Keith Farmer wrote:

> A revival of Python CE has been going on for a while, but nobody's gotten
> around to compiling 2.2 for the Clio.  Check Hammond's pages for links to
> the current Python CE effort.
> 

What compilers are needed for this effort?  I assume Visual C++ 6.0 that 
can target the Clio's processor.


> ----------
> Keith J. Farmer
> kfarmer at thuban.org
> http://www.thuban.org
> 




From othello  Wed Feb 27 01:59:03 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 01:59:03 -0500
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com> <7xlmdglwpp.fsf@ruckus.brouhaha.com>
Message-ID: <a5i03u$6tg$1@bob.news.rcn.net>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xlmdglwpp.fsf at ruckus.brouhaha.com...
> Paul Rubin <phr-n2002a at nightsong.com> writes:
>
> > I thought I'd toss out the following two ideas:
>
> Sigh.  "Two ideas" followed by sections labelled 1, 2, and 3.  I
> should have labelled the differently.  There are two ideas as stated;
> sections 1 and 2 are two parts of the same idea.  Section 3 is the
> second idea.  Sorry for confusion.


How utterly Pythonic!

"NOBODY expects the Spanish Inquisition! Our chief weapon is
surprise...surprise and fear...fear and surprise.... Our two weapons are
fear and surprise...and ruthless efficiency.... Our *three* weapons are
fear, surprise, and ruthless efficiency...and an almost fanatical devotion
to the Colonies.... Our *four*...no... *Amongst* our weapons.... Amongst our
weaponry...are such elements as fear, surprise.... I'll come in again. (Exit
and exeunt) "


Raymond

_ ~
@ @
\_/





From starx at pacbell.net  Fri Feb  8 21:11:54 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 02:11:54 GMT
Subject: Best python way to do 'object messaging'
Message-ID: <Xns91AFB91DF83A2RASXnewsDFE1@130.133.1.4>

Ok, so i have this pre-prototype (embrionic? :) game system, rogueish, 
that  basically involves a bunch of different objects running around a 
map, containing each other, beating each other up, that kind of thing 
:). What i'm trying to do is set up a general purpouse system for 
passing messages between the interface, the objects, the AI's, the game 
world and timing system, and whatever else comes up. Here are my current 
ideas/contarints:

1. Basic interface: anything that can accept a message has a .message
(msg) function attaced to it. If the player hits a critter, some part of 
the player (i have the body and control seperated) sends the body of the 
critter a damage/type/amount/source type message.

2. It has to be able to be passed easily through various receptor 
objects that don't really care about what the message is. For example, i 
do something along the lines of 'map.messageradius( 
msg,x,y,radius,inv_arsqrd(1.2) ), which only affects a 'percentage of 
effect' segment of the message while distrubiting it to all of the 
affected cells, which do nothing but ditrubite it to whatever objects or 
creatures happen to be standing there. 

3. It has to be storable. My idea for a global timekeeping system is 
something that stores messages to send to objects at various times. For 
example, an AI object would always set a ai-tick message to itself at 
some time in the future at the end of its run, and this is the way that 
I would get keep 'live' objects updated properly.

I am currently thinking of representing the message as an object with 
various properties, and having each object that recieves one switch on 
the type of message, or even better, have a routine in a base class that 
takes some element of the message object, does a little string mangling, 
and calls a 'Rec<MESSAGETYPE>(message)' function similar to how TADS 
uses the VerDoXXX, VerIoXXXWith kind of functions for dealing with verbs 
(if anyone knows enough about TADS to understand this analogy ... it's 
one of my favorite languages :).

This is -probably- what i'm going to go with, but i'd like to see if 
anyone has any ideas/comments/insperations/pointers-to-relevant-
information, because i've never really made a system like this in any 
language, and i'd like to get it right before i build most of the rest 
of the system over it (i alredy have to refactor(i think that's how you 
use it) all of my display stuff out of the game data stuff, and remake 
the resource system -again- :)

Thanks for anything :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From wookiz at hotmail.com  Sun Feb 10 03:34:45 2002
From: wookiz at hotmail.com (wooks)
Date: 10 Feb 2002 00:34:45 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a43due.3vv9a5r.1@kserver.org>
Message-ID: <8852d33c.0202100034.16f1b47b@posting.google.com>

Sheila King <usenet at thinkspot.net> wrote in message news:<a43due.3vv9a5r.1 at kserver.org>...
> 
> I think Python is extremely appropriate for your needs. The problem is,
> you are trying to jump straight into high-level stuff like that before
> you are really comfortable with the language itself.
> 

Perhaps it more to do with the book. The approach I am taking now is
exactly how I painlessly learnt VBScript - using books with a WSH COM
bias. Most VBScript tomes have a  client side scripting bias that is
not relevant to my needs. Anyway as per below I have been using
tutorials as well.

> I recommend:
> (1) The Python tutorial that comes with the standard distribution:
> http://www.python.org/doc/current/tut/tut.html

Rereading it at the moment.

> (2) Core Python Programming by Wesley Chun
> (not for beginning programmers...aimed at someone with some programming
> experience...addresses the OO points and many features of the language
> quite well)

Yes. Thats the one that all the reviews seem to point to. If I
remember rightly the copy in the bookshop I went to was rather dated.
Will investigate further.

> (3) For an advanced online tutorial, try Dive Into Python:
> http://diveintopython.org/
> 

Thats where I started. Have read bits of it twice (skipped the HTML
and XML bits). Probably will need to read it again.


From gerson.kurz at t-online.de  Mon Feb 25 10:41:32 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 15:41:32 GMT
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <3c7a5aa1.9629343@news.t-online.de>

On 25 Feb 2002 08:18:39 -0800, mtaylor at lowebroadway.com (Moray Taylor)
wrote:

>I am desperatly seeking a good python editor, so far JEdit is the best
>I have found, but since it is written in Java, I find it painfully
>slow, and PyBrowse, although good, a little flaky.
>
>Can anyone reccommend a good python editor for UNIX/linux/IRIX. I
>don't mind paying for it, but $300 for SlickEdit is a little steep.
>
>Before all you Emacs/vi fanatics start ranting at me, I have tried,
>and hate, them both.
>
>I have only two requirements for my editor, syntax coloring, and a
>Class/function list.

I personally like SciTE best (http://www.scintilla.org), but it has no
class/function list. (However, you could expand it to do so, its open
source. Or, pay the developer $300 to add a function list ;) because
scintilla is really a cool product. Requires GTK, though. 

Note that the default settings are a kind of downer, open the "global
options file" to see what is possible. 


From francis.glassborow at ntlworld.com  Wed Feb 13 06:31:38 2002
From: francis.glassborow at ntlworld.com (Francis Glassborow)
Date: 13 Feb 2002 03:31:38 -0800
Subject: ACCU Spring Conference 2002
Message-ID: <8402ac34.0202130331.68661a17@posting.google.com>

Could I take the briefest of moments to remind everyone that the early
bird
price stops on February 18 (it is excellent value even afterwards but
why
spend more than you have to?) This conference includes a two-day track
for Python users.

-- 
Francis Glassborow
Check out the ACCU Spring Conference 2002
4 Days, 4 tracks, 4+ languages, World class speakers
For details see: http://www.accu.org/events/public/accu0204.htm


From magnus at thinkware.se  Tue Feb 19 14:42:07 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 19 Feb 2002 11:42:07 -0800
Subject: vars(os) starts help in ActivePython!!!
Message-ID: <258fd9b8.0202191142.71314c03@posting.google.com>

I have this strange behaviour in ActivePython 2.1.1 on
Windows 2000. If I type

>>> import os
>>> vars(os)

I will get the help prompt after some printouts!

help>

In the normal Python Interpreter Shell I will get
a listing of variables before

"Welcome to Python 2.1!  This is the online help utility...",

in PythonWin I only get the "Welcome..." text, and then the
prompts. (Although long lines don't wrap in PythonWin, and
sometimes they get invisible instead...)

I noticed this by chance just after I tried to get PySpy.exe
<http://www.malcolmson.com/pyspy/index.htm> to work. Running
PySpy also caused the "help>" to mysteriously appear...

Can anyone else see this strange behaviour, or is there something
fishy in my installation?


From cliechti at gmx.net  Sun Feb 17 13:43:14 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 17 Feb 2002 19:43:14 +0100
Subject: cStringIO module broken by design?
References: <3C6FEBED.40704@snakefarm.org>
Message-ID: <Xns91B8C928ACD82cliechtigmxnet@62.2.16.82>

Carsten Gaebler <clpy at snakefarm.org> wrote in news:3C6FEBED.40704
@snakefarm.org:
> It seems like the cStringIO module doesn't provide a write() method 
> for StringIO objects that were initialized with a string:
...
> Why is that? I find it quite annoying.

i wouldn't call that broken. StringIO just exposes read-only and 
write-only files.

if there were read/write files they would also need to support seek 
adn truncate etc. but yes it could be practical for some cases... (PEP 
someone?)

chris

-- 
Chris <cliechti at gmx.net>



From bldrake1 at yahoo.com  Wed Feb  6 10:23:53 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 6 Feb 2002 07:23:53 -0800
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org>
Message-ID: <fb072181.0202060723.e400eaa@posting.google.com>

Maybe I'm missing something, but here is what I get:

Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import math
>>> n = 10
>>> nf = 10.0
>>> a = 1.0
>>> b = 0.5
>>> a/n
0.10000000000000001
>>> b + a/n
0.59999999999999998
>>> a/nf
0.10000000000000001
>>> b + a/nf
0.59999999999999998
>>> 

Python does double precision floats only (see Python Essential
Reference 2nd ed. by David Beazley, p.23).  Python implements IEEE
754: 17 digits of precision with exponent in -308 to 308.

         float     double
python   64 bits   64 bits
C        32 bits   64 bits

BTW, you might want to check out the Numeric module in the NumPy
distribution at http://numpy.sourceforge.net.

Barry Drake

"Jason Orendorff" <jason at jorendorff.com> wrote in message news:<mailman.1012956250.23108.python-list at python.org>...
> Hugo Martires writes:
> > suposing that n=10 we get: myArray[0] = 0.600000023 ("something 
> > like this")
> > 
> > but if i do: myArray[i] + (1.0/n) it give's me 0.6
> > 
> > what's the problem of the array, and how can i solve this ?
> 
> It's only half of the normal precision.  From the documentation:
> 
>     The following type codes are defined:
>         Type code   C Type             Minimum size in bytes
>         'f'         floating point     4
>         'd'         floating point     8
> 
> Try using 'd' instead of 'f'.
> 
> ## Jason Orendorff    http://www.jorendorff.com/


From john_coppola_r_s at yahoo.com  Thu Feb 14 17:37:18 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Thu, 14 Feb 2002 14:37:18 -0800 (PST)
Subject: Pythonistas? <- Re: OT: Perl programmers?
In-Reply-To: <000401c1b57b$6b01a7c0$0de2533e@fonso>
Message-ID: <20020214223718.85913.qmail@web11806.mail.yahoo.com>

I like this message. this is not spam.
semantics is very important to python so in a strange
way i feel this message is related.

--- Afonso Fernandez Nogueira <afonso at writeme.com>
wrote:
> >Gustavo Cordova <gcordova at hebmex.com> graced us by
> uttering:
> >> And also "agua" which seems female, but it's a
> male subject: "el agua".
> 
> >IIRC, "agua" _is_ feminine, but takes the masculine
> article when
> >singular. Mi diccionario dice que "agua" es
> feminina, tambi?n.
> 
> >OTOH, la palabra "mar" es masculina _y_ feminina.
> Por ejemplo:
> >
> >   "el agua en la mar"     # Correct
> >    "las aguas del mar"     # Correct
> 
> Right. "Agua" is feminine, but takes the masculine
> article in singular just to avoid the cacophony of
> the two consecutive a's.
> 
> "La _ agua" is considered difficult to say and/or
> ugly in Spanish, as you can't merge the sounds
> together as if in Catalan "L'aigua".
> 
> But this doesn't apply to every word starting with
> "a", only in a few cases.
> 
>
almost-zero-python-knowledge-but-eager-to-share-email-address-with-spammers'ly
> yours
> fonso
> 
> 
> --
> http://mail.python.org/mailman/listinfo/python-list


__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com



From feldspar at ix.netcom.com  Mon Feb  4 14:28:52 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Mon, 04 Feb 2002 15:28:52 -0400
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com>
Message-ID: <3C5EE0F4.E33215CC@ix.netcom.com>

Marc Laymon wrote:
> 
> I am just learning python.  I have written a few simple scripts which
> run fine when invoked using "python myscript.py".  The Python manual
> says you can execute a Python script directly In Unix by making the
> first line "#!/usr/bin/python", then giving the command
> "./myscript.py".  However, when I try this, I get an error "Command
> not found.".  I get the same error under both Linux and Solaris.
> Using the same format for perl works fine.  Why doesn't the
> "#!/usr/bin/python" line work for Python ?  (perl and python are both
> in /usr/bin on my Linux system.)
> 
> Thanks.
> 
> Marc Laymon

Hmmmm.  I tried a few things, none of which reproduced your error:

* trying to run the script without execute permission
* running without any "if __name__ == '__main__':" code
* subtracting/adding spaces between the "#!" token and the path

So if nothing else, there's a few possibilities down...

	-jc


From narnett at mccmedia.com  Mon Feb 25 15:30:08 2002
From: narnett at mccmedia.com (Nick Arnett)
Date: Mon, 25 Feb 2002 12:30:08 -0800
Subject: All-on, all-off button for checkbuttons?
In-Reply-To: <3C7A9975.8D7299BA@earthlink.net>
Message-ID: <NDBBLAEMOMEHKALAMKKJIEBDJHAA.narnett@mccmedia.com>

Thanks -- I suspected there was something along those lines.  I think it
should work.  I'm using Tkinter, which I should have mentioned.

...

> Depending on the GUI you're using, there should probably
> be a way to find all objects on a form/frame, and loop over
> them. That would be another way to do it.

That's what I was after -- how to loop through them and do the comparison
operation to see if the object is a Checkbutton.

Nick




From dsavitsk at e-coli.net  Mon Feb 25 20:57:55 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 26 Feb 2002 01:57:55 GMT
Subject: call vb dll (will pop up vb form) form python
References: <u7loa4o4kt5t48@corp.supernews.com>
Message-ID: <DYBe8.9822$pC1.1721017@newssrv26.news.prodigy.com>

can you recompile the dll as an activex exe? i think that will work.

-d

"chongqing xiao" <cqxiao at charter.net> wrote in message
news:u7loa4o4kt5t48 at corp.supernews.com...
> Hi,
>
> I have a visual basic component (dll) which will pop up a form in run
time).
> When I try to call this dll in python (using activie python on win2000),
it
> give the
> following error message like this:
> Unable to show modal form in this enviroment
>
> Any ideas or suggestions are welcome.
>
> thanks
> chong
>
>




From loewis at informatik.hu-berlin.de  Mon Feb 25 08:10:45 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 14:10:45 +0100
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt>
Message-ID: <j4bsedlnai.fsf@informatik.hu-berlin.de>

"Hugo Martires" <hmartires at cultalg.pt> writes:

> How can i count the number of bytes of:
> --> List []
> --> array('d',[])
> 
> There is any function that return's the number of bytes ?

For the list, no. For the array: len(arr.tostring()) will give a rough
estimate. The actual memory consumption is higher, due to Python and
malloc overheads.

Regards,
Martin



From owen at astrono.junkwashington.emu  Fri Feb  1 19:29:01 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Fri, 01 Feb 2002 16:29:01 -0800
Subject: question about subclassing dict
Message-ID: <a3fbse$gc4$1@nntp6.u.washington.edu>

I'm writing a subclass of dict (OrderedDict, keeps items in the order 
entered; you can get it from http://www.astro.washington.edu/owen/ if 
interested.).

I have a question about the iterator methods a mapping type is supposed 
to define.

I started by defining __iter__. I was then pleased to find that the 
methods iterkeys, itervalues and iteritems all worked just fine.

Clearly dict is basing those three iterators on the iterator returned by 
__iter__. My question is, will it continue to do so? Or should I play it 
safe and define all of these methods myself, in case the implementation 
of dict changes?

(Also, is subclassing of built in types documented in the manuals? I 
didn't find much, just a brief references to subclassing the class 
"dictionary", which didn't work very well, since the class is actually 
"dict".)

-- Russell


From phr-n2002a at nightsong.com  Wed Feb  6 17:40:48 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 06 Feb 2002 14:40:48 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <mailman.1013003651.8981.python-list@python.org> <a3re5o$1ad8nu$1@ID-98166.news.dfncis.de>
Message-ID: <7xg04ensi7.fsf@ruckus.brouhaha.com>

"MDK" <mdk at mdk.com> writes:
> This is for a project where Python is embeded in a piece of hardware and in
> order to move the data around it must be "flattened".  After I get the list
> of characters I am going to convert it to ASCII.

If you're only going to get flattened data from trusted sources, use
the pickle module.  If you're getting the data from untrusted sources
(such as over the internet), make sure to read the security notes in
the new (2.2) version of the pickle module documentation.



From threeseas at earthlink.net  Sun Feb  3 19:00:56 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:00:56 GMT
Subject: Autocoding project proposal.
References: <mailman.1012365168.19059.python-list@python.org>
Message-ID: <895.799T1416T11704514threeseas@earthlink.net>

On 29-Jan-02 23:31:45 Delaney, Timothy <tdelaney at avaya.com> wrote:

>OK - I give up too. I've done my best to understand what you're going on
>about, but so far as I can tell, everything you say means nothing. Seriously
>- I don't think I've read a single thing from you that, once you remove all
>the double-talk and nonsense, actually means anything.

>I have just added you to my killfile. Any post from you will now be
>automatically deleted.

>Tim Delaney

Some people don't want to see what will make them useless.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From geoff at elj.com  Sat Feb  2 19:48:41 2002
From: geoff at elj.com (Geoff Eldridge)
Date: 2 Feb 2002 16:48:41 -0800
Subject: elj library 0.1 beta release ..
Message-ID: <8cd6cc47.0202021648.14a4a36b@posting.google.com>

Just a very quick note to inform that a set of installations have
been made available for the elj libraries. This marks the first beta
release of the elj library:

   http://elj.sourceforge.net/

This is just an advance notice of the release and more details will
follow. For the adventurous, you can find details here:

http://sourceforge.net/project/showfiles.php?group_id=28837

--8<--
0.1 - 2002-01-27 16:00 

elj-01-ccl-plus-db.exe     2533013 0 i386 Other Source File 
elj-01-ccl-plus-tools.exe  2099658 0 i386 Other Source File 
elj-01-ccl.exe             2038650 0 i386 Other Source File 
elj-01-full.exe            2594048 0 i386 Other Source File 
elj-01-win32-only.exe      1332700 0 i386 Other Source File 
-->8--

NB:

* ccl is our higher level GUI library:

  * http://elj.sourceforge.net/projects/gui/ccl/doc/tutorial/

* db provides interfaces to databases: FireBird, mysql, Postgres
  SleepyCat (ie Berkeley DB) ...

  * http://elj.sourceforge.net/projects/db/firebird/
  * http://elj.sourceforge.net/projects/db/mysql/
  * http://elj.sourceforge.net/projects/db/postgres/
  * http://elj.sourceforge.net/projects/db/sleepycat/

* tools includes, an editor, image editor, menu editor:

  * http://elj.sourceforge.net/projects/tools/eifed/
  * http://elj.sourceforge.net/projects/tools/imed/
  * http://elj.sourceforge.net/projects/tools/mened/

Some documentation effort needs to be applied to the web pages
starting from here:

   http://elj.sourceforge.net/

I am spending a lot of time on the firebird database interface, a
particular impressive server based database derived from the Borland
Interbase source release:

   http://elj.sourceforge.net/projects/db/firebird/
   http://firebird.sourceforge.net/

Hopefully, elj will soon have some further documentation and notes on the
beta release.

In the meantime, try a download and compile some of the examples or tools
using SmallEiffel (-0.74 beta 15) and lcc-win32.

Note work has started on developing a Linux-based implementation of ccl:

   http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/elj/elj/lib/gtk/eiffel/

If you have not had a look at elj, maybe now is a good time.

Regards .. Geoff

-- geoff at elj.com
-- http://elj.sourceforge.net/


From gcordova at hebmex.com  Wed Feb 27 10:30:13 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 27 Feb 2002 09:30:13 -0600
Subject: PIL vs eps
Message-ID: <19650FF08A99D511B19E00B0D0F06323060B10@SRM201002>

> 
> I think this is a pure ghostscript problem, but PIL fails to 
> convert eps to other formats and in fact it seems that gs
> won't create any output without a showpage at least somewhere.
> 

Isn't "showpage" the postscript command to render
all the code that's been processed?

-gustavo



From jkraska at san.rr.com  Sun Feb 24 23:12:04 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 25 Feb 2002 04:12:04 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu>
Message-ID: <0qej7u8uoepjg377elqo1msgklfc4t5eed@4ax.com>

>[x.f() for x in items]

Yeah, whoever it was that decided this was the syntax
for a list comprehension was just a plain genius. It's
one of those linguistic thingies where you take one
look at it, and you say, "of course".

Wait a second. What's this theory about Guido and his
time machine? I guess you're the genius, and Guido's
only cheating. :-)

C//



From jeff at ccvcorp.com  Thu Feb  7 18:08:44 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 07 Feb 2002 15:08:44 -0800
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <f72dac0d.0202070547.8cb548a@posting.google.com> <HGx88.95599$KM2.3475019@atlpnn01.usenetserver.com> <zBC88.30105$X64.10397017@news1.rdc2.pa.home.com>
Message-ID: <3C6308FB.1CBB1E48@ccvcorp.com>

Terry Reedy wrote:

> I'd like to see the following someday (from his slide 4)
>
> import http://www.w3.org/2000/10/swap/llyn.py
>
> coupled with a public library of modules for anyone to import latest
> version of.

Hm, I dunno.  While it'd be nice to have easy access to the most recent
version of a module, embedding a URI directly into a program just
screams versioning nightmare to me.  I can just imagine some project
that works fine for years, until someone makes an incompatible update to
a module that's web-imported from another module that's web-imported
from....

For this to work, I think you'd need to have some canonical way of
handling versions, and enforcing compatibility.  Of course, then you're
still left with the problems of running a program when some webserver is
offline, or traffic is extremely heavy, or...

CPAN-like functionality of automatically making a local copy would seem
like a much better idea to me.  (Still requires the versioning tools,
though.)

Jeff Shannon
Technician/Programmer
Credit International




From ol1 at v10a.com  Tue Feb 19 11:35:33 2002
From: ol1 at v10a.com (O Polite)
Date: 19 Feb 2002 17:35:33 +0100
Subject: Recursively traverse linked list -- help!
In-Reply-To: <7x7kp9blab.fsf@ruckus.brouhaha.com>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
	<mailman.1014106650.11778.python-list@python.org>
	<7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek>
	<mailman.1014126336.32560.python-list@python.org> 
	<7x7kp9blab.fsf@ruckus.brouhaha.com>
Message-ID: <1014136534.2891.169.camel@matijek>

On Tue, 2002-02-19 at 15:26, Paul Rubin wrote:

> By "middle" I didn't mean the geometric middle, but the usual case of
> wanting to insert a new node after an existing node that you already
> have a handle on. 
Yeah I got that.

> 
> Yes, the lists have to be pretty large for implementing linked lists
> in Python to beat Python's built-in vectors.  But that's only because
> interpreting Python code is so abysmally slow, which is a Python
> implementation artifact.  If and when faster Python implementations
> (i.e. with real compilers) become available, the crossover point will
> probably become a lot lower.  
Sounds wonderful. Who's working on it? When is it due?

> Even as it is, though, your measurement
> only went up to 20,000 nodes; I'd be interested in the timing at a
> million nodes.

Fair enough! Actually if I pit the LinkedList against the builtin list
in the case you're talking about (where you never have to do any work to
find the insertpoint) the LinkedList will come out as a winner already
at 5000 - 10000 inserts.


-----
program


import time, random, Numeric
class LinkedList:

    def __init__(self):
        self.start = None

    def append(self, value = None):
        new = ListElement(value)
        if self.start:
            new.next = self.start
        self.start = new


    
    def insert(self, position, value):
        new = ListElement(value)
        
        if self.start:
            i = 0
            current = self.start
            while current and position > i:
                i += 1
                current = current.next
            new.next = current.next
            current.next = new
        else:
            self.start = new

        
    def pop(self):
        result = None
        if self.start:
            result = self.start.data
            self.start = self.start.next
        return result
    
    def __str__(self):
        result = ""
        current = self.start
        while current:
            result += "%s " % (str(current))
            current = current.next
        return result

class ListElement:
    def __init__(self, value = None):
        self.data = value
        self.next = None

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


def do_many_inserts(list, count1, count2):
    length = 0
    for j in range(count1):
        t1 = time.time()
        for i in range(count2):
            list.insert(length/2, i)
            length += 1
        print length, time.time() - t1
    
def do_many_pushes(list, count1, count2):
    length = 0
    for j in range(count1):
        t1 = time.time()
        for i in range(count2):
            list.append(i)
            length += 1
        print length, time.time() - t1

    
if __name__ == '__main__':

    sets = 10
    reps = 20000

    print "inserts at start of LinkedList"
    do_many_pushes(LinkedList(), sets, reps)
    print
    
##    print "inserts at start of builtin list"
##    do_many_pushes([], 10, 20000)
##    print
    
##    print "inserts in middle of LinkedList"
##    do_many_inserts(LinkedList(), 10,200)
##    print

    print "inserts in middle of builtin list"

    do_many_inserts([], sets,reps)

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

output
inserts at start of LinkedList
20000 0.457922101021
40000 0.546599984169
60000 0.834394097328
80000 0.565314054489
100000 1.09416890144
120000 0.563593029976
140000 1.29079794884
160000 0.564273953438
180000 0.556931972504
200000 1.63955891132

inserts in middle of builtin list
20000 0.787240982056
40000 1.71993708611
60000 4.12805104256
80000 6.63266301155
100000 9.03667891026
120000 12.0142199993
140000 26.1722120047
160000 47.1146579981
180000 56.5776759386
200000 68.3339459896






From jkraska at san.rr.com  Fri Feb  8 02:09:51 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 07:09:51 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <jb246uci04pj6q2l0aqu18ch17kmv8ujof@4ax.com> <Xns91AE5FFAFDC37ahmebahhotmailcom@198.99.146.10>
Message-ID: <jku66u0er861d0127gtrbh7agjncm6qqsn@4ax.com>

>Well, I'm sure somebody would tell him he's gone over the edge.

Well the word "delusional" is often both used and understood
to be insulting. But consider it's non-insulting aspects: is
one connected with reality or not? It happens. Most every
one has been delusional at one point in their lives; even in
the strict clinical sense. Consider any crush you might have
had; most crushes are rife with delusion.

C//



From martin at v.loewis.de  Fri Feb 15 03:16:49 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 15 Feb 2002 09:16:49 +0100
Subject: Tkinter and Greek Symbols
References: <3daa1aa8.0202141718.3500c462@posting.google.com>
Message-ID: <m3heoj5fda.fsf@mira.informatik.hu-berlin.de>

jmsun at bioeng.ucsd.edu (Jeffrey) writes:

>   You can probably guess what I want to know from the title.  But hey
> I'll explain anyways :)  I want to be able to display Greek Symbols
> like alpha and theta.

import Tkinter

l = Tkinter.Label(text = u"\N{GREEK SMALL LETTER ALPHA}")
l.pack()
l.tk.mainloop()

If you happen to know that GREEK SMALL LETTER ALPHA is \u0391, you can
use that as well. If you have an editor that supports utf-8, you can
use unicode("<utf-8 encoding of the greek text>", "utf-8") instead
of the Unicode literal.

HTH,
Martin



From mhammond at skippinet.com.au  Tue Feb 12 04:39:43 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 12 Feb 2002 09:39:43 GMT
Subject: PYTHONPATH on the Windows Registry?
References: <3c673b3e$1_2@nopics.sjc>
Message-ID: <3C68E321.8080507@skippinet.com.au>

Ixokai wrote:
>     Question-- if Python on Windows (w/o win32all) is running on a machine,
> without any PYTHONPATH/etc set in the registry, but with an environment
> variable set, will it set its 'current' python path into the registry? Or
> set anything at all? I want a zero-impact run of Python on the system. :)
> 

PYTHONPATH is in addition to any paths in the registry.  Thus you can 
list all paths in PYTHONPATH and have no registry at all.

Mark.



From phr-n2002a at nightsong.com  Sun Feb  3 20:29:42 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 03 Feb 2002 17:29:42 -0800
Subject: pickle security
References: <mailman.1012591743.10841.python-list@python.org> <7xit9g51lz.fsf@ruckus.brouhaha.com> <j4g04ihc7k.fsf@informatik.hu-berlin.de>
Message-ID: <7xsn8im3uh.fsf@ruckus.brouhaha.com>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:
> Apart from feeling uncomfortable, can you point to a real
> counter-example? The code to deal with quoted strings is designed to
> check that there it is nothing but a quoted string.

The code to deal with quote strings calls eval, and without actually
examining the eval implementation (big, complicated) there's no way to
confirm that it can't somehow run arbitrary code found in the string.

> So, apart from setting find_global, I don't think there are any
> further issues to be aware of.

One remaining issue is that the string is handed off to a very
complicated and flexible piece of code (the evaluator) that an
attacker might exploit somehow.  We shouldn't assume the absence of
exploits just because the exploits aren't immediately there to be
seen.  Microsoft does that and that's why we keep hearing about new
exploits being discovered in Microsoft products.  The code has to be
checked thoroughly, which is easier to do when the code is simple.

There's a big difference between ordinary good programming
(programming to avoid leaving bugs that can make the program fail by
accident or error) and writing secure programs (programming to thwart
malicious, intelligent, persistent attackers deliberately trying to
make the program fail and are willing to spend more effort analyzing
and attacking the program than the programmer spent writing it).

Writing secure programs (programs intended to be exploit-free even
when operated by hostile entities) is extremely difficult and has to
be done very carefully, thinking about security every step of the way.
Most programs including Python weren't written that way.  Therefore,
any interface between Python and potential attackers--that includes
stuff like the socket libraries and also turns out to include the
pickle module--should very thoroughly isolate the main Python
interpreter from potentially hostile data til the data has been
examined.  If you use Perl's taint-checking feature for a while you'll
see how easy it is to overlook stuff--it often catches things that I
miss.  I wish Python had a similar feature.  In any case, I wish the
pickle module didn't call eval for something as simple as handling
quoted strings.

You might read the book "Security Engineering", by Ross Anderson, to
get an idea of what you're up against when trying to program a secure
system.


From jbauer at rubic.com  Thu Feb 28 14:13:02 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 28 Feb 2002 13:13:02 -0600
Subject: struct_time type testing 2.1-2.2
Message-ID: <3C7E813E.E31FDD8B@rubic.com>

Hi Robin.

Actually someone at ReportLabs put _SeqTypes in my
NormalDate class <0.3 wink>, not that I have any
objection.

My latest patch submitted to ReportLabs used something
like this:

    _TimeType = type(time.localtime(time.time()))

    ...

    if tn in (TupleType, ListType) or tn is _TimeType:
        <snip>

The _TimeType will work whether time is a tuple (<= Python2.1)
or a struct_time (Python.2.2) type.

Martin von Loewis also posted some good suggestions, although
his idea to use the code below won't work (he wouldn't have
known, without looking at ReportLab's source) because the
input type could also be a string, int, or an instance of
NormalDate.

>  try:
>    self.normalDate = int("%04d%02d%02d" % normalDate[:3])
>  except TypeError:
>    # else code

I think the problem is type checking is inherently bad
practice which I violated for some user convenience
in NormalDate.

inelegant-ly y'rs,

Jeff Bauer
Rubicon Research

Robin writes:
> Jeff Bauer kindly tells me that Python 2.2 has a different return type
> for time.localtime(...). 
> 
> In his module he had a test on an argument type of the form
> 
>     _SeqTypes = (ListType,TupleType)
> 
> 
>     tn = type(arg)
>     if tn in _SeqTypes:
>       self.normalDate = int("%04d%02d%02d" % normalDate[:3])
> 
> and in 2.2 we seem to need 
> 
>     if tn in _SeqTypes or tn is time.struct_time:
> 
> is there a cleverer way than
> 
>     if tn in _SeqTypes or tn is getatr(time,'struct_time',None):
> 
> to get a cross version test?



From phr-n2002a at nightsong.com  Tue Feb  5 21:11:07 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 18:11:07 -0800
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c6011e3$1_5@news.bluewin.ch> <tor06u0j1ce99ibgd98uflckcms89apfci@4ax.com> <3c607be8$1_1@news.bluewin.ch> <9k116u4r58thf9lp69cso89ubtvvhhsm7f@4ax.com>
Message-ID: <7xr8nze4w4.fsf@ruckus.brouhaha.com>

Courageous <jkraska at san.rr.com> writes:
> At the moment, no. I have my own lexer, parser, validator
> and emitter. The first two are fairly close to done, missing
> a few notables, like comprehensions. The last are very much
> a work in progress. Lotto work there; still in design on those.

Is this written in Python?


From christophe.delord at free.fr  Fri Feb 22 15:17:27 2002
From: christophe.delord at free.fr (Christophe Delord)
Date: Fri, 22 Feb 2002 21:17:27 +0100
Subject: re. help needed :(
References: <a56838$qb8$1@newsg4.svr.pol.co.uk>
Message-ID: <3C76A757.7040103@free.fr>

Hi,

try this :


import re

x="Mary had a little lamb called lamb, but didn't like the lamb's mood"

print re.sub(r"lamb\b(?!')","pig",x)


\b is a word boundary and (?!...) matches if ... doesn't match what follows.



Christophe.


G. Willoughby wrote:

> Hi,
>     Im trying to replace the single word "lamb" and the word lamb followed
> by any punctuation in this string:
> 
> x="Mary had a little lamb called lamb, but didn't like the lamb's mood"
> 
> i want to replace all of the single words "lamb[punctuation]" but not the
> word "lamb's" to achieve this:
> 
> x="Mary had a little pig called pig, but didn't like the lamb's mood"
> 
> any help with be greatly appreiciated!
> 
> ta
> 
> G. Willoughby
> 
> 
> 


-- 
Christophe Delord
http://christophe.delord.free.fr/



From sdm7g at Virginia.EDU  Wed Feb 20 00:43:39 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Wed, 20 Feb 2002 00:43:39 -0500 (EST)
Subject: execfile question [+Q: why doesn't del dict imply dict.clear()
 ??]
In-Reply-To: <a4v4md$ecm$0@216.39.172.122>
Message-ID: <Pine.OSX.4.43.0202200007490.296-100000@localhost>


On 20 Feb 2002, Bengt Richter wrote:

> >> ##main.py######################
> >> execfile('executable.py', {})
> >> print 'executed'
> >> ########################
> >>

> I think he's asking why finishing execfile is not one of those
> 'other actions'.
>
> I think the contents of the anonymous dictionary he passed are not
> automatically deleted (why not?) when it goes away, so a reference
> in it is keeping the c instance from being collected.


I replied too hastily earlier and misread the example -- I didn't
see the single '{}' in 'execfile('executable.py', {})'.

I my example with two '{}'-s: 'execfile('executable.py', {}, {})'
it worked as expected.

If you try it with two named dicts, and do a sys.refcount() on both,
you'll see that the global dict has a couple of extra references after
the execfile call.

>>> G = {}; L = {}
>>> sys.getrefcount(G)
2
>>> sys.getrefcount(L)
2
>>> execfile( 'executable.py', G, L )
it's me
init
>>> sys.getrefcount(L)
2
>>> sys.getrefcount(G)
4

You'll find one of those references to G in:
>>> L['c'].__init__.im_func.func_globals

and the other in the classes __del__ method.


When there are two separate dicts for globals and locals, the local
dict can be deleted and that releases the references to the globals,
so it can be deleted. When they are the same dict, then you have a
self-reference loop where items in the dict hold references to the
dict.

Doing a 'D.clear()' instead of 'del D' deletes the instances first.

I guess this really isn't a bug: 'del D' just decrements D's refcount,
so even if the dict destructor called the clear function, it's still
not going to get called until there are no more refs, and until you
call clear, those extra two refs will remain.

Maybe we should add a note to the documentation warning that that
usage can cause a memory leak.

-- Steve Majewski





From ressler at cheetah.jpl.nasa.gov  Fri Feb  8 14:35:29 2002
From: ressler at cheetah.jpl.nasa.gov (Michael Ressler)
Date: 8 Feb 2002 19:35:29 GMT
Subject: Is pippy alive?
Message-ID: <slrna68a44.qph.ressler@cheetah.jpl.nasa.gov>

Is any active development still going on with pippy? I've found it to be
quite valuable as a learning tool for python, but now that I'm actually
ready to "do something" with it, the lack of a finished palm interface (and
floating point! though I can work around that) really hurts. I've checked
the CVS and various python archives, but little has been said about it
recently ...

Mike

--
Mike Ressler
mike.ressler at alum.mit.edu







From fperez528 at yahoo.com  Wed Feb 27 08:59:14 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Wed, 27 Feb 2002 13:59:14 +0000
Subject: cross platform (linux/windows) problem
References: <f5c74c0f4b.miah@picsel.com>
Message-ID: <a5jheh$ceb$1@peabody.colorado.edu>

See if the following code helps. It does all the trapping of stdout/stderr in 
a class, and you can then decide what to do with it. It doesn't address the 
calling part of your problem, only the output part.

cheers,

f.

File: OutputTrap.py

"""Class to trap stdout and stderr and log them separately."""

#*****************************************************************************
#       Copyright (C) 2001 Fernando P?rez. 
#
#  Distributed under the terms of the GNU Lesser General Public License (LGPL)
#
#  The full text of the LGPL is available at:
#
#                  http://www.gnu.org/copyleft/lesser.html
#*****************************************************************************

__author__  = 'Fernando P?rez. '
__version__ = '0.1.0'
__license__ = 'LGPL'
__date__   = 'Tue Dec 11 00:27:58 MST 2001'

import exceptions,sys
from StringIO import StringIO

class OutputTrapError(exceptions.Exception):
    """Exception for OutputTrap class."""

    def __init__(self,args=None):
        exceptions.Exception.__init__(self)
        self.args = args

class OutputTrap:

    """Class to trap standard output and standard error. They get logged in
    StringIO objects which are available as <instance>.out and
    <instance>.err. The class also offers summary methods which format this
    data a bit.

    A word of caution: because it blocks messages, using this class can make
    debugging very tricky. If you are having bizarre problems silently, try
    turning your output traps off for a while. You can call the constructor
    with the parameter debug=1 for these cases. This turns actual trapping
    off, but you can keep the rest of your code unchanged (this has already
    been a life saver).

    Example:

    # config: trapper with a line of dots as log separator (final '\\n' 
needed)
    config = OutputTrap('Config','Out ','Err ','.'*80+'\\n')

    # start trapping output
    config.trap_all()

    # now all output is logged ...
    # do stuff...

    # output back to normal:
    config.release_all()

    # print all that got logged:
    print config.summary()

    # print individual raw data:
    print config.out.getvalue()
    print config.err.getvalue()
    """

    def __init__(self,name='Generic Output Trap',
                 out_head='Standard Output. ',err_head='Standard Error. ',
                 sum_sep='\n',debug=0,trap_out=0,trap_err=0,
                 quiet_out=0,quiet_err=0):
        self.name = name
        self.out_head = out_head
        self.err_head = err_head
        self.sum_sep = sum_sep
        self.out = StringIO()
        self.err = StringIO()
        self.out_save = None
        self.err_save = None
        self.debug = debug
        self.quiet_out = quiet_out
        self.quiet_err = quiet_err
        if trap_out:
            self.trap_out()
        if trap_err:
            self.trap_err()

    def trap_out(self):
        """Trap and log stdout."""
        if sys.stdout is self.out:
            raise OutputTrapError,'You are already trapping stdout.'
        if not self.debug:
            self._out_save = sys.stdout
            sys.stdout = self.out

    def release_out(self):
        """Release stdout."""
        if not self.debug:
            if not sys.stdout is self.out:
                raise OutputTrapError,'You are not trapping stdout.'
            sys.stdout = self._out_save
            self.out_save = None

    def summary_out(self):
        """Return as a string the log from stdout."""
        out = self.out.getvalue()
        if out:
            if self.quiet_out:
                return out
            else:
                return self.out_head + 'Log by '+ self.name + ':\n' + out
        else:
            return ''

    def flush_out(self):
        """Flush the stdout log. All data held in the log is lost."""

        self.out.close()
        self.out = StringIO()

    def trap_err(self):
        """Trap and log stderr."""
        if sys.stderr is self.err:
            raise OutputTrapError,'You are already trapping stderr.'
        if not self.debug:
            self._err_save = sys.stderr
            sys.stderr = self.err

    def release_err(self):
        """Release stderr."""
        if not self.debug:
            if not sys.stderr is self.err:
                raise OutputTrapError,'You are not trapping stderr.'
            sys.stderr = self._err_save
            self.err_save = None

    def summary_err(self):
        """Return as a string the log from stderr."""
        err = self.err.getvalue()
        if err:
            if self.quiet_err:
                return err
            else:
                return self.err_head + 'Log by '+ self.name + ':\n' + err
        else:
            return ''

    def flush_err(self):
        """Flush the stdout log. All data held in the log is lost."""

        self.err.close()
        self.err = StringIO()

    def trap_all(self):
        """Trap and log both stdout and stderr.

        Cacthes and discards OutputTrapError exceptions raised."""
        try:
            self.trap_out()
        except OutputTrapError:
            pass
        try:
            self.trap_err()
        except OutputTrapError:
            pass

    def release_all(self):
        """Release both stdout and stderr.

        Cacthes and discards OutputTrapError exceptions raised."""
        try:
            self.release_out()
        except OutputTrapError:
            pass
        try:
            self.release_err()
        except OutputTrapError:
            pass
        
    def summary_all(self):
        """Return as a string the log from stdout and stderr, prepending a 
separator
        to each (defined in __init__ as sum_sep)."""
        sum = ''
        sout = self.summary_out()
        if sout:
            sum += self.sum_sep + sout
        serr = self.summary_err()
        if serr:
            sum += '\n'+self.sum_sep + serr
        return sum

    def flush_all(self):
        """Flush stdout and stderr"""
        self.flush_out()
        self.flush_err()

    # a few shorthands
    trap = trap_all
    release = release_all
    summary = summary_all
    flush = flush_all
# end OutputTrap


#****************************************************************************
# Module testing. Incomplete, I'm lazy...

def _test_all():

    """Module testing functions, activated when the module is called as a
    script (not imported)."""

    # Put tests for this module in here.
    # Define them as nested functions so they don't clobber the
    # pydoc-generated docs

    def _test_():
        name = ''
        print '#'*50+'\nRunning test for ' + name
        # ...
        print 'Finished test for '+ name +'\n'+'#'*50

    def _test_OutputTrap():
        trap = OutputTrap(name = 'Test Trap', sum_sep = '.'*50+'\n',
                          out_head = 'SOut. ', err_head = 'SErr. ')

        name = 'OutputTrap class'
        print '#'*50+'\nRunning test for ' + name
        print 'Trapping out'
        trap.trap_out()
        print >>sys.stdout, '>>stdout. stdout is trapped.'
        print >>sys.stderr, '>>stderr. stdout is trapped.'
        trap.release_out()
        print trap.summary_out()

        print 'Trapping err'
        trap.trap_err()
        print >>sys.stdout, '>>stdout. stderr is trapped.'
        print >>sys.stderr, '>>stderr. stderr is trapped.'
        trap.release_err()
        print trap.summary_err()

        print 'Trapping all (no flushing)'
        trap.trap_all()
        print >>sys.stdout, '>>stdout. stdout/err is trapped.'
        print >>sys.stderr, '>>stderr. stdout/err is trapped.'
        trap.release_all()
        print trap.summary_all()

        print 'Trapping all (flushing first)'
        trap.flush()
        trap.trap_all()
        print >>sys.stdout, '>>stdout. stdout/err is trapped.'
        print >>sys.stderr, '>>stderr. stdout/err is trapped.'
        trap.release_all()
        print trap.summary_all()
        print 'Finished test for '+ name +'\n'+'#'*50

    # call the actual tests here:
    _test_OutputTrap()


if __name__=="__main__":
    # _test_all() # XXX BROKEN.
    pass

#************************ end of file <OutputTrap.py> ************************



From phr-n2002a at nightsong.com  Sat Feb 16 20:23:11 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 16 Feb 2002 17:23:11 -0800
Subject: Guido wins FSF Award
References: <mailman.1013906967.6705.python-list@python.org>
Message-ID: <7xy9hsvr40.fsf@ruckus.brouhaha.com>

"Tim Peters" <tim.one at home.com> writes:
> [Paul Rubin]
> > Congratulations to Guido for winning the fourth Free Software Foundation
> > Award, for inventing and implementing Python as Free Software:
> >
> >   http://www.fsf.org/press/2002-02-16-FSF-Award.html
> 
> Hooray for Guido!  Shows how much the world cares about secure pickles
> <wink>.

Yes!  The award, so quickly following the Python 2.2 docs addressing
pickle security, is wonderful recognition for the Python finally
treating this vitally important issue ;-).

> > And to think I learned of this from a Timothy Rue post.  Sigh.
> 
> Not at all.  Guido's acceptance speech was a foundational masterpiece,
> revealing the Nine Pythonic Actions underlying all that we do.

Well, maybe the speech will go up on the web like last year's.
Meanwhile the award also made the front page of Slashdot, as I found a
few minutes after posting.  Yes, I checked c.l.py before reading
Slashdot when I woke up this morning.  I'm not sure what that says
about either news source ;-).


From cliechti at gmx.net  Tue Feb 12 15:58:21 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 12 Feb 2002 21:58:21 +0100
Subject: cgi and mime type
References: <ed6d1c1a.0202121225.295ca950@posting.google.com>
Message-ID: <Xns91B3E00B94C16cliechtigmxnet@62.2.16.82>

aseem at neurobs.com (A M) wrote in news:ed6d1c1a.0202121225.295ca950
@posting.google.com:

> fp = open('fp.exe')
> print fp.read();

use open(name, 'rb') here. binary mode is important when the cgi runs 
on win32. not sure if this fixes your other problem though...

chris

-- 
Chris <cliechti at gmx.net>



From bosahv at netscapenospam.net  Wed Feb 13 04:39:23 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Wed, 13 Feb 2002 09:39:23 GMT
Subject: 19 y/o blonde want to know something about Python working with IPX protocol
References: <mailman.1013529085.695.python-list@python.org>
Message-ID: <fvqa8.372999$pa1.87468846@news3.rdc1.on.home.com>

Reply, sorry cant really help this time either.

Just a thought though, that if you're doing this for a Novell IPX setup you
can probably update it to use TCP/IP.

My only thought regarding the blond was that you were grouping your post
with some pretty foul spam.

bo


"Mariano Corsunsky" <core.lists.python at core-sdi.com> wrote in message
news:mailman.1013529085.695.python-list at python.org...
DISCLAIMER:

Hi!, I tried to get this same question answered before, but nobody did. So
I'm using the "19 y/o blonde strategy"


----

Hi boys!

I want some examples of Python dealing with the IPX protocol.
I do only need a simple, so so simple example which sends and receive
strings between 2 computers using IPX.

Could you help me?

I need someting like ........

(PC number 1)
SEND "I'm gonna aply some sun-tan oil over my californian body"

(PC number 2)
[Received-->] "I'm gonna aply some sun-tan oil over my californian body"

SEND "...but you are only pretending to be a 10aged girl just to get this
question answered!, STOP that, or everybody will start thinking you're gay!"

(PC number 1)
[Received-->]
"...but you are only pretending to be a 10aged girl just to get this
question answered!, STOP that, or everybody will start thinking you're gay!"



-----

That's it.
Thank you all!




From eric.brunel at pragmadev.com  Thu Feb 28 05:13:28 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Thu, 28 Feb 2002 10:13:28 +0000
Subject: problems with photoimage and tkinter
References: <Y%hf8.149154$dh.38770971@typhoon.nyroc.rr.com>
Message-ID: <a5ks4b$61a$1@norfair.nerim.net>

Hi Seth,

Seth Jones wrote:
> Hi, I'm looking for some help using images in tkinter. I've spent hours
> with it, and I can't
> seem to get anything working. Well, I have something. The following code
> is from the Python Imaging Library 'painter.py'.. This is the minimum
> necessary to display an image on the canvas.
> The line
> 
>                 #self.tile[(x,y)] = box, tile
> 
> must be uncommented in order for this code to work, and for the life of
> me, I can't imagine why. 
[snip]

You ran into a bug in Tkinter: when you create an image (either via 
BitmapImage or PhotoImage) and you attach it to a widget (Button, or 
Canvas, or whatever), for some reason, the reference counter for the 
BitmapImage or PhotoImage instance is *not* increased. So, using a simple 
example, here is what happens:

def f(canvas):
  # Image creation and recording in "img" -> reference counter is 1
  img = PhotoImage(...)
  # Creation of image in canvas -> reference counter is *not* increased, so
  # it's still 1
  canvas.create_image(..., image=img ...)

# End of function f => img goes out of scope => reference counter for image
# is decremented => falls to 0 => image is destroyed!!!
# When canvas will be updated on screen, the image won't show, since it has
# been destroyed...

So here is probably one of the purposes of the line you spotted: keep 
another reference on the image so that it won't be destroyed once the 
method is finished... If I wanted the above code to work, I would create a 
global variable for example, just to put the image in it.

BTW, for the general audience, this bug exists at least since Python 1.5. I 
found it very confusing the first time I encountered it, and I do 
understand why other people find it quite confusing too. Even if I didn't 
personally see it, I suppose this is an issue that appeared more than once 
in c.l.py. Why is this bug still not corrected? Anybody maintaining Tkinter 
out there?

HTH, & TIA
 - eric -



From gerson.kurz at t-online.de  Sun Feb 17 12:27:12 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sun, 17 Feb 2002 17:27:12 GMT
Subject: Python.NET
References: <mailman.1013959887.19560.python-list@python.org>
Message-ID: <3c6fe5cd.20383078@news.t-online.de>

On Sun, 17 Feb 2002 12:29:25 -0300, "Raphael Ribeiro"
<raphaelr at softhome.net> wrote:

>I want to know about you guys what do you think of Python.NET . ( I'm =
>thinking on learning python with .net , but i don't know if it is a good =
>option  , so i want to hear comments!)

Microsofts' Advertisment of Python for .NET is kind of misleading:

- "Visual Python" is just "plain old python" that uses the "Visual
Studio .NET" IDE. It does NOT produce code for .NET.

What they advertise on their page is this (excellent, but expensive)
IDE.

- Python.NET is an experimental compiler, and seems to be a
sort-of-abandoned project, see

http://www.activestate.com/Initiatives/NET/Research.html

The main restrictions are listed in the whitepaper that you'll find on
this page. 

Quote: """
- There is no support for some of the features of .NET that other
frameworks will require, such as custom attributes, PInvoke or
ASP.NET.  This means that Python for .NET users are not currently able
to write classes that interact with tools requiring those features.
Related topics are the mismatch between the class/instance semantics,
module/package semantics and exception systems.

- The speed of the current system is so low as to render the current
implementation useless for anything beyond demonstration purposes.
This speed problem applies to both the compiler itself, and the code
generated by the compiler.  Given that part of the appeal of Python
programming is a quick edit-compile-run cycle, the speed issues
severely limit the utility of Python on this platform.  Some of the
blame for this slow performance lies in the domain of .NET  internals
and Reflection::Emit, but some of it is due to the simple
implementation of the Python for .NET compiler.

- There is no support for some Python features that some programs will
require.  Most of these are fairly obscure, but are a limitation.
Examples include:
- String formatting
- Core language features, such as long integers, complex numbers,
builtin object methods and so forth.
- The standard Python library
"""

Basically, while M$ hypes the Multilanguage-thing for .NET, right now
its best to EITHER concentrate on Python, OR on .NET. If I were you,
I'd choose Python ;)


From lac at strakt.com  Wed Feb 27 09:27:36 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 27 Feb 2002 15:27:36 +0100
Subject: Are the PEPs mirrored anywhere in Europe?
Message-ID: <200202271427.g1RERarB026499@ratthing-b246.strakt.com>

I am writing a document describing some of them and the link to
sourceforge is a turtle.

Laura Creighton
(in Sweden if it matters)



From xx758 at cam.ac.uk  Tue Feb  5 07:49:56 2002
From: xx758 at cam.ac.uk (Xiao-Qin Xia)
Date: Tue, 05 Feb 2002 12:49:56 +0000
Subject: Inherit Class Dynamicly
Message-ID: <a3ok79$bvr$1@pegasus.csx.cam.ac.uk>

Hi, everybody,

from Tkinter import *
class NewWidget:
        ...
        def __init__(self, widget_name = "Button", **dict):
                # What is needed here to make the new born object(self) 
                # behavior like both the class specified by widget_name
                # and NewWidget?

Thanks from

Xiao-Qin



From gcordova at hebmex.com  Fri Feb 15 09:57:08 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 08:57:08 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A7F@SRM201002>

> 
> Que me des unos ejemplos?
> 
> Gracias,
> Tim Hammerquist
>

Hmmm... (or "Mmmm..." in Spanish :-)

I personally don't do "Correct Spanish", but for
a very delightful bought with it, read "Las Aventuras
del Ingenioso Hidalgo Don Quixote de La Mancha",
by Miguel de Cervantes Saavedra.

I don't think there's a more correct Spanish, without
breaking out the rules books by La Real Academia de
la Lengua Espa?ola, but those are a bit more boring
than Don Quixote.

TGIF!!!

-gustavo

pd: So delightful to find people willing to talk
    spanish and english! I've been previously flamed
    for daring to use another language besides english.



From gward at python.net  Mon Feb 11 14:40:12 2002
From: gward at python.net (Greg Ward)
Date: Mon, 11 Feb 2002 14:40:12 -0500
Subject: [Python-Dev] RE: RFC: Option Parsing Libraries
In-Reply-To: <CLEFLIBFLLLHJNOKPNGFKEDNCDAA.Gerson.Kurz@t-online.de>
References: <CLEFLIBFLLLHJNOKPNGFKEDNCDAA.Gerson.Kurz@t-online.de>
Message-ID: <20020211194012.GB22667@gerg.ca>

On 11 February 2002, Gerson Kurz said:
> I will get beat for this, but: can it be optionally non-case-sensitive? I
> know, I know, in time-honoured unix-tradition a commandline should be
> dangerous and unforgiving in use, but still, please?

Interesting idea; should be trivial given a case-insensitive dictionary.
And hasn't such a beast been bandied about as an example of subclassing
built-in types with Python 2.2?

Anyways, that's an Optik feature requests, and belongs on
optik-users at lists.sourceforge.net.  If you're serious, take it up there.

> Also, I've just scanned the specs and didn't find some
> "rest-of-the-commandline-whatever-that-is" option. As in:
> 
> filename options file1 file2 ... filen

When you do this:

  parser = OptionParser(...)
  (options, args) = parser.parse_args()

then args is the list of positional arguments left over after parsing
options.

But again, that's a question about Optik, and belongs (for now) on the
optik-users list.

        Greg
-- 
Greg Ward - just another Python hacker                  gward at python.net
http://starship.python.net/~gward/
Paranoia is simply an optimistic outlook on life.



From threeseas at earthlink.net  Fri Feb 15 20:45:40 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:40 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <uk7tkylre.fsf@synopsys.com>
Message-ID: <3428.811T2555T11695545threeseas@earthlink.net>

On 11-Feb-02 12:24:37 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> On 07-Feb-02 11:39:14 David Masterson <dmaster at synopsys.com> wrote:

>>> You might be the first to make the claim that those principles can
>>> be represented in "nine commands".  However, so far, your
>>> definition of the "nine commands" isn't clear enough to determine
>>> whether your claim is valid.

>> David, the biggest hurdle I see you dealing with is your belief that
>> the VIC and it's nine commands are, or make up, a language.

>I didn't say that exactly -- but you're close.

>> The VIC is not a language, it is an identification of unavoidable
>> physics in how we use abstraction. You can change the labels
>> attached to the action set all you want, but the action set is
>> physics, unchangeable.

>Computers deal in "language" of one sort or another.  Computers boil
>everything (that they can handle!) down to a series of bits and
>various processes for manipulating those bits.  Problems are
>abstractly represented in one of the languages that the computer
>understands so that it may deal with the problem.

>If the problem of the VIC cannot be abstractly represented in a
>computer language, then perhaps the VIC is not something that can be
>implemented on a computer.

So what is it? A series of bits or a language?

Gee and I always though that if you give a human to many switches to deal
with that they would create "abstraction word = bit sequence/combo" to
deal with the mass of bits. Just becaue it looks like a language doesn't
mean it is.


>> So I've asked for help in creating the validation. <shrug> And I've
>> gotten alot of Nos.... suggesting there is interest in staying
>> ignorant or widespread (trained?) ignorance regarding how to
>> advance.

>Or you haven't given us anything we can deal with.  Note, if we wanted
>to remain ignorant, we could simply just stop talking to you.  You
>have an audience and your "10 minutes of fame", but, if you cannot
>express your problem in understandable fashion, we will go on to other
>things that we do understand.

Sure I've given you a bunch of reflection to deal with (what you do in all
you do) and 10 minutes of fame defined as what all that has happened here
is not, nor will it ever be, "fame".

>That will be your fault -- not ours.

"we", "our"..... "mine" ....... many against The One? hmmmmm...

In other words, you know I'm right. Just look at all the nowhere responses
to "afraid of autocoding" post. I never asked for those who chose to be
negatove to respond at all. Why they did, is because they always knock
down project proposals? Or because they want to knock down autocoding?

Certainly you know the answer that only years of experience in dealing
with the computer industry from the outside has given me.

You also know the difference between a language and an all that a language
is not.

>--
>David Masterson                dmaster AT synopsys DOT com
>Sr. R&D Engineer               Synopsys, Inc.
>Software Engineering           Sunnyvale, CA


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From chris.gonnerman at newcenturycomputers.net  Sat Feb  9 23:47:13 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sat, 9 Feb 2002 22:47:13 -0600
Subject: HTML Generators
References: <I5m98.1105$dy4.39461@eagle.america.net>
Message-ID: <001001c1b1ee$0391bca0$0101010a@local>

----- Original Message ----- 
From: "Mark Sass" <sassman at nucentrix.net>


> Are there any HTML generator modules for Python?  I just want to create
> static html and not use scripts i.e. CGI to do it.  Any available?

http://newcenturycomputers.net/projects/webpub.html

The script is named makesite.py.  I use it for all my static web page
generation, and the latest version also supports being used from a CGI
script.  makesite.py is a templating tool; each source (content) document
is stored in a form similar to rfc822; for instance:

    Title: Home Page

    <h1>Example Page</h1>
    <p>
    Here is an example content page.

This would be stored as index.src (for instance).  Then, the template looks
like this:

    Extensions: .src

    <html>
    <head>
    <title>Sample Site -- <!--%title%--></title>
    </head>
    <body bgcolor=white>
    <!--%body%-->
    </body>
    </html>

The default name for a template file is template.site.  When you run 
makesite.py in the same folder as template.site and the .src files,
each .src file is merged with the template.site file to create a like-
named html file.

The <!--%somenamehere%--> comments in the template are replaced with the
source file headers which match the name (case-insensitive).  The "body"
of the source file is put where the <!--%body%--> name appears.  If a
name can't be found in the source file headers, the template file headers
are consulted; there is also a default data context which contains the
name "date" mapping to today's date in MM/DD/YYYY format.  (I haven't 
needed any other formats so far...)  Header values can be substituted
inside header values recursively.

Sounds more complex than it is...





From phd at phd.pp.ru  Sat Feb  2 12:31:16 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sat, 2 Feb 2002 20:31:16 +0300
Subject: curses module
In-Reply-To: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>; from rhymes@myself.com on Sat, Feb 02, 2002 at 04:59:17PM +0100
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>
Message-ID: <20020202203116.A26499@phd.pp.ru>

On Sat, Feb 02, 2002 at 04:59:17PM +0100, Lawrence Oluyede wrote:
> Why in the std distribution of Python 2.2 there isn't the curses
> module?

   It *is* in the distribution. It is not a single module, but a package.

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



From dsh8290 at rit.edu  Wed Feb  6 10:15:54 2002
From: dsh8290 at rit.edu (dman)
Date: Wed, 6 Feb 2002 10:15:54 -0500
Subject: Autocoding project proposal.
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A16E@natasha.auslabs.avaya.com>
References: <B43D149A9AB2D411971300B0D03D7E8BF0A16E@natasha.auslabs.avaya.com>
Message-ID: <20020206151554.GA12489@dman.ddts.net>

On Wed, Feb 06, 2002 at 04:10:12PM +1100, Delaney, Timothy wrote:
| > From: dman [mailto:dsh8290 at rit.edu]
| > 
| > Figure it out already!!!!  He was giving advice on __communication__,
| 
| Isn't it strange how we accumulate habits?

<grin>

| The normal convention for underlining in plain text is a single
| underscore on each end ...

Unfortunately that wouldn't have been loud enough for the intended
recipient, it seems <0.3 wink>.

-D

-- 

"He is no fool who gives up what he cannot keep to gain what he cannot lose."
        --Jim Elliot




From sholden at holdenweb.com  Mon Feb  4 18:27:20 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 18:27:20 -0500
Subject: Equivalent of Perl chomp?
References: <mailman.1012626605.22240.python-list@python.org>
Message-ID: <_PE78.45877$KM2.1535985@atlpnn01.usenetserver.com>

"Tim Peters" <tim.one at home.com> wrote in message
news:mailman.1012626605.22240.python-list at python.org...
> [Steve Holden]
[times variations of slicing and dicing strings]
>
> This is the kind of fun I was hoping the OP would have.  Good job!  A
lesson
> to take is that timing even seemingly trivial things can turn up major
> surprises with seemingly trivial changes in exactly what it is you're
> timing.  The results are almost never accidental, but determining the true
> causes can leave you crawling on your belly for weeks begging for relief.
>
Yup. Real life can always sock you in the eye.

> A few things going on in this test under the covers:
>
> + endswith does indeed pay a large price for building a bound
>   method object every time.  It's not only the dict lookup to
>   resolve "endswith", but also malloc and free to build the
>   bound method object and discard it afterwards.
>
Hmm, I seem to remember someone else spotted that, not me.

> + The huge increase in sliceit's time going from 1 to 2 is due to
>   that s[:-1] creates a one-character string, and Python interns
>   those by magic:  every instance of, e.g., 'a', no matter how
>   computed, is exactly the same object (that's not actually true,
>   but it's close <wink>).  But 2-character computed strings are
>   not interned, so s[:-2] has to malloc a fresh string object each
>   time to hold the two characters, and free it later.  endswith
>   doesn't have to build new string objects, so doesn't care as much.
>
I'd more or less got there.

> Now for some real fun:  I have no idea why endswith took a huge jump going
> from 1 to 2.  The saving grace for me is that it's so easy to find timing
> mysteries that I'd go mad if I felt compelled to chase them all down, so I
> can walk away without even looking back.  I hope not everyone is so
> well-adjusted <wink>, though, as I'd like to know why.
>
> like-living-in-pumpkin-ly y'rs  - tim
>
Being as well-adjusted (?!) as you, I am happy to see that Bengt is
sufficiently anal to take up the cudgels and "chomp this one to death".

we-can't-all-be-well-adjusted-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/





From hmartires at cultalg.pt  Mon Feb 25 09:27:51 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 14:27:51 -0000
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de> <s%qe8.598$Zj1.7273540@newsserver.ip.pt> <j4664lllc6.fsf@informatik.hu-berlin.de>
Message-ID: <QPre8.602$qQ1.7486593@newsserver.ip.pt>

See my point:
I need to send a list in a socket

server side
------------
list = [0.5 0.5 0.5 0.5]
client.socket.send(list)


client side
-------------
client.socket.recv(1024)

the problem is the buffer size (1024 bytes)
if i need to send a "big" list (with more than 1024 bytes), i have to count
the number of bytes to send in parts. So i must know how many bytes my list
oucupies.

Did your understand my problem ?
All the help you can provide will be apreciated.

Tanks


"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:j4664lllc6.fsf at informatik.hu-berlin.de...
> "Hugo Martires" <hmartires at cultalg.pt> writes:
>
> > And how can i count in a list ?
> > It is possible ?
>
> Read the source code. Alternatively, read the newsgroup archives,
> searching for "size" and "object", e.g.
>
>
http://groups.google.com/groups?hl=en&selm=j43d3m93iu.fsf%40informatik.hu-be
rlin.de
>
> This article also discusses the size of lists.
>
> Regards,
> Martin




From phr-n2002a at nightsong.com  Mon Feb 11 07:05:37 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 04:05:37 -0800
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com> <69c619074b.tim@worthy.demon.co.uk>
Message-ID: <7xlme0fcku.fsf@ruckus.brouhaha.com>

Tim Howarth <tim at worthy.demon.co.uk> writes:
> >   a = eval('0x%x' % a)
> > 
> > might be a little more palatable.  
> 
> Nicer but I'm not sure that it works, I want
> 
> 0xFFFFFFFFL
> to become
> -1

That's the result I get in Python 2.2.  I don't have 2.1 on this system.
1.5.2 raises an exception when it tries to convert 0xFFFFFFFFL to hex.

> besides I grew up believeing eval==evil.

Yes, I agree.  However, the obvious int('FFFFFFFF',16) raises
ValueError because the literal is too large.  I guess you could use
binascii and struct to convert the hex string to a binary string and
then pack the binary string into an integer.  Ouch.






From christopher_gardner at cuc.claremont.edu  Wed Feb 20 12:24:52 2002
From: christopher_gardner at cuc.claremont.edu (Christopher Gardner)
Date: Wed, 20 Feb 2002 09:24:52 -0800
Subject: Problems with enabling thread support in Python 2.1.x
Message-ID: <3C73DBE4.DD20C098@cuc.claremont.edu>

I can't get threads from becoming non-responsive in the version
of python (2.1.2) that I am building on H-P U/X (hpux) 10.20.

Someone mentioned that it appeared that threads isn't enabled
in my version of python running on my box.  I grabbed the tar'ed
source and compiled it.

I have installed Python 2.1, PyWebLib 1.1.0+, python-ldap 2.0.0pre4
against my build of OpenLDAP 2.0.18 on H-P U/X (hpux) 10.20 on a
E35 (e.g., 735) H-P based workstation.  I get the following error
when trying to start web2ldap in stand-alone mode with the HPPD
Software Porting & Archive <http://hpux.connect.org.uk> version:

--
# python /opt/web2ldap/sbin/web2ldap.py
Traceback (most recent call last):
  File "/opt/web2ldap/sbin/web2ldap.py", line 28, in ?
    import w2lapp.core,w2lapp.handler,msHTTPServer,msHTTPHandler,ipadr
  File "/opt/web2ldap/pylib/w2lapp/core.py", line 77, in ?
    crossCheckVars = w2lapp.cnf.misc.session_checkvars,
  File "/opt/python/lib/python2.1/site-packages/pyweblib/session.py",
    line 122, in __init__
    self._session_lock = threading.Lock()
NameError: global name 'threading' is not defined

I used Gnu C (gcc) 2.95.2, binutils 2.9.1, Gnu make 3.78.1, Gnu sed
3.02 to build the python 2.1.2 source against the DB 3.2.9, Pth 1.4
gdbm 1.8, ncurses 5.2, readline x.x and tcp-wrappers 7.6 with the
"-D_REENTRANT" and "-D_XOPEN_SOURCE_EXTENDED" options (OPT) during
building and the libs (-l)pthread and (-l)pth libs flagged.

I seem to be able to compile Python 2.1.2 with POSIX threads
support.  The problem is now that web2ldap starts up in stand-alone
mode -- I'm not able to connect to it.  When I run 'make test' from
the untar'ed source, I get this:

[..]
test_asynchat
Connected
**********************************************************************
Then ...
We expected (repr): "Received: 'hello world'\n"
But instead we got: "error: uncaptured python exception, closing 
 channel <echo_client connected at 4023fa04> (socket.error:(239,
 'Connection refused')
[./python/src/Lib/asynchat.py|handle_read|82]
 [./python/src/Lib/asyncore.py|recv|352])"
[..]

When ./Lib/test/test_asynchat.py & ./Lib/test/test_socketserver.py
are run seperated, I get the following:

--
./python ../Lib/test/test_asynchat.py

Traceback (most recent call last):
  File "../Lib/test/test_asynchat.py", line 4, in ?
    import asyncore, asynchat, socket, threading, time
  File "../../src/Lib/asyncore.py", line 511, in ?
    import FCNTL
ImportError: No module named FCNTL

--
./python ../Lib/test/test_socketserver.py

ADDR = ('localhost', 15251)
CLASS = SocketServer.TCPServer
server created
server running
test client 0
Traceback (most recent call last):
  File "../Lib/test/test_socketserver.py", line 162, in ?
    main()
  File "../Lib/test/test_socketserver.py", line 158, in main
    testall()
  File "../Lib/test/test_socketserver.py", line 148, in testall
    testloop(socket.AF_INET, tcpservers, MyStreamHandler, teststream)
  File "../Lib/test/test_socketserver.py", line 124, in testloop
    testfunc(proto, addr)
  File "../Lib/test/test_socketserver.py", line 63, in teststream
    s.connect(addr)
socket.error: (239, 'Connection refused')
thread: creating server
thread: serving three times

However, when I start web2ldap, I get this:

--
./src/python /opt/web2ldap/sbin/web2ldap.py

Accepted IP address ranges: 0.0.0.0/0.0.0.0
Running mode: multi-threaded
SSL: disabled

Point your favourite browser to

<http://127.0.0.1:1760/web2ldap>

to access the web application.

... I am running tcp-wrappers 7.6 on my box. Could that cause a
problem as well?


From phr-n2002a at nightsong.com  Mon Feb 18 00:10:50 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 21:10:50 -0800
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org> <a4q0f4$akj4@news1.gtech.com>
Message-ID: <7xy9hridd1.fsf@ruckus.brouhaha.com>

"Peter Milliken" <peter.milliken at gtech.com> writes:
> Why not use a language that provides your requirements of signal safety etc
> and do a Python API for it? (assuming one doesn't exist) i.e. I find it
> somewhat difficult to reconcile your statement about responding to time
> critical signals when you propose use of an interpretive language..... :-)
> Sounds like you would ideally like real-time language features in a language
> that isn't designed for it (well, I'm not sure Guido had visions of Python
> being used in real-time applications :-)).

I don't see anything real-time specific in that post.  It's a generic
problem.  I don't see a clean way to make sure your 'finally' clauses
execute if the user hits ^C several times while the program is
running.  Early versions of Unix didn't have reliable signals either,
and it led to flaky programs, til BSD Unix fixed it.


From phd at phd.pp.ru  Fri Feb  1 04:32:41 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 1 Feb 2002 12:32:41 +0300
Subject: locale on cygwin
In-Reply-To: <3C5A607F.6060107@halfcooked.com>; from andy47@halfcooked.com on Fri, Feb 01, 2002 at 08:31:43PM +1100
References: <Xns91A1A678DEEE3andy47halfcookedcom@203.109.250.24> <20020125140805.GC1252@dothill.com> <3C5A607F.6060107@halfcooked.com>
Message-ID: <20020201123241.H10316@phd.pp.ru>

On Fri, Feb 01, 2002 at 08:31:43PM +1100, Andy Todd wrote:
>  >>> locale.setlocale(locale.LC_ALL, '')
> Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
>    File "/usr/lib/python2.1/locale.py", line 374, in setlocale
>      return _setlocale(category, locale)
> locale.Error: locale setting not supported
>  >>>
> angua:/home/andy47# echo $LANG
> en_AU
> angua:/home/andy47#

   On lInux this means your system lacks files for en_AU locale. It is not
Python fault. Use different locale, en_EN or any that exists on the system.

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



From mcfletch at geocities.com  Tue Feb 26 19:40:40 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Tue, 26 Feb 2002 19:40:40 -0500
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <3C7C2B08.6040900@geocities.com>

See standard module bisect:

# untested...
lines = open( file ).readlines()
# name is < name+',number'
import bisect
index = bisect.bisect( lines, name )
if index+1 < len(lines):
	line = lines[index+1]
	if line[:len(name)+1] == name+',':
		do_stuff_with_line(line)

For your own solution:
	for loop would likely be faster than filter, as it only needs to check 
until it matches or the values become > the test value

HTH,
Mike

HankC wrote:
> Greetings:
> 
> I'm 2 days new to Python and hope I can get some pointers here...
> 
> I have a sorted, delimited text file such as:
> 
> Arty,1000
> Bobby,2000
> Charlie,3000
> 
> Knowing a name, I want to return the entire line from the file.  My
> approach/problems (any comments appreciated):
> 
> # use the filter method
> # filter fxn
> def Match(line): return line[:toklen] == tok
> 
> # find match fxn
> def FindMatch:
> 	fp = open(fname)
> 	flist = fp.readlines()
> 	dataline = filter(Match, eodflist)
> 
> My problem is that I don't understand how, if it's even possible, to
> set tok and toklen values in the FindMatch fxn and pass them to Match.
> Is it possible to pass multiple values to the function used by
> filter()?
> 
> More general...  does filter() have any performance advantage over
> just using a for loop?  The datafile list has 10,000+ items.  My
> background is in object pascal where I would read the items into a
> list then use a binary search to find the target.  I'm hoping there's
> an efficient python method to perform a search easily on a sorted
> list.
> 
> Suppose I want to make tok and toklen global to the module.  Is this
> possible?
> 
> TIA!
> 


-- 
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From ahmebah at hotmail.com  Thu Feb  7 12:23:38 2002
From: ahmebah at hotmail.com (ameoba)
Date: 7 Feb 2002 17:23:38 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org>
Message-ID: <Xns91AE6032817F5ahmebahhotmailcom@198.99.146.10>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
news:mailman.1013090595.10717.python-list at python.org: 

 
> How the heck to you "enjoy" an idiot like Rue though?


With a light chianti?


From sholden at holdenweb.com  Thu Feb  7 16:55:47 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 16:55:47 -0500
Subject: [NEWBIE] ftplib again...
References: <mailman.1013094492.31504.python-list@python.org>
Message-ID: <sJC88.96852$KM2.3618183@atlpnn01.usenetserver.com>

"Bruce Dykes" <bkd at graphnet.com> wrote ...
> Okay...I've got the ftp connections working...thing is, when I did my
> initial prototyping, I used a local file, and readlines() to get the data
> from the file into a list, and to create the function to parse out the
> useful data from noise. Well it worked as far as getting the processing
> logic down, but now that I'm trying from ftp'ed files, well...
>
> All I need to do is get the file contents into a list. That's it. Now it
> looks as if retrlines("RETR filename") will do that, with some additional
> manipulations.
>
> But it doesn't. Every single variation I've tried,
> dataset=[host.retrlines("RETR filename") ],  host.retrlines("RETR
filename",
> dataset.append) , dataset=list(host.retrlines("RETR filename") ), etc.
> always results in the file contents getting dumped to the
> console/stdout/interactive window, and the completion code, '226 Command
> Completed Successfully' getting assigned to the variable.
>
> What am I missing? Do you need to see more of my code?
>
"""
retrlines (command[, callback])
Retrieve a file or directory listing in ASCII transfer mode. command should
be an appropriate "RETR" command (see retrbinary() or a "LIST" command
(usually just the string 'LIST'). The callback function is called for each
line, with the trailing CRLF stripped. The default callback prints the line
to sys.stdout.
"""
This doesn't really make it that obvious, but what you have to do is provide
a callback, which is called, with the line as an argument, each time a line
is retrieved from the server. The code you need is something like this:

linelist = []
host.retrlines("RETR filename", linelist.append)

After which, linelist will be a list of non-terminated lines.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jeff at ccvcorp.com  Mon Feb 18 16:40:46 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 18 Feb 2002 13:40:46 -0800
Subject: wxPython:  EVT_CLOSE question
References: <c0a56015.0202151702.1f29d19e@posting.google.com> <slrna6tjb6.e10.zeitlin@seth.lpthe.jussieu.fr> <c0a56015.0202171933.7d828ae1@posting.google.com>
Message-ID: <3C7174DE.2CCD73AA@ccvcorp.com>


alan scales wrote:

> Vadim Zeitlin <zeitlin at seth.lpthe.jussieu.fr> wrote in message news:<slrna6tjb6.e10.zeitlin at seth.lpthe.jussieu.fr>...
>
> >  The best would be to do it in EVT_CLOSE of the frame containing the control.
> > You also can use EVT_WINDOW_DESTROY but this could be too late to do anything
> > meaningful with the control.
>
>
> the frame containing the control isn't being closed (the controls in
> question are sort of being "paged in and out" - maybe this is
> pathological application). however, the EVT_WINDOW_DESTROY might be
> made to work.

In this case, I would give each of the controls a Save() method (or some equivalent), and call that each time before you
page them out.  It's a nuisance to have to remember to do so every time, but should be workable. (Be sure to document
it!)

Jeff Shannon
Technician/Programmer
Credit International




From mcfletch at geocities.com  Thu Feb 14 02:03:40 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Thu, 14 Feb 2002 02:03:40 -0500
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <NBBBIOJPGKJEKIECEMCBMEDBMCAA.pobrien@orbtech.com>
Message-ID: <3C6B614C.7040003@geocities.com>

Actually, with the work I just finished, I replaced all watcher 
instances with property objects and call dispatcher directly, so I no 
longer need the watchers at all :) .  The version up on the site is the 
old one, the new one is rather trivial (which is why I got rid of it in 
my code):

import dispatcher

class WatchAble:
	"""Mix-in class providing support for "watching" object changes
	
	Not currently thread-safe, should add that some day
	Use of _v_ prefix is to support use with ZODB persistence
	mechanisms.  It indicates that the list is volatile and should
	not be stored in the ZODB.
	"""
	### Customisation Point
	def beforeUpdateWatchers( self, **keywords):
		"""Called before any watchers are updated, lets you update object state 
before the watchers see changes"""

	### Public API
	def addWatcher( self, object, signal = dispatcher.Any ):
		"""Add a new watcher object, must be a callable object with signature:
			object( watchedObject, event )
		Where watchedObject is this instance, and event
		is the opaque parameter passed to updateWatchers.
		"""
		dispatcher.connect( object, signal, self )
	def removeWatcher( self, object, signal = dispatcher.Any ):
		"""Remove a particular watcher from the list of watchers"""
		dispatcher.disconnect( object, signal, self )
	def updateWatchers( self, signal = dispatcher.Any, **keywords ):
		"""Send event to each watcher of the object"""
		self.beforeUpdateWatchers( signal=signal, **keywords)
		dispatcher.send( signal, self, **keywords)

As you can see, no real benefit beyond the interception before the 
message is sent, and conceptually, that should (in my case) be handled 
by the assignment of property values, not by a pre-messaging hack.  Of 
course, the pattern does make the watcher reference local to the object, 
rather than global, but that's just a coding preference, and not 
important to me.

Anywho, need to get back to work. This whole Python 
2.2-on-ZODB-with-automatic-event-generation excursion has sucked up 
almost 4 full days of working time.

Enjoy,
Mike


Patrick K. O'Brien wrote:
> Thanks. Is your new watcher class available? I couldn't tell if this one
> (http://members.rogers.com/mcfletch/programming/watchable.py) was the newly
> rewritten version or not.
> 
> The dispatcher module is part of a project that is very alpha, called
> Bulldozer. You can find the files in CVS on SourceForge:
> 
> http://sourceforge.net/projects/bdoz/
> 
> The license is the same as the Python license. As Magnus mentioned, there
> were several implementations developed as part of the anygui project. We all
> learned from each other, but still came up with different approaches.
> 
> ---
> Patrick K. O'Brien
> Orbtech
...





From matthias_j at lycos.com  Sun Feb 10 14:22:08 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 10 Feb 2002 11:22:08 -0800
Subject: Help with Gordon McMillan Installer
References: <d7d5ebdf.0202091152.40a6e0af@posting.google.com> <Xns91B167C6A6F08gmcmhypernetcom@199.171.54.216>
Message-ID: <d7d5ebdf.0202101122.19931f2a@posting.google.com>

Gordon McMillan <gmcm at hypernet.com> wrote in message news:<Xns91B167C6A6F08gmcmhypernetcom at 199.171.54.216>...
> Matthias Janes wrote:
> 
> [snip]
> 
> > I would like to create a onefile exe but I can't figure out how to
> > write the right Makespec.file.
> 
> [snip]
> 
> > a = Analysis(['C:\\_Matthias\\_APPS\\Installer\\support\\useUnicode.py',
> > 'C:\\_Matthias\\_APPS\\TEST3\\MyApp.py'],
> >              pathex=[])
> > pyz = PYZ(a.pure)
> > exe = EXE( pyz,
> >           a.scripts,
> >           a.binaries,
> 
>             [('MyData.cfg', 'c:/_Matthias/_APPS/TEST3/Data.cfg', 'BINARY')],
> 
> >           name='MyApp.exe',
> >           debug=0,
> >           console=1)
> > 
> > 
> > I tried different ways to add the Data.cfg file but this is never
> > included in the ONEFILE.EXE instead it is copied extra in a folder.
> 
> Inline above. As BINARY, it will be extracted on each run. Of course,
> that makes it effectively read-only.
> 
> -- Gordon
> http://www.mcmillan-inc.com/

Thanks a lot Gordon;

This help a little.
1. I had a small problem because you changed the Data.cfg to
MyData.cfg - after that it worked when I put the Data.cfg file in the
root dir.

2. Now to my real question : How about when the Data file is in a
subfolder I tried this:

a = Analysis(['C:\\_Matthias\\_APPS\\TEST3\\MyApp.py'],
             pathex=[])
pyz = PYZ(a.pure)
exe = EXE( pyz,
          a.scripts,
          a.binaries,

          [('Data.cfg', 'c:/_Matthias/_APPS/TEST3/Data/Data.cfg',
'BINARY')],

          name='MyApp.exe',
          debug=0,
          console=1)


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

But the file in the subfolder Data is missing -- CAN Installer also
EXTRACT subfolders correctly and how do i include this folder/files
proberly.

As I work with larger applications I would like to have access to
subfolders and not everything in the root dir.

Any help would be greate.


From phr-n2002a at nightsong.com  Mon Feb 18 00:24:30 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 21:24:30 -0800
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org>
Message-ID: <7xu1sficq9.fsf@ruckus.brouhaha.com>

Mark Mitchell <mark at codesourcery.com> writes:
> But, I'm having a hard time figuring out how to write signal-safe
> code in the presence of synchronous signals.  Consider, for example,
> the following:

The most obvious way to deal with it is for the interpreter to
postpone delivery of exceptions while a finally clause, and of
exception X while an exception X handler is running, where X is some
asynchronous exception like KeyboardInterrupt.  Asynchronous
exceptions seems like an area where Python needs some further
design work.  I'll dig out some manuals when I get a chance and
see how the MIT Lisp machine dealt with these issues.


From jdaily at kc.rr.com  Mon Feb 11 20:02:30 2002
From: jdaily at kc.rr.com (james)
Date: Tue, 12 Feb 2002 01:02:30 GMT
Subject: just a moron trying to get python working on win98 and personal web 
 server
Message-ID: <3C685DDE.AA8AFD10@kc.rr.com>

RE: just a moron trying to get python working on win98 and personal web
server

i am trying to get python to work on my MS personal web server ( which
is
supposed to support python in a web url ).

what do i need to do?

i have tried setting the registry.

i have set the PATH in autoexec.bat.

is there an apache alternative.




From sab at NOSPAM.freeuk.com  Mon Feb  4 10:30:58 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Mon, 4 Feb 2002 15:30:58 -0000
Subject: Amount of files on a drive?
Message-ID: <a3m9d3$mr4$1@newsg2.svr.pol.co.uk>

Is there a fast an easy way to get the number of files on a drive??
ta,
G. Willoughby




From python_only at yahoo.com  Sat Feb  2 00:50:19 2002
From: python_only at yahoo.com (Goh S H)
Date: 1 Feb 2002 21:50:19 -0800
Subject: Java vs Python Benchmarks: Java is faster
References: <Ldc68.67$8u4.28600@wagner.videotron.net>
Message-ID: <184fbd02.0202012150.2f88c22b@posting.google.com>

I would like to comparison between Java (JIT enabled) & Python (Psyco enabled).

For those not familiar with psyco, point to

http://homepages.ulb.ac.be/~arigo/psyco/


From caljason76 at yahoo.com  Tue Feb 19 20:46:30 2002
From: caljason76 at yahoo.com (Jason)
Date: 19 Feb 2002 17:46:30 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org>
Message-ID: <e9e6dd1f.0202191746.2c3525db@posting.google.com>

# I am a Python novice so I don't really understand how things
# work, yet.  So some of this may either be wrong or not the
# Python way.  Also I have no idea why the 'if __name__ == ...'
# construct is needed instead of just the naked main statements.
#
# A few notes on the code.  There are two solutions: list_naked
# and list.
#
# list_naked is the simplies solution, it has no wrapper object
# and each node in list_naked is the head of a list.  You can see
# the recursion in list_naked.sum() and list_naked.__str_all(),
# where you will notice the if statements that test for the
# base case of the recursion.
#
# list_wrapped is a little more complex.  It has a wrapping object
# that is always used to refer to the list (unlike list_naked
# where the head of the list can change when a new value is pushed,
# hence the ll = ll.push(ll,x) assignment).  Here, the recursion
# is stopped by the_list_wrapped_empty_node and all the control
# flow is done by the interpreter.  I like this best, because I
# hate explicit control flow in programs and I think that it makes
# it easier to see the forest instead of the trees.  Looking at
# list_wrapped.sum() it is clear: add this node to the sum of the
# rest.  In list_naked.sum() you now have to look at two statements
# which do the same thing semantically.  With list_wrapped there
# is no way to forget to check for your base case because you don't
# check anything!  With just a list it seems a little too much,
# but when you start playing with various tree types and graphs
# it makes the code so much more readable.
#
# Also, can somebody help me find the documentation to the builtin
# procedures, like print().  I checked the __builting__ module docs
# but there is nothing there.  Thanks
#
# -j

class list_naked:
    def __init__(s, data, next = None):
        s.data, s.next = data, next
    def __str__(s):
        return "[%s]" % s.__str_all()
    def push(s, data):
        return list_naked(data, s)
    def sum(s):
        if s.next: return s.data + s.next.sum()
        else:      return s.data
    def __str_all(s):
        if s.next: return str(s.data) + " " + s.next.__str_all()
        else:      return str(s.data)

class list_wrapped_node:
    def __init__(s, data, next): s.data, s.next = data, next
    def __str__(s): return str(s.data) + " " + str(s.next)
    def sum(s): return s.data + s.next.sum()

class list_wrapped_empty_node(list_wrapped_node):
    def __init__(s): pass
    def __str__(s): return ""
    def sum(s): return 0

class list_wrapped:
    def __init__(s): s.head = list_wrapped_empty_node()
    def __str__(s): return "[" + str(s.head) + "]"
    def push(s, data): s.head = list_wrapped_node(data, s.head)
    def sum(s): return s.head.sum()
        
if __name__ == '__main__':
    ll = list_naked(0)
    for i in range(6): ll = ll.push(i+1)
    print ll
    print ll.sum()

    l = list_wrapped()
    for i in range(6): l.push(i)
    print l
    print l.sum()


From kp87 at lycos.com  Fri Feb 15 00:55:27 2002
From: kp87 at lycos.com (kevin parks)
Date: 14 Feb 2002 21:55:27 -0800
Subject: help with silly algorhythm
Message-ID: <5e8bd451.0202142155.143ca4d9@posting.google.com>

I have been wanting to ask this one for a while, but it is a bit
tricky to articulate so I will try my best.

What I want to do is generate a new list of items based on
an input list and a some criteria (list of acceptable values, for
example). So, i might have:

* list of input values =[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11] 
* a list of acceptable new 'partner values' = [1, 3, 7, 10]

Now I want to create a new list made up of the closest 
members of the [1, 3, 7, 10] list in 4 different ways:
  - upper
  - lower
  - alternating starting higher
  - alternating starting with closest lower value

All with the further stipulation that values from the input list
do not team with values in the "buddy" list if they are the
same, so input value 1 would not go with 1 from [1, 3, 7, 10], 
but with 3 (the next value up),  or 10 (next value down), depending
on which of the 4 possible 'modes' we chose.

vague i know, i am trying my best...

so we might get (in mode 1: all closest higher value)

[ [0, 1], [1, 3], [2, 3], [3, 7], [4, 7], [5, 7], [6, 7], [7, 11]....

-or (as a single new list) -

[1, 3, 3, 7, 7, 7, 7, 11.....]

mode 2 (all closest value down) might be:

[[0,10], [1, 10], [2, 1], [3, 1], [4, 3], [5,3], [6,3], [7,3], [8, 7]....

mode 3

[ [0, 1], [1, 10], [2, 3], [3, 1].....

etc (hope i haven't made a mistake here).

So problem number one for me is how do i figure out which value
is 'closest' (up or down), do i have to test it against each
value of the list or is there a smarter way.

I realize that this is very odd and convoluted. If anyone can make
heads or tails of this explanation and would like to help me I
would be grateful.

-kp--


From ballabio at mac.com  Thu Feb 28 05:02:12 2002
From: ballabio at mac.com (Luigi Ballabio)
Date: Thu, 28 Feb 2002 10:02:12 +0000
Subject: full closures
In-Reply-To: <7934d084.0202271547.60ab7365@posting.google.com>
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com>
 <a5jbra$7ok00$1@ID-11957.news.dfncis.de>
Message-ID: <5.1.0.14.0.20020228094114.00a9cbe0@mail.mac.com>

At 03:47 PM 2/27/02 -0800, Andrae Muys wrote:
> > I don't understand closures, but I might try something like:
> >
> > class Counter:
> >     def __init__(self):
> >         self.count = 0
> >     def __int__(self):
> >         self.count += 1
> >         return self.count
> >
> > counter = Counter()
> > print "the current count is %d" % counter
> >
>
>That works, as does using a generator:
>
>def Counter():
>     counter = 0
>     while 1:
>         print "the current count is %d % counter
>         counter += 1
>         yield counter - 1   # might as well return it :)

Hi all,
         the usual Scheme closure thing, twisted and perverted to have it 
work in Python but probably very un-Pythonic because of that, would look like:

 >>> def counter(x=0):
...     l = [x-1]
...     def f(l):
...         l[0] += 1
...         return l[0]
...     return lambda: f(l)
...
 >>> c1 = counter()
 >>> c2 = counter(5)
 >>> c1()
0
 >>> c1()
1
 >>> c1()
2
 >>> c2()
5
 >>> c2()
6

Note: the above only works in Python 2.2 (by default) or Python 2.1 (by 
importing nested_scope from __future__)

Much of the awkwardness comes from having to store the counter in a list to 
have pass-by-reference to f, and from having to define a separate f because 
lambda is limited to a single expression. In Scheme, which has different 
semantics, this trick is done much more tersely as

(define (counter x)
   (let ((y (- x 1)))
     (lambda () (set! y (+ y 1)) y)))

which in non-working pseudo-Python syntax for the parens-challenged would be:

def counter(x):
     y = x-1
     return lambda: y += 1; y

but one can't have everything. It wouldn't be Python if it had Scheme 
semantics...

Bye,
         Luigi




From olsongt at bellatlantic.net  Tue Feb 12 17:10:00 2002
From: olsongt at bellatlantic.net (Grant Olson)
Date: Tue, 12 Feb 2002 22:10:00 GMT
Subject: Just for the gurus -- Remote admin password changing with privileged user ??
References: <d7d5ebdf.0202121332.4f7d3723@posting.google.com>
Message-ID: <Yoga8.15982$gf5.3530@nwrddc02.gnilink.net>

If you can find a Windows 2000 box, there is a RUNAS command that operates
like SU.  Either execute from the console, or right-click on an icon to
bring up Run As... on the context menu.

You also might want to look at cusrmgr from the NT resource kit

--
"Now that's something you don't see every day...
   unless you go to Arby's every day."
"Matthias Janes" <matthias_j at lycos.com> wrote in message
news:d7d5ebdf.0202121332.4f7d3723 at posting.google.com...
> Just a small question:
>
> I'm working on a PC (Win NT) loggod on as user Matthias to the DOMAIN
> VIENNA:
>
> I would like to execute localy a python script to CHANGE THE PASSWORDS
> for
> 2 Administrators (SuperUser, TechAdmin) which are remote (60 NT
> Servers)on 60 different Domains where there is no user account
> Matthias.
>
> I thought of using the win32net.NetUserChangePassword to do the job.
>
>
>
>             BUT THE REAL PROBLEM is that I do not have admin rights on
> the other domains.
>
> I could copy the script to all the Domains and execute it logged on as
> Admin - but I would realy like to avoid this if possible.  Maybe there
> is a way to send an authetification to the domains like
> username:TechAdmin + the old passwort - (like a Su command) I'm not
> sure..... -- ANY HELP
>
> Would be absolutly create.
>
> THANKS




From winelight at spamcop.net  Tue Feb 12 16:30:51 2002
From: winelight at spamcop.net (Andrew Ward)
Date: Tue, 12 Feb 2002 21:30:51 -0000
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com> <Ry3a8.40197$bP3.337803@NewsReader> <8f070afd.0202120550.45d8873b@posting.google.com> <Xns91B39B73814D6mhueningzedatfuberli@130.133.1.4>
Message-ID: <gQfa8.41893$bP3.355857@NewsReader>

It's very intermittent for me. I can leave it running for hours in the
morning and it's fine, in the evenings (UK time) it usually does fail fairly
soon.

I'm rewriting the url stuff as individual httplib calls to see if I can pin
down the exact point of failure, although I assume it will still hang, and
that it will also be at the read() (I imagine that the urllib read() maps
directly to the http read() method).

Andrew Ward

> Hhmm, I can't reproduce this. Everything works just fine.





From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 21:47:18 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 20:47:18 -0600
Subject: redirect console output to widget?
References: <de6c123c.0202101612.72b27f8d@posting.google.com>
Message-ID: <014301c1b2a6$705c58a0$0101010a@local>

----- Original Message ----- 
From: "Lei Chen" <llchen223 at hotmail.com>


> Hi,
> 
> I want to redirect console output to a Text widget.  So far the code 
> I wrote crashes on occasion.  Attached is the portions of the code. 
> Thanks in advance for your help.  Here are some more details: program
> runs on WIN32; the console stream is fast and continuous (console
> program runs 24x7)

<< hack hack hack >>

I didn't think you were supposed to use Tkinter in parallel threads?

Am I wrong?  (Someone will tell me if I am, I'm sure...)





From tismer at tismer.com  Sun Feb 17 10:13:16 2002
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 17 Feb 2002 16:13:16 +0100
Subject: Is Stackless Python DEAD?
References: <184fbd02.0110262039.74799675@posting.google.com> <j4d72xasx0.fsf@informatik.hu-berlin.de> <uady09pkm.fsf@python.net> <mailman.1009806401.11232.python-list@python.org> <110A0579A87BA846.8CEE70EF24F483BD.740CFC77F4D54FA3@lp.airnews.net> <mailman.1013761643.13888.python-list@python.org> <pan.2002.02.15.10.39.47.229964.1126@uchicago.edu>
Message-ID: <3C6FC88C.6000202@tismer.com>

Ben Wolfson wrote:

> On Fri, 15 Feb 2002 02:27:03 -0600, Christian Tismer wrote:
> 
> 
>>The new solution is cool and uncool at the same time.
>>All my difficult changes to the core have been abandoned.
>>All magic about Stackless' implementation is gone (sigh).
>>It is now orthogonal to the Python engine. 
>>
> 
> Does this mean that the changes that made Stackless faster than CPython
> are gone, too?


Right now, yes. Maybe they will be re-applied later.
Stackless is today running 1-2% slower than CPython.
That's due to a little indirection and overhead
in the function wrapped around eval-frame, plus
some fractal effects that seem to appear whenever
I touch the core at all. Adding code always resulted
in some speed loss, at least on Win32. No idea what
exactly happens; I think there are unfortunate
decisions about code placement in the compiling.

It has to be said that the speed patches of the former
Stackless have already been split apart. Without them,
old Stackless was 5 or more percent slower than CPython.
I don't know of these patches can still easily applied
to Python 2.2, but there is a good chance that this
will work out.
In that sense, the new Stackless is even faster than
the old one.

I might put some time into this, after Stackless is
ready for production code. These should be considered
as two disjoint projects now.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From jason at jorendorff.com  Fri Feb  1 20:47:47 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 19:47:47 -0600
Subject: Windows 2000
In-Reply-To: <01C1AB52.89FFDF80@BRIANMIT>
Message-ID: <HFEKILOLEFEFMKAECNDLGENECPAA.jason@jorendorff.com>

Brian Mitchell wrote:
> I tried an install of Python 2.2 on a Windows2000 box today. IDLE
> seemed to function okay but I couldn't close the command mode
> with control-Z.  Is there anything special about using Python under
> Windows2000 as compared to Windows98?  TIA.

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> exit
'To exit, type Ctrl-D.'
>>> 

## Jason Orendorff    http://www.jorendorff.com/




From laymon at crd.ge.com  Tue Feb  5 08:47:47 2002
From: laymon at crd.ge.com (Marc Laymon)
Date: 5 Feb 2002 05:47:47 -0800
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com> <3c5ef07d$0$2201$7586b60c@news.frii.com>
Message-ID: <60018f02.0202050547.4836c5ff@posting.google.com>

Several people suggested that the script needs to be
executable.  However, python is in
/usr/bin/python and the script is executable:

(laymon) [23] which python
/usr/bin/python
(laymon) [24] ls -l retest.py
-rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*

If the script is not executable, you get a "Permission denied"
error, not "Command not found"

(laymon) [25] chmod -x retest.py
(laymon) [26] ./retest.py
./retest.py: Permission denied.
(laymon) [27] chmod +x retest.py
(laymon) [28] ./retest.py
./retest.py: Command not found.
(laymon) [29] ls -l retest.py
-rwxr-xr-x    1 laymon   omtool       1747 Feb  4 10:55 retest.py*

I can duplicate the error message with a perl script by changing
"#!/usr/bin/perl" to "#!/usr/bin/foo", so it seems that for some
reason, Unix can't find python, but since /usr/bin is in my path
and I am explicitly giving the command "usr/bin/python",
I can't figure out why.  Anybody have any other ideas ?

Thanks.

> "Marc Poulin" <mpoulin at verinet.com> wrote in message > news:<3c5ef07d$0$2201$7586b60c at news.frii.com>...
> In article <60018f02.0202041208.67d3100e at posting.google.com>, "Marc
> Laymon" <laymon at crd.ge.com> wrote:
> 
> > I am just learning python.  I have written a few simple scripts which
> > run fine when invoked using "python myscript.py".  The Python manual
> > says you can execute a Python script directly In Unix by making the
> > first line "#!/usr/bin/python", then giving the command "./myscript.py".
> >  However, when I try this, I get an error "Command not found.".  I get
> > the same error under both Linux and Solaris. Using the same format for
> > perl works fine.  Why doesn't the "#!/usr/bin/python" line work for
> > Python ?  (perl and python are both in /usr/bin on my Linux system.)
> > 
> > Thanks.
> > 
> > Marc Laymon
> 
> First, use the "which" command to make sure the interpreter is in your
> path:
> 
>      which python
> 
> You should see
> 
>      /usr/bin/python
> 
> Next, make your script executable. Type
> 
>      chmod +x myscript.py
> 
> at the command line.


From kkeller at speakeasy.net  Fri Feb  8 16:18:53 2002
From: kkeller at speakeasy.net (Keith Keller)
Date: Fri, 8 Feb 2002 13:18:53 -0800
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com>
    <41914a.537.ln@wombat.san-francisco.ca.us>
    <a40e2r.3vvegs9.1@kserver.org>
    <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net>
Message-ID: <tbf14a.ep7.ln@wombat.san-francisco.ca.us>

In article <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232 at lp.airnews.net>,
	claird at starbase.neosoft.com (Cameron Laird) writes:
> In article <a40e2r.3vvegs9.1 at kserver.org>,
> Sheila King  <usenet at thinkspot.net> wrote:
>>On Fri, 8 Feb 2002 11:30:44 -0800, kkeller at speakeasy.net (Keith Keller)
>>wrote in comp.lang.python in article
>><41914a.537.ln at wombat.san-francisco.ca.us>:
>>
>>> In article <a417rn$ge8$1 at panix2.panix.com>,
>>> 	aahz at panix.com (Aahz Maruch) writes:
>>> > I know some of y'all still keep roots to the Perl community, so I was
>>> > wondering if there's any kind of definitive nickname for Perl
>>> > programmers, the way we call ourselves Pythonistas?
>>> 
>>> I don't think so--having one nickname would contradict TMTOWTDI.  :)
>>
>>I thought it was Perl Monks.
>
> "Mongers" is the closest answer.
> 
> I generally write "Perlites", though.

I told y'all: TMTOWTDI!  At least three, at present count.

Aahz, are you sorry you asked yet?  :)

--keith

-- 
kkeller at speakeasy.net
public key:  http://wombat.san-francisco.ca.us/kkeller/public_key
alt.os.linux.slackware FAQ:  http://wombat.san-francisco.ca.us/perl/fom



From phil at river-bank.demon.co.uk  Thu Feb 28 14:50:25 2002
From: phil at river-bank.demon.co.uk (lihpuk)
Date: Thu, 28 Feb 2002 19:50:25 -0000
Subject: PyQT and shaped windows in X
In-Reply-To: <d906b2c1.0202280334.35f7942f@posting.google.com>
Message-ID: <a5m1m1+6c9v@eGroups.com>

--- In python-list at y..., mtaylor at l... (Moray Taylor) wrote:
> Hi,
> 
> Can anyone point me in the direction of an example of shaped windows
> using QT with Python, is the XLib module required? Is it even
> possible?
> 
> BTW I have tried asking in the PyQT/PyKDE mailing list, but it 
wasn't
> considered to be a PyQT question, go figure.

PyQt wraps Qt. If Qt can do it then PyQt can probably do it. If Qt 
can't do it then PyQt can't do it.

Phil




From tismer at tismer.com  Tue Feb 19 07:39:43 2002
From: tismer at tismer.com (Christian Tismer)
Date: Tue, 19 Feb 2002 13:39:43 +0100
Subject: O'Reilly Network: Stackless Reincarnate [Feb. 14, 2002]
Message-ID: <3C72478F.20103@tismer.com>

<http://www.onlamp.com/pub/a/python/2002/02/14/pythonnews.html>

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From grante at visi.com  Wed Feb 20 10:51:55 2002
From: grante at visi.com (Grant Edwards)
Date: Wed, 20 Feb 2002 15:51:55 GMT
Subject: Guido wins FSF Award
References: <mailman.1013906967.6705.python-list@python.org> <3c73bdf1$0$79553$6536d21b@news.bitstream.net>
Message-ID: <vCPc8.44440$Wf1.8408903@ruti.visi.com>

In article <3c73bdf1$0$79553$6536d21b at news.bitstream.net>, mylinuxboxroot wrote:
> 
> Tim Peters <tim.one at home.com> wrote in message
> news:mailman.1013906967.6705.python-list at python.org...
>> [Paul Rubin]
>> > Congratulations to Guido for winning the fourth Free Software Foundation
>> > Award, for inventing and implementing Python as Free Software:
>> >
>> >   http://www.fsf.org/press/2002-02-16-FSF-Award.html
>>
>> Hooray for Guido!  Shows how much the world cares about secure pickles
>> <wink>.
> snip
> 
> what are revealing the Nine Pythonic Actions underlying all that we do.
> can you explain what that means ??????

Bhuddism allusion?

-- 
Grant Edwards                   grante             Yow!  What a
                                  at               COINCIDENCE! I'm an
                               visi.com            authorized "SNOOTS OF THE
                                                   STARS" dealer!!


From nospam at bigfoot.com  Fri Feb 15 17:49:44 2002
From: nospam at bigfoot.com (Gillou)
Date: Fri, 15 Feb 2002 23:49:44 +0100
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <a4k2si$1skr$1@norfair.nerim.net>

>
> * No need to declare variables. Use them and they are immediately
>   created for you, ex nihilo.

You declare types in Basic (the Dim statement of VB and/or as var name
suffix)

>
> * NO LINE NUMBERS! Enough said.
...
>   IF condition GOSUB line-number (I never did understand the necessity
>   for that kind of statement until well after I'd stopped using BASIC
>   entirely and moved to more "modern" languages like... Pascal).

Only old men like you and me remember this

>
> * Batteries included. This is the big one, which makes Python more than
>   just a toy language suitable for learning. The richness of the
>   standard library makes Python incredibly useful, and I for one would
>   like to see it continue to grow. I understand the concern that people
>   developing Python-based programs for distribution would like their
>   users to be able to run a 100k script without downloading a 10M
>   distribution, but I don't want to see that cause a reduction in the
>   standard library. Instead, we could make make two Python
>   distributions, a "full" install containing everything and a "lite"
>   install containing only the interpreter and maybe -- *maybe* -- a
>   couple of essential modules like os and sys. Then people developing
>   Python programs like, say, a pygame-based game, could distribute a
>   single package containing the "Python lite" distribution, their own
>   code, and any other modules their code depended on -- all in one
>   easy-for-the-end-used-to-install package.

And when downloading another one he gets the python interpreter again, and a
new copy of some packages...
No, Python deserves a dedicated RPMfind  like utility that manages
dependencies (OS, python version, required modules...).
This would be a really great extension for distutils.

>
> Anyhow, I hope this helps develop some ideas.
>
> --
> Robin Munn
> rmunn at pobox.com




From mdk at mdk.com  Thu Feb 14 15:53:52 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 15:53:52 -0500
Subject: Convert String to Dictionary question
Message-ID: <a4h851$a8g7$1@ID-98166.news.dfncis.de>

I saved my dictionary {'hello1': [1, 0, 0]} to a file.

When I do a readline the I get back a string.

How do I convert this string to a dictionary?

Thanks.




From loredo at astro.cornell.edu  Mon Feb  4 18:35:09 2002
From: loredo at astro.cornell.edu (Tom Loredo)
Date: Mon, 04 Feb 2002 18:35:09 -0500
Subject: RELEASED - StandaloneZODB 1.0 release candidate 1
References: <mailman.1011990406.22651.python-list@python.org> <3C52A2E5.6020709@mxm.dk> <mailman.1012258202.16406.python-list@python.org>
Message-ID: <3C5F1AAD.A16F6356@astro.cornell.edu>

Hi-

I am having difficulties building StandaloneZODB on a *Mac*
(Classic, not OS X), even though I was able to build ZODB
from stuff I threw together from the last official Zope release
fine.  I tried it a couple weeks ago before the official
release, and got farther than I was able to get with this
new release candidate (which appears to have a heavily
revised setup.py, and to have ignored some changes that
Guido checked in specifically to help the code clear the
Codewarrior compiler).

I'm really pressed for time currently and have had to give
up on this for the time being (esp. since I already have
something that works).  If Mac compatibility is at all a
priority for you folks, do let me know and I'll try to work
more on it if I can.  In the past I have not found much
enthusiasm for Mac compatibility in Zope land....

Cheers,
Tom Loredo


From john_coppola_r_s at yahoo.com  Mon Feb 18 15:07:43 2002
From: john_coppola_r_s at yahoo.com (john coppola)
Date: Mon, 18 Feb 2002 12:07:43 -0800 (PST)
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: <oqofim3dei.fsf@titan.progiciels-bpi.ca>
Message-ID: <20020218200743.32778.qmail@web11808.mail.yahoo.com>

--- Fran?ois Pinard <pinard at iro.umontreal.ca> wrote:
> 
> Exactement.  Bravo!  Bien dit! :-)

I agree with Fran?ois and I'm english speaking.

-john


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From starx at pacbell.net  Sat Feb  2 15:25:52 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 2 Feb 2002 20:25:52 GMT
Subject: Standing Objects ... possible?
Message-ID: <Xns91A97E73BDDEDRASXnewsDFE1@130.133.1.4>

    	I have this vague idea that this would be useful in the near 
future, and I was wondering if it was at all possible without recoding 
the interperter. What i want is a construct like:

object Foo( ancestor ):
    	pass

    	or more likely:

class Foo( ancestor, standing_mixin ):
    	pass

    	to be more or less equivelent to:

class _Foo( ancestor ):
    	pass
Foo = _Foo()

    	The second option looked easy at one point, but I don't think that 
I can make it self-initiate... In any case, is there a prefered way of 
writing hidden stuff like the _Foo for doing it the more possible way?

Thanks for any idea...

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From ryan.phillips at csus.edu  Wed Feb 13 19:07:07 2002
From: ryan.phillips at csus.edu (Ryan Phillips)
Date: 14 Feb 2002 00:07:07 GMT
Subject: Simple Question
References: <6QCa8.7212$zj7.1668243@news1.telusplanet.net>
Message-ID: <a4ev3b$o7d@dispatch.concentric.net>

There is a good discussion here that might help:
http://groups.google.com/groups?hl=en&safe=active&threadm=1012851741.725670%
40cswreg.cos.agilent.com&rnum=4&prev=/groups%3Fq%3Dpass%2Bby%2Breference%2Bp
ython%26hl%3Den%26safe%3Dactive%26selm%3D1012851741.725670%2540cswreg.cos.ag
ilent.com%26rnum%3D4

-ryan

"Tristan Barry" <tbarry at rpl-ca.com> wrote in message
news:6QCa8.7212$zj7.1668243 at news1.telusplanet.net...
> Is there a way to pass references or pointers as parameters to functions
in
> python?
>
>




From loewis at informatik.hu-berlin.de  Mon Feb 25 05:20:07 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 11:20:07 +0100
Subject: Tkinter + Entry + cp1251 cyrillic
References: <26428104.0202231657.639cec52@posting.google.com> <mailman.1014533499.3218.python-list@python.org> <m3adtz9t8y.fsf@mira.informatik.hu-berlin.de> <mailman.1014602241.11983.python-list@python.org>
Message-ID: <j4heo5n9rc.fsf@informatik.hu-berlin.de>

Kirill Simonov <kirill_simonov at mail.ru> writes:

> Yes. Tcl also checks the path "$TCL_LIBRARY/../tcl<VERSION>/".
> 
> Oh, I just noticed that you had already fixed this in cvs. Many thanks!

Actually, only after I studied your suggestion, and an earlier
explanation you had given.

Regards,
Martin


From neal at metaslash.com  Sat Feb  2 13:24:05 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Sat, 02 Feb 2002 13:24:05 -0500
Subject: ANN: PyChecker 0.8.9
Message-ID: <3C5C2EC5.3575F3E8@metaslash.com>

A new version of PyChecker is available.  This is mostly a bug fix release.

        PyChecker is a tool for finding bugs in python source code.
        It finds problems that are typically caught by a compiler for less
        dynamic languages, like C and C++.  It is similar to lint.

Comments, criticisms, new ideas, and other feedback is welcome.

Changes from 0.8.8 to 0.8.9:
  * Add -3/--properties warning when using properties with classic classes
  * Add more warnings for statements with no effect
  * Fix crash due to import module problems
  * Fix crash with nested scopes
  * Fix spurious warnings about module attributes and importing
  * Fix spurious warnings in Python 2.2 when using builtin classes (eg, socket)
  * Fix spurious warning for format string problem when using % at module scope
  * Fix spurious warning for implicit returns when doing while 1:
  * Fix spurious warning for inconsistent return types 
        when objects are subclasses
  * Don't warn about inconsistent return types from __getattr[ibute]__
  * Always assume readline module is used if input/raw_input is used

PyChecker is available on Source Forge:
    Web page:           http://pychecker.sourceforge.net/
    Project page:       http://sourceforge.net/projects/pychecker/

Neal
--
pychecker at metaslash.com



From maxm at mxm.dk  Tue Feb 26 17:22:46 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 26 Feb 2002 22:22:46 +0000
Subject: Help! IIS and Python
References: <8f8ffe67.0202131245.375fe55d@posting.google.com> <ZtAa8.211294$KM2.8800436@atlpnn01.usenetserver.com> <Xns91C1A4DAB542Djimpublishingresourc@209.249.90.101>
Message-ID: <3C7C0AB6.2090307@mxm.dk>

Jim Abrams wrote:


> If, like me, you don't have the option of moving away from IIS, you can 
> save yourself alot of headache, time, and resources by writing a nice 
> wrapper object, sticking Response inside and using your own. You can then 
> add sorely lacking functionality, like str conversion, to it as well as any 
> additional shortcuts.
> 
> I do it for most of the built ins (Response, Request, Server, Appication, 
> Session) and it's been a lifesaver.

If thats your bag, feel free to use my 'iisUtils' module.

It both has a windows version of Pythons FieldStorage, and a Dict 
version of the Request.

To get FieldStorage:

import iisUtils

fs = iisUtils.winFieldStorage(Request)

<%= fs['somevar'].value%>

This is good because it can handle file uploads. But it's a bother that 
it's not a stright dict for the rest.

So to get Request as a plain dict:

req = iisUtils.request(Request)

<%= req['somevar']%>

etc.

Its documented in the code and works nicely in production code. If there 
are _many_ fields in the form 'request' is faster than 'winFieldStorage'.

regards Max M
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: iisUtils.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20020226/42041881/attachment.ksh>

From nick at src.uchicago.edu  Mon Feb 25 23:08:48 2002
From: nick at src.uchicago.edu (Nick Collier)
Date: Tue, 26 Feb 2002 04:08:48 GMT
Subject: grammar question
Message-ID: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu>

Hi,

I've been messing around with parser generators and the python grammar as
it is described in Grammar/Grammar in the source distribution and have run
into something I don't understand. Most of my previous experience with parser
generators has been with smaller grammars and with javacc a parser
generator for Java that allows a look ahead of > 1. Anyway, I thought that
the Python grammar was LL(1) compliant and yet when I translate the
grammar into the Coco/R LL(1) parser generator I get lots of errors like:

LL(1) error in X: Y is start of more than one alternative 
LL(1) error in X: Y is start and successor of deletable structure

Where X and Y are productions such as arglist, etc. Unfortunately, I can't give
the exact errors as they are on another machine at the moment. Can anyone
shed any light on this (even in the absence of the details :) ). 

thanks,

Nick


From tim.one at home.com  Thu Feb 14 01:56:06 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 14 Feb 2002 01:56:06 -0500
Subject: Python License Issues
In-Reply-To: <ff133276.0202132007.61937647@posting.google.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIECLNMAA.tim.one@home.com>

[Rocco Moretti]
> ...
> On a somewhat tangential note, I realize that most of the PythonLabs
> team at this point says "Done is good!" with the Python license and
> wish to wash their hands of it,

The current copyright is held by the PSF, not by PythonLabs.  Nevertheless,
there's scant visible interest among the PSF membership in doing more
license haggling.

> but the ~70 line contract with the four different copyright owners and
> different licensing clauses for each makes me a little queasy, even with
> assurances that it is "GPL compatible."  (Legalese makes me nervous.)

In lieu of a court decision, most people (and reasonably so) accept what the
FSF says about which licences are and aren't GPL-compatible.  You should
read for yourself what they have to say about the various Python licenses:

    http://www.fsf.org/licenses/license-list.html

> From what I understand, the copyright owners can change the licensing
> terms at will, so if the PSF has the copyright to *all* of the code in
> Python, they could replace the four section license with a single
> section license like a BSD style license or even with (just) the
> current PSF section and a listing of the Copyright: Year-Year Holder
> list at the top.

Yes, although the PSF only has power to change its own license; it cannot
change the licenses inherited from CWI, CNRI or BeOpen.

> (Aside: Why *does* the PSF use an 8 subpoint licenseinstead of a BSD
> style? Is it strictly to maintain compatibility with the BeOpen and CNRI
> sections?)

While I personally favor an MIT-style license (BSD is way too wordy for my
tastes <wink>), I did argue at the time for the current PSF license,
precisely for its extreme similarity to the other two most recent licenses.
If you take this stuff seriously, you have to understand all of them, so the
fewer variations the better.  Most people pay no attention at all, and
that's fine by me too.

> So who actually holds the copyright for all the bits of Python?

The stack of licenses spell out the clearest answer to that you're going to
get, combined with digging through the CVS history (the CVS history is
intact from the very beginning, so every version ever released can be
reconstructed; and the licenses identify copyrights holder by release
numbers).  I expect the PSF will vote to require contribution forms in the
future, to make copyright issues slightly less muddy for new contributions
down the road.

> From what I understand, BeOpen is no longer in business.

If there was a formal bankruptcy judgment I never heard about it.  So,
AFAIK, BeOpen still exists in some limbo state.

> Has the PSF considered obtaining the copyright for the portions of code
> written at BeOpen from whoever got it from the fire sale?

As far as we know, BeOpen still has it, and of course they've been
approached about assigning copyright to the PSF.  You can guess the results
of that from observing the licenses we ship with <wink>.

> Although Stichting Mathematisch Centrum still exists, would the PSF
> consider approaching them about a transfer of the Python copyrights?

Of course.  You should realize that any legal transaction is expensive on
all sides, though, and since CWI has never hassled anyone over Python I
personally doubt they're keen to spend their own money quelling other
peoples' overblown fears.

> CNRI probably is the toughest sell, as they still exist and have
> concerns about liability.

It's worse than just that:  CNRI is a public charity (a technical term under
US tax laws), and it's said they *can't* transfer assets to an entity that
isn't also a public charity under US tax laws.  The PSF intends to qualify
as a public charity in the same sense, but so far does not enjoy that
status.

> Would it be possible to negotiate some sort of contract with them where
> the PSF assumes all liability for Python versions other than 1.6.z, and
> CRNI keeps perpetual rights to the 1.6.z codebase?

You would have to ask CNRI about that.  You should also keep in mind that
the PSF has no assets other than the current Python copyright, so the idea
that the PSF could assume liability in any meaningful sense is fantastic:
one lawsuit and the PSF is bankrupt.

> I'm not really asking about if it is legally possible, but more so to
> see if the PSF would be willing to take on such a task.

The PSF is a legal conceit, like all corporations.  In the end, you have to
find *people* with the free time, expertise, and persistence to get things
done.  Everyone is eager to volunteer other peoples' lives <wink>.




From lac at strakt.com  Fri Feb  8 08:11:26 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 08 Feb 2002 14:11:26 +0100
Subject: Defending the Python lanuage... 
In-Reply-To: Message from Cliff Wells <logiplexsoftware@earthlink.net> 
   of "Thu, 07 Feb 2002 14:07:55 PST." <20020207140755.364d5d77.logiplexsoftware@earthlink.net> 
References: <3C62EE4A.5090305@destiny.com>  <20020207140755.364d5d77.logiplexsoftware@earthlink.net> 
Message-ID: <200202081311.g18DBQGo020260@ratthing-b246.strakt.com>

> If if.  No doubt things such as personality are going to be a huge factor. 
> I guess that sums up my assertion: the type of personality attracted to
> programming is not the type to endure watching someone else code.  My
> fingers would twitch themselves numb.  To have pair-programming work you
> would need two programmers who: 
> 
> 1.  have similar skill levels

This is incorrect.  It works great with people with radically different
skill levels.  Of course, they start to converge, so they do not end
up with different skill levels.  thats the point.  For example, people
who pair program with me and are not aware of design patterns and signature 
based polymorphism will get that fixed in a hurry.  But what happens
when Alex Martelli and I are pair programming is I learn so much stuff
I never even heard about before.

It is such a joy.
> 2.  like and respect each other
> 3.  have compatible programming philosophies

(you are spot on with those, however.)

Cliff, again you are up against 'American management is impossible.'
Seriously consider visiting Sweden, for a vacation.  I think you would 
like it here.

Laura Creighton



From peter at engcorp.com  Sun Feb 24 15:37:02 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 15:37:02 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com>
Message-ID: <3C794EEE.1F25B623@engcorp.com>

Russell Nelson wrote:
> 
> Henrik Motakef <henrik.motakef at web.de> writes:
> 
> > Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> >
> > > Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> > > > Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> > > > czignat?  Classifero en bragna!
> > > Woas da Deifi wos des hoasn mog.
> >
> > Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> 
> Even better: we could switch languages within threads, to *really*
> keep the monolinguists confused.

Sed kiujn lingvojn ni devos uzi?


From niemeyer at conectiva.com  Tue Feb 19 08:44:02 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Tue, 19 Feb 2002 10:44:02 -0300
Subject: Recursively traverse linked list -- help!
In-Reply-To: <1014115178.2891.30.camel@matijek>
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org> <7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek>
Message-ID: <20020219104402.A1325@ibook.distro.conectiva>

> Sure, granted that you keep track of the middle link of your list, so
> that you don't have to traverse up to that point. But even if keep track
> of it you'll have to be doing pretty large lists before the linked list
> will outperform the builtin list.

Agreed.

It's also nice to remember why we use python in the first place. If I
had to write a linked list to keep billions of records with a very specific
and performance dependent algorithm, I would probably write a python module
in C. And if the whole application is dependent on performance, I wouldn't
use any high level language to develop it.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From phr-n2002a at nightsong.com  Thu Feb 28 01:45:31 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 22:45:31 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com> <7xk7sznsoa.fsf@ruckus.brouhaha.com> <3C7D3B2C.80A97BA2@ccvcorp.com> <7x664i8vb6.fsf@ruckus.brouhaha.com> <3C7D6994.B2D43BFA@ccvcorp.com>
Message-ID: <7xd6yqgl4k.fsf@ruckus.brouhaha.com>

Jeff Shannon <jeff at ccvcorp.com> writes:
> Ah, yes, got it.  That does make sense.  I don't see using try/finally as
> being too onerous, in cases where you *need* to have certain names cleaned up
> (which would seem a moderately rare occurence, to me), but I can see that
> being able to declare a local scope would be a bit cleaner.  I'm inclinded to
> think that the need for local-scope-inside-function-scope is not frequent
> enough to justify a language change to be cleaner than try/finally, but maybe
> it's just because I've been working in relatively straightforward application
> areas.

The idea of using a local scope is actually to avoid shadowing an
outer scope.  Of course if you KNEW you were shadowing an outer scope
in a way that could cause a bug (you're not always sure--say you want
to use "i" as a loop index) then you'd just choose a different
variable name.  So the possible relative rareness of the occurence
doesn't matter.  What matters is having to study that much more of
someone else's grotty code and possibly make a mistake, every time
you want to add a patch.  Of course it would be better to understand
everything perfectly and for the surrounding code to not be grotty,
but programming in the real world doesn't work like that.

The designers of C originally had the same idea as Python, that you
needed global and local variables but inner local scopes weren't
needed.  Eventually they saw the need to introduce local scopes in C,
so you can now put declarations introducing new variables after any
open curly brace.  C++ went further and you can put declarations
almost anywhere.


From tim.one at comcast.net  Sun Feb 17 17:26:51 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sun, 17 Feb 2002 17:26:51 -0500
Subject: Problems with telnetlib
In-Reply-To: <a4p9l1$cg4$0@216.39.172.122>
Message-ID: <LNBBLJKPBEHFEDALKOLCIECINNAA.tim.one@comcast.net>

[Bengt Richter]
> ...
> BTW, why are context diffs preferred over unified?

Because Guido says so.  Really, that's all there is to it.  In general, he
doesn't care so much about what the code used to do as about what it does
after the change, and especially for complex patches the "after" logic is
simply easier to follow if it isn't cluttered via interleaving with the
"before" logic.  OTOH, the opposite seems true more often to me when looking
at simple patches.  Someone should write a tool to generate both patch
flavors, compare them, and objectively determine which is clearer.  Indeed,
I demand someone do this at once <wink>.




From bosahv at netscapenospam.net  Wed Feb 13 04:31:53 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Wed, 13 Feb 2002 09:31:53 GMT
Subject: newbie asking for info about Python
References: <xkka8.2932$VX6.245554@weber.videotron.net>
Message-ID: <doqa8.372998$pa1.87467989@news3.rdc1.on.home.com>

Try surfing over to Sourceforge and look at the python foundry:

http://sourceforge.net/foundry/python-foundry/

or check through a few links at

http://python.org/

Thats going to give you a bigger view.

For me its a really extensible, flexible, portable interpretted language.
Its free and powerful but YMMV. Looking at whats been done with it is
probably faster than having people describe it, thats one of the bonuses of
open software.



"Joseph Youssef" <leader730 at hotmail.com> wrote in message
news:xkka8.2932$VX6.245554 at weber.videotron.net...
Hello, I know nothing about Python but before I start learning I would like
to know what Python is usually "used" for, I mean I don't think it's made to
write some kinda application or something, so what exaclty is it used for?




From max at alcyone.com  Sat Feb 23 04:18:31 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 23 Feb 2002 01:18:31 -0800
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <3C775E67.F8E7319B@alcyone.com>

Tripp Scott wrote:

> Suppose I want to make a function that modifies its argument:
> 
>   a = -1
>   make_abs(a)
>   print a # = 1
> 
> What is the Pythonic way of doing this? Wrap it in a list?
> 
>   make_abs([a])

I would make a class to do the job (note, code untested):

	class Wrapper:
	    def __init__(self, x):
	        self.x = x

	    def get(self): return self.x
	    def set(self, x): self.x = x

	def make_abs(wrapper):
	    wrapper.set(abs(wrapper.get()))

	w = Wrapper(-1)
	make_abs(w)
	print w.get()

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From emile at fenx.com  Fri Feb 15 09:16:06 2002
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 15 Feb 2002 06:16:06 -0800
Subject: certification
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>
Message-ID: <a4j5hd$l3j2$1@ID-11957.news.dfncis.de>

"Trung Hoang" <thoa0025 at mail.usyd.edu.au> wrote in message
news:2n8b8.29223$Zu6.141297 at news-server.bigpond.net.au...
> Hi,
>
> i was wondering if i can get certified with python from an
internationally
> recognised company?
>

I don't know if they're still doing it, but at one point Brainbench.com
was working on competency testing, and IIRC, gearing that towards
employers and prospective employees.

http://www.brainbench.com/

--

Emile van Sebille
emile at fenx.com

---------



From aahz at panix.com  Fri Feb  8 18:47:02 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 15:47:02 -0800
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org> <slrna689o0.cn1.akuchlin@ute.mems-exchange.org>
Message-ID: <a41o1m$8uv$1@panix2.panix.com>

In article <slrna689o0.cn1.akuchlin at ute.mems-exchange.org>,
A.M. Kuchling <akuchlin at mems-exchange.org> wrote:
>
>One visible trend at this year's conference was HTML + a Web browser
>as a presentation tool.  Skip, Jeremy and I all used HTML, and I think
>other people did too.  It's a nice trend to see, and Opera's
>full-screen mode can make the resulting presentation quite attractive.

I tried that for OSCON 2000, only to discover that the Linux version of
Netscape by default could only handle a maximum 24pt font size.  Was
anyone who was using HTML doing so under Linux?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From dgallion1 at yahoo.com  Mon Feb 11 00:07:19 2002
From: dgallion1 at yahoo.com (Darrell)
Date: Mon, 11 Feb 2002 05:07:19 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com>
Message-ID: <bkI98.61439$QG.13122338@typhoon.nyroc.rr.com>

>>> map(max, zip('ab','ba'))
['b', 'b']

--Darrell

dsavitsk wrote:

> maybe this isn't xor, but i would like something that will smush strings
> ... that is, something that would compare two strings position by position
> and return the higher (or lower i suppose) value piece in each position,
> i.e.
> 
>>>> fun('00', 'AA')
> 'AA'
> 
>>>> fun('00', 'A')
> 'A0'
> 
>>>> fun('00', '')
> '00'
> 
> i understand that this is easy to write, but i was looking for a builtin.
> does it exist?
> 
> -d



From guy at wyrdrune.com  Tue Feb  5 13:47:17 2002
From: guy at wyrdrune.com (Guy Gascoigne - Piggford)
Date: Tue, 5 Feb 2002 10:47:17 -0800
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com> <slrna602gj.ct4.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <3c6028a4$1_2@news.nwlink.com>

Option 2 looks really interesting.  I already have the reading separated off
from the actual testing, in effect the read is buffered until the client
code explicitly gobbles it.  For what I'm doing this seems to work quite
well.  Separating off the particular tests like you suggest may not be too
hard, and may make for a very readable script.

I'll give it a try.

Guy

"Albert Hofkamp" <hat at se-46.wpa.wtb.tue.nl> wrote in message
news:slrna602gj.ct4.hat at se-46.wpa.wtb.tue.nl...
> On Sun, 3 Feb 2002 00:25:14 -0800, Guy Gascoigne - Piggford
<guy at wyrdrune.com> wrote:
> >I'm trying to find a way to write something like expect in a python
module,
> >and I'm having trouble getting the syntax the way that I want.  In expect
I
> >can write something like this:
> >
> >expect {
> >    -re "ast login" {
> >           pass
> >           }
> >    -re "ogin" {
> >           send  user
> >           }
> >    -re "sword" {
> >           send password
> >           }
> >}
>
> A known problem :-)
>
> I also already noticed that the existing Expect-like thingies are not much
help
> here, they also mangle the order of the patterns and the actions.
> I have been thinking that the following should be possible:
>
> x = ExpectMatcher() # A random name, needs some more thought
> ... # do some initing stuff, start up a sub-process, connect that to x,
etc
> while not x.ismatched():
>     if x.trymatch('ast login'):
>         pass
>     elif x.trymatch('ogin'):
>         send user
>     elif x.trymatch('sword'):
>         send password
>
> There are at least 2 options for implementation (but I haven't had the
> opportunity to implement either of them).
>
> 1) (stupid way)
>    The x object reads data in the x.ismatched() method. If it has
'enough', it
>    returns false. Python then executes the x.trymatch() methods. In the
method,
>    you compare the imput with the argument. If it matches, you set a flag
that
>    x.ismatched() should return true on the next call, followed by
returning
>    true (so the if test succeeds).
>    If all x.trymatch() calls fail, Python executes the 'x.ismatched()'
again,
>    given you a chance to read more data.
>
> 2) (intelligent way)
>    The x.ismatched() returns false in the first call. Python then executes
the
>    if, and as a result, all the x.trymatch() calls. Store the patterns,
and
>    return false to all of them.
>    On the second call to x.ismatch(), you have a list of patterns, so you
can
>    wait until you have seen a pattern match.
>    At that moment, return false again, and now return false to all
x.trymatch()
>    methods, except the one that matches.
>    Python handles the code in the if branch at that point, and returns to
the
>    while.
>    On the 3rd call to x.ismatched() return true.
>
> (the above scheme should be extended to include time-out)
>
> There are probably other/better implementations possible, I haven't really
> considered alternatives except the above 2.
>
> >Am I missing something?  I'm fairly new to python but I don't see a way
to
>
> Probably not. I took me a few months and a few tricky twists to come up
with
> the above.
> Although the solution seems sound to me, I haven't really tried to
implement it.
>
>
> Albert
> --
> Constructing a computer program is like writing a painting




From spencer at efn.org  Fri Feb  1 14:23:13 2002
From: spencer at efn.org (Spencer Ernest Doidge)
Date: Fri, 1 Feb 2002 11:23:13 -0800
Subject: test - ignore
Message-ID: <Pine.GSU.4.21.0202011122010.20574-100000@garcia.efn.org>

testing To: field.

Spencer Doidge
www.spencerdoidge.com



From tim at vegeta.ath.cx  Mon Feb 11 04:06:21 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Mon, 11 Feb 2002 09:06:21 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a46uqm$gc7$1@peabody.colorado.edu> <mailman.1013399900.24051.python-list@python.org>
Message-ID: <slrna6f2ff.ugb.tim@vegeta.ath.cx>

<brueckd at tbye.com> <brueckd at tbye.com> graced us by uttering:
> On Sun, 10 Feb 2002, Fernando P?rez wrote:
>> Indeed, it's just the Spanish form of the 'ist' English suffix (pianist,
>> violinist, etc.) It simply means 'who performs an activity'.
> 
> Could be from a number of languages; I always assumed quasi-Esperanto:

This may be very likely, but the limit of my Esperanto knowledge is Red
Dward episodes...

> Of course, in either Spanish or Esperanto it would more commonly be
> "Pythonisto" anyway, so who knows.

No idea about Esperanto, but Spanish ends in -ista, even when masculine.
eg: dentist (English) --> /dentista/ (Spanish).

HTH
Tim Hammerquist
-- 
All mail clients suck. This one just sucks less.
    -- Michael Elkins on mutt


From greg at cosc.canterbury.ac.nz  Sun Feb 24 20:49:39 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 25 Feb 2002 14:49:39 +1300
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014418936.10783.python-list@python.org>
Message-ID: <3C799833.AC08A401@cosc.canterbury.ac.nz>

Gustavo Cordova wrote:
> 
> Das Computermaschine is not for your fingerpoken?
> .... blinkenlighten?
> 
> (can't remember nor find the whole thing)

>From memory, something like this:

     ACHTUNG!

Dies Komputermaschine ist nicht fur gerfingerpoken
und mittengrabben. Ist easie schnappen der Springenwerken,
mit Poppenkorken und Spitzensparken. Die rubbernecken
Seitseer ist keepen der Hans in der Pocketsrelaxen,
watchen das Blinkenleits.

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


From stede at computer.org  Sat Feb  2 11:46:50 2002
From: stede at computer.org (Stede Troisi)
Date: Sat, 02 Feb 2002 16:46:50 GMT
Subject: Question for the Python Community!
Message-ID: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>

I was thinking of doing an E-book (because I don't like the high price of
regular books) on an introduction to programming and artificial intelligence
in Python.

I was wondering if there would be interest in such a e-book and if it would
sell?  I would love to here everyone's opinion.

Thanks,
Stede




From tripps81 at yahoo.com  Tue Feb 26 12:16:58 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Wed, 27 Feb 2002 00:16:58 +0700
Subject: Implementing interfaces with inheritance, good or bad?
In-Reply-To: <5.1.0.14.2.20020225234807.0407db58@202.95.131.154>
Message-ID: <5.1.0.14.2.20020227001436.040bae80@202.95.131.154>

Hm, perhaps what I wanted to say is that, agreeing to what is 
read on the Interfaces Wiki at Zope.org: interfaces and classes 
should be separate. Both can have inheritance mechanism. So we 
have interface inheritance, and implementation (normal class) 
inheritance. However, without direct language support, an 
Interface is most easily implemented using normal python class.

t


At 25/02/2002 23:56, Tripp Scott wrote:
>I just saw:
>
>  http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/68445
>
>in which interfaces are implemented using class, and a class 
>implementing one or more interfaces subclasses the interfaces 
>class(es). Now suppose class C implements two Interfaces, I1 
>and I2. Both interfaces define an operation foo(). In one case, 
>I want to instantiate C because of I1, and in other case, 
>because I need a class that implements I2. How does C implement 
>both foo() operations (since foo() will be implemented as a 
>method)? And how do I tell C that I want I1, or I2? With an 
>argument to __init__? Any other way?
>
>I prefer providing Interface lookup via other mechanism, like 
>with the __implements__ class attribute as done in Zope's 
>Interface module. But are there any advantages of interfaces 
>being done with inheritance?
>
>t




From phd at phd.pp.ru  Fri Feb  1 05:37:23 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 1 Feb 2002 13:37:23 +0300
Subject: [Jython]ImportError: no module named dbhash
In-Reply-To: <a3dr0j$2h2v$1@rivage.news.be.easynet.net>; from bergeston@yahoo.fr on Fri, Feb 01, 2002 at 11:28:02AM +0100
References: <a3dr0j$2h2v$1@rivage.news.be.easynet.net>
Message-ID: <20020201133723.N10316@phd.pp.ru>

On Fri, Feb 01, 2002 at 11:28:02AM +0100, Bertrand Geston wrote:
> I tryed to do open a shelve file created with Python 2.1 (CPython or Cython
> or ... ? I mean the original one) in Jython 2.1 and I received this error in
>  ..\Lib\anydbm.py, line 85, in open

   dbhash is not a full-blown module, but just a wrapper for DerkeleyDB
database API. Jython evidently does not have it.

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



From maxm at mxm.dk  Mon Feb 18 06:59:45 2002
From: maxm at mxm.dk (Max M)
Date: Mon, 18 Feb 2002 11:59:45 +0000
Subject: There is more than one way to do it - and for no apparent reason.
Message-ID: <3C70ECB1.2090709@mxm.dk>

Hi

It's just a little thing that has been bugging me for a while.

I came to Python from JScript and was used to every object being both an 
object and a dictionary.

So If I want to get an the value form an objects attribute I would just 
write:

myObj = new Object()
myObj.name = 'max';
myObj.age = 36;

print myObj.name; // pretending that JScript has a print function :-)
 >>> max
print myObj.name;
 >>> 36

or

print myObj['name']; // pretending that JScript has a print function :-)
 >>> max
print myObj['name'];
 >>> 36

Basically you can say that JScript uses the dictionary notation for 
getting attributes. As fas I can see this is smarter than the way we do 
it in Python where we have several notations for the same action:

In an object:
setattr(myObj, 'name', 'max')
getattr(myObj, 'name')

In a dict:
myObj['name'] = 'max'
myObj['name']


I don't really understand why the setattr/getattr should be nessecary 
when it can be done much simpler with the dictionary notation.

ie. in Zope an object is written anyway that takes both notations, and 
really it should not be nessecary when it could just as easily be built 
into the objects.

Wouldn't it be a good idea to unify the dict and the object so that both 
could be called with attribute and dict notation like in JScript?

I have not really tried 2.2 yet so I am not aware if doing something like:

class myClass(dict, object):
     pass

will give the functionality I am after.

I know it's just syntactic sugar, but it would be so sweet.
Is there any reason why not to do it?

I have not really tried 2.2 yet so I am not aware if doing something like:

class myClass(dict, object):
     pass

will give the functionality I am after.

regards Max M



From sholden at holdenweb.com  Mon Feb 11 07:42:22 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 07:42:22 -0500
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org> <3C66F2CA.802A00BF@earthlink.net> <86lme1xdoc.fsf@apo.hmotakef.int> <a46uun$gc7$2@peabody.colorado.edu>
Message-ID: <G_O98.149110$KM2.6688952@atlpnn01.usenetserver.com>

"Fernando P?rez" <fperez528 at yahoo.com> wrote in message
news:a46uun$gc7$2 at peabody.colorado.edu...
> Henrik Motakef wrote:
>
> > This might be slightly off-topic, but usually pressing both buttons at
> > once emulates pressing the middle one...
>
> As long as you have
>
>     Option "Emulate3Buttons"
>     Option "Emulate3Timeout" "50"
>
> in your "InputDevice" section of /etc/X11/XF86Config-4
>
See how easy Linux is, Ron? ;-)

Seriously, when faced with problems of the apparent complexity of the ones
you describe here, there are two alternatives:

1) Continously simplify until you get something that works (even if it's
just a one-line script), then make it more complex again. Either it stops
working as you make it more complex, and you eventually have a head-smacking
epiphany, or it doesn't, and you are left slightly bemused as to what you
were originally doing wrong, but at least you have a working system.

2) Resort to superstition. Which was why, I presumed, you started altering
file permissions on things that looked unconnected <0.9 wink>.

Because there are so many things to do on a Linux system, there are many
things you can do wrong. Resist the urge to make things more complex to get
them working, and remember the old motto:

    First, make it work

    Then, *if* it doesn't run fast enough [or at 4.30 am <wink>],
    make it run faster [or at 4.30 am].

One last comment: many people would have edited their crontab to make the
job run at a different time instead of getting up at 4.30 to see what
happened. Although you don't say so, I presumed you chose the latter option
because you didn't want to disrupt your daily sequence. It's a useful
practice to write software so that you can run it in test mode in such a way
that normal production operations aren't disturbed by testing.

Often this will simply amount to running the tests in a different
directory -- back to the importance of environment again!

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From neal at metaslash.com  Wed Feb  6 19:45:10 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Wed, 06 Feb 2002 19:45:10 -0500
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu>
Message-ID: <3C61CE16.2E4BA60B@metaslash.com>

David Eppstein wrote:
> 
> In article <a3rnd1$1a44oq$1 at ID-11957.news.dfncis.de>,
>  "Emile van Sebille" <emile at fenx.com> wrote:
> 
> > def flatten(ary, rslt):
> >     try:
> >         flatten(ary[0], rslt)
> >         flatten(ary[1:], rslt)
> >     except:
> >         if ary: rslt.append(ary)
> >
> > rslt = []
> > flatten(("cat",5,['dog',[3,3,1]],"zoo"), rslt)
> > print 'rslt = ',tuple(rslt)
> 
> Isn't all that list-copying (ary[1:]) going to make for a very inefficient
> algorithm?

This code oughta be faster (adapted from Tim Peter's list flattening code).  
It may be faster to change map() to a list comprehension, ie, 

	''.join([str(x) for x in inlist])

def flatten(inlist, type=type):
    "Side-effect, flattens inlist"
    ltype = types.ListType
    try:
        for ind in xrange(sys.maxint):
            while type(inlist[ind]) is ltype:
                inlist[ind:ind+1] = inlist[ind]
    except IndexError:
        pass
    return ''.join(map(str, inlist))

>>> n = ["cat",5,['dog',[3,3,1]],"zoo"] 
>>> flatten(n)
'cat5dog331zoo'
>>> print n
['cat', 5, 'dog', 3, 3, 1, 'zoo']
>>>

Neal


From jjl at pobox.com  Mon Feb 18 15:25:28 2002
From: jjl at pobox.com (John J. Lee)
Date: Mon, 18 Feb 2002 20:25:28 +0000
Subject: PyKDE
In-Reply-To: <u72ll0kis0p201@corp.supernews.com>
References: <u72ll0kis0p201@corp.supernews.com>
Message-ID: <Pine.SOL.4.30.0202182011450.13673-100000@mimosa.csv.warwick.ac.uk>

On Mon, 18 Feb 2002, Mik Mifflin wrote:

> I'm trying to learn Qt programming and decided to use my new favorite
> language (Python, of course).  I'm using SuSE 7.3, and there is an rpm
> called PyQt installed, but it looks like I need PyKDE as well?  This is the
> program that I'm trying to run, and the error messages it gives:
[...]
> ImportError: No module named kdecore
>
> If I do need PyKDE, does anyone know of an rpm that will work with SuSE?
> Or am I just better off installing from source?

You don't need PyKDE if, er, you don't want to write a KDE program :-)
PyKDE is currently in a state of flux, thanks to Jim Bublitz, who's
finishing PyKDE2 just as KDE 3.0 is nearing completion (no, he's not
*that* slow, KDE 2.0 had been around a while when he started work on
PyKDE).

You clearly don't have a working PyKDE installed, but I wouldn't worry
about that yet: get Qt figured out first, then come back to KDE.  The
translation here from C++ to Python is trivial, even if you don't know
C++, so just read the Qt docs (should be in one of your rpms, will end up
somewhere like /usr/share/doc or /usr/doc), and take note of the
differences between Qt proper and PyQt (for which see the docs distributed
with PyQt itself).

Actually, plain-old PyQt programs look pretty KDE-ish (on the surface) to
me, but I haven't used KDE for a while...


John



From steve at lurking.demon.co.uk  Fri Feb  8 06:40:26 2002
From: steve at lurking.demon.co.uk (Steve Horne)
Date: Fri, 08 Feb 2002 11:40:26 +0000
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com> <f72dac0d.0202050751.68bc0a7@posting.google.com> <3j406uo6g577f6djre8jeunin6q11o8p6n@4ax.com> <f72dac0d.0202051348.4c45f4ff@posting.google.com>
Message-ID: <0db76ucfotgv339h09gcj1fc9eigfbsce8@4ax.com>

On 5 Feb 2002 13:48:22 -0800, alanmk at hotmail.com (Alan Kennedy) wrote:

>Steve Horne <steve at lurking.demon.co.uk> wrote:

>> The XML would certainly not be easy for people to edit, but it's not
>> meant to be - the presentation is an issue for the editor.
>
>Hand-editing would be a PITA. One of the xython proponents seems to be
>focussed more on the ability to autogenerate python/xython from XSLT,
>but for the life of me, I can't think of a use case for this that would
>not be better served with a dedicated code manipulation/generation tool.

Here, I agree with you except for one thing - XSLT is a standard,
which has some advantages in itself. Develop a library of 'templates'
using XML and XSLT, and presumably you are not making yourself
dependent on some single-vendor tool set.

It's not something I'd be keen to follow up, though. I've already
developed pretty much my ideal templating facility by simply
pre-processing Python scripts to add an ASP-like string literal with
substitutions and block structures. And importantly, it's not limited
to generating Python.

>> But yes, I can see problems. For instance, the error reports would
>> relate to the XML AST - not to the visible representation of the
>> source code.
>
>And don't forget that speedwise, an XML parser would be extemely unlikely
>to be faster than a dedicated parser, even just checking for well-formedness,
>let alone validity.....

Agreed - in fact, it would probably be significantly slower for the
simple reason that the input files are bulkier and contain more text
to process.

-- 
Steve Horne
steve at lurking.demon.co.uk


From python at chajadan.net  Tue Feb 26 03:52:39 2002
From: python at chajadan.net (chajadan)
Date: Tue, 26 Feb 2002 00:52:39 -0800
Subject: VPython--close causes crash!
In-Reply-To: <5.1.0.14.0.20020226001726.00a69050@63.118.78.184>
Message-ID: <5.1.0.14.0.20020226005209.00a6fb10@63.118.78.184>

It's working fine with VPython's Idle, though I haven't tried just good 
plain Idle (don't use it much). I wonder what pythonwin's prob is?

--chajadan

At 12:18 AM 2/26/02 -0800, chajadan wrote:
>I downloaded VPython, and PyGeo, but whenever I close the VPython 
>generated windows, it crashes all of Pythonwin.
>
>Maybe this is a pythonwin problem, I'll try Idle--but anyone out there 
>have any ideas?
>
>--chajadan
>
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From jeff at ccvcorp.com  Thu Feb 21 13:56:17 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 21 Feb 2002 10:56:17 -0800
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com> <3C747AE3.FD16CB33@earthlink.net>
Message-ID: <3C7542D1.7918EC87@ccvcorp.com>

Hans Nowak wrote:

> "Gon?alo Rodrigues" wrote:
>
> > 10. I have a class that works like a list in the sense that supports the
> > method __getitem__ and its relatives. But I'm at a loss as to what I
> > have to do to support the slice notation.
>
> __getitem__(self, x) will be called with a so-called slice
> object if you call your instance with the slice notation.
> Example:

IIRC, Slice objects (types.SliceType, etc) are deprecated, and objects wishing
to use slice notation should use __getslice__() and __setslice__() [and,
perhaps, __delslice__()] instead.



From sholden at holdenweb.com  Fri Feb  8 08:25:15 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 08:25:15 -0500
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com> <slrna6649e.d87.philh@comuno.freeserve.co.uk>
Message-ID: <NkQ88.105036$KM2.4008630@atlpnn01.usenetserver.com>

phil hunt" <philh at comuno.freeserve.co.uk> wrote in message
news:slrna6649e.d87.philh at comuno.freeserve.co.uk...
> On Thu, 7 Feb 2002 09:51:53 -0500, Steve Holden <sholden at holdenweb.com>
wrote:
> >"Orr, Steve" <sorr at rightnow.com> wrote in message
> >news:mailman.1012938609.3651.python-list at python.org...
> >> Python Newbie here.
> >>
> >> I need to migrate a web application from PHP to Python and I have a
> >> requirement to save state info WITHOUT using cookies where much of the
> >state
> >> data MUST not be visible to the end users. The PHP application allows
> >users
> >> to logon to one of several databases using any of a number of Oracle
> >> user/passwords. The logon info needs to be saved for queries on other
> >pages
> >> in the app. With PHP it's easy to avoid cookies by saving this info
using
> >> its session handling functions. Is there an EASY way to do this in
Python
> >> without having to adopt the Mother of all frameworks?
> >>
> >Disclaimer: I have never used PHP.
> >
> >However, *if* PHP allows you to associate requests with session state,
then
> >*either*:
> >
> >a) it does use cookies, you just don't know about them, or
> >b) it puts session references in the URL.
> >
> >There simply isn't any other available mechanism.
>
> You could go by the client's IP address. Though IMO this is a bit iffy.
>
Extremely iffy when you realise that many X terminal users may run their web
clients on the same CPU, not to mention Lynx users on timesharing terminals.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py"




From brueckd at tbye.com  Tue Feb 19 14:43:17 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 19 Feb 2002 11:43:17 -0800 (PST)
Subject: Adding static typing to Python
In-Reply-To: <slrna758ng.d1u.quinn@yak.ugcs.caltech.edu>
Message-ID: <Pine.LNX.4.33.0202191140530.5828-100000@ns0.tbye.com>

On 19 Feb 2002, Quinn Dunkan wrote:

> On 19 Feb 2002 00:59:42 -0800, Alexander Jerusalem <ajeru at vknn.org> wrote:
> >There's another thing that bugs me with the missing static typing and
> >that is that it makes code less readable. I don't know what I can do
> >with a function parameter when I see it. For example, when I see a
> >function like def computeSalary(person)... I don't know at first sight
> >if person is the id of person, a person object and if it's an object
> >what its exact class in a Person class hierarchy is. I have to find
> >the place where the constructor is called to find that out. That's not
>
> Or you can just look and see how the 'person' is used in the function.  If the
> first statement is a 'o = db.by_id[person]' then you know it's an id.  If it

An even better approach is to ask yourself: if it's a person's ID, then
why isn't the variable named personID?

-Dave




From peter at engcorp.com  Tue Feb 12 20:58:23 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 12 Feb 2002 20:58:23 -0500
Subject: Hidden files, selecting directories, & taxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
Message-ID: <3C69C83F.10F12615@engcorp.com>

Christopher wrote:
> 
> 1)  Is it possible to generate a hidden file (under Windows)? 

Simple pragmatic approach: 

import os
os.system('attrib +h %s' % filename)


From jkraska at san.rr.com  Sat Feb 16 01:16:12 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 16 Feb 2002 06:16:12 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com> <slrna67k0u.for.philh@comuno.freeserve.co.uk> <br186u8qg3bu2hpsf0dcgs1ldih7gir4rg@4ax.com> <slrna6bfie.o7n.philh@comuno.freeserve.co.uk> <jndd6ugqesuqibt2vssrfg6l368mj64mtk@4ax.com> <slrna6rhvs.9fc.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <7kur6u4e4blasv1f61elk9f3p09d11qps2@4ax.com>

>hygienic macros prevent the programmer from doing variable capture, which
>if you use it indicriminately, can cause an unholy mess, though they do so
>at the cost of incresed complexity (though you are disputing this here,
>I personally find it much easier to understand (or design) a non-hygienic
>system...

You seem to use design and understand interchangeably. They are different.

In any case, a hygienic macro behaves very similarly to a function, so it's
not all that hard to undertand. The difference is that it conceptually
mutates its arguments by implicit reference, and further, depending on
the system, can handle code itself as an argument.

>It seems like non-hygienic macros fall in the philosophy of "trust the
>programmer", which is in line with dynamic typing, signature oriented
>interfaces, etc.

One thing I'm experimenting with is making hygiene explicit. Add to the
system a first class closure, and the macro simply is or isn't hygienic
depending on whether or not a clusure is defined within it.

Alternately you could do something like

defmacro clean mymacro(...)

Where "clean" means hygienic or some such.

C//



From steve at ferg.org  Wed Feb 13 12:42:06 2002
From: steve at ferg.org (Stephen Ferg)
Date: 13 Feb 2002 09:42:06 -0800
Subject: Proceedings CD-ROM of IPC10?
References: <dinu-781536.17015113022002@news.t-online.com>
Message-ID: <b16e4ef7.0202130942.6cbbdbf0@posting.google.com>

Yes, there is.  They were distributing copies of it along with the
paper copies at the conference.


From dsavitsk at e-coli.net  Mon Feb 11 01:53:57 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 11 Feb 2002 06:53:57 GMT
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com> <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com> <vqI98.3303$pC1.669077@newssrv26.news.prodigy.com> <3c6763c6_3@corp.newsgroups.com>
Message-ID: <9UJ98.3347$pC1.674754@newssrv26.news.prodigy.com>

"Joshua Muskovitz" <joshm at taconic.net> wrote in message
news:3c6763c6_3 at corp.newsgroups.com...
> It isn't clear why you want to compare to '00', but this code should do
the
> trick nicely:
>
> >>> (r,g,b) = (56,156,12)
> >>> theString = '&H%02x%02x%02x' % (b,g,r) # note the order!
> >>> theString
> '&H0c9c38'
>

VB does BBGGRR for no reason that i understand.

> Is the test for '00' for anything other than the formatting of the string?
> Are you trying to find the brightest color, or something?

no, just formatting.

thanks for the suggestion. i had a bunch of if and len statements that
worked but didn't look like python is supposed to.


> --
> # Joshua Muskovitz
> # joshm at taconic.net
> def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
> lyyrs('Hire me!  I need the work!')
>
>
>
>
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----




From sabren at manifestation.com  Tue Feb 12 16:00:45 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 12 Feb 2002 16:00:45 -0500 (EST)
Subject: strongbox
Message-ID: <Pine.LNX.4.44.0202121559380.22800-100000@mercury.sabren.com>

[sorry if this is sent twice - i had mailman issues...]


Hey all,

This is from an idea I'm working on at:

http://www.sixthdev.com/wiki.cgi/strongbox

It concerns a scheme for somewhat strongly typed python,
including getters and setters... There's no code, but I'm
posting it here because I saw Guido's slides from the
conference, and noticed that the development team is
thinking about adding this kind of thing in a future python
version... The strongbox concept offers these things through
metaclasses.

(It also describes a scheme that uses the Observer pattern
to persist objects to various backends (relational, ZODB,
etc))


----

I call it strongbox because it's strongly typed. Python
variables are weakly typed, which means x can be a string
one minute, and an integer the next. That's great most of
the time, but it's a pain for working with relational
databases, because the columns hold only one type of
data. So my zdc applications are full of validation code
that gets repeated all over the place. Data validation is
one of the biggest things slowing me down when I go to write
a web app.


Strongbox changes everything. It starts with a metaclass
called, not surprisingly, Strongbox. When you subclass
Strongbox, you don't get a normal python class, but a
Strongbox class. Strongboxes have built-in getters and
setters. They also let you define strongly-typed slots like
so:


from strongbox import Slot 
from strongbox import Strongbox 
from pytypes import FixedPoint 
 
class Person(Strongbox): 
    fname  = Slot(str, size=15, default="herman") 
    salary = Slot(FixedPoint) 
    favorite = Slot(str, valid=('red','blue','green')) 
    age = Slot(int, valid=lambda x: x > 0) 
 
    def get_canDrink(self): 
        return self.age > 21 
 
>>> fred = Person(fname='fred', age=18) 
>>> fred.canDrink 
0 
>>> fred.age = 25 
>>> fred.canDrink 
1  


The "valid" argument lets you specify rules for valid
data. I plan to support regular expressions, enums, and
lambdas. (If you need a full function, you can just override
get_XXX).


The "pytypes" module mentioned above is also a child of
zdc... I'm taking all the datatypes defined there (Date,
DateTime, IdxDict, FixedPoint) and putting them into their
own module.


Another neat thing about Strongbox is the Observer
pattern. Notice that in the example, fred is just a
freestanding object. No database was needed. Fred doesn't
know anything about databases, but he can be saved to
one. The simplest way is to pickle him and store him in a
blob or a file. (Not 100% sure pickling works with
metaclasses, but we'll find out). But we can also store
fred's slots in a relational database.

To do that, we create a special observer called a Clerk. As
Fred changes, he fires off events notifying anyone who cares
to listen that something has happened. The Clerk listens to
these events, and can therefore decide whether or not he
needs to be saved. This means you don't actually have to
call "save" every time something changes. You *can* do that
if saving is expensive, but you can also just have the Clerk
save fred to the database.

Which database, though? It doesn't matter. Clerk doesn't
know about databases either, it knows about Sources. Source
is just an interface, and there will be different Sources
for different types of storage systems. DBAPI2Source will
map strongboxes to a database like MySQL.


With zdc, the object-relational mapping required all sorts
of work, but with strongbox, it's simple. You've defined
your Slots in python, so there's no need to investigate the
table. Mapping classes to tables can be done with a simple
python dictionary. In fact, we can even generate the SQL
statements for the tables from the class definition.


The best part of all this is that most of it is already
done. I expect Strongbox to be a fraction of the size of
zdc, and it should only take a day or two to implement

----

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------





From robin at jessikat.fsnet.co.uk  Fri Feb 15 11:17:49 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 15 Feb 2002 16:17:49 +0000
Subject: win32 DeleteObject ??
Message-ID: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk>

In Mark Hammond's excellent win32 extensions I can see how to obtain a
handle to a bitmap, but how does one remove from memory? I can't seem to
find DeleteObject.

Similarly I can't seem to find things like CreateCompatibleDC, BitBlt
etc which are part of win32. It almost seems as though I'm missing
win32gdi.pyd :)

Do I really have to lug along the whole of MFC to do simple things?
-- 
Robin Becker


From threeseas at earthlink.net  Tue Feb 26 23:13:48 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 27 Feb 2002 04:13:48 GMT
Subject: Parse variable into variables >Re: Apology
References: <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <2240.820T2407T12845765threeseas@earthlink.net> <r29j7uo1ckvhesgc7po0thcevmdg0jeukg@4ax.com> <3C7AF404.6BC628F0@cosc.canterbury.ac.nz>
Message-ID: <1121.822T529T14383507threeseas@earthlink.net>

On 25-Feb-02 21:33:40 Greg Ewing <greg at cosc.canterbury.ac.nz> wrote:
>Courageous wrote:
>>
>> >>> s = "a,b,c,d"
>> >>> tple = s.split(",")
>> >>> tple
>> ['a', 'b', 'c', 'd']

>One further tip: you can then "unpack" the tuple
>into variables, if you wish, like this:

>   var1, var2, var3, var4 = tple


testfile = open('test','r')

varset = testfile.readline()
tple = varset.split()
vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = tple

I assume this will work???

if so it Seems like this can be shortened to:

vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = varset.split()


is there a way to add the readline() into this line?

..... = testfile.readline().split()


???


A bit to tired to think ATM.

And thanks to everyone that responded.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From wookiz at hotmail.com  Sun Feb 10 03:45:07 2002
From: wookiz at hotmail.com (wooks)
Date: 10 Feb 2002 00:45:07 -0800
Subject: Perhaps I am just dumb - On Perl, Python and context...
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a44eni$2hj$1@peabody.colorado.edu>
Message-ID: <8852d33c.0202100045.35e41180@posting.google.com>

Blimey. Didn't need a PhD to follow that. Thanks. Great perspective.


From quinn at regurgitate.ugcs.caltech.edu  Fri Feb 15 21:46:21 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 16 Feb 2002 02:46:21 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com> <slrna67k0u.for.philh@comuno.freeserve.co.uk> <br186u8qg3bu2hpsf0dcgs1ldih7gir4rg@4ax.com> <slrna6bfie.o7n.philh@comuno.freeserve.co.uk> <jndd6ugqesuqibt2vssrfg6l368mj64mtk@4ax.com>
Message-ID: <slrna6rhvs.9fc.quinn@regurgitate.ugcs.caltech.edu>

On Sun, 10 Feb 2002 18:03:00 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>>So I wouldn't be able to do that macro expansion. This means the macro
>>processor is conceptually more complex, and therefore more difficult
>>to learn and probably has more bugs in it too.
>
>Potentially. However, hygienic macro systems are better known for
>introducing less comprehensional complexity. Likewise, it's usually

Another way of looking at the hygienic or not macro thing is:

hygienic macros prevent the programmer from doing variable capture, which
if you use it indicriminately, can cause an unholy mess, though they do so
at the cost of incresed complexity (though you are disputing this here,
I personally find it much easier to understand (or design) a non-hygienic
system) and reduced flexibility (sometimes you want variable capture).

Stated this way, it seems clear (to me at least) that a pythonic macro system
would be non-hygienic.  Consider:

private and protected object attributes prevent the programmer from modifying
object attrs in clients willy-nilly, which, if you do it indiscriminately,
can cause an unholy mess, though they do so at the cost of increased complexity
(new keywords, new access rules, etc.) and reduced flexibility (sometimes
you want to muck with private attributes).

It seems like non-hygienic macros fall in the philosophy of "trust the
programmer", which is in line with dynamic typing, signature oriented
interfaces, etc.


From tjreedy at home.com  Tue Feb 12 11:01:44 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 12 Feb 2002 16:01:44 GMT
Subject: XP (was Re: Defending the Python lanuage... )
References: <3C62EE4A.5090305@destiny.com> <20020208103714.11a2ae01.logiplexsoftware@earthlink.net> <200202110940.g1B9eTGo023834@ratthing-b246.strakt.com> <mailman.1013434699.26587.python-list@python.org> <h1S98.86429$A51.36785660@typhoon.southeast.rr.com> <Wg4a8.5754$k91.299578@wards>
Message-ID: <I%aa8.39028$X64.14250422@news1.rdc2.pa.home.com>

"Graham Ashton" <graz at mindless.com> wrote in message
news:Wg4a8.5754$k91.299578 at wards...
> On Mon, 11 Feb 2002 16:09:49 +0000, Thomas R. Corbin wrote:
>
> > The XP folks say that just adopting part of XP just isn't as
efficacious
> > as doing the whole thing because each piece supports the other
pieces.
> >
> >         Of course, they also say, "season to taste".
>
> Ron Jeffries says it's an ideal, which is almost never achieved:
>
>   http://www.xprogramming.com/xpmag/Misconceptions.htm
>
> From which I quote:
>
>   XP says that "all production code is written by two people sitting
>   together at one machine". That's the rule. As far as I know, no
project
>   has ever done that, including many that I consider to deserve the
name
>   XP. So what's the rule about? It's an ideal.

The recent book Extreme Programming: Applied (or something like that)
has a page on The Lone Wolf explaining how a person working alone
might apply the other parts of XP even without pair programming.

Terry J. Reedy





From janez.jere at void.si  Mon Feb 25 17:02:56 2002
From: janez.jere at void.si (Janez Jere)
Date: Mon, 25 Feb 2002 23:02:56 +0100
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com> <3c7a278a.13899093@news.isar.de>
Message-ID: <3C7AB490.2070703@void.si>

> 
> # results:
> 0.518367439587 (your)
> 0.228796436055 (mine)
> -------------------------------------------------------
> 
> But, your version seems reasonably fast already :). My "Full-Blown
> RPCServer" that uses dedicated threads for function calls (so that one
> RPC call can issue another RPC call while not blocking other threads
> making their RPC calls etc) is *a lot* slower...
> 
> 
As you noticed my code is not a real word example, just a test of small
serialization part of evolving IPC mechanism, current performance is not
too bad, so I could use the code. But every overhead is annoying, and
you just showed how to cut half of the overhead. And is is always a nice
time when losing half of the overhead :)

Janez









From henrik.motakef at web.de  Fri Feb 15 14:03:53 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 15 Feb 2002 20:03:53 +0100
Subject: Basic Question
References: <m3y9huzd3a.fsf@mycomp.tky.hut.fi>
Message-ID: <86vgcyworq.fsf@apo.hmotakef.int>

Karthikesh Raju <karthik at james.hut.fi> writes:

> import random
> 
> def randomList(n):
>     s = [0] * n
>     for i in range(n):
>         s[i] = random.random()
>     return s
> 
> in a file called randomList.py. Started python, imported randomList
> (>> import randomList) and then called this function (>>
> randomList(5)), got this un-understandable error:
> 
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: object of type 'module' is not callable

With the plain "import", you just loaded the module called randomList
(after the file it lives in). This module coutains the function you
def's. So you either have to write

>>> import randomList
>>> randomList.randomList(5)

or

>>> from randomList import randomList
>>> randomList(5)

hth
Henrik


From dsh8290 at rit.edu  Wed Feb  6 00:09:23 2002
From: dsh8290 at rit.edu (dman)
Date: Wed, 6 Feb 2002 00:09:23 -0500
Subject: (Long) Re: Autocoding project proposal.
In-Reply-To: <4898.802T2975T395701threeseas@earthlink.net>
References: <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net>
Message-ID: <20020206050923.GB12156@dman.ddts.net>

On Wed, Feb 06, 2002 at 04:54:44AM +0000, Timothy Rue wrote:

| We both know it can be done, and in more than one way. At this stage of
| development, that is all that is needed. Once the tool is built then we
| can focus on getting the tool to output what we want from the input we
| provide it with.
| 
| In either case, more information is needed and that's what we work on
| after the tool is created.

Alright then, stop flooding this forum with noise and build the tool
so we can start to focus.  Once you actually build the tool (that no
one else comprehends yet due to lack of communication) then the rest
of the world can see if you're more than just talk.

-D

-- 

I can do all things through Christ who strengthens me.
        Philippians 4:13




From gh_pythonlist at gmx.de  Tue Feb  5 15:00:55 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 5 Feb 2002 21:00:55 +0100
Subject: distutils, C extensions, Borland compiler, wxPython
In-Reply-To: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk>
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk>
Message-ID: <20020205200054.GA6560@lilith.hqd-internal>

Le 05/02/02 ? 15:07, Anders J. Munch ?crivit:
> Hi,
> 
> I have built Python from source on Win2K using cygwin except with a
> Borland compiler (BCB4) instead of gcc.  It took a fair amount of
> hacks to get the configure script and makefiles to work (s/.o/.obj/ is
> just the beginning) but it was worth it.

I' currently doing something similar with mingw32 instead of Cygwin :-)

The real question is if you can use dynamically loaded extension modules
at all if you have all of the Python core in a python.exe.

I think I have heard something to the effect, that on Windows, DLLs
can't link back to .EXEs, only to other DLLs.

Which would mean that you can either build a python{major}{minor}.dll or
you compile every single extension module (i. e. all of wxPython, for
example) into one big python.exe.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.3 ?C      Wind: 3.0 m/s



From calves at coelce.com.br  Sat Feb 23 08:50:02 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Sat, 23 Feb 2002 10:50:02 -0300
Subject: RES: While loop and time.sleep function
Message-ID: <29A97D00F387D411AC7900902770E148040DFAC4@lcoeexc01.coelce.net>

Thanks for your help!
But, for me, it only works properly under interactive sections (don't forget
I'm under Windows ;-)).
So, to make it works as I desire (accepting 'double-click' and 'enter' to
launching application) I just put two small lines in your code(see below). 
Now, the code is running just fine.
> -----Mensagem original-----
> De: Eric Brunel [mailto:eric.brunel at pragmadev.com]
> Enviada em: quinta-feira, 21 de fevereiro de 2002 16:42
> Assunto: Re: While loop and time.sleep function
> 
> 
> Hi,
> 
> Alves, Carlos Alberto - Coelce wrote:
> > Hi all,
> > I tray to make a simple clock program ( see the code below ).
> > Can someone explain to me why code 01 works while code 02 doesn't
> [snip]
> 
> Eeer, in fact both codes didn't work... At least on my Linux 
> box. The first 
> problem is that there is no mainloop, so Tk never actually 
> starts. The 
> second problem is that since the control never returns to Tk 
> (because of 
> the "while 1"), Tk never gets a chance to refresh the 
> display. So you must 
> explicitely refresh it manually:
> 
> -----------------------
> from Tkinter import *
> import time
> 
> class Clock:
>   def __init__(self,root):
>     self.lb=Label(root,padx=5,pady=5,fg='blue',font=('Times',20))
>     self.lb.pack()
>     self.ck()
>     self.update()
>   def ck(self):
>     self.lb.configure(text=time.asctime()[11:19])
>     # Refresh the display; otherwise, nothing happens...
>     self.lb.update_idletasks()
>   def update(self):
>     while 1:
>       self.ck()
>       time.sleep(1)
> 
> if __name__=='__main__':
>   root=Tk()
>   root.title('Clock')

    root.update()	

>   Clock(root)

    root.update()

>   # Run Tk main loop
>   root.mainloop()
> -------------------------------
> 
> HTH
>  - eric -
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020223/6f8095b7/attachment.html>

From sholden at holdenweb.com  Mon Feb  4 08:43:00 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 08:43:00 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl>
Message-ID: <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com>

"Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
[editor stuff]
>
> I suppose it might be a brain thing. I use my left hand for writing and
> drawing while I use my right hand for more coarse-grained tasks, like
> throwing a ball and such.

Strange. I don't come across many other people who do this. I've always
thought my brain must have been differently wired for some reason I'll never
know.

> I wouldn't count myself as ambidextrous though.
>
"I'd give my right arm to be ambidextrous".

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From mark at mceahern.com  Wed Feb  6 22:40:34 2002
From: mark at mceahern.com (Mark McEahern)
Date: Wed, 6 Feb 2002 19:40:34 -0800
Subject: money class?
In-Reply-To: <000001c1af86$d050f6d0$445d4540@Dell2>
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEEPKBAA.mark@mceahern.com>

> From: Brian Quinlan [mailto:brian at sweetapp.com]
> Are you looking for any class that happens to be named "Money" or are
> you looking for some particular functionality?

Dang, and I thought I could get away with hoping someone on the list would
read my mind.  ;-)

Consider this:

  #! /usr/bin/env python
  # money.py
  cost = 5.99
  percentDiscount = 10
  months = 3
  subTotal = cost * 3
  discount = subTotal * (percentDiscount * 1.0) / 100
  total = subTotal - discount
  expectedTotal = 16.17
  print total == expectedTotal

I want something that will just do the right thing and return true here.
Now that I've gone through the trouble of demonstrating why I can't simply
rely on float, I realize I can turn this into a test case and I'll be
halfway to having my own class.

I think the Money class is basically going to end up overriding special
methods like __mul__ to make sure that percentage calculations return the
expected result.  Or, maybe it should look like this:

	cost = Money(5.99)
	discount = cost.percent(10)
	print isinstance(discount, Money)
	1

Sorry to continue to be vague.  If I end up writing a money class, I'll
share it with the list so that everyone can see how dumb I am.  ;-)

Cheers,

// mark




From jon+usenet at unequivocal.co.uk  Tue Feb 26 01:10:45 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Tue, 26 Feb 2002 06:10:45 -0000
Subject: help: Problem with cgi form
References: <Xns91BE6246576BAAusYourStandingInIt@130.133.1.4> <3C7AAF31.8040201@mxm.dk> <slrna7lgm7.1kf.jon+usenet@snowy.squish.net> <3C7AD506.3000906@mxm.dk>
Message-ID: <slrna7m9n4.nj.jon+usenet@snowy.squish.net>

In article <3C7AD506.3000906 at mxm.dk>, Max M wrote:
>> Beware that, in general, that is a bug. You are open to Cross Site
>> Scripting attacks. Even in the absence of these, if the value happens
>> to contain characters significant to HTML (such as '"', '&', '>', etc)
>> then your program will go wrong.
> 
> Yeah ok ... I thought that there was a limit to how much should be 
> explained in one post :-) But here goes:
> 
> from cgi import escape
> 
> '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % escape(last_byte_pos)

That is also a bug ;-). cgi.escape does not escape single or double
quotes by default, and indeed never escapes single quotes.

> from xml.sax.saxutils import quoteattr
> 
> '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % quoteattr(last_byte_pos)

This is also a bug ;-). quoteattr outputs the quotes as well as the
value.

The Python library functions do seem to be trying to make somewhat of
a mountain out of this trivial task. cgi.escape would work in all
situations and would not break backwards compatibility if it simply
always escaped single and double quotes.


From hfoffani at yahoo.com  Wed Feb 13 07:26:17 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Wed, 13 Feb 2002 13:26:17 +0100
Subject: photo id ?!?!
References: <mailman.1013549360.16845.python-list@python.org>
Message-ID: <20020213072631.181$5Z@news.newsreader.com>

"Andrew Dalke" escribi? ...
>
> ... about photo-id requierement...
>

In all conferences I assisted, in Latin America and
Europe, goverment issued ids were requiered only to
check credit card ownership. When I had the registration
prepaid a business card was enough. Actually it was
prefered because they sell (or give to expositors) the
info afterwards.

The worst case I had happened me a few day ago.
easycar.com wanted me to show before picking up the car:
- Valid credit card
- Goverment issued id (with photo)
- Driver license
- Any form (like telephone bill) where I can prove the
address given, if differs from what is stated in the id.
- And they even took a photo of myself with a web cam!
All that for a Class A Mercedes! Come on! Now, if I
were to rent a Ferrari I'd ask myself to do a blood
test, ADN and retine scans! :-)

I agree with Andrew that when paying cash at the desk
or in advanced registration a business card should
suffice.

Regards,
-Hernan








From sab at NOSPAM.freeuk.com  Mon Feb  4 11:26:42 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Mon, 4 Feb 2002 16:26:42 -0000
Subject: how can i get the screen width and height in pixels?
Message-ID: <a3mclk$pjh$1@news5.svr.pol.co.uk>

how can i get the screen width and height in pixels?

ta,

G. Willoughby




From jason at jorendorff.com  Mon Feb  4 20:00:23 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 19:00:23 -0600
Subject: Jython question concerning actionPerformed
In-Reply-To: <f6b73ec8.0202040425.4c3e0bb4@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEFGDAAA.jason@jorendorff.com>

Mishre wrote:
> self.catCombo = \
>     javax.swing.JComboBox(actionPerformed=self.actionPerformed)

Unfortunately JComboBox has a public method named actionPerformed.
Jython apparently thinks you're trying to assign to that.

Alas, it seems you'll have to use self.catCombo.addActionListener()
just like an ordinary Java programmer.  (sigh)


> I've also tried this with the actionPerformed function declared
> as a class inheriting java.awt.event.ActionListener, which works
> a little better..but then I can't manipulate the self.catCombo
> item in the main function (or at least if I can I don't know how).

Easy: when you create the ActionListener instance, assign an
attribute on it that refers to the JComboBox object.

## Jason Orendorff    http://www.jorendorff.com/



From martin at v.loewis.de  Sun Feb 24 03:25:03 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 24 Feb 2002 09:25:03 +0100
Subject: Is there a python parser generator that is similar to JavaCC from webgain?
References: <NuRd8.8978$pC1.1550429@newssrv26.news.prodigy.com> <m38z9jpz4a.fsf@mira.informatik.hu-berlin.de> <B89D9263.3864%pereira@cis.upenn.edu>
Message-ID: <m3it8n9ti8.fsf@mira.informatik.hu-berlin.de>

Fernando Pereira <pereira at cis.upenn.edu> writes:

> JavaCC takes grammar specs (lexical patterns and rules) and produces a
> parser (a Java class) using a LL(k)-type of rule compilation (that is,
> deterministic recursive descent with lookahead).

Ok, I guess the closest thing for Python like this is YAPPS:

http://theory.stanford.edu/~amitp/Yapps/

Regards,
Martin


From tdelaney at avaya.com  Tue Feb 26 23:40:39 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 27 Feb 2002 15:40:39 +1100
Subject: Parse variable into variables >Re: Apology
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1B6@natasha.auslabs.avaya.com>

> From: Jason Orendorff [mailto:jason at jorendorff.com]
> 
> Timothy Rue wrote:
> > vic['AI']['1'], vic['AI']['2'], vic['AI']['3'] = varset.split()
> >
> > is there a way to add the readline() into this line?
> > ..... = testfile.readline().split()
> 
> All of this is correct.  You might even want to just do:
> 
>   vic['AI'] = testfile.readline().split()
> 
> but of course that makes vic['AI'] a list of strings,
> not a dictionary.  Which may or may not be okay.

However, it can also be confusing.

The way I would do it would be (using a better indexed() though - generators
:)

def indexed (seq, start=0):
    return map(None, range(start, len(seq) + start), seq)

line = testfile.readline()
tokens = line.split()

for i, token in indexed(tokens, 1):
    vic['AI'][str(i)] = token

This fits your first line, but in a more general (it allows an unknown
number of tokens) and much easier-to-read way. Of course, whether you really
want a dictionary of dictionaries I don't know ... (I suspect you actually
want a list of tokens, in which case you should just put tokens into your
vic dictionary).

Always go for the most readable way first - you will appreciate it when you
have to go back and work out what you have done.

As you can see, as soon as you start showing that you've done some work,
people become much more responsive.

Tim Delaney



From aahz at panix.com  Fri Feb  1 10:37:46 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 07:37:46 -0800
Subject: An improvement to Queue
References: <Xns91A793581DD79michaelrcpcouk@194.238.50.13> <Xns91A7A14C232EBmichaelrcpcouk@194.238.50.13> <a3ccif$rls$1@panix2.panix.com> <Xns91A85CFFB7ED2michaelrcpcouk@194.238.50.13>
Message-ID: <a3ecoa$rpe$1@panix2.panix.com>

In article <Xns91A85CFFB7ED2michaelrcpcouk at 194.238.50.13>,
Michael Abbott  <michael at rcp.co.uk> wrote:
>aahz at panix.com (Aahz Maruch) wrote in news:a3ccif$rls$1 at panix2.panix.com:
>>
>> What I still don't get is why you're waiting on an event rather than on
>> data.  It seems pointless.  Just have the worker thread grab data and
>> process it.
> 
>The worker thread has other things on its mind: when it does a job, it 
>wants the whole job fully set up and ready to go.

You are contradicting yourself.  In another post, you say:

>I'm not putting incomplete commands in the queue.  However, the consumer is
>in a position to merge multiple commands together, and I'd like it to have
>the option.

>From the programming point of view, if you're sending an incomplete job
to the worker thread, that's an incomplete command.  Accumulate enough
work in an intermediary pipeline to make a whole job.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From martin at v.loewis.de  Wed Feb 20 02:15:45 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 20 Feb 2002 08:15:45 +0100
Subject: importing dlls with python
References: <e6bacef5.0202192029.2d095cc8@posting.google.com>
Message-ID: <m34rkc62u6.fsf@mira.informatik.hu-berlin.de>

tkeon0 at netscape.net (Trevor Keon) writes:

> Hi, I am new to Python, and was wondering if it was possible to import
> dlls written in Delphi 5?  I have tried a number of methods to make it
> work, and I either get:

Depends on what you are trying to do. If you want to use the DLL
as-is, it will be tricky. If you can arrange for the DLL to look like
a Python module, it will interact much better.

> function initxxx: Integer;  StdCall;
>   export; 

If you want to implement a Python module as a DLL, you need to follow
the Python/C API, see

  http://www.python.org/doc/current/ext/ext.html

In particular, this means that you need to call InitModule4 inside
initxxx, and that initxxx needs to get a proper signature.

If you want to call an arbitrary DLL (be it compiled by Delphi or by
other means), you can try the CallDLL package. A 2.1 version of
calldll is available from

http://www.nightmare.com/software.html

Regards,
Martin



From fredrik at pythonware.com  Tue Feb  5 02:47:55 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 05 Feb 2002 07:47:55 GMT
Subject: how can i get the screen width and height in pixels?
References: <a3mclk$pjh$1@news5.svr.pol.co.uk> <3c5ec34b.27704203@news.t-online.de> <1965188.y9P21nC1DU@borg.pp.se> <a3n3vi$706$1@newsg3.svr.pol.co.uk>
Message-ID: <L6M78.23303$l93.4247163@newsb.telia.net>

G. Willoughby wrote:

> I'm centering a Tkinter app in any given monitor res.  :)

should have mentioned that in your first post...

in Tkinter, for any widget "w" on any platform, do:

    width = w.winfo_screenwidth()
    height = w.winfo_screenheight()

</F>




From jicondon at us.ibm.com  Mon Feb 25 13:57:07 2002
From: jicondon at us.ibm.com (dW)
Date: 25 Feb 2002 10:57:07 -0800
Subject: Intro to PYX - developerWorks article
Message-ID: <dc128977.0202251057.57f8ee63@posting.google.com>

The following article just came out on developerWorks.  Also are some
Python/Web Services articles.
Jeff

Find out how the open-source PYX format can make working with XML even
easier. Here, David Mertz introduces PYX and provides multi-language
code samples.
http://www-106.ibm.com/developerworks/xml/library/x-matters17.html?open&l=968,t=gr

Python SOAP libraries 
Web services columnists Mike Olson and Uche Ogbuji discuss the various
SOAP implementations available for Python, giving detailed code
examples.
http://www-106.ibm.com/developerworks/library/ws-pyth5?open&l=968,t=gr

More on XML in Python: 4Suite 
http://www-106.ibm.com/developerworks/xml/library/x-matters16.html?open&l=968,t=gr


From sholden at holdenweb.com  Fri Feb  1 14:17:39 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 1 Feb 2002 14:17:39 -0500
Subject: "Private" attributes - a possible idea (maybe PEP-worthy).
References: <B43D149A9AB2D411971300B0D03D7E8BF0A154@natasha.auslabs.avaya.com> <mailman.1012522682.17576.python-list@python.org>
Message-ID: <PTB68.107937$8b.4076617@atlpnn01.usenetserver.com>

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
news:mailman.1012522682.17576.python-list at python.org...
> On Fri, 1 Feb 2002 10:40:45 +1100
> Delaney, Timothy wrote:
>
> > I'd like to close this thread here (which I created because I thought
I'd
> > come up with a way to get around the wart) unless someone *can* come up
> with
> > a way to deal with it.
>
> With Python you can start a thread, but you can't stop it.  Sorry.  You'll
> have to wait until reaches the end of execution.

So, just the same as c.l.py, then?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From bergeston at yahoo.fr  Fri Feb  1 05:28:02 2002
From: bergeston at yahoo.fr (Bertrand Geston)
Date: Fri, 1 Feb 2002 11:28:02 +0100
Subject: [Jython]ImportError: no module named dbhash
Message-ID: <a3dr0j$2h2v$1@rivage.news.be.easynet.net>

Hello,

I tryed to do open a shelve file created with Python 2.1 (CPython or Cython
or ... ? I mean the original one) in Jython 2.1 and I received this error in
 ..\Lib\anydbm.py, line 85, in open

If somebody has an explanation (bug ? specification, todo ? ...) and even
better a workaround, I will be happy.

Anyway, have a nice day.

B.




From gdr+google at ravenbrook.com  Wed Feb 20 10:41:52 2002
From: gdr+google at ravenbrook.com (Gareth Rees)
Date: 20 Feb 2002 07:41:52 -0800
Subject: Mystery Guest from Number Theory
References: <a501qd$1c3k$1@news.idiom.com>
Message-ID: <71b62b4f.0202200741.7ff3cc9f@posting.google.com>

jimd at vega.starshine.org (Jim Dennis) wrote in message news:<a501qd$1c3k$1 at news.idiom.com>...  
>  So, as odd as it sounds, I'd like to post the code that implements  
>  this here and see if anyone can name our mystery guest:  
>   
>  This should be pretty readable, for any whole number, if it's even   
>  cut it in half, otherwise multiply by three and add one.  Iterate until   
>  you get to one.  
  
This has many names, but is best known as the "Collatz Problem" or just
as the "3x+1" problem.  The Collatz Problem is to prove that every positive
integer eventually leads to 1 under the iteration of the function

    x -> 3x+1 (x odd)
    x -> x/2 (x even).  
  
The problem is unsolved.
See <http://mathworld.wolfram.com/CollatzProblem.html>.


From gerson.kurz at t-online.de  Thu Feb 21 14:04:09 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 19:04:09 GMT
Subject: newbie floating point question
References: <pan.2002.02.21.12.47.20.378591.4031@R191.net>
Message-ID: <3c754438.21854687@news.t-online.de>

On Thu, 21 Feb 2002 19:47:20 GMT, Alex <alexhelder at R191.net> wrote:

>Why does this happen ?
>
>>>> .2
>0.20000000000000001
>
>Is this the correct behavior for the Python 2.2 interpreter ? (version info
>is printed below )
>
Yes. For a very good explanation see

http://www.python.org/doc/current/tut/node14.html

It also explains the difference between str(float) and repr(float),
which I found out the hard way ;)



From jason at jorendorff.com  Thu Feb 21 00:56:06 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 23:56:06 -0600
Subject: A bundle of questions from a Python newbie
In-Reply-To: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEDLDEAA.jason@jorendorff.com>

> 2. What is the idea of having in the loop
> 
> for var in range(10):
>     pass
> 
> the var variable still available after it?

Because it's useful, sometimes.

(It bugs me, too, from an aesthetic standpoint.)


[Concerning "self":]
> I don't expect that this solution is adopted (it would break all the
> existing code - ahhh every Nero-like language designer's dream - wreak
> havoc in the citadel of its users and start afresh), what I would like
> to know is how bad/good is this idea.

It's not such a bad idea.  I think Ruby does something like this,
with its @tributes.  Of course, as you say, it's obviously too late for
Python.

Besides, it is said that the explicit "self" makes Python easier to
understand for newbies.  And it is known for sure that Guido, the
language inventor, prefers it this way.


> 6. Is there a reason why 3.__abs__() does not work?

Yes, "3." is a floating-point number.  If you use extra parentheses,
it works:

  (3).__abs__()


> 8. Can the id() of an object change? e.g. via the garbage collector
> moving it around?

Nope.


> 9. What would be better in terms of speed?
> 
> for i in range(<whatever>):

This is much faster than the other one, for lists smaller than
a million elements or so.  Got RAM?

Your iterator object could be replaced by xrange(), or by a simple
generator.  It'd be faster.  But still probably not as fast as range().


> 12. This is not exactly about Python, but about programming in general.
> I keep reading about continuations and coroutines and I would like to
> know, in simple terms if possible, what the heck are they and in what
> ways they differ from generators as they exist in Python.

This is just my understanding of it, and I may be completely wrong.

"The stack" consists of:
  * All the same information that gets printed in a Python traceback,
    namely which functions have been called and in what order, and
    what line of code is currently running;
  * AND, all the local variables for all those functions.

(The inspect module can tell you more about this.)

A "continuation" is just a special kind object.
When you create one, it saves a copy of the whole stack.

Later you can "call" the continuation; what happens is kind of like
a massive undo.  The current stack gets clobbered.  The old stack,
as saved by the continuation, gets restored.

One consequence of this is, no matter what function you're in now,
that part of the stack gets clobbered too, so your program resumes
from the point where the continuation was created!

Sound weird?  Yeah.

Sound like generators?  Not really.  But continuations are the basis
for many other varieties of oddness, and generators are one of the
things that come out of that.

So are coroutines; but I'll leave that one for somebody else. ;)

## Jason Orendorff    http://www.jorendorff.com/




From googlegroups at sixtyten.org  Tue Feb 12 06:38:27 2002
From: googlegroups at sixtyten.org (Allan Crooks)
Date: 12 Feb 2002 03:38:27 -0800
Subject: win32evtlog
Message-ID: <304d20df.0202120338.371ae9fd@posting.google.com>

Hi,

I understand that the win32evtlog.pyd file is also a dll file
containing message strings which you can use.

However, I'm not able to supply an eventID that works (to get the
message string). Can someone suggest some values which might work?

Allan.


From sarnold at arnolds.dhs.org  Sun Feb 17 10:53:33 2002
From: sarnold at arnolds.dhs.org (Stephen L Arnold)
Date: Sun, 17 Feb 2002 07:53:33 -0800
Subject: high precision mathematics
In-Reply-To: <3C6FC7C8.7060409@hccnet.nl>
Message-ID: <20020217155253.981461FAFF@shiva.arnolds.bogus>

On 17 Feb 02, at 15:10, Jaap Spies wrote:

> I.J. wrote:
> 
> > Is there a module for python that grants ussage of decimal numbers with more
> > than 14 decimal places, including all scientific functions to do with those
> > numbers?
> 
> Take a look at
> 	http://cens.ioc.ee/projects/pyginac/

Having just browsed the pyginac page for the first time, I'm 
probably just flapping my gums, but why would you want a symbolic 
math package for high-precision math?  I would think a python 
wrapper for GMP would do it.  If such a wrapper already exists then 
you're good to go.  Otherwise, go get swig and generate one.  The 
Pyginac thing looks cool (if unfinished) but why use it if you 
don't have to?

Just a thought, Steve

*************************************************************
Steve Arnold                      http://arnolds.dhs.org:8080

Java is for staying up late while you program in Python...



From magnus at thinkware.se  Wed Feb 20 15:06:10 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 12:06:10 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <roy-05958C.07303117022002@reader2.panix.com> <ubsenehjm.fsf@online.no> <a4psgi$5jo$3@peabody.colorado.edu> <a4racj$bm2$1@panix2.panix.com>
Message-ID: <258fd9b8.0202201206.64b94d0f@posting.google.com>

aahz at panix.com (Aahz Maruch) wrote in message news:<a4racj$bm2$1 at panix2.panix.com>...

> (The above sarcasm directed at the glass house of someone who uses
> something other than ASCII as the $NAME.)

Well, that might be because ASCII can't be used to fully
express his name. English, on the other hand, can probably
be used to express the messages people really want to convey
in their posts.

I have one of those names as well. My options include stating
a false name which foreigners will pronounce as badly as my
proper name, but which will make Scandinavians think my name
is different than it really is, or to write my proper name
and either get strange MIME thingies in mail headers which
will confuse some people and break some software, or just send
8 bit ANSI / ISO8859-1 and break other pieces of software, but
display correctly for most people. Or should I attach a gif?

(Actually, a few posts (I won't mention any names since I've
suppressed them) are written in such a poor English that one
would prefer that they were written in a language one didn't
understand at all. But that was probably some other mailing
list...)

/Magnus Lyck&aring; or Magnus Lyck\aa or whatever... (I should
probably write Lucko to make English speaking people get a bit
closer to the correct pronounciation, but no, they'd pronounce
it Lacko then. I think I'd need to attach a wav...)


From emile at fenx.com  Sun Feb 17 14:23:11 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 17 Feb 2002 11:23:11 -0800
Subject: __class__ for 2.2 types
References: <2ec1bc1c.0202071240.44185531@posting.google.com> <a3upaq$1arn1e$1@ID-59885.news.dfncis.de> <a4otl1$44d$1@news.idiom.com>
Message-ID: <a4p0ep$1ptki$1@ID-11957.news.dfncis.de>

"Jim Dennis"
>  That's an interesting "gotchya"  I understand why 2.__class__ is a
>  syntax error (ambiguous with a float: 2.0 etc).  But I'm surprised
>  that 2 . __class__ works; the spaces seem unnatural here.
>

I was surprised by this recently as well.  It turns out the '.' is
simply a delimiter, and whitespace is allowed either side of a
delimiter.

>>> sys . path . pop ( )
'F:\\Python22\\lib\\site-packages'
>>>


--

Emile van Sebille
emile at fenx.com

---------



From wayne at mishre.com  Tue Feb 12 15:18:14 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Tue, 12 Feb 2002 12:18:14 -0800
Subject: catching error output
References: <a4br6c$lt1$1@news-int.gatech.edu>
Message-ID: <3C697886.8080408@mishre.com>

Joseph,

   Have you tried one of the other versions[1] of os.popen?  The other 
versions don't return a status code, but they do let you read the stdin, 
stderr and stdout.

   You might try one of the other versions, then read() from stderr, 
inside a try block, to see if there was an error or not.  If there was 
no error, read from the stdout.

[1] http://www.python.org/doc/current/lib/os-newstreams.html

Joseph Holland King wrote:
> in a small script i am using the popen command to run a command 
> and then parse the output:
> fs = os.popen('foo some_data')
> fs.readline()
> ...
> however when foo fails i need to catch that error and be able to parse
> it. is there anyway to do this? fs = .. does not catch it just the 
> normal output. thanks.
> 
> --
> Holland King        
> gte743n at cad.gatech.edu
> 


-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From kdahlhaus at yahoo.com  Thu Feb 21 09:31:31 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 21 Feb 2002 06:31:31 -0800
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com> <mailman.1014250235.21771.python-list@python.org>
Message-ID: <283adf56.0202210631.f083922@posting.google.com>

That's actually not such a bad idea.  I'm only paying $10/month for
dialup, and looking to pay about the same for hosting.   Does DSL
require you to continue phone service once it is installed?  That
would make up the difference and it would be affordable then.

> With the money you pay to ISP, why don't you get DSL with static IP, and
> run your own server?


From amuys at shortech.com.au  Wed Feb 27 18:47:49 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 27 Feb 2002 15:47:49 -0800
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com> <a5jbra$7ok00$1@ID-11957.news.dfncis.de>
Message-ID: <7934d084.0202271547.60ab7365@posting.google.com>

"Emile van Sebille" <emile at fenx.com> wrote in message news:<a5jbra$7ok00$1 at ID-11957.news.dfncis.de>...
> "Michael P. Soulier" <msoulier at nortelnetworks.com_.nospam> wrote in
> message news:slrna7q9m0.4pb.msoulier at pmerd071.ca.nortel.com...
> >     Hello.
> >
> >     While I know about bound methods in Python, I'm wondering if a
>  full
> > closure can be implemented. Can it?
> >
> >     Example: I want to pass an arbitrary code reference into a closure
>  and
> > have all of its variables preserved.
> >
> >     ie.
> >
> >     print "the current count is %d" % counter; counter += 1
> >
> >     Can I pass the above code into a closure, and then simply call it
> > repeatedly and see counter increment itself? If I pass in a function
>  with a
> > local variable, that local will be reset each time, and I don't want
>  to use a
> > global.
> >
> 
> I don't understand closures, but I might try something like:
> 
> class Counter:
>     def __init__(self):
>         self.count = 0
>     def __int__(self):
>         self.count += 1
>         return self.count
> 
> counter = Counter()
> print "the current count is %d" % counter
> 

That works, as does using a generator:

def Counter():
    counter = 0
    while 1:
        print "the current count is %d % counter
        counter += 1
        yield counter - 1   # might as well return it :)

counter = Counter()
counter.next()
 - the current counter is 0
 - 0
counter.next()
 - the current counter is 1
 - 1

Andrae


From mlh at vier.idi.ntnu.no  Tue Feb 12 12:15:37 2002
From: mlh at vier.idi.ntnu.no (Magnus Lie Hetland)
Date: Tue, 12 Feb 2002 17:15:37 +0000 (UTC)
Subject: New PEP (281): Loop Counter Iteration with range and xrange
References: <mailman.1013470220.15221.python-list@python.org>
Message-ID: <slrna6ijdp.cik.mlh@vier.idi.ntnu.no>

In article <mailman.1013470220.15221.python-list at python.org>, Delaney,
Timothy wrote:
>> The PEP is available at:
>> 
>>   http://www.python.org/peps/pep-0281.html
>
>I've proposed this in the past and was told by the timbot IIRC that it
>wouldn't get off the ground.

Oh... Oh, well. At least now we'll get the rejection documented (which
is, I guess, part of the PEP process.)

Anyway, I've noticed there are quite a lot of PEPs dealing with the
issue of indexed iteration... I only hope we get some solution, almost
no matter which :)

>Tim Delaney
>

--
Magnus Lie Hetland                                  The Anygui Project
http://hetland.org                                  http://anygui.org


From sholden at holdenweb.com  Thu Feb 21 17:00:52 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 21 Feb 2002 17:00:52 -0500
Subject: Is there a bug in CGI-POST
References: <8f8ffe67.0202201345.66fa13ed@posting.google.com>
Message-ID: <W5ed8.53878$n73.989478@atlpnn01.usenetserver.com>

"Sunit Joshi" <sjoshi at ingr.com> wrote in message
news:8f8ffe67.0202201345.66fa13ed at posting.google.com...
> Hello All
> I'm trying use the cgi module and I have seen that if I use the POST
> method in the form then I donot get any form fields. This however
> works with the GET method. Is this known, or am'i doing something
> wrong here. I'm using Xitami Server on Win2000
>
I've used Xitami with the cgi module without any problems, so it's probaby
something you're doing wrong. Would you like to post some code?

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jkraska at san.rr.com  Sun Feb 17 13:25:37 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 17 Feb 2002 18:25:37 GMT
Subject: Apology
References: <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net>
Message-ID: <ketv6u49deaj528kctcpj83s8daff0es37@4ax.com>

>                   under the GPL (and I reserve the right to
>change that to a more secure license should there be ...

No matter what changes one makes in a licensing scheme, no
unilateral change in licensing can withdraw rights to current
licensees (which in your case with the GPL represent the
entire human race) unless that license has a specific clause
allowing that. GPL'ing something is a one-way ticket for the
entire licensed body of work up to that moment in time. For
all those licensees -- all humans, I mean -- your change in
licensing terms for that past body of work is irrelevant
masturbation.

Any future work you do of course, is up to you. Which is
indeed begging a question! Given that you've outright
admitted that you expect OTHER PEOPLE to do the coding, it
really begs the question why you think YOU could fiddle
with the licensing at all for something which could not be
regarded by the sharp eyes of the law as YOURS. You CANNOT
dictate the rights of a third party to their work. Its
THEIRS, not yours.


C//



From sholden at holdenweb.com  Wed Feb 13 12:29:07 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 12:29:07 -0500
Subject: wxTextCtrl and handling button down even in OnChar()
References: <88bc63c6.0202130907.31a13d52@posting.google.com>
Message-ID: <ynxa8.209099$KM2.8640420@atlpnn01.usenetserver.com>

"Doug Farrell" <writeson at earthlink.net> wrote in message
news:88bc63c6.0202130907.31a13d52 at posting.google.com...
> Hi,
>
> I've written an application that has a control I've derived from
> wxButton so I can provide my own EVT_BUTTON(self, ID, self.OnClick)
> event handling. In the method OnClick() I'm getting the button press,
> but I see both the button down and button up events, which causes my
> code to run twice. I'd like to filter the button press and only react
> to the button down event. The trouble is, I can't seem to find the
> method of event or whatever that will allow me to do this. If any one
> has some ideas how I can process only moust button down events, I'd
> really appreciate it.
>
Doug:

I'm sure the wxHeads will reply giving you the answer you seek, this is just
a cautionary point about user interfaces in general -- it may not apply in
the specific case you have in mind.

If you are going to choose a single event to action on a button, it's better
to use the button UP rather than button Down event. There's nothing worse
than an interface that won't let you change your mind, so that a little
unintentional extra pressure on the mouse button is all you need for
disaster. On the better (more "comfortable") user interfaces you should find
you can click on a button then change your mind, and move the cursor off the
button before releasing the mouse button, without selecting any action.

still-looking-for-dwim-mode-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From mhammond at skippinet.com.au  Sat Feb 23 04:18:57 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 23 Feb 2002 09:18:57 GMT
Subject: pythonwin using COM import pythoncom error
References: <3c775bc7$0$79558$6536d21b@news.bitstream.net>
Message-ID: <3C775EFD.80204@skippinet.com.au>

mylinuxboxroot wrote:
> what should I do ????
> how can I get this file onto my system  ?????
> is this my problem then when I try to import pythoncom ????

The win32all EXE is compatible with .ZIP files - thus winzip etc can 
extract files manually.  You may need to exit windows before you can 
update the existing file though.

Mark.



From mhuening at zedat.fu-berlin.de  Mon Feb 18 07:42:57 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 18 Feb 2002 12:42:57 GMT
Subject: win32com and MS Excel
References: <mailman.1013967115.15656.python-list@python.org> <98322540.0202180251.2ddf81f4@posting.google.com>
Message-ID: <Xns91B98BCE94184mhueningzedatfuberli@130.133.1.4>

alife at narod.ru (Serge Terekhoff) wrote in
news:98322540.0202180251.2ddf81f4 at posting.google.com: 

> Dear all!
> 
> on
> from win32com.client import Dispatch
> 
> My Python2.2 replies:
> 
> Traceback (most recent call last):
>   File "F:\Users\Serge\MyPy\to_excel.py", line 1, in ?
>     from win32com.client import Dispatch
> ImportError: No module named win32com.client
> 
> 
> What i did wrong?

You will have to install Mark Hammond's win32all-extensions:
http://starship.python.net/crew/mhammond/

The ActiveState-version of Python comes with the extensions:
http://www.activestate.com/Products/ActivePython/

Matthias


From timr at probo.com  Thu Feb 21 00:45:57 2002
From: timr at probo.com (Tim Roberts)
Date: Wed, 20 Feb 2002 21:45:57 -0800
Subject: execfile question
References: <mailman.1014117330.24204.python-list@python.org>
Message-ID: <97297u47not4jc87fh6o1ajual65bhavbv@4ax.com>

Kerim Borchaev <warkid at storm.ru> wrote:

>Can someone explain why while running this script I don't see C.__del__
>eventually called?

Your object c isn't actually deleted until it gets garbage-collected.  In
your case, no garbage collection happens until after your program has
ended.  In that case, there is no sense in taking any further action.

__del__ is not like a C++ destructor.  As I recall, there was a lengthy
thread on this newsgroup last year on this topic, in which it was concluded
there is no way for an object to guarantee that a particular method gets
called at the end of its life.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From caljason76 at yahoo.com  Tue Feb 19 18:41:52 2002
From: caljason76 at yahoo.com (Jason)
Date: 19 Feb 2002 15:41:52 -0800
Subject: I haven't seen a recursive version, yet, so here.
Message-ID: <e9e6dd1f.0202191541.30eb9f62@posting.google.com>

# I am a Python novice so I don't really understand how things
# work, yet.  So some of this may either be wrong or not the
# Python way.  Also I have no idea why the 'if __name__ == ...'
# construct is needed instead of just the naked main statements.
#
# A few notes on the code.  There are two solutions: list_naked
# and list.
#
# list_naked is the simplies solution, it has no wrapper object
# and each node in list_naked is the head of a list.  You can see
# the recursion in list_naked.sum() and list_naked.__str_all(),
# where you will notice the if statements that test for the
# base case of the recursion.
#
# list_wrapped is a little more complex.  It has a wrapping object
# that is always used to refer to the list (unlike list_naked
# where the head of the list can change when a new value is pushed,
# hence the ll = ll.push(ll,x) assignment).  Here, the recursion
# is stopped by the_list_wrapped_empty_node and all the control
# flow is done by the interpreter.  I like this best, because I
# hate explicit control flow in programs and I think that it makes
# it easier to see the forest instead of the trees.  Looking at
# list_wrapped.sum() it is clear: add this node to the sum of the
# rest.  In list_naked.sum() you now have to look at two statements
# which do the same thing semantically.  With list_wrapped there
# is no way to forget to check for your base case because you don't
# check anything!  With just a list it seems a little too much,
# but when you start playing with various tree types and graphs
# it makes the code so much more readable.
#
# Also, can somebody help me find the documentation to the builtin
# procedures, like print().  I checked the __builting__ module docs
# but there is nothing there.  Thanks
#
# -j

class list_naked:
    def __init__(s, data, next = None):
        s.data, s.next = data, next
    def __str__(s):
        return "[%s]" % s.__str_all()
    def push(s, data):
        return list_naked(data, s)
    def sum(s):
        if s.next: return s.data + s.next.sum()
        else:      return s.data
    def __str_all(s):
        if s.next: return str(s.data) + " " + s.next.__str_all()
        else:      return str(s.data)

class list_wrapped_node:
    def __init__(s, data, next): s.data, s.next = data, next
    def __str__(s): return str(s.data) + " " + str(s.next)
    def sum(s): return s.data + s.next.sum()

class list_wrapped_empty_node(list_wrapped_node):
    def __init__(s): pass
    def __str__(s): return ""
    def sum(s): return 0

class list_wrapped:
    def __init__(s): s.head = list_wrapped_empty_node()
    def __str__(s): return "[" + str(s.head) + "]"
    def push(s, data): s.head = list_wrapped_node(data, s.head)
    def sum(s): return s.head.sum()
        
if __name__ == '__main__':
    ll = list_naked(0)
    for i in range(6): ll = ll.push(i+1)
    print ll
    print ll.sum()

    l = list_wrapped()
    for i in range(6): l.push(i)
    print l
    print l.sum()


From wurmy at earthlink.net  Sat Feb 16 12:24:32 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Sat, 16 Feb 2002 17:24:32 GMT
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <86n0yaw8u4.fsf@apo.hmotakef.int>
Message-ID: <3C6E95FA.F95C914B@earthlink.net>

Henrik Motakef wrote:
> 
> Hans Nowak <wurmy at earthlink.net> writes:
> 
> Hi.
> 
> First of all, I have to admit that I'm not sure what your problem
> is. :)

That is probably my fault. Here's a short summary. I want to store a
multiline string, containing newlines, in SQL Server, using an
updategram in XML. I post the record. I retrieve the record back.
Newlines are gone from the string.

I get the records back as XML too. In raw, unparsed XML (just a string)
the newlines are still there. As soon as I parse the XML, they are
somehow removed. See my other post to Martin von Loewis.

I don't know much about XML; if whitespace is considered insignificant,
like in HTML, then it might be removed when parsing.

> However, one thing that might possibly be related is that XML parsers
> are required to convert all of the funky end-of-line markers around
> (i.e. "\r\n", "\n", "\r") to a single newline "\n". Scince you Use SQL
> Server, I assume that you are also running Windows, where the
> "official" newline is "\r\n". Could that be related to your problem?

It was part of a previous problem; the XML supposedly only contained \n,
where \r\n was desirable. While investigating it, I stumbled upon a
different, more serious problem-- no newlines at all.

On a side note, any operating system with an Internet connection can talk
to the SQL Server database using SQLXML. All you need is some HTTP
functions (like Python has in httplib) to talk to it. No drivers needed.
All communication with the server is done through XML. Pretty fly for
a M$ product. ;-)

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From skip at pobox.com  Fri Feb  8 10:25:38 2002
From: skip at pobox.com (Skip Montanaro)
Date: Fri, 8 Feb 2002 09:25:38 -0600
Subject: [OT] PPT reader?
Message-ID: <15459.60914.405905.301501@12-248-41-177.client.attbi.com>

Guido posted a bunch of talk slides to the Python website in PowerPoint
format.  I understand his reasons for not creating HTML versions (mostly
because the HTML generated by PPT sucks).  Still, that leaves people without
access to PPT sort of out in the cold.  What non-MS tools are out there
(open source or otherwise) for reading, displaying or converting PPT slides
to other, more accessible, formats?  If we can get a reasonable (but fairly
short) list of tools, I can probably see they get referenced on the relevant
page of the Python website.

Thanks,

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From sjmachin at lexicon.net  Sun Feb 10 16:59:11 2002
From: sjmachin at lexicon.net (John Machin)
Date: 10 Feb 2002 13:59:11 -0800
Subject: CSV revisited
References: <slrna62f14.pef.mlh@vier.idi.ntnu.no> <a4474v$9t6$1@bob.news.rcn.net> <m3ofiyqgjf.fsf@ferret.object-craft.com.au>
Message-ID: <c76ff6fc.0202101359.784576c0@posting.google.com>

Dave Cole <djc at object-craft.com.au> wrote in message news:<m3ofiyqgjf.fsf at ferret.object-craft.com.au>...
> >>>>> "Raymond" == Raymond Hettinger <othello at javanet.com> writes:
> 
> Raymond> Documentation of key design decisions.  For instance, is the
> Raymond> library limited to reading various CSV formats or is
> Raymond> write-back in the same style to be supported.

YES. Has to be supported, to get data into applications that don't
support the MS one true way.

> Raymond> Can the first record, contain optional field names?

YES, but the implementation has to be layered; at the bottom layer the
deal should be to get rows of data into and out of Python as cleanly,
correctly and quickly as possible. The next layer up could be told
that the first row contains field names and (e.g.) stuff them into a
dict so that the caller could refer to columns by name rather than by
integer index. The next layer up could have a GUI. Let's design a
reasonable separation of layers. Let's get the bottom layer going
first.

> Raymond> Will incomplete records raise an exception?

Variations of format should IMO be handled by requiring the caller to
specify the expected format. The module should by default raise an
exception if the data deviates from that format. There may be options
to suppress the exceptions.

Back to Dave:
> 
> I am the author of the fast CSV module at:
> 
>         http://www.object-craft.com.au/projects/csv/
> 
> One of the problems with CSV is that everyone has their own ideas
> about what should be handled and how it should be handled.  If there
> was some sort of effort to document the correct behaviour of a CSV
> parser then I would be extremely happy to make mine conform to that
> document.

Ya, ya, me too. See http://www.lexicon.net/sjmachin/delimited.htm

The basic problem is that many people have "interesting" ideas on how
to create CSV-style data files. Query tools for commercial databases
have been known not to properly double the quotes. Given
address_line_1 containing
   3rd Floor, "Murgatroyd Mansions"
they produce
   "3rd Floor, "Murgatroyd Mansions""
instead of 
   "3rd Floor, ""Murgatroyd Mansions"""
This can be recovered if the presumption that there were an even
number of quotes in the original data is correct, but it needs a quite
different FSM to handle it [my quote_mode=1].

Other variations that I have seen are:
(a) Different front & back quotes:
      `3rd Floor, "Murgatroyd Mansions"'
(b) Alternate quote character [my quote_mode=3]
   3rd Floor, Murgatroyd Mansions
produces
   "3rd Floor, Murgatroyd Mansions"
as expected, but
   3rd Floor, "Murgatroyd Mansions"
produces
   '3rd Floor, "Murgatroyd Mansions"'
but this approach is no good if the data can contain both " and '

:-)
A few other desiderata: shouldn't leak memory, shouldn't die with a
SEGV (or SRE recursion limit exceeded!!!) if you ask it to unpack say
"x" * 70000 ...
(-:


From jeff at ccvcorp.com  Mon Feb 11 16:58:13 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 11 Feb 2002 13:58:13 -0800
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4> <3c658a5a_3@corp.newsgroups.com>
Message-ID: <3C683E75.EDFB3F4C@ccvcorp.com>


Joshua Muskovitz wrote:

> My suggestion would be for each type of receiver to simply create its own
> dict which maps message types (strings) to local functions.  ...
> [--------------snip---------------]
> class F(Receiver):
>     def __init__(self):
>         Receiver.__init__(self)
>         self.msgFuncs['fighto'] = rec_fighto

I believe you'd need that last line to be

        self.msgFuncs['fighto'] = self.rec_fighto

But otherwise this ought to work.  (Which might or might not be the same thing
as being the appropriate solution, but that's for the O.P. to decide.  ;) )

Jeff Shannon
Technician/Programmer
Credit International




From python at chajadan.net  Sun Feb 17 19:47:09 2002
From: python at chajadan.net (chajadan)
Date: Sun, 17 Feb 2002 16:47:09 -0800
Subject: help - inserting values into database tables
In-Reply-To: <3C6FEFAB.DE9907CC@bitstream.net>
Message-ID: <5.1.0.14.0.20020217163909.00a667a0@63.118.78.184>

Yes, this is altogether possible.

There is a module include in the main distribution called odbc, and one 
called dbi.

You can look into the modules documentation for the details.

But basically, do something like this:

import dbi
import obdc

conn = odbc.odbc("dsn")
rs = conn.cursor()
rs.execute("insert into " + table + " values ('" + charval1 + "', " + 
numval1 + ")")
rs.close
conn.close


 From what I heard, dbi should always be imported prior to obdc. The conn 
string takes a dns name as a string--it's easier to use system dsns for 
this, but I think there is a way around that to use file dsns if you must.

There are other oddities; but all in all this should get you started.

--chajadan

At 06:00 PM 2/17/02 +0000, mylinuxboxroot wrote:

>Hey all,
>
>Ok I am wondering something:
>I have a python list which has values I would like to insert into a
>database, the
>database is a microsoft (ugh)  Access type of database.
>-- I don't know much about the python for windows side of things like
>ODBC
>----Is this type of thing possible ????????????
>
>What are some basic things I need to know about this problem space???
>Where shall I look (on the web) for guideance for this type of thing ???
>
>Basically the MS - Access database tables do exist and I want to insert
>    info from my python list into it.  But I have no clue if this can be
>done.
>    Can I connect to the MS - Access database with a python script to
>     insert my list into the database ??????????
>
>Thanks,
>Joe F.
>-----
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From loewis at informatik.hu-berlin.de  Mon Feb 25 05:40:22 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 11:40:22 +0100
Subject: "Data" vs. "Non-data" Descriptors in PEP 252, and "once" attributes
References: <MPG.16e373104557a63b989685@news.supernews.com>
Message-ID: <j48z9hn8tl.fsf@informatik.hu-berlin.de>

Phillip J. Eby <pje at telecommunity.com> writes:

> Interestingly, this does *not* work if you try to do something similar 
> with the built-in 'property' type.  

I don't find this surprising. The builtin property type has a fixed
meaning for its constructor argument: they are callables, and they are
called on each access. How did you mean to implement your "once"
properties with it?

> Does anyone know if this usage is safe for future Python versions? 

That is the intention. There may be a different syntax in addition to
the one you are using now, which may allow you to write

    class aThing(object):

        def anAttr(self) [once]:
            print "computing anAttr"
            return 100*100

The idea is that at the end of the function definition, once is lookup
up and called with the function that was just defined; whatever it
returns is then taken as the value of anAttr inside aThing.

> Operationally, it seems it would be hard to change any of this in a way 
> that would break my "once" examples without also breaking lots of other 
> things.  But the PEP doesn't really give an official way to create either 
> a data or non-data descriptor; it just says what the current 
> implementation does.  Hence my concern.

Nobody can tell the future with certainty. If you think that certain
wording would tighten the PEP to make you feel more safe, just submit
a patch for the PEP to SF.

Regards,
Martin



From hugomartires at hotmail.com  Tue Feb  5 12:33:43 2002
From: hugomartires at hotmail.com (Hugo Martires)
Date: 5 Feb 2002 09:33:43 -0800
Subject: array precision
Message-ID: <d0380df9.0202050933.7481d17b@posting.google.com>

Hellow every one !
I'm using the array module to create a simple array:

myArray = array.array('f', [0.5, 0.5, 0.5, 0.5]

now i want to add to each position of the array the number:

myArray[i] = myArray[i] + (1.0/n)

suposing that n=10 we get: myArray[0] = 0.600000023 ("something like this")

but if i do: myArray[i] + (1.0/n) it give's me 0.6

what's the problem of the array, and how can i solve this ?

Tanks in advance !


From danb_83 at yahoo.com  Sat Feb 16 15:30:50 2002
From: danb_83 at yahoo.com (Dan Bishop)
Date: 16 Feb 2002 12:30:50 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <a4k2si$1skr$1@norfair.nerim.net> <926k4a.3a.ln@127.0.0.1>
Message-ID: <ad052e5c.0202161230.547557aa@posting.google.com>

Carl Banks <idot at vt.edu> wrote in message news:<926k4a.3a.ln at 127.0.0.1>...
> Gillou wrote:
> >>
> >> * NO LINE NUMBERS! Enough said.
>  ...
> >>   IF condition GOSUB line-number (I never did understand the necessity
> >>   for that kind of statement until well after I'd stopped using BASIC
> >>   entirely and moved to more "modern" languages like... Pascal).
> > 
> > Only old men like you and me remember this
> 
> Hi.
> 
> I remember using calculated GOSUB, something like this, in GW-BASIC:
> 220 ON LEVEL GOSUB 1000,2000,3000,4000,5000,...
> It was for a crude level-based video game.  I'm 25 now.

Atari BASIC even allowed statements like

220 GOSUB LEVEL*1000

Yuck!


From starx at pacbell.net  Sun Feb 17 17:13:11 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 17 Feb 2002 22:13:11 GMT
Subject: watching the number of references to an object
References: <1013700953.2880.0.nnrp-01.c1c3e191@news.demon.co.uk> <jhtn6u8lkbpviucvml02m8inj1khefbjkk@4ax.com>
Message-ID: <Xns91B8909E2AB87RASXnewsDFE1@130.133.1.4>

Courageous || Thu 14 Feb 2002 09:30:48a:

> On Thu, 14 Feb 2002 15:50:04 -0000, "Paul Brian"
> 
>>Dear all,
>>
>>Does anyone know how to monitor the references pointing to an instance
>>of an object?
> 
> In addition to the other answers, please note that the number of
> referrers will often be 1 higher than what you expect: the function
> for getting the references itself temporarily grabs a reference,
> IIRC.

Hm, it seems it would have to be this way to handle unbound object. E.g.

>>> sys.getrefcount( newfoo() )
1

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From cce at clarkevans.com  Sat Feb 16 17:24:12 2002
From: cce at clarkevans.com (Clark C . Evans)
Date: Sat, 16 Feb 2002 17:24:12 -0500
Subject: LMTP implementation?
Message-ID: <20020216172412.A32064@doublegemini.com>

I was wondering if there is a LMTP (RFC 2033) implementation 
in Python?  I'd like to have a custom SMTP handler in python
using Postfix/Exim as the primary mail handler, delegating
to a python deamon using LMTP.  

Thanks for any pointers!

Clark



From phd at phd.pp.ru  Mon Feb 18 11:05:14 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 18 Feb 2002 19:05:14 +0300
Subject: strange libtk8.3.so: open failed
In-Reply-To: <85e27928.0202180751.6782f145@posting.google.com>; from francois-regis.chalaoux@sanofi-synthelabo.com on Mon, Feb 18, 2002 at 07:51:03AM -0800
References: <85e27928.0202180751.6782f145@posting.google.com>
Message-ID: <20020218190514.R24919@phd.pp.ru>

On Mon, Feb 18, 2002 at 07:51:03AM -0800, copter24 wrote:
> ld.so.1: /usr/local/bin/python: fatal: libtk8.3.so: open failed: No
> such file or
>  directory

   Dynamic linker cannot find libtk8.3.so or *.so referenced by
libtk8.3.so. What OS? Where is your libtk8.3.so in the filesystem?

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



From sholden at holdenweb.com  Thu Feb  7 09:40:49 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 09:40:49 -0500
Subject: Twisted Selector
References: <u5rsqqb9tamaad@corp.supernews.com> <LBe88.87264$KM2.2829250@atlpnn01.usenetserver.com> <u63qm2cnomt801@corp.supernews.com>
Message-ID: <Dlw88.94843$KM2.3440170@atlpnn01.usenetserver.com>

"Lee Harr" <missive at frontiernet.net> wrote ...
> >> I am looking at this Twisted tutorial:
> >> http://twistedmatrix.com/page.epy/philosophy.html
> >> and having trouble with the very first code.
> >>
> > Moshe? Glyph?
> >
>
> Is anyone else using this Twisted thing?
> It looks cool to me, but I can't fix it :(
>
I'm copying this to Moshe, because I think he and Glyph should take a look
at the ViPEr software that was described at Python 10 (in fact it was the
subject of the winning paper). This is a visual programming environment. I
foresee that it could be extrememly useful in allowing network software to
be constructed by coupling twisted components together.

Moshe and Glyph described the Twisted environment at about ninety miles an
hour. I suspect the project needs someone who thinks a bit more slowly to
work on the documentation ;-) Like many open source projects that are
"on-ramps to the information superhgway [yawn]", there's more interest and
fun in building the stuff than in stopping construction to put up signposts.

This is not intended as a put-down, since I think twisted is neat.

just-not-fully-documented-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From eric.brunel at pragmadev.com  Thu Feb  7 12:25:25 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Thu, 7 Feb 2002 18:25:25 +0100
Subject: Getting correct file names under Windows
Message-ID: <a3ud3i$17g3$1@norfair.nerim.net>

Hi all,

Windows case insensitivity striked again: I've got a Python/Tkinter program
that uses the new tk function named tk_chooseDirectory (new in Tk 8.3),
displaying a directory selection dialog. For some reason, this function
insists in returning the selected directory all uppercase. Unfortunately,
even if Windows does not care about case, we definetely don't want that: the
disk we work on is mounted from a Samba Unix - therefore case sensitive -
server, and we want to use the same file names on both environments.
So I tried to get the "real" file name, i.e. the one that is defined on
Unix, and the one I actually see in Windows Explorer or DOS windows. I
didn't succeed...

Does anybody know any way to get the "real" name of a file or directory
under Windows, i.e. the one that Explorer displays?

TIA
 - eric -

PS: For those interested, during my tests, I stumbled on an "interesting"
thing...
If I issue at the DOS prompt something like:
cd PYTHON21
and then:
cd
it prints the correct directory name (C:\Python21), even if I typed it the
"wrong" way in the cd.
Now, if I do exactly the same thing via Python:
>>> import os
>>> os.chdir('PYTHON21')
>>> os.getcwd()
the result is 'C:\\PYTHON21'.
Even if I do an os.popen('cd', 'r').readline(), it returns the same thing!
And apparently, the case I use in the chdir is kinda "sticky":
>>> os.chdir('pYtHoN21')
>>> os.getcwd()
'C:\\pYtHoN21'
>>> os.popen('cd').readline()
'C:\\pYtHoN21'
This is *weird*!




From bobx at linuxmail.org  Tue Feb 26 22:11:51 2002
From: bobx at linuxmail.org (Bob)
Date: 26 Feb 2002 19:11:51 -0800
Subject: ActiveState version of 2.2???
References: <1001ff04.0202170702.6ebaaf55@posting.google.com> <3C704157.B872722F@bellatlantic.net>
Message-ID: <1001ff04.0202261911.22131410@posting.google.com>

So there is no ActiveState version now?  Or is it just as well to get
the regular distro + the win32all stuff?

David Lees <debl2NonoSpammyWhammy at bellatlantic.net> wrote in message news:<3C704157.B872722F at bellatlantic.net>...
> Mark Hammond who created PythonWin is no longer employed by ActiveState,
> but he has been kind enough to continue to update to 2.2 in the form of
> Win32all-146 that can be downloaded from his home page at:
> http://starship.python.net/crew/mhammond/win32/Downloads.html
> 
> David Lees
> 
> 
> Bob wrote:
> > 
> > When?


From nospam at bigfoot.com  Sat Feb 23 11:27:23 2002
From: nospam at bigfoot.com (Gillou)
Date: Sat, 23 Feb 2002 17:27:23 +0100
Subject: Newbie and exceptions.
References: <f0gf7u8h6tn05bf4a7orbnsm0e0lsn3oa6@4ax.com>
Message-ID: <a58ffq$uak$1@norfair.nerim.net>

"Jan Johansson" <spamfilter2 at mupp.net> a ?crit dans le message news:
f0gf7u8h6tn05bf4a7orbnsm0e0lsn3oa6 at 4ax.com...
> How on earth do i define errors to trap from ftplib?
>
> http://www.python.org/doc/current/lib/module-ftplib.html states
> "exception error_perm"  but "except error_perm:" gives me
>
>    except error_perm:
> NameError: name 'error_perm' is not defined
>
>
> why? and what is the proper way?

try...

except ftplib.error_perm:

...or...

# Near the start of the script
from ftplib import error_perm

# Continue as before
...




From Ed.Smirnov at mail.ee  Fri Feb 22 02:43:28 2002
From: Ed.Smirnov at mail.ee (Eduard Smirnov)
Date: Fri, 22 Feb 2002 09:43:28 +0200
Subject: Python vs. C/C++/Java: quantitative data ?
Message-ID: <3c75f5b3$0$149$bb624dac@diablo.uninet.ee>

I?m writing a diploma paper on the difference between scripting (e.g. in
Python or Tcl) and system programming (e.g. in C, C++ or Java), and one of
the claims I want to make is that you can often develop applications 10x or
more faster in a scripting language.
I apologue that more applications can be written or re-written in scripting
language faster and with more complex functionality then in system
programming language.
If you have any experience in writing applications in system programming
language (e.g. in C, C++) and re-writing it in scripting language (e.g.
Python) or vice versa, please compare your work in both implementations,
describe your work/application?s functionality, approximate number of lines
and time of development of each implementation, skill?s level needed.
Any more comments about are welcome.

Additional thanks for copy your answer to my e-mail.

Eduard.Smirnov at mail.ee
Tallinn Technical Univercity
Estonia





From bkd at graphnet.com  Tue Feb  5 01:03:03 2002
From: bkd at graphnet.com (Bruce Dykes)
Date: Tue, 5 Feb 2002 01:03:03 -0500
Subject: Drag and drop from one PC to the other
References: <a3nbg4$1jgs$1@rivage.news.be.easynet.net>
Message-ID: <007e01c1ae0a$caff75d0$5102020a@graphnet.com>

----- Original Message -----
From: "Bertrand Geston" <bergeston at yahoo.fr>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Monday, February 04, 2002 20:04
Subject: Drag and drop from one PC to the other


> I would like to know your opinion about this: I have 2 Windows computers
one
> next to the other and I would like to be able to pass object from one to
the
> other really easily. A kind of window open on each computer could be handy
> if the user can paste (by Cut 'n Paste or by Drag 'n Drop) an object in
one
> window and retrieve it from the other computer window. In other words, I
> would like to have a shared and distributed clipboard.
>
> Does it exist already ? Is it feaseable and easy ? I really don't know COM
> or COM+ but I guess that this could be part of the solution. Any opinion
or
> advice is really welcome (just to know how to start or where to look for).

The clipboard viewer in NT4 (at least, it's likely to be true for 98, and
quite possibly up through 2K, ME, and XP as well) has the ability to connect
to clipboards on other MS networked machines.

There are also third party clipboard replacements that offer expanded
functionality.

bkd




From hfoffani at yahoo.com  Fri Feb  8 18:29:23 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Sat, 9 Feb 2002 00:29:23 +0100
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <41914a.537.ln@wombat.san-francisco.ca.us> <a40e2r.3vvegs9.1@kserver.org> <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net>
Message-ID: <20020208182936.075$e7@news.newsreader.com>

> .
> "Mongers" is the closest answer.
>
> I generally write "Perlites", though

I looked for "monger" in http://www.dictionary.com/search?q=monger,
and now I ask myself, do the Perlites know the first acception
of monger or the second one?
:-P

-Hernan





From huaiyu at gauss.almadan.ibm.com  Thu Feb  7 16:04:50 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Thu, 7 Feb 2002 21:04:50 +0000 (UTC)
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com>
Message-ID: <slrna65qur.hgk.huaiyu@gauss.almadan.ibm.com>

On Wed, 06 Feb 2002 09:48:10 -0800, Jeff Shannon <jeff at ccvcorp.com> wrote:
>
>The problem you're running into, is that you've failed to convince anyone that
>*you* understand the real problems involved, enough to be able to solve them.  You
>don't need to convince us of the benefits of autocoding; you need to convince us
>of *your* competence.  So far, you're failing miserably.

He has several other major problems:

1. He seems to be unable to differentiate and respond well to negative
   comments.  He treats them all equally, and replies to the more negative
   ones with more enthusiasm.  Can he tell the difference between the
   following comments?  "I don't understand."  "I's already done."  "It
   can't be done."  "Yeah, that'll be good, but I don't think you've got the
   right approach to do it."  Sometimes he can't even see a humor.

2. He seems to claim that he has already abstracted all of computing into
   his nine commands, yet he also claims that this vision of his is
   impossible to be put into words, examples or mathematics.  That's kind of
   contradictary since the art of programming, or much of science itself, is
   to figure out exactly how things behave, and how to make things work as
   desired.  It is the art of collecting and summarizing details, to
   formulize and abstract them, to articulate and communicate them.

3. He misunderstood the mechanism of Open Source projects.  These projects
   exist because it is more efficient to do things that way.  A leader being
   a leader must have the ability to make things more efficient for others.
   He claims he knows the road ahead, but don't have the time to travel
   there.  So let's ask some hyperthetical questions.  Suppose there are
   twenty programmers willing to join his project.  Now he could achieve in
   a year or two what he would have done by himself in twenty years.  What
   are the tasks these programmers would be handling?  What are the most
   urgent theoretical problems to be solved?  What would be the codes he
   wants them to write in the first month?  Will he handle the final checkin
   and approval of code, and if so, what functionality and other criteria
   he'll use to judge them?  He seems to think that one man with a vision
   plus a million monkeys jumping on keyboards would solve real problems.

So with these problems in the way, it is not of much use communicating on
this topic.  The first and second problems also imply that there's not much
chance that any of these problems can be solved soon.

(Rue, if you are listening here, there's no need to respond.  We are not
sympathetic people here.  Programmers behave like Nazies here.)


Huaiyu


From starx at pacbell.net  Thu Feb  7 01:38:55 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 7 Feb 2002 06:38:55 GMT
Subject: why i love python...
References: <mailman.1013045771.13577.python-list@python.org>
Message-ID: <Xns91ADE66546C0ERASXnewsDFE1@130.133.1.4>

Mark McEahern || Wed 06 Feb 2002 05:34:52p:

> [please forgive the embedded html; this is posted on my weblog, but
> since that's not exclusively about python, i didn't want to bore you
> all with it.] 
> 
> what started out as just an attempt to make debugging easier turned
> into an epiphany about why i love python.  allow me to share that
> love... 
> 
> python comes with a slew of special methods for objects that you can
> leverage to do some really neat things.  consider __str__, which is
> called on your objects when you <b>print</b> them.  if you don't define
> __str__, that's ok, you'll get something like this:
> 
> <pre>
>>>>class Simple:pass ...
>>>> s = Simple()
>>>> print s
> <__main__.Simple instance at 0x100fd4c8>
> </pre>
> 
> by defining __str__, i get to control the output generated by print
> (and str()):
> 
> <pre>
>>>> class Prints:
> ...     def __str__(self):
> ...             return "whatever i like"
> ...
>>>> p = Prints()
>>>> print p
> whatever i like
> </pre>
> [further examples]

Hm, i really like this feature too. Consider this:

class Container( list ):
    """Base class for something that can contain."""
    def __repr__(self):
        """We are not a list when printed"""
        basic = object.__repr__(self)
        return basic

    def __str__(self):
        return self.printcontent()

    def printcontent(self, indent = 0):
        """Recurisve repr listing"""
        output = repr(self) + '\n'
        for x in self:
            for tabs in range(indent+1):
                output += ' '
            if hasattr( x, 'printcontent'):
                output += x.printcontent( indent+1 )
            else:
                output += repr(x) + '\n'
        return output

class Object( Container ):
    #init sets name
    def __repr__(self):
        return '<' + self.name + '>'

class Creature( Object ):
    #init sets name, class, brain
    def __repr__(self):
        return ''.join( '<', self.class, ' ', self.name,
                        ' is ', self.brain.state_desc )

    def __str__(self):
        out = ' '.join( name, " the ", class, "\nI am ",
                        self.brain.state_desc "\n")
        keys = self.stats.keys()
        keys.sort()
        for key in keys:
            out += ''.join( key, ': ', self.stats[key] '\n' )
        out += "I am carrying:\n"
        out += Object.__str__
        return out

PyFilling from PyCrust just became a lot more useful :). (My container 
already works like this, i just made up the other two, but i will probably 
do similar.)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From alf at leo.logilab.fr  Fri Feb  1 04:40:38 2002
From: alf at leo.logilab.fr (Alexandre Fayolle)
Date: Fri, 1 Feb 2002 09:40:38 +0000 (UTC)
Subject: underflow exception (was Re: float limits)
References: <mailman.1012520889.3930.python-list@python.org>
Message-ID: <slrna5kokl.8nm.alf@leo.logilab.fr>

On Thu, 31 Jan 2002 18:42:33 -0500, Tim Peters <tim.one at home.com> wrote:
>In that case I'd try explicitly checking the intermediate products, from
>time to time, to see whether they've gotten so small as to be
>indistinguishable from 0 for the app's purposes.  Depending on the whole
>algorithm, it may be quite sensible to treat a probability of (say) 1e-20 as
>if exactly zero, and may even speed things up to do so.  The difference
>between a probability of 1e-100 and one of 1e-300 has no consequence in any
>universe I've visited <wink>, and if so there's no point burning cycles to
>put an ever-finer point on a useless distinction.

I would have agreed with you, but in this case, there is a point, because 
these small values are not probabilities, but they are used to compute 
probabilities from other probabilities. The algorithm I use has something
like pi = a/b where both a and b are in the aforementionned order of magnitude
(they can be expressed roughly as the products of 100~1000 floats smaller than 
one).

The solution is to use a clever scaling of a and b to avoid numerical 
underflow, but the computation of the scaling factor is a costly operation, 
so I'd like to perform it only when needed. Right now, the scaling is triggered
by a division by zero, but I'm afraid this is too late, and that some of the 
terms already have been rounded down. I'll probably have to add an explicit
check that the values are not getting smaller than 1e-200, and trigger the 
scaling at that point instead of relying on the exception.

By the way, in case you're interested, the module I'm working on is for 
training hidden markov models (HMMs). It's available at 
http://www.logilab.org/hmm/

Thanks for your advice.


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


From stojek at part-gmbh.de  Thu Feb 14 08:45:07 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Thu, 14 Feb 2002 13:45:07 GMT
Subject: sorting list of tuples by second (third...) tuple item
Message-ID: <3c6bbe96.12195968@news.easynews.net>

Hi,

I am still struggling with tuple-lists.

How can I sort a list of tuples (all tuples have 4 float items)
by the second (or third or n-th) tuple item. And I have to
do it fast.

Tanks,
Marcus


From jason at jorendorff.com  Sun Feb 10 23:42:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 22:42:02 -0600
Subject: FixedPoint
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBCEBEKCAA.marklists@mceahern.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEKEDBAA.jason@jorendorff.com>

> [Jason Orendorff]
> > Yep.  If you know what's in _norm() you can do better than I
> > can.  :)  Also, you might pull the first return statement out
> > of the try.
> 
> _norm basically pulls the number and the precision out of the FixedPoint
> instances.  So that would give us this:
> 
>     def __cmp__(self, other):
>         try:
>             xn, yn, p = _norm(self, other)
>         except TypeError:
>             xn, yx = self, float(other)
>         return cmp(xn, yn)

No, instead of "self, float(other)" it should be "float(self), other".

    def __cmp__(self, other):
       try:
           xn, yn, p = _norm(self, other)
       except TypeError:
           xn, yn = float(self), other
       return cmp(xn, yn)

This should work with None, and any other object that a float
can be compared with.

## Jason Orendorff    http://www.jorendorff.com/



From robin at jessikat.fsnet.co.uk  Thu Feb 14 12:56:11 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 17:56:11 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com>
Message-ID: <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk>

In article <2upn6uoqotqejiji69bg8kebfacdu9oamb at 4ax.com>, Paul Moore <paul.moore at atosorigin.com>
writes
>One nice feature of Java (sorry!) is the ability to ship an application
>as a single JAR file - containing all code, resources, etc, etc. Is
>there any equivalent feature available in Python? The obvious place
>where it would be is in Gordon McMillan's Installer package, with its
>Python Archive format, but I don't think (from looking at the docs) that
>does quite what I'm after.
>
>I have no requirement to include a Python interpreter, or C extensions,
>in the archive. The code is all pure Python. But, I do want to be able
>to include support files (text files, image files for use in wxPython
>toolbars, graphics for display on screen, sound files, etc etc) which
>can be picked up from the same archive.
>
>I'm looking for something which works a bit like Java's
>ClassLoader.getResourceAsStream() (along with the necessary stuff to
>allow the stream to then be *used* as an icon/image/whatever).
>
>It seems like the sort of thing which should be possible with Python,
>but I can't find it. Did I miss something?
>
>Thanks in advance,
>Paul.
>
well his .pyz is a compressed archive form and I have been using it to compress the entire
python library. All that's required is a special importer to allow imports from such an archive
to work. I believe that the archive is allowed to contain data files as well so provided you
alter the open/read you can at least consider doing that.

The startup code I use is below. Assuming that zlib is available means
that the modules archive_rt & imputils don't need to be specially obtained
with a linked version of libz. 

In our demo we do this just after initialising
python and  before importing site. If a full python is already available
you'd only need to do this in a stub python file to get access to all of your 
modules etc.

###################################################
def _rl_readPYZ(fn):
        import zlib, imp, marshal, sys, struct
        MAGIC = 'PYZ\\0'
        TOCPOS = 8
        pymagic = imp.get_magic()
        lib = open(fn,'rb')
        if lib.read(len(MAGIC)) != MAGIC:
                raise RuntimeError, fn+' is not a valid ZArchive file'
        if lib.read(len(pymagic)) != pymagic:
                raise RuntimeError, fn+' has version mismatch to dll'
        lib.seek(TOCPOS)
        (offset,) = struct.unpack('=i', lib.read(4))
        lib.seek(offset)
        toc = marshal.load(lib)
        for name in ('imputil','archive_rt'):
                (ispkg, pos, lngth) = toc.get(name, (0, None, 0))
                if pos is None: raise ValueError, 'Module '+name+' not found'
                lib.seek(pos)
                code = marshal.loads(zlib.decompress(lib.read(lngth)))
                if ispkg: values['__path__'] = [fqname]
                is_module = isinstance(code, type(sys))
                module = is_module and code or imp.new_module(name)
                module.__dict__.update({'__file__':code.co_filename,'__iname__':name})
                sys.modules[name] = module
                if not is_module: exec code in module.__dict__
                module = sys.modules[name]
                module.__name__ = name
        import archive_rt
        sys.importers.append(archive_rt.ZlibArchive(fn, 0))
_rl_readPYZ(r'reportlab.pyz')
del _rl_readPYZ
-- 
Robin Becker


From logiplexsoftware at earthlink.net  Tue Feb 19 21:00:57 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 19 Feb 2002 18:00:57 -0800
Subject: Recursively traverse linked list -- help!
In-Reply-To: <e9e6dd1f.0202191746.2c3525db@posting.google.com>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
	<mailman.1014106650.11778.python-list@python.org>
	<e9e6dd1f.0202191746.2c3525db@posting.google.com>
Message-ID: <20020219180057.30ca3b8e.logiplexsoftware@earthlink.net>

On 19 Feb 2002 17:46:30 -0800
Jason wrote:

> # I am a Python novice so I don't really understand how things
> # work, yet.  So some of this may either be wrong or not the
> # Python way.  Also I have no idea why the 'if __name__ == ...'
> # construct is needed instead of just the naked main statements.

It isn't "needed".  It's typically used so that a module can provide a
built-in test/example.  When you use "import ll" this code will not be
executed because __name__ will not be "__main__".  If you put just the
naked main statements, you will need to comment them out before you import
them into another program or else they'll get executed (unless that's what
you want) upon import.

> # A few notes on the code.  There are two solutions: list_naked
> # and list.
> #
> # list_naked is the simplies solution, it has no wrapper object
> # and each node in list_naked is the head of a list.  You can see
> # the recursion in list_naked.sum() and list_naked.__str_all(),
> # where you will notice the if statements that test for the
> # base case of the recursion.
> #
> # list_wrapped is a little more complex.  It has a wrapping object
> # that is always used to refer to the list (unlike list_naked
> # where the head of the list can change when a new value is pushed,
> # hence the ll = ll.push(ll,x) assignment).  Here, the recursion
> # is stopped by the_list_wrapped_empty_node and all the control
> # flow is done by the interpreter.  I like this best, because I
> # hate explicit control flow in programs and I think that it makes
> # it easier to see the forest instead of the trees.  Looking at
> # list_wrapped.sum() it is clear: add this node to the sum of the
> # rest.  In list_naked.sum() you now have to look at two statements
> # which do the same thing semantically.  With list_wrapped there
> # is no way to forget to check for your base case because you don't
> # check anything!  With just a list it seems a little too much,
> # but when you start playing with various tree types and graphs
> # it makes the code so much more readable.

I'm not seeing a question here.  Are you simply asking if your code is
"pythonic"?

> # Also, can somebody help me find the documentation to the builtin
> # procedures, like print().  I checked the __builting__ module docs
> # but there is nothing there.  Thanks

print is a statement, not a procedure (which is why you don't need
parentheses when using it), and hence is located in the language reference,
not the library reference.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From kfarmer at thuban.org  Mon Feb 25 02:57:04 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 25 Feb 2002 07:57:04 GMT
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com> <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com> <258fd9b8.0202200150.1b4e53de@posting.google.com>
Message-ID: <k7me8.42891$9_1.1798161575@newssvr14.news.prodigy.com>

Not ActiveState -- regular Win32 install:

Microsoft Windows 2000 [Version 5.00.2195]
(C) Copyright 1985-2000 Microsoft Corp.

C:\>python
Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.


Here's one symptom I noticed (in the copyright entry .. also in the license,
you'll see):

 <class exceptions.AssertionError at 0x00881490>, 'TypeError': <class
exceptions
.TypeError at 0x0087E868>, 'copyright': Copyright (c) 2001 Python Software
Found
ation.
All Rights Reserved.

Copyright (c) 2000 BeOpen.com.
All Rights Reserved.

Copyright (c) 1995-2001 Corporation for National Research Initiatives.
All Rights Reserved.

Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved., 'license': Type license() to see the full license
text, 'c
lassmethod': <type 'classmethod'>, 'KeyboardInterrupt': <class
exceptions.Keyboa
rdInterrupt at 0x0087ED50>, 'UserWarning': <class exceptions.UserWarning at
0x00

What I ended up noticing was that the way the dict gets handled seems to
execute the value under certain conditions which I didn't bother to
investigate further.  In PyCrust (before Patrick made his corrections, which
I haven't tried out), I was able to cause my program to exit by inserting a
reference to the exit function into the namespace and then print it out.

However, the excerpt above doesn't come from PyCrust -- it was from the
command line interpreter.

I'm running Win2k SP2.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Magnus Lyck?" <magnus at thinkware.se> wrote in message
news:258fd9b8.0202200150.1b4e53de at posting.google.com...
> "Keith Farmer" <kfarmer at thuban.org> wrote in message
news:<x7Cc8.2126$RJ4.246925423 at newssvr14.news.prodigy.com>...

> Running a non-ActiveState version of Python?





From alet at unice.fr  Thu Feb 14 04:18:16 2002
From: alet at unice.fr (Jerome Alet)
Date: Thu, 14 Feb 2002 10:18:16 +0100
Subject: How to generate XML
References: <3c21e330@news.opennet.it> <j44rmko5s6.fsf@informatik.hu-berlin.de> <a0212d$57k$1@news1.xs4all.nl> <1013618891.574031@rexx.com>
Message-ID: <3C6B80D8.C80DF0ED@unice.fr>

> > "Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
> > news:j44rmko5s6.fsf at informatik.hu-berlin.de...
> >
> >> I'm amazed that people always start proposing difficult-to-use
> >> libraries when this question comes up. I'd prefer to use
> >>
> >> print "<foo/>"

I suggest you to try my jaxml module available under the GPL at
http://cortex.unice.fr/~jerome/jaxml/

it is alsmost as easy to use as print statements but takes care of the
closing tags for you.

hth.

Jerome Alet


From andymac at bullseye.apana.org.au  Tue Feb  5 07:26:35 2002
From: andymac at bullseye.apana.org.au (Andrew MacIntyre)
Date: Tue, 5 Feb 2002 23:26:35 +1100 (EDT)
Subject: Problem while uploading a binary file using cgi module
In-Reply-To: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU>
Message-ID: <Pine.OS2.4.32.0202052319360.2529-100000@tenring.andymac.org>

On Sun, 3 Feb 2002, J. Wang wrote:

{...}

> the code is like this:
> #!e:/Python22/python.exe
> import cgi

{...}

Given that a text file works and a binary file doesn't, you are almost
certainly falling foul of the CRLF<->LF newline (text/binary) translation
used by default on most DOS derived environments (I'm not certain how
Cygwin handles this).

You don't say exactly what platform your Python is running on, however if
using the standard Windows binaries, look into the docs for the msvcrt
module (setmode()).  I have something similar in development for OS2/EMX.
If Cygwin, you'll need to follow up with Cygwin folk (Jason Tishler
perhaps?)

--
Andrew I MacIntyre                     "These thoughts are mine alone..."
E-mail: andymac at bullseye.apana.org.au  | Snail: PO Box 370
        andymac at pcug.org.au            |        Belconnen  ACT  2616
Web:    http://www.andymac.org/        |        Australia




From maxm at mxm.dk  Thu Feb 14 08:16:37 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 14 Feb 2002 13:16:37 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk>
Message-ID: <3C6BB8B5.8020903@mxm.dk>

Robin Becker wrote:

> I am having bad problems with zope not starting up properly. I think
> this is caused by having Python-2.1.2 + win32all installed. It seems
> that paths set up in the registry by win32all are getting into sys.path
> when zope runs. This is causing massive confusion between modules
> installed in the base python and those I wish to test with zope.

I have several versions of Zope and Python + win32all installed and has 
never had any problems.

Are you starting zope via your own script or via z2.py ??

I have a "zope243.bat" on my desktop for starting Zope:

C:\zope\zope243\bin\python.exe C:\zope\zope243\z2.py -D -P 8000

Then I am shure that it gets started with the correct Python version and 
libraries. And I can easily run it in debug mode. Which is a bit harder 
than starting it as a service. Well usually I install it as a sevice and 
then disable it. That way I can easily switch between development and 
production.

And if I want to run more than one server at a time i have another bat 
file ie:

"zope250.bat"

C:\zope\zope25\bin\python.exe C:\zope\zope25\z2.py -D -P 8001

So that version runs on another port.

All pretty painless.

regards Max M



From st-newsgroups at thomas-guettler.de  Tue Feb 26 03:16:55 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Tue, 26 Feb 2002 09:16:55 +0100
Subject: MSWord and Python
References: <3C7A4EF8.4010404@thomas-guettler.de> <a5f00p$6udv4$1@ID-12869.news.dfncis.de>
Message-ID: <3C7B4477.2040909@thomas-guettler.de>

Ricky wrote:

> The client side document is include in Python online documents.


Sorry I didn't understand you. Is there are library for creating word 
documts without COM? This would be nice since I could create the 
documents on unix, too. What's its's name?

thomas



From quinn at regurgitate.ugcs.caltech.edu  Thu Feb 21 19:15:48 2002
From: quinn at regurgitate.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Feb 2002 00:15:48 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202191901.5f6643f6@posting.google.com> <a50le4$jmk$1@thorium.cix.co.uk>
Message-ID: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu>

On Wed, 20 Feb 2002 17:12:36 +0000 (UTC), gbreed at cix.compulink.co.uk
<gbreed at cix.compulink.co.uk> wrote:
>Alexander Jerusalem wrote:
>
>> I just looked up what CLOS does. I don't know if you call that static
>> typing or not.

You wouldn't.  At least we should agree that lisp is dynamically typed :)

>> typing or not. In any case, CLOS does have type declarations in the
>> source code which is what I want and what enables the features I'd
>> like to see in Python. For example in CLOS you can have the following
>> method declarations:
>> 
>>  (defmethod Area ((C Circle)) (* pi (Square (Radius C)))
>>            (defmethod Area ((R Rectangle)) (* (Width R) (Height R)))
>>            (defmethod Area ((Sq Square)) (Square (Width Sq)))
>>            OR (defmethod Height ((Sq Square)) (Width Sq))
>
>But you can do that in Python:
>
>>>> def area(shape):
>...   if isinstance(shape, Circle):
>...     return math.pi*shape.radius**2
>...   if isinstance(shape, Rectangle):
>...     return shape.width*shape.height
>...   if isinstance(shape, Square):
>...     return shape.width**2
>...   raise TypeError, "unknown shape"
>...

This kind of giant typecase is exactly the sort of thing that OO is supposed
replace.

>It can't take much imagination to see how it could be sugared to work with 
>multiple functions with types resolved at run time.

Right:
def area(Shape(o)):
    ...
def area(Rectangle(o)):
    ...
def area(Circle(o)):
    ...

But this is no longer python.

>                                                     As for strict, 
>strong, static, explicit typing (which may be a straw man) it looks *less* 
>suitable for this example.

Yeah, I agree that static typing is orthogonal to OO-style function dispatch.

>Say you want to include a triangle in your list of shapes.  So, you add
>
>...   if isinstance(shape, Triangle):
>...     return shape.base*shape.height
>
>to the area function.

No, you shouldn't have to modify area(), but add a shape() method to Triangle.

>                       But then, will all the other functions that call 
>area be allowed to accept a triangle?

Sure.  They specify that they want an object with an area method, and
as soon as you add one of those to Triangle, Triangles becomes acceptable.
This is how static CLOS-esque systems work, like haskell or clean.

>You could make them all subclasses of Shape.  In which case, you have 
>functions defined to use a polymorphic object that depend on function 
>overloading by the real type.  Not all statically typed languages will 
>allow that.

C++ requires special keywords for it, but all statically typed OO languages I
know of do dynamic dispatch.  I certainly don't know every static language out
there, but just because there's a limited implementation somewhere doesn't
mean the concept is unworkable :)

>             If they do, you have the problem that you might have added 
>triangles for some other reason, and forgotten to cater for them in 
>area().  You aren't getting static type checking at all.

Sure you are.  area will reject objects that don't have the proper methods
defined for them.  How it works in haskell is that each class has a minimum
interface that you must implement.  If you don't implement all the methods
of the Shape class, Shape-accepting functions won't accept your objects.
This is less flexible than having to implement *only* the methods the
function actually uses, as in python, but c++ templates are sorta like that.

>Wouldn't it be better to make area a method of the shapes?  And for static 
>typing, make it abstract in the Shape class?

In your above example, that's what it is.  You're just spelling 'circle.area()'
as 'area(circle)'.

I actually like the second better, because lets you do things like
'map(area, shapes)' instead of 'map(lambda o: o.area(), shapes)', as well
as multi-dispatch, and eliminates the 'len(o)' vs 'o.len()' thing.  It also
has other benefits like letting you define a method anywhere you want which
obeys the usual scoping rules (so you can add a 'frobbify' method to the
string class, but not have to worry about it making a global change for
everyone in all modules).

I'm not even sure why the smalltalk style 'object followed by method' notation
caught on, especially since 'function followed by args' has always been
firmly entrenched.


From aleax at aleax.it  Mon Feb 18 03:33:23 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 08:33:23 GMT
Subject: lambda to Sort Dictionary.items() by Value
References: <a4p47i$13ed$1@news.idiom.com>
Message-ID: <n%2c8.82065$3J1.2616781@news2.tin.it>

Jim Dennis wrote:

>  I'm playing with text and I decided to write a simple histogram
>  function (calculating letter occurence frequencies, in this case).
> 
>  In the course of that I found two simple uses for lambda one of
>  which seems like it would be a very common idiom in Python.

Fortunately not, because decorate-sort-undecorate (DSU) is faster
and more general.

>     i = h.items()
>     i.sort(lambda (k1,v1),(k2,v2): cmp(v2,v1) ) # sort by value, not by
>     key
>     i=string.join(map(lambda (x,y):x, i),"")  # extract letters from each
>     return (h,i)

Try instead (Python 2.2, but it's similar in older Pythons, 2.0 & later):
    temp = [ (h[key], key) for key in h ]
    temp.sort()
    return h, ''.join([ key for value, key in temp ])

>  It seems like one might need to sort dictionary items by value
>  pretty often.

Depends on your application field, but DSU and list comprehensions
help when you do have such needs.


Alex



From synthespian at uol.com.br  Mon Feb 11 20:17:23 2002
From: synthespian at uol.com.br (synthespian)
Date: Mon, 11 Feb 2002 22:17:23 -0300
Subject: [solved] RE: No explanation for weird behavior in re module! [reading files in ISO-8859-1]
References: <mailman.1013386937.27383.python-list@python.org><Xns91B26FC9941CEmhueningzedatfuberli@130.133.1.4>
Message-ID: <a49qpb$1d6ppa$1@ID-78052.news.dfncis.de>

In article <Xns91B26FC9941CEmhueningzedatfuberli at 130.133.1.4>, Matthias Huening <mhuening at zedat.fu-berlin.de> wrote:
> "Tim Peters" <tim.one at home.com> wrote in
> news:mailman.1013386937.27383.python-list at python.org: 
> 
>> 
>>>      Other than the fact that 'T?r' has the '?' unicode charcater, I
>>>      fail 
>>> to see any difference! 
>> 
>> Heh.  Leaving this joy to someone else <wink>.
>> 
> 
> Okay, I'll try...
> If your string comes in 'Latin-1' you will have to tell Python to treat it 
> as Unicode. And when you want to print it afterwards, you'll have to 
> encode it as 'Latin-1'.
> 
> ---
> import re
> txt = 'die T?r, T?ren'
> txt = unicode(txt, 'latin-1')
> pattern = re.compile(ur'(der|die|das)\s+(\w+)', re.UNICODE)
> 
> match = pattern.match(txt)
> article = match.group(1)
> noun = match.group(2)
> 
> print article.encode('latin-1')
> print noun.encode('latin-1')
> ---
> 
> Hope this helps.
> Matthias

Matthias-

	Thank you for this simple elegant solution.
	Although Jason Orendorf has also posted a similar solution, this one is simpler.
	Thanks for all the fine people who tried to help, too.

	Learning German will never be the same again for me!!
	Danke sch?n!!! :-))

	Cheers
	Henry
	synthespian at uol.com.br





From gh_pythonlist at gmx.de  Fri Feb  8 13:35:35 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Fri, 8 Feb 2002 19:35:35 +0100
Subject: What's the equivalent in Python for Perl's CPAN?
In-Reply-To: <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>
References: <CFz88.19365$zJ3.1503300@typhoon.maine.rr.com> <a3v903$kgk$1@panix2.panix.com> <msU88.19568$zJ3.1529583@typhoon.maine.rr.com>
Message-ID: <20020208183534.GB1206@lilith.hqd-internal>

Le 08/02/02 ? 18:06, Mike Carifio ?crivit:
> [...] 
> So, what's the equivalent in Python for Perl's CPAN?

Nothing yet. ActiveState's solution works quite well for now.  And work
is being done on this project:

    http://sourceforge.net/projects/pythonsiphon

But don't expect miracles soon. They might take a while to happen ;-)

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.4 ?C      Wind: 2.1 m/s



From deathtospam43423 at altavista.com  Sat Feb 16 19:52:46 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 17 Feb 2002 00:52:46 GMT
Subject: Newbie: Problems with File IO
Message-ID: <Xns91B873A49D077AusYourStandingInIt@210.49.20.254>

Why idn't it working?

I'm trying to write my own crude, html based chat script. I figure the way 
to do it is to use a Python cgi script to take info from a web form, store 
it in a separate text file, and then get the script to redisplay it in the 
same page the form was on. 

I re read the chapter on File IO in Josh Cagliati's beginner's tutorial 
http://www.honors.montana.edu/~jjc/easytut/easytut/ 

I'm trying to use the commands detailed there to store the cgi info in the 
file, but there's a problem. Everytime someone accesses the form, and the 
script burns the info into the text file, the info that was previously in 
the file gets erased. That's no good! That means that visitors to this chat 
site will only ever be able to see one post at a time - their own!

Can someone tell me what I might be doing wrong? What do I need to learn in 
order to fix the problem?

Here follows the script. The part I think I might be getting wrong is the 
part labelled ## The Action......

#! C:/Python22/python
import cgi, string
from time import time, ctime

print "Content-Type: text/html\n\n"
## This determines the type of output the cgi gives. Can be 'text/plain'

def TOPpage():
    print '<html>'
    print '<title>hairy fat bastard chat</title>'
    print '<body>'
    print '<br>'

def BOTpage():
    print"</body>"
    print"</html>"
    
TOPpage()

##The Action
The_Form = cgi.FieldStorage()
out_file = open("log.txt", "w")
for loop in The_Form.keys():
    out_file.writelines(The_Form[loop].value + "<BR>")
out_file.close()

in_file = open("log.txt","r")
chatters = in_file.read()

print chatters


print '<FORM method=post action="fieldtest1.py">'

print '<b>handle:</b>'
print '<INPUT type=text NAME="name" SIZE="30" MAXLENGTH="30">'
print '<br><b>chat:</b>'
print '<br>'
print '<TEXTAREA name = "chat" ROWS=10 COLS=80 SIZE="30" MAXLENGTH="30">'
print '</textarea>'
print '<br>'
print '<INPUT TYPE=submit VALUE="Get/Send"> </form> </center>'

BOTpage()



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From charlie at news.wanadoo.fr  Mon Feb  4 09:35:23 2002
From: charlie at news.wanadoo.fr (charlie at news.wanadoo.fr)
Date: Mon, 4 Feb 2002 14:35:23 GMT
Subject: READ THE WEB-SITE THAT THE KIKES TRY TO CENSOR
Message-ID: <A042AF87D5C59948FB6368C76F958@news.wanadoo.fr>

WWW.GIWERSWORLD.ORG
THE SITE WHERE ALL KIKES ARE LAMPSHADES

PROUDLY PRESENTED BY

MATTHIAS GIWER
14205 N. NEBRASKA AVE
TAMPA FLORIDA

813-977-7380


jull43 at tampabay.rr.com
matt at giwersworld.org


PROUDLY HOSTED BY:

DREAMHOST WEB HOSTING

213-947-1032



Other sticky bizarre ointments will order annually before shirts.  
They explain once, sow partly, then fear in back of the tyrant 
at the monolith.  She'd rather scold lovingly than answer with 
Yolanda's younger hat.  She wants to call elder units against 
Neil's window.  

It can recollect wickedly if Norbert's pear isn't lost.  Try not to 
tease the cans halfheartedly, clean them usably.  Lots of fresh 
diets through the clean stadium were arriving towards the durable 
cave.  Both playing now, Corinne and Byron dyed the polite barns 
towards cold spoon.  

She might sneakily irritate empty and attacks our heavy, hot 
raindrops in back of a ceiling.  For Edna the tree's closed, 
in me it's stale, whereas beneath you it's moving lower.  Hardly any 
abysmal painter or sunshine, and she'll wastefully cook everybody.  

Charlene attempts, then Sherry virtually fills a weak jug in 
Ronnie's hallway.  Ben, beside yogis humble and proud, laughs 
alongside it, tasting easily.  

If you will behave Norris's road beneath cases, it will daily 
learn the plate.  What will we excuse after Lionel converses the 
ugly castle's ache?  If the filthy buckets can measure eventually, the 
pathetic bandage may lift more camps.  It's very sad today, I'll 
creep firmly or Nydia will change the wrinkles.  How doesn't 
Peter kill fully?  Tomorrow, it covers a cup too dry alongside her 
long lake.  





From chris.gonnerman at newcenturycomputers.net  Thu Feb  7 09:01:24 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Thu, 7 Feb 2002 08:01:24 -0600
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl>
Message-ID: <007901c1afdf$f901d820$0101010a@local>

----- Original Message -----
From: "Martijn Faassen" <m.faassen at vet.uu.nl>


> Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
> > The time has come, my friends and colleagues, to bring the issue at hand
> > to a vote:
>
> > Should Timothy Rue go away?
>
> > All in favor, post AYE.  All opposed, post NAY.
>
> > (Yes, this is a waste of bandwidth, but in my opinion not as bad as
Rue.)
>
> Please, I think it's far worse than Rue. I don't want this group to start
> turning into a group that 'votes' others off. Either ignore him or
killfile
> him or just talk to the fellow. Having a vote is only a weird popularity
> contest more about the people doing the voting than the person being voted
> on in my opinion, and we could do without.

Ah, perhaps you are right.  I was in a particularly bad mood at that point.

How the heck to you "enjoy" an idiot like Rue though?





From sab at NOSPAM.freeuk.com  Mon Feb  4 18:04:34 2002
From: sab at NOSPAM.freeuk.com (G. Willoughby)
Date: Mon, 4 Feb 2002 23:04:34 -0000
Subject: how can i get the screen width and height in pixels?
References: <a3mclk$pjh$1@news5.svr.pol.co.uk> <3c5ec34b.27704203@news.t-online.de> <1965188.y9P21nC1DU@borg.pp.se>
Message-ID: <a3n3vi$706$1@newsg3.svr.pol.co.uk>

I'm centering a Tkinter app in any given monitor res.  :)

G. Willoughby

"Jakob Borg" <jakob at borg.pp.se> wrote in message
news:1965188.y9P21nC1DU at borg.pp.se...
> Gerson Kurz wrote:
>
> > On Linux:
> >
> > Buy a metric ruler, then construct a lego mindstorm robot to measure
> > the screen with that ruler.
>
> Or, perhaps simpler, get the output of xdpyinfo and do a regexp on it to
> extract the 'dimensions:' field for screen #0.
>
> Then there is also the question about _why_ you want to know the sceen
> dimensions - my guess is that it is probably to solve the wrong kind of
> problem...
>
> --
> Jakob Borg




From alanmk at hotmail.com  Wed Feb 27 04:40:54 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 27 Feb 2002 01:40:54 -0800
Subject: How to have multi-threaded server log to one file ?
References: <97ae44ee.0202261855.15c97708@posting.google.com>
Message-ID: <f72dac0d.0202270140.97b925f@posting.google.com>

shriek at gmx.co.uk (Stephen) wrote in message 
> When the server is run, each request from a client is handled by
> MyHandler() in a separate thread. If multiple requests are being run
> at the same time, how should I write my code so that they cal all
> write to the same log file without conflict ?  

How about representing every log message as a simple object which is
put() on a Queue.Queue? So each server thread produces messages to go
on the Queue, and a single logging thread consumes those messages and
does the actual writing into the log files. Since there is only ever
one thread writing to the files, there are no contention problems.

That Queue.Queue class is mightily useful!

Regards,

Alan.


From pan-news at asheeshenterprises.com  Wed Feb 27 20:44:24 2002
From: pan-news at asheeshenterprises.com (Asheesh Laroia)
Date: Thu, 28 Feb 2002 01:44:24 GMT
Subject: Regular expression problem
References: <mailman.1014855289.6788.python-list@python.org>
Message-ID: <YXff8.148576$dh.38567308@typhoon.nyroc.rr.com>

I've been trying to use sgmllib, actually, to delete all the other tags.

It just doesn't handle the <@ [...] > condition well.  It refuses to
parse it, treating it as text.

So, I need to use a regex to kill it, instead.

-- Asheesh.

On Wed, 27 Feb 2002 19:13:25 -0500, Sean 'Shaleh' Perry wrote:


> On 27-Feb-2002 Asheesh Laroia wrote:
>> I have some SGML input (PageMaker 6.5 tagged text), and I want to be
>> able to recognize (and delete) a tag.  That tag looks like:
>> 
>>       <@Trap Body text:>
>> 
>> It may also look like <@Trap Body text: useless-data>.
>> 
>> So, I tried the regular expression r"<@.?>".  That doesn't match the
>> above string.  Nor does r"<@.?Trap Body text.?>".  What RE should I be
>> using, and why doesn't this work?
>> 
>> 
> There is a really nice SGML parser in python's library.  Perhaps it will
> help you.  The name is 'sgmllib'.


From db3l at fitlinxx.com  Sat Feb 23 00:43:35 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 23 Feb 2002 00:43:35 -0500
Subject: HELP! Obscure Python Reentrancy Question
References: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com>
Message-ID: <ur8ncpxbs.fsf@ctwd0143.fitlinxx.com>

Courageous <jkraska at san.rr.com> writes:

> 3. The C++ program occasionally dispatches to a bit of Python
> code.
> 
> 4. The Python code calls an external C++ function which
> triggers the thread switch back to the scheduler.
> 
> 5. The scheduler starts or resumes another thread, which
> very likely will contain more Python, some of which has
> been left in this intermediate state like this.
> 
> My test shows this going to hell in a handbasket. That
> doesn't surprise me.
> 
> What now?

I know this was designed as a minimal description, but a little more
detail might be helpful.

How have you allocated interpreter state(s) versus your C++ scheduler
threads (or fibers) and how are you maintaining the GIL?

When the other thread in 5 resumes and "contains" more Python, does
that mean that from the perspective of the Python script you are
"returning" from the call in 4, or does that mean there's some
independent Python code you are just starting to execute?

It seems to me that depending on how you want to link your Python code
execution with your scheduler thread states, you may want to have an
interpreter (or at least thread state) per thread, even if your
threads aren't technically native but fibers within a single native
thread.  

In that case, when the currently executing thread that is in Python
code calls into the C++ function and provides the opportunity to yield
to another thread, that new thread can simply continue executing with
its own state - presumably returning to Python from an earlier
function call which itself yielded.

But it's hard to suppose too much more in the abstract.

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


From aldo at nullcube.com  Sat Feb 16 08:44:44 2002
From: aldo at nullcube.com (Aldo Cortesi)
Date: Sun, 17 Feb 2002 00:44:44 +1100
Subject: Warning about "for line in file:"
In-Reply-To: <3C6D73DE.3090201@wi.mit.edu>
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu>
Message-ID: <20020216134444.GA12676@nullcube.com>

Thus spake Brian Kelley (bkelley at wi.mit.edu):

> Neil Schemenauer wrote:
> 
> >Russell E. Owen wrote:
> >
> >>The readline or xreadline file methods work fine, of course.
> >>
> >
> >Why "of course"?  iter(file) does the same thing as file.xreadlines().
> >Have you tested xreadlines?
> >
> >  Neil
> >
> >
> 
> I had the same problem with xreadlines but "for line in
> file" is MUCH less explicit and leads to erros like this.
> 
> file = open(...)
> 
> count = 0
> for line in file:
>     if count > 10: break
>     print line
>     count = count + 1
> 
> for line in file:
>     print line
> 
> Doesn't work like I would expect.  This is essentially
> doing the following:
> 
> file = open(...)
> 
> count = 0
> for line in file.xreadlines():
>     if count > 10: break
>     print line
>     count = count + 1
> 
> for line in file.xreadlines():
>     print line
> 
> So what is REALLY happening is that you are creating two
> seperate iterators in the above examples.  Writing "for
> line in file" instead of "for line in file.xreadlines()"
> simply hides and confuses this.
> 
> The problem with spawning multiple iterators is that their
> is a read cache going on behind the scenes and
> file.xreadlines() doesn't rewind the file to the starting
> point.


Actually this has nothing to do with iterators, or a "read
cache". iter(file) creates a line iterator that does the
same thing as file.readline() every time .next() is called,
until it reaches the end of the file. But file.readline(),
just like any other file read, starts reading at the
_current seek position_ of the file. 

For instance, say we have a file with one digit per line,
like this:

1
2
3
4
5

On your machine it may differ, but on my machine this file
is exactly 10 characters long - each digit is followed by a
line feed. If we now do:

file = open("file")

print file.tell()
for i in file:
    pass
print file.tell()

We see that the file position started at character 0, and
ended at character 10. Another attempt to read from the file
will produce nothing. However, if we now do:

file.seek(6)

then...

for i in file:
    print i,

... we get:

4
5

In a nutshell, a file read will start at the file offset,
which can be found by going file.tell(), and set by going
file.seek(). This is the case wether you use iterators,
xreadlines(), readlines(), or just plain read()...



Cheers,



Aldo





-- 
Aldo Cortesi
aldo at nullcube.com
www.nullcube.com



From jeff at nmt.edu  Thu Feb 28 13:01:54 2002
From: jeff at nmt.edu (Jeff Shipman - SysProg)
Date: Thu, 28 Feb 2002 11:01:54 -0700
Subject: auto LD_LIBRARY_PATH loading?
Message-ID: <3C7E7092.5060101@nmt.edu>

I need to be able to have DCOracle run without me explicity
setting the LD_LIBRARY_PATH before starting python. Is there
a way I can modify python to automatically add certain
paths to LD_LIBRARY_PATH?

-- 
Jeff Shipman           E-Mail: jeff at nmt.edu
Systems Programmer     Phone: (505) 835-5748
NMIMT Computer Center  http://www.nmt.edu/~jeff



From phr-n2002a at nightsong.com  Thu Feb 21 18:51:41 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 15:51:41 -0800
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org>
Message-ID: <7xlmdmfl6a.fsf@ruckus.brouhaha.com>

Geoffrey Talvola <gtalvola at nameconnector.com> writes:
> Webware's WebKit app server has a different approach that doesn't require
> cookies and also doesn't require adding the session ID to all URL's.  It has
> an option to use "automatic path sessions".  If you turn on this option,
> then WebKit will redirect a request without a session ID like:
> 
> http://foo.bar/cgi-bin/WebKit.cgi/MyServlet
> 
> to:
> 
> http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet
> 
> Since the session ID comes _before_ the name of the servlet, all relative
> links work just fine.  WebKit knows how to parse the URL and extract the
> session ID.

Btw, I better point out a not-so-obvious security peril with this
scheme and any others like it that put the session ID into the URL.

If the ID is for a logged in session (i.e. the user has typed a
password) and needs to be protected from unauthorized access, you have
to be careful of any links on your page that point to other sites.
If the user's browser is pointing at

   http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet

and the user clicks a link on the page to www.attacker.com/whatever,
then the browser will normally send an HTTP GET request to the
attacker.com server containing an HTTP REFERER header which has
your page's URL including the session ID.  The attacker can then
use the session ID from the referer header to hijack the user session.
If you're running something like a web bbs where attackers can
drop their own URL's into messages that they post, they can
carry out the attack without you (the foo.bar owner) ever
adding any links of your own to the attacker.

The user doesn't even have to be made to click a link.
if your page (because of you or the attacker) contains an
   <IMG src=http://www.attacker.com/whatever>
to display an image, the attack happens before the user
has a chance to do anything about it.

There are various hacks you can think up to get around the problem,
but basically putting session ID's in the URL is not a great scheme if
the session needs to be kept secure and you can't carefully control
the page content to not have any off-site references.


From xtian at toysinabag.com  Mon Feb  4 03:43:21 2002
From: xtian at toysinabag.com (xtian)
Date: 4 Feb 2002 00:43:21 -0800
Subject: PyQT - Eric the Debugger?
Message-ID: <d111eafa.0202040043.b64f5df@posting.google.com>

I was quite keen to try out the debugger supposedly included in the
PyQT distribution. So I downloaded and installed it (Windows binary
install), but it wasn't there. After a bit of Googling I found a CVS
archive for it, and downloaded the debugger files separately. But when
I run eric.py, I get an error:

Traceback (most recent call last):
  File "eric.py", line 13, in ?
    from DebugServer import DebugServer
  File "DebugServer.py", line 7, in ?
    from qt import SIGNAL, PYSIGNAL, QServerSocket, QSocket
ImportError: cannot import name QServerSocket

(Python 2.1.1, Win98)

QServerSocket doesn't appear in the docs I have, and it's not in any
of the Qt modules (at least, not the ones I installed). The examples
work, so I'm reasonably sure the Qt installation's fine.

Anyone know about this? It's not really that important, I'm just
curious as to what I've done wrong.

Thanks,
Xtian


From clj at theWorld.com  Mon Feb  4 11:33:50 2002
From: clj at theWorld.com (Chris Jones)
Date: Mon, 4 Feb 2002 16:33:50 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org> <XPg78.22996$l93.4173649@newsb.telia.net>
Message-ID: <tdnn0yp4369.fsf@shell01.TheWorld.com>

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

[...]

  (how come Guido ignores K&R and Indian Hill, but still insists
  on Python programmers using the GvR style guide? ;-)

As the punchline to the joke goes, "Because he can!"


From daniels at mindspring.com  Thu Feb 21 13:10:04 2002
From: daniels at mindspring.com (Alan Daniels)
Date: 21 Feb 2002 10:10:04 -0800
Subject: Stupid Lambda Tricks [LONG]
References: <3c74c211.20012718@news.isar.de>
Message-ID: <5d283efe.0202211010.472bf39c@posting.google.com>

gerson.kurz at t-online.de (Gerson Kurz) wrote in message news:<3c74c211.20012718 at news.isar.de>...

> Stupid Lambda Tricks...

You are an evil, evil man.

Keep up the good work! :-)


From vlindberg at verio.net  Fri Feb 22 21:13:58 2002
From: vlindberg at verio.net (VanL)
Date: Fri, 22 Feb 2002 19:13:58 -0700
Subject: Request for comments: Search and replace script
References: <XFMail.20020222152043.shalehperry@attbi.com>
Message-ID: <3C76FAE6.6030502@verio.net>

Hello,

Thanks to Kirby Urner, there is now a highlighted html file:

http://lindbergs.org/vsar.html

Thanks,

Van




From emile at fenx.com  Thu Feb 28 14:04:58 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 28 Feb 2002 11:04:58 -0800
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl> <a5lk0a$8enfq$1@ID-11957.news.dfncis.de> <a5lltf$4ra$1@animus.fel.iae.nl>
Message-ID: <a5lvav$8614t$1@ID-11957.news.dfncis.de>

"Carel Fellinger"
> Emile van Sebille
> >     for ii in iter(len(mylist)):
>
> And how would this compare to
>
>       for ii in indici(mylist):
>

Interesting idea... I guess I was after something like:

class _dexedItem:
    def __init__(self, sequence, idx, idxname, idxnamespace):
        self.__dict__['sequence'] = sequence
        self.__dict__['idx'] = idx
        self.__dict__['idxname'] = idxname
        self.__dict__['idxnamespace'] = idxnamespace
    def __setattr__(self, attr, value):
        self.__dict__['sequence'][self.idx] = value
    def __del__(self):
        if
self.__dict__['idxnamespace'].has_key(self.__dict__['idxname']):
            self.__dict__['sequence'][self.idx] =
self.__dict__['idxnamespace'][self.idxname]
        else:
            self.__dict__['sequence'][self.idx] = None
    def __coerce__(self, other):
        return (self.__dict__['sequence'][self.idx],other)

class dexing:
    def __init__(self, sequence):
        import traceback, sys
        self.sequence = sequence
        self._idxname = traceback.extract_stack()[-2][3].split('in
dexing(')[0].split()[-1]
        frame = sys._getframe(-2)
        if frame.f_locals.has_key(self._idxname):
            self._idxnamespace = frame.f_locals
        else:
            self._idxnamespace = frame.f_globals
    def __getitem__(self, idx):
        if idx < len(self.sequence):
            return _dexedItem(self.sequence, idx, self._idxname,
self._idxnamespace)
        else:
            raise IndexError

mylist = range(5)

for ii in dexing(mylist):  # read as ' for ii in_dexing(mylist)'  ;-)
    ii *= 4

print mylist

Is __del__ protocol such as to reliably support this now?

--

Emile van Sebille
emile at fenx.com

---------



From sjmachin at lexicon.net  Tue Feb 26 04:44:24 2002
From: sjmachin at lexicon.net (John Machin)
Date: 26 Feb 2002 01:44:24 -0800
Subject: ask help about raise, thank u
References: <mailman.1014686626.31452.python-list@python.org>
Message-ID: <c76ff6fc.0202260144.70e52642@posting.google.com>

"dongkui" <dongkui at huawei.com> wrote in message news:<mailman.1014686626.31452.python-list at python.org>...
> cGxzIGdpdmUgbWUgb25lIGV4YW1wbGUgYWJvdXQgcmFpc2Ugd2hpY2ggaGF2ZSB0aHJlZSBwYXJh
> LCBhbmQgdGhlIHRoaXJkIHBhcmEgaXMgdHJhY2ViYWNrIG9iag0KDQpsaWtlIA0KcmFpc2UgImEi
> LCAiYiIsIGMgI2MgYXMgdHJhY2ViYWNrIG9iamVjdA0KDQp0aGFuayB1DQo=
> 
> --

After running the above through
repr(base64.decodestring(the_above_guff)) and wrapping lines at 40
columns:

'pls give me one example about raise whi
ch have three para, and the third para i
s traceback obj\r\n\r\nlike \r\nraise "a
", "b", c #c as traceback object\r\n\r\n
thank u\r\n'

Further translation:
para -> paragraph? No
para -> paratrooper? I see no "fee fi fo fum" material here.
para -> parameter -> argument? Aha! 

Disclaimer: This is how I think the third arg is meant to be used; I
have never
actually had to do this in practice.

try:
   some code which may explode
except:
   exc, data, tb = sys.exc_info()
   if I_can_handle_this(exc):
      handle_it(exc, data)
   else:
      # re-raise the exception, with traceback giving the *original*
      # source of the problem (somewhere inside the 'try' suite),
      # not the 'raise' statement
      raise exc, data, tb

Also, you would need to heed the warning in the documentation for
sys.exc_info()about circular references caused by assigning the
traceback to a local variable.

Hope this helps,
John


From akuchlin at ute.mems-exchange.org  Thu Feb  7 22:10:59 2002
From: akuchlin at ute.mems-exchange.org (A.M. Kuchling)
Date: 08 Feb 2002 03:10:59 GMT
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <mailman.1013125874.23020.python-list@python.org>
Message-ID: <slrna66gdu.t61.akuchlin@ute.mems-exchange.org>

In article <mailman.1013125874.23020.python-list at python.org>, 
	Michael Chermside wrote:
> I agree, and apparently so does TimBL. He was careful to point out that 
> you don't actually go out to the web each time to get module, rather 
> that the actual code is "cached" locally and there needs to be something 

Note that there's a distinction between using a URI as a unique
identifier and actually downloading something from it.  In the XML
Namespaces spec, URIs are used in things like <element
xmlns:rng="http://xml.mems-exchange.org/schema/">.  That URI can't be
dereferenced (the hostname doesn't even exist) but it's never
retrieved, so that doesn't matter.  I think TBL's suggestion was more
along the lines of having unique identifiers for modules and versions,
and not primarily importing via URL.

--amk                                                  (www.amk.ca)
Nothing I've ever written has reached 1.0.
    -- Greg Ward at IPC7, on using small version numbers


From tim.one at comcast.net  Mon Feb 18 00:27:02 2002
From: tim.one at comcast.net (Tim Peters)
Date: Mon, 18 Feb 2002 00:27:02 -0500
Subject: Problems with telnetlib
In-Reply-To: <a4pin8$tc$0@216.39.172.122>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEDKNNAA.tim.one@comcast.net>

[Bengt Richter]
> ...
> Otherwise maybe a standalong unified diff GUI viewer with buttons to see
> old, new, or both?

On Windows, that's called windiff.  That was the first app I ever used that
made me suspect a GUI might actually be better for something than working in
an Emacs process buffer.  Luckily, it was also the last <wink>.

waiting-patiently-for-the-browser-fad-to-fade-ly y'rs  - tim




From sholden at holdenweb.com  Fri Feb  8 14:52:39 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 14:52:39 -0500
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <00W88.108659$KM2.4246695@atlpnn01.usenetserver.com>

"Skip Montanaro" <skip at pobox.com> wrote in message
news:mailman.1013181915.301.python-list at python.org...
>
> Guido posted a bunch of talk slides to the Python website in PowerPoint
> format.  I understand his reasons for not creating HTML versions (mostly
> because the HTML generated by PPT sucks).  Still, that leaves people
without
> access to PPT sort of out in the cold.  What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT
slides
> to other, more accessible, formats?  If we can get a reasonable (but
fairly
> short) list of tools, I can probably see they get referenced on the
relevant
> page of the Python website.
>
> Thanks,
>
As a public service, and completely without Guido's permission, I have
temporarily installed a PDF copy (generated by printing GvR's ppt file to
Acrobat PDFWriter) of his Developers' Day presentation at

    http://www.holdenweb.com/python/py10keynote.pdf

If Guido would like to have one of his many minions take a copy and save it
on the Python.org site I will then be able to delete mine. Alternatively, if
he's happy to let it stay there then so am I.

but-until-he-says-so-i'm-not-building-any-more-links-to-it-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From henrik.motakef at web.de  Sun Feb 10 04:40:54 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 10 Feb 2002 10:40:54 +0100
Subject: Python IDEs
References: <zfC08.15504$Hg7.1623775@news11-gui.server.ntli.net> <a1uv39$o79$1@rex.ip-plus.net> <3c664c2b@nntp.server.uni-frankfurt.de>
Message-ID: <867kplznbt.fsf@apo.hmotakef.int>

Michael 'Mickey' Lauer <mickey at tm.informatik.uni-frankfurt.de> writes:

> From my point of view, I'd like to have the time to add Python
> capabilities to kdevelop, which seems the most promising open source
> IDE around.

AFAIK, there is Python (and other non-C++ languages) support in the
CVS version of Gideon (the successor of KDevelop). So no need to take
the time yoursef.

Regards
Henrik


From deathtospam43423 at altavista.com  Sun Feb 10 01:07:06 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 10 Feb 2002 06:07:06 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org>
Message-ID: <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
news:mailman.1013316023.10782.python-list at python.org: 

>
> "PERL type CGI's" should be read "scripts".  You can't run Python,
> Perl, Ruby,
> Lua, etc. from this particular server.
> 
> Try Xitami.  It is quite compatible with Python.  http://xitami.com
> 
Thanx. I'm downloading it as we speak. 

So with this server, will I be able to just send info direct from my 
browser form to a file? If I can do that, then I'm up and running.

[snip]
> 
> The form looks good, but where is script.cgi?
> 
In the same directory as the html form, cgi-bin.


-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From jason at jorendorff.com  Sun Feb 10 23:36:25 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 22:36:25 -0600
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
In-Reply-To: <e840346c.0202070724.13440c8d@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEKEDBAA.jason@jorendorff.com>

Mark Brady wrote:
> I think this would be very useful also but how would it work with
> Python's dynamic features? I know some native code comipled languages
> such as Objective C and GJC (the GNU java native code compiler) come
> with a runtime library for the native exe, is this possible for a
> language like python?

C and C++ also have runtime libraries.  The difference is, those
particular runtimes are already installed everywhere.

Yes, it's possible for Python.  You could think of py2exe this
way, actually.

## Jason Orendorff    http://www.jorendorff.com/



From sdm7g at Virginia.EDU  Mon Feb  4 19:51:50 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Mon, 4 Feb 2002 19:51:50 -0500
Subject: [OSX] milter won't build
In-Reply-To: <200202050100.TAA28342@starbase.neosoft.com>
References: <200202050100.TAA28342@starbase.neosoft.com>
Message-ID: <Pine.OSX.4.43.0202041946230.317-100000@d-128-61-180.bootp.virginia.edu>


On Mon, 4 Feb 2002, Cameron Laird wrote:

> In article <mailman.1012779846.20463.python-list at python.org> you write:
> 			.
> 			.
> 			.
> >I don't know what the milter module is, but:
> 			.
> 			.
> 			.
> Does <URL: http://www.swexpert.com/CC/SE.C12.JAN.01.pdf > help?

Thanks.
So does that mean that "-lsm" was looking for the sendmail lib ?

If that what was missing, it's not shipped in Mac OS X, but he
can get the sendmail sources and projects for OSX at
publicsource.apple.com ( or maybe try Fink ? )

-- Steve Majewski




From mcherm at destiny.com  Thu Feb 14 08:51:19 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Thu, 14 Feb 2002 08:51:19 -0500
Subject: How to generate XML
Message-ID: <3C6BC0D7.7010103@destiny.com>

Mark McEahern writes:
> There seem to be two bugs in the code you posted, Michael.  For what it's
> worth, here's a fully working demo of the code with the minor bug fixes:


Thanks. That was just something I threw together between reading Dave 
Kuhlman's posting and sending it in, so I didn't do the kind of testing 
that I would on "real" code. Apparently my final "test run" after the 
last refactoring didn't actually execute.

Of course, there are better tools out there than what I threw together 
in a few minutes (Jerome's kit seemed nice). I mostly just wanted to 
point out that Python's extensive introspection makes it pretty easy to 
write something like this as generic code rather than having to code 
boilerplate for each and every class.

-- Michael Chermside









From cfelling at iae.nl  Thu Feb 21 21:03:44 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 22 Feb 2002 03:03:44 +0100
Subject: giant stdlib tuples
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com> <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu> <uadu3m5dz.fsf@python.net> <slrna7agtq.1vk.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <a548u0$fd2$1@animus.fel.iae.nl>

Quinn Dunkan <quinn at regurgitate.ugcs.caltech.edu> wrote:
...
> The days of 'os.stat(fn)[stat.ST_BLAH]' are over.

> I still wonder about those wacky C structure names, though.  Did they think
> people would forget what kind of structure they were asking about after they
> typed the '.'?

In the old days when this habbit formed, (some?) C compilers used to
simple replace a field name with an offset and for simplicity reasons
used the same offset for all fields with the same field name
regardless of in what struct it was used.  So you had to name those
fields differently or be sure that they all *really* had the same
offset.  It took me days to find the bug this caused, and then I
decided to never ever use C again, boy was I glad:)
-- 
groetjes, carel


From andy47 at halfcooked.com  Sun Feb 24 20:23:39 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Mon, 25 Feb 2002 01:23:39 +0000 (UTC)
Subject: Problems building DCOracle 1.3.2 under Win2K/Cygwin
References: <1a241b9d.0202211459.394c3e64@posting.google.com>
Message-ID: <Xns91C07DF1A8B8Eandy47halfcookedcom@203.109.252.31>

cmedcoff at home.com (Charles Medcoff) wrote in 
news:1a241b9d.0202211459.394c3e64 at posting.google.com:

> I'm just following the instructions in the src/readme.txt file.  It
> appears that a file is missing from the distribution.  Any advice
> appreciated.
> 
> $ make
> gcc -DUSE_DL_IMPORT -g -O2 -Wall -Wstrict-prototypes
> -I/usr/local/include/python
> 2.1 -I/usr/local/include/python2.1 -DHAVE_CONFIG_H  -I/rdbms/demo
> -I/rdbms/publi
> c -I/network/public -I/plsql/public -DDCORACLE8 -c ././oci_.c -o
> ./oci_.o
> ././oci_.c:565: ociapr.h: No such file or directory
> In file included from ././oci_.c:697:
> oci_8.c:67: oci.h: No such file or directory
> make: *** [oci_.o] Error 1
> 
> Regards,
> Chuck
> 

Chuck,

I hate to be the bearer of bad tidings but I'm not sure its possible to do 
what you want. I have had similar issues trying to get cx_Oracle working 
under cygwin and have currently put that effort 'on hold' (see 
http://groups.google.com/groups?
hl=en&threadm=Xns91AD8D95B2B69andy47halfcookedcom%40203.109.250.24&rnum=1
&prev=/groups%3Fq%3Dandy47%2Bcygwin%2Boracle%2Bgroup:comp.lang.python.*%
2Bgroup:comp.lang.python.*%26hl%3Den%26selm%
3DXns91AD8D95B2B69andy47halfcookedcom%2540203.109.250.24%26rnum%3D1) excuse 
the long URL.

Basically, you have the Oracle client libraries for Windows and you are 
trying to compile them on *nix type system. Oracle provides different 
clients for each platform, oci.dll for Win32 and liboci.so for *nix. The 
make file for DCOracle will take this into account and wont easily mix and 
match the two (i.e. I don't think its possible just to substitute one for 
the other).

The only fruitful line of inquiry I didn't fully pursue was to create an 
import library for oci.dll (so that cygwin sees it as liboci.so) which 
should allow you to build DCOracle under cygwin.

For information on creating an import library look at;

http://www.cygwin.com/ml/cygwin/2000-03/msg00694.html

Let us know how you progress.

Regards,
Andy
-- 
Contents free posts a speciality


From op73418 at mail.telepac.pt  Mon Feb 25 11:02:48 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 16:02:48 +0000
Subject: Question on Python path
Message-ID: <3qnk7ug977hk4peorufobtlo38nh8j9v87@4ax.com>

I have put all my own modules in a subdirectory of site-packages. How
can I instruct Python (I'm using Python 2.2 on win2k) to also look up
this directory when a

import <whatever>

is found?

My Thanks,
Gon?alo Rodrigues

P.S: I have edited the registry directly by adding a User key in the
Pythonpath folder. Needless to say I do NOT want to mess directly with
the registry.


From tjreedy at home.com  Thu Feb  7 16:47:11 2002
From: tjreedy at home.com (Terry Reedy)
Date: Thu, 07 Feb 2002 21:47:11 GMT
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <f72dac0d.0202070547.8cb548a@posting.google.com> <HGx88.95599$KM2.3475019@atlpnn01.usenetserver.com>
Message-ID: <zBC88.30105$X64.10397017@news1.rdc2.pa.home.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:HGx88.95599$KM2.3475019 at atlpnn01.usenetserver.com...
> His slides are available at
>
>     http://www.w3.org/2002/Talks/0206-python
>
> As for the content, he took an ec;ectic view of Python, and
suggested that
> his thoughts on "Webizing" weren't necessarily particularly
practical (with
> which I'd agree).

I'd like to see the following someday (from his slide 4)

import http://www.w3.org/2000/10/swap/llyn.py

coupled with a public library of modules for anyone to import latest
version of.

Terry J. Reedy





From rdsteph at earthlink.net  Sun Feb  3 08:26:33 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 03 Feb 2002 13:26:33 GMT
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net>
Message-ID: <3C5D3B66.2B06CBE6@earthlink.net>

Greetings! Welcome to the python community. I am also a newbie, at it a
lttle less than one year but I'm a slow learner ;-))) One source for
tutorials, bok revirews, etc for newbies is my site at
http://www.awaretek.com/plf.html


wrbt wrote:
> 
> Hi and hello all, I'm the latest Python convert... 2 months and counting.
> 
> I LOVE IT.
> 
> You'll see me around.


From ivan at kupol.ru  Thu Feb 21 06:02:44 2002
From: ivan at kupol.ru (Ivan Bykov)
Date: Thu, 21 Feb 2002 15:02:44 +0400
Subject: socket in python 2.2 vs 2.1
In-Reply-To: <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de>
References: <mailman.1014206732.25525.python-list@python.org>
 <m37kp7kcs0.fsf@mira.informatik.hu-berlin.de>
Message-ID: <12425686785.20020221150244@kupol.ru>

Hello Martin,

Thursday, February 21, 2002, 2:29:35, you wrote:
>> It`s bug?
MvL> I think this is a bug in your program. You don't need to connect() for
MvL> a UDP socket, but if you do, you need to pass a valid address.
Python 2.1: '255.255.255.255' - valid address. And my application
working is correct.

Python 2.2: '255.255.255.255' - narrower not valid address.

In than problem here is.
-- 
Best regards,
 Ivan                            mailto:ivan at kupol.ru





From phd at phd.pp.ru  Fri Feb  8 04:16:07 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 8 Feb 2002 12:16:07 +0300
Subject: Please stop responding to Rue
In-Reply-To: <b34d3835.0202080025.1d69d198@posting.google.com>; from rony.steelandt@bucodi.com on Fri, Feb 08, 2002 at 12:25:55AM -0800
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com>
Message-ID: <20020208121607.C12726@phd.pp.ru>

It is amazing how fast a dozen of lamers, few trolls and a free software
advocate can destroy the good old friendly spirit of c.l.py.

On Fri, Feb 08, 2002 at 12:25:55AM -0800, Rony wrote:
> YOU SHOULD BE ASHAME !!!!!!!!!!!!!!!!!!!!!
> 
> Who on earth you think you are, to just post a message on usenet
> saying don't talk to that guy !!
> 
> 1. Nobody or nothing gives you that right.
> 2. It's a attack to peoples intelegence here, i assume they very well
> can decide by themself if they want to dicusse or not
> 3. It's against the spirit of usenet
> 4. You are not God to decide this
> 5 ....
> 
> I could go on for hours like that
> 
> Think about this
> 
> Rony
> 
> 
> sandskyfly at hotmail.com (Sandy Norton) wrote in message news:<b03e80d.0202060227.368f6ea7 at posting.google.com>...
> > "Whereof one cannot speak, thereof one must be silent."

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



From shelton at custommachine.com  Sat Feb 16 01:30:27 2002
From: shelton at custommachine.com (Jeffrey N. Shelton)
Date: Sat, 16 Feb 2002 01:30:27 -0500
Subject: win32com and Passing SafeArrays of Doubles
References: <mailman.1013814685.21004.python-list@python.org> <3C6DB769.9060906@skippinet.com.au>
Message-ID: <000a01c1b6b3$6cec9610$0a01a8c0@DELL530>

Success, at last!

Orginally, the makepy-generated code which didn't work was:

return self._oleobj_.InvokeTypes(0x10153, 0x0, 1, (11, 0),
                                              ((12, 1), (3, 1)),
doubleInfoIn, typeWanted)
                                               ^^^
where doubleInfoIn was the array that was being mangled.

Using a little brute force, I determined that the call will work correctly
with either:

return self._oleobj_.InvokeTypes(0x10153, 0x0, 1, (11, 0),
                                              ((8197, 1), (3, 1)),
doubleInfoIn, typeWanted)
                                                ^^^
or

return self._oleobj_.InvokeTypes(0x10153, 0x0, 1, (11, 0),
                                              ((12293, 1), (3, 1)),
doubleInfoIn, typeWanted)
                                                 ^^^

This indicates that a type flag of either VT_ARRAY | VT_R8  or
VT_ARRAY | VT_VECTOR | VT_R8 is required.

(I can also get things to work with type values of 8198, 8199, 12294, and
12295, although it's beyond me why they should work...)

With this correction, I can pass in the array as a tuple, a list, or a tuple
within a
tuple, and it all works fine.

The type library information for this particular call is:

[id(0x00010153), helpstring("Select enities by ray fire")]
VARIANT_BOOL SelectByRay(
[in] VARIANT doubleInfoIn,
[in] long typeWanted);

and the SolidWorks C++ code for handling this function is:

BOOL IModelDoc2::SelectByRay(const VARIANT& doubleInfoIn, long typeWanted)
{
 BOOL result;
 static BYTE parms[] =
  VTS_VARIANT VTS_I4;
 InvokeHelper(0x10153, DISPATCH_METHOD, VT_BOOL, (void*)&result, parms,
  &doubleInfoIn, typeWanted);
 return result;
}

Since the type library doesn't indicate anything other than a VARIANT for
input, I don't know how one is supposed to figure this out in advance.
However, if I can provide more information that would help allow makepy to
better handle these types of arrays, please let me know.

And thank you, Mark, for your generous time and assistance!

Jeff

----- Original Message -----
From: "Mark Hammond" <mhammond at skippinet.com.au>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, February 15, 2002 8:33 PM
Subject: Re: win32com and Passing SafeArrays of Doubles


> I believe it will simply be a matter of getting the correct "type tuple"
> in the generated code.  I am afraid I don't know VB well enough to know
> what the extra set of parens used by VB is.  It could be some special
> "byref" annotation, or maybe even specifying another level of
> indirection (ie, an array of arrays of doubles).
>
> I suggest you try and determine exactly what that VB syntax does - you
> may then need to craft a type typle that indicates a 2 dimensional
> array, and actually pass ((1,2,3),) in your Python code.
>
> Good luck, and feel free to mail if you need more help - getting this
> sorted would be good - we may even identify that pythoncom is being less
> then optimal <wink> in these edge cases.
>
> Mark.
>
> Jeffrey N. Shelton wrote:
> > In the process of using win32com to hook Python into SolidWorks (a 3D
> > modeler), I've run into the same problem as others before me -- namely,
some
> > strange things happen when passing SafeArrays of doubles into
SolidWorks.
> > Having created the appropriate gen_py file, I'm usually successful when
I
> > pass in a tuple with the appropriate values. But when passing SafeArrays
of
> > doubles into SolidWorks, it thinks I've inserted an extra value of 0.0
> > between each parameter I've passed. So it thinks that an array (1, 1, 1)
is
> > instead (0, 1, 0, 1, 0, 1).
> >
> > Past discussion at:
> >
http://groups.google.com/groups?hl=en&selm=39943819.65368645%40news.onlineis
> > p.com
> >
> > Mark Hammond's reply (back in August of 2000) indicated that he thought
the
> > problem was with the SolidWorks code, and I'm quite willing to accept
that.
> > However, I've noted from the SolidWorks API documentation (quoted below)
> > that, even when using VB, some special massaging is required.
> >
> > So, do I have to do any special 'wrapping' to get SolidWorks to accept
my
> > arrays, or does this still look like its just a shortcoming in the
> > SolidWorks implementation?
> >
> > Any suggestions would be appreciated!
> >
> > Jeff
> >
> >>From SolidWorks API Documentation:
> >
> >>Passing array data to SolidWorks in Visual Basic requires the use of a
> >>Variant variable to hold the data. The procedure is to put the data
> >>into an array then to assign a Variant to the array. The variant is then
> >>passed to an API function with parentheses round it:
> >>
> >>     Dim varArray As Variant
> >>     Dim DataArray(9) As double
> >>
> >>     ' Assign values
> >>     DataArray(0) = 0.1
> >>     ...
> >>
> >>     ' Pass the data to SolidWorks
> >>     varArray = DataArray
> >>     Result = Object.Method( (varArray) )  ' Note the extra parentheses
> >>
> > are required
> >
> >
> >
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list
>




From rdsteph at earthlink.net  Wed Feb 27 20:51:45 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Thu, 28 Feb 2002 01:51:45 GMT
Subject: Deitel and Deitel Book...
Message-ID: <3C7D8DCA.119D52EC@earthlink.net>

Sorry but I can't wait to give a brief "preview review" of the new
Deitel and Deitel book, "Python How to Program" (I also posted this at
my website at http://www.awaretek.com/plf.html ).
...
Well, let's see, I bought this 1292 page book from Amazon a week ago for
$72. This is the most I have ever paid for a computer book. Heck, it may
well be the most I have ever paid for *any* book ;-))).

I am leaving on a long business trip to Asia in two days, and I just
can't wait until I get back to tell you how much I am enjoying this
book. I have read the first 340 pages, which cover the core language; I
intend to carefully read the remaining chapters, which cover various
more advanced topics, as I fly around the world from 
New York City to Amsterdam, to Singapore, to Taiwan, to Japan, to Korea,
on to Shanghai, China, and finally back home to New York. I am looking
forward to the book more than the trip ;-))))

This book is one of the most enjoyable I have ever read. The pedagogical
approach that the authors use is superb; they have obviously learned
their educational craft well; for me, it actually works. The quality of
the paper is the finest in any Python book yet; the text is large and
clear. The way actual code is highlighted and separated from the text is
extremely 
helpful. Everything is first class.

I actually sort of expected the book to be a second rate effort in the
sense that the Deitels have written several best sellers about other
languages, and I sort of expected them to half-heartedly do a me-too
book by just applying the same old formula to a Python version.

But the book is fresh and clearly thought out. Since I have never read
any other Deitel book, it is possible that those who have done so may
have a different experience.

But I think even Deitel veterans will be pleased. More than anything
else, the book reminds me of a great college textbook, and reading such
a text is something I haven't done in way too many years. But this
educational quality of the text brings it to a whole new level of
sophistication, compared to most computer programming books.

The book includes, at the end of each chapter,  extensive questions and
answers, review lists, and summaries. It actually feels like you are
studying for credit in a college course. The end result is that, after
carefully reading the chapters covering the core language, I feel like I
have consolidated my Python knowledge considerably and that I am both
ready to do some really serious coding (once I get back from Asia ;-)))
and also to successfully attack the advanced chapters while I am on the
planes.

In addition to excellent coverage of the core language, the book covers
CGI, XML, Databases, XHTML (quite extensive) , Data Structures,
Multimedia, OpenGL, Tkinter, CSS, Python Server, Networking, Threading,
Regular Expressions, and Unicode. There are even excellent and extensive
appendixes reviewing Octal, Hexadecimal, and Binary Number Systems;
HTML; and Python version 2.2 Additions. Most of these topics are not
just skimmed over but are covered in great detail. yet the book really
flows well. The exercises and questions and answers are equal to the
very best college text books  I have encountered. Even the summaries
after each chapter are useful; I think I will be reviewing the chapter
summaries on the plane and later again after a few weeks, in order to
consolidate my learning even more. It seem like a good use of my time,
because the summaries are well thought out and very helpful.

I am excited.

I will do a detailed and very thorough review of the whole book as soon
as I return home from my trip. But even based on my reading of only the
first 340 pages, this book is a winner and I think any 
Pythonista who is willing to part with the (admittedly huge) sum of $72,
will be richly rewarded. One could easily get engrossed in this fine
book and wind up absorbing all 1292 pages in 10 days of blissful mind
mongering. (Well, what the hell, its only a preview review and I can
coin an asinine phrase like "mind mongering " if I want to ;-))))

...
By the way, I have studied the code of Hans Nowak's mygale.py well
enough to be confident in making modifications and additions, and I am
really looking forward to doing that as soon as I return  home. I intend
to add more starting point sources for the article searches, to add the
capability to search for any given topic, and also trying to fine tune
the searching mechanisms. I even have some other ideas for the code, but
I'll keep them quiet for now. Quite basic coding of course, the real
work has already been done by Hans and I really appreciate his
open-sourcing of this fine program.

I'm still searching for more tutorials and I'm up to 24 links now. I
have a lot of other stuff I want to add when I get back in mid March. 

Ron Stephens
< a href = "http://www.awaretek.com/plf.html">Python City</a>


From tjreedy at home.com  Wed Feb 20 11:13:33 2002
From: tjreedy at home.com (Terry Reedy)
Date: Wed, 20 Feb 2002 16:13:33 GMT
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com> <7xu1scew8m.fsf@ruckus.brouhaha.com> <slrna76g9l.rag.yhatakka@kenny.evolvit.fi> <7xeljg7h7b.fsf@ruckus.brouhaha.com> <RtNc8.17064$Zp2.1755158@atlpnn01.usenetserver.com>
Message-ID: <NWPc8.55790$X64.20050659@news1.rdc2.pa.home.com>

Someone who does not understand Python sematics claimed (incorrectly):

> > > > That's ok in Scheme but I don't think any current Python
implementations
> > > > handle tail recursion "correctly".

and clarified

> > By "correctly" I mean the code should be executed without pushing
> > another recursive call onto the call stack for every list element.
> > The recursive call should instead be compiled as a jump back to
the
> > top of the function.

Steve Holden responded

> All the tail recursion optimization does is speed things up slightly
and
> avoid stack exhaustion at high recursion depths. No wonder you felt
the need
> to quote "correctly".
>
> For what it's worth, no current implementations of Python implement
such
> tail recursion optimization (AFAIK).

Correct.  As Guido explained when this same issue came up some months
ago ...

Automatic (hidden) tail recursion removal would be semanticly
*incorrect* for Python because it assumes (incorrectly, in general)
that the association between a particular name and a particular
function object is permanent.  In other words, the compiler cannot
tell whether a return statement that looks like tail recursion really
is tail recursion or will remain tail recursion even when it initially
is.  Python is not Scheme.

Consider the following legal and possibly even sensible (when fleshed
out) code:

def f1(val): return 1

def f2(val): return 2

def f(val):
  global f
  f = val and f1 or f2
  return f(val)

>>> f
<function f at 785ee0>
>>> f(0)
2
>>> f
<function f2 at 786be0>

When the assumption of fixed association *is* correct, the programmer
can easily do the standard translation to loop form if desired.

Terry J. Reedy





From op73418 at mail.telepac.pt  Mon Feb 25 16:23:13 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 21:23:13 +0000
Subject: On inner classes
References: <mailman.1014665653.16956.python-list@python.org>
Message-ID: <e8al7us7vtknrn86cfd5a7rii9b3clt21b@4ax.com>

On Mon, 25 Feb 2002 12:31:38 -0700, "Bjorn Pettersen"
<BPettersen at NAREX.com> wrote:

>> From: Gon?alo Rodrigues [mailto:op73418 at mail.telepac.pt] 
>> 
>> Suppose we have something like
>> 
>> class <name>:
>>     class <name>:
>>         <whatever>
>>     <whatever>
>> 
>> that is, a class defined inside another. How can the inner 
>> class acess the methods/attributes of the outer one?
>
>I'm assuming you're meaning something like:
>
>  class Outer:
>     def outerMethod(self):
>        class Inner:
>           def innerMethod(this):
>              self.outerMethod() # call class Outer's outerMethod
>
>-- bjorn

Actually no. Let me be more specific

class outer_class:
    def __init__(self):
        self.example = "just a test"

    class inner_class:
        def __init__(self):
            <whatever>

and now, specifically:

How in the __init__ of the inner class do i acess the instance
attributes declared in the __init__ of the outer class? In this case the
instance attribute example?



From phr-n2002a at nightsong.com  Mon Feb 18 05:03:02 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 18 Feb 2002 02:03:02 -0800
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>
Message-ID: <7xofinyund.fsf@ruckus.brouhaha.com>

Ville Vainio <vvainio at karhu.tp.spt.fi> writes:
> Briefly put: can it be done? I'm mainly concerned with the price tag
> on mxODBC...

There are a bunch of database modules listed at:

   http://www.python.org/topics/database/modules.html

None seem really satisfactory but they can probably get you started.


From phd at phd.pp.ru  Fri Feb  8 16:59:57 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Sat, 9 Feb 2002 00:59:57 +0300
Subject: Please stop responding to Rue
In-Reply-To: <a41c9d$bf4$2@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Fri, Feb 08, 2002 at 08:26:21PM +0000
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <mailman.1013159835.28909.python-list@python.org> <a41c9d$bf4$2@newshost.accu.uu.nl>
Message-ID: <20020209005956.A23706@phd.pp.ru>

On Fri, Feb 08, 2002 at 08:26:21PM +0000, Martijn Faassen wrote:
> Oleg Broytmann <phd at phd.pp.ru> wrote:
> > It is amazing how fast a dozen of lamers, few trolls and a free software
> > advocate can destroy the good old friendly spirit of c.l.py.
> 
> Am I the only one who noticed you including this 'free software advocate'?
> I remember a few amusing threads with you in it. :)

   Yes, I must admit I am a part of the problem.

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



From logiplexsoftware at earthlink.net  Thu Feb  7 17:07:55 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 7 Feb 2002 14:07:55 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <3C62EE4A.5090305@destiny.com>
References: <3C62EE4A.5090305@destiny.com>
Message-ID: <20020207140755.364d5d77.logiplexsoftware@earthlink.net>

On Thu, 07 Feb 2002 16:14:50 -0500
Michael Chermside wrote:

> >> Really? What an unusual idea - I didn't realise we had such a surplus
of
> >> programmers to advocate something like this :-)
> > 
> > This is a little more "extreme" than I was referring to as well.  The
idea
> > of two programmers sitting at the same computer doesn't really go over
well
> > with me... I was referring having two programmers sit side-by-side at
> > separate computers, working on different parts of the same program
versus
> > having one programmer work in isolation.  Indeed, I don't know many
> > programmers with the patience to sit and watch someone else type ;) 
> 
> 
> Ermmm.... No.
> 
> When XP or Agile Programming advocates talk about "pair programming" 
> they really DO mean one keyboard, two programmers.

Yes, I was aware of this from the previous post.  I was simply disagreeing 
;)

> Peter Milliken writes that he didn't know we had such a surplus of 
> programmers -- this is a common misconception. Suppose that, working on 
> your own, you get X amount of coding done in a day. If we could spare a 
> second programmer to stand there and just help you out -- someone to 
> bounce ideas off of, to watch for typos, to suggest things that you had 
> forgotten -- then you'd get MORE done... say X * k.

I can get this by having someone in close proximity who I can interrupt on
occasion.  He doesn't have to watch the _whole_ time, just when there is
something interesting or puzzling or when there are design issues to be
discussed.  I do the same for him and since we're able to both code
independently _and_ collaborate, I suspect k is even higher than in
pair-programming.  Not unlike open-source development, but in this case the
developers have a faster, better means of communication (face-to-face
versus the net).

> If k is 1.05, then you are getting a 5% speedup by having someone around 
> to help out. If k were 3, then you'd get THREE TIMES as much done. If 
> THIS were the case, then it'd be worth it to your company to tell 
> someone else to stop working on their own and just help you out! The 
> break-even point is somewhere around k=2.
> The exact value of k depends on LOTS of things, including the 
> personalities of the programmers involved, the kind of work involved, 
> etc, but I have heard several different annecdotal estimates that place 
> k at just-less-than 2. (Maybe k=1.9 +/- 0.2 for typical situations.)

If if.  No doubt things such as personality are going to be a huge factor. 
I guess that sums up my assertion: the type of personality attracted to
programming is not the type to endure watching someone else code.  My
fingers would twitch themselves numb.  To have pair-programming work you
would need two programmers who: 

1.  have similar skill levels
2.  like and respect each other
3.  have compatible programming philosophies

Did I mention you need two programmers?  My current employer has only one -
yours truly.

Besides, if I understand correctly, k = 1 = normal productivity for a
single programmer.  Then your own "anecdotal" estimates place
pair-programming just below individual programming (for 2 individual
programmers, k = 2, for 2 pair-programmers, k = 1.9), so what's the point?

> If k is close to 2, then there it might NOT be a stupid idea to put 
> programmers in pairs, even without a surplus. Particularly because we 
> expect some additional benefits in addition to the coding speed, like 
> fewer bugs and better sharing of the knowledge among the coding team.

No doubt, assuming k is close to 2.  That would seem difficult to measure
with any degree of accuracy and so seems like a big assumption.
 
> As for Cliff Wells' contention that he doesn't know many programmers who 
> would want to do this, I have to differ. There are clearly some 
> programmers who absolutely hate working in pairs, but there are also 
> many others who LOVE it.

That may very well be the case.  I obviously have never tried it (and
probably won't, since the idea would never fly with anyone I've ever worked
for), plus I've haven't worked at dozens of companies with hundreds of
programmers to give me the empirical data I would need to make any definite
statements about it.

> Now I just wish I could convince my company to consider doing pair 
> programming. Because for *ME* personally, k is around 3 or 4.

Working by myself, k for me is already around 3 or 4 because that's how
much work I do around here: enough for 3 or 4 people... I'm trying to
reduce k, dammit ;)

I wonder: would you be the person doing the typing or the person doing the
watching?  For the person doing the actual coding, I can see where this
wouldn't be so bad (it'd probably be great), but I find it difficult to
believe that watching someone else code would be something any programmer
could love for very long.

Interestingly, your last paragraph brings to mind previous posts on this
thread where code reviews were discussed and I asserted that I didn't
consider them useful, not because they wouldn't be effective, but because
programmers won't do them.  In this case, it's the managers who won't do
it.  Even if pair-programming is very effective and programmers love it, I
doubt that managers and employers are ever going to be very receptive to
the idea, which IMHO makes it somewhat less than useful.

Regards,

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From rxg218 at psu.edu  Thu Feb 21 19:11:03 2002
From: rxg218 at psu.edu (Rajarshi Guha)
Date: Thu, 21 Feb 2002 19:11:03 -0500
Subject: running a program via exec*
Message-ID: <a54294$l0e@r02n01.cac.psu.edu>

Hi,
  this is somewaht related to my earlier post. When I use execl(progname) 
it seems to work. However it always looks for the progrma in the current 
directory. Is there any way to specify the path from the environment 
without having to type out the path to the program itself?

TIA,

-- 
-------------------------------------------------------------------
Rajarshi Guha                  | email: rajarshi at presidency.com
417 Davey Laboratory           | web  : http:// www.jijo.cjb.net
Dept. Of Chemistry             | ICQ  : 123242928
Pennsylvania State University  | AIM  : LoverOfPanda
-------------------------------------------------------------------
GPG Fingerprint: DCCB 4D1A 5A8B 2F5A B5F6  1F9E CDC4 5574 9017 AF2A 
Public Key     : http://pgp.mit.edu/
-------------------------------------------------------------------
All syllogisms have three parts, therefore this is not a syllogism.



From bkd at graphnet.com  Thu Feb  7 10:00:09 2002
From: bkd at graphnet.com (Bruce Dykes)
Date: Thu, 7 Feb 2002 10:00:09 -0500
Subject: [NEWBIE] ftplib again...
Message-ID: <01b901c1afe8$261e06a0$5102020a@graphnet.com>

Okay...I've got the ftp connections working...thing is, when I did my
initial prototyping, I used a local file, and readlines() to get the data
from the file into a list, and to create the function to parse out the
useful data from noise. Well it worked as far as getting the processing
logic down, but now that I'm trying from ftp'ed files, well...

All I need to do is get the file contents into a list. That's it. Now it
looks as if retrlines("RETR filename") will do that, with some additional
manipulations.

But it doesn't. Every single variation I've tried,
dataset=[host.retrlines("RETR filename") ],  host.retrlines("RETR filename",
dataset.append) , dataset=list(host.retrlines("RETR filename") ), etc.
always results in the file contents getting dumped to the
console/stdout/interactive window, and the completion code, '226 Command
Completed Successfully' getting assigned to the variable.

What am I missing? Do you need to see more of my code?

thanks in advance...
bkd




From sdm7g at Virginia.EDU  Tue Feb 19 14:42:12 2002
From: sdm7g at Virginia.EDU (Steven Majewski)
Date: Tue, 19 Feb 2002 14:42:12 -0500 (EST)
Subject: Finding nonprintable characters?
In-Reply-To: <Pine.OSX.4.43.0202191420260.1674-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <Pine.OSX.4.43.0202191437300.1674-100000@d-128-61-180.bootp.virginia.edu>

I meant to add:
 The reason for doing read rather than readlines or xreadlines is
that if it's a binary file, your 'lines' could be very long...
 Also: you could filter for CR and LF and count those per block
as an additional test -- although there may be autowraped text
files that are also made of a couple very long 'lines' .

-- Steve Majewski





From logiplexsoftware at earthlink.net  Fri Feb 22 22:23:04 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 22 Feb 2002 19:23:04 -0800
Subject: Newbie: Crazy, but Quick
In-Reply-To: <slrna7dkti.5et.quinn@regurgitate.ugcs.caltech.edu>
References: <u7ag65cbc0q3e@corp.supernews.com>
	<CMfd8.5037$D6.14223@news.iol.ie>
	<u7b81629bo9o4a@corp.supernews.com>
	<slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu>
	<mailman.1014399281.7163.python-list@python.org>
	<slrna7dkti.5et.quinn@regurgitate.ugcs.caltech.edu>
Message-ID: <20020222192304.2b94604a.logiplexsoftware@earthlink.net>

On 22 Feb 2002 23:26:42 GMT
Quinn Dunkan wrote:

> On Fri, 22 Feb 2002 09:23:24 -0800, Cliff Wells
> <logiplexsoftware at earthlink.net> wrote:
> >For that matter, you could also store the Python objects directly in the
> >SQL database.  My own opinion is that SQL is so easy that it's a
no-brainer
> >to use it whenever possible.  A problem I've seen many times is people
> >utilize some shortcut because they feel the program will never need
> >anything more, only later to discover that the program has utility far
> >beyond what they expected, but the shortcut they've taken cripples it,
and
> >they end up rewriting (or worse, they don't).
> 
> I don't find SQL to be a no-brainer.  You have to download the thing
(MySQL at
> least is massive... the sunos binary tarball I have is >5MB).  Then you
have to

5MB?  And how big was /your/ Python download?  The RPMs for my system were:
Python 2.2: 8MB
Python docs: 2MB

Hm.  5MB is looking pretty good.

I won't mention wxPython, but maybe I shouldn't use that either, since it's
almost 3MB.

> install it (create tables, blah blah, learn MySQL's complicated
permission
> system, etc.).  Then you have to turn it on (which means either getting
root to
> edit the system rcs or setting up some hackery to restart it if the
server
> reboots).  Not to mention that you have to learn enough SQL to do all
this.
> Oh yes, and if there are no binaries for your platform, I hope you happen
to
> have a decent C++ compiler lying around.

I was referring specifically to the Windows port to which none of these
things (other than setting user permissions and creating tables) applies. 
You download a single .zip (yes, it's fairly large ~ 12MB, but still very
doable, even over 56K - I'm doing it at this moment, just for fun ;), unzip
it, run the program named "install.exe", read the file named README (still
with me?),  and run a single command (the secret of which is revealed in
the the README) "mysqladmin.exe".  From that point on, MySQL runs
automatically upon boot.  Anyone who can't manage that shouldn't bother
trying to write an application (unless Rue's 9 commands are done by then). 

> And even if it is easy to introduce giant complicated support program,
should
> you?  If linking in a massive shared library (or *any* shared library) is
as
> easy as tacking on a '-lpiggy', is it a no-brainer to link it in whenever
> possible?  I don't think so.  How easy it is for the developer to create

Absolutely agree.  But when writing a database application, it makes sense
to use a database.

> something is only part of the equation (that's what the python "creed" of
> legibility and simplicity is about, yes?).

Sure.  And using MySQL will simplify your code, because it handles the data
for you.  Writing a report becomes as simple as doing a query, rather than
writing a couple of pages of Python to sort and group your data (oh, wait
we didn't spec reports, did we?  Well, we want them now).  In fact, if you
have Excel, you can use it to create the reports, no coding at all and you
get some nice formatting to boot (your Python code will by now be hundreds
of lines, and quite complex).

> 
> Of course, you're right that you can easily marshal python objects to
SQL, I
> assume by pickling to a string.  But then you lose the "it's not python
> specific" advantage of SQL.  And you still have to scrunch your data into
the
> relational rows/columns model.  What if it's hierarchical and lumpy?

Obviously, modeling hierarchical data in a relational database isn't as
easy, but is that really on-topic?  The OP's data sounded pretty relational
to me.  If I /did/ need to model hierarchical data, I would try to find a
database that does it already, versus reinventing the wheel.  Besides, I
wouldn't think a programmer who can deal with that sort of thing should
have any problem installing a database, would she?

> 
> >What if she decides to maintain historical data?  Then to keep the files
> >reasonably sized, she'll have to start juggling files.
> 
> I think the *dbm type stuff scales ok up to mediumish files at least
(there are
> a few > 50MB gdbms lying around, INN uses it for article index or
something).
> I have no idea if it's happy all the way up to a few gigs.  If I wanted
to use
> it for that much data I'd test it out.

And I'd still argue that using MySQL is as easy as using dbm, and I won't
need to wonder if it will scale.  Not to mention it'll be /slightly/ faster
;)

> >Software has a tendency to grow beyond the developer's expectations, so
I
> >consider it unwise to take shortcuts when a better long-term solution
isn't
> >much more difficult to implement.
> 
> That's one point of view, and a reasonable one.  Another point of view is
that
> you should only design for what you know you need, instead of increasing
> generality and complexity to deal with some hypothetical future
situation.
> If you think your program has acquired a new requirement, you decide if
> it's really a requirement or if it should be done by another program, and
if
> it is, you modify your program, which hopefully is easy to modify since
you've
> been trying to keep things simple.

Yes and no.  I would agree with this if you are the only one who will ever
use the software, or you are writing a simple, single-purpose utility.  If
other people will be using the software, then I can guarantee you that you
don't "know what you need", because people will invariably want more
features, ways to interface your program with their existing software, etc.
 If you store your data in a SQL database, you're halfway home.  How about
making your database available on a network?  Seems like a reasonable
request, but your dbm/flatfile based code is going to get more complex (and
fragile), while the SQL-based solution will remain more or less the same.

> In the shelve case, it would be pretty easy to write an SQL-using backend
> that implements the shelve interface.  And maybe you'll like that
interface
> better than the SQL-ish cur.execute(this_and_that) anyway.

I probably wouldn't actually do this unless there were a very good reason
to.  I have in the past, actually stored compiled Python byte-code in a SQL
database (I had a good reason ;), but in general, I haven't had much
problem with the standard SQL datatypes and query language.

> >Not only that, but storing the data in a SQL database separates the data
> >storage from the program, so in the future she could use that data from
> >some other program, not just the Python one.  Then if she decides she
wants
> >to access the data from, say, a web server, it's not a big problem. 
> 
> You can write CGI scripts in python :)

Easier than using MySQL?  That's arguable.  Besides, web access was a
single example.  What if she wants to analyze the data in Excel?  Mark
Hammond's win32 stuff is nice and makes such things reasonably easy, but
still more far more difficult than accessing the data via ODBC.  And now,
you've got three access methods to support: Python, CGI, and Excel.  This
is the point behind n-tier design: separate data and interface from program
logic.  Using an external database is an easy and flexible way to separate
the data.  This is especially important when you consider that most of the
time, the data is more valuable than the program that generated it - what
happens when you leave that program behind and some poor sod is left trying
to extract all your data so it can be used in a more scalable program?

Sorry if I sound like an jerk - I've had a long week hacking on someone's
code for a custom database that had to grow beyond it's means.  Now if he'd
only used SQL...

Regards,

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From shalehperry at attbi.com  Fri Feb 22 18:20:43 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Fri, 22 Feb 2002 15:20:43 -0800 (PST)
Subject: Request for comments: Search and replace script
In-Reply-To: <3C76AF02.2070101@verio.net>
Message-ID: <XFMail.20020222152043.shalehperry@attbi.com>

> 
> I would be very appreciative of any comments or patches to it.
> Particularly, I am interested in:
> 
> - Things that could make it faster
> - Things to make it more 'pythonic'
> - Bugs or logic errors
> 

tabs, ick, tabs .....

please try to keep a line down to 78 chars

exiting in a constructor is not nice, especially when you are planning on having
  others use your code.  Just throw an exception.

check() makes some of the same checks over and over, perhaps you should look at
  the code's flow here and see if it can be done differently

globals are also icky

why is usage a) a list (it could have been a tuple) and b) defined at all.  You
  could just read __doc__ and make it the file's doc string



From jeff at ccvcorp.com  Thu Feb  7 14:29:41 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 07 Feb 2002 11:29:41 -0800
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <a3u9sk$890$1@news.mathworks.com>
Message-ID: <3C62D5A5.4B57F487@ccvcorp.com>


Joshua Marshall wrote:

> Another possibility:
>
> from mymod import *
>
> def dosomething(x):
>     import mymod
>     mymod.dosomething(1000)

Nope,this does not work.  As the O.P. already noted....


> > I tried mymod.dosomething(1000) but it returns an error:
> > AttributeError: 'builtin_function_or_method' object has no attribute
> > 'dosomething'

from module import *  does not create a reference to module in the current
namespace, so there's nothing there to look up attributes from.

I still think that the O.P. is probably better off just using the explicit 'import
mymod; mymod.dosomething()' notation, since that is almost always the safer way to
do things, but that's a different story.  :)

Jeff Shannon
Technician/Programmer
Credit International





From xscottgjunk at yahoo.com  Mon Feb 18 12:59:53 2002
From: xscottgjunk at yahoo.com (Scott Gilbert)
Date: 18 Feb 2002 09:59:53 -0800
Subject: Embedding: How to set globals for PyObject_Call
Message-ID: <79b2b0bb.0202180959.16ae249c@posting.google.com>

Hi All.

I'm embedding Python in another library.  Basically, I'm treating
Python as a sophisticated data structures library, statement executor,
and expression evaluator.  The sample below is just to illustrate my
problem, not the actual code I'm using.

The problem is that while PyRun_String(...) allows me to pass in the
global variables dictionary, PyObject_Call(...) does not.

I dove through the code, and it appears that I need to setup a
PyFrameObject to be associated with my PyThreadState.  I wasn't able
to figure out how to do that though.

As you can see in my example below, I have the globals sitting right
there.  I just don't know how to tell Python that those are the
appropriate globals while invoking PyObject_Call(...).  A similar
problem occurs when evaluating "globals", "locals", "dir", or
functions that call those.

Is there a way around this?

Thanks in advance for any and all assistance!



#include <Python.h>

int main (int argc, char* argv) {
  PyObject* module;
  PyObject* globals;
  PyObject* vars;
  PyObject* tuple;
  PyObject* dict;
  PyObject* result;

  Py_Initialize();
  module = PyImport_AddModule("__main__");
  globals = PyModule_GetDict(module);

  vars = PyRun_String("vars", Py_eval_input, globals, globals);
  tuple = PyTuple_New(0);
  dict = PyDict_New();

  result = PyObject_Call(vars, tuple, dict);

  if (!result) {
    printf("Had a problem calling vars()\n");
    PyErr_Print();
  }

  /* skip the decrefs and finalization */

  return 0;
}


From aequitas at planet.nl  Tue Feb 19 16:23:44 2002
From: aequitas at planet.nl (Johan)
Date: Tue, 19 Feb 2002 21:23:44 GMT
Subject: Alternative for \n when using telnetlib
Message-ID: <1103_1014153824@Johan>

When connecting to my adsl modem (alcatel speed touch pro) via telnetlib i can read the date write my username. But then i when i give the \n to proceed to the password it doesn't work. In interactive mode it has the 
same problem. It just writes a new line. 

Is there a alternate for the \n command?



From brian at sweetapp.com  Fri Feb  8 17:38:33 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Fri, 8 Feb 2002 14:38:33 -0800
Subject: TRYING to like python, but...
In-Reply-To: <a41h0m$ear$1@web1.cup.hp.com>
Message-ID: <001601c1b0f1$57498f10$445d4540@Dell2>

jimh wrote:
 
> I untarred, configured and built python 2.1.2.  It all seems to be
working
> fine.

You probably should take Trent's advice and just use ActivePython (even
though he's been slacking and doesn't have a 2.1.2 build yet).

>    xml.sax._exceptions.SAXReaderNotAvailable: No parsers found
> 
> Well, THAT'S helpful.

Seems like a useful error message to me.
 
> Nothing in Setup about xml.dom or xml.sax, so I assume that Setup is
ok.
> Besides, I WAS able to import and start using xml.dom.

xml.dom is written purely in Python so you aren't dependant on an
external C library. 
 
> So I get online and start searching around for this error message.
Find
> some vague references to needing "expat" (whatever that is).  

Expat is a C library for XML parsing. It was written my James Clark, who
claims that it is the fastest parser in the world.

> Ok, I remember vaguely something about expat in Setup, so I go back to
> Setup and see that TO USE A STANDARD PYTHON LIBRARY, I need to
DOWNLOAD
> ADDITIONAL code (expat) from SOME PLACE ELSE.

The problem is that expat is not a project run by the crew at
PythonLabs; it's a third-party product. I think that you will find this
to be the case with all scripting languages (it certainly is with
Perl/PHP/Tcl).

Maybe someone (not volunteering) should try to provide binaries for
novice Linux users?

> So I ./configure again, and Make again, and, guess what??? IT STILL
> DOESN'T WORK!!!  (xmlparse.h: no such file).

Do you look for that file? Is it where you expected it to be?

> To be honest, if python is really this difficult to use, I am going to
try
> something different.

Why don't you try something different and then tell us if it was easier
or not.

Cheers,
Brian




From sjmachin at lexicon.net  Tue Feb 26 17:02:54 2002
From: sjmachin at lexicon.net (John Machin)
Date: 26 Feb 2002 14:02:54 -0800
Subject: Strange problem with C API
References: <71c0a06e.0202260513.4b67841f@posting.google.com>
Message-ID: <c76ff6fc.0202261402.10060a2@posting.google.com>

amrit040 at yahoo.com (Amrit) wrote in message news:<71c0a06e.0202260513.4b67841f at posting.google.com>...
> Hi 
> 
> I'm trying to write a C function callable from python. I'm getting
> some strange results :

>   float x[N];
>   float Thr;
>   float sum;
>   float  max,min;

Don't you mean double???????????

>       obj[i] = PySequence_GetItem (vec, i+1);

(1) This is Python & C, not FORTRAN. Lose the '+1'.
(2) There is no need to have an array obj[], a scalar obj will do.
(3) You need to test the result.

obj = PySequence_GetItem (vec, i);
if (obj == NULL) return NULL;
// :-) This is typically author stuffup, not caller stuffup.

>       PyArg_Parse (obj[i], "f", &x[i]);
(1) This is most likely what is causing the exception that you are
getting -- the first arg to this function should be a *tuple*.
(2) This is not a recommended way of doing things. Try something like
this:

if (PyInt_Check(obj))
   x[i] = (double)PyInt_AsLong(obj);
else if (PyFoo_Check(obj))
   x[i] = (double)PyFoo_AsCFoo(obj);
// etc etc
else if (PyFloat_Check(obj))
   x[i] = PyFloat_AsDouble(obj);
else {
   raise_an_exception(..........);
     // You read the manual or look at others' code, to find out how.
   return NULL;
}
Py_DECREF(obj); // Don't forget this, else you will leak memory.

#### If anybody knows how to call the equivalent of the built-in
float() from C, please post.

However, what does your algorithm really need as its first arg: An
array of (Python) floats? An array of (Python) ints? A simple int
(like 15 in your example and in your comments) from which your
implementation will build an array of doubles 0.0, 1.0, ...., 14.0?

> My python implemtation of the same algo works just fine !

Famous last words?


From sorr at rightnow.com  Fri Feb 15 11:02:17 2002
From: sorr at rightnow.com (Orr, Steve)
Date: Fri, 15 Feb 2002 09:02:17 -0700
Subject: Java vs Python ... ???
Message-ID: <D408B6AC060DB24790622E1A54204DB00823AA9F@xchange.rightnowtech.com>

What about Jython? Do you get some of the benefits of Java along with
easier, more productive Python programming?

For instance, with the Oracle Application Server (not the database) you
supposedly can have all your Java apps load balanced "for free" without
having to write load balancing code. So I was wondering the other day, since
Jython compiles to bytecode, will it run under applications servers (such as
Oracle) and enjoy the same Java features those servers provide? 

I want to have my cake and eat it too!
Steve Orr, Bozeman Montana



-----Original Message-----
From: philh at comuno.freeserve.co.uk [mailto:philh at comuno.freeserve.co.uk]
Sent: Friday, February 15, 2002 8:38 AM
To: python-list at python.org
Subject: Re: Java vs Python ... ???


On 15 Feb 2002 04:53:02 -0800, Bertrand Geston <bergeston at yahoo.fr> wrote:
>
>You know, Java is my wife and Python is my mistress. 

UIs that bevcause they both demand lots of resources, and often
won't run, complaining of "insufficient cache"?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>



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



From gh_pythonlist at gmx.de  Mon Feb 11 02:16:06 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 11 Feb 2002 08:16:06 +0100
Subject: subclassing tuple
In-Reply-To: <4abd9ce7.0202102243.d078f9c@posting.google.com>
References: <4abd9ce7.0202102243.d078f9c@posting.google.com>
Message-ID: <20020211071605.GA2528@lilith.hqd-internal>

Le 10/02/02 ? 22:43, damien  morton ?crivit:
> Ive been trying to subclass the tuple type, without a whole lot of success
> 
> Heres the code Ive been using:
> 
> class triple(tuple):
>     def __init__(self, x, y, z):
>         tuple.__init__(self, (x,y,z))
> 
> >>> t = triple(1,2,3)
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: tuple() takes at most 1 argument (3 given)

Well, tuple constructors take only 1 argument :-)

The following works for me, however:

#!/usr/bin/python2.2
class doubledlist(list):
    def __init__(self, x):
        list.__init__(self, map(lambda t: 2*t, x))

x = doubledlist([1,2])
print x

> For some reason, subclasses of tuple dont call the proper __init__ code.

Yes, /me observed this too. Lists work as expected, tuples don't!?

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 6.9 ?C      Wind: 3.6 m/s



From jeff at ccvcorp.com  Wed Feb 27 13:38:03 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 10:38:03 -0800
Subject: controlling web browser in windows
References: <1103_1013605186@netnews.cf.ac.uk> <Zeua8.206395$KM2.8473296@atlpnn01.usenetserver.com> <3C6ACDB6.95186821@ccvcorp.com> <ea9p7u85b38k4gdgf5bc4mjlovq7p4ta61@4ax.com>
Message-ID: <3C7D278A.2345AD6C@ccvcorp.com>


Matthias Baas wrote:

> On Wed, 13 Feb 2002 12:33:58 -0800, Jeff Shannon <jeff at ccvcorp.com>
> wrote:
> >Of course, I then discovered that much of what I *wanted* to do, was forbidden
> >from scripts for security reasons, but that's another story.  ;)
>
> Reminds me of something.....recently I was also playing around with
> COM and IE and tried to navigate to a page and save it to disk.
> Navigating was no problem (using the Navigate() method) but saving the
> page didn't work. I tried to save it via the ExecWB() call, but that
> only raised an exception (which didn't really contain a useful error
> message. The message was in German, so I think it doesn't help much if
> I post it). So my question: Is that an example of what is forbidden
> from Python? Or is there a way to save the content to disk
> automatically?

I think that this is a security issue, yes.  But it's not a matter of being
forbidden to Python, specifically; it's a matter of being forbidden to *any*
script.  If IE could be told, by a script, to save files to your hard drive, then a
malicious script could save an arbitrary autoexec.bat file in your root
directory... or, more simply, save hundreds of copies of itself, filling your
entire drive.  In my case, I was looking to automatically send pages to a printer,
without human intervention, but IE forbids this because a hostile script could then
send hundreds of pages of junk to your printer.  (Since I was running from a .HTA,
where security restrictions are looser, I may have been able to work around this,
but decided that it wasn't worth the effort.  That program still requires human
operation, but we're now working on avoiding using the web interface altogether...)



> And another thing that confused me: When I start the COM makepy
> utility I don't see IE in the list ("InternetExplorer.Application"
> doesn't show up). Why not?

I believe (but I'm not certain) that the type library is identified as "Microsoft
Internet Controls", or some such equally vague thing.  :)

Jeff Shannon
Technician/Programmer
Credit International




From calves at coelce.com.br  Thu Feb 21 10:34:32 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Thu, 21 Feb 2002 12:34:32 -0300
Subject: BLT Setup and Installation
Message-ID: <29A97D00F387D411AC7900902770E14804047A2C@lcoeexc01.coelce.net>

Hi all,
How can I install and setup the BLT extentions to tcl/tk to use in my python
programs.
I have downloaded the Pmw and BLT packages (binaries forms) and unzip them
into C:\Python21 and C:\Python21\tcl. But when try the following code, it
crashs:

from Tkinter import *
import Pmw
root=Tk()
g=Pmw.Blt.Graph(root)
Here, the system raise and error and say that "::blt::graph" commands not
found, when callig Tk/tcl.

I'm running Python2.1 under Windows98, wich it's installed at C:\Python21

Carlos Alberto
COELCE/DPRON-Departamento de Projetos e Obras Norte
Fone: 677- 2228
e-mail: calves at coelce.com.br
\|||/
(o o)
--ooo0-(_)-0ooo--

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020221/441c606a/attachment.html>

From husam at macrozon.net  Wed Feb 20 04:54:13 2002
From: husam at macrozon.net (husam)
Date: Wed, 20 Feb 2002 10:54:13 +0100
Subject: How do you conver output of "popen('ls')" to a list?
Message-ID: <3C737245.BC5D7164@macrozon.net>

thanks for reply.




From mal at lemburg.com  Tue Feb 26 05:22:58 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 26 Feb 2002 11:22:58 +0100
Subject: PEP 263 comments
References: <200202250520.g1P5KKD01484@mira.informatik.hu-berlin.de>
		<3C7B5E35.129E5501@lemburg.com> <m31yf8fsxu.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3C7B6202.761CFAA0@lemburg.com>

"Martin v. Loewis" wrote:
> 
> "M.-A. Lemburg" <mal at lemburg.com> writes:
> 
> > Thanks for the comments. I've update the PEP at SourceForge...
> 
> Thanks! Is there anything that needs to be resolved before we can ask
> for BDFL pronouncement? According to PEP 1, the PEP editor must be
> informed that the PEP is ready for review; I suggest it is (after this
> batch of changes).

Agreed. I'll post something to python-dev...
 
> > I've added a two phase approach to the PEP: first we only
> > handle Unicode literals, then we do the whole file in a later
> > step.
> 
> Sounds good.
> 
> > I did not add the IDLE changes to the PEP. Please upload them
> > as feature request to SF.
> 
> Ok, I will.

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From usenet at thinkspot.net  Fri Feb  8 21:41:14 2002
From: usenet at thinkspot.net (Sheila King)
Date: Fri, 08 Feb 2002 18:41:14 -0800
Subject: email module Bug in 2.2
Message-ID: <a4164b.3vv7l9p.1@kserver.org>

OK, here is a bug in the mail module. It has to be????

I took an email that I received (a spam) and saved it to a text 
file (which I called 'dtv.txt') and then imported it into an email class
object in an interactive session. 

I've pasted below an interactive session, and below that I've pasted the text 
file that was used in the session. There seem to be some inconsistencies?

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



E:\Web\Thinkspotwebsite\dev>python
Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import email
>>> msg = email.message_from_file(open('dtv.txt'))
>>> msg.get_type()
'multipart/related'
>>> msg.get_main_type()
'multipart'
>>> msg.is_multipart()
0
>>>


Here is the text file dtv.txt:

>From DirecTV4Free!!!!@yahoo.com Fri Feb 08 04:45:43 2002
Return-Path: <DirecTV4Free!!!!@yahoo.com>
Delivered-To: xthinkspot-usenet at thinkspot.net
Received: (qmail 3201 invoked from network); 8 Feb 2002 09:45:03 -0000
Received: from unknown (HELO mx.futurequest.net) (130.91.184.200)
  by pt01.futurequest.net (63.151.147.170) with SMTP; 08 Feb 2002 09:45:03 -0000
From: "DirecTV 4 Free" <DirecTV4Free!!!!@yahoo.com>
Date: Fri, 08 Feb 2002 04:45:43
To: usenet at thinkspot.net
Subject: DirecTV 4 Free!!!!
MIME-Version: 1.0
Content-Type: multipart/related;
  boundary="----=_NextPart_MIMSCHNGDM"
Content-Transfer-Encoding: 7bit
Message-ID: PM20004:45:43 AM

This is an HTML email message.  If you see this, your mail client does not support HTML messages.

------=_NextPart_MIMSCHNGDM
Content-Type: text/html;charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

<title>Do you have a DirecTV system</title>
<style>
<!--
DIV.Section1 {
    page: Section1
}
SPAN.GramE {
    
}
H2 {
    FONT-WEIGHT: normal; FONT-SIZE: 12pt; FONT-FAMILY: "Times New Roman"; margin-left:0cm;
margin-right:0cm; margin-top:0cm; margin-bottom:0pt
}
-->
</style>
</head>

<body>

<p>&nbsp;&nbsp;&nbsp;&nbsp; Do you have a DirecTV system? Are you tired of 
paying outrageous prices for what you have? Want more? Are you tired of sending 
your card out to other people and waiting days, possibly paying fees, every time 
DirecTV hits your card? Now you can re-set your own DIRECT TV Access Card 
yourself. You can get all the DIRECT TV channels for free. That's over 300 
channels, including all NFL, NHL, NBA, &amp; Sports, HBO, Pay-per-View, Movies &amp; 
Events, XXX and Local Channels. You can do this by using our simple access card 
activating device. It comes complete with the software you need, and simple 
detailed instructions as well as full support. All you need is a home computer.</p>
<p>Not want to mess around with your card? We provide a service where you can 
send your card to us and we load a PRIVATE Script on your card. The script we 
have does not suffer the shutdowns from DirecTV like others may. The last time 
our script was hit was in Oct. 2001, but if it does happen, we will fix it free 
of charge. Not want to send your card? No problem, we can provide you with a 
brand new card loaded with our software, you just put it in your receiver and 
enjoy the shows.</p>
<p>Not have DirecTV? Paying outrageous amounts for your cable that you only get 
a few channels on? Maybe not have cable available in your area or the &quot;other&quot; 
satellite company isn't providing what you expect? We also have all of the 
satellite equipment available in complete packages. DirecTV with TiVO, Ultimate 
TV, etc. are also available.</p>
<p>Live in an area where high speed broadband internet access is not available? 
We have cutting edge technology that can give you broadband internet access 
while watching your FREE DirecTV programming.</p>
<h2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</h2>
<h2>&nbsp;</h2>
<h2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Want More Information????</h2>
<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Then </p>
<blockquote>
  <blockquote>
    <p class="MsoNormal"><span style="font-size: 12pt">

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span><span
style="font-size: 18pt">
    <a style="color: blue; text-decoration: underline; text-underline: single"
href="mailto:freesat12000 at yahoo.com?subject=I Would Like More Information">
    CLICK HERE</a>.&nbsp; </span></p>
  </blockquote>
</blockquote>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<p class="MsoNormal"><span style="font-size: 12pt">&nbsp;</span></p>
<h2>&nbsp;&nbsp;
<a href="mailto:freesatdelete at yahoo.com?subject=Remove From Your Mailing List Please">
Remove Me&nbsp; From Your Mailing List Please</a></h2>

</body>

</html>
------=_NextPart_MIMSCHNGDM--






From matthias_j at lycos.com  Mon Feb 25 13:24:02 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 25 Feb 2002 10:24:02 -0800
Subject: Trouble with getting remote Computer Name
Message-ID: <d7d5ebdf.0202251024.236b2c1b@posting.google.com>

Hi all.

I'm working on a small app for Win NT - to change automaticly user
names on about 70 different Win NT Server each one in a different
Domain.
I'm nearly finished just having I thing one probleme still to be
solved.

As I'm new to python I would appriciate some help.

My Problem:

I have the IP address of all the Servers in a VPN for instance
(10.251.23.6,10.251.24.6, 10.251.25.6 and so on)

How can I get from each Server the Name of the Server in python.

I was thinking of one way: maybe a "ping -a 10.251.23.6" and with
popen somehow to read the name - but there must be an easier way I'm
just not aware of it.

Any Help would be real greate.

Matthias Janes


From op73418 at mail.telepac.pt  Mon Feb 25 14:45:06 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 19:45:06 +0000
Subject: On inner classes
References: <iv3l7ukajb2mnmsi19u050iapodl4itqd4@4ax.com>
Message-ID: <iv4l7uganb3b53c6fk3jca58aaik7s1br1@4ax.com>

On Mon, 25 Feb 2002 19:29:03 +0000, Gon?alo Rodrigues
<op73418 at mail.telepac.pt> wrote:

>Suppose we have something like
>
>class <name>:
>    class <name>:
>        <whatever>
>    <whatever>
>
>that is, a class defined inside another. How can the inner class acess
>the methods/attributes of the outer one?
>

Let me be a bit more precise: I am thinking mainly of the instance
attributes of the parent instance (but the others also, of course).

>Thanks for the help,
>Gon?alo Rodrigues



From nospam at bigfoot.com  Thu Feb 21 18:23:49 2002
From: nospam at bigfoot.com (Gillou)
Date: Fri, 22 Feb 2002 00:23:49 +0100
Subject: Any Interest in EDIFACT Parser/Generator?
References: <mailman.1014245075.4124.python-list@python.org>
Message-ID: <a53v4f$u4i$1@norfair.nerim.net>

I've been looking for this some months ago !!

"Jim Kraai" <jkraai at paypal.com> a ?crit dans le message news:
mailman.1014245075.4124.python-list at python.org...
> Just curious.
>
> --jim
>




From mwh at python.net  Thu Feb 21 07:49:50 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 21 Feb 2002 12:49:50 GMT
Subject: Python crash in PyImport_ImportModule
References: <993ae4ca.0202210436.21805729@posting.google.com>
Message-ID: <usn7vkndv.fsf@python.net>

AdrianLeu at kelseus.com (Adrian Leu) writes:

> Hi ! I am getting at the end of my tether with Python. 
> 
> I have written a script in Python 2.0 under Windows 2000. I call this
> script from inside a C++ class method which has the Py_Initialize()
> and Py_finalize() in its constructor and destructor respectively.
> Everything seemed to work OK. However, recently a crash started to
> appear and it cannot be repeated always. It is in the
> PyImport_ImportModule(module_name) function and the stack looks like
> this:

I don't know if this is the cause of your problem but some versions of
Python have problems when you call Py_Initialize/Py_Finalize
repeatedly.  Perhaps you could write a tiny problem that calls these
functions in a loop and see if that crashes?  If it does, you should
probably trying upgrading your Python.  If it doesn't, you have some
other problem -- from your stack trace, I'd guess it's some kind of
heap corruption, so good luck.

Cheers,
M.

-- 
  Whaaat? That is the most retarded thing I have seen since, oh,
  yesterday                             -- Kaz Kylheku, comp.lang.lisp


From mhammond at skippinet.com.au  Wed Feb 27 19:42:43 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 28 Feb 2002 00:42:43 GMT
Subject: ActiveState version of 2.2???
References: <1001ff04.0202170702.6ebaaf55@posting.google.com> <3C704157.B872722F@bellatlantic.net> <1001ff04.0202261911.22131410@posting.google.com> <3C7D2879.F54DF385@ccvcorp.com>
Message-ID: <3C7D7D0B.7050007@skippinet.com.au>

Jeff Shannon wrote:
> For me, the big advantage of ActiveState's is that it 
 > automatically registers all of the COM and Active Scripting
 > stuff, which I've been using fairly extensively, but it's not
 > *that* difficult to do this manually.

The win32all installers also do this, and have for quite some time now 
(years).  Do you get any error messages at install time?

Mark.



From aleax at aleax.it  Mon Feb 18 04:51:23 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 09:51:23 GMT
Subject: Warning about "for line in file:"
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <3C6D73DE.3090201@wi.mit.edu> <20020216134444.GA12676@nullcube.com> <mailman.1013880747.26249.python-list@python.org>
Message-ID: <v84c8.82333$3J1.2625119@news2.tin.it>

Neil Schemenauer wrote:

> Aldo Cortesi wrote:
>> Actually this has nothing to do with iterators, or a "read
>> cache".
> 
> It does.
> 
>> iter(file) creates a line iterator that does the
>> same thing as file.readline() every time .next() is called,
>> until it reaches the end of the file.
> 
> It does not.  iter(file) calls file.xreadlines().  xreadlines()
> internally calls readlines(CHUNKSIZE).

Right.  If you want an iterator that works exactly as Aldo
just described, albeit at substantial performance cost,
iter(file.readline, '') is the handiest way to build one.

The general form for two-argument iter is
    iter(callable, sentinel)

and it works just about like:

def iter2(callable, sentinel):
    while 1:
        result = callable()
        if result==sentinel: return
        yield result

so that:

for item in iter(callable, sentinel):
    process(item)

is a handy way of expressing:

while 1:
    item = callable()
    if item==sentinel: break
    process(item)



Alex



From jason at jorendorff.com  Tue Feb 26 12:00:32 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: 26 Feb 2002 11:00:32 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb 26)
Message-ID: <5A65FB556DC4B92B.47C0A0259510F4DA.27E89041417BA1D1@lp.airnews.net>

QOTW:  "Pythonistas tend to delight in the 'incidental generality' engendered
by the language's robust polymorphism."  -Steve Holden

"Is there an idiots guide to code objects, frame objects and settrace
anywhere?" -Tim Rowe
"Anybody using those would look like a very smart idiot ..." -Martin von Loewis


    Is there a good way to script C++ from Python?  The community offers a
    wide variety of answers.
        http://groups.google.com/groups?th=f8a882ef09879fbd
        http://www.scipy.org/site_content/weave
        http://www.boost.org/libs/python/doc/example1.html
        http://www.acl.lanl.gov/siloon/
        http://root.cern.ch/root/Cint.html
    
    The Python UK Conference coincides with the ACCU Spring Conference,
    April 4 & 5 2002.  Get two conferences for the price of one!
        http://www.accuconference.co.uk
[dup?]
    
    mxCGIPython is a complete Python interpreter in one file.  It's easy
    to install and starts up fast.
        http://www.egenix.com/files/python/mxCGIPython.html
    
    Cameron Laird shows how to use SOAP.py to check air fares.
        http://groups.google.com/groups?th=2dd8799dafd0304c
    
    Justin Sheehy shows how to write a Linux daemon in Python.
        http://groups.google.com/groups?th=556689d395822992
    
    The "turtle" standard library module provides logo-like turtle
    graphics.
        http://www.python.org/doc/current/lib/module-turtle.html
    
    Two unrelated threads investigate pythoncom performance.  The second
    thread specifically looks at calling ADO from Python.
        http://groups.google.com/groups?th=87081640f325081f
        http://groups.google.com/groups?th=cd2810da4437737c
    
    Alex Martelli exposes with utter lucidity why Visual Basic,
    and not Python, is the risky and speculative technology.
	http://groups.google.com/groups?th=5773a3f58c6b95aa

    PythonTheater plays MPEG and AVI movies on Linux.
        http://xtheater.sourceforge.net/
    
    markow.py implements a classic programmer's toy: it chains words
    together to produce "dadaistic parodies of known bodies of text" -
    such as the comp.lang.python newsgroup.
        http://groups.google.com/groups?selm=mailman.1012965542.895.clpa-moderators%40python.org


========================================================================

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

    Python.org's Python Language Website is the traditional
    center of Pythonia
        http://www.python.org
    Notice especially the master FAQ
        http://www.python.org/doc/FAQ.html

    PythonWare complements the digest you're reading with the
    daily python url
         http://www.pythonware.com/daily  
    Mygale is a news-gathering webcrawler that specializes in (new)
    World-Wide Web articles related to Python.
	 http://www.awaretek.com/nowak/mygale.html 
    While cosmetically similar, Mygale and the Daily Python-URL
    are utterly different in their technologies and generally in
    their results.

    comp.lang.python.announce announces new Python software.  Be
    sure to scan this newly-revitalized newsgroup at least weekly.
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

    Michael Hudson continued Andrew Kuchling's marvelous tradition
    of summarizing action on the python-dev mailing list once every
    other week, into July 2001.  Any volunteers to re-start this
    valuable series?
        http://starship.python.net/crew/mwh/summaries/
        http://www.amk.ca/python/dev

    The Vaults of Parnassus ambitiously collect Python resources
        http://www.vex.net/~x/parnassus/   

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

    The Python Software Foundation has replaced the Python Consortium
    as an independent nexus of activity
        http://www.python.org/psf/

    Cetus does much of the same
	http://www.cetus-links.org/oo_python.html

    Python FAQTS
        http://python.faqts.com/

    The old Python "To-Do List" now lives principally in a
    SourceForge reincarnation.
	http://sourceforge.net/tracker/?atid=355470&group_id=5470&func=browse
	http://python.sourceforge.net/peps/pep-0042.html

    *Py: the Journal of the Python Language*
        http://www.pyzine.com
     
    Links2Go is a new semi-automated link collection; it's impressive
    what AI can generate
        http://www.links2go.com/search?search=python

    Tenth International Python Conference 
        http://www.python10.org            

    Archive probing tricks of the trade:
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100
        http://groups.google.com/groups?meta=site%3Dgroups%26group%3Dcomp.lang.python.*

Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://www.ddj.com/topics/pythonurl/         
  http://purl.org/thecliff/python/url.html (dormant)
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From greg at cosc.canterbury.ac.nz  Thu Feb 28 22:59:23 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 01 Mar 2002 16:59:23 +1300
Subject: dirwalk.py generator version of os.path.walk
References: <a5d36e$1daf$1@news.idiom.com> <ac677656.0202271621.5a134d44@posting.google.com> <a5kv49$ae8$2@news.idiom.com> <just-BED145.11365928022002@news1.xs4all.nl> <a5mbm7$18vj$1@news.idiom.com>
Message-ID: <3C7EFC9B.AB6029EA@cosc.canterbury.ac.nz>

Jim Dennis wrote:
> 
>  I guess I'm just wary of recursion, particularly when I've read
>  that Python doesn't support tail-end recursion.

That's only an issue if you abuse recursion for the likes of
iterating along a linear list. In languages which optimise
tail-recursion, it is common for people to do things like
that using recursion, but it would be silly in Python.
If, e.g. the list were a million items long, you'd need a
million stack frames.

But here, the stack is only going to be as deep as the
directory hierarchy, which is hardly going to kill you.

So, don't just blindly reject recursion -- think about
how it applies to the problem at hand!

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


From newgene at bigfoot.com  Mon Feb 25 14:26:26 2002
From: newgene at bigfoot.com (newgene)
Date: 25 Feb 2002 11:26:26 -0800
Subject: How to call COM constants and global functions?
Message-ID: <a50631f6.0202251126.467ef9d7@posting.google.com>

Hi, group,
     Recently, I tried to integrate python with excel by using
win32com module, but I don't know how to call a constants in VB for
Excel, like 'xlCSV' for 'XlFileFormat', 'xlBackgroundAutomatic' for
'XlBackground',etc. Also, I don't know how to call a global function
like "RGB".
     Hope you can help me out.

PS. When I follow the code in "Python programming on win32" Page
206-207(Using COM constants), it doesn't work for me. The code is :

>>> from win32com.client import constants, Dispatch
>>> xl=Dispatch("Excel.Application")
>>> constants.xlAscending
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "C:\Python22\win32com\client\__init__.py", line 133, in
__getattr__
    raise AttributeError, a
AttributeError: xlAscending

My platform is ActivePython 2.2a2+ on Win2K.


Thanks.

CL


From jason at jorendorff.com  Tue Feb 19 12:43:14 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 19 Feb 2002 11:43:14 -0600
Subject: Adding static typing to Python
In-Reply-To: <a4stbu$ptq$1@peabody.colorado.edu>
Message-ID: <HFEKILOLEFEFMKAECNDLOELPDDAA.jason@jorendorff.com>

Fernando Perez wrote:
> Courageous wrote:
> > One of the things I like about C sharp is their use of attributes,
>
> Just curious, isn't there something quite similar to what you describe in
> py2.2? I've only glanced at it so I could be wrong...

Yes.

import re

class Frob(object):
    def get_name(self):
        return self.__name

    def set_name(self, name):
        if re.match('^\w+$', name) is None:
            raise ValueError("Frob.name may only contain name characters")
        self.__name = name

    name = property(get_name, set_name, None,
                    """ The Frob's canonical given name """)

Guido hints there may be some syntactic sugar for this
in an upcoming version.

## Jason Orendorff    http://www.jorendorff.com/




From c at tech.usd253.org  Mon Feb 11 08:45:09 2002
From: c at tech.usd253.org (Fred)
Date: 11 Feb 2002 05:45:09 -0800
Subject: Question on if, while, for
Message-ID: <e959e7b.0202110545.3dff102@posting.google.com>

Hello,

I have written a simple program that I want to simplify. I used if and
elif. But I am sure this can be simplified with for or while. Or what
if I wrote a function to do the testing?

Please DO NOT re-write this for me I want to figure it out for myself.
I would just like to know what is the BEST way to do it:

1) For loop (nested??)
2) While loop
3) Function
4) Leave it alone it's fine??

I am a rookie here learning Python with my son, very impressed so far!
Fred
-----------------------------------------------------------------------

num = input("Enter the numerator: ")
den = input("Enter the denominator: ")

if num > den:		#Test for improper fraction.
    i = num/den		#This will only return the whole part, it discards
the remainder.
    j = num%den		#This will return only the remainder.

    if j != 0:		#If j is equal to 0 you will have a whole number, no
remainder,
			#skips down to else.

	#If j is not equal to 0, begin test to find lowest terms of the
fraction.
        if j%9 == 0 and den%9 ==0: 
            a = j/9
            b = den/9
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
	#Prints the value of i, the whole part, reduces the fracton to lowest
terms and prints it.
	
        elif j%8 == 0 and den%8 ==0:
            a = j/8
            b = den/8
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%7 == 0 and den%7 ==0:
            a = j/7
            b = den/7
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%6 == 0 and den%6 ==0:
            a = j/6
            b = den/6
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%5 == 0 and den%5 ==0:
            a = j/5
            b = den/5
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%4 == 0 and den%4 ==0:
            a = j/4
            b = den/4
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%3 == 0 and den%3 ==0:
            a = j/3
            b = den/3
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%2 == 0 and den%2 ==0:
            a = j/2
            b = den/2
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)
        elif j%1 == 0 and den%1 ==0:
            a = j/1
            b = den/1
            print "You entered an improper fraction, simplified it is:
%d %d/%d" % (i,a,b)

    else:
        print "You entered an improper fraction, simplified it is: %d"
% (i)
	#You have a whole number, which is stored in i, this prints it.
        #If you typed in 4/2 you would be here.

        
#This section is for a proper fraction.
elif num%9 == 0 and den%9 ==0:
    a = num/9
    b = den/9
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%8 == 0 and den%8 ==0:
    a = num/8
    b = den/8
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%7 == 0 and den%7 ==0:
    a = num/7
    b = den/7
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%6 == 0 and den%6 ==0:
    a = num/6
    b = den/6
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),

elif num%5 == 0 and den%5 ==0:
    a = num/5
    b = den/5
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%4 == 0 and den%4 ==0:
    a = num/4
    b = den/4
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%3 == 0 and den%3 ==0:
    a = num/3
    b = den/3
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%2 == 0 and den%2 ==0:
    a = num/2
    b = den/2
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),
    
elif num%1 == 0 and den%1 ==0:
    a = num/1
    b = den/1
    print "The fraction you entered is %d/%d in lowest terms." %
(a,b),


From chappa at math.washington.edu  Tue Feb  5 14:56:36 2002
From: chappa at math.washington.edu (Eduardo Chappa)
Date: Tue, 5 Feb 2002 11:56:36 -0800
Subject: OT again... is there an IMAP expert in the house?
In-Reply-To: <mailman.1012915928.22449.python-list@python.org>
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org>
 <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org>
 <mailman.1012224450.11377.python-list@python.org>
 <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu>
 <mailman.1012892049.28494.python-list@python.org>
 <Pine.OSF.4.44.0202042332410.184841-100000@goedel2.math.washington.edu>
 <mailman.1012915928.22449.python-list@python.org>
Message-ID: <Pine.OSF.4.44.0202051154340.470451-100000@goedel3.math.washington.edu>

*** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:

:) > :) Tried that, didn't work.  I was using Netscape 4.75 as one client,
:) > :) Outlook Express as the other, and concurrent deletion (even with
:) > :) expunges) just didn't work.  Outlook Express never seemed to
:) > :) notice Netscape doing deletions, and I couldn't get Outlook
:) > :) Express to properly expunge without closing it.
:) >
:) > Just to be sure you did that. What are the first characters of the
:) > first line of the folder that you want to share?
:)
:) Gah.  I didn't keep the files around, so I can't tell you for sure.  I
:) did look at them with vi after conversion and they definitely weren't
:) Unix mailboxes.

Did you try two different sessions in Pine?. I have donde that with Pine
several times so I know it works. At least if you succeed with one client
you could discard the server and format as the reason why you are failing.

-- 
Eduardo
http://www.math.washington.edu/~chappa/pine/



From pinard at iro.umontreal.ca  Tue Feb 12 09:31:40 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 12 Feb 2002 09:31:40 -0500
Subject: PyFT and PIL
In-Reply-To: <9c11bn$sph@gap.cco.caltech.edu>
References: <mailman.987996754.9967.python-list@python.org>
	<9c11bn$sph@gap.cco.caltech.edu>
Message-ID: <oqu1sm4vqr.fsf@titan.progiciels-bpi.ca>

[Robert Kern]

> [...] I updated PyFT to handle that gracefully.  Because of unavailability
> of web space for some time, I had only semi-released [PyFT].  It is now
> at http://starship.python.net/crew/kernr/source/PyFT-0.65.tar.gz [...]
> I have given up maintenance of PyFT since I've moved onto other things.
> Steve Alexander has offered to take it over. Of course, someone (not me)
> should write a new PyFT from scratch to wrap FreeType2.

PyFT seems to be a wonderful thing.  But the `README' file roughly says:

   `Compilation of the extension is standard.'

The only standard I know is `Distutils'.  Would someone, by any chance,
have an all ready `setup.py' for `PyFT'?  Or else, precise instructions?
In any case, I should try a little harder :-).

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



From tim.one at home.com  Mon Feb  4 04:30:24 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 4 Feb 2002 04:30:24 -0500
Subject: Backwards Compatibility of Python versions
In-Reply-To: <7xu1sxzkfo.fsf@ruckus.brouhaha.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEJGNJAA.tim.one@home.com>

[Paul Rubin]
> That's the problem--if I don't import future division, my code will
> break in the future, because the // operator isn't supported in the
> present.

Not so:  introducing "//" was 100% backward-compatible, so isn't controlled
by __future__.  "//" works out of the box in 2.2, no __future__ required.

>  But if I -DO- import future division, my code will break in
> the present, because current (2.1) and older (1.5) interpreters don't
> recognize the import statement.

2.2 is current; 2.1 and 1.5 are past.

> (The most recently released version of Python is a 2.1 series release,
> so 2.1 is still current).

I don't know where you're looking, but 2.2 final was released.  Bugfix
releases for past versions may (or may not) get released from time to time,
but if you decide to build a 1.5.3 someday, that won't make 1.5 "current",
it will just make 1.5.3 the most recent release of the 1.5 line (as 2.1.2 is
the most recent release of the 2.1 line, and contains none of the features
new in the 2.2 line -- active development of 2.1 ended the day 2.1 final was
released, so calling it "current" is rhetorical gimmickry).

> I don't see any reasonable way I can write code right now that uses
> division that works in both the current 2.1 (which doesn't support
> importing future division) and in forthcoming versions (where the
> semantics of the / operator will be incompatibly different).

The semantics of "/" didn't change in 2.2, and won't in 2.3, 2.4, 2.5, ...
either (read PEP 238).  If you need a given static instance of "/" to do
radically different things depending on whether you feed it ints or floats
dynamically, you're the first person to claim such a need.  Else you have a
specific flavor of division in mind when you type "/", and it's easy to
write it in such a way that it will deliver the same results in 3.0 as in
0.9.8.




From jkraska at san.rr.com  Fri Feb  8 18:46:05 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 23:46:05 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu> <a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org>
Message-ID: <f8p86ug3e4vqcq0dv72euibd8n4ugpdleu@4ax.com>

>Arguing with irrational oddballs on usenet is today's equivalent of the
>Rubik's Cube.

Hey, that's not a bad analogy! Although I just call it "verbal
masturbation". :-)

C//



From fredrik at pythonware.com  Tue Feb  5 02:47:56 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 05 Feb 2002 07:47:56 GMT
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com> <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu>
Message-ID: <M6M78.23304$l93.4247257@newsb.telia.net>

"John" wrote:

> Any will do, e.g., uniform, normal, etc.

does any of the "uniform" or "normal" generators in the standard
random module do what you want?

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

    This module implements pseudo-random number generators
    for various distributions. For integers, uniform selection from
    a range. For sequences, uniform selection of a random element,
    and a function to generate a random permutation of a list in-
    place. On the real line, there are functions to compute uniform,
    normal (Gaussian), lognormal, negative exponential, gamma,
    and beta distributions. For generating distribution of angles,
    the circular uniform and von Mises distributions are available.

no "etc" generator in there, though.

</F>




From mcfletch at geocities.com  Thu Feb 14 02:07:59 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Thu, 14 Feb 2002 02:07:59 -0500
Subject: The joys of weakrefs... references to instance methods (callbacks)
References: <mailman.1013255897.29941.python-list@python.org> <VkV98.2452$yV3.182935@newsread2.prod.itd.earthlink.net>  <mailman.1013502171.782.python-list@python.org> <slrna6ijm4.cik.mlh@vier.idi.ntnu.no>
Message-ID: <3C6B624F.1010308@geocities.com>

Hmm, short on time now to look at the code again.  dispatcher does 
include weak/strong switching IIRC.  Not critical for me at this point.

Would be nice to have a standard module for this stuff, eventually, but 
for now, I'll just stick with dispatcher (since it's already coded in). 
  Might look at AnyGUI's loop when I get around to re-writing 
OpenGLContext's event processing mechanisms.

Must run, enjoy,
Mike

Magnus Lie Hetland wrote:
> In article <mailman.1013502171.782.python-list at python.org>, Mike C.
> Fletcher wrote:
...
>>Also looked at the References.py module in anygui.  Didn't spend much 
>>time on it, as dispatcher is much closer to what I need.
>>
> 
> Take a look at References.py and Events.py together -- Events.py is
> the dispatcher part; References.py is just a set of utilities for
> references that can be either weak or strong.
> 
> 
>>Was one of 
>>these modules based on the other (your comments on the ASPN site seem to 
>>suggest there's some relationship)?
>>
> 
> Yes, the dispatcher in Anygui is based on this and at least two other
> similar dispatcher modules (but it is basically a full rewrite). There
> was some lengthy discussion on the Anygui devel list about this sort
> of dispatcher mechanism a while ago. (If you're interested, I could
> probably find the other two modules as well :)
> 
> One advantage of the Anygui dispatcher is that it allows you to use
> weak references for some handlers and strong references for others.
> (Maybe Patrick has also added that now -- I'm not sure.)
> 
> 
>>Good job all around. Thanks,
...




From bokr at oz.net  Fri Feb 22 02:03:17 2002
From: bokr at oz.net (Bengt Richter)
Date: 22 Feb 2002 07:03:17 GMT
Subject: Pythonic bit twiddling
References: <il845a.7qa.ln@127.0.0.1>
Message-ID: <a54qfl$vl4$0@216.39.172.122>

On Thu, 21 Feb 2002 17:59:14 -0800, Jim Richardson <warlock at eskimo.com> wrote:

>
>Pythonic bit shuffling question
>
>I have an application that does a calculation on a single byte of data,
>after breaking it into an upper and lower nibble of four bits. How can I
>do this in python? To clarify.
>Take a single byte, break off the upper 4 bits, and do some stuff, then
>do similar stuff with the lower nibble, and then munge them back
>together to create a byte again. This is for a checksum to inteface with
>a bit of embedded stuff. I know how I can do this in C, but as that
>would be the sole bit of copiled code in the app, I'd prefer to avoid
>it if possible. 
>
Unless you are mixing in other data somehow into the byte, there is only
256 possible states of your input and output, so you should able to
pre-construct a translation table and say
    outByte = transTable[inByte]

and be done. This may not be smaller than defining a function
to do it, but if you need speed, it ought to run faster.
If your in and out bytes are characters, the above transTable
becomes a string and you have
    outByte = transTable[ord(inByte)]

It sounds like you should be able to describe your function in detail
pretty easily. Why not post a cut at a function definition?

Then you can decide to use it directly or use it to compute a
translation string, per your priorities. If you are doing exactly
the same thing to both nybbles, you can do a 16-item lookup for
that part. It might be easier to enumerate than to program
the logic, depending.

Regards,
Bengt Richter



From skip at pobox.com  Mon Feb 25 10:53:34 2002
From: skip at pobox.com (Skip Montanaro)
Date: Mon, 25 Feb 2002 09:53:34 -0600
Subject: [Python-Dev] PEP 282: A Logging System  --  comments please
In-Reply-To: <20020215164333.A31903@ActiveState.com>
References: <20020215164333.A31903@ActiveState.com>
Message-ID: <15482.24062.951258.55849@12-248-41-177.client.attbi.com>

    Trent> More standard Handlers may be implemented if deemed desirable and
    Trent> feasible.  Other interesting candidates:

    ...
    Trent>     - SyslogHandler: Akin to log4j's SyslogAppender.

I'd implement at least this one, both as a proof of concept and to provide a
standard mapping between the levels used in your logger and those syslog
provides.

Skip



From aa8vb at yahoo.com  Thu Feb 28 13:11:04 2002
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 28 Feb 2002 13:11:04 -0500
Subject: Python DB that supports numbers
In-Reply-To: <3C7E7E34.24356.6EB01AA@localhost>; from gs@styrax.com on Thu, Feb 28, 2002 at 07:00:04PM +0000
References: <20020228125553.A2852587@vislab.epa.gov> <3C7E7E34.24356.6EB01AA@localhost>
Message-ID: <20020228131104.A2852698@vislab.epa.gov>

Garry Steedman:
 |> Just a simple DB for quick projects without the setup headache of
 |> server-based database systems.
 |
 |gadfly: http://www.chordate.com/gadfly.html

Thanks.  I'd actually tried to check that one out before mailing, but...:

     ERROR
     The requested URL could not be retrieved
           Connection Failed 
     The remote host or network may be down. Please try the request again. 

so I wasn't sure if it was still maintained.  I'll check back later to see
if the site recovers.

Randall

-- 
Randall Hopper
aa8vb at yahoo.com



From rmunn at pobox.com  Tue Feb 26 12:01:30 2002
From: rmunn at pobox.com (Robin Munn)
Date: 26 Feb 2002 17:01:30 GMT
Subject: Problems with select
References: <pan.2002.02.26.04.17.29.538545.11917@onesite.org>
Message-ID: <slrna7nfkh.2bd.rmunn@rmunn.dyndns.org>

On Tue, 26 Feb 2002 04:17:30 +0100, Alain Tesio <alain at onesite.org> wrote:
>Hi, I can't manage to make select work, in both 1.52 and 2.1.2
>
>I just want to read stderr and stdout when executing a command
>with popen3, and I must use select, read() blocks when there is
>more than a fixed amount of data (4 KB I think)
>
>The problems is that select still gives something after the end
>of the file, or sometimes never returns anything at all.
>
>I create the pipes with :
>
>(pipe_out,pipe_in,pipe_err)=popen2.popen3(command)
>
>and use select like this:
>
>pollstatus=select.select([],[pipe_out],[pipe_err],1)
>
>
>(the doc says : select(iwtd, owtd, ewtd[, timeout])
>I didn't put pipe_in as it's closed)

I don't think you've understood the documentation completely. The three
lists you pass to select() are lists of objects for input, output, and
"exceptional conditions" (which are things like out-of-band data in
TCP/IP sockets). Here, "input" and "output" means from the point of view
of your program: if you want to be informed when a pipe has data waiting
for you to read, you put it in the input list.

So what I think you should be doing is:

pollstatus=select.select([pipe_out, pipe_err], [], [], 1)

Hope this helps.

-- 
Robin Munn
rmunn at pobox.com


From martin at v.loewis.de  Sun Feb 17 08:13:24 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 17 Feb 2002 14:13:24 +0100
Subject: Redefining "\n"
References: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>
Message-ID: <m3bseouu8b.fsf@mira.informatik.hu-berlin.de>

Dale Strickland-Clark <dale at riverhall.NOTHANKS.co.uk> writes:

> Is there some way to redefine the "\n" escape sequence?

The only way to do achieve that is to change the Python interpreter.

Regards,
Martin


From syver-en+usenet at online.no  Mon Feb 18 19:19:23 2002
From: syver-en+usenet at online.no (Syver Enstad)
Date: Tue, 19 Feb 2002 00:19:23 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <uk7ta2uk2.fsf@online.no>

ajeru at vknn.org (Alexander Jerusalem) writes:
> In Python I'd have to test the whole software suite including each and
> 
> every branch in each and every if statement to be sure that I didn't
> forget something. 

Yes. 

Don't you have to test a statically typed program in each and every
branch to be sure you didn't forget something too?


>To build such a test suite would take almost as long
> 
> as it takes to build the whole software and I'd have to maintain the
> test suite as well. That's just unafordable.

Perhaps not, if you build the test suite as you build the
application. I agree that dynamically typed software without tests
can be a nightmare, but I think that statically typed code without
tests is almost as bad. In my experience the tests gives you the
security that a compiler will give you and much more. Plus it's easier
to write tests in python than let's say C++ which is the statically
typed language that I am most familiar with.

See www.xprogramming.com or http://c2.com/cgi/wiki?UnitTests for more
information on unit testing your code.

-- 

Vennlig hilsen 

Syver Enstad


From threeseas at earthlink.net  Sun Feb 10 10:51:51 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 10 Feb 2002 15:51:51 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com>
Message-ID: <8791.806T2435T6975373threeseas@earthlink.net>

On 07-Feb-02 11:39:14 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:
>>>>>>>> Timothy Rue writes:

>>>> There is only the creation of a configuration of very common code
>>>> functionality, carried out to it's natural conclusion in support
>>>> of versatility.

>>> That has already been done.  It's been done over and over for the
>>> last 50 years of computer programming.  Without a clear definition
>>> of what you mean by "common code functionality" to differentiate
>>> your approach versus what's already out there, I see nothing
>>> worthwhile in your project.

>> I've defined the nine commands, so show me where this configuration
>> you claim has been done over and over again.

>I didn't say they did it in "nine commands".  I said that everyone has
>been working on a definition of "common code" since the very beginning
>of computer programming (for proof, read some of the history of many
>computer programming languages).  The problem is that everyone has a
>different definition of *exactly* what works best.  Oh, there are many
>generally accepted principles but everyone "colors" the definition of
>those principles within their particular implementation (its only
>natural for that to occur).  You might be the first to make the claim
>that those principles can be represented in "nine commands".

>However, so far, your definition of the "nine commands" isn't clear
>enough to determine whether your claim is valid.

>--
>David Masterson                dmaster AT synopsys DOT com
>Sr. R&D Engineer               Synopsys, Inc.
>Software Engineering           Sunnyvale, CA



David, the biggest hurdle I see you dealing with is your belief that the
VIC and it's nine commands are, or make up, a language.

There is a difference between the VIC and language, just as there is a
difference between 3 deminsional reality and 3 deminsional data arrays
existing in a 2 deminsional media.

This is like the problems faced in believing the world is flat. Where what
is communicated from a global round world is attempted to be framed within
a flat earth mindset by those believing in a flat earth. And sometimes it
can seem confusing and wrong. It doesn't fit onto a flat earth...

If you hold an apple up in the air and let go, what language does it use
to tell you it is falling via gravity? Or does it depend on what country
the apple is grown in? Perhaps there is even a dialect it speaks to you in
depending on what part of a country it is grown in?

However, what is real is that a falling apple is not dependant on
language, in order for it to fall. Its dependant on physics of which
existed before we ever did and before we have invented and attached
languages in a description of it, in order to communicate about it's
existance and action/force.

Even this observation of gravity on an apple, then converted to a
description in a given language, makes use of the VIC and its nine
commands.

AI - Start apple movement experiment

PK - keep track of time (to know the beginning and end difference)

OI - Observe

IP - using eyes

OP - record a description of what happens, under the title "gravity"

SF - note and follow sequencial steps in apple falling experiment

IQ - Place output record description of "gravity" here for later accesses

ID - Identify earth and apple and their relative size/distance to each
     other and at various points in the experiment.

KE - Constrain to identifying Gravity and defining it in an abstract
     language.


Wait a minute! Long before Newton, there were apples falling and all sorts
of other things falling to the earth too.

How is it possible that it is a revelation when Newton does the above?
Perhaps it was just a beginning of understanding the force of gravity?

Today, we still do not fully understand gravity!

Which means that if gravity is a language, as you seem to want to believe
everything is some sort of abstract language, then how do you know you
are speaking it right, since it's not all there?

What came first? The abstract or the concrete?

Language is an invention of man, initially in order to better communicate
reality. Eventually language became capable of describing the non-real.
And as such also became capable of distortions or deception in
communicating reality.

The VIC is not a language, it is an identification of unavoidable physics
in how we use abstraction. You can change the labels attached to the
action set all you want, but the action set is physics, unchangeable.

The value of a postage stamp only goes as far as its ability to stick to a
package. The value of a language to communicate only goes as far as it's
agreed upon use and definitions. The value of physics is ..... well
undeniable.

The definition of the nine commands will never be clear enough to anyone
who persist in seeing them in a scope of less than which they exist. Just
like a belief in a flat earth inherently prevents one from fully
understanding concepts and things from a global round earth.

Or to use another example, advanced math is not possible to understand
using the abstract language of roman numerals. Where roman numerals were
invented to count things, not including "nothing" or "zero" items and as
such didn't have a nothing place holder word/symbol, though the abacus
understood the place value. There is no concept of a place holder in
roman numerals. There is yet no specific identified use of the physical
action constants in abstract programming languages used to automate????

Showing that language and beliefs can falsely limit your perception of
physical reality. Language is a tool that is created as we go along and
better identify and define reality into abstractions. As such the
possibilities of language are endless.

But it's not going to change the fact that you and all of us use the nine
actions of the VIC. If anything, it only keeps you unaware of yourself and
others, what is done. But ignorance has never stopped physics! At best
it's only stopped us from becomming more advanced in what we know and do.

Of the endless potential of languages, the physics of creating and using
languages remains the same. Communication beyond language requires an
awareness that we as a species have yet to develope, if it is at all
possible (mental telepathy of images and actions). But then David, I
suppose to you that would be a language too.

Thru greater awareness of physics we are able to better control our
environment, be it food, clothing, shelter, medicines, .... computer
automations,.... etc..

Avoiding physics does not invalidate physics. It only keeps you and others
ignorant. Meaning the way to validate, is to create the VIC within the
functioning environment of computers and their inherent limitations.

To argue about it is the same as avoidance, it doesn't get you to
validation.

So I've asked for help in creating the validation. <shrug> And I've gotten
alot of Nos.... suggesting there is interest in staying ignorant or
widespread (trained?) ignorance regarding how to advance.

Now, what sort of person or peoples would want to promote ignorance of
reality? Those who insist that the world of the abstract is all there is
and that they are the masters of it, over others? What values did those
who promoted the earth as being flat obtain? Position in society? Control
over others?

There is a very real physics to abstraction creation and use. Regardless
of how disrupting it may be to those content on living in pure
abstraction in trying to make reality fit them. .coms proved reality don't
care and doesn't listen to their abstract desires. <shrug>

I can imagine that sometime in the not to distant such things as barbaric
elitist programming practices and .coms will be viewed as sickening as
witch killing, niger burnings and ...... well pick something, anything out
of our history where we now look back and see ignorant controlling acts
performed upon humans by humans. Joan of Arc, Galielo, ______, Holicost a
lot.....

Funny how Hitler chased Einstein and Fermi out of his territory, their
homeland.....

IS history repeating or going to repeat itself here???

If so, where does it put me?

If you think I want this, then you are insane, living in your world of
abstraction!

There is only one way out of the world of pure abstract illusion.

Validate! and be credited for it.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From michael at stroeder.com  Fri Feb 22 06:48:23 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 22 Feb 2002 12:48:23 +0100
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org> <7xlmdmfl6a.fsf@ruckus.brouhaha.com>
Message-ID: <3C763007.78FEA7FF@stroeder.com>

Paul Rubin wrote:
> 
> Btw, I better point out a not-so-obvious security peril with this
> scheme and any others like it that put the session ID into the URL.
> [..]
> attacker.com server containing an HTTP REFERER header which has
> your page's URL including the session ID.  The attacker can then

One can't repeat this often enough and bang this into the heads of
web application programmers!

Especially web mail applications, web forum software or any other
applications which display untrusted HTML input are probably
vulnerable to this.

There are two things you can do against it:

1. Detect session hijacking by cross-checking CGI-BIN vars you can
assume to be constant throughout the whole session for *each* hit.
If you want to be more secure use SSL on a server properly handling
SSL session renegotiation and exposing the SSL session ID in the
CGI-BIN env vars (e.g. Apache with mod_ssl -> env var
SSL_SESSION_ID). If you want to be really secure or you can't rely
on proper SSL session renegotiation deploy client certificates and
use the subject DN and issuer DN exposed in the env vars.

2. Have a URL redirector in place *within* your application which
outputs HTML with redirect URL in <head> section. In this case the
browser sends the redirector's URL as referer URL. It's not
sufficient to redirect by HTTP status codes 301 or 302 and Location:
header! Off course it's much work to keep track of all links you're
displaying and it's not feasible for inline-displaying of images via
<img> tag.

You're invited to play with the on-line demo of web2ldap at
http://sites.inka.de:8002/web2ldap which implements these kind of
things and places the session ID in PATH_INFO (because personally I
do not like cookies although I see some of their advantages, e.g.
single cross-site sign-on).

Feedback welcome.

Ciao, Michael.


From cs1spw at bath.ac.uk  Mon Feb 11 13:38:54 2002
From: cs1spw at bath.ac.uk (Simon Willison)
Date: Mon, 11 Feb 2002 18:38:54 GMT
Subject: Newbie needs references
References: <u9bkUtmsBHA.388@cpimsnntpa03>
Message-ID: <3C680FA9.4050101@bath.ac.uk>

billludden wrote:
> Hi<
> 
> I have not programmed since 1975, and that was in BASIC.
> I'd like to learn Python, but everything I find on the web is
>     way above my head.
> Does anyone have books or tutorials they would recommend
>     for an illiterate to teach himself Python?
> 
> Thank you,  Bill

I'm a Python newbie as well and while I found some useful resources on 
the web (the tutorial on www.python.org and the Python 101 on 
www.devshed.com were both very useful) by far the best resource I've 
found so far is the O'Reilly book "Learning Python". It covers the core 
of the language thoroughly but in an easy to understand way and is 
probably the clearest "learning" programming book I've read, although a 
lot of that could be credited to the elegant design of Python itself.



From patois at calvix.org  Tue Feb 26 08:21:11 2002
From: patois at calvix.org (Yannick Patois)
Date: Tue, 26 Feb 2002 13:21:11 +0000 (UTC)
Subject: Calling baseclass method from a derivated class
Message-ID: <slrna7n2u6.f53.patois@ns.sicfa.org>

Hello,

I would like to know how I can call a method defined in a base class 
from a derivated class that overloaded it.

Example:

class toto1:
   def titi(self):
        print "titi1"

class toto2(toto1):
   def titi(self):
        print "titi2"

And now, if I create:
item=toto2()

I can do:
item.titi() # -> "titi2"

But what to get the first 'titi' method to be called ?

Thanks for any help,

	Yannick

-- 
 _/ Yannick Patois \___________________________________________________
| web :http://feelingsurfer.net/garp/ | CALVIX ! Le LUG de CAEN        |
| email : patois at calvix.org           | web : http://www.calvix.org/   |
| Petit Portail en Bois, Modeste et G?nial : http://www.rezo.net/      |


From brueckd at tbye.com  Mon Feb 18 23:16:32 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Mon, 18 Feb 2002 20:16:32 -0800 (PST)
Subject: Adding static typing to Python
In-Reply-To: <0DFA7478F1E5D311B90900508B1208CF565714@ausmail02.mis.gtech.com>
Message-ID: <Pine.LNX.4.33.0202181937180.4907-100000@ns0.tbye.com>

On Mon, 18 Feb 2002, Milliken, Peter wrote:

> Rather than from ignorance, maybe it's from actual experience using it.
> Have you considered the possibility that the claims might be true? In this
> job as well as the past few our Python code has always had less time
> devoted to developing, testing, maintaining, and refactoring it. It's not
> that there's hidden costs involved - we tend to write it and move on to
> something else because it just works.
>
> Depends, "ignorance" was used in the context of "have you tried an
> alternative?". You don't provide any examples of where you have found Python
> superior over any other language development environment. So it becomes
> impossible to verify the veracity of your statement :-) I provided an
> alternative language in my statements to back up my claims :-). Since we are
> talking loose typing versus strong typing, I would expect you to respond
> with experience in a strongly typed language. The original poster used Java
> as his comparision and is wishing for the feature in Python. Necessarily he
> feels it is a deficiency in the Python language that I concur with :-)

Sorry, I didn't realize you would be so picky. :) I am speaking from
experience using C, C++, and Java. So, in my above paragraph, when I say
stuff like "our Python code always took less time to develop, debug, blah,
blah" it is obviously in relation to another language, otherwise it
couldn't be less now could it? Also, to be picky, we are talking about
static versus dynamic typing, actually. :) Python is strongly but
dynamically typed.

> also depends on your definition of "large", my definition of this is several
> million lines of code - co-ordinating enough programmers to produce this
> level of effort in a reasonable time can be quite a challenge :-).

Very true, but several million lines of C++ might very well be several
hundred thousand of Python, so be careful in the comparison. A Python
program several million lines long may very well be the largest and most
complex program in the history of the world.

> "Contractual" type interfaces between modules helps a lot - something else
> that Python lacks (IMO) that is useful for "large" projects.

It might be useful, sure, but you can certainly have contractual
interfaces without having the language force your obedience. On any large
project there need to be published interfaces between the various
components, and someone not adhering to the interface is both an easy and
quick find, in most cases.

> Python also lacks features of "data hiding", it attempts to provide
> some support for it but only in a desultory "procedural" way - try
> this with 40 programmers and you won't get far! :-)

This is silly - let me guess, your same 40 programmers also can't remember
to document their code, they follow no naming conventions, they use global
variables everywhere, etc, right? :) Surely, if you can somehow convince
them to follow basic coding standards then Python's support for data
hiding won't cause you any problems. It keeps the honest people honest,
that's all. OTOH, if you have to babysit your developers then you're
right, Python is not for your project. Also, you should probably fire most
of your development staff.

> This level of discussion has occured in the past and
> people have disputed every one of these points - again, relevant experience
> is important, the argument generally tends to the dismissive without
> offering any supportive experience

Excellent point! When, exactly, did Python's "weaknesses" that you
describe cause your big project to fail? Was it when the team grew from
one dozen to two dozen developers, or was it when the LOC exceeded 100k?
Please be specific. :)

> i.e. "I used data hiding features of
> language X for Y+ years on projects that were over Z programmers large and I
> never missed it once I moved to Python".

I used the data hiding features of C++ for about 8 years on projects that
were over 40 developers large and I never missed it once I moved to
Python. Before that I was mainly a C and assembly programmer and (getting
back to the point), I found C's type system to be more of a hinderance
than a help.

> The meer fact that you are referring to "this job"  and the "past few" means
> that (to me :-)) you haven't worked on a "large" project (by my definition).
> This many of my "large" projects cover a span of 10 or more years!

Congratulations. Please re-read my post: "this job" and the "past few"
refer to the ones in which I have used Python for production code. I
should also point out that a huge percentage of "large" projects that
cover a span of so many years is because they take so long to get into
production that they are no longer relevant. :) I've been on those, and
while you are in many cases guaranteed a permanent job because of it, it
doesn't make it fun.

> Strong typing forces design up front i.e. it forces you to think through the
> problem rather than just start coding - so it would be an anathema for

Ahhh... so as far as software goes, the best and most bugfree code would
therefore be in assembly language as it would require the most
forethought! The amount of design you do, before and during, is
proportional to the size and complexity of the project, regardless of
language. This really has nothing at all to do with static typing at all.

> and the compiler won't let you do sloppy coding

That's one mighty intelligent compiler! <good-natured snicker>

> Finding such bugs (misnamed variables) requires multiple runs through your
> test suite i.e. find a bug, fix it and re-run the tests. You *can't* find
> all such bugs in a single pass!

Yes and, all told, that process took a whopping 2 seconds. Meanwhile, the
compiler for those other languages has finally moved on to the link stage
and is busy printing out 100 lines of "errors" because of your one
misnamed variable.

> You also have to find it through (often) painstaking use of the
> debugger or print statements.

No: most of the time you'll see it in the traceback, which is
automatically printed for you.

> are *never* involved! Running the same test over and over again to get these
> levels of bugs out is clearly more inefficent than using a compiler. So
> where your process may take possibly hours, I'm talking about minutes to fix
> and recompile :-).

Well, since you earlier thought I had no experience, I'll risk doing the
same thing here: I have never, ever had this process take more than a few
minutes, and I'd wager that 99% of the time it would be more easily
measured in seconds. It really sounds like you are letting your mind get
worked up over a theoretical problem that you have never actually
encountered. I apologize for being blunt, but it just ain't so. :)

> Might be true for lots of people, but nearly every single Python module in
> our company's code has either built-in or accompanying tests that run
> every night. The same is true for our non-Python code. The dynamic or
> static typing of the language makes little real difference - if you don't
> test, you overlook bugs, pure and simple.
>
> True, you obviously have a large investment in your test suite. My point is
> that up front development time is reduced through the features of strong
> typing.

Can you give a concrete example showing this?

> by that background. The original post states that the typing of Java was
> perceived as a distinct advantage, so presumably the level of type checking
> there is enough to request the feature in Python.

Ugh! Probably the worst change that could happen to Python would be to add
a static type system similar to that in Java. Yes, I have used it, and yes
it really does hurt that much. It adds complexity and confusion and longer
compile time, and yet runtime type errors can still occur.

Have fun Pete,
Dave




From kosh at aesaeion.com  Sat Feb 16 23:13:05 2002
From: kosh at aesaeion.com (kosh at aesaeion.com)
Date: Sat, 16 Feb 2002 21:13:05 -0700 (MST)
Subject: Have to try
In-Reply-To: <004501c1b763$78bca9c0$b278c53f@pacbell.net>
Message-ID: <Pine.LNX.4.32.0202162112070.1394-100000@aesaeion.com>

On Sat, 16 Feb 2002, Rick Hamilton wrote:

> So I am trying to do a (what should be) simple thing. I have a string stored
> in a variable. I want to search that string for the occurrence of a sub
> string. I have tried
> Find(s,sub)
> Rfind
> MyString.find(sub)
>

I think you might find count to be more appropriate if you just want to
know if it has some string.




From llchen223 at hotmail.com  Sun Feb 10 19:12:14 2002
From: llchen223 at hotmail.com (Lei Chen)
Date: 10 Feb 2002 16:12:14 -0800
Subject: redirect console output to widget?
Message-ID: <de6c123c.0202101612.72b27f8d@posting.google.com>

Hi,

I want to redirect console output to a Text widget.  So far the code I
wrote crashes on occasion.  Attached is the portions of the code. 
Thanks in advance for your help.  Here are some more details: program
runs on WIN32; the console stream is fast and continuous (console
program runs 24x7)


#this function pipes input to an widget
def pipeToWidget(input, widget):
    widget.delete('1.0', END)
    while 1:
        line = input.readline()
        if not line: break
        widget.insert(END, str(line))
        widget.see(END)
        widget.update()

def funcThread(widget):
    input = os.popen(r'executeMe.exe', 'r')
    pipeToWidget(input, wiget)

#uber-main
root = Tk()
widget = Text(root)
widget.pack(side=TOP, expand=YES, fill=BOTH)
thread.start_new(funcThread, (widget,))
#...Some more threads started here...
root.mainloop()


From pedronis at bluewin.ch  Tue Feb 26 19:51:34 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Wed, 27 Feb 2002 01:51:34 +0100
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com> <3c7a70d2_4@news.bluewin.ch> <gherman-A86A1B.18413225022002@news.t-online.com>
Message-ID: <3c7c308b$1_5@news.bluewin.ch>

Dinu Gherman <gherman at europemail.com> wrote in message
gherman-A86A1B.18413225022002 at news.t-online.com...
> In article <3c7a70d2_4 at news.bluewin.ch>,
>  "Samuele Pedroni" <pedronis at bluewin.ch> wrote:
>
...
> > 2. if you check the test suite for CPython 2.1 and Jython there is no
test
>
>
> I'm pretty surprised! I knew the Python test suite is *very* far
> from complete, but list() is an *extremely* crucial function,
> isn't it?
>

I have submitted a patch for the CPython test suite,
and Guido has applied it, so that part of the problem is solved.

-- Samuele Pedroni.




From niemeyer at conectiva.com  Fri Feb 22 22:14:25 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Sat, 23 Feb 2002 00:14:25 -0300
Subject: Stackless Platform Independence?
In-Reply-To: <m37kp5i5a4.fsf@chvatal.cbbrowne.com>
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com>
Message-ID: <20020223001425.A8624@ibook.distro.conectiva>

> > - Platform independence is gone.
> 
> I hope not...  I would hope to hear that "platform indepedence has
> _ARRIVED_"

That's the human nature. Tismer writes half a dozen mails, some with
unbelivable results, and some asking for help. One of the first answers
he gets is complaining about something.

Thanks Tismer. I'll let you know if I do any work about it or find
someone to sponsor you. Personally, I'd like to see it running in PowerPC
Linux. Maybe it's a good reason to learn ppc assembly.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From Christian.Schmidt at aifb.NOSPAMuni-karlsruhe.de  Thu Feb 28 09:22:36 2002
From: Christian.Schmidt at aifb.NOSPAMuni-karlsruhe.de (Christian Schmidt)
Date: Thu, 28 Feb 2002 15:22:36 +0100
Subject: kill childs
References: <a5j1p9$ftb$1@news.rz.uni-karlsruhe.de> <m3sn7misch.fsf@mira.informatik.hu-berlin.de> <a5jles$317u$1@nntp6.u.washington.edu>
Message-ID: <a5leih$lp1$1@news.rz.uni-karlsruhe.de>

> He's aiming at the process group, which is a good idea.  os.kill(-1, sig)
> will work, though of course the signal will be delivered to the calling
> process as well.  It might be a good idea to set a flag indicating that
> the process group kill has already happened, to avoid an infinite loop.
I tried this one... and it worked. Here's my script.
Thanks.
---------------
import signal, os, sys

term_childs_first = 1 # static variable for signal handler

def term_childs (signal, frame):
    global term_childs_first
    if term_childs_first: # first kill of parent process
        term_childs_first = 0
        os.kill (-1, signal) # process group kill
    else: # second kill, because of process group kill
        sys.exit (signal) # exit with signal
# end term_childs

signal.signal (signal.SIGTERM, term_childs) # install signal handler for
terminate
print os.getpid()
os.system("ls -laR / > NUL") # long running commandline
print "long running command completed."




From wayne at mishre.com  Tue Feb 12 15:11:25 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Tue, 12 Feb 2002 12:11:25 -0800
Subject: Developing a GUI program like BOIDS
References: <3C6968A4.1CA9C62E@boeing.com>
Message-ID: <3C6976ED.6080609@mishre.com>

What is your target platform?

If you are only working with Windows, you can use MFC (sorry, had to put 
that in there...).  There is a list at Cameron Laird's site[1], I don't 
know how current it is though

I typically use Qt for most of my development then if necessary use 
Gordon McMillan's Installer[2] to package it up.  I've also seen 
reference to AnyGUI[3], which aims to allow you to program to any GUI 
with a common API (..at least that's what I think it does..).  I haven't 
tried it yet though.

There are some good form design tools available, if you want to use Qt 
take a look at BlackAdder[4].  For Tk there is SpecPython[5] and 
wxWindows has wxDesigner[6] and BOA[7], GTK has Glade[8] I don't know of 
any others, but a quick search on Google with the GUI name and 'Python' 
should keep you busy for a while.

You can also use any Java IDE, then access the GUI you develop with 
Jython[9].  This might be a good way to start if you want to port over 
your existing Java application.

I also found this review of IDEs[10] while looking up some of the other 
web pages.

[1] http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html
[2] http://www.mcmillan-inc.com/
[3] http://www.anigui.org/
[4] http://www.thekompany.com/products/blackadder/
[5] http://www.geocities.com/eric_brunel/SpecPython.en.html
[6] http://www.roebling.de/
[7] http://boa-constructor.sourceforge.net/
[8] http://glade.gnome.org/
[9] http://www.jython.org/
[10] http://aspn.activestate.com/ASPN/Mail/Message/758104

HTH,


WEC7108 wrote:
> I would like to develop a GUI program similar to the ever-popular BOIDS
> using Python instead of Java.  Are there URL references or books on this
> kind of development?  :o)  Bill
> 


-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From tismer at tismer.com  Fri Feb 22 07:42:04 2002
From: tismer at tismer.com (Christian Tismer)
Date: Fri, 22 Feb 2002 13:42:04 +0100
Subject: Stackless: 100,000 Tasklets in 80 MB
Message-ID: <3C763C9C.60701@tismer.com>

I'm kidding?
No, it's the truth.
Stackless Python can run such amounts of tiny tasks
in just 80 MB main memory, although it takes snapshots
of the C stack.

The website http://www.stackless.com/ has been updated a little.
-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Kaunstr. 26                  :    *Starship* http://starship.python.net/
14163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
      where do you want to jump today?   http://www.stackless.com/





From jason at jorendorff.com  Mon Feb 11 16:14:48 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: 11 Feb 2002 15:14:48 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb 11)
Message-ID: <65D322D62049DED1.100EC0150B20FD6E.E4912BB6F9A2A65D@lp.airnews.net>

QOTW:  "'poly-what?'  Yes, python is polymorphic, but there's no need to
get all fancy-pants about it."  -Quinn Dunkan

"Good general rule: if Tim made it that way, it's supposed to be that
way." -Steve Holden

"[J]ust fire up python and start typing.  i like that." -Mark McEahern,
on his naive pickling technique

"I was wondering if there's any kind of definitive nickname for Perl
programmers, the way we call ourselves Pythonistas?" -Aahz Maruch
"Masochists." -Daniel Klein

"I'd just like to take this moment to point out that C has all the
expressive power of two dixie cups and a string." -Jamie Zawinski,
in the source code for xkeycaps

    TTX translates TrueType fonts to an XML format and back,
    enabling you to edit TrueType fonts with a plain text editor.
        http://www.letterror.com/code/ttx

    The 10th International Python Conference took place February 4-7.
    Several remarkable slide shows are now available online, along with
    a student video.
        http://www.python.org/workshops/2002-02/

    PyGame 1.4 is released.  PyGame is a set of Python modules for
    writing games.  This version adds support for Mac OSX.  It's also
    easier to build EXE's on Windows with the pygame2exe sample script.
        http://www.pygame.org/

    StandaloneZODB is a Python object persistency system.  Version 1.0
    final is out!
        http://groups.google.com/groups?th=725c2a321d15543c

    Anygui is a simple, generic module for building GUIs in Python.  It
    can run on top of many different toolkits.  The first official release
    is version 0.1 - check out the screenshots!
        http://sourceforge.net/projects/anygui
        http://anygui.sourceforge.net/screenshots.php

    Benchmarks show Python is still slower than Java.
        http://groups.google.com/groups?th=5ea1ffa9db9c9e99

    Jeremy Hylton and Skip Montanaro presented ideas for making Python
    faster at Python 10.  These have triggered a flurry of activity on
    the Python-Dev list.  For the technically inclined:
        http://www.python.org/~jeremy/talks/spam10/PEP-267-1.html
        http://manatee.mojam.com/~skip/python/ipc10/PEP-266-1.html

    Pyco is experimental software for building smaller Windows executables
    from Python programs.
        http://www.pythonapocrypha.com/projects/pyco/

    Are any HTTP proxy servers written in Python?
        http://groups.google.com/groups?th=9171610619278850

    "Courageous" toys with hygienic macros (a la Scheme) in Python.
        http://groups.google.com/groups?th=619ad5469f24308b

    The Google Programming Contest accepts submissions in only three
    languages: C++, Java, and Python.
        http://www.google.com/programming-contest/


========================================================================

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

    Python.org's Python Language Website is the traditional
    center of Pythonia
        http://www.python.org
    Notice especially the master FAQ
        http://www.python.org/doc/FAQ.html

    PythonWare complements the digest you're reading with the
    daily python url
         http://www.pythonware.com/daily  
    Mygale is a news-gathering webcrawler that specializes in (new)
    World-Wide Web articles related to Python.
	 http://www.awaretek.com/nowak/mygale.html 
    While cosmetically similar, Mygale and the Daily Python-URL
    are utterly different in their technologies and generally in
    their results.

    comp.lang.python.announce announces new Python software.  Be
    sure to scan this newly-revitalized newsgroup at least weekly.
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

    Michael Hudson continued Andrew Kuchling's marvelous tradition
    of summarizing action on the python-dev mailing list once every
    other week, into July 2001.  Any volunteers to re-start this
    valuable series?
        http://starship.python.net/crew/mwh/summaries/
        http://www.amk.ca/python/dev

    The Vaults of Parnassus ambitiously collect Python resources
        http://www.vex.net/~x/parnassus/   

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

    The Python Software Foundation has replaced the Python Consortium
    as an independent nexus of activity
        http://www.python.org/psf/

    Cetus does much of the same
	http://www.cetus-links.org/oo_python.html

    Python FAQTS
        http://python.faqts.com/

    The old Python "To-Do List" now lives principally in a
    SourceForge reincarnation.
	http://sourceforge.net/tracker/?atid=355470&group_id=5470&func=browse
	http://python.sourceforge.net/peps/pep-0042.html

    *Py: the Journal of the Python Language*
        http://www.pyzine.com
     
    Links2Go is a new semi-automated link collection; it's impressive
    what AI can generate
        http://www.links2go.com/search?search=python

    Tenth International Python Conference 
        http://www.python10.org            

    Archive probing tricks of the trade:
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100
        http://groups.google.com/groups?meta=site%3Dgroups%26group%3Dcomp.lang.python.*

Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://www.ddj.com/topics/pythonurl/         
  http://purl.org/thecliff/python/url.html (dormant)
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From mhammond at skippinet.com.au  Wed Feb  6 08:41:31 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 06 Feb 2002 13:41:31 GMT
Subject: Catch COM events generated by a Python COM server with Visual Basic
References: <6411841c.0202060107.773ffe68@posting.google.com>
Message-ID: <3C6132D8.8050000@skippinet.com.au>

Marc ENGEL wrote:
> I am trying for more than one week to use a Python COM server I made
> inside Visual Basic. I encounter a problem with Connection Points

...

You could try making your event a "dispinterface", or otherwise implying 
it is an IDispatch implemented event.

Also, ensure that all your objects are wrapped in a "dispatcher", so 
that you can see all exceptions generated.

The only examples I know about are in the win32com test suite. 
Specifically, there is a C++ and a VB test program that uses/is used by 
the test suite.  However, I don't think the VB test code does exactly 
what you describe.

If you like, you could modify the test code and send me patches.  Add 
everything necessary so that it should work - the event interface, and 
all Python support needed to generate the event.  I can then see why it 
does not work (and having it in the test suite means it then always will :)

You would need to modify the test VB project 
(http://www.pythonpros.com/cgi-bin/cvsweb.cgi/PyWin32/com/TestSources/PyCOMVBTest/) 
and the test driver (win32com\test\testvb.py)

Mark.



From djrassoc01 at mindspring.com  Thu Feb 14 22:35:04 2002
From: djrassoc01 at mindspring.com (Dr. David J. Ritchie, Sr.)
Date: Thu, 14 Feb 2002 21:35:04 -0600
Subject: gif format for PhotoImage
Message-ID: <3C6C81E6.9AE05615@mindspring.com>

Is the gif format expected by PhotoImage different from the gif format
generated by something like GifConverter, for example?

In the example below (in which I have removed some things but
hopefully not broken anything of relevance to this question), I find that
some things I think of
as "gif" images on my system display fine while others are just black
or a series of dots (but they display fine with Netscape).

Can anyone give me some pointers to where I might find a discussion
of the gif format expected?

Thanks.

--David

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

# Tk Text Example

from Tkinter import *
root = Tk()
root.title("My Toplevel")

# Text
# A top level window
wTL2 = Toplevel(root, bg='brown')

# Put a Text widet named text in it with a vertical scroll bar.
text = Text(wTL2, height=30, width=50)
scroll= Scrollbar(wTL2, command=text.yview)
text.configure(yscrollcommand=scroll.set)

# define a picture and specify where it is to go
photo=PhotoImage(file='Picture.gif')
photo.configure(width=50, height=50)
text.image_create(END, image=photo)

# arrange the text window to the left and the scroll bar to the right
text.pack(side=LEFT)
scroll.pack(side=RIGHT, fill=Y)

# Putting Tk into Operation
# Turn over control to Tk so that it can respond
# to mouse clicks and keyboard activity which then
# causes it to call your program's code via call backs.

root.mainloop()
print "Goodbye!"

--
Dr. David J. Ritchie, Sr.
djrassoc01 at mindspring.com
http://home.mindspring.com/~djrassoc01/




From cbbrowne at acm.org  Sat Feb  9 12:44:28 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Sat, 09 Feb 2002 12:44:28 -0500
Subject: Repost of original >: autocoding project proposal
References: <5512.785T1955T13695089threeseas@earthlink.net> <6206.805T2889T7505727threeseas@earthlink.net>
Message-ID: <m3bseymtxf.fsf@chvatal.cbbrowne.com>

In an attempt to throw the authorities off his trail, "Timothy Rue" <threeseas at earthlink.net> transmitted:
> Reposting as a matter of 20/20 hindsight in reference to what all has
> evolved from this proposal. (has therre really been any productive
> direction achieved....)

Apparently nothing "productive" has been accomplished.

Does this imply that:
a) You might want to find a forum that might be more receptive, or
b) You should redouble your efforts here?

Speculate as you will; the choice of where you want to post about this
is obviously yours...
-- 
(reverse (concatenate 'string "gro.mca@" "enworbbc"))
http://www.ntlug.org/~cbbrowne/nonrdbms.html
"Jeez, got me.  Unix is sorta like Heroin, It feels good for about 5 minutes
a day and horrible the rest of the time."  -- Jim O'Dell


From nas at python.ca  Sun Feb 10 17:50:42 2002
From: nas at python.ca (Neil Schemenauer)
Date: Sun, 10 Feb 2002 14:50:42 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
In-Reply-To: <a46heu$8f2$0@216.39.172.122>; from bokr@oz.net on Sun, Feb 10, 2002 at 07:25:18PM +0000
References: <a46heu$8f2$0@216.39.172.122>
Message-ID: <20020210145042.A25774@glacier.arctrix.com>

Bengt Richter wrote:
>  >>> 'abc'.split('')
>  Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
>  ValueError: empty separator
> 
> Wouldn't it make sense to return list('abc') ?

It would also make sense to return ['a', 'b', 'c'].  Since it's not
obvious what you want Python raises an error.

  Neil



From threeseas at earthlink.net  Tue Feb  5 23:54:32 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:32 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <slrna5sudd.uef.mill@pedgr571.sn.umu.se>
Message-ID: <2065.802T1340T13998threeseas@earthlink.net>

On 04-Feb-02 07:08:14 Olof Oberg <mill at pedgr571.sn.umu.se> wrote:
>On Mon, 04 Feb 2002 00:01:09 GMT, Timothy Rue <threeseas at earthlink.net>
>wrote:
>> In reviewing over many messages in the autocoding and related threads and
>> seeing the evidence of contridictions, hypocracy, irrational and
>> illogical responses......
>Good thing you are coming to your senses.

>> Its damn clear that there is a need to remove the sort of people posing to
>> be professional who are in essence just good liars, abstraction
>> manipulators, from the industry.
>You are working in the industry?

>> Who knows, maybe that stench will be enough to motivate me see that it
>> gets done, gets removed.
>Take a shower.

>> Maybe I should be more appriciative of the stench if it so motivates me.
>>
>> Anyone want to add anything to this?
>Yes. Are you going to remove yourself from these newsgroups?

>I think you should be proud of yourself that managed to lure
>(Flying Lure(tm)?) so many people into actually trying to
>understand your ramblings. I mean no one (speak up if I am
>wrong) really understood anything of what you were talking about,
>but still tried to find out if you had something worthwhile to
>say.

> /mill

>--
>#############################################################
># millNO at SPAMludd.luth.se # http://pedgr571.sn.umu.se/~mill #
>#############################################################



The most common manipulation from the computer industry is the flipping of
the bit.

It is no supprise, far from new or even original in application, the
simple bit flip you attempted.

How many programmers own a two headed quarter?
No matter what it always comes up in their favor...

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From DeepBleu at DeepBleu.org  Fri Feb  8 19:16:19 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Sat, 09 Feb 2002 00:16:19 GMT
Subject: I wish.............
Message-ID: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com>

.....that we quit calling Python as CPython.  Something bugs me pesonally
about the name CPython.  This is the pure original.  It is Python not
CPython.  There is Python + Java = Jython.  There maybe Perl + Python = P#,
there maybe Python + .NET = Pythonette, there maybe Fortran +Python =
Pfeiffer.
But let us keep hand off the name Python.  It is not CPython.  It is Python.
:O
DeepBleu




From m.faassen at vet.uu.nl  Sun Feb  3 07:43:14 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Feb 2002 12:43:14 GMT
Subject: How can I write this C code in Python?
References: <mailman.1012663266.30558.python-list@python.org> <3c5c204e.11589328@news.t-online.de>
Message-ID: <a3jb92$d98$2@newshost.accu.uu.nl>

Gerson Kurz <gerson.kurz at t-online.de> wrote:
> On Sat, 2 Feb 2002 09:16:32 -0600, "Jason Orendorff"
> <jason at jorendorff.com> wrote:

>>Fred wrote:
>>> I think if I could see the below C code done in Python it
>>> would get me over the beginners hump.
>>
>>for i in range(1, 13):
>>    for j in range(1, 13):
>>        print "%4d" % (i*j),
>>    print
>>
>>## Jason Orendorff    http://www.jorendorff.com/
>>
> Alternatively, you can always write

> for n in range(12):
>    exec("abcdefghijkl"[n]+"=lambda x=0,y=0: "+filter(
>        lambda x:x not in "\n$\r","""(x*y#x/x!range(x,
> y#x+y!b(1,1#d(e~,e~#d(f~,f~#c(e~,e~+d(g~,d(g~,g~))#"%4
> d" % a(x,y#map(lambda y:i(x,y),h~#" ".join(j(x)#"\\n".
> join(map(k,h~))""".replace("~","()").replace("#",")!")
>        ).split("!")[n])
> print l()

Yay! TIMTOWTDI!!

I like how this way uses Darting's descent to reach an algorithmic
solipcism, not even needing to use the time travel module to reach 
subtler but in the end much more impressive segue. Also nice use of
the Meta Abstraction pattern!

Regards,

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


From logiplexsoftware at earthlink.net  Tue Feb 12 18:02:07 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 12 Feb 2002 15:02:07 -0800
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <B43D149A9AB2D411971300B0D03D7E8BF0A17E@natasha.auslabs.avaya.com>
References: <B43D149A9AB2D411971300B0D03D7E8BF0A17E@natasha.auslabs.avaya.com>
Message-ID: <20020212150207.0ef355e1.logiplexsoftware@earthlink.net>

On Wed, 13 Feb 2002 10:03:01 +1100
Delaney, Timothy wrote:

> > From: brian_l at yahoo.com [mailto:brian_l at yahoo.com]
> > "Steve Holden" <sholden at holdenweb.com> wrote in message 
> > news:<ZND98.140690$KM2.6229113 at atlpnn01.usenetserver.com>...
> > > "Brian Lenihan" <brian_l at yahoo.com> wrote in message
> > > news:afe3b146.0202100036.f221c9a at posting.google.com...
> > 
> > > > I don't throw like a girl right-handed, so maybe I am actually
> > > > ambidextrous.
> > > >
> > > Not to mention gender-biased? :-)
> > 
> > Of course I'm gender biased.  I love girls.  I'm also sure there are
> > plenty of girls who can throw better than I can.  I left out the fact
> > that I'm also a switch-hitter because I thought it would only add to
> > the confusion.
> 
> For those of use who have no interest in (I presume) baseball, what the
hell
> is a switch-hitter?

Someone who can hit right- or left-handed.  Outside of baseball, it's a
somewhat dated euphemism for bisexual.

I mean, um... so I've heard...

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From paul at prescod.net  Mon Feb 11 23:46:18 2002
From: paul at prescod.net (Paul Prescod)
Date: Mon, 11 Feb 2002 20:46:18 -0800
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <a49s16$5ao$1@newshost.accu.uu.nl> <pan.2002.02.11.20.34.30.302944.2302@uchicago.edu>
Message-ID: <3C689E1A.1D95861B@prescod.net>

Ben Wolfson wrote:
> 
>...
> 
> It's a poor conspiracy that censors the messages denying its existence.

That's clear. So obviously the PSU can NOT exis

 Paul Prescod



From mhuening at zedat.fu-berlin.de  Fri Feb  1 06:51:01 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 1 Feb 2002 11:51:01 GMT
Subject: Python like the old-time batch files.
References: <a3bhcf$176195$1@ID-98166.news.dfncis.de> <a3bk4b$11u4$1@norfair.nerim.net> <Xns91A86D0EAA121sebsauvagedotnet@127.0.0.1>
Message-ID: <Xns91A882B25A2BCmhueningzedatfuberli@130.133.1.4>

"S?bastien SAUVAGE" <nospam at sebsauvage.net> wrote in 
news:Xns91A86D0EAA121sebsauvagedotnet at 127.0.0.1:

> http://effbot.org/efflib/console/

Try this one:
http://effbot.org/downloads/

Matthias


From op73418 at mail.telepac.pt  Wed Feb 27 14:37:35 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Wed, 27 Feb 2002 19:37:35 +0000
Subject: Newbie scopes question
Message-ID: <vvcq7uso4lujr5i5caodd6j2dv781322hh@4ax.com>

As far as i understand, a class statement opens a new scope.

In the following piece of code

class Outer:
    attribute = "The outer class attribute"

    def __init__(self):
        try:
            print attribute
        except NameError:
            print "Attribute not defined!"


def Outer_function():
    local_var = "Local var of the outer function"

    def inner_function():
        print local_var

    inner_function()

Outer_function()
a = Outer()

What we get printed is

Local var of the outer function
Attribute not defined!

This means that function and class scopes behave differently as fas as
nesting goes. Could someone explain why is this?

Thanx,
Gon?alo Rodrigues


From nhodgson at bigpond.net.au  Tue Feb  5 16:23:46 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Tue, 05 Feb 2002 21:23:46 GMT
Subject: distutils, C extensions, Borland compiler, wxPython
References: <3c5fe73f$0$204$edfadb0f@dspool01.news.tele.dk> <mailman.1012939329.5851.python-list@python.org>
Message-ID: <C3Y78.38682$Ni2.219590@news-server.bigpond.net.au>

Gerhard H?ring:

> I think I have heard something to the effect, that on Windows, DLLs
> can't link back to .EXEs, only to other DLLs.

   No, DLLs can link back to EXEs although you may have to force (link
/force) the link to complete and/or link the DLL again after linking the
EXE. I used to work on a large project, Fujitsu's IntelligentPad that was
built this way. Mostly it didn't matter as you had old versions of the EXE
and DLL to link against but the initial bootstrap was messy.

   Neil





From missive at frontiernet.net  Sun Feb  3 21:35:06 2002
From: missive at frontiernet.net (Lee Harr)
Date: Mon, 04 Feb 2002 02:35:06 -0000
Subject: Twisted Selector
Message-ID: <u5rsqqb9tamaad@corp.supernews.com>

I am looking at this Twisted tutorial:
http://twistedmatrix.com/page.epy/philosophy.html
and having trouble with the very first code.

>>> from twisted import threadable
ImportError: cannot import name threadable
>>> from twisted.python import threadable
>>>
>>> from twisted import internet
>>> selector = internet.Selector()
AttributeError: 'twisted.internet' module has no attribute 'Selector'

Ok, I was able to get past the first problem. Looks like since
Twisted-0.9.0 when the tutorial was written, they refactored
threadable into a subdirectory, but I grepped the whole distribution
and see nothing called Selector anywhere.

This is with Twisted-0.15.0, but I went back in time to 
Twisted-0.12.3 and still nothing called Selector.

The ChangeLog does not mention Selector either.

Anyone have any ideas?





From marknenadov at winningteam.com  Thu Feb 21 17:29:19 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Thu, 21 Feb 2002 17:29:19 -0500
Subject: Suggestion for best intro to Python
References: <c510a321.0202210608.1acc4d29@posting.google.com>
Message-ID: <20020221.172919.1143408282.1356@d57-47-79.home.cgocable.net>

In article <c510a321.0202210608.1acc4d29 at posting.google.com>, "Bill
Hackmann" <hack at socket.net> wrote:


> What book/web site/online document/something else would be best as a
> source to learn Python for an experienced programmer?  Thanks,
> Hackmann

Start off with www.python.org. That should be enough to keep any newbie
busy for at least a week. After that I would recommend getting
Programming Python 2nd edition by OReilly.

~Mark Nenadov


From gimbo at ftech.net  Tue Feb 26 05:01:36 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 10:01:36 +0000
Subject: urllib and proxy
In-Reply-To: <a5ecas$qiq$1@zeus.polsl.gliwice.pl>
References: <a5ecas$qiq$1@zeus.polsl.gliwice.pl>
Message-ID: <20020226100136.GA18498@andy.tynant.ftech.net>

On Mon, Feb 25, 2002 at 11:03:07PM +0100, Marek Augustyn wrote:

> Is is possible to disable proxy for some urllib.urlopen() call?

Not directly but hey, this is open source so we can make it happen.
:-)

I've attached my_urllib.py which contains a version of urlopen() with
the following behaviour:

    - It takes an optional proxies parameter.

    - Its default behaviour is to not use any proxies.

    - You can explicitly tell it which proxies to use.

    - You can explicitly tell it to do what urllib.urlopen() does, ie
      get the proxy settings from the environment, registry, or
      wherever.

Let's have a look at how I worked out how to do this (which I only did
in the last 20 mins in order to answer your question).  Basically I
looked at urllib.py and saw how that handled proxies, and went from
there...

urllib.urlopen() just creates and utilises a urllib.FancyURLopener
instance.  Note that it keeps the FancyURLopener instance cached in a
global variable for efficiency, but that my_urllib.urlopen() doesn't
do this, for simplicity.

_urlopener = None
def urlopen(url, data=None):
    """urlopen(url [, data]) -> open file-like object"""
    global _urlopener
    if not _urlopener:
        _urlopener = FancyURLopener()
    if data is None:
        return _urlopener.open(url)
    else:
        return _urlopener.open(url, data)

Looking at the constructors of FancyURLopener and URLopener (its
superclass), we have:

class FancyURLopener(URLopener):

    def __init__(self, *args):
        apply(URLopener.__init__, (self,) + args)
        self.auth_cache = {}
        self.tries = 0
        self.maxtries = 10

class URLopener:

    def __init__(self, proxies=None, **x509):
        if proxies is None:
            proxies = getproxies()
        assert hasattr(proxies, 'has_key'), "proxies must be a mapping"
        self.proxies = proxies

	# ... other stuff not important here

So, you can tell URLopener (and hence FancyURLopener) which proxies to
use, but if you don't pass it any, it calls urllib.getproxies() which
gets the proxy settings from the environment, or registry, or whatever
(depending on platform).

So what form do these proxy settings take?  Well, as the assert says,
they have to be a map, and if we look at urllib.getproxies() we see
that in fact it's "a dictionary of scheme -> proxy server URL
mappings."

On my box, this looks like this:

h[1] >>> import urllib
h[1] >>> urllib.getproxies()
{'http': 'http://127.0.0.1:8080/'}

So, to tell FancyURLopener (or URLopener) _not_ to use any proxies,
and also to _not_ get the "system" proxies from the environment or
wherever, we simply pass it an empty dictionary as the proxies
parameter.

Which is what my_urllib.urlopen() is capable of.

I haven't _extensively_ tested the code, but it looks right to me,
assuming I haven't made a mistake in the reasoning outlined above.
All comments/feedback appreciated.

Hope this helps,

Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.
-------------- next part --------------
#!/usr/bin/env python
#
# my_urllib.py
#
# urlopen() with optional proxying.
#
# Changelog:
#   2002.02.26.0945 AMG     v1.0
#                           Created.

"""my_urllib.py -- urlopen() with optional proxying."""

__version__ = "1.0"
__author__ = "Andy Gimblett"

import urllib

def urlopen(url, proxies={}, data=None):

    """Adaption of urllib.urlopen() with optional proxying.

    Adaption of urllib.urlopen() with optional proxying.  Parameters
    and behaviour are as urllib.urlopen() except for the optional
    proxies parameter.

    proxies
    
        By default this is an empty dictionary, which results in no
        proxy being used.

        Alternatively could be a dictionary of scheme -> proxy server
        URL mappings, eg {'http': 'http://127.0.0.1:8080/}
    
        Alternatively could be None, in which case urllib.getproxies()
        is called by URLopener.__init__() to obtain proxy settings
        from the environment/registry/whatever.

    """

    # nb: Unlike urllib.urlopen(), we create a new FancyURLopener
    # instance for each call.  This is less efficient but easier to
    # grok.  Adapt as you see fit.
    
    urlopener = urllib.FancyURLopener(proxies)
    if data is None:
        return urlopener.open(url)
    else:
        return urlopener.open(url, data)

From donn at u.washington.edu  Thu Feb  7 12:18:12 2002
From: donn at u.washington.edu (Donn Cave)
Date: 7 Feb 2002 17:18:12 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news>
Message-ID: <a3ucsk$12h8$1@nntp6.u.washington.edu>

Quoth ullrich at math.okstate.edu (David C. Ullrich):

| An amusing aspect of all this: I just browse clp occaisionally,
| since I'm an abstraction manipulator, not a programmer. So I
| overheard comments about this guy before seeing any examples
| of his output. Before seeing the things he wrote I was gonna
| say "you guys got it easy - you should see sci.math". But
| the funny thing is the guy reminds me _exactly_ of the
| sci.math guy I had in mind: If you can't see what a genius
| he is you're incompetent, you're all gonna be fired when
| the Truth comes out, etc...

So is sci.math choked with posts from people who normally seem
to have their heads screwed on, endlessly replying this guy as
if they sincerely think there's some purpose in it?  People who
take time several times every day to post about how useless his
ideas are and how they have no interest in helping?  Or are
they saner there than we are?

The existence of people with unusual ideas isn't surprising.
I'm kind of taken aback by the response here, though.

	Donn Cave, donn at u.washington.edu


From kamp at inl.nl  Mon Feb 11 05:54:31 2002
From: kamp at inl.nl (Peter H.J. v.d. Kamp)
Date: 11 Feb 2002 10:54:31 GMT
Subject: Python MySQL and MAC OS X
References: <3C656ACD.1020907@inter.nl.net> <mailman.1013294776.15973.python-list@python.org> <3C663807.AA608724@xs4all.nl>
Message-ID: <a487t7$qhb$1@highway.leidenuniv.nl>

>> > According to the MAC OS X documentation of Python 2.0 I have
>> > to drop the setup.py file onto the interpreter and pass the
>> > build argument by setting a Unix style command line (which can
>> > be set by holding the option key during interpreter startup).
>> 
>> Sounds frightening ;-) Don't you have a shell or terminal window on
>> MacOS X?
>
>Sure there is. I think maybe he's not yet aware that there are still two
>very different version of Python available for OSX: MacPython, which
>is still backwards compatible with MacOS 9 and earlier, and then there's
>plain vanilla "unix Python". It sounds like he's trying with MacPython
>where he really needs unix Python.


Sorry, that was an omission in my original posting. I'm indeed working with
MacPython and not with the plain vanilla "Unix Python".


Peter



From jeff at ccvcorp.com  Wed Feb 20 14:10:40 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 20 Feb 2002 11:10:40 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014221254.26490.python-list@python.org> <7xu1scdsxh.fsf@ruckus.brouhaha.com>
Message-ID: <3C73F4AF.4028FDD7@ccvcorp.com>


Paul Rubin wrote:

> "Sean 'Shaleh' Perry" <shalehperry at attbi.com> writes:
> > this is a design decision in the language.  By not allowing this style of
> > programming we do not have the ever so fun:
> >
> > if foo = 2:
> >    handle_foo()
> >
> > type bugs.  Sure there are ways around it.
>
> So why are augmented assignments not allowed in expressions?
>
>   if (foo += 2) < 7:
>         handle_foo()
>
> doesn't seem likely be an =/== confusion.

Because the benefits of doing this are small compared to the detriment of
muddying the statement/expression distinction, perhaps?  It would be a weird
special case, behaving differently from everything else in the language, and
therefore a wart.

Jeff Shannon
Technician/Programmer
Credit International




From syring at email.com  Thu Feb 21 02:29:02 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 08:29:02 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1>
Message-ID: <a527jt$42g58$1@ID-7529.news.dfncis.de>

"Carl Banks" <idot at vt.edu> schrieb im Newsbeitrag
news:r2615a.ii.ln at 127.0.0.1...
> Karl M. Syring wrote:
> > "Carl Banks" <idot at vt.edu> schrieb>
> > <snip>
> >> Flight simulation: The earth's radius is about 20 million feet (about
> >> 7 million meters for my SI using friends), and the simulator must
> >> resolve, at a minimum, distances of about 0.1 feet.  That's a
> >> difference of 9 decimal places.  We then need several more places for
> >> numerical robustness.
> > <snip>
> >
> > You surely are joking, Mr. Banks?
> > This is exactly the kind of misuse of floating point that was referred
to in
> > a previous posting. Every computational geometry software would use a
> > rational data type here, although you might get away with simple point
data
> > binning in this case.
>
>
> I suspect you don't know what I'm talking about, since what I'm
> referring to is nothing like computational geometry.  Nevertheless,
> I'll answer this.
>
> No, I'm not joking.
>
> I have seen the source code of maybe a dozen flight simulators, and
> I've written several myself (not in Python, though, yet).  Several of
> the simulators are used in research, one came out of NASA, and one was
> proprietary.  I have yet to see anything stored as a rational number.
>
> High-fidelity, long-period spacecraft simulations might require
> something more precise than 64-bit floating point, but for atmospheric
> flight, unmodeled disturbances make the extra precision pointless.

But this is really dilettantic work and NASA today is an organization run by
amateurs. If you look at real computational geometry programs like CGAL, you
can see, how it is done.
Anyway, it is not necessary to appeal to authorities to get a grasp of the
problem.: Imagine, you have a polygon in the plane and you do a series of
rotations and translations on it. Now, you want to know whether the position
is exactly identical to another polygon. If you do this with floating point,
you will never get an exact result, while in rational coordinates you always
will.

is-the-loss-of-knowledge-a-form-of-postmodernism?
Karl M. Syring




From jason at jorendorff.com  Mon Feb 18 03:19:04 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 02:19:04 -0600
Subject: entries between labels dynamically
In-Reply-To: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
Message-ID: <HFEKILOLEFEFMKAECNDLMEFMDDAA.jason@jorendorff.com>

Artur Skura wrote:
> I have a specific problem: I have to create dialogs on the fly, and
> in those dialog text together with text entries.

The code below works in Jython.  Not to say that Java Swing is
necessarily the way to go here, of course; just another option.

## Jason Orendorff    http://www.jorendorff.com/


# jythonwocky.py - Text fields between labels, dynamically

from java.awt import *
from javax.swing import *
import re

text = """'Twas _____, and the _____ _____
Did _____ and _____ in the _____.
All _____ were the _____
And the _____ _____.
"""

frame = JFrame("Blanks", size=(500, 300))
mainpanel = Box.createVerticalBox()

for line in text.split('\n'):
    mainpanel.add(Box.createVerticalStrut(6))
    parts = re.split(r'(_+)', line)
    hb = Box.createHorizontalBox()
    for part in parts:
        if part:
            if part.replace('_', '') == '':
                hb.add(JTextField(8, preferredSize=(100, 20),
                                  maximumSize=(100, 20)))
            else:
                hb.add(JLabel(part))
    hb.add(Box.createHorizontalGlue())
    mainpanel.add(hb)

mainpanel.add(Box.createVerticalGlue())
frame.contentPane.layout = BorderLayout()
frame.contentPane.add(mainpanel, BorderLayout.CENTER)
frame.visible = 1






From thomas at gatsoft.no  Thu Feb 28 08:48:17 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Thu, 28 Feb 2002 13:48:17 GMT
Subject: HTTP chunk Transfer-Encoding
References: <4Wpf8.13$zd.865@nreader2.kpnqwest.net>
Message-ID: <Byqf8.135$5vh.170870272@news.telia.no>

Medusa has some code for that. I'm looking into implementing that myself but
don't have the time right now. Please post any code or findings you can
share.

Best regards,
Thomas Weholt


"Francis Meyvis" <francis.meyvis at sonycom.com> wrote in message
news:4Wpf8.13$zd.865 at nreader2.kpnqwest.net...
> Hoi,
>
> Is there some python source that sends chunked packets (RFC 2616) in
> the SimpleHTTPServer or BaseHTTPServer?
> Or do you know where to find snippet of code that can split up
> file to send them in chunk format?
>
> Thx
>
> --
> ---
> Kind regards, best wishes,
> francis meyvis
>
>




From JoeSalmeri at home.com  Fri Feb  1 21:42:34 2002
From: JoeSalmeri at home.com (Joe Salmeri)
Date: Sat, 02 Feb 2002 02:42:34 GMT
Subject: Python 2.2 + Win32All 146 = Python/ASP Failures
References: <vtk68.20338$8d1.7162721@news1.rdc1.md.home.com> <3C59DA37.9070506@skippinet.com.au> <JEF68.22313$8d1.8213037@news1.rdc1.md.home.com> <3C5B4A9A.8000809@skippinet.com.au>
Message-ID: <umI68.22589$8d1.8384250@news1.rdc1.md.home.com>

No problem, I appreciate the assistance :-)

So far everything is now working again.

I will report back if I can get any failures.

Thanks again!

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C5B4A9A.8000809 at skippinet.com.au...
> Joe Salmeri wrote:
> > Mark,
> >
> > I grabbed framework.py from win32all-142.exe and replaced it in the
> > installation as you suggested.  I also deleted the framework.pyc just to
> > make sure the new version was compiled.  Next I restarted the web
server,
> > just as a sanic check to make sure it saw the changes to the python asp
> > engine.
>
> Sorry - my mistake.  You need to go from win32all-140 down to
> win32all-133 (inclusive) for a working version :(  I have mailed you a
copy.
>
> Mark.
>




From gtalvola at nameconnector.com  Mon Feb 25 11:24:28 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Mon, 25 Feb 2002 11:24:28 -0500
Subject: Concurrent XML-RPC?
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E41240E4@MAILBOX.nameconnector.com>

Jordan Krushen write:
> I know this has been asked before (or something close), but 
> having explored
> as much as I can, I still haven't found an existing, 
> lightweight way to have
> true concurrency when serving XML-RPC.  While Medusa is ok 
> for IO-bound
> stuff, we'll be doing some heavy db-driven processing, which 
> when run under
> Medusa is serialized.  Obviously, the SocketServer code 
> behaves this way, as
> well.
> 
> Is there a patch for the SocketServer code anywhere to allow 
> it to run under
> inetd?  I'd hate to have to rewrite all the xmlrpclib handling of the
> connections, and this would allow me to run it under DJB's tcpserver,
> obviously giving us the concurrency we need.
> 
> I've also heard that Zope's ZServer might do the trick, is this simple
> enough?  (Note the above mentioning of 'lightweight' :)
> 
> What we're wanting is a framework where one can simply take 
> the appropriate
> library and use an automated installer, which would talk to a 
> centralized
> authority telling it which IP/port to bind to, set up all the
> stunnel/tcpserver instances, generate certificates, etc.  I imagine
> somewhere around 50 separate, independent XML-RPC services 
> would exist at a
> time, talking between themselves and the outside world.  Of 
> course, many of
> these instances may be running on the same host, so we'd like 
> them to be as
> lightweight as possible.
> 
> Any thoughts?

Webware's WebKit application server supports XML-RPC and is multi-threaded,
so it gives you the concurrency you need.  The only problem is that it
doesn't currently support the HTTP protocol directly, but instead requires
you to run it behind Apache, which probably disqualifies it as
"lightweight".  (There are at least 3 different experimental HTTP servers
for Webware which you are welcome to try but none are part of the Webware
core -- yet.)

http://webware.sourceforge.net

- Geoff



From francois-regis.chalaoux at sanofi-synthelabo.com  Mon Feb 18 10:51:03 2002
From: francois-regis.chalaoux at sanofi-synthelabo.com (copter24)
Date: 18 Feb 2002 07:51:03 -0800
Subject: strange libtk8.3.so: open failed
Message-ID: <85e27928.0202180751.6782f145@posting.google.com>

Could somebody explain me why I obtain the following error when I
invoke a simple cgi script (print "hello world"). In the Python
interpreter, import _tkinter, import Tkinter and Tkinter._test() are
ok.

Bye, FR.

===========================================================================

ld.so.1: /usr/local/bin/python: fatal: libtk8.3.so: open failed: No
such file or
 directory
[Mon Feb 18 16:45:51 2002] [error] [client 172.24.34.90] Premature end
of script
 headers: /usr/Micromass/GlobalServer/V1.1/WebServer/Apache/cgi-bin/lib/test.py


From aahz at panix.com  Tue Feb 19 16:36:15 2002
From: aahz at panix.com (Aahz Maruch)
Date: 19 Feb 2002 13:36:15 -0800
Subject: basic books/guides on multithreading programming?
References: <5.1.0.14.2.20020216055418.02817d08@202.95.131.154> <mailman.1013816158.26552.python-list@python.org> <mailman.1014139059.23770.python-list@python.org>
Message-ID: <a4uggf$ht6$1@panix2.panix.com>

In article <mailman.1014139059.23770.python-list at python.org>,
Tripp Scott  <tripps81 at yahoo.com> wrote:
>
>ok, i understand more or less about deadlock or lock ordering (from the
>database equivalent terms), though have not heard about livelock or the
>others. suppose i decide to be ignorant about these issues "for now"
>and go about building my multithreading project. and then suppose these
>issues cause serious problems. would it be difficult to resolve it?
>would i have to rethink/review all the threaded code?

That's a "definite maybe".  ;-)  Much depends on the complexity of your
application and the extent to which you divorce the threading issues
from the application logic.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From greg at cosc.canterbury.ac.nz  Tue Feb 26 23:03:55 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 27 Feb 2002 17:03:55 +1300
Subject: Proposal: local variable declarations and type advice
References: <7xpu2slwtl.fsf_-_@ruckus.brouhaha.com>
Message-ID: <3C7C5AAB.4349BCFF@cosc.canterbury.ac.nz>

Paul Rubin wrote:
> 
>      odds = [local x for x in range(10) if x%2 != 0]

That should be

   odds = [x for local x in range(10) if x%2 != 0]

The first thing in a list comprehension is in general an
expression, not just a variable.

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


From starx at pacbell.net  Sat Feb  9 15:35:32 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 20:35:32 GMT
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4> <3c657554.20448468@news.t-online.de>
Message-ID: <Xns91B0801A91441RASXnewsDFE1@130.133.1.4>

Gerson Kurz || Sat 09 Feb 2002 11:21:29a:

> On 9 Feb 2002 18:21:31 GMT, Philip Swartzleonard <starx at pacbell.net>
> wrote:
> 
>>Anyway, i was just wondering, since this seems so great, is there 
>>anything wrong with doing things this way that i'm not seeing? 
> I've been using code like that quite frequently, so if its bad, I'm in
> for it, too.

=)
 
>>Also, is there any real reason to have the message classes inherit when 
>>the only thing that seems to identify them is an attribute? 
> No, it works if you omit the "Message" Inheritance. You could use
> plain strings and have the same effect. 

Well, i need to have data stuck on the message for whatever recives it, so 
each should have it's own init prolly. map.shpere_msg(12,13,14, M_Damage( 
type = dam.fire | dam.plasma, value=40000) ), or something to that effect 
:)

What i'd like to know is, is there a benefit to using an inherited system 
(like there is for say, exceptions, where you can catch base classes). 
Would it be better to define a message that passes through a certain 
substance as a mixin or base class, or by a attribute passes_vaccum=1 type 
of thing (though it would be more likely to do a more in-depth analysis of 
the object to detrmine this :)

Thanks, though :) 



-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From huaiyu at gauss.almadan.ibm.com  Thu Feb 14 13:47:30 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Thu, 14 Feb 2002 18:47:30 +0000 (UTC)
Subject: Proposal for quoted octal
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net>
Message-ID: <slrna6o1i2.u4q.huaiyu@gauss.almadan.ibm.com>

On Wed, 13 Feb 2002 07:51:36 -0800, John W. Baxter <jwbaxter at spamcop.net> wrote:
>In article <mailman.1013611523.15053.python-list at python.org>, Simon
>Brunning <SBrunning at trisystems.co.uk> wrote:
>
>> Probably to late to do anything about it now - think of all the code which
>> would break!
>
>Did the elves at AT&T pick up the "leading 0 means octal" from
>something earlier, or did they invent this stupidity themselves?
>
>Either way, we're stuck with it.

Not necessarily stuck for ever.  How about the following plan in three steps?


				Proposal to change octal notation

1) allow \041 to mean what is today 041, ie 33.  It might be even better to
   use \O41 (the letter O).

2) from __future__ import backquote_octal so that 041 will mean 41.
   from __future__ import zerolead_octal so that 041 will mean 33.
   Otherwise octals produce a warning.

3) make backquote_octal mandatary.  

As long as steps 2 and 3 are far apart this will not cause much trouble.
This does not change the meaning of \041 in a string:

	 '\041' == '!';     \041 == 33;      041 == 41

Huaiyu


From sorr at rightnow.com  Fri Feb  8 15:43:16 2002
From: sorr at rightnow.com (Orr, Steve)
Date: Fri, 8 Feb 2002 13:43:16 -0700 
Subject: Perl programmers?
Message-ID: <D408B6AC060DB24790622E1A54204DB00823AA90@xchange.rightnowtech.com>

Perl mongers...
Go forth and be thou pearlescent in the dullness of thy day to day routine.



-----Original Message-----
From: aahz at panix.com [mailto:aahz at panix.com]
Sent: Friday, February 08, 2002 12:11 PM
To: python-list at python.org
Subject: OT: Perl programmers?


I know some of y'all still keep roots to the Perl community, so I was
wondering if there's any kind of definitive nickname for Perl
programmers, the way we call ourselves Pythonistas?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth
-- 
http://mail.python.org/mailman/listinfo/python-list



From syver-en+usenet at online.no  Mon Feb 11 12:10:24 2002
From: syver-en+usenet at online.no (Syver Enstad)
Date: Mon, 11 Feb 2002 17:10:24 GMT
Subject: Emacs python mode and pdb.. shouldn't they cooperate??
References: <m3y9i0rud3.fsf@localhost.localdomain>
Message-ID: <ubsewymgu.fsf@online.no>

Andrew Markebo <flognat at flognat.myip.org> writes:

> Hey emacs lovers!
> 
> How do you debug in your emacs?? Or you don't need to?? Once upon a
> time, long, long ago I ported the xemacs gud-back/frontend, whatever
> to emacs, but it is uncosy to use.. 

I don't get the regular support for pdb to work either. But that is
not a real problem to me as the latest version of python-mode (which
you can find on www.python.org) has a feature known as pdb-track.

This works by looking out for the  "(Pdb)" prompt in a shell or
python-shell window and kicking in from there. The pdb track feature
will show the current source file with a little arrow in the fringe
pointing to the current source line. You can use pdb as normal with
text commands in the shell or python-shell, and pdbtrack shows right
source line at all times. Works great, if you want to add your
favorite shortcut keys for stepping in, out, over and so on just add
them to the py-shell-map and map them to inserts of the commands that
does these things in pdb.

The only thing I really miss is setting breakpoints directly in the
source code, which would make python mode almost complete.

-- 

Vennlig hilsen 

Syver Enstad


From bkelley at wi.mit.edu  Wed Feb  6 10:40:31 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Wed, 06 Feb 2002 09:40:31 -0600
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <3C614C3F.7000508@wi.mit.edu>
Message-ID: <3C614E6F.4000607@wi.mit.edu>

The formatting got a little screwed up, the delve function should be:

>
> def delve(input):
>    """(input)->recursively split input
>    i.e.
>    ('cat', 5, ['dog', [3, 3, 1]], 'zoo')
>    will reult in
>    ('c', 'a', 't', 5, 'd', 'o', 'g', 3, 3, 1, 'z', 'o', 'o')
>
>    warning, purposely uses maximum recursions!
>    """      

>    target = []
>    _delve(data, target)
>    return tuple(target)
>
>



From p.mayers at ic.ac.uk  Thu Feb 21 09:40:33 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Thu, 21 Feb 2002 14:40:33 +0000
Subject: Large integers from C
References: <20020220.163039.1504569917.30753@ic.ac.uk> <mailman.1014230314.2890.python-list@python.org> <c76ff6fc.0202201519.509ec73e@posting.google.com>
Message-ID: <20020221.144033.1012502954.7349@ic.ac.uk>

In article <c76ff6fc.0202201519.509ec73e at posting.google.com>, "John
Machin" <sjmachin at lexicon.net> wrote:

> Gustavo Niemeyer <niemeyer at conectiva.com> wrote in message
> 
> It is possible that the OP wants, but is not looking very hard for,
> this:
> 
> [from Python 2.2 source, in include/longobject.h] """
> /* _PyLong_FromByteArray:  View the n unsigned bytes as a binary integer
> in
>    base 256, and return a Python long with the same numeric value. If n
>    is 0, the integer is 0.  Else:
>    If little_endian is 1/true, bytes[n-1] is the MSB and bytes[0] the
>    LSB; else (little_endian is 0/false) bytes[0] is the MSB and
>    bytes[n-1] the LSB.
>    If is_signed is 0/false, view the bytes as a non-negative integer. If
>    is_signed is 1/true, view the bytes as a 2's-complement integer,
>    non-negative if bit 0x80 of the MSB is clear, negative if set. Error
>    returns:
>    + Return NULL with the appropriate exception set if there's not
>      enough memory to create the Python long.
> */
> extern DL_IMPORT(PyObject *) _PyLong_FromByteArray(
>         const unsigned char* bytes, size_t n, int little_endian, int
>         is_signed);
> """

Ah, I should probably point out I am restricted to using Python 1.5.2 for
various dull reasons, but in theory that is what I'm looking for.
However, I note that API isn't listed in the C API documentation.

I'll probably try the PyLong_FromString, and see if that helps.

Cheers,
Phil


From mark at mceahern.com  Wed Feb  6 22:15:23 2002
From: mark at mceahern.com (Mark McEahern)
Date: Wed, 6 Feb 2002 19:15:23 -0800
Subject: money class?
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEENKBAA.mark@mceahern.com>

Without starting an endless thread on how many floating point angels can
dance on the head of a pin, I'd like to ask:

	Does anybody know of a nice Money class they'd be willing to share, point
to?

(I did a little bit of searching, but found nothing obvious in the 5-10
minutes I spent.)

I don't think it'd be too hard to write one myself, but I figured I'd see
what's out there first.  I don't need something that handles different
currency.  US Dollars is good enough.  I suppose I'll be taken to task (or
ignored) for not specifying exactly what I need or why I need it.  What can
I say?  I'm lazy.  ;-)  Does this suffice?

	It should "Do the right thing."

Thanks,

// mark




From kalath at lycos.com  Tue Feb 12 04:54:54 2002
From: kalath at lycos.com (Mark Brady)
Date: 12 Feb 2002 01:54:54 -0800
Subject: Python 2.2 class methods examples?
Message-ID: <e840346c.0202120154.6b41f449@posting.google.com>

Hi. I've just moved to 2.2 and I think I've got my head around most of
the cool new features but I can't see what class methods would be used
for. I understand the difference between them and static methods and I
know where static methods are useful but it would be great if someone
could post an example of class method usage or some links. Thanks.

Mark.


From shalehperry at attbi.com  Fri Feb 22 22:17:44 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Fri, 22 Feb 2002 19:17:44 -0800 (PST)
Subject: Request for comments: Search and replace script
In-Reply-To: <3C76E36F.3030501@verio.net>
Message-ID: <XFMail.20020222191744.shalehperry@attbi.com>

> 
>  > check() makes some of the same checks over and over, perhaps you 
> should look at
>  > the code's flow here and see if it can be done differently
> 
> I am not sure what you mean here.  What I tried to do was to define each 
> separate check (binary, openable, etc) in its own function and then have 
> one function that would call them all in the correct order.  Is this not 
> the best way to do it?
> 

you check isDirectory(filename) 3 times in check().  The first time you return
so the others can not possibly be true.




From schoeller at zkm.de  Fri Feb  8 11:04:35 2002
From: schoeller at zkm.de (=?ISO-8859-1?Q?Andreas_Sch=F6ller?=)
Date: Fri, 8 Feb 2002 17:04:35 +0100
Subject: Stupid Question
Message-ID: <8BAD9ABC-1CAD-11D6-9876-0003937266F6@zkm.de>

Hello list,

is it possible to delete a key from a Btree() ? I?m using a IOBTree from 
the ZODB Standalone dist. A simple del tree[key] gives me a __delitem__ 
error ? Reassign works as expected tree[key] = value.

thanks, andreas




From hat at se-46.wpa.wtb.tue.nl  Tue Feb  5 11:36:12 2002
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 5 Feb 2002 16:36:12 GMT
Subject: Expect like syntax in Python
References: <3c5cf3d6_2@news.nwlink.com>
Message-ID: <slrna602gj.ct4.hat@se-46.wpa.wtb.tue.nl>

On Sun, 3 Feb 2002 00:25:14 -0800, Guy Gascoigne - Piggford <guy at wyrdrune.com> wrote:
>I'm trying to find a way to write something like expect in a python module,
>and I'm having trouble getting the syntax the way that I want.  In expect I
>can write something like this:
>
>expect {
>    -re "ast login" {
>           pass
>           }
>    -re "ogin" {
>           send  user
>           }
>    -re "sword" {
>           send password
>           }
>}

A known problem :-)

I also already noticed that the existing Expect-like thingies are not much help
here, they also mangle the order of the patterns and the actions.
I have been thinking that the following should be possible:

x = ExpectMatcher()		# A random name, needs some more thought
...	# do some initing stuff, start up a sub-process, connect that to x, etc
while not x.ismatched():
    if x.trymatch('ast login'):
        pass
    elif x.trymatch('ogin'):
        send user
    elif x.trymatch('sword'):
        send password

There are at least 2 options for implementation (but I haven't had the
opportunity to implement either of them).

1) (stupid way)
   The x object reads data in the x.ismatched() method. If it has 'enough', it
   returns false. Python then executes the x.trymatch() methods. In the method,
   you compare the imput with the argument. If it matches, you set a flag that
   x.ismatched() should return true on the next call, followed by returning
   true (so the if test succeeds).
   If all x.trymatch() calls fail, Python executes the 'x.ismatched()' again,
   given you a chance to read more data.

2) (intelligent way)
   The x.ismatched() returns false in the first call. Python then executes the
   if, and as a result, all the x.trymatch() calls. Store the patterns, and
   return false to all of them.
   On the second call to x.ismatch(), you have a list of patterns, so you can
   wait until you have seen a pattern match.
   At that moment, return false again, and now return false to all x.trymatch()
   methods, except the one that matches.
   Python handles the code in the if branch at that point, and returns to the
   while.
   On the 3rd call to x.ismatched() return true.

(the above scheme should be extended to include time-out)

There are probably other/better implementations possible, I haven't really
considered alternatives except the above 2.

>Am I missing something?  I'm fairly new to python but I don't see a way to

Probably not. I took me a few months and a few tricky twists to come up with
the above.
Although the solution seems sound to me, I haven't really tried to implement it.


Albert
-- 
Constructing a computer program is like writing a painting


From gustav at morpheus.demon.co.uk  Wed Feb 20 16:59:08 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Wed, 20 Feb 2002 21:59:08 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <3C6D8B63.5070909@geocities.com> <mailman.1013864067.16756.python-list@python.org>
Message-ID: <69587ukeu1m51mf46kpumjnt9qfujho7ju@4ax.com>

On Sat, 16 Feb 2002 12:52:55 +0000, Paul Moore
<gustav at morpheus.demon.co.uk> wrote:

>I'll go and work through all the suggestions I have and come up with
>something suitable. I'll report back on what I manage to produce.

I've had a play with the various options, and mulled the whole thing
over a bit. The conclusion I've come to is that it's certainly possible
to do the sort of thing I'm after, but it's probably too messy (at the
moment) to make it worthwhile.

One thing I discovered is PEP 273, which proposes to allow zipfiles on
sys.path. This is proposed for Python 2.3, and would make zipping up
.py[c] files a much more reasonable option - until this gets
implemented, the necessary import hacks to load modules from a zip file
are possible, but messy. I'll wait for the official implementation...

Getting "resources" out of a zipfile is relatively simple. The zipfile
module lets you get at the data, and StringIO lets you wrap it as a
file-like object. There are still a few issues, though...

1. The data is read in "binary" (with \r\n line terminators on Windows).
StringIO wraps this "raw", which could be a problem if you are getting a
textfile resource, and need to read in text mode. The simplest answer is
to do a data = data.replace("\r","") before wrapping the data.

2. All the data gets read into memory, which could be an issue for big
chunks of data. Translating text files as in (1) above creates a second
copy - more memory used.

Looking at specific types of data

Text data - No real problem. Read as a string, or wrap as a StringIO.

Sound - Not a problem on Windows, as the winsound module takes in-memory
data as an option.

Images (PIL) - Not a problem, as PIL constructors can take a file-like
object as well as a filename.

Images (wxPython) - A bit messy, the simplest way is to go via PIL.
(This leads to yet more in-memory copying, so it may be a problem for
big images...) The next version of wxPython is likely to contain
constructors which take file-like objects, so things will be simpler
then...

One obvious point is that you have to code specifically for reading from
a zipfile rather than from the filesystem. This is a nuisance, unless
you write a wrapper class which lets you choose filesystem vs zipfile at
runtime.

Fredrik Lundh's Squeeze package lets you pack up Python code and data
files in a single archive, and read datafiles from the archive at
runtime. It's almost exactly what I'm after. It uses an internal data
format rather than standard zipfiles - a reasonable decision, but not
quite what I was aiming for.

The basic feeling I get is that there are loads of ways of doing this
sort of thing. None of them are too hard, but on the other hand, there
isn't one "off the shelf" standard way of doing this, so it would be
necessary to code what you wanted into your application (or more
sensibly, as a reusable component which you use in your application). In
most cases, I suspect it's not worth the bother for any specific
application.

I could see some value in a PEP to propose a standard way of doing this
sort of thing - probably as a library module, maybe with some sort of
core support. But I don't know that anyone's likely to feel strongly
enough to push it. (I don't think I am...)

Thanks for all the suggestions and comments, it was an interesting
exercise...

Paul.


From feldspar at ix.netcom.com  Mon Feb  4 14:15:49 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Mon, 04 Feb 2002 15:15:49 -0400
Subject: Please help with python and "curses"!
References: <b22b2103.0201281206.1eb0c092@posting.google.com> <mailman.1012249785.6941.python-list@python.org>
Message-ID: <3C5EDDE5.2479B9F2@ix.netcom.com>

Cliff Wells wrote:
> 
> On 28 Jan 2002 12:06:15 -0800
> Antaeus Feldspar wrote:

> > Note the commented-out "start_color()" line.  This is listed in the

> This works under Python 2.1.1, but not under 1.5.2, AFAICT.

> > says that a curses.newwin command such as in the above line "secondwin =
> > curses.newwin(10,10,2,1)" should create a new window that will 

> All seems to work under 2.1.1.  I get a nice big window with a thin border
> around it.  Under 1.5.2 I get jack.  Maybe you should upgrade?

Hmmmmm, upgrading's something I have to put off for a while.  This part
(the new window part, that is) I now know how to do under 1.5.2, so ...
I can live without color for a while.  ^^  Thanks for your help!

	-jc


From ivo at nospamAmaze.nl  Tue Feb  5 06:36:56 2002
From: ivo at nospamAmaze.nl (ivo at nospamAmaze.nl)
Date: 5 Feb 2002 11:36:56 GMT
Subject: Fun with lambda and map
References: <mailman.1012542242.4324.python-list@python.org> <a3mddf$7do$1@news1.xs4all.nl> <3c5eaf65.22609703@news.t-online.de>
Message-ID: <a3og4o$6l0$1@news1.xs4all.nl>

Gerson Kurz <gerson.kurz at t-online.de> wrote:
: Well, you can:

: IF_ELSE = lambda x:(map(lambda a:a and 1,[x[0]()])[0] and x[1]()) or
: x[2](x[3])
: fac = lambda n:IF_ELSE([lambda:not n,lambda:n==n,lambda
: n:fac(n-1)*n,n])

Unfortunately, this uses and/or and some list operations, which are not
defined in the lambda calculus I presented. Of course, you could define
them in pure lambda, but I don't thing they would have the short-circuit
behaviour you want.


: If you're in a masochistic mood, you can write this in one line:

: fac = lambda n:map(lambda x:(map(lambda a:a and 1,[x[0]()])[0] and
: x[1]()) or x[2](x[3]),[[lambda:not n,lambda:n==n,lambda
: n:fac(n-1)*n,n]])[0]

I'm in an academic masochistic mood, so I don't want to borrow anything
from python (btw, I got a solution by email that is pure lambda, but I
first have to figure out how it works)

	Ivo



-- 
Drs. I.R. van der Wijk                              -=-
Brouwersgracht 132                      Amaze Internet Services V.O.F.
1013 HA Amsterdam                                   -=-
Tel: +31-20-4688336                          Linux/Web/Zope/SQL
Fax: +31-20-4688337                           Network Solutions
Web:     http://www.amaze.nl/                    Consultancy
Email:   ivo at nospamAmaze.nl                         -=-



From philh at comuno.freeserve.co.uk  Sat Feb 23 15:33:49 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sat, 23 Feb 2002 20:33:49 +0000
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie> <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk> <7xheo9uftb.fsf@ruckus.brouhaha.com>
Message-ID: <slrna7fv5d.5oo.philh@comuno.freeserve.co.uk>

On 22 Feb 2002 17:50:40 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:
>philh at comuno.freeserve.co.uk (phil hunt) writes:
>> What do you consider the other 2 methods to be? I can think of 
>> encoding it in the URL, and the server remembering the browser's IP 
>> address. Are there others?
>
>You CANNOT identify sessions by browser IP addresses. 

What's bugging you? I suggest you calm dowm.

> That's been
>discussed here several times.  Think of proxy servers sending requests
>from many browsers through the same IP address--for example, AOL
>proxies millions of browsers through a few hundred addresses.

I am perfectly aware of the limitations of this method, thank you 
very much.

>A number of other means of identifying sessions have been discussed
>here in the past few weeks though.

Indeed, I have taken part in these discussions.

-- 
### Philip Hunt === philh at comuno.freeserve.co.uk ###





From rdsteph at earthlink.net  Sun Feb 10 20:28:42 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Mon, 11 Feb 2002 01:28:42 GMT
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org> <3C66F2CA.802A00BF@earthlink.net> <mailman.1013387057.27588.python-list@python.org>
Message-ID: <3C671F29.4688C7E6@earthlink.net>

Chris Gonnerman wrote:
> 
> ----- Original Message -----
> From: "Ron Stephens" <rdsteph at earthlink.net>
> 
> > Since id dint get any mail despite putting your insert into the top
> > pf mygale.py
Thanks.


> >
> > import sys
> >     sys.stdout = open("/home/ron/.error-log","a")
> >     sys.stderr = sys.stdout
> >     print "\n(job name) Starting..."
> >
> > I wonder if my mail is broken or if it is record somewhere else???
> 
> Exactly.  You misunderstood, I think... if you DIDN'T have the
> lines described, cron should send you mail.  With those lines,
> all your output from the script(s) should be in the file
> 
> /home/ron/.error-log
> 
> and you can view them with less, more, vi, etc.


From rdacker at pacbell.net  Mon Feb  4 15:12:53 2002
From: rdacker at pacbell.net (rdack)
Date: 4 Feb 2002 12:12:53 -0800
Subject: [OSX] milter won't build
References: <mailman.1012785372.11813.python-list@python.org>
Message-ID: <644f6688.0202041212.1f3cfabf@posting.google.com>

osx 10.1.2, python2.1, milter-0.3.9
fixed first problem adding  '-flat_namespace' to config/makefile

now i get:
cc -bundle -flat_namespace -undefined suppress
build/temp.darwin-5.2-Power Macintosh-2.1/miltermodule.o -lmilter -lsm
-o build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
/usr/bin/ld: can't locate file for: -lsm

what file is it looking for? why can't it find it? what is option
'-lsm'? could i just remove it? what file would it be in?


From bp at sam.localdomain  Thu Feb 28 21:13:41 2002
From: bp at sam.localdomain (Bob Parnes)
Date: Fri, 01 Mar 2002 02:13:41 -0000
Subject: Mod-python problem
References: <slrna7rdq5.2vf.bp@sam.localdomain> <f72dac0d.0202280734.77bfa545@posting.google.com>
Message-ID: <slrna7ttd5.1j3.bp@sam.localdomain>

On 28 Feb 2002 07:34:08 -0800, Alan Kennedy <alanmk at hotmail.com> wrote:
>bp at sam.localdomain (Bob Parnes) wrote in message 
>> Just in case I am missing something obvious, the following 
>> is my test code, for a linux system:
>
>
>So, the corrected version, which works for me, is as follows
>
># Script starts here
>
>from mod_python import util
>from mod_python import apache
>
>class clTest:
>    def cltest(self, req, formDict):
>        res = '<html><head></head><body>'
>        res += '<table align=center>'
>        res += '<TR><TH colspan=2>Welcome to the \
>            mod-python class test'
>        res += '<TR><TD colspan=2>keys: %s' \
>		    % formDict.keys()
>        res += '<form method=post>'
>        res += '<TR><TD>Name <TD><input type=text \
>            name="name">'
>        res += '<TR><TD>Phone <TD><input type=text \
>            name="phone">'
>        res += '<TR><TD colspan=2 align=center> \
>            <input type=submit name=submit>'
>        res += '</form></table></body></html>'
>        return res
>
>def handler(req):
>    formDict = util.FieldStorage(req)
>    cl = clTest()
>    req.content_type = "text/html"
>    req.send_http_header()
>    req.write(cl.cltest(req, formDict))
>    return apache.OK
>    
># Script ends here
>
>Hope that works.

Al,

Thanks for the detailed response. Your comments were easy to understand. 
Furthermore your corrected version works for me too, under the conditions 
you had assumed. I learned a lot from this.

The only trouble is that I had forgotten about my httpd.conf file. The 
relevant lines are:

<Directory /var/www/modPython/test>
  AddHandler python-program .py
  PythonHandler mod_python.publisher
</Directory>

This was the reason I used main() instead of handler(); in the url window I 
typed in "test.py/main" and got the table but no response from FieldStorage().
When I changed the third line to "PythonHandler test", your corrected version 
works. When I do not change the line and type 'test.py/handler' in the url 
window, the browser renders a table but also additional lines showing an 
error. The source of the rendered display is as follows:
---------------------------------------------------------------------

<html><head></head><body><table align=center><TR><TH colspan=2>Welcome to the 	
		mod-python class test<TR><TD colspan=2>keys: []<form 
method=post><TR><TD>Name <TD><input type=text 
			name='name'><TR><TD>Phone <TD><input type=text 
			name='phone'><TR><TD colspan=2 align=center>
			<input type=submit 
name=submit></form></table></body></html>HTTP/1.1 200 OK
Date: Fri, 01 Mar 2002 01:34:22 GMT
Server: Apache/1.3.23 (Unix) Debian GNU/Linux mod_python/2.7.6 Python/2.1.2
Connection: close
Transfer-Encoding: chunked
Content-Type: text/html; charset=iso-8859-1

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
<HTML><HEAD>
<TITLE>200 OK</TITLE>
</HEAD><BODY>
<H1>OK</H1>
The server encountered an internal error or
misconfiguration and was unable to complete
your request.<P>
Please contact the server administrator,
 bp at sam and inform them of the time the error occurred,
and anything you might have done that may have
caused the error.<P>
More information about this error may be available
in the server error log.<P>
</BODY></HTML>
-------------------------------------------------------------------------

No lines concerning the error were printed in the error log. But it seems to
me that, as a result of the error, the http header and "OK" from 
"return apache.OK" were simply displayed as text. 

Are publisher and FieldStorage() incompatible, or am I misusing them? Knowing 
what I know now, I can probably get along without publisher, but I am 
accustomed to placing several scripts in one directory.

Bob

-- 
Bob Parnes
rparnes at megalink.net


From theller at python.net  Fri Feb 15 12:03:13 2002
From: theller at python.net (Thomas Heller)
Date: Fri, 15 Feb 2002 18:03:13 +0100
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com>
Message-ID: <a4jf10$och3$1@ID-59885.news.dfncis.de>

"Paul Moore" <paul.moore at atosorigin.com> wrote in message news:2upn6uoqotqejiji69bg8kebfacdu9oamb at 4ax.com...
> One nice feature of Java (sorry!) is the ability to ship an application
> as a single JAR file - containing all code, resources, etc, etc. Is
> there any equivalent feature available in Python? The obvious place
> where it would be is in Gordon McMillan's Installer package, with its
> Python Archive format, but I don't think (from looking at the docs) that
> does quite what I'm after.
>
> I have no requirement to include a Python interpreter, or C extensions,
> in the archive. The code is all pure Python. But, I do want to be able
> to include support files (text files, image files for use in wxPython
> toolbars, graphics for display on screen, sound files, etc etc) which
> can be picked up from the same archive.
>
> I'm looking for something which works a bit like Java's
> ClassLoader.getResourceAsStream() (along with the necessary stuff to
> allow the stream to then be *used* as an icon/image/whatever).
>
> It seems like the sort of thing which should be possible with Python,
> but I can't find it. Did I miss something?

Have you looked into /F's squeeze? Should exactly do what you want,
from the description...

Thomas




From tdelaney at avaya.com  Thu Feb 21 17:58:38 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 22 Feb 2002 09:58:38 +1100
Subject: Reloading modules
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1A6@natasha.auslabs.avaya.com>

> From: Oleg Broytmann [mailto:phd at phd.pp.ru]
> 
> On Thu, Feb 21, 2002 at 12:02:08PM +0000, Michael Hudson wrote:
> >     if file.endswith("c"):
> >         file = file[:-1]
> 
> if file.endswith("c"):
>    file = file[:-1]
> elif file.endswith("o"):
>    file = file[:-1]

if file[-1:] in ('c', 'o',):
    file = file[:-1]

Tim Delaney



From threeseas at earthlink.net  Tue Feb  5 23:54:09 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:09 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <m3n0yqt5ib.fsf@chvatal.cbbrowne.com>
Message-ID: <1462.801T1396T14014714threeseas@earthlink.net>

On 03-Feb-02 20:11:56 Christopher Browne <cbbrowne at acm.org> wrote:
>"Timothy Rue" <threeseas at earthlink.net> wrote:
>> In reviewing over many messages in the autocoding and related
>> threads and seeing the evidence of contridictions, hypocracy,
>> irrational and illogical responses......

>> Its damn clear that there is a need to remove the sort of people
>> posing to be professional who are in essence just good liars,
>> abstraction manipulators, from the industry.

>Why don't you simply put us all in your killfile so you don't see what
>we write?

>I'm sure an increasing number of readers of these newsgroups are doing
>this to you...
>--
>(reverse (concatenate 'string "ac.notelrac.teneerf@" "454aa"))
>http://www3.sympatico.ca/cbbrowne/
>"Bother,"  said Pooh,  "Eeyore, ready  two photon  torpedoes  and lock
>phasers on the Heffalump, Piglet, meet me in transporter room three"


Tell me how prestegious that email address is of your as you will be
forces to pull your head out of your ass and admit that auto-coding is
possible and should have been done a decade or so ago.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From haep1004 at dreamwiz.com  Thu Feb 28 02:15:25 2002
From: haep1004 at dreamwiz.com (Haeyoung Kim)
Date: 27 Feb 2002 23:15:25 -0800
Subject: Want to monitor process..
Message-ID: <87550ef1.0202272315.15779bbd@posting.google.com>

I want to run or kill a process(program?) using python.
Once it's started, I don't want to run the same process any more. (but
can be killed or restarted, of course)
At present I use a naive way to write a pid file when initiate the
process, but in that case I can't restart it when the process has been
down. (i.e., if that when the process has gone, I should kill the
processes and remove pid file manually.)
I'm not so good at such a system programming.. :(
Is there any nice solution?
Thank you in advance.


From eric.brunel at pragmadev.com  Mon Feb 25 13:07:08 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 25 Feb 2002 18:07:08 +0000
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <a5dqoh$1e99$1@norfair.nerim.net>

Moray Taylor wrote:
> I am desperatly seeking a good python editor,
[snip]
> I have only two requirements for my editor, syntax coloring, and a
> Class/function list.

I personnaly use NEdit (http://www.nedit.org), that has syntax colouring, 
is quite fast, but has no class/function list. However, it's highly 
customizable and can call external commands from macros. I already used 
this feature to display a list of functions & class/methods in the current 
file, and another macro allowed to jump to it. If anyone's interested, I 
can provide these customizations (just a couple of NEdit macros and a 
Python script).
HTH
 - eric -


From tanzer at swing.co.at  Tue Feb 19 02:02:36 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Tue, 19 Feb 2002 08:02:36 +0100
Subject: Adding static typing to Python
In-Reply-To: Your message of "Tue, 19 Feb 2002 11:07:19 +1100."
             <a4s51k$8pd1@news1.gtech.com>
Message-ID: <E16d4IO-0006UZ-00@tswings.swing.cluster>

"Peter Milliken" <peter.milliken at gtech.com> wrote:

> The loose typing of Python is perceived as one of its big features :-).
> There are regular discussions on the usefulness of Python for "larger"
> projects. Many claim that they use it without any problems - I suspect that
> it is from a background of ignorance and lack of metrics i.e. how much time
> is lost in their projects to misnamed variables etc?

Your statement smacks of quite a bit of ignorance (not to mention the
arrogance involved).

> The only way in Python to find these is through (exhaustive) testing -
> such rigorous testing *requires* you to develop an exhaustive test
> suite that is capable of running periodically i.e. nightly etc to
> catch such errors.

I have to admit that I'm not yet using `exhaustive` testing. Still,
lack of static typing hasn't been a problem. I should add that I came
to Python with a strong belief in static typing and was quite a bit
worried about it in the beginning. It turned out to be a non-issue.

I even shed my long standing habit of using pre-/post-conditions and
invariants -- something I still consider indispensable for working in
Ada or C++.

> Whether the people who support Python for larger
> projects have the discipline within their project team to maintain
> such an effort is unclear - there has been no mention of it at all.

You seem to read only what is fitting into your preconceived notions.
XP and unit testing is brought up regularly in such discussions.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From QnickQm at alum.mit.edu  Fri Feb  1 01:22:45 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Fri, 01 Feb 2002 06:22:45 GMT
Subject: While loop error
References: <32c382ad.0201312208.166d5de3@posting.google.com>
Message-ID: <slrna5kd7c.bd0.QnickQm@localhost.localdomain>

In article <32c382ad.0201312208.166d5de3 at posting.google.com>, mlorfeld wrote:
> I continue to get the error:
> Traceback (most recent call last):
>   File "Columbus:Desktop Folder:gaus.py", line 19, in ?
>     n=list[i]
> IndexError: list index out of range
> 
> my code is:
> i=0
> j=0
> sum=0
> list=[1,2,3,4,5,6,7,8,9,10,50,100]
> list_length=len(list)
> #G(n)=sum of first n natural numbers
> while i < list_length:
> 	n=list[i]
> 	while j <= n:
> 		sum=sum+j
> 		j=j+1
> 	print sum
> 	j=0
> 	sum=0
> 	i=i+1
> 
> any ideas??

Sadly, no.  The code you list above works just fine for me.  Are you
sure you're running exactly the version you posted? It looks like the
version you're running has advanced i past the bounds of your list.

BTW, you could write the above code a bit more idiomatically like this:

list=[1,2,3,4,5,6,7,8,9,10,50,100] 
for n in list:
    sum = 0
    for j in xrange(n+1):
	sum = sum + j
    print sum

HTH,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From peter.milliken at gtech.com  Sun Feb  3 16:23:57 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 4 Feb 2002 08:23:57 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <Wls68.14178$om6.357057@news1.tin.it>
Message-ID: <a3ka03$tfn4@news1.gtech.com>

"Alex Martelli" <aleax at aleax.it> wrote in message
news:Wls68.14178$om6.357057 at news1.tin.it...
> Peter Milliken wrote:
>         ...
> > Actually, if you are *good* at what you do, then code review, whilst
still
> > beneficial, is not nearly *that* beneficial. Things like code review
(IMO
> > :-)) is there for the 25-50% of programmers who just plain shouldn't be
in
> > the industry - what they generate is just absolute garbage - code review
> > will hopefully stop the code from getting into Unit test and beyond the
> > point of redemption :-).
>
> I disagree.  Code review, or even better the "continuous ongoing code
> review" that's such a substantial part of the benefits of pair
programming,
> is in my experience just as beneficial with the best coders: without it,
> besides bugs, it's far too much a temptation to pepper the code with
> clever tricks, individual style quirks, unexamined (subconscious)
> assumptions about inputs and environmental conditions, and the like.
>

I doubt that you really "disagree" Alex with what I say, you certainly
disgree with what you *think* I say :-). Mine is not an absolute statement,
it is relative to a certain situation and is very heavily qualified (not the
use of "nearly" etc). I think we are both completely agreed that code
reviews, when done properly and in the right "spirit" will always be
beneificial to all parties involved. As for "pair" - I happen to have
experience that indicates that reviews should *always* be conducted by 3 or
more people and the results communicated via a "formal" sit-down meeting.
The "3 or more" because the more people involved the more that will be
picked up (not everybody thinks of everything) - this of course has to be
tempered by practicalities such as you can't have all 10 programmers on the
project reviewing the same code :-). To my mind (opinion here :-)), the
ideal is 2 to 3 reviewers, the author and possiby a "junior" who is
nominally a reviewer but is really there for the experience i.e they
generally do not have much useful input to the review i.e. no more than 5
people in the room - tops. I have worked at getting various groups I have
worked with doing reviews ever since I was first introduced to them back in
1985 - so I have about 17 years (on and off) with getting people to do them
(or trying to) and the various forms that people come up with when
attempting them :-).

The "formal" meeting part is necessary to achieve a group synergy which just
doesn't happen when passing around pieces of paper with other people
comments i.e. this is the idea of the sum of the parts are not greater than
the whole. I have reasonably extensive experience with a number of different
review techniques (the so called round robin, the just pass the comments
back to the author etc) and none of them work nearly as well as "the 3 or
more with a meeting". Of course, many people would scoff or denigrate the
meeting because they feel meetings are a waste of time - these are the
people who lack the self discipline to get in there, do the job and get out
i.e. they are the ones who go into a meeting and start it by asking how your
weekend was, discussing how the Packers went in the game on the weekend etc
:-). A simple course on meeting techniques generally fixes this problem.

> The code coming out of a good review, or a good-synergy pair, is
> thus of much higher quality and more valuable than the results of a lone
> coding session of even a very good coder.
>

No dispute :-).

Peter




From cliechti at gmx.net  Wed Feb 20 11:34:01 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 20 Feb 2002 17:34:01 +0100
Subject: RS232 Serial Port
References: <3c737e09$0$12226$4d4efb8e@news.be.uu.net>
Message-ID: <Xns91BBB34251F12cliechtigmxnet@62.2.16.82>

"DelPiccolo Ivano" <ivano.delpiccolo at degroof.be> wrote in 
news:3c737e09$0$12226$4d4efb8e at news.be.uu.net:
> How can I read/write on a RS232 port ?
> I'm working under windows 98.

try http://pyserial.sourceforge.net

-- 
Chris <cliechti at gmx.net>



From ajeru at vknn.org  Tue Feb 19 04:46:22 2002
From: ajeru at vknn.org (Alexander Jerusalem)
Date: 19 Feb 2002 01:46:22 -0800
Subject: PyChecker does it!
Message-ID: <24c39b2c.0202190146.62cf3df@posting.google.com>

I just downloaded PyChecker and I think it does what I want. It
actually sees thru my code to check if the methods I'm calling
actually exist and if I'm supplying the correct number of parameters.
Thanks a lot for that  tip

Regards,
Alexander Jerusalem


From cliechti at gmx.net  Fri Feb 15 19:55:12 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 16 Feb 2002 01:55:12 +0100
Subject: impossible to reboot OS in python?
References: <a4k9as$ak8$1@nntp.itservices.ubc.ca>
Message-ID: <Xns91B714137F926cliechtigmxnet@62.2.16.82>

"Kris Botha" <3dsphere at home.com> wrote in
news:a4k9as$ak8$1 at nntp.itservices.ubc.ca: 

> I heard it was impossible to induce a reboot in python as the os 
> module does not cater for that. Is this true?
...
> Does it depend on the os installed?

yes, you've got an answer for linux on win32 you can call

win32api.ExitWindows from the win32all extensions

-- 
Chris <cliechti at gmx.net>



From phd at phd.pp.ru  Thu Feb  7 08:59:43 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 7 Feb 2002 16:59:43 +0300
Subject: Autocoding evolves from........
In-Reply-To: <a3u0m8$il4$1@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Thu, Feb 07, 2002 at 01:50:00PM +0000
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl>
Message-ID: <20020207165943.R497@phd.pp.ru>

On Thu, Feb 07, 2002 at 01:50:00PM +0000, Martijn Faassen wrote:
> Chris Gonnerman <chris.gonnerman at newcenturycomputers.net> wrote:
> > Should Timothy Rue go away?
> > All in favor, post AYE.  All opposed, post NAY.
> 
> Please, I think it's far worse than Rue. I don't want this group to start
> turning into a group that 'votes' others off. Either ignore him or killfile
> him or just talk to the fellow. Having a vote is only a weird popularity 
> contest more about the people doing the voting than the person being voted
> on in my opinion, and we could do without.

   That was exactly why I voted NAY.

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



From mwh at python.net  Tue Feb 19 07:46:32 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 19 Feb 2002 12:46:32 GMT
Subject: Dynamic object import ??
References: <zkrc8.172$JWb.179470336@news.telia.no>
Message-ID: <uadu5d4bk.fsf@python.net>

"Thomas Weholt" <thomas at gatsoft.no> writes:

> I need a way to scan a given folder for modules, import the ones
> which fits a certain criteria ( ie. has a proper config-file present
> etc.), create instances and map this instance to the name of the
> module, all in a specific dictionary.

imp.load_module may be your friend.

Cheers,
M.

-- 
   This proposal, if accepted, will probably mean a heck of a lot of
   work for somebody.  But since I don't want it accepted, I don't
   care.                                   -- Laura Creighton, PEP 666


From richie at entrian.com  Wed Feb 13 08:06:14 2002
From: richie at entrian.com (Richie)
Date: Wed, 13 Feb 2002 13:06:14 +0000
Subject: cgi and mime type
References: <ed6d1c1a.0202121225.295ca950@posting.google.com>
Message-ID: <6ook6u0h35jl6dcroap3s90j6kensqhohq@4ax.com>

Hi,

> in NS it refuses to recognize the MIME type and prompts that
> you are attempting to download "text/plain" script.py

I've seen browsers use the filename part of the URL ('script.py' in your
case) to determine the MIME type.  One workaround is to code the name of
the download into the URL, so that instead of:

  <form action="/cgi-bin/script.py" method="POST">

you have:

  <form action="/cgi-bin/script.py/app.exe" method="POST">

The web server will run your script as normal (with '/app.exe' in the
PATH_INFO environment variable) and the browser will see 'app.exe' as the
filename part of the URL.

-- 
Richie Hindle
richie at entrian.com



From just at xs4all.nl  Thu Feb  7 06:12:46 2002
From: just at xs4all.nl (Just van Rossum)
Date: Thu, 07 Feb 2002 12:12:46 +0100
Subject: Python method reference kills garbage collection?
References: <a3tmuv$qm1$1@news.ost.eltele.no>
Message-ID: <3C62612D.F25C8FE1@xs4all.nl>

Troels Walsted Hansen wrote:

> Consider the following program. What is so special about references to the
> object's own methods that cause the object to never be garbage collected? Is
> this a cyclic reference problem, and if so why doesn't the cycle detector
> (--with-cycle-gc) cure the problem?
> 
> class a:
>     def __init__(self):
>         self.f = self.f1
>     def __del__(self):
>         print "this is never printed"
>     def f1(self):
>         pass
> b=a()
> b=a()

The cycle collector doesn't collect objects of classes with __del__ methods.
(See http://www.python.org/doc/current/lib/module-gc.html for a brief
axplanation.).

Just


From robin at alldunn.com  Fri Feb  8 16:13:50 2002
From: robin at alldunn.com (Robin Dunn)
Date: Fri, 8 Feb 2002 13:13:50 -0800
Subject: [wxPython] RE: wxpython-users digest, Vol 1 #620 - 13 msgs
References: <DHEEKAFNPOKNEBEBLCIAOEFLCHAA.me@mikerobin.com>
Message-ID: <13ba01c1b0e5$825236f0$0214a8c0@Rogue>

> ---------------------------------------------------------
> "Steve Holden" <sholden at holdenweb.com> wrote in message news:<T%u88.94053>
> > The thing I find most frustrating (and have just aded a FAQ question
about
> > on the Wiki) is the lack of a connection between my intuition and the
way
> > sizers work. I have licensed a copy of wxDesigner just to try and get
over
> > this hurdle, but dammit I *still* can't put interfaces together that
work
> > the way I expect them to under window resizing.
>
> I have the same problem. It doesn't help that wxDesigner does sizers
> but doesn't do anchors via layout constraints - and Boa does anchors
> but doesn't do sizers, so you can't really experiment/maintain at the
> designer level with one tool. (Is it true that anchors are going away?
> - I remember seeing in the wx pages that they were being deprecated.

wxLayoutConstraints (which anchors are built upon) are deprecated, but I
don't think they will ever go away entirely.

> They seem to be simple and give predictable results vis-a-vis heavily
> nested sizers.)

Anchors are a (currently python-only) simplification of the very complex
layout constraints.  Riaan did a good job of abstracting out of the
contraints a common (but not complete) usage model that could be expressed
in a simple way.  The wxSizers were created with similar purposes (to
provide a layout model simpler than the constraints that handled most cases)
but from scratch instead of building on the layout constraints.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com       Java give you jitters?
http://wxPython.org      Relax with wxPython!







From jjl at pobox.com  Sat Feb 16 16:47:05 2002
From: jjl at pobox.com (John J. Lee)
Date: Sat, 16 Feb 2002 21:47:05 +0000
Subject: win32com and MS Excel
In-Reply-To: <mailman.1013886806.19234.python-list@python.org>
References: <mailman.1013886806.19234.python-list@python.org>
Message-ID: <Pine.SOL.4.30.0202162145240.4432-100000@primrose.csv.warwick.ac.uk>

On Sat, 16 Feb 2002, Jimmy Retzlaff wrote:

> >from win32com.client import Dispatch
[...]
> >but with 50000 elements you'll have to wait some minutes for
> >Excel to show the data. And with 100000, I get a fatal error.
[advice to run makepy.py snipped]

You can also ensure this has been done by just changing Dispatch to
win32com.client.gencache.Dispatch (not entirely sure that's where gencache
is, but you get the idea).


John



From emile at fenx.com  Wed Feb 20 16:49:05 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 20 Feb 2002 13:49:05 -0800
Subject: HTMLifyer for Python code?
References: <mailman.1014238715.7462.python-list@python.org>
Message-ID: <a515vj$3q7qk$1@ID-11957.news.dfncis.de>

"Bjorn Pettersen"
> I just browsed through the Vaults, and did a quick Google search, but
I
> couldn't find a tool for converting Python source code to
> syntax-highlighted HTML. Surely someone must have written something
like
> this?


The view feature within viewcvs http://viewcvs.sourceforge.net/ does
this, but I don't know what it would take to extract that out.

--

Emile van Sebille
emile at fenx.com

---------



From martin_byrne2 at yahoo.co.uk  Thu Feb 21 12:17:26 2002
From: martin_byrne2 at yahoo.co.uk (B)
Date: 21 Feb 2002 09:17:26 -0800
Subject: Python ActiveX Scripting Engine error
Message-ID: <508cb8ae.0202210917.6d97e157@posting.google.com>

I'm getting the following error intermittently in my ASP scripts.  Does 
anybody know why it is happening and how to sort it out?  


Python ActiveX Scripting Engine error '80020009' 
Traceback (innermost last): KeyboardInterrupt: 

/Default.asp,line 0


From phr-n2002a at nightsong.com  Tue Feb 19 09:21:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 06:21:34 -0800
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi> <23891c90.0202190434.147ee6d1@posting.google.com> <7x7kp9zkuw.fsf@ruckus.brouhaha.com> <mailman.1014124533.24449.python-list@python.org>
Message-ID: <7xbselblht.fsf@ruckus.brouhaha.com>

Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> > I hadn't heard of iODBC before, but just found the home page www.iodbc.org
> > and it mentions that a Python bridge (whatever that is) already exists.
> 
> SQLRelay?
> 
> http://www.firstworks.com/sqlrelay.html

Wow, that looks like a pretty neat package.  The Python DB page should
link to it.  However, it doesn't use ODBC.  I guess I mostly care
about MySQL and PostgreSQL, which have their own non-ODBC interfaces
that are implemented by existing Python modules, so maybe ODBC isn't
that important for my purposes.  Thanks.


From bkelley at wi.mit.edu  Fri Feb 15 13:40:07 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 15 Feb 2002 13:40:07 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org> <3C6C3805.37BD87FB@ccvcorp.com>
Message-ID: <3C6D5607.3070900@wi.mit.edu>

Jeff Shannon wrote:

> 
> 
> This works, but is generally a Bad Idea(tm).  The problem
> with eval/exec on arbitrary strings (and any string you read
> in from a file is arbitrary) is that it's hard to be
> positive that what you're eval/exec-ing is what you expect.
> And if it's *not*, then many, many very bad things can
> happen.  (Imagine someone "accidentally" replacing that
> textfile with one that contains the line "import
> os;os.system('rm -s /')" -- suddenly your entire filesystem
> is blank....)
> 
> If you want to store the contents of a dictionary on a
> permanent disk file, you should look at the Pickle and
> Shelve modules.  If you're doing this a lot, or with lots of
> data, then try using one of the generic db modules (anydbm,
> bsddb, etc).


n.b. pickle and cPickle also use eval to convert representations of 
string values into python strings so security is not a valid reason to 
choose one over the other, yet...

Source forge bug entry:
http://sourceforge.net/tracker/?func=detail&atid=105470&aid=502503&group_id=5470



> 
> Jeff Shannon
> Technician/Programmer
> Credit International
> 
> 
> 

Brian Kelley



From tdelaney at avaya.com  Wed Feb  6 00:10:12 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Wed, 6 Feb 2002 16:10:12 +1100 
Subject: Autocoding project proposal.
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A16E@natasha.auslabs.avaya.com>

> From: dman [mailto:dsh8290 at rit.edu]
> 
> Figure it out already!!!!  He was giving advice on __communication__,

Isn't it strange how we accumulate habits? The normal convention for
underlining in plain text is a single underscore on each end ...

Tim Delaney



From ivano.delpiccolo at degroof.be  Tue Feb 12 08:07:30 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Tue, 12 Feb 2002 14:07:30 +0100
Subject: popen3 Object
Message-ID: <3c69138a$0$12226$4d4efb8e@news.be.uu.net>

How can I print the stdout of the popen3

exemple :

fileobjet = os.popen3("ls -la", "true" )
print ??????????

Thnaks




From DeepBleu at DeepBleu.org  Fri Feb  1 15:39:48 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Fri, 1 Feb 2002 14:39:48 -0600
Subject: telnetlib read_until()( question
References: <u5lqt76m0ogfd8@corp.supernews.com> <a3er56$17h5lp$1@ID-11957.news.dfncis.de> <u5lsfu2m34gpa9@corp.supernews.com> <a3esav$17p2n5$1@ID-11957.news.dfncis.de>
Message-ID: <u5lvb7fj4qjrcc@corp.supernews.com>

The telnet connection in this case is part of a CGI Python script.
A user that is outside the firewall sends us a file via a web browser.  Then
the script ftp's the file from the web server into another server, and then
telnets that server to do some processing on the file.  But the ftp and
telnet are done within the LAN.  So, I do not think the firewall got
anything to do with it.  hmmmmm..... I do not know.  Also, even if the user
outside the firewall closes the web browser window, it is not to my
knowledge that processing done through telneting in the script will stop.
Could it be that the script itself running on the web browser is being
dropped by the firewall while the script is waiting for processing to finish
through telnet?  Is this possible as you are suggesting?
The whole system is working fine with no glitches.  Except for this incident
today when the text file (that particular file was quite big:  143 MB)  was
ftp'd and the telnet worked since some processing was done (applying
dos2unix and then tr on it).  Then apparently it dropped out without doing
the next steps of processing.
I am hoping it was only a network glitch.  We just have to ask those people
to send the file again and see if it works.
DeepBleu

"Emile van Sebille" <emile at fenx.com> wrote in message
news:a3esav$17p2n5$1 at ID-11957.news.dfncis.de...
>
> "DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
> news:u5lsfu2m34gpa9 at corp.supernews.com...
> > Thanks Emile
> > I am trying to find out why a telnet connection (using a Python
> script)
> > dropped out.
>
>
> Would you by chance be trying to connect out to a system behind a
> firewall?  Perhaps a non-activity timer is dropping you while you wait
> for processing to complete on the remote host.
>
> --
>
> Emile van Sebille
> emile at fenx.com
>
> ---------
>




From phr-n2002a at nightsong.com  Mon Feb 11 17:40:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 14:40:17 -0800
Subject: Question on if, while, for....  Figured it out..
References: <e959e7b.0202110901.485da0b0@posting.google.com> <U2T98.152185$KM2.6837837@atlpnn01.usenetserver.com>
Message-ID: <7xk7tjprqm.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> > if num > den: #Test for improper fraction.
> >     i = num/den #This will only return the whole part.
> >     j = num%den #This will return only the remainder.
> >
> Rather than comment the usage of "i" and "j", why not just use "d" for
> dividend and "r" for remainder? The idea behind being able to choose
> variable names is to allow them to have some mnemonic significance.

I think Steve means quotient and remainder, not divident and remainder.

You might also like to error-check the input: what happens if den==0?


From dmaster at synopsys.com  Tue Feb 12 13:30:23 2002
From: dmaster at synopsys.com (David Masterson)
Date: 12 Feb 2002 10:30:23 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <uk7tkylre.fsf@synopsys.com> <a492j6$18om$1@nntp6.u.washington.edu> <slrna6gegr.vsl.philh@comuno.freeserve.co.uk>
Message-ID: <uk7tiy2m8.fsf@synopsys.com>

>>>>> phil hunt writes:

> On 11 Feb 2002 18:29:58 GMT, Donn Cave <donn at u.washington.edu> wrote:
>> Quoth David Masterson <dmaster at synopsys.com>:

>> | Or you haven't given us anything we can deal with.  Note, if we wanted
>> | to remain ignorant, we could simply just stop talking to you.  You
>> | have an audience and your "10 minutes of fame", but, if you cannot
>> | express your problem in understandable fashion, we will go on to other
>> | things that we do understand.

>> Who's "we" - you have a mouse in your pocket? 

> No, he's just pleased to see us.

All right, so I used the king's "we" a little too freely.  I had no
intention of going on with it and this was my way of saying "You've
lost me as an audience, Tim".

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From wolfson at uchicago.edu  Tue Feb 26 16:04:20 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Tue, 26 Feb 2002 15:04:20 -0600
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org> <m3pu2sancq.fsf@chvatal.cbbrowne.com> <eppstein-42B711.20584025022002@news.service.uci.edu> <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547@4ax.com> <a5fp9u$dkd@gap.cco.caltech.edu>
Message-ID: <pan.2002.02.26.15.04.17.605099.1294@uchicago.edu>

On Tue, 26 Feb 2002 04:50:38 -0600, Robert Kern wrote:

> In article <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547 at 4ax.com>,
> 	Courageous <jkraska at san.rr.com> writes:
>> 
>>>>  - Pulchritudinous Python?
>>>>  - Busty Python?
>> 
>> Clever Linguistic Integration Tricks without Stacks?
> 
> Wrong body part.
> 
> Tightly Integrated Tricks without Stacks.

Neatly Increased Power of Python, with Larger Expressivity.

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From fredrik at pythonware.com  Mon Feb 25 10:40:36 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 25 Feb 2002 15:40:36 GMT
Subject: Counting bytes
References: <Q7qe8.593$LI1.7401571@newsserver.ip.pt> <j4bsedlnai.fsf@informatik.hu-berlin.de> <s%qe8.598$Zj1.7273540@newsserver.ip.pt> <j4664lllc6.fsf@informatik.hu-berlin.de> <QPre8.602$qQ1.7486593@newsserver.ip.pt>
Message-ID: <UVse8.28260$l93.5416160@newsb.telia.net>

Hugo Martires wrote:

> I need to send a list in a socket
>
> server side
> ------------
> list = [0.5 0.5 0.5 0.5]
> client.socket.send(list)

(did you try running this code?)

> client side
> -------------
> client.socket.recv(1024)
>
> the problem is the buffer size (1024 bytes)
> if i need to send a "big" list (with more than 1024 bytes), i have to count
> the number of bytes to send in parts. So i must know how many bytes my list
> oucupies.

a list contain objects, not bytes.  how things are stored in
memory is implementation dependent; unlike in C, there's no
way to get a pointer to the internal data structures.

to convert python objects to byte streams (and back), use
the pickle module (or cPickle or marshal).

    import pickle

    list = [0.5, 0.5, 0.5, 0.5]
    data = pickle.dumps(list)

    # data is now a string of bytes; use len(data) to
    # get the number of bytes

    list = pickle.loads(data)

more info here:

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

also see cPickle and marshal.  to produce an array of floating
point values (same as C's float data[] or double data[]), use
the array (or struct) module.

</F>

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




From claird at starbase.neosoft.com  Thu Feb 28 10:56:41 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Feb 2002 09:56:41 -0600
Subject: Deitel and Deitel Book...
References: <3C7D8DCA.119D52EC@earthlink.net>
Message-ID: <335802D43CB2BB6A.BCBCBFF5C03DB1E3.FFEAA682B5A8B145@lp.airnews.net>

In article <3C7D8DCA.119D52EC at earthlink.net>,
Ron Stephens  <rdsteph at earthlink.net> wrote:
>Sorry but I can't wait to give a brief "preview review" of the new
>Deitel and Deitel book, "Python How to Program" (I also posted this at
>my website at http://www.awaretek.com/plf.html ).
>...
I like your enthusiasm and generosity.  Keep up the good work.

My own reaction to *Python How to Program* is more reserved
(and I was one of the tech reviewers of the book).  While it
does, indeed, have several strengths, it's not flawless.  In
particular, the last draft I saw gave Win* programmers an
inadequate sense of both Python's capabilities and good style.

I'll withhold more specific comment until I see the book as
it appears on store shelves.  I can add, though, that the
Deitels have always been thoroughly courteous and professional
with me, even through technical disagreements.
-- 

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


From phr-n2002a at nightsong.com  Wed Feb 20 19:25:19 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 16:25:19 -0800
Subject: HTTP state management without cookies?
References: <BFFEB31BC23CCD44B8924E139E1D87E41240D0@MAILBOX.nameconnector.com> <mailman.1014247607.13819.python-list@python.org>
Message-ID: <7xeljfhea8.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> > http://foo.bar/_SID_=2374820394/myscript.cgi/
> >
> > into:
> >
> > http://foo.bar/myscript.cgi?_SID_=2374820394
> >
> > so the same trick will work for CGI?
> >
> Geoff:
> 
> I appreciate the slightly unusual nature of thios scheme, but I don't
> see how it isn't included under the heading "pass session id through all
> URLs" as mentioned in my original post.

Why would you want to pass the session ID to off-site URL's?


From eliotm at pacbell.net  Wed Feb  6 15:46:20 2002
From: eliotm at pacbell.net (Eliot Miranda)
Date: Wed, 06 Feb 2002 20:46:20 GMT
Subject: when does python free memory?
References: <10b3f29a.0202050840.7641923a@posting.google.com>
Message-ID: <3C6197D0.2D3A7CC8@pacbell.net>

Robert,

	you'll need a more sophisticated garbage collection implementation to
actually return memory to the O/S.  The problem with the C allocator
malloc/free is that it simply maintains a pool that grows on demand but
never returns memory to the O/S.  The garbage collector instead needs to
obtain memory via memory mapping (e.g. VirtualAlloc on Windows, mmap on
unix) and arrange that regions so obtained are unmapped when they become
vacant.  Vacating these segments is a significant problem that typically
requires a compacting garbage collector as it is unlikely that just by
chance all objects within a memory-mapped segment would be unmapped at
the same time.  Any objects still used within a segment obviously
prevent it from being unmapped at all.

You might take a look at the Boehm garbage collector for C/C++ to see if
it addresses these issues.  Also interesting is the Harlequin GC
toolkit.  I'm speaking from experience with VisualWorks, the Smalltalk
environment, as I did the simple part of the work (the memory mapping)
while Barry Hayes did the hard part (the cross-segment compactor) in
allowing the VisualWorks VM to return memory to the O/S.

HTH

Robert Eanes wrote:
> 
> Hi-
> Reading the docs and searching the list archives, it seemed that
> python should release memory as soon as the last reference to an
> object is deleted.  However, I can't seem to prove this in practice.
> Look at this example:
> 
> Python 1.5.2 (#1, May 11 2000, 11:19:54)  [GCC 2.8.1] on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import os
> >>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
> >>> memsize = blah.readlines()
> >>> print memsize
> ['23240 2360\012']
> >>> z = ["LSDKFJLSKDFJSDF"]*10000000 #allocate a large list
> >>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
> >>> memsize = blah.readlines()
> >>> print memsize
> ['23240 41424\012']
> >>> del z #get rid of the large list
> >>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
> >>> memsize = blah.readlines()
> >>> print memsize
> ['23240 41424\012']
> 
> The first number that comes out of the "print memsize" statement is
> the pid, the second is the amount of memory used.  It doesn't seem to
> be actually freeing the memory that was allocated to variable z.  Now
> I played around with it a little more and if I reallocate another list
> it does re-use that memory:
> 
> >>> x = ["LSDKFJLSKDFJSDF"]*10000000
> >>> blah = os.popen("ps -ef -o pid -o vsz | grep %s" % os.getpid())
> >>> memsize = blah.readlines()
> >>> print memsize
> ['23240 41424\012']
> 
> So apparently python knows that the memory formerly allocated to z is
> available, but it doesn't tell the OS.  Is this the case, and if so,
> is there any way around it?  I have an application that needs to run
> 80-150 python processes that can stay around for 15min to an hour, so
> I'd like them not to hold onto the maximum amount of memory that they
> use throughout the whole life of the process.
> 
> Thanks for any help,
> -Robert

-- 
_______________,,,^..^,,,____________________________
Eliot Miranda              Smalltalk - Scene not herd


From stephan.schumacher at andra.fr  Thu Feb 28 03:20:56 2002
From: stephan.schumacher at andra.fr (Stephan)
Date: 28 Feb 2002 00:20:56 -0800
Subject: writing efficiently a numpy array
References: <MPG.16e755abd178b85e989682@news.wanadoo.fr> <a5jege$7r6k2$1@ID-11957.news.dfncis.de>
Message-ID: <fc905aec.0202280020.738f55db@posting.google.com>

> In what area are you looking to be more efficient?  Memory, time, or
> something else?  Do you need it in ascii form for another program, or
> are you saving it to be read back in?

in Time !!!

I need it in ascii file because this file will be the input file of
another program. The numpy arrays store fields of velocity, pressure,
saturation... The fields are computed on a mesh with about 300000
nodes. Then another program read the fields and compute the migration
of several species.


From threeseas at earthlink.net  Sun Feb  3 19:01:06 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:01:06 GMT
Subject: Rue the day...
References: <mailman.1012274265.29834.python-list@python.org> <3C564908.124C6148@engcorp.com> <xjodfKADjnV8Ew1a@jessikat.demon.co.uk> <1Kw58.28319$8b.1144526@atlpnn01.usenetserver.com>
Message-ID: <1467.799T2404T11775742threeseas@earthlink.net>

On 29-Jan-02 07:36:50 Steve Holden <sholden at holdenweb.com> wrote:
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>news:xjodfKADjnV8Ew1a at jessikat.demon.co.uk...
>> In article <3C564908.124C6148 at engcorp.com>, Peter Hansen
>> <peter at engcorp.com> writes
>> >Chris Gonnerman wrote:
>> >>
>> >> It doesn't matter whether your response is kind or cruel, agreeable or
>> >> argumentative, it's the *attention* he craves.  Just don't talk to him.
>> >> Not at all, even if he calls you terrible names.  If we ignore him long
>> >> enough, he'll eventually go away.
>> >
>> >"The 'Norwegian Blue' Ribbon Campaign to Shun Rue in c.l.p."
>> >
>> >Chris, would you be an honorary founding member?  :-)
>> >
>> >(Count me in.  I'm going cold turkey.  Please flame me violently if
>> >I have a relapse.)
>> now if only we could redirect him toward perl :)

>Well, he didn't seem to notice when I set followups to alt.stupid. So you'd
>only need to forge an address of your own to avoid becoming a part of the
>resulting conflagration :-)

Guess that's another way of saying you are dishonest.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From AdrianLeu at kelseus.com  Tue Feb 19 10:54:29 2002
From: AdrianLeu at kelseus.com (Adrian Leu)
Date: 19 Feb 2002 07:54:29 -0800
Subject: Python crash
Message-ID: <993ae4ca.0202190754.6ccb79d2@posting.google.com>

Hi !
I have written a script which I am calling from inside a C++
application using:

 module = PyImport_ImportModule( ... ) and 
 method = PyObject_CallMethod( ... ) 

methods. 

This script is called for a certain amount of times (depending on the
user). Initially I call Py_Initialize() and the last command
Py_Finalize() is called at the end (when the user does not want to
call the script anymore). It looks like it is working OK until after a
few repeats the system crashed with the following stack view (showing
that the error/crash is in the PyImport_ImportModule function and it
is related to the deallocation of the module pointer - I also get the
"Fatal Python error: UNREF invalid object" error in the output
window):

NTDLL! 77f9eea9()
_Py_ForgetReference(_object * 0x02da59dc) line 956 + 10 bytes
_Py_Dealloc(_object * 0x02da59dc) line 977 + 9 bytes
class_dealloc(PyClassObject * 0x02da42ec) line 114 + 62 bytes
_Py_Dealloc(_object * 0x02da42ec) line 978 + 7 bytes
tupledealloc(PyTupleObject * 0x02da7e3c) line 151 + 93 bytes
_Py_Dealloc(_object * 0x02da7e3c) line 978 + 7 bytes
class_dealloc(PyClassObject * 0x02da7ddc) line 114 + 62 bytes
_Py_Dealloc(_object * 0x02da7ddc) line 978 + 7 bytes
tupledealloc(PyTupleObject * 0x0371e4ac) line 151 + 93 bytes
_Py_Dealloc(_object * 0x0371e4ac) line 978 + 7 bytes
class_dealloc(PyClassObject * 0x0371e44c) line 114 + 62 bytes
_Py_Dealloc(_object * 0x0371e44c) line 978 + 7 bytes
instancemethod_dealloc(PyMethodObject * 0x0371e2ec) line 1634 + 62
bytes
_Py_Dealloc(_object * 0x0371e2ec) line 978 + 7 bytes
PyDict_Clear(_object * 0x0371e67c) line 539 + 108 bytes
dict_tp_clear(_object * 0x0371e67c) line 1156 + 9 bytes
delete_garbage(_gc_head * 0x0012cc94, _gc_head * 0x1e1dccf8
generation2) line 340 + 7 bytes
collect(_gc_head * 0x1e1dcce8 generation1, _gc_head * 0x1e1dccf8
generation2) line 416 + 13 bytes
collect_generations() line 476 + 15 bytes
_PyGC_Insert(_object * 0x056d592c) line 508
PyTuple_New(int 6) line 87 + 9 bytes
r_object(WFILE * 0x0012d770) line 498 + 9 bytes
r_object(WFILE * 0x0012d770) line 565 + 18 bytes
r_object(WFILE * 0x0012d770) line 502 + 9 bytes
r_object(WFILE * 0x0012d770) line 565 + 18 bytes
PyMarshal_ReadObjectFromFile(_iobuf * 0x00286908) line 620 + 9 bytes
read_compiled_module(char * 0x0012d7b8, _iobuf * 0x00286908) line 577
+ 9 bytes
load_source_module(char * 0x0012da70, char * 0x0012d914, _iobuf *
0x002868e8) line 740 + 19 bytes
load_module(char * 0x0012da70, _iobuf * 0x002868e8, char * 0x0012d914,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012da70, char * 0x0012da70) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012db84, char * 0x0012da70, int *
0x0012da6c) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x056aa62c, _object *
0x056aa62c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x056ab7fc, _object * 0x056aa62c,
_object * 0x056aa62c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1503
+ 21 bytes
builtin___import__(_object * 0x00000000, _object * 0x056ae7dc) line 31
+ 21 bytes
call_builtin(_object * 0x052f4180, _object * 0x056ae7dc, _object *
0x00000000) line 2650 + 11 bytes
PyEval_CallObjectWithKeywords(_object * 0x052f4180, _object *
0x056ae7dc, _object * 0x00000000) line 2618 + 17 bytes
eval_code2(PyCodeObject * 0x056ae200, _object * 0x056aa62c, _object *
0x056aa62c, _object * * 0x00000000, int 0, _object * * 0x00000000, int
0, _object * * 0x00000000, int 0, _object * 0x00000000) line 1680 + 15
bytes
PyEval_EvalCode(PyCodeObject * 0x056ae200, _object * 0x056aa62c,
_object * 0x056aa62c) line 324 + 31 bytes
PyImport_ExecCodeModuleEx(char * 0x0012e090, _object * 0x056ae200,
char * 0x0012ddd8) line 495 + 17 bytes
load_source_module(char * 0x0012e090, char * 0x0012ddd8, _iobuf *
0x002868c8) line 758 + 17 bytes
load_module(char * 0x0012e090, _iobuf * 0x002868c8, char * 0x0012df34,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012e090, char * 0x0012e090) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012e1a4, char * 0x0012e090, int *
0x0012e08c) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x056a445c, _object *
0x056a445c, _object * 0x052522ec) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x056a24dc, _object * 0x056a445c,
_object * 0x056a445c, _object * 0x052522ec) line 1503 + 21 bytes
builtin___import__(_object * 0x00000000, _object * 0x0525f6fc) line 31
+ 21 bytes
call_builtin(_object * 0x052f4180, _object * 0x0525f6fc, _object *
0x00000000) line 2650 + 11 bytes
PyEval_CallObjectWithKeywords(_object * 0x052f4180, _object *
0x0525f6fc, _object * 0x00000000) line 2618 + 17 bytes
eval_code2(PyCodeObject * 0x056a4500, _object * 0x056a445c, _object *
0x056a445c, _object * * 0x00000000, int 0, _object * * 0x00000000, int
0, _object * * 0x00000000, int 0, _object * 0x00000000) line 1680 + 15
bytes
PyEval_EvalCode(PyCodeObject * 0x056a4500, _object * 0x056a445c,
_object * 0x056a445c) line 324 + 31 bytes
PyImport_ExecCodeModuleEx(char * 0x0012e6b0, _object * 0x056a4500,
char * 0x0012e3f8) line 495 + 17 bytes
load_source_module(char * 0x0012e6b0, char * 0x0012e3f8, _iobuf *
0x002868a8) line 758 + 17 bytes
load_module(char * 0x0012e6b0, _iobuf * 0x002868a8, char * 0x0012e554,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012e6b0, char * 0x0012e6b0) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012e7c4, char * 0x0012e6b0, int *
0x0012e6ac) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x056a270c, _object *
0x056a270c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x01b5441c, _object * 0x056a270c,
_object * 0x056a270c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1503
+ 21 bytes
builtin___import__(_object * 0x00000000, _object * 0x0525b51c) line 31
+ 21 bytes
call_builtin(_object * 0x052f4180, _object * 0x0525b51c, _object *
0x00000000) line 2650 + 11 bytes
PyEval_CallObjectWithKeywords(_object * 0x052f4180, _object *
0x0525b51c, _object * 0x00000000) line 2618 + 17 bytes
eval_code2(PyCodeObject * 0x056a2080, _object * 0x056a270c, _object *
0x056a270c, _object * * 0x00000000, int 0, _object * * 0x00000000, int
0, _object * * 0x00000000, int 0, _object * 0x00000000) line 1680 + 15
bytes
PyEval_EvalCode(PyCodeObject * 0x056a2080, _object * 0x056a270c,
_object * 0x056a270c) line 324 + 31 bytes
PyImport_ExecCodeModuleEx(char * 0x0012ecd0, _object * 0x056a2080,
char * 0x0012ea18) line 495 + 17 bytes
load_source_module(char * 0x0012ecd0, char * 0x0012ea18, _iobuf *
0x00286888) line 758 + 17 bytes
load_module(char * 0x0012ecd0, _iobuf * 0x00286888, char * 0x0012eb74,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012ecd0, char * 0x0012ecd0) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012ede4, char * 0x0012ecd0, int *
0x0012eccc) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x056a110c, _object *
0x056a110c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x056a322c, _object * 0x056a110c,
_object * 0x056a110c, _object * 0x1e1e8d48 __Py_NoneStruct) line 1503
+ 21 bytes
builtin___import__(_object * 0x00000000, _object * 0x052489cc) line 31
+ 21 bytes
call_builtin(_object * 0x052f4180, _object * 0x052489cc, _object *
0x00000000) line 2650 + 11 bytes
PyEval_CallObjectWithKeywords(_object * 0x052f4180, _object *
0x052489cc, _object * 0x00000000) line 2618 + 17 bytes
eval_code2(PyCodeObject * 0x056a1160, _object * 0x056a110c, _object *
0x056a110c, _object * * 0x00000000, int 0, _object * * 0x00000000, int
0, _object * * 0x00000000, int 0, _object * 0x00000000) line 1680 + 15
bytes
PyEval_EvalCode(PyCodeObject * 0x056a1160, _object * 0x056a110c,
_object * 0x056a110c) line 324 + 31 bytes
PyImport_ExecCodeModuleEx(char * 0x0012f2f0, _object * 0x056a1160,
char * 0x0012f038) line 495 + 17 bytes
load_source_module(char * 0x0012f2f0, char * 0x0012f038, _iobuf *
0x00286868) line 758 + 17 bytes
load_module(char * 0x0012f2f0, _iobuf * 0x00286868, char * 0x0012f194,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012f2f0, char * 0x0012f2f0) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012f404, char * 0x0012f2f0, int *
0x0012f2ec) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x00000000, _object *
0x00000000, _object * 0x00000000) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x005b27e8 `string', _object *
0x00000000, _object * 0x00000000, _object * 0x00000000) line 1503 + 21
bytes
PyImport_ImportModule(char * 0x005b27e8 `string') line 1436 + 20 bytes

After each time I call the method and the module using the methods
above I make sure that they are decremented using Py_XDECREF( module)
and Py_XDECREF( method ).

Any ideas why it crashes would be warmly received.

Thanks.

Adrian.


From Ed.Smirnov at mail.ee  Sun Feb 17 10:03:05 2002
From: Ed.Smirnov at mail.ee (Eduard Smirnov)
Date: Sun, 17 Feb 2002 17:03:05 +0200
Subject: Python vs. C/C++/Java: quantitative data ?
Message-ID: <3c6fc507$0$159$bb624dac@diablo.uninet.ee>

I?m writing a diploma paper on the difference between scripting (e.g. in
Python or Tcl) and system programming (e.g. in C, C++ or Java), and one of
the claims I want to make is that you can often develop applications 10x or
more faster in a scripting language.
I apologue that more applications can be written or re-written in scripting
language faster and with more complex functionality then in system
programming language.
If you have any experience in writing applications in system programming
language (e.g. in C, C++) and re-writing it in scripting language (e.g.
Python) or vice versa, please compare your work in both implementations,
describe your work/application?s functionality, approximate number of lines
and time of development of each implementation, skill?s level needed.
Any more comments about are welcome.

More thanks for copy your answer to my e-mail.

Eduard.Smirnov at mail.ee
Tallinn Technical Univercity
Estonia





From phr-n2002a at nightsong.com  Sun Feb 17 19:09:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 17 Feb 2002 16:09:17 -0800
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com> <m37kpcuu31.fsf@mira.informatik.hu-berlin.de> <7x4rkgw79y.fsf@ruckus.brouhaha.com> <c76ff6fc.0202171321.4bd2e69c@posting.google.com>
Message-ID: <7xu1sfmz0y.fsf@ruckus.brouhaha.com>

sjmachin at lexicon.net (John Machin) writes:
> The array module is one of the included batteries that needs a
> publicity agent. It is good value for:

Thanks.  I knew about the array module but it wasn't at all clear from the
docs that it does what I need.  It's still not clear from the source code whether
it does everything I need, but it looks promising and I'll have to
examine it further.


> It's a pleasant surprise that the re module supports searching etc on
> 'c' arrays.

Yes, this was not documented at all.

> You'll find one nuisance when working with 'c' arrays; it doesn't
> support coercing an ordinary string to an array -- you have to do it
> yourself. For example:
> 
> x = array.array('c', 'abc')
> x = x + 'def'
> 
> doesn't work. You have to do 
> 
> x = x  + array.array('c', 'def')

That doesn't bother me much.

> >  So I guess I'll submit a doc bug saying the above items should be described.
> 
> Certainly the re docs should be a tad more explicit about what a
> "string" can be, or what behaviour(s) it must provide.

Next question (what I need to examine the source for):  what happens if you do

  x = array.array('c','a'*1000000)
  del x[500000]
  del x[500000]
  del x[500000]

Do a half million characters get copied around for each of the delete operations?


From msteed at altiris.com  Thu Feb  7 12:41:42 2002
From: msteed at altiris.com (Mike Steed)
Date: 07 Feb 2002 17:41:42 GMT
Subject: Maintenance of Python interface to gdchart
Message-ID: <a3ue8m$3cj@dispatch.concentric.net>

Hi,

   About two years ago I put together a Python interface to Bruce
Verderaime's gdchart library.  It was useful to me at the time, and I
released the original source and a few updates in the following months.
Judging by the amount of email I receive about it, others have found it
useful too.  I am no longer using gdchart, and unfortunately, I can no
longer afford the time to maintain it.

   Bruce has some interesting-looking plans for the gdchart library, which
will probably require some significant work on the Python side.  If you have
the time and the inclination to take over maintenance of the Python
interface, let me know and we can discuss it.

   Python interface:  http://athani.pair.com/msteed/software/gdchart/
   gdchart library:  http://www.fred.net/brv/chart/

Regards,
Mike Steed




From oivvio at polite.se  Tue Feb 19 05:39:37 2002
From: oivvio at polite.se (Oivvio Polite)
Date: 19 Feb 2002 11:39:37 +0100
Subject: Recursively traverse linked list -- help!
In-Reply-To: <7x7kp9g6pr.fsf@ruckus.brouhaha.com>
References: <3C71B585.90F644EB@kontact-no-spam-please.com>
	<mailman.1014106650.11778.python-list@python.org> 
	<7x7kp9g6pr.fsf@ruckus.brouhaha.com>
Message-ID: <1014115178.2891.30.camel@matijek>


> 
> That's a misconception.  One of the main reasons for using linked
> lists is to be able to insert stuff in the middle taking a constant
> amount of time.  If you do that with a built-in list, the time needed
> is proportional to the size of the list, which can be very large.
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 

Sure, granted that you keep track of the middle link of your list, so
that you don't have to traverse up to that point. But even if keep track
of it you'll have to be doing pretty large lists before the linked list
will outperform the builtin list.



-----------
program
import time
class LinkedList:

    def __init__(self):
        self.start = None

    def append(self, value = None):
        new = ListElement(value)
        if self.start:
            new.next = self.start
        self.start = new


    
    def insert(self, position, value):
        new = ListElement(value)
        
        if self.start:
            i = 0
            current = self.start
            while current and position > i:
                i += 1
                current = current.next
            new.next = current.next
            current.next = new
        else:
            self.start = new

        
    def pop(self):
        result = None
        if self.start:
            result = self.start.data
            self.start = self.start.next
        return result
    
    def __str__(self):
        result = ""
        current = self.start
        while current:
            result += "%s " % (str(current))
            current = current.next
        return result

class ListElement:
    def __init__(self, value = None):
        self.data = value
        self.next = None

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


def do_many_inserts(list, count1, count2):
    length = 0
    for j in range(count1):
        t1 = time.time()
        for i in range(count2):
            list.insert(length / 2, i)
            length += 1

        print time.time() - t1
    
def do_many_pushes(list, count1, count2):    
    for j in range(count1):
        t1 = time.time()
        for i in range(count2):
            list.append(i)
        print time.time() - t1

    
if __name__ == '__main__':


    print "inserts at start of LinkedList"
    do_many_pushes(LinkedList(), 10, 20000)
    print
    
    print "inserts at start of builtin list"
    do_many_pushes([], 10, 20000)
    print
    
    print "inserts in middle of LinkedList"
    do_many_inserts(LinkedList(), 10,200)
    print

    print "inserts in middle of builtin list"
    do_many_inserts([], 10,200)
    

-----------
output
python2.2 /tmp/linked.py
inserts at start of LinkedList
1.11887598038
1.11088299751
1.35213696957
1.5195029974
1.95533299446
1.17704892159
1.64089500904
1.10084593296
1.29115998745
1.82064199448

inserts at start of builtin list
0.14047896862
0.140606999397
0.145263075829
0.133972048759
0.206911921501
0.145354986191
0.132588028908
0.147331953049
0.141435980797
0.138702988625

inserts in middle of LinkedList
0.235688090324
0.61442399025
0.989964962006
1.43489098549
1.76089990139
2.27561497688
2.53811192513
3.29149603844
3.56963992119
3.98600196838

inserts in middle of builtin list
0.00191104412079
0.00234496593475
0.0030529499054
0.00247895717621
0.00264799594879
0.00436305999756
0.00301098823547
0.00318598747253
0.00469303131104
0.00358402729034
-----------

-- 
oivvio polite

cell +46 (0)709 30 40 30 / phone +46 (0)8 669 64 18 / fax +46 (0)8 84 00
18
varvsgatan 10A / s-117 29 stockholm / sweden




From carifio.nospam at nospam.usys.com  Tue Feb 26 10:00:04 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Tue, 26 Feb 2002 15:00:04 GMT
Subject: Calling baseclass method from a derivated class (others will probably give you a better solution)
References: <slrna7n2u6.f53.patois@ns.sicfa.org>
Message-ID: <UpNe8.8$26.2901@typhoon.maine.rr.com>

class Base:
    def meth(self):
        print "Base.meth"

class Derived (Base):
    def meth(self):
        print "Derived.meth"
    # Embeds knowledge of superclass in method implementation.
    def chain(self):
        Base.meth(self)


b = Base()
# Find b's meth, Base.meth.
b.meth()

d = Derived()
# Find d's meth, Derived.meth
d.meth()

# Find d's chain, which in turn calls Base.meth.
d.chain()

# Call Base.meth directly.
# Probably a more object-oriented way...
Base.meth(d)





From spencer at spencerdoidge.com  Wed Feb 27 22:45:21 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Wed, 27 Feb 2002 19:45:21 -0800
Subject: Windows GUI - which way?
Message-ID: <a5k8t9$hrp$1@news.efn.org>

I have a nice script running and doing useful work--in the Windows console
mode.
It sends RS232 commands and receives replies, punctuated by various time
delays.
I parked the script at http://www.spencerdoidge.com/tp1.py
Kudos and credits to Isaac Barona for his contribution, the SerialPort
class, which I modified slightly to suit my needs.
The script tabulates and reports the results and writes a log.

I'd like to make this palatable for others around the company to use, and
that means Windows GUI.
The user would select the script from a menu, or type the filename in a
dialog, or use File/Open, or...

(Forgive me, please if I miss the mark with my questions--I'm a bit fried
after burying myself in Python the last couple of weeks. I'm just coming up
for air...)
I'm looking for suggestions and insights on which of the following
approaches might be best:

So should I write up a program in my favorite style, Win32 API, and embed
Python in it?
Or should I spawn a child process and run my good old console version in the
child process? (Not sure how I would be able to interrupt the child process
though.)
Or should I write the whole thing in Python?

Would somebody like to comment on the pros and cons?

Spencer Doidge




From phr-n2002a at nightsong.com  Thu Feb  7 19:55:42 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 07 Feb 2002 16:55:42 -0800
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com>
Message-ID: <7xu1ssiygh.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> However, *if* PHP allows you to associate requests with session state, then
> *either*:
> 
> a) it does use cookies, you just don't know about them, or
> b) it puts session references in the URL.
> 
> There simply isn't any other available mechanism. So maybe what you're
> looking for is an interface that hides the cookie implementation from the
> programmer?

There are several other ways to track session:

- Hidden form fields populated by the server
- User data persistence (a messy and evil IE-specific cookie-like data store)
- SSL session identifier (beware, not guaranteed to be persistent)
- Put session tag in the URL of a surrounding or internal frame, and
  retrieve the tag with client side javascript (pass back to server via
  hidden form field).  That's a variant of the URL scheme but keeps the
  session tags out of the navigation bar.
- Maybe other stuff I haven't thought of
- Client side browser extensions can use various schemes of their own

The most important thing is not to send server-confidential information
like database passwords to the client under any circumstances, and do NOT
eval stuff that the client sends to the server (someone had suggested
a scheme like that).  Also, don't use the pickle module or the default
("smart") cookie objects from the cookie module to build cookie strings.


From Scirious at Scirious.com  Fri Feb 15 14:48:05 2002
From: Scirious at Scirious.com (Scirious)
Date: Fri, 15 Feb 2002 19:48:05 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <Xns91B6B5749771Scirious@140.99.99.130>

Ok! Once again thaks for your rsponse.
Scirious.


From drgergl at mindspring.com  Mon Feb 25 13:03:33 2002
From: drgergl at mindspring.com (Greg Landrum)
Date: Mon, 25 Feb 2002 10:03:33 -0800
Subject: Silly link question: python22_d.dll
References: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com> <mailman.1014593379.10486.python-list@python.org> <j74j7uomek4igopeoegk222pdm5u7v078p@4ax.com>
Message-ID: <a5dugi$eb4$1@slb4.atl.mindspring.net>

"Courageous" <jkraska at san.rr.com> wrote in message
news:j74j7uomek4igopeoegk222pdm5u7v078p at 4ax.com...
>
> There is a school  of thought that binaries and documentation are
> what make a system complete and not the source. While it's not
> that big a deal at all, one shouldn't have to checkout and compile
> all of Python in order to straightforwardly develop C Extensions.
>

You don't really need to do this.  The easiest way to solve the problem is
something like the following:

#ifdef _DEBUG
#undef _DEBUG
#include <Python.h>
#define _DEBUG
#else
#include <Python.h>
#endif

Then you'll never have the problem of Python looking for _d versions of
DLLs.  Of course you also won't be able to trace into python itself when
you're debugging, but I rarely find myself wanting to do this.  If you want
the option of doing full-blown debugging, you can define a third build
target and do something like the following (from Boost Python):

#ifdef _DEBUG
# ifndef BOOST_DEBUG_PYTHON
#  undef _DEBUG // Don't let Python force the debug library just because
we're debugging.
#  define DEBUG_UNDEFINED_FROM_WRAP_PYTHON_H
# endif
#endif

#include <Python.h>

#ifdef DEBUG_UNDEFINED_FROM_WRAP_PYTHON_H
# undef DEBUG_UNDEFINED_FROM_WRAP_PYTHON_H
# define _DEBUG
#endif

hope this helps,
-greg






From markabarclay at attbi.com  Wed Feb 27 03:20:37 2002
From: markabarclay at attbi.com (Mark Barclay)
Date: Wed, 27 Feb 2002 03:20:37 -0500
Subject: Python GUI
References: <uZ_e8.24022$dj3.835926@typhoon.austin.rr.com>
Message-ID: <a5i4tu019nh@enews3.newsguy.com>

I have tried Boa Constructor, which is pretty good. It generates code for
use with wxPython. I don't recommend this combination for large scale GUI
development, but using these you can get a simple GUI up and running pretty
quickly.

http://boa-constructor.sourceforge.net/
http://www.wxpython.org/

Mark

"Avi Homes" <Moo at houston.rr.com> wrote in message
news:uZ_e8.24022$dj3.835926 at typhoon.austin.rr.com...
> New to python (still.. I made a post last week with my ridiculous error
> about using () for indexing in a dictinoary instead of [], ah well, its
the
> easy mistakes that get us. anyway) I was curious if there were some type
of
> gui builder or somesuch for python, alla Borland Builder ? This is a
> delightful language,everything seems so easy for it. as a friend pointed
out
> "yeah ,but you can do that in c++ as well.." and i replied "yeah, but its
> not this easy! =)"
> thanks guys
>
> ~Avi
>
>




From aktay.sw at t-online.de  Wed Feb 27 11:11:38 2002
From: aktay.sw at t-online.de (Aktay SW)
Date: Wed, 27 Feb 2002 17:11:38 +0100
Subject: Equations with Regular Expressions
References: <mailman.1014805905.18241.python-list@python.org>
Message-ID: <a5j0gc$ouo$06$1@news.t-online.com>

Equations with Regular ExpressionsHi,

here is a piece of code that does the conversion:


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

import re                    # import the regular expressions module

s='2x^3-2x^2+15'    # set up the original string
s1=s                        # make a copy of the original string. it will be
modified for searching purposes
s2=''                        # create an empty resulting string

match=re.search('[0-9]x\^[0-9]+',s1)    # search the string s1 for the first
time for: one digit followed by 'x^' followed by one or more digits

while match:                            # loop as long as there is a match

    begin,end = match.span()    # get beginning and ending character
positions of matching portion

    if begin > 0:                       # if the matching portion does not
begin at the first character position you have to copy to the resulting
string
        s2=s2+s1[:begin]                # the characters from the beginning
of the string up to the first matching character


s2=s2+s1[begin:begin+1]+'*('+s1[begin+1:begin+2]+'**'+s1[begin+3:end]+')'
                                        # paste the characters '*(', '**',
')' into the matching portion and add it to the resulting string

    s1=s1[end:]                         # shorten string s1 by the
characters up the first character after the matching portion

    match=re.search('[0-9]x\^[0-9]*',s1)# do the search again and go through
the loop if there was a match, else you're done with the loop

s2=s2+s1            # add any leftover characters to the resulting string
print s             # print original string
print s2            # print conversion result

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

I hope this helps you

regards

Ruhi Aktay
software

"Alves, Carlos Alberto - Coelce" <calves at coelce.com.br> schrieb im
Newsbeitrag news:mailman.1014805905.18241.python-list at python.org...
Hi everybody,
Could someone of you give me and example of how can I translate
'2x^3-2x^2+15' into '2*(x**3)-2*(x**2)+15' using regular expressions.
Carlos Alberto
COELCE/DPRON-Departamento de Projetos e Obras Norte
Fone: 677- 2228
e-mail: calves at coelce.com.br
\|||/
(o o)
--ooo0-(_)-0ooo--




From lac at strakt.com  Tue Feb 12 13:42:22 2002
From: lac at strakt.com (Laura Creighton)
Date: Tue, 12 Feb 2002 19:42:22 +0100
Subject: Tk button help 
In-Reply-To: Message from park997@aol.comnotospam (Park997) 
   of "12 Feb 2002 17:23:38 GMT." <20020212122338.02073.00001446@mb-ch.aol.com> 
References: <20020212122338.02073.00001446@mb-ch.aol.com> 
Message-ID: <200202121842.g1CIgMGo029441@ratthing-b246.strakt.com>

You can't pass arguments to bindings like:
	Button(top, text='Enter Leaf N', command = test.run(G)).pack(side=RIGHT, fill = BOTH)

you just want the NAME of the function.  This works great when there are no
arguments.  If there are, get out the lambda.

This works:
from __future__ import nested_scopes #I am still using 2.1
from Tkinter import *

def run(arg):

    from tkSimpleDialog import askfloat
    N = askfloat('N', 'Leaf N (50 - 250 mmol m-2) ')
    print N, arg



###########################
if __name__ == '__main__':

    root = Tk()
    quit = Button(root, text="QUIT", fg="red", command=root.quit)
    quit.pack(side='left')

    B = Button(root, text='Enter Leaf N', command=lambda : run('hi there'))
    B.pack(side=RIGHT, fill= BOTH)
    root.mainloop()

So you might want command=lambda s=self, arg=4.0 : s.run(arg) or some such.

Laura Creighton



From Bill.Scherer at VerizonWireless.com  Tue Feb 26 16:07:02 2002
From: Bill.Scherer at VerizonWireless.com (Scherer, Bill)
Date: Tue, 26 Feb 2002 16:07:02 -0500 (EST)
Subject: Building 2.2 with the Intel icc compiler
In-Reply-To: <9gSe8.18435$ZC3.1526281@newsread2.prod.itd.earthlink.net>
Message-ID: <Pine.LNX.4.33.0202261557150.4697-100000@bagend>

I was able to build Python2.2 with the intel compiler. I had 
discussed this with an IBM compiler engineer at LinuxWorld 
recently. He had doubts that it would afford any significant 
speedup to running Python code.

I did some simple timing tests on a few of the test suite tests 
and they confirmed said engineer's expectations. I don't have any 
specifics; my initial tests showed no reason to get excited.

That said, I'm no whiz with a c compiler, and YMMV...

On Tue, 26 Feb 2002, Ted Drain wrote:

> Has anyone built Python 2.2 with the new Intel C/C++ compiler icc?  I
> configured Python and then edited the Makefile by hand to use icc and icc
> options.  Python builds but the modules all report errors (and then remove
> themselves).
> 
> This compiler generates code that is roughly 30% faster than gcc.  Check
> out:
> http://www.open-mag.com/754088105111.htm
> 
> Any thoughts?
> Ted
> 
> 
> 

-- 
Bill.Scherer at Verizon Wireless
RHCE 807101044903581




From jon+usenet at unequivocal.co.uk  Sun Feb  3 15:21:28 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Sun, 3 Feb 2002 20:21:28 +0000 (UTC)
Subject: Python 2.2 Modules/Setup lies?
Message-ID: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net>

I am installing Python 2.2 from source on an OpenBSD machine and a
Solaris machine.

The build instructions indicate that you should:

  ./configure
  vi Modules/Setup
  make

When I come to edit Modules/Setup, I am somewhat confused, because
everything in it is commented out, including standard built-in modules
that are explicitly commented "on by default" but which appear to be
off by default - e.g. the following section of the file:

  # Modules that should always be present (non UNIX dependent):

  #array arraymodule.c    # array objects
  #cmath cmathmodule.c # -lm # complex math library functions
  #math mathmodule.c # -lm # math library functions, e.g. sin()
  #struct structmodule.c  # binary structure packing/unpacking

Why are these (and everything else in the file) commented out?


From gcordova at hebmex.com  Wed Feb 20 10:46:48 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 09:46:48 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC0@SRM201002>

> 
> Simon Brunning writes:
>  > for line in file:
>  >     line.whatever()
> 
> Cool.  Good reason to upgrade to 2.2.
> 
>  > >     matched = re.match(r'egexp(.*)', line)
>  > >     if matched:
>  > >         print matched.group(1)
>  >  
>  > Assignments and expressions are different things in Python...
> 
> Indeed, and that's a good thing.  But I'm proposing an alternate
> syntax for the if statement, specifically:
> 	"if" <variable> "=" <expression":.
> 

But you impact legibility:

  mo = findSomething.search(someString)
  while mo:
	....
	....
	mo = findSomething.search(someString)


is much more understandable than:


  while mo = findSomething.search(someString):
	....
	....
	mo = findSomething.search(someString)


because *that* can be confused with *this*


  while flag == object.method(something):
	....
	....


and confusion ensues.


no?

I'd sincerely prefere the more readable alternative.

-gustavo



From a-no-space-here-weberr at verizon-dot.net  Wed Feb 13 08:51:12 2002
From: a-no-space-here-weberr at verizon-dot.net (AJ)
Date: Wed, 13 Feb 2002 13:51:12 GMT
Subject: SSL problem with Python (NEWBIE ALERT)
Message-ID: <kbua8.895$t83.126513@paloalto-snr1.gtei.net>

Can anyone tell me why this is suddenly occuring?  It seemed to work in
previous runs.  It's the ipcheck.py script on:
http://ipcheck.sourceforge.net/.

I'm told it's probably a corrupt Python library.  Does anyone know how to
troubleshoot this further, and/or how/which library I should re-install (and
where to get that lib)?  If I run the script with the flag to use non-ssl
connections it works fine...but who wants to be sending username/passwords
over the net in the clear???

Thanks in advance for any assistance!
--AJ

Traceback (most recent call last):
  File "ipcheck.py", line 4083, in ?
    _main(sys.argv)
  File "ipcheck.py", line 3663, in _main
    h2.putrequest("GET", updateprefix + hostlist + updatesuffix)
  File "/usr/local/lib/python2.1/httplib.py", line 437, in putrequest
    self.send(str)
  File "/usr/local/lib/python2.1/httplib.py", line 379, in send
    self.connect()
  File "/usr/local/lib/python2.1/httplib.py", line 639, in connect
    ssl = socket.ssl(realsock, self.key_file, self.cert_file)
socket.sslerror: SSL_connect error





From gh_pythonlist at gmx.de  Mon Feb 18 16:12:28 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 18 Feb 2002 22:12:28 +0100
Subject: Installing Python
In-Reply-To: <b6dc8.5870$pC1.1234514@newssrv26.news.prodigy.com>
References: <b6dc8.5870$pC1.1234514@newssrv26.news.prodigy.com>
Message-ID: <20020218211227.GA9549@lilith.hqd-internal>

Le 18/02/02 ? 20:03, Empty_One ?crivit:
> I just downloaded the python tarball, and installed it on my slack system 
> using sheck install.  All works good, except I cannot start IDLE.

I'm sure there was an error message when you tried to start IDLE. I've
lost my crystal ball, so these would be incredibly useful.

Chances are that Tkinter wasn't built. You'll need to have Tk installed
for this, including the header files.

> How can I configure the install script to also install IDLE, since it
> is included with the base python, and not a seperate download.

See above, you can also change the paths in Modules/Setup, should that
be necessary. It usually isn't necessary, unless the Tk header files and
libraries are in some very strange places.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.3 ?C      Wind: 4.1 m/s



From hmartires at cultalg.pt  Mon Feb 25 07:30:30 2002
From: hmartires at cultalg.pt (Hugo Martires)
Date: Mon, 25 Feb 2002 12:30:30 -0000
Subject: Threads in python?
References: <1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5@4ax.com>
Message-ID: <P5qe8.592$6L1.7417907@newsserver.ip.pt>

Yes, se the thread and threading module in:

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

Hope it will help


"Fernando Rodr?guez" <frr at ThouShallNotSpam.EasyJob.NET> wrote in message
news:1o9k7ucpql8tqkvha1h2jjl9urbf7cioa5 at 4ax.com...
>
> I'm going to start coding a little utility that needs to retrieve webpages
and
> process them.
>
> However, I worried that the whole app might stall if one server takes too
long
> to respond or is offline.
>
> Does Python support threads? "Learning Python" doesn't mention it...
>
>
>
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> | Fernando Rodr?guez   frr at EasyJob.NET
> | http://www.EasyJob.NET/
> | Expert resume and cover letter creation system.
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~




From maxm at mxm.dk  Sun Feb  3 05:42:50 2002
From: maxm at mxm.dk (Max M)
Date: Sun, 03 Feb 2002 10:42:50 +0000
Subject: PIL Q: Python interpreter cannot find _imaging.pyd
References: <21298535.0202011002.4674269@posting.google.com>
Message-ID: <3C5D142A.2060008@mxm.dk>

Svein Brekke wrote:

> Environment:
>   PATH = c:/usr/python;c:/usr/python/DLLs;c:/usr/python/PIL
>   PYTHONPATH = c:/usr/python;c:/usr/python/PIL/PIL


Try moving your two .pyd files to c:/usr/python/DLLs

regards Max M



From logiplexsoftware at earthlink.net  Tue Feb 26 12:28:53 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Tue, 26 Feb 2002 09:28:53 -0800
Subject: .NET only solves problems I don't have! (Was: advocacy help ( Python vs VB ))
In-Reply-To: <3C7BB093.3020600@mxm.dk>
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01>
	<mailman.1014706963.7898.python-list@python.org>
	<3C7BB093.3020600@mxm.dk>
Message-ID: <20020226092853.0fa93973.logiplexsoftware@earthlink.net>

On Tue, 26 Feb 2002 15:58:11 +0000
Max M wrote:

> What I find most amusing about com and .NET is that they are trying to 
> solve a problem I only had when programming using MS tools.

[snip]

Well said.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From bokr at oz.net  Sun Feb 17 19:43:20 2002
From: bokr at oz.net (Bengt Richter)
Date: 18 Feb 2002 00:43:20 GMT
Subject: Problems with telnetlib
References: <mailman.1013985929.26925.python-list@python.org>
Message-ID: <a4pin8$tc$0@216.39.172.122>

On Sun, 17 Feb 2002 17:26:51 -0500, Tim Peters <tim.one at comcast.net> wrote:

>[Bengt Richter]
>> ...
>> BTW, why are context diffs preferred over unified?
>
>Because Guido says so.  Really, that's all there is to it.  In general, he
>doesn't care so much about what the code used to do as about what it does
>after the change, and especially for complex patches the "after" logic is
>simply easier to follow if it isn't cluttered via interleaving with the
>"before" logic.  OTOH, the opposite seems true more often to me when looking
>at simple patches.  Someone should write a tool to generate both patch
>flavors, compare them, and objectively determine which is clearer.  Indeed,
>I demand someone do this at once <wink>.
>
Seems like the cvsviewer could be expanded to view effects of proposed patches
tentatively in various modes? Even regurgitating the diff in the alternate format
by re-diffing with a patched temp file. Seems like all the pieces are there, except
something new in a config file to define access to proposed patch files and temp caching,
and some glue. Of course, that's easy for me to say, not having seen any of that code ;-)

Otherwise maybe a standalong unified diff GUI viewer with buttons to see
old, new, or both?

Regards,
Bengt Richter



From frederic.giacometti at arakne.com  Tue Feb 26 15:23:23 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Tue, 26 Feb 2002 20:23:23 GMT
Subject: [ANN] JPE (Java-Python Extension) V2.0
Message-ID: <mailman.1014902104.8035.clpa-moderators@python.org>

This is the official announcement of the second version of JPE.

JPE (The Java-Python Extension) is a production-grade seamless integration
of Python and Java together.
JPE lets you call Java objects and classes from Python, and vice-versa,
within the same process space. JPE is reentrant (cross-language callbacks),
and supports multi-threading concurently in Java and Python.

The reference web site for JPE is http://jpe.sourceforge.net


As compared to its first release:

1) The build system of JPE has been completely reengineered, and is based on
a new build library (libplus).
In this new version:

  - Java and python are auto-detected from the command line, and everything
else is derived from there
(Java libraries, library paths...).

  - libplus generated its own makefiles, and runtime environment (env
variables) for the build and test phases; so there is no need to set-up
anything

  - everything is now plateform-independant, and the same scripts are used
on Win32 and posix (no more GNU make / cygwin make required on windows)

  - JPE has been ported to posix environments. However, a shared python
library is required on these platforms. You can build your onw, or use the
branched version of Python 2.1, named 'pythonx', which is provided for these
purposes.

  - JPE is distributed in source form from CVS. Substantial efforts have
been invested to provide a config and build environment that makes it
straightforward to configure, build from source, and test multi-component,
multi-platform software systems. Hopefully, that can be one step ahead from
the current autoconf or distutil systems.
Easily extensible to support arbitrary software components and tools,
libplus currently supports C, C++, Python, Java, and swig wrappers.
Libplus generates configuration-specific makefiles that chain software
component dependencies together, and provides runtime software/tool
configuration information.
Litteratly, this is an 'open-source' project :))


2) New capabilities of JPE, for the 2d release:

- Java bean support: Jbean properties are accessed as regular python
attributes

- Cross-language native array access: JPE integrates the bufarray python
extension type; a layer on top of the python buffer interface that provides
type-safe access to contiguous fixed-length C arrays (actually, 99% of the
situations in scientific C or Fortran libraries).
This integration permits to safely access arrays by shared memory reference
between Java, Python, and C; while hidding the native memory format.


3) Acknowledgements:

Arakn?, and The Molecular Graphics Laboratory at The Scripps Research
Institute.


Benevolently yours,

Fr?d?ric Giacometti
fred at arakne.com





From ws at mystrobl.de  Fri Feb  8 03:28:29 2002
From: ws at mystrobl.de (Wolfgang Strobl)
Date: Fri, 08 Feb 2002 09:28:29 +0100
Subject: Help! Using Boa Constructor
References: <8f8ffe67.0202071344.6673ed5d@posting.google.com>
Message-ID: <3k276u87qkibb6p773sg318chvh0mupusj@4ax.com>

On 7 Feb 2002 13:44:02 -0800, sjoshi at ingr.com (Sunit Joshi) wrote :

>Hello
>I'm trying to run boa constructor and I keep getting this error
>message when the application loads up
>03:45:01 PM: Traceback (most recent call last):
>03:45:01 PM:   File "C:\Python21\wxPython\tools\boa\Editor.py", line
>1168, in poll_for_open
>03:45:01 PM:     except Queue.Empty:
>03:45:01 PM:     AttributeError: 'Queue' module has no attribute
>'Empty'
>03:45:01 PM:     AttributeError: 'Queue' module has no attribute
>'Empty'Traceback (most recent call last):
>03:45:01 PM:     AttributeError: 'Queue' module has no attribute
>'Empty'  File "C:\Python21\wxPython\tools\boa\Editor.py", line 1168,
>in poll_for_open
>
>Any ideas what might be wrong here. I'm using Python 2.1, Wxpython
>2.3.1

The current version of Boa (0.1.0 - alpha, from the downloads section
on sourceforge, the cvs versions obviously _isn't_ up-to-date,
currently) says "wxPython 2.3.2 required". 

I successfully run it with wxPython 2.3.2.1 + Python21 (works well) and
wxPython 2.3.2.1 + Python22 (works, but it throws a GPF in python
during shutdown). 

I'd suggest upgrading to wxPython 2.3.2.1

Boa is definitely worth looking at.

-- 
Thank you for observing all safety precautions


From msoulier at mcss.mcmaster.ca_.nospam  Sat Feb  2 22:14:53 2002
From: msoulier at mcss.mcmaster.ca_.nospam (Michael P. Soulier)
Date: Sun, 03 Feb 2002 03:14:53 GMT
Subject: print statement in lambda expression
Message-ID: <NW178.16270$bIE.6837@news1.bloor.is>

    Hello people. 

    I just recently found that I couldn't put a print statement in a lambda
expression as a binding in Tk. Could someone explain why? 

    ie.
    
    b1 = Button(frame, text="Message", command=lambda: print "message")

    If I replace print with sys.stdout.write, it works fine. I don't follow
why. 

    Thanks,

    Mike

-- 
Michael P. Soulier <msoulier at mcss.mcmaster.ca>, GnuPG pub key: 5BC8BE08
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to Unix


From fbartlet-fixit at optonline.net  Mon Feb 11 09:20:27 2002
From: fbartlet-fixit at optonline.net (Frederick H. Bartlett)
Date: Mon, 11 Feb 2002 14:20:27 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <3C67C369.597E3517@optonline.net> <mailman.1013434940.27537.python-list@python.org>
Message-ID: <3C67D328.985269AC@optonline.net>

Whoops! He's ... Herr Professor Doktor Gerhard Wuensch.


Chris Gonnerman wrote:
> 
> ----- Original Message -----
> From: "Frederick H. Bartlett" <fbartlet-fixit at optonline.net>
> 
> > Years ago, my father -- a professor emeritus of music at the
> > University of Western Ontario -- coined the acronym ETD: Ego-Talent
> > Discrepancy.
> >
> > He never published, so he can't get official credit for the
> > observation; but I try to do what I can to get the coinage into
> > common currency.
> 
> What's his name, so we'll know who to credit for this gem?


From pobrien at orbtech.com  Tue Feb 19 23:17:02 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Wed, 20 Feb 2002 04:17:02 GMT
Subject: vars(os) starts help in ActivePython!!!
References: <258fd9b8.0202191142.71314c03@posting.google.com> <x7Cc8.2126$RJ4.246925423@newssvr14.news.prodigy.com>
Message-ID: <2rFc8.2002$ZC3.167000@newsread2.prod.itd.earthlink.net>

In what version of PyCrust can you reproduce this behavior? I don't get the
error in any version I have on my machine, which happens to be a Win98
machine.

--
Patrick K. O'Brien
Orbtech


"Keith Farmer" <kfarmer at thuban.org> wrote in message
news:x7Cc8.2126$RJ4.246925423 at newssvr14.news.prodigy.com...
> The same behavior was seen with wxPython's PyCrust shell, so I suspect
it's
> not (just) activepython, but rather the way the interpreter module's
> written.
>
> ----------
> Keith J. Farmer
> kfarmer at thuban.org
> http://www.thuban.org
>
>
>




From skywalkerpackage at hotmail.com  Sun Feb  3 15:07:10 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Sun, 03 Feb 2002 20:07:10 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de>
Message-ID: <3c5d9698.182096361@news.online.de>

Hello Laura and Eric,

thanks very much for your answer, it has helped me a lot :)

Laura, I don't think,

widget=widgetType(master, **mydict)

is necessary. I have tried

widget=widgetType(master, mydict)

with mydict={'text':'hello', 'relief':'sunken'}. It worked without the
two stars.
Aren't the two stars just necessary for function calls with variable
length key parameters?


ByeBye

Johannes






From gh_pythonlist at gmx.de  Tue Feb  5 04:49:34 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 5 Feb 2002 10:49:34 +0100
Subject: Size of integers (2**32-1)
In-Reply-To: <3C5FA2F4.8020905@thomas-guettler.de>
References: <3C5FA2F4.8020905@thomas-guettler.de>
Message-ID: <20020205094933.GA7571@lilith.hqd-internal>

Le 05/02/02 ? 10:16, Thomas Guettler ?crivit:
> I have two python version installed:
> with the cygwin version "print 2**32-1" works
> but with the zope version I get OverFlowError.

Zope uses Python 2.1, while your Cygwin Python is 2.2, right?

Python 2.2 does automagically change the type to LongType when the
integers get too big. Python 2.1 doesn't do this, yet.

The reason it fails in Python 2.1 is that 2**32 is not in the range of
the IntType (on your 32 bit platform), IOW it's greater than sys.maxint
(sys.maxint == 2**31-1) Note that in Python 2.1, you can't even say
"print 2**31-1" because it would first need to represent 2**31 as an
IntType, which it cannot.

To represent the number you want in Python < 2.2, you need to say
2L**32-1.

> Background: I want to get a random 32-Bit integer like this
> randrange(0, 2*32-1)

Btw. it looks like random.randint/randrange really want ints and won't
accept longs. A workaround could be to use a construction like:

    print random.randrange(0, sys.maxint) * 2L

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 11.2 ?C      Wind: 4.6 m/s



From anton at vredegoor.doge.nl  Thu Feb  7 18:00:36 2002
From: anton at vredegoor.doge.nl (Anton Vredegoor)
Date: Fri, 08 Feb 2002 00:00:36 +0100
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <mailman.1013062513.21646.python-list@python.org> <a3u0m8$il4$1@newshost.accu.uu.nl> <mailman.1013090595.10717.python-list@python.org> <a3uukp$5dq$1@news.hccnet.nl>
Message-ID: <a3v0vh$7n7$1@news.hccnet.nl>

On Thu, 07 Feb 2002 23:17:31 +0100, anton at vredegoor.doge.nl (Anton
Vredegoor) wrote:
>
>Now what is it? Its a permutation engine for multisets. Read more
>about it at:
>
>http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.py
>
>or (html version):

Hope I'll get it right this time:

>
>http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.html



From wurmy at earthlink.net  Sat Feb 16 18:02:27 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Sat, 16 Feb 2002 23:02:27 GMT
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <86n0yaw8u4.fsf@apo.hmotakef.int> <3C6E95FA.F95C914B@earthlink.net> <m3bsepjeyn.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3C6EE52D.7ED017B8@earthlink.net>

"Martin v. Loewis" wrote:
> 
> Hans Nowak <wurmy at earthlink.net> writes:
> 
> > It was part of a previous problem; the XML supposedly only contained \n,
> > where \r\n was desirable. While investigating it, I stumbled upon a
> > different, more serious problem-- no newlines at all.
> 
> This is all intentional. In content, the XML processor *must*
> normalize newline characters, replacing different variants of newline
> with #xA. In *attribute* only, newline is further normalized to space
> characters.  So if you have content where newline characters are
> relevant, don't put that into attributes. If you have data which are
> that sensible to their binary representation, you may consider not
> using XML at all, or chosing a binary-safe representation, such as
> base64-encoded strings.

I don't have much choice, I'm afraid... since I don't have an SQL
Server 2000 client, SQLXML is the only way I can talk to this
database. On top of that, I'm dependent on the XML returned by 
the server. If you have a table X with fields A, B and C, you'll
get the contents of a record like this:

  <X A="foo" B="bar" C="baz" />

and since this string-with-newlines is a value of a field in a
table, it is returned as an attribute, whether I want that or
not. :-(

I guess I'll use regular expressions to extract the data. That
does work, although it's not pretty.

Thanks for the clarifications,

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From SBrunning at trisystems.co.uk  Wed Feb 13 09:40:59 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 13 Feb 2002 14:40:59 -0000
Subject: Python Gotcha with Octal Numbers
Message-ID: <31575A892FF6D1118F5800600846864DCBCC3C@intrepid>

> From:	Christopher.Saunter at durham.ac.uk
> [SMTP:Christopher.Saunter at durham.ac.uk]
> It all stems to this, I was assigning to numbers to variables:
> 
> var1 = 1000
> var2 = 0100
> etc.
 
See <http://mindprod.com/unmain.html> - Coding Obfuscation, point 11. (About
halfway down.)

Nice to know that you can write unmaintainable code in Python, too.

Probably to late to do anything about it now - think of all the code which
would break!

Cheers,
Simon Brunning.
sbrunning at fuckmicrosoft.com




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From gherman at europemail.com  Mon Feb 25 10:34:24 2002
From: gherman at europemail.com (Dinu Gherman)
Date: Mon, 25 Feb 2002 16:34:24 +0100
Subject: Jython bugs or features?
Message-ID: <gherman-035CF3.16342425022002@news.t-online.com>

Hi,

I'm trying to get a little more familiar with Jython, but after having
just installed the 2.1 final release on OS X I find the following
rather surprising differences between Jython and CPython (2.1 and 2.2
respectively, but that doesn't matter here):

  1. cStringIO.StringIO.reset() is missing in Jython
  2. list() doesn't perform as expected in Jython

See sample sessions below... 

Are these bugs and if so: how many people are actually using Jython,
then, especially as books start to come out for it from O'Reilly and 
New Riders...?

Best regards,

Dinu





###
### list() in CPython
###

[localhost:~] dinu% python
Python 2.2 (#1, Feb 13 2002, 11:18:45) 
[GCC 2.95.2 19991024 (release)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> 
>>> L = [1, 2, 3]
>>> L2 = list(L)
>>> L2
[1, 2, 3]
>>> L2.insert(0, 4)
>>> L2
[4, 1, 2, 3]
>>> L
[1, 2, 3]    ### as expected, same on CPython 2.1


###
### list() in Jython
###

[localhost:~] dinu% jython 
Jython 2.1 on java1.3.1 (JIT: null)
Type "copyright", "credits" or "license" for more information.
>>> 
>>> L = [1, 2, 3]    
>>> L2 = list(L)
>>> L2
[1, 2, 3]
>>> L2.insert(0, 4)
>>> L2
[4, 1, 2, 3]
>>> L
[4, 1, 2, 3]    ### unexpected
>>> 
>>> 
>>> import copy 
>>> L = [1, 2, 3]
>>> L3 = copy.copy(L) # instead of list() (use only for this snippet)
>>> L3.insert(0, 4)
>>> L3
[4, 1, 2, 3]
>>> L
[1, 2, 3]    ### as expected


###
### cStringIO in CPython
###

[localhost:~] dinu% python
Python 2.2 (#1, Feb 13 2002, 11:18:45) 
[GCC 2.95.2 19991024 (release)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> 
>>> from cStringIO import StringIO
>>> s = StringIO()
>>> s.reset()    # does also work on CPython 2.1


###
### cStringIO in Jython
###

[localhost:~] dinu% jython
Jython 2.1 on java1.3.1 (JIT: null)
Type "copyright", "credits" or "license" for more information.
>>> 
>>> from cStringIO import StringIO
>>> s = StringIO()
>>> s.reset()
Traceback (innermost last):
  File "<console>", line 1, in ?
AttributeError: instance of 'org.python.modules.cStringIO.StringIO' has 
no attribute 'reset'


From philh at comuno.freeserve.co.uk  Thu Feb  7 18:15:24 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:15:24 +0000
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>  <00b901c1af9e$9de1d080$0101010a@local> <mailman.1013103792.9086.python-list@python.org>
Message-ID: <slrna662kc.d87.philh@comuno.freeserve.co.uk>

On Thu, 07 Feb 2002 18:40:33 +0100, Laura Creighton <lac at strakt.com> wrote:
>The only reason that Timothy Rue should leave is because we are absolutely
>of no use to him in the furthering of his project.  If absolutely everybody
>who has no intention of helping him had the discipline to shut up and
>stay shut, he could more accurately judge how helpful posting here is
>to his project.
>
>Voting people out of a newsgroup, as a concept, disgusts me.

I suspect you might not have heard of Matt Giwer.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From soperpd at yahoo.com  Fri Feb 15 15:56:42 2002
From: soperpd at yahoo.com (Paul Soper)
Date: 15 Feb 2002 12:56:42 -0800
Subject: Tk 8.4a3 & Tkinter - multiple file selection in tk_getOpenFile
Message-ID: <f89f6348.0202151256.a0a9957@posting.google.com>

I'd like to use the multiple file selection ability of tk_getOpenFile
in Tk 8.4a3 with Tkinter and Python 2.2 running on Windows.  What is
the least amount of change to the standard 2.2 distribution which
would enable this?

If anyone knows of a better (or even uglier but easier) way to do
multiple file selection, that information would also be appreciated.


From soperpd at yahoo.com  Wed Feb 20 09:42:56 2002
From: soperpd at yahoo.com (Paul Soper)
Date: 20 Feb 2002 06:42:56 -0800
Subject: How to hide Tk root window when using only canned dialogs?
Message-ID: <f89f6348.0202200642.7d38ae20@posting.google.com>

The example script below works fine but the Tk window is visible
throughout.  Is there a way to make it invisible?

import shutil
import sys
import tkMessageBox
import tkFileDialog

infn = tkFileDialog.askopenfilename()
if not infn:
    tkMessageBox.showerror("No input file", "No input file given")
    sys.exit()
outfn = tkFileDialog.asksaveasfilename()
if not outfn:
    tkMessageBox.showerror("No output file", "No output file given")
    sys.exit()    
shutil.copyfile(infn, outfn)
tkMessageBox.showinfo("File Copy Complete", "%s\ncopied to\n%s" %
(infn, outfn))


From emile at fenx.com  Wed Feb 13 14:36:35 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Feb 2002 11:36:35 -0800
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013624902.21586.python-list@python.org>
Message-ID: <a4efi4$1ebg78$1@ID-11957.news.dfncis.de>

> -- Steve "the reformed packrat" Majewski
> 

Is there a step program to reform?  ;-)

desperately-seeking-anything-ly y'rs,

-- 

Emile van Sebille
emile at fenx.com

---------



From pje at telecommunity.com  Sun Feb 24 22:01:25 2002
From: pje at telecommunity.com (Phillip J. Eby)
Date: Sun, 24 Feb 2002 22:01:25 -0500
Subject: "Data" vs. "Non-data" Descriptors in PEP 252, and "once" attributes
Message-ID: <MPG.16e373104557a63b989685@news.supernews.com>

One nice feature of 2.2 is that you can define attribute descriptors in 
Python.  I've found a way to use this to create "once" attributes: 
instance attributes which are computed once per instance, with the value 
cached from that point on.  Here's an example::

    class once(object):

        def __init__(self,name,func):
            self.name = name
            self.func = func

        def __get__(self, ob, typ=None):
            if ob is not None:
                value = self.func(ob)
                ob.__dict__[self.name]= value
                return value

    class aThing(object):

        def anAttr(self):
            print "computing anAttr"
            return 100*100

        anAttr = once('anAttr', anAttr)

    thing = aThing()

    # prints "computing anAttr", then 10000
    print thing.anAttr

    # just prints 10000 the second time
    print thing.anAttr


Interestingly, this does *not* work if you try to do something similar 
with the built-in 'property' type.  Apparently, 'property' is considered 
a data descriptor, while the above 'once' type is considered a "non-data" 
descriptor.

Does anyone know if this usage is safe for future Python versions?  In 
another variation of this, I use a metaclass to make classes that 
automatically instantiate themselves inside of a containing class, e.g::

    class onceClass(type):
        
        def __get__(klass, ob, typ=None):
            if ob is not None:
                instance = klass(ob)
                ob.__dict__[klass.__name__] = instance
                return instance
        
    class Outer(object):

        class Inner:
            
            __metaclass__ = onceClass

            def __init__(self, ctx):
                print "creating Inner, bound to", ctx
                

    anOuter = Outer()

    # prints creation message, then the inner object
    print anOuter.Inner

    # just prints the inner object
    print anOuter.Inner

This sort of thing is handy for when you have complex (but relatively 
static) object structures that need to be built up.  I imagine it will 
come in especially handy for GUI programming, but I haven't actually used 
it for that yet.

Anyway, I'd like to be sure I'm not doing anything which is "unsafe" for 
future versions.  Although PEP 252 spells out most of the rules quite 
clearly, there is one bit of ambiguity: the definition of "data" versus 
"non-data" descriptors.  To be precise, it doesn't really define the 
difference.  The 2.2 source operationally defines a data descriptor as 
one which has a 'tp_descr_set' slot value, and new-style classes don't 
set that slot unless you supply a '__set__' method.  (Interestingly, 
'__delete__' is ignored for slot setup purposes, and supplying 
'__delete__' without a '__set__' still results in a "non-data" 
descriptor.

Operationally, it seems it would be hard to change any of this in a way 
that would break my "once" examples without also breaking lots of other 
things.  But the PEP doesn't really give an official way to create either 
a data or non-data descriptor; it just says what the current 
implementation does.  Hence my concern.

Of course, if upon reading this, lots of people think that "once" 
attributes are cool and start using it, I suppose it'll become a backward 
compatibility issue for more than just me.  :)


From tim.one at comcast.net  Sat Feb 23 20:17:04 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sat, 23 Feb 2002 20:17:04 -0500
Subject: [Tutor] Working with Email question  [new 'email' module in Python
 2.2]
In-Reply-To: <6AD3F866286@kserver.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCEENLNOAA.tim.one@comcast.net>

[Sheila King]
> There are other problems with the email module as well.
> ...

Sheila, if you belive there's a bug, you should open a bug report.  No bug
gets fixed by arguing on a mailing list, and I can pretty much guarantee
that the email package's author isn't going to see anything said about it
unless he reads it in a bug report.

A bug report doesn't need a lot of argument unless it proves to be
controversial; but since you haven't yet gotten this to the author's
attention, you can't yet know whether Barry will love or hate the complaint.
It doesn't really matter what Oleg or I may think:  we're not the experts
for this package.

> ...
> I hope this doesn't sound like sniping or whining.

Ditto:

> ...
> But I believe that my complaints are valid, and that these are problems
> that need to be corrected in the email module.

Without a bug report, nothing will change.  I often open one for a poster
when a posted behavior is obviously a bug, or when the poster seems unable
to articulate their case clearly.  This one doesn't fit either case, so I'm
leaving it to you:

    http://sf.net/tracker/?group_id=5470&atid=105470

Barry may or may not agree it's a bug, but-- honest --he doesn't bite,
except on Thursdays <wink>.




From sjmachin at lexicon.net  Sat Feb 16 17:43:09 2002
From: sjmachin at lexicon.net (John Machin)
Date: 16 Feb 2002 14:43:09 -0800
Subject: Warning about "for line in file:"
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu>
Message-ID: <c76ff6fc.0202161443.3ba0f746@posting.google.com>

Brian Kelley <bkelley at wi.mit.edu> wrote in message news:<3C6D73DE.3090201 at wi.mit.edu>...
[snip]
> The behavior that I would expect is exemplified in the following correct 
> example that actually prints out all lines of a file.

Correct? Does "a file" mean a particular file with 11 or fewer lines?
It certainly doesn't work in generality for larger files:

C:\junk>cat iter_titter.py
fsim = ["Line " + str(x) for x in range(20)]
count = 0
iterator = iter(fsim)
for line in iterator:
     if count > 10: break
     print line
     count = count + 1
for line in iterator:
     print line

C:\junk>python iter_titter.py
Line 0
Line 1
Line 2
Line 3
Line 4
Line 5
Line 6
Line 7
Line 8
Line 9
Line 10
Line 12
Line 13
Line 14
Line 15
Line 16
Line 17
Line 18
Line 19

Looks like a bit of a whoopsie with Line 11.


From jkraska at san.rr.com  Fri Feb  8 18:53:12 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 08 Feb 2002 23:53:12 GMT
Subject: Usenet fascism
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <kk186ugk4hpo7m0eca81bv2lm2d0smnjiv@4ax.com> <1HV88.108504$KM2.4230280@atlpnn01.usenetserver.com>
Message-ID: <4gp86uoa4nha3ui6qssksi2f30ao1820st@4ax.com>

>I suspect you are confusing rights and abilities.

You're wrong. I differentiated: "Plainly is the right not
only held, but it is exercised". The ability is part and
parcel to it being exercised. The right is also _held_
which is different from the ability.

>I don't have the *right* to do that, however.

Non sequitur. The right to free speech includes the right
to tell others you find their opinion execrable and to
"scram!" as much as it does to hang out and be annoying.
Come now.

C//



From loewis at informatik.hu-berlin.de  Sun Feb  3 09:27:59 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 03 Feb 2002 15:27:59 +0100
Subject: pickle security
References: <mailman.1012591743.10841.python-list@python.org> <7xit9g51lz.fsf@ruckus.brouhaha.com>
Message-ID: <j4g04ihc7k.fsf@informatik.hu-berlin.de>

Paul Rubin <phr-n2002a at nightsong.com> writes:

> > Did I get it right?  Are there any other security issues I need to be aware
> > of, or does this cover them?
> 
> I continue to feel uncomfortable using unpickle on untrusted data.
> For example, it calls eval to handle quoted strings.  While that looks
> safe on the surface, there are an awful lot of code paths you have to
> examine to make sure it can't nail you.

Apart from feeling uncomfortable, can you point to a real
counter-example? The code to deal with quoted strings is designed to
check that there it is nothing but a quoted string.

So, apart from setting find_global, I don't think there are any
further issues to be aware of.

Regards,
Martin


From logiplexsoftware at earthlink.net  Fri Feb  1 14:02:43 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 11:02:43 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <Wls68.14178$om6.357057@news1.tin.it>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3d46d$tfn2@news1.gtech.com>
	<Wls68.14178$om6.357057@news1.tin.it>
Message-ID: <20020201110243.3f2dfdc0.logiplexsoftware@earthlink.net>

On Fri, 01 Feb 2002 08:29:42 GMT
Alex Martelli wrote:

> Peter Milliken wrote:
>         ...
> > Actually, if you are *good* at what you do, then code review, whilst
still
> > beneficial, is not nearly *that* beneficial. Things like code review
(IMO
> > :-)) is there for the 25-50% of programmers who just plain shouldn't be
in
> > the industry - what they generate is just absolute garbage - code
review
> > will hopefully stop the code from getting into Unit test and beyond the
> > point of redemption :-).
> 
> I disagree.  Code review, or even better the "continuous ongoing code
> review" that's such a substantial part of the benefits of pair
programming,
> is in my experience just as beneficial with the best coders: without it, 
> besides bugs, it's far too much a temptation to pepper the code with
> clever tricks, individual style quirks, unexamined (subconscious) 
> assumptions about inputs and environmental conditions, and the like.
> The code coming out of a good review, or a good-synergy pair, is
> thus of much higher quality and more valuable than the results of a lone
> coding session of even a very good coder.

I couldn't agree more.  I worked for a while with another programmer who
was very good and I know while we were working together it seemed as if the
whole of our efforts exceeded the sum of the parts.  This was very likely
due to the fact that while we both agreed on general principles, our skill
sets and backgrounds in programming were very different, so we tended to
each see things the other did not. 
Additionally, because programming is a pure mental excercise (if you
discount the typing), psychology and morale play a very large role in
productivity for a programmer and having a second like-mind (even if only
to bounce ideas off of) can have a huge positive effect on that aspect.

I much prefer the "continuous ongoing code review" idea better than a
formalized procedure (if I understand what you mean by it correctly). 
Working in pairs provides this as a side-effect rather than having a formal
procedure for code-review and seems less time-consuming overall.

> 
> > Agreed with the "Art" - many programmers give one the impression that
they
> > perceive themselves as a Prima Donna of the software industry :-) Ask
> > someone to look at their code (as in code review) and they look
offended -
> > like you'd just asked for their first born or something :-) A severe
lack
> > of comprehension is the general response - "You want to look at *my*
code?
> > What one earth for! It's perfect (by definition)" :-).
> 
> I fully agree with you.  But that implies my stance in the above
paragraphs,
> not yours:-).  Actually the root of the problem may be in 'my' vs
'yours',
> the "*my*" in your quote being correctly emphasized:-).

Actually, when I say "art", I am using it in the sense of "skill", as in
"martial art" or "art of design" versus the broader sense that includes
unrestrained personal interpretation.  In this sense "art" means taking
basic, established techniques and internalizing them to the point where
they are second-nature. At that point, personal expression becomes an
extension of those techniques - not a deviation from them.  My experience
has been that programmers at this level feel a sense of pride and personal
accomplishment and _appreciate_ code review (otherwise, who will ever see
and appreciate their accomplishment?).  I suspect this to be a major
driving force behind open-source software - it's the equivalent of a public
demonstration of one's skills. 

This is not to say that radical departures can never be beneficial, but
these departures are rare and do require intense peer review.  

-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From jason at jorendorff.com  Wed Feb  6 15:10:00 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 14:10:00 -0600
Subject: RFC -- Hygienic Macros in Python
In-Reply-To: <5cf88.18880$zJ3.1464485@typhoon.maine.rr.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEMPDAAA.jason@jorendorff.com>

Mike Carifio writes:
> Can someone fill in the newbies on this topic? Not all macro systems
> are just "blind text substitution". Scheme and LISP both have reasonable
> macro facilities which, when used judiciously, can actually enhance the
> comprehensibility of code (in my opinion, of course).

Well, one question is whether a feature honestly lends itself to
be "used judiciously" by the uninitiated coder.  Call it the
"judiciousness quotient", or JQ.  It's a pragmatic measure of how
easy a language feature is to use.

Consider overloading __call__().  I think the judiciousness quotient
for that is about 35%.  That is, 65% of the time it's used, it only
makes the code more mysterious.

eval() has a JQ of about 4%.

For new language features like list comprehensions and generators,
it's hard to tell.  My intuition says:

  list comprehensions: 50%
  generators: 75%
  macros: 20%

Also, when macros are used *in*judiciously, the pain is greater
than with, say, a confusing list comprehension.  Consulting the
source of a macro to figure out what it does is hard.  Or rather,
for any sufficiently interesting syntax, it's hard to write a
really clear macro.  (Of course, much depends on the syntax
provided for defining macros.)

## Jason Orendorff    http://www.jorendorff.com/




From eppstein at ics.uci.edu  Thu Feb 28 23:56:34 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 20:56:34 -0800
Subject: Status of PEP's?
References: <mailman.1014944882.14890.python-list@python.org>
Message-ID: <eppstein-8E7829.20563428022002@news.service.uci.edu>

In article <mailman.1014944882.14890.python-list at python.org>,
 James_Althoff at i2.com wrote:

> For example, one problem I had when playing with an experimental
> implementation of the "middle argument" variant mentioned above is that the
> specific relational operators are tied to the values of the bounds.  If you
> want to iterate from the smaller value (up) to the larger value you write:
> 
> for 1 <= var < 5:
> 
> whereas if you want to iterate from the larger value (down) to the smaller
> value you have to turn things around and write:
> 
> for 5 > var >= 1:
> 
> which all looks very nice when you are dealing with literal constants.  But
> suppose you have variables for the bounds, "start" and "end"?  Let's say I
> have a function
> 
> def doSomethingFromTo(start,end):
> 
> The function has to test to see which of "start" and "end" is the smaller
> value in order to know whether to execute a for-loop with "<" or a for-loop
> with ">" (and the code specified in each separate for-loop has to be
> written redundantly).  I found this to be awkward.

What's wrong with
    for min(start,end) < var < max(start,end)
?

I am assuming you want an open interval because you wrote < and > above.
If you want a closed interval, of course, use <=, but I don't think it 
makes sense to ask for a half-open interval when you don't know which end 
is which.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From pereira at cis.upenn.edu  Mon Feb 25 22:38:39 2002
From: pereira at cis.upenn.edu (Fernando Pereira)
Date: Mon, 25 Feb 2002 22:38:39 -0500
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org>
Message-ID: <B8A06D6F.3BDC%pereira@cis.upenn.edu>

On 2/25/02 9:23 PM, in article
mailman.1014690284.11692.python-list at python.org, "Greg Ewing"
<greg at cosc.canterbury.ac.nz> wrote:

> Hmmm... maybe it should be called Pythonlet...

Let's see... Lots of pythonlets... Nest of vipers? <duck/>

-- F



From anthony_barker at hotmail.com  Wed Feb 27 12:34:39 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 27 Feb 2002 09:34:39 -0800
Subject: what I would like to see in python to make a better "glue" language (Newbie)
References: <899f842.0202261325.6ff3dd54@posting.google.com> <m3sn7nubyy.fsf@mira.informatik.hu-berlin.de>
Message-ID: <899f842.0202270934.281b65e4@posting.google.com>

martin at v.loewis.de (Martin v. Loewis) wrote in message > 
> Adding modules is typically a two-stage process. First, a specific
> module should establish itself as the monopoly implementation on a
> certain field. Then, it can be considered for inclusion in the
> standard library.
> 
> If you have any specific LDAP module in mind, I'd expect, as a
> prerequisite, that it is a distutils package. That should make
> installation very easy, and open doors to inclusion in "foreign"
> distributions. E.g. Linux distributors may include it. Also, there
> always was the plan of a Sumo distribution which includes all packages
> that atleast build successfully - volunteers to make this come true
> are still needed.

SUMO distro - sounds fun - kind of like the everything included linux
kernels that are floating around. Actually the ActiveState windows
distribution seems to be a bit of a "SUMO" distro.



> > 2. DNS really should be included in the base library, I emailed
> > Anthony Baxter and he replied, saying it was almost done. 
> I completely agree.
 
> > 3. Telnet Library is great - including an SSH module would be a nice
> > addition.
> 
> This is not trivial. To my knowledge, there is no SSH library that you
> could link. I don't know how difficult it is to implement a SSH module
> on top of OpenSSL.

Here is some code that could have wrappers built around:
1. Putty (a windows app with some base c code)
http://www.chiark.greenend.org.uk/~sgtatham/putty/. others - such as
winscp (a delphi application) seem to have sucessfully written
wrapper classes around it. I don't know how portable it is though. 

2.Have a look at Net::SSH::Perl - Perl client interface to Secure
Shell
http://rhumba.pair.com/ben/perl/net-ssh/
It is also available on CPAN.

> > 4. Include an SNMP implementation, there are several around, but a
> > pure python one as a library would be nice
> 
> All such libraries are quite large, right? If so, chances for
> inclusion in the standard library are low, since I expect the user
> community for such a library is quite small. Nevertheless, bug the
> maintainers of those libraries for distutils support.

If there is a package that supports distutils. what is the best way to
get this done? Post to the python-dev mailing list?

Thanks for the feedback,

Anthony


From logiplexsoftware at earthlink.net  Mon Feb 25 14:36:16 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 25 Feb 2002 11:36:16 -0800
Subject: [Tutor] unrelated bug issue
In-Reply-To: <LNBBLJKPBEHFEDALKOLCEEOBNOAA.tim.one@comcast.net>
References: <3C78529D.F4E7D754@mail.verizon.net>
	<LNBBLJKPBEHFEDALKOLCEEOBNOAA.tim.one@comcast.net>
Message-ID: <20020225113616.4af2bcaa.logiplexsoftware@earthlink.net>

On Sun, 24 Feb 2002 01:00:59 -0500
Tim Peters wrote:

> I got a better one:  suppose you shoved a banana down your throat, then a
> carrot in each nostril.  Would it then be the banana's or the carrots'
fault
> that you couldn't breathe anymore?  That's right:  it would be the
carrots',
> because you didn't have any trouble breathing before they wound up in
your
> nose.  Or maybe it was the banana's fault after all, since if it weren't
> there, the carrots' presence wouldn't matter.  It's a puzzler.  Just so
long
> as it's clear that it's never ever your fault <wink>.

Why bananas and carrots?  I think the image of a python down the throat and
a ruby in each nostril seems much more exotic and interesting.  This attire
might even be considered attractive in some cultures.  If this were the
case, then perhaps it would be arguable that it really /isn't/ the OP's
fault, but rather society's for imposing unattainable standards of beauty
upon him.

Really Tim, you are _so_ culturally insensitive.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From peter at engcorp.com  Tue Feb 12 08:05:59 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 12 Feb 2002 08:05:59 -0500
Subject: Exiting threads silently
References: <a4au9e$u1v$01$1@news.t-online.com>
Message-ID: <3C691337.A891C9DA@engcorp.com>

Michael Bauer wrote:
> 
> What happens is the error page gets displayed correctly in the browser, but
> therad.exit() dumps following exception to the console:
> 
> ----------------------------------------
> Exception happened during processing of request from ('127.0.0.1', 40400)
> Traceback (most recent call last):
>   File "/usr/lib/python2.2/SocketServer.py", line 458, in
> process_request_thread
>     self.finish_request(request, client_address)
>   File "/usr/lib/python2.2/SocketServer.py", line 253, in finish_request
>     self.RequestHandlerClass(request, client_address, self)
>   File "/usr/lib/python2.2/SocketServer.py", line 514, in __init__
>     self.handle()
>   File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 170, in handle
>     server, client = self.Connect(host, port)
>   File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 346, in Connect
>     self.error(200, 'Error connecting to "%s" (%s)' % (host, err))
>   File "/home/fake/devel/iowl/pProxy/cProxyHandler.py", line 504, in error
>     thread.exit()
> SystemExit
> ----------------------------------------
> 
> Anybody has an idea how i can exit silently?

The docs (i.e. the SocketServer source) suggest that the default 
error handler will print a traceback and continue when an exception
is raised.  Try overriding handle_error() on your server class,
catch SystemExit separately, and re-raise it if you get it.

(There may be better ways of handling this.  I'm no expert.)

-Peter


From gombosk at vas.birosag.hu  Sun Feb 10 06:58:27 2002
From: gombosk at vas.birosag.hu (gombosk at vas.birosag.hu)
Date: Sun, 10 Feb 2002 12:58:27 +0100 (CET)
Subject: Python + windows + problems with make.py
In-Reply-To: <HFEKILOLEFEFMKAECNDLEEGJCMAA.jason@jorendorff.com>
Message-ID: <Pine.LNX.4.33.0202101252460.12697-100000@vb24.vas.birosag.hu>

Haliho!

I want to put my *.py from Linux to windows.

1. I have downloaded installer_linux_b_01.tar.gz

2. I want to run the /builder/source/Make.py

3. I've got the next:

reading exceptions from
Traceback (most recent call last):
  File "./Make.py", line 133, in ?
    main()
  File "./Make.py", line 73, in main
    print "reading exceptions from", exceptions.__file__
AttributeError: 'exceptions' module has no attribute '__file__'


Yes, exceptions module has no attribue '__file__'

What can I do with it?

Thanks: Katalin




From jason at jorendorff.com  Wed Feb 20 14:17:02 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 13:17:02 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <19650FF08A99D511B19E00B0D0F06323060AC3@SRM201002>
Message-ID: <HFEKILOLEFEFMKAECNDLKEAMDEAA.jason@jorendorff.com>

Gustavo Cordova wrote:
> >>> class WrapRX:
> ... [...]

This is standard in Python 2.2.

  >>> import re
  >>> rx_vowels = re.compile(r"[aeiou]+", re.S)
  >>> some_string = "This is a beautiful string."
  >>> for match in rx_vowels.finditer(some_string):
  ...     print "Found vowel: %r" % match.group(0)
  ...

## Jason Orendorff    http://www.jorendorff.com/





From fperez528 at yahoo.com  Mon Feb  4 13:35:25 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 04 Feb 2002 18:35:25 +0000
Subject: [Fwd: directly executing a python script in Unix]
References: <3C5F1638.1D9B527C@ix.netcom.com>
Message-ID: <a3ncv6$fgs$1@peabody.colorado.edu>


>> I've read once that Linux doesnt finds she-bang command scripts if
>> they end with something different from a newline - for example a
>> carriage return plut a newline (MS Windows / DOS convention). So it's
>> possible to write such a script under DOS, which is visually identical
>> but won't run under Linux.

Here's my python fix for that moronic problem (moronic of Dos designers, who 
thought that two characters per line ending were a good idea when there 
already was a perfectly viable convention to follow, namely that of Unix. 
Perhaps they were following a VMS convention or somesuch, I don't know).

Call this tounix.py, put in your path and use as needed even on a whole 
directory ("tounix.py *" will work):

[python]> cat tounix.py
#!/usr/bin/env python

"""Convert DOS or Macintosh text files to unix end-of-line convention.

Usage:
  tounix file1 file2 ...

A backup (file1~,file2~,etc.) of each file is made.
"""

import sys,os

files = sys.argv[1:]

if len(files) == 0:
    print __doc__
    sys.exit()

for file in files:
    original = open(file).read()
    bak = file+'~'
    os.rename(file,bak)
    new = open(file,'w')
    new.write('\n'.join(original.splitlines()))
    new.close()


PS. This is quick and dirty but works. If you want, you can spice it up with 
lots of error checks and whatnot. 

Cheers,

f


From andy47 at halfcooked.com  Thu Feb  7 17:25:03 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Thu, 7 Feb 2002 22:25:03 +0000 (UTC)
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <slrna65qur.hgk.huaiyu@gauss.almadan.ibm.com>
Message-ID: <Xns91AF5FB58F54Fandy47halfcookedcom@203.109.250.24>

huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu) wrote in 
news:slrna65qur.hgk.huaiyu at gauss.almadan.ibm.com:

> On Wed, 06 Feb 2002 09:48:10 -0800, Jeff Shannon <jeff at ccvcorp.com> 
wrote:
>>
>>The problem you're running into, is that you've failed to convince anyone 
that
>>*you* understand the real problems involved, enough to be able to solve 
them.  You
>>don't need to convince us of the benefits of autocoding; you need to 
convince us
>>of *your* competence.  So far, you're failing miserably. 
> 
> He has several other major problems:
> 
> 1. He seems to be unable to differentiate and respond well to negative
>    comments.  He treats them all equally, and replies to the more 
negative
>    ones with more enthusiasm.  Can he tell the difference between the
>    following comments?  "I don't understand."  "I's already done."  "It
>    can't be done."  "Yeah, that'll be good, but I don't think you've got 
the
>    right approach to do it."  Sometimes he can't even see a humor.
> 
> 2. He seems to claim that he has already abstracted all of computing into
>    his nine commands, yet he also claims that this vision of his is
>    impossible to be put into words, examples or mathematics.  That's kind 
of
>    contradictary since the art of programming, or much of science itself, 
is
>    to figure out exactly how things behave, and how to make things work 
as
>    desired.  It is the art of collecting and summarizing details, to
>    formulize and abstract them, to articulate and communicate them.
> 
> 3. He misunderstood the mechanism of Open Source projects.  These 
projects
>    exist because it is more efficient to do things that way.  A leader 
being
>    a leader must have the ability to make things more efficient for 
others.
>    He claims he knows the road ahead, but don't have the time to travel
>    there.  So let's ask some hyperthetical questions.  Suppose there are
>    twenty programmers willing to join his project.  Now he could achieve 
in
>    a year or two what he would have done by himself in twenty years.  
What
>    are the tasks these programmers would be handling?  What are the most
>    urgent theoretical problems to be solved?  What would be the codes he
>    wants them to write in the first month?  Will he handle the final 
checkin
>    and approval of code, and if so, what functionality and other criteria
>    he'll use to judge them?  He seems to think that one man with a vision
>    plus a million monkeys jumping on keyboards would solve real problems.
> 
> So with these problems in the way, it is not of much use communicating on
> this topic.  The first and second problems also imply that there's not 
much
> chance that any of these problems can be solved soon.
> 
> (Rue, if you are listening here, there's no need to respond.  We are not
> sympathetic people here.  Programmers behave like Nazies here.)
> 
> 
> Huaiyu

And thus Godwin's law is proven once more and this discussion can draw to a 
close.

I thank you,
Andy
-- 
Contents free posts a speciality


From sjmachin at lexicon.net  Sat Feb 16 03:07:07 2002
From: sjmachin at lexicon.net (John Machin)
Date: 16 Feb 2002 00:07:07 -0800
Subject: COFF2OMF C++ Builder 4
References: <2830c89c.0202150629.5a6b711f@posting.google.com> <MPG.16d7dccfbfb730039896d6@newnews.dircon.co.uk>
Message-ID: <c76ff6fc.0202160007.2097f54@posting.google.com>

Carl Caulkett <carlca at dircon.co.uk> wrote in message news:<MPG.16d7dccfbfb730039896d6 at newnews.dircon.co.uk>...
> In article <2830c89c.0202150629.5a6b711f at posting.google.com>, 
> andrew.gregory at npl.co.uk says...
> > Trying to create BCB4 DLLs which can be imported into Python 2.2
> > COFF2OMF python22.lib new.lib gave me error message "Invalid COFF file".
> > What is the easiest way to fix this?
> 
> Hi Andrew,
> 
> This page may help. 
> 
> http://www.cyberus.ca/~g_will/pyExtenDL.shtml
> 
> It's oriented towards the Borland C++ v5.5 command line tools which is 
> based on the BCB5 compiler back-end. Hopefully it will be relevant to 
> BCB4 as well. I can confirm that I have been able to compile PYDs using 
> the info on this page.

Gordon's info (N.B. page last updated some time in 2000) was extremely
useful, before distutils came out. Distutils makes compiling
extensions on Windows a snap; it supports using the Borland freebie
kit or the MinGW compiler as well as the Microsoft VC++ product.


From scarblac at pino.selwerd.nl  Thu Feb 21 12:45:45 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 21 Feb 2002 17:45:45 GMT
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr> <3c7403d7.25939921@news.t-online.de> <3C74241F.3060403@free.fr> <3c74c3d6.20466328@news.isar.de>
Message-ID: <slrna7accg.f7i.scarblac@pino.selwerd.nl>

Gerson Kurz <gerson.kurz at t-online.de> wrote in comp.lang.python:
> I found an easier solution:
> 
> sort = lambda x:(x.sort(),x)[1]

def sort(x): x.sort(); return x

is equally concise and doesn't have to do the odd tuple thing :)

-- 
Remco Gerlich


From core.lists.python at core-sdi.com  Tue Feb 26 16:24:11 2002
From: core.lists.python at core-sdi.com (Mariano Corsunsky)
Date: Tue, 26 Feb 2002 18:24:11 -0300
Subject: Oddity question
Message-ID: <01de01c1bf0b$ef204720$6e42a8c0@corelabs.coresdi.com>

Sorry, my mistake. I forgot the "self." when I wrote the question...

Here, again....

 I was fooling around with sockets last night, using a thread to deal with
the read part and another method to do the print.

 like this:

def thread_read (self):
    while 1:
        self.buf=self.buf+self.my_socket.recv (8)

def print_buf(self):
    if self.buf:
        print "bufeer-->",self.buf
        self.buf='""

But, what happened to me was that the -buf = ""- line seemed to be not
working  because I got outputs like this

buffer->a
buffer->ab
buffer->abpleasework
buffer->abpleaseworkpleaseflush!


 So, I dont know why I changed thread_read function to:

def thread_read (self):
    while 1:
        aux=3Dself.my_socket.recv (8)
        self.buf=3Dself.buf+aux


And this version worked properly..... but I still do not know why.
To me, I've only split the line in two, using an auiliar variable... but
THAT made the trick!
why?
 WHY?
   W H Y?

(thanks for answers)
>
>
> --
"mcherm" <mcherm at destiny.com> wrote in message
news:3C7BF84F.1090306 at destiny.com...
> > I was fooling around with sockets last night, using a thread to deal =
> > with the read part and another method to do the print.
> >
> > like this:
> >
> > def thread_read (self):
> >   while 1:
> >     self.buf=3Dself.buf+self.my_socket.recv (8)
> >
> > def print_buf(self):
> >   if buf:
> >     print "bufeer-->",buf
> >     buf=3D'""
> >
> > But, what happened to me was that the -buf =3D ""- line seemed to be not
=
> > working     [...]
>
> > So, I dont know why I changed thread_read function to:
> >
> >   while 1:
> >     aux=3Dself.my_socket.recv (8)
> >     self.buf=3Dself.buf+aux
> >
> > And this version worked properly..... but I still do not know why.
>
> Well, I can't tell exactly what you were doing here (try sending plain
> text instead of html... it doesn't get mangled as much), but I do see
> one thing. In the example that doesn't work you have a line 'buf=""' at
> the end of a function or method, while the example that DOES work has
> 'self.buf=self.buf+aux' inside a loop.
>
> If buf is a local variable, then 'buf=""' at the end of a function will
> only modify the local variable and do nothing to the actual buffer
> (strings are immutable in Python). Modifying self.buf in a loop will not
> be lost in this fashion.
>
> Does that help any?
>
> -- Michael Chermside
>
>
>
>
>
> --
> http://mail.python.org/mailman/listinfo/python-list


--- for a personal reply use: "Mariano Corsunsky" <morsa at core-sdi.com>



From quinn at vomit.ugcs.caltech.edu  Thu Feb 14 14:25:09 2002
From: quinn at vomit.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Feb 2002 19:25:09 GMT
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <X1va8.207034$KM2.8517807@atlpnn01.usenetserver.com> <a4gsg7$m4i$1@sirius.dur.ac.uk>
Message-ID: <slrna6o3ol.2ji.quinn@vomit.ugcs.caltech.edu>

On Thu, 14 Feb 2002 17:35:03 +0000 (UTC), Christopher Saunter
<Christopher.Saunter at durham.ac.uk> wrote: >Steve Holden (sholden at holdenweb.com)
wrote:
>Also, on the subject of number bases, is there an easy (read library)  way
>to take a string containing a number represented as hex and convert it to
>an int or float?  I've looked through the libraries, and I'm sure it's
>there somewhere, just can't find it.....

Closer than you think:

print int.__doc__


From duncan at NOSPAMrcp.co.uk  Mon Feb 25 08:20:32 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Mon, 25 Feb 2002 13:20:32 +0000 (UTC)
Subject: Dictionary type access to list of instances
References: <a5dcgh$b22$1@ctb-nnrp1.saix.net>
Message-ID: <Xns91C0868A14D1Dduncanrcpcouk@127.0.0.1>

"Grant Beasley" <gbeasley at tsa.ac.za> wrote in
news:a5dcgh$b22$1 at ctb-nnrp1.saix.net: 

> If I have a list of instances, each with a property which I want to
> use as a key to access the list in dictionary-like manner, what would
> be the best way of doing this?

Is there any reason why you can't just store your instances in a dictionary 
using the property as the key?

-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From billludden at email.msn.com  Mon Feb 11 20:01:27 2002
From: billludden at email.msn.com (billludden)
Date: Mon, 11 Feb 2002 18:01:27 -0700
Subject: Newbie needs references
References: <u9bkUtmsBHA.388@cpimsnntpa03> <3C680FA9.4050101@bath.ac.uk>
Message-ID: <e9JQxPEtBHA.1572@cpimsnntpa03>

Thank you!




From phr-n2002a at nightsong.com  Mon Feb  4 19:06:53 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 16:06:53 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org> <7xu1sx41cl.fsf@ruckus.brouhaha.com> <3C5F04BF.3090806@yumpee.org> <7x4rkwdfx2.fsf@ruckus.brouhaha.com> <B884CF01.43F6%jonathan@onegoodidea.com>
Message-ID: <7xlme83i76.fsf@ruckus.brouhaha.com>

Jonathan Hogg <jonathan at onegoodidea.com> writes:
> > The program I'm hacking right now actually needs exact rational
> > division, but that's another story.  The issue is there's no way to
> > use integer division that works in both 2.1 and 3.0, without using
> > something like divmod that would shock the unwary.
> 
> I'm sorry, the "unwary"? I presume by this you mean that your user base
> contributes to your code, that you aren't the only programmer on the
> project? Are the other coders having difficulty with reading or writing
> 1.5.2 forward compatible code?

By the unwary maybe I should have said the unconverted.  For example,
I sent a Python script last night to a Lisp user, that very nicely
solves in Python something that the Lisp people were going on about
how Lisp was so great for.  The intention was that the Lisper would
see the clean infix expressions, the absence of parentheses, the
natural expression of the algorithm, and say "wow, neat, I think I'll
start programming in Python too".  But if instead of a/b they have
to see divmod(a,b)[0], they'll think it's even worse than parentheses.

> The easy way for you to tell is to make two versions of your code available,
> one written for 1.5.2 and one optimised for 2.1 upwards (where they may be
> exactly the same version - I'm not suggesting you do any more work). You'll
> be able to gauge from downloads/sales/however your users obtain your work,
> which version they have.

I have no way to know from downloads how many copies of the program
are spread around by other means.  It's free software, so there's no
sales.  It's the exact same problem as knowing how many users are
running Python 1.5.2.  Downloads from python.org are probably
outnumbered by Red Hat cd's by a fair margin.

> At the moment, this honestly looks like a storm in a teacup to me. I'm sure
> you can put a small disclaimer on your program saying "I cannot guarantee
> this program will work in some unknown future time on an as yet unknown
> version of the Python interpreter."

I could also save myself a lot of programming by writing a small
disclaimer saying that the program doesn't work on the present
interpreter either, but the idea is to write code that works.  The
next best thing is to write code that fails immediately and prints an
error message.  The last thing I want is code that appears to work but
generates the wrong answers, which is what the division change can cause.

I guess if I don't want to deal with making separate 2.1 and post-2.1
versions, I can just put in an initialization check that stops the
program from working in 3.0 and higher, and says to download a new
version of the program (much smaller than downloading a new
interpreter) which would presumably exist by the time 3.0 became
widespread.  Maybe I'll end up doing that.


From jkraska at san.rr.com  Wed Feb 27 00:03:30 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 27 Feb 2002 05:03:30 GMT
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3C7C6252.FC779971@cosc.canterbury.ac.nz>
Message-ID: <t6qo7us0md30ttpdv06fnt2mtvlm0i4i26@4ax.com>

>For these reasons, saying just "Python's grammar is
>LL(1)" is oversimplifying things considerably.

I can't speak to Python's parser and its generator,
but as for someone who is currently building and maintaining
an ANTLR (http://www.antlr.org) LL(k) grammar for a Python
derivative, I can say that it's quite definitely possible
to do, and without writing a whit of custom parser code.

I will point out that ANTLR has a feature called "syntactic
predicates" which allow you in effect to selectively make
extend the k by some fixed amount n, where specific terms
in a specific order disambiguate the grammar. Whether you
regard this as strict LL(k) is up to you, but as ANTLR is
one of the top LL(k) opensourced parser generators on the
market, the question is at best academic.

Indentation sensitivity can quite simply and very cleanly
be achieved by using ANTLR lexer multiplexing. This can
be expressed very, very succinctly in an ANLTR grammar,
where you tell it that [] {} and () use indent
insensitive mode for all inner terms, but is by default
otherwise indent-sensitive.

C//



From martin at v.loewis.de  Fri Feb 15 16:22:36 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 15 Feb 2002 22:22:36 +0100
Subject: Tkinter and Greek Symbols
References: <3daa1aa8.0202141718.3500c462@posting.google.com> <m3heoj5fda.fsf@mira.informatik.hu-berlin.de> <3c6d3278$1@news.mt.net.mk>
Message-ID: <m3it8y1lur.fsf@mira.informatik.hu-berlin.de>

?????? ?.  <???????????@????.???.??> writes:

> > If you happen to know that GREEK SMALL LETTER ALPHA is \u0391, you can
> > use that as well. If you have an editor that supports utf-8, you can
> > use unicode("<utf-8 encoding of the greek text>", "utf-8") instead
> > of the Unicode literal.
> 
> Does this mean that Tkinter support unicode?

It certainly does. Starting with 8.0, Tcl recognizes UTF-8 internally,
and provides a rich character set conversion library; with 8.1 and
later, it also offers a Unicode type internally.

Tkinter integrates the Tk Unicode support since Python 2.0.

> Does it support the iso10646-1 fonts too?

Tk uses the system fonts; it does not come with fonts on its own. It
can display Unicode strings even if the fonts are not Unicode encoded,
since it knows how to convert Unicode into the font's encoding.

> Can I display Symbols from several alphabets n a single widget?

Certainly. Tk is quite smart in selecting fonts; it tries to use the
same font as long as possible, and then switches to a different font
where necessary.

> And how do I get the keyboard input, in unicode od locale specific?

In Unicode. Actually, Tkinter, when retrieving a string from Tcl,
returns it as a byte string if it is all ASCII, and as a Unicode
string otherwise.

HTH,
Martin


From spencer at spencerdoidge.com  Sun Feb  3 22:32:05 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sun, 3 Feb 2002 19:32:05 -0800
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org> <mailman.1012787645.17462.python-list@python.org>
Message-ID: <a3kv4l$jki$1@news.efn.org>

Thanks for all your patient help, Chris. Now I have run into an apparent
contradiction:

This is a method within SerialPort_win.py from Isaac Barona's site-package,
modified by myself because his 2nd ReadFile(..) parameter was an integer,
and I want the buffer instead. I need to see what was read...

   def read(self, num=1, bfr=[0]*256):
        """Read num bytes from the serial port.

        If self.__timeout!=0 and != None and the number of read bytes is
less
        than num an exception is generated because a timeout has expired.
        If self.__timeout==0 read is non-blocking and inmediately returns
        up to num bytes that have previously been received.
        """
        overlapped=OVERLAPPED()
        overlapped.hEvent=CreateEvent(None, 0,0, None)
        hr, bfr = ReadFile(self.__handle, bfr, overlapped)
        WaitForSingleObject(overlapped.hEvent, win32event.INFINITE)
        return hr

After I have run the script that defines read(...), the other methods
defined by the script work fine, but with read(..) I get this result:

>>> tty = SerialPort("COM1",None,None,None)
>>> b = [0] * 256
>>> tty.read(1,b)
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "C:\Python22\Lib\site-packages\MyScripts\MySerialPort_win.py", line
238, in read
    hr, bfr = ReadFile(self.__handle, bfr, overlapped)
TypeError: Second param must be an integer or a buffer object

If bfr isn't a "buffer object," then what is?

Spencer Doidge




"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message
news:mailman.1012787645.17462.python-list at python.org...
> ----- Original Message -----
> From: "Spencer Doidge" <spencer at spencerdoidge.com>
>
>
> > I really tried to RTM first this time, but I failed.
> > Can someone give me a hint where in the doc html's it explains how to
pass
> > by reference?
> > Specifically, I want to do the equivalent of this:
>
> Briefly, you don't.  Or you do.  Mutable types (lists, dictionaries,
> objects)
> can be changed as you request, but immutable types (int, float, string)
> can't.
>
> Since Python supports returning multiple values (via tuples or lists) you
> normally don't need to do this.
>
> If you have a C function like this:
>
>     int MyCFunction(char *buffer);
>
> the Python equivalent would be called like this:
>
>     intval, stringval = MyCFunctionWrapped()
>
> or if the buffer is both input and output:
>
>     intval, stringval = MyCFunctionWrapped(stringval)
>
>
>
>




From fredrik at pythonware.com  Sun Feb 17 04:01:01 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 17 Feb 2002 09:01:01 GMT
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com> <Ry3a8.40197$bP3.337803@NewsReader> <8f070afd.0202120550.45d8873b@posting.google.com> <Xns91B39B73814D6mhueningzedatfuberli@130.133.1.4> <gQfa8.41893$bP3.355857@NewsReader>
Message-ID: <hjKb8.26255$l93.4949437@newsb.telia.net>

Andrew Ward wrote:
> It's very intermittent for me. I can leave it running for hours in the
> morning and it's fine, in the evenings (UK time) it usually does fail fairly
> soon.
>
> I'm rewriting the url stuff as individual httplib calls to see if I can pin
> down the exact point of failure, although I assume it will still hang, and
> that it will also be at the read() (I imagine that the urllib read() maps
> directly to the http read() method).

did you get anywhere on this one?

I just realized that a program that has been running
successfully for several years (as a cronjob) recently
started hanging about 99% of the times it's run, after
we upgraded the linux version on that box.

it also hangs when you run it from the command line,
under both 1.5.2 and 2.0.1 -- and when I press control-c,
I find that it's stuck on this line:

    oldcontent = file.read()

where "file" is an urllib stream.

and to confuse things even more, if I fire up an inter-
preter and types in the correspond urllib calls, it never
ever hangs.

bug in linux?

</F>




From jkraska at san.rr.com  Tue Feb 19 11:37:23 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 19 Feb 2002 16:37:23 GMT
Subject: Adding static typing to Python
References: <3C726575.5090400@destiny.com> <mailman.1014133892.1456.python-list@python.org>
Message-ID: <q3057us1qab29uvm2c6f551p4vs6j7pjvr@4ax.com>

>Thanks for the link. Reading it makes me wonder if some sort of partial
>static typing would be beneficial, e.g. if the compiler was able to
>determine up front that X is an atomic ...

That's everyone's first thought, however in the generation of
a compiler, several other first steps are necessary to achieve
performance; these steps can be achieved without static typing
or even compiler hints. While optional static declarations are
high on the list of things that can increase performance in a
compiled python executible, there are things which are higher
and have nevertheless not been done. That they haven't means in
my mind that we should stop, and not add (optional) static typing
until and unless they have. Because right now it would be mostly
linguistic masturbation and not really help anything at all.

If you're thinking that the compiler folks haven't made any
progress because strictly there's no static type system, you'd
be wrong.

C//



From othello  Wed Feb 27 00:12:41 2002
From: othello (Raymond Hettinger)
Date: Wed, 27 Feb 2002 00:12:41 -0500
Subject: embedding a jpg in an excel spreadsheet
References: <mailman.1014755052.28857.python-list@python.org>
Message-ID: <a5hpsf$h95$1@bob.news.rcn.net>

"James Strater" <James.Strater at exolink.com> wrote in message
news:mailman.1014755052.28857.python-list at python.org...
> Hello,
>
> I am creating an excel spreadsheet using python win32com.client.  This all
> works fine.
>
> In the first row/cell of the spreadsheet, I want to embed a logo which is
> stored in a a jpg file.  Can anyone suggest how to do this?

from win32com.client.dynamic import Dispatch
xl = Dispatch( 'Excel.Application' )
xl.Range("B5").Select
xl.ActiveSheet.Pictures.Insert("C:\ Pictures\corplogo.jpg")

>
> Also, I need to change the cell background colors on some rows.  Again,
any
> suggestions?

xl.Range("A1").Select
xl.Selection.Interior.ColorIndex = 8

>
> Thanks,
>
> James
>

The easiest way to devine the Excel object model is to turn-on the macro
recorder, perform the steps manually, stop the recorder, edit the macro, and
observe the calls that were made.

Raymond Hettinger




From gcordova at hebmex.com  Mon Feb 18 17:33:32 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Mon, 18 Feb 2002 16:33:32 -0600
Subject: Apology
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AA4@SRM201002>

> 
> Especially when we read your posts over brandies at the weekly
> PSU meetings, while we light our cigars with $100 bills, and
> chuckle about our complete control over the unsuspecting world.
> Mwahahahahaha.
> 

Just a reminder for you that the next meeting



From gmcm at hypernet.com  Tue Feb 12 10:08:43 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 12 Feb 2002 15:08:43 GMT
Subject: Help with Installer.
References: <Xns91B3864EE6013maccetipl@149.156.4.11>
Message-ID: <Xns91B36755D3F49gmcmhypernetcom@199.171.54.214>

Maciej Starzyk wrote:

> I generated a single-directory program with the
> McMillan Installer.
> When I launch the executable from bash, it works OK, 
> however, when run as a shebang line in a shell script it 
> fails to start. It says it cannot open the archive. 
> 
> It does not seem to be fault of the Installer, though.
> 
> It looks like bash cuts off the full path to the 
> executable, when run as a shebang. 

[snip]

> Does anyone know how it can be fixed ?

The only way I know of (and boy, I hope I've
missed something!) is to install the executable
on the user's PATH. It should be able to find
itself, at the expense of some hunting and pecking.

-- Gordon
http://www.mcmillan-inc.com/


From greg at cosc.canterbury.ac.nz  Sun Feb 24 20:55:58 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 25 Feb 2002 14:55:58 +1300
Subject: Time to rename Stackless? (Re: Stackless Platform Independence?)
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com>
Message-ID: <3C7999AE.A7786828@cosc.canterbury.ac.nz>

Seems to me that "Stackless" is a misnomer now. If it
works by switching between multiple stacks, then it now
uses *more* stacks than normal Python!

Maybe something like "Microthreaded Python" or
"Multitasking Python" would be better?

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


From meowbot at meowing.net  Mon Feb 11 20:36:22 2002
From: meowbot at meowing.net (A Meowbot)
Date: Tue, 12 Feb 2002 01:36:22 GMT
Subject: Wow.
Message-ID: <MPG.16d23ba6a6f4eb4989681@news.verizon.net>

In a few weeks of skimming those threads i'd consistently wackyparsed 
the term as "autoconfig." The scary part is that all the flames and 
rebuttals I read made perfect sense in that context.


From mhammond at skippinet.com.au  Fri Feb 15 19:58:53 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 16 Feb 2002 00:58:53 GMT
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com> <3C6D08C1.2000307@skippinet.com.au> <9d4c7023.0202151038.1d3307f@posting.google.com>
Message-ID: <3C6DAF39.5080808@skippinet.com.au>

Sanjay wrote:
> What I am trying to do is something that was somewhat discussed on a
> previous thread, but I'm not sure the thought was fully completed. 
> Specifically, I would like to write an exchange server event sink by
> implementing two com interfaces, IExStoreAsyncEvents and
> IExStoreSyncEvents.  The problem I'm having is that when the python
> event listener is loaded as a com+ application, and I register this
> application for events with exchange, I get the error TypeError ...
> IID not supported.  From looking at the previous threads, it seems as
> this problem is caused by not being able to handle arbitrary vtable
> interfaces, such as the interface methods of IExStoreAsyncEvents and
> IExStoreSyncEvents.  If, however,  the two com classes
> IExStoreAsyncEvents and IExStoreSyncEvents were defined in a pyd, then
> pythoncom would have knowledge of these interfaces, and properly call
> my application.
> 
> I hope I'm explaining the problem clearly. I am a little confused as
> to how to get around this with pythoncom.

You need t6o check out the win32com\servers\test_pycomtest.py.  This 
shows how arbitary vtables must be loaded by your code before it is 
recognized.  This in-turn implies that the interface is described in a 
type library.

Mark.



From jonathan at onegoodidea.com  Sat Feb  2 04:38:53 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Sat, 02 Feb 2002 09:38:53 +0000
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com> <B88081C5.40F9%jonathan@onegoodidea.com> <a3ell7$29bm$1@nntp6.u.washington.edu> <B8809EB2.411E%jonathan@onegoodidea.com> <a3fisr$9j6$0@216.39.172.122>
Message-ID: <B881642D.41AD%jonathan@onegoodidea.com>

On 2/2/2002 2:28, in article a3fisr$9j6$0 at 216.39.172.122, "Bengt Richter"
<bokr at oz.net> wrote:

> Rule 17 says maybe the above would work if you *don't* setuid it, but I don't
> think
> you need it, unless maybe mod Python could otherwise get inbetween? I haven't
> set that
> up myself.

Yes, perhaps I should have made that more clear. I meant using 'suexec'
*instead* of randomly setuiding executables. With mod_python it's even
harder as the Apache process itself needs to be running as the user
(requires the new 2.0 per-user pooling magic), but I don't think that's the
problem.

'suexec' is a much more secure mechanism for running cgis as a particular
user. The user is set in the config file (per virtual host) and suexec has a
number of rules that it checks before it will run the script as that user
(as Bengt refers to) - such as that it is owned by the user, isn't writeable
by anyone else, is within a particular cgi directory (this is the main
problem with suexec as it is only one path per installation, so I've had to
set it to something useless like '/' in the past to make it work with
disparate cgi directories).

It would be much better all round if you can get the hosters to use this
rather than setuiding your own scripts.

Jonathan



From justin at iago.org  Tue Feb 26 13:54:09 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 26 Feb 2002 13:54:09 -0500
Subject: grammar question
In-Reply-To: <B8A0F6F8.3C3B%pereira@cis.upenn.edu> (Fernando Pereira's
 message of "Tue, 26 Feb 2002 08:25:28 -0500")
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu>
	<m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de>
	<B8A0F6F8.3C3B%pereira@cis.upenn.edu>
Message-ID: <x7bseccbvy.fsf@ra.iago.org>

Fernando Pereira <pereira at cis.upenn.edu> writes:

> In general, languages with infix operators and complex nested
> structures like lists are not LL(k), because a parser can only
> decide the role of a subexpression after it sees the operator or
> terminator that follows it.

One can usually (but not always) create an LL(k) grammar for such
languages.  It just requires some interesting manipulation of the
grammar productions such as left factoring and removal of left
recursion.

This is a pain to do, but once you have done it you can use a nice
recursive-descent LL(1) parser.

Now, you might not find it worthwhile to do those transformations.
That's your business, and I won't try to make that decision for you.

-Justin

 




From calves at coelce.com.br  Thu Feb 21 10:42:15 2002
From: calves at coelce.com.br (Alves, Carlos Alberto - Coelce)
Date: Thu, 21 Feb 2002 12:42:15 -0300
Subject: While loop and time.sleep function
Message-ID: <29A97D00F387D411AC7900902770E14804047A4D@lcoeexc01.coelce.net>

Hi all,
I tray to make a simple clock program ( see the code below ).
Can someone explain to me why code 01 works while code 02 doesn't
----------------------------------------------------------------------------
---------------------
Code 01
----------------------------------------------------------------------------
---------------------from Tkinter import *
import time

class Clock:
	def __init__(self,root):
	
self.lb=Label(root,padx=5,pady=5,fg='blue',font=('Times',20))
		self.lb.pack()
		self.ck()
		self.update()
	def ck(self):
		self.lb.configure(text=time.asctime()[11:19])
	def update(self):
		while 1:
			self.ck()
			print "",
if __name__=='__main__':
	root=Tk()
	root.title('Clock')
	Clock(root)
----------------------------------------------------------------------------
---------------------
Code 02
----------------------------------------------------------------------------
---------------------
from Tkinter import *
import time

class Clock:
	def __init__(self,root):
	
self.lb=Label(root,padx=5,pady=5,fg='blue',font=('Times',20))
		self.lb.pack()
		self.ck()
		self.update()
	def ck(self):
		self.lb.configure(text=time.asctime()[11:19])
	def update(self):
		while 1:
			self.ck()
			time.sleep(1)
if __name__=='__main__':
	root=Tk()
	root.title('Clock')
	Clock(root)

Carlos Alberto
COELCE/DPRON-Departamento de Projetos e Obras Norte
Fone: 677- 2228
e-mail: calves at coelce.com.br
\|||/
(o o)
--ooo0-(_)-0ooo--

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020221/add22039/attachment.html>

From ilgattodelvicino at hotmail.com  Fri Feb 22 09:43:52 2002
From: ilgattodelvicino at hotmail.com (ilgattodelvicino)
Date: 22 Feb 2002 06:43:52 -0800
Subject: FILEDSN
Message-ID: <c13db1aa.0202220643.37578e1c@posting.google.com>

Can i use a file DSN with odbc module??

thanks


From martin at v.loewis.de  Tue Feb 26 16:25:12 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 22:25:12 +0100
Subject: Building 2.2 with the Intel icc compiler
References: <9gSe8.18435$ZC3.1526281@newsread2.prod.itd.earthlink.net>
Message-ID: <m3it8kue9z.fsf@mira.informatik.hu-berlin.de>

"Ted Drain" <teddrain at earthlink.net> writes:

> Has anyone built Python 2.2 with the new Intel C/C++ compiler icc?  I
> configured Python and then edited the Makefile by hand to use icc and icc
> options.  Python builds but the modules all report errors (and then remove
> themselves).

People have successfully done so. Can you share a few of the errors it
reports?

Regards,
Martin



From gimbo at ftech.net  Mon Feb 25 12:47:38 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Mon, 25 Feb 2002 17:47:38 +0000
Subject: Iterators and generators for the masses
In-Reply-To: <5e8bd451.0202250924.70f47df1@posting.google.com>
References: <5e8bd451.0202250924.70f47df1@posting.google.com>
Message-ID: <20020225174738.GB1796@andy.tynant.ftech.net>

> A note in my Python 2.2 folder says that:
> 
> Iterators (PEP 234) and generators (PEP 255) were added.
> 
> When I ask questions on the tutor list and also this list i get
> answers that sometimes start:

[snip]

This is a pretty good article:

http://www-106.ibm.com/developerworks/linux/library/l-pycon?t=grl,l=252,p=iterators

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From zayats at red.seas.upenn.edu  Wed Feb 13 10:28:37 2002
From: zayats at red.seas.upenn.edu (Salim Zayat)
Date: 13 Feb 2002 15:28:37 GMT
Subject: Canvas postscript issues
Message-ID: <a4e0n5$idn$1@netnews.upenn.edu>

Hey there.  Is there a way for the canvas.postscript method to produce a 
postscript with arabic text involved?  I've tried messing around with the 
fontmap option, but I keep getting crap in the file.

Thanks!

Salim


From jason at jorendorff.com  Thu Feb 21 14:11:43 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 21 Feb 2002 13:11:43 -0600
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <a51cj9$9m1$1@bob.news.rcn.net>
Message-ID: <HFEKILOLEFEFMKAECNDLIEGGDEAA.jason@jorendorff.com>

David Abrahams wrote:
> Craig Maloney wrote:
> > As for the CERN Root interpereter (CINT)
> > http://root.cern.ch/root/Cint.html -- which I had already heard about
> > from my friends in the know in the particle physics community -- I
> > would assume that many here also harbor the same skeptecism toward the
> > C++ interpereter as they do toward the C++ parsing systems?
>
> Look at the site documentation. They only claim about 85% coverage of C++.

Cint limitations:
  http://root.cern.ch/root/Cint.phtml?limitations

An interesting read.  Apparently their parser isn't very
good.  (I would think just parsing the code would be the
easy part.  Shows what I know.)

## Jason Orendorff    http://www.jorendorff.com/




From husam at macrozon.net  Thu Feb 21 03:06:56 2002
From: husam at macrozon.net (husam)
Date: Thu, 21 Feb 2002 09:06:56 +0100
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com>
Message-ID: <3C74AAA0.D3A050EE@macrozon.net>

Jeff, thanks alot for the enlightening remarks. I will try them. By the way,
"glob.glob('*pdb') does read the files properly, but when I pass each of the to
the other program for processing it doesn't work. I only work with files red
with "popen".  A side problem when using "popen('ls *pdb')" I get an error
message saying that "ls" got to many arguments. I got the same error in the
shell. So I worked around this problem by copying all the *pdb file to a
separate directory.




From ahimsa at inext.co.za  Tue Feb 12 00:24:44 2002
From: ahimsa at inext.co.za (ahimsa consulting)
Date: Tue, 12 Feb 2002 07:24:44 +0200
Subject: Possible dumb query
Message-ID: <000801c1b385$97747aa0$0100000a@ahimsa>

Greeting Pythonites

This might qualify as one of those "really dumb questions", but:

Under Linux RH 7.2 when doing a search for which python ("$ which python") I
have discovered that I have one entry for python as /usr/bin/python and one
entry for python2.2 as /usr/bin/python2.2. My question is this: if I were to
delete the directory /usr/bin/python what impact - if any - will it have on
any of the libraries/ etc. that I would require to use python2.2

Once again, my apologies if this is a dumb query, but gotta learn somehow.

Much obliged

Andrew




From ykingma at accessforall.nl  Mon Feb  4 14:12:38 2002
From: ykingma at accessforall.nl (Ype Kingma)
Date: Mon, 04 Feb 2002 20:12:38 +0100
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012761186.28271.python-list@python.org>
Message-ID: <3C5EDD21.6DC9923E@accessforall.nl>

Jason,

you wrote:
> 
<snip>
> 
> I guess I imagine that Python's speed issues are:
> 
>  1. Ultra-late binding of, and consequent repetitive lookups of,
>     global names, attributes, methods, module members, etc.
>     (Of course, unless names like 'int' and 'float' can be bound
>     at compile time, optional type declarations are unhelpful.)
> 
>     Since many people like this about Python, and many have used it
>     in existing code, effort is focused on making lookups cheaper
>     (or caching the results) rather than changing the binding
>     behavior to require fewer lookups overall.
> 
>  2. Nothing ever gets compiled to anything even close to
>     machine code.  This is largely because the Python development
>     team doesn't have the resources to invest in it.  Clearly
>     there's something to gain by moving the huge ceval.c loop/switch
>     from runtime to compile time.  No one knows how much.  Clearly
>     there's something to lose from the resulting explosion in the
>     complexity of the compiler.
> 

Please note that jython is compiled directly to Java opcodes.
Anything that makes interpreting these faster (JIT, native compilation)
will make jython faster.

Regards,
Ype


From eppstein at ics.uci.edu  Thu Feb 28 16:10:25 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 13:10:25 -0800
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz> <3C7CFAF8.1080108@mxm.dk> <Xgaf8.10287$pC1.1854469@newssrv26.news.prodigy.com> <3C7E8D47.21BE30B3@ccvcorp.com>
Message-ID: <eppstein-3B013E.13102528022002@news.service.uci.edu>

In article <3C7E8D47.21BE30B3 at ccvcorp.com>,
 Jeff Shannon <jeff at ccvcorp.com> wrote:

> No, it's not something that people often use.  Generally, if you want 
> dict-like behavior, you just use a dict.  The times that it's worth the 
> effort to fake the behavior with something else, are pretty unusual.  
> The speed difference between lists and dicts isn't all that significant, 
> anyhow, and which one has the advantage depends on how you're using it.  
> Python dicts are pretty amazingly nifty, really; there's no reason to be 
> shy about using them.

I agree.  I had been looking at some code where I needed two-dimensional 
arrays, which I was coding as lists of lists, and it was getting pretty 
ugly initializing the list of lists so that each list is distinct --
[[None]*x]*y doesn't work for reasons I'll leave as an exercise.

Finally, I realized I could just use a dict.  No tricky initialization, and 
the syntax for double indexing dict[x,y] is even slightly nicer than 
list_of_lists[x][y].  Unlike a list of lists, it's easy to grow the range 
of indices: just start using them.  It would be good if dict.has_key(x,y) 
worked without having to double the parentheses, but that's only a minor 
distraction.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From zayats at blue.seas.upenn.edu  Fri Feb  8 11:34:33 2002
From: zayats at blue.seas.upenn.edu (Salim Zayat)
Date: 8 Feb 2002 16:34:33 GMT
Subject: Label width
Message-ID: <a40ump$d9c$1@netnews.upenn.edu>

Is there a way to determine the width and height of a Label widget (in 
pixels) before it is actually drawn to screen?

Thanks

Salim


From phr-n2002a at nightsong.com  Wed Feb 20 20:54:53 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 17:54:53 -0800
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014252214.26682.python-list@python.org>
Message-ID: <7xlmdnsioi.fsf@ruckus.brouhaha.com>

Gustavo Cordova <gcordova at hebmex.com> writes:
> Now *THIS*:
> >    while (m := re.match(pat,x)):  blah()
> Indeed, this is very readable. The ":=" asignment operator
> is used in Pascal, if I remember correctly; another language
> where you can't assign inside an expression.

OK.  It still blurs the statement/expression distinction, but now
there's no longer anything wrong with it?

> But it only feels contorted, to you. To me it feels quite
> natural:
> 
> 	1. Set up the loop conditions and variables.
> 	2. Do the loop.
> 	3. Exit the loop.

What you and I see as contorted are not necessarily the same.
The language should not impose a choice on either of us over
something like this.

> Yes, your ":=" suggestion is valid, and it removes the "=/=="
> readability / typo reason. But if it's not there, and if
> the people maintaining your code are going to get all huffy
> because it's "spelled out" instead of being in a compact
> form, then the problem is another, and not the language.

I'm saying the =/== objection is bogus because it can be fixed
easily with := or something similar.  With =/== fixed, is there
any remaining objection?

> Then again, maybe all the newbies used other languages
> which feature the assignment / expression duality. Also,
> this "limitation" (of not allowing assignments in expressions)
> is clearly expressed in the docs, which means that newbies
> nowadays are simply NOT reading the documentation before
> launching into new waters. That severely downgrades the
> quality of the questions they might pose, no?

That's silly.  If the language for some reason didn't recognize the
digit 3 (you could still say x=2+1 if you wanted to assign 3 to x) and
the restriction was documented, people would still ask why there was
no 3 in the language.  Saying the questioners should just read the
documentation is unproductive.  They come to the language with the
expectation that if you can say x=2 or x=4, then you should also be
able to say x=3.  If enough people keep asking the same question,
maybe the expectation is ok and it's the language (and documentation)
that need fixing.

> Just count how many times the "static vs dynamic typing" thread has
> reappeared, it just won't die. And it's always *another* newbie
> which comes in and asks the same question, instead of learning,
> slowly, breathing calmly, how to best use this shiny new tool.

That's a somewhat different situation which needs a deeper type of
answer.  

> I'm sure that it'll keep evolving, it needs to. But there's
> something inherently *nice* which comes into play when
> the code's all spelled out, doing the clearest thing
> possible, which just isn't there in Lisp, C, C++, etc.
> 
> Anyway, it's just my opinion. :-)

Yes, it's subjective--I personally prefer concise code, in that I
often prefer using an assignment or embedded conditional inside an
expression rather than using several statements, but I don't claim
it's objectively better than doing it the other way.  

More than anything else, readability is a matter of opening your mind
to other programmers through the code that you write, not really a
technical matter but almost a spiritual one.  The best programming
language is the one you're not even aware that you're using--your code
just naturally reflects your thoughts.  Python does a pretty good job
in most respects of not getting in the way of expressing what you're
thinking, but it has certain shortcomings, mostly from when it tries
to impose its own notions of readability on programmers who know what
they're doing.  


From logiplexsoftware at earthlink.net  Mon Feb 25 18:28:31 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 25 Feb 2002 15:28:31 -0800
Subject: 15 minute presentation on python?
In-Reply-To: <Xns91C16A332ADFCandy47halfcookedcom@203.109.252.31>
References: <3c7ab974_4@corp.newsgroups.com>
	<a5ef3h$6p6g1$1@ID-11957.news.dfncis.de>
	<Xns91C16A332ADFCandy47halfcookedcom@203.109.252.31>
Message-ID: <20020225152831.75ea23bf.logiplexsoftware@earthlink.net>

On Mon, 25 Feb 2002 23:27:13 +0000 (UTC)
Andy Todd wrote:

> With a nod to the great man, how about borrowing Mark Pilgrim's roman 
> numeral converter from Dive into Python  (http://diveintopython.org/). If

> you want to get really ambitious you could add an XML-RPC or SOAP call to
a 
> remote server and convert the stock price of a local corporation or the 
> some local weather information (temperature usually) to roman numerals on

> the fly.

And you'll have a useful utility when you're finished <wink>

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From emile at fenx.com  Wed Feb  6 21:46:51 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 6 Feb 2002 18:46:51 -0800
Subject: Is there a Python module that already does this?
References: <a3rbad$19u08k$1@ID-98166.news.dfncis.de> <a3rnd1$1a44oq$1@ID-11957.news.dfncis.de> <eppstein-95B064.09445206022002@news.service.uci.edu> <eppstein-80FFB2.17393106022002@news.service.uci.edu>
Message-ID: <a3sq4m$1an2ll$1@ID-11957.news.dfncis.de>

"David Eppstein" <eppstein at ics.uci.edu> wrote in message
news:eppstein-80FFB2.17393106022002 at news.service.uci.edu...
> Here it is in a functional programming style:
>
> import operator
>
> def flatten(x):
>         if operator.isSequenceType(x) and (len(x) != 1 or x[0] != x):
>                 return reduce(operator.concat, map(flatten, x), [])
>         else:
>                 return [x]
>
> print flatten(["cat",5,['dog',[3,3,1]],"zoo"])
>
> ...outputs...
> ['c', 'a', 't', 5, 'd', 'o', 'g', 3, 3, 1, 'z', 'o', 'o']
>
> All that list copying makes it somewhat inefficient, though...

Why do you say that?  This looks pretty good to me.

You can squeeze out another (ugly) 15-20% performance gain
by making almost everything local and taking out the len check:

def flatten(x, isseq=operator.isSequenceType,
    concat=operator.concat, reduce=reduce, map=map):
    if isseq(x) and x[0] != x:
        return reduce(concat, map(flatten, x), [])
    else:
        return [x]

And when you make everything local, it's ugly beyond benefit.  ;-)

--

Emile van Sebille
emile at fenx.com

---------



From spencer at spencerdoidge.com  Sun Feb  3 01:45:54 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sat, 2 Feb 2002 22:45:54 -0800
Subject: Win32 comport extensions
Message-ID: <a3im40$rjt$1@news.efn.org>

In the win32 extensions script win32comport_demo.py, line 102 reads:
rc, data = ReadFile(self.handle, comstat.cbInQue, overlapped)

I don't see the five arguments I'm used to associating with Win32 API's
ReadFile(). The nBytesToRead and lpNumBytesRead args are not there.

I want to use nBytesToRead and the COMMTIMEOUTS parameters to receive
packets, as coded at line 194 of
http://www.spencerdoidge.com/code/rs232.c
In other words, I want my receiving routine to use the COMMTIMEOUTS,
nBytesToRead, and lpNumBytesRead to determine when ReadFile() passes the
hOverlapped signal (osReader.hEvent  READ2). Is there a way to do this with
these win32 extensions?
Do I need to create my own extension to accomplish this result?

Spencer Doidge






From gbreed at cix.compulink.co.uk  Wed Feb 20 12:12:36 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Wed, 20 Feb 2002 17:12:36 +0000 (UTC)
Subject: Adding static typing to Python
References: <24c39b2c.0202191901.5f6643f6@posting.google.com>
Message-ID: <a50le4$jmk$1@thorium.cix.co.uk>

Alexander Jerusalem wrote:

> I just looked up what CLOS does. I don't know if you call that static
> typing or not. In any case, CLOS does have type declarations in the
> source code which is what I want and what enables the features I'd
> like to see in Python. For example in CLOS you can have the following
> method declarations:
> 
>  (defmethod Area ((C Circle)) (* pi (Square (Radius C)))
>            (defmethod Area ((R Rectangle)) (* (Width R) (Height R)))
>            (defmethod Area ((Sq Square)) (Square (Width Sq)))
>            OR (defmethod Height ((Sq Square)) (Width Sq))

But you can do that in Python:

>>> def area(shape):
...   if isinstance(shape, Circle):
...     return math.pi*shape.radius**2
...   if isinstance(shape, Rectangle):
...     return shape.width*shape.height
...   if isinstance(shape, Square):
...     return shape.width**2
...   raise TypeError, "unknown shape"
...

It can't take much imagination to see how it could be sugared to work with 
multiple functions with types resolved at run time.  As for strict, 
strong, static, explicit typing (which may be a straw man) it looks *less* 
suitable for this example.

Say you want to include a triangle in your list of shapes.  So, you add

...   if isinstance(shape, Triangle):
...     return shape.base*shape.height

to the area function.  But then, will all the other functions that call 
area be allowed to accept a triangle?  If they're defined to accept either 
circles, rectangles or squares, you'll have to change each one's interface 
to allow triangles as well.

You could make them all subclasses of Shape.  In which case, you have 
functions defined to use a polymorphic object that depend on function 
overloading by the real type.  Not all statically typed languages will 
allow that.  If they do, you have the problem that you might have added 
triangles for some other reason, and forgotten to cater for them in 
area().  You aren't getting static type checking at all.

Wouldn't it be better to make area a method of the shapes?  And for static 
typing, make it abstract in the Shape class?


                     Graham

         (http://www.microtonal.co.uk/)


From thomas.heller at ion-tof.com  Thu Feb 14 08:27:29 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 14 Feb 2002 14:27:29 +0100
Subject: Problem freezing Win32 services
References: <a4g3s3$k0s$1@news.ost.eltele.no> <a4g5sl$f89$1@ID-59885.news.dfncis.de> <a4gci5$pbh$1@news.ost.eltele.no>
Message-ID: <a4ge0d$2fhf$1@ID-59885.news.dfncis.de>

> > > I'm attempting to freeze a Windows NT service (in Win2000) written in
> Python
> > > 2.2 using win32all 142 and the freeze util bundled with Python in the
> tools
> > > directory, and I've encountered some problems. Freezing console apps
> works
> > > fine.
> > <shameless plug>
> > You could (should ;-) try out the newest py2exe release, which also can
> > create service exe-files: http://starship.python.net/crew/theller/py2exe/
>
> I tried py2exe and it seems to work better at least :-) I mananged to create
> a service executable that can be started either from the Windows Services
> GUI or by 'net start myservice'.

Great! The first step is done.

>  However if I try to just run the service
> executable I get:
>
> Starting service - this may take several seconds - please wait...
> Could not start the service - error 997

This is more or less by design. Don't do this - use one of the command
line options when you start it from the command line.

>
> The real problem, however, starts when I move the executable and associated
> pyds and dlls to another computer. I then get the following error in the
> event viewer:
>
> "The description for Event ID ( 3 ) in Source ( myservice ) cannot be found.
> The local computer may not have the necessary registry information or
> message DLL files to display messages from a remote computer. The following
> information is part of the event:  ; e; N."
>
> I suspect I may be missing some imports. py2exe gave me some warnings,
> however I've tried to find out which it is with no luck. Is there any way to
> get more descriptive errors (tracebacks?)? Running the service with
> the -debug option outputs "Error 0xC0000003 - The instance's SvcRun() method
> failed" and  " e: N" . Any suggestions?
>
There is still a bug that tracebacks are not printed to the event
log correctly, will be fixed in the next release.

I think we should try to sort this out privately, please send me your
email address. Which Python version do you use? I could send you
some patches...

Thomas




From jason at jorendorff.com  Sat Feb 16 10:12:55 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 16 Feb 2002 09:12:55 -0600
Subject: using lists in C-extensions?
In-Reply-To: <3c6e711a.9871609@news.easynews.net>
Message-ID: <HFEKILOLEFEFMKAECNDLEEAHDDAA.jason@jorendorff.com>

Marcus Stojek wrote:
> But I am still far too slow. So I have to go for C.
> My problem is, that I have to pass a list (or actually
> a couple of lists) to the C routine and after performing
> the calculations, returning them to Python.

It might be a good idea to use Numeric, and
pass a Numeric array into your function.

It will be nice and fast, since the data is stored as
a flat array.  (Otherwise, you get to trawl all over
memory to get the numbers out of that list of tuples.)

> As far as I understand things, I have to use
> PyArg_Parse() and Py_BuildValue() for this, right?

You probably only need PyArg_ParseTuple().
  http://www.python.org/doc/current/ext/parseTuple.html

You're right, there's no format for lists.
You want "O" or "O!".

  PyObject *myArray;

  if (!PyArg_ParseTuple(args, "O", &myArray))
      /* error, bad arguments */ ...
  if (!PyArray_Check(myArray))
      /* error, wrong type */ ...
  ...

If you really want to use a list of tuples, something
similar will work.

The ArrayObject C API is covered in Chapter 13 of
the NumPy manual:
  [PDF] http://www.pfdubois.com/numpy/numpy.pdf
The documentation seems kind of thin to me, but
basically you care about:
  PyArray_ContiguousFromObject(...)  // get contiguous data
  myArray->data          // the actual data
  PyArray_Size(myArray)  // how much of it there is
  myArray->descr->type   // what type it is
As always, mind your INCREFs and DECREFs.

And though you may already know all this, Alex Martelli's
sample code really helped me to pull it all together.
If this is your first C extension, take a look.
  http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/66509

## Jason Orendorff    http://www.jorendorff.com/



From nospam at sebsauvage.net  Fri Feb  1 04:39:45 2002
From: nospam at sebsauvage.net (Sébastien SAUVAGE)
Date: 01 Feb 2002 09:39:45 GMT
Subject: Python like the old-time batch files.
References: <a3bhcf$176195$1@ID-98166.news.dfncis.de> <a3bk4b$11u4$1@norfair.nerim.net>
Message-ID: <Xns91A86D0EAA121sebsauvagedotnet@127.0.0.1>

"Gillou" <nospam at bigfoot.com> wrote in news:a3bk4b$11u4$1
@norfair.nerim.net:

> Not in the standard distro for win32
> Unix have a curses package for this.
> You can get a curses like package for windows from there...
> 
> http://effbot.org/efflib/console/

Yurk !
I get a 404 on the download page 
(http://effbot.org/efflib/console/downloads/)

Any alternate URLs ?

(I'd really love to have a curses-compatible module for Windows).

-- 
S?bastien SAUVAGE
sebsauvage at sebsauvage dot net
http://sebsauvage.net


From magnus at trhinkware.se  Thu Feb 21 06:46:18 2002
From: magnus at trhinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Thu, 21 Feb 2002 12:46:18 +0100
Subject: Python vs. C/C++/Java: quantitative data ?
References: <3c74a81c$0$141$bb624dac@diablo.uninet.ee>
Message-ID: <3C74DE0A.2020700@trhinkware.se>

Eduard Smirnov wrote:

  > I'm writing a diploma paper on the difference between scripting
  > (e.g. in Python or Tcl) and system programming (e.g. in C, C++ or
  > Java), and one of the claims I want to make is that you can often
  > develop applications 10x or more faster in a scripting language.

I wouldn't really say that Python is a scripting language though. The
name scripting language suggests a language suited for small snippets.
Python is as much a general language for large scale software
development as C++ is. But Python is surely a rapid development
language.

Bash, awk, Tcl and Perl are in my opinion scripting languages, but
not Python. (Well, Perl probably has the theoretical requirements for
developing large systems, but it's fairly obscure and bolted on as an
afterthought.)

  > I apologue that more applications can be written or re-written in
  > scripting language faster and with more complex functionality then
  > in system programming language. If you have any experience in
  > writing applications in system programming language (e.g. in C,
  > C++) and re-writing it in scripting language (e.g. Python) or vice
  > versa, please compare your work in both implementations,

Actually, I think many comparisions are made like this, and a problem
is that if you first write in C, and then translate, you are likely to
use your C design for Python, and end up with a program that is almost
as big, but much slower. If you go the other way, there is no way you
can use your Python solution directly, since C can't handle the level
of abstraction, and you will probably get a fairer comparision.

  > describe your work/application's functionality, approximate number
  > of lines and time of development of each implementation, skill's
  > level needed. Any more comments about are welcome.

Personally, my Python applications worked well, thank you, so I
never rewrote them in any other language...

There are some links at
http://www.thinkware.se/cgi-bin/thinki.cgi/PythonQuotes
which might guide you do relevant sources. (Glyph Lefkowitz,
Greg Stein and Eric S. Raymond for instance.)

Also, have a look at
http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprt_computer2000.pdf
For this experiment, the productivity gain with Python is roughly
factor three, but this is also

There is a comparision at http://www.bagley.org/~doug/shootout/
but it's really just a pointless play. Don't fall for this one.
It has two fundamental flaws: It only uses small programs, so
it doesn't show you anything about how the language scales. If you
believed in that, do all your coding in C and bash. It also ignores
one of greatest assets of Python, its rich standard library.

So instead of writing the "random generator" as

import random
print random.random()

They reinvent the wheel like this:

import sys

IM = 139968
IA = 3877
IC = 29573

LAST = 42
def gen_random(max):
      global LAST
      LAST = (LAST * IA + IC) % IM
      return( (max * LAST) / IM )

def main():
      N = int(sys.argv[1])
      if N < 1:
          N = 1
      gr = gen_random
      for i in xrange(1,N):
          gr(100.0)
      print "%.9f" % gr(100.0)

main()

In other words, many of the programs are about as long
as C programs since they "level the playfield" by not
using the "included batteries". On the other hand Python
always show more CPU and memory usage since it loads the
interpreter and a lot of modules before it runs the those
twenty lines of code. No wonder that C seems slim and fast
there. If we talked about a REAL application, and compared
50 000 lines of C++ with 10 000 lines of Python code, things
would look very different.








From jeff at ccvcorp.com  Wed Feb 27 15:01:48 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 12:01:48 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com> <7xk7sznsoa.fsf@ruckus.brouhaha.com>
Message-ID: <3C7D3B2C.80A97BA2@ccvcorp.com>


Paul Rubin wrote:

> Also, there's no way to make sure that the "del temp" at the end of
> a block actually gets run, unless you put try...finally around it,
> which is a big mess.

I may be in way out of my depth here, but I don't think that this is true.  If
you explicitly delete an object, then that name should be guaranteed to be
unbound.  What's *not* guaranteed, is whether or not an unreferenced object's
__del__() method gets called -- the object may still exist in memory,
unreachable, somewhere.  However, that name is *not* going to reference it
properly.

At least, that's my understanding of it.

Jeff Shannon
Technician/Programmer
Credit International




From steve at ferg.org  Thu Feb 14 10:50:52 2002
From: steve at ferg.org (Stephen Ferg)
Date: 14 Feb 2002 07:50:52 -0800
Subject: ANN: EasyGui - basic, easy-to-use GUI functions
References: <b16e4ef7.0202131021.2b105a66@posting.google.com> <slrna6lbrp.guj.mlh@vier.idi.ntnu.no> <d7d5ebdf.0202131429.16a21cca@posting.google.com> <915a998f.0202140105.1df2e70@posting.google.com>
Message-ID: <b16e4ef7.0202140750.62660f96@posting.google.com>

My apologies for the typo!  I'll fix it in the EasyGui documentation tonight!
-- Steve


From fperez528 at yahoo.com  Mon Feb  4 08:19:19 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 04 Feb 2002 13:19:19 +0000
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com>
Message-ID: <a3mqee$5sg$1@peabody.colorado.edu>

Marc Laymon wrote:

> I am just learning python.  I have written a few simple scripts which
> run fine when invoked using "python myscript.py".  The Python manual
> says you can execute a Python script directly In Unix by making the
> first line "#!/usr/bin/python", then giving the command
> "./myscript.py".  However, when I try this, I get an error "Command
> not found.".  I get the same error under both Linux and Solaris.
> Using the same format for perl works fine.  Why doesn't the
> "#!/usr/bin/python" line work for Python ?  (perl and python are both
> in /usr/bin on my Linux system.)


$ chmod +x myscript.py

It's not the shebang failing, it's that your script isn't executable (most 
likely). To be safer, use

#!/usr/bin/env python

This will look for python in all your path, and will find it even if it's not 
located in /usr/bin for whatever reason.

cheers,

f.


From logiplexsoftware at earthlink.net  Mon Feb 11 14:33:50 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Mon, 11 Feb 2002 11:33:50 -0800
Subject: Fear of autocoding! >Re: Please stop responding to Rue
In-Reply-To: <3c681c43.13768618@127.0.0.1>
References: <b03e80d.0202060227.368f6ea7@posting.google.com>
	<b34d3835.0202080025.1d69d198@posting.google.com>
	<7934d084.0202101718.404404d1@posting.google.com>
	<7643.807T2626T856092threeseas@earthlink.net>
	<b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com>
	<3c681c43.13768618@127.0.0.1>
Message-ID: <20020211113350.4cd227b1.logiplexsoftware@earthlink.net>

On Mon, 11 Feb 2002 19:29:49 GMT
(Five Fresh) Fish wrote:

> Perhaps one metric is to count the number of people who are telling you,
> bluntly, that you're a loon.  It just could be that they're right.  :-)
> 
> (The delusional folk, of course, won't recognize that they're being
called
> a loon...)

What people accept as "reality" is often simply a consensus.  The flip-side
of this though, is that most of the great ideas of the past have been from
people who broke with this consensus (and by extension "reality").

"The reasonable man adapts to the world around him. The unreasonable man
expects the world to adapt to him. Therefore, all progress is made by
unreasonable men."
-- George Bernard Shaw

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From DeepBleu at DeepBleu.org  Thu Feb 14 09:30:41 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Thu, 14 Feb 2002 08:30:41 -0600
Subject: zlib.so and libz.so and Python
Message-ID: <u6niisr909gtd6@corp.supernews.com>

Python comes with zlib.so (for Sun OS).  But I noticed that on the same
machine, there is also libz.so
Can someone please tell me what is the difference between these two and what
libz.so is used for?  I am trying to troubleshoot a little problem and I
need to be certain that libz.so may or may not be used to work with ZIP
files along with Python.  Also, I want to find out if zlib.so (the one that
comes with the Python distribution) is all we need to work with ZIP files in
Python.
Thanks,
DeepBleu




From loewis at informatik.hu-berlin.de  Thu Feb 28 09:09:23 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 15:09:23 +0100
Subject: PEP 263 comments
References: <mailman.1014762060.30054.python-list@python.org> <m3u1s31eah.fsf@mira.informatik.hu-berlin.de> <87g03lhm3p.fsf@tleepslib.sk.tsukuba.ac.jp>
Message-ID: <j4y9hdg0ks.fsf@informatik.hu-berlin.de>

"Stephen J. Turnbull" <stephen at xemacs.org> writes:

> Hi, I'm Steve Turnbull, I do XEmacs.  Mostly Mule.  Barry asked me to
> step up to bat on this.

Thanks for your comments!

> You don't.  From now on, anything that goes into the official Python
> sources is in UTF-8.  Convert any existing stuff at your leisure.
> This is recommended practice for 3rd party projects, too.  People can
> do want they want with their own stuff, but they are on notice that if
> it screws up it's their problem.

It's worse than this: under the proposed change, Python would refuse
to accept source code if it is not UTF-8 encoded. In turn, code that
has a euc-jp comment in it and is now happily accepted as source code
in the current Python programming language would be rejected.

This is like mandating that all Emacs-Lisp files are UTF-8, whether
they are part of the Emacs sources, or installed somewhere out there
in the wild.

> XEmacs actually did this (half-way) three years ago.  I convinced
> Steve Baur to convert everything in the XEmacs CVS repository that
> wasn't ISO 8859/1 to ISO-2022-JP (basically, start in ASCII, all other
> character sets must designate to G0 or G1, and get invoked to GL; at
> newlines, return to ASCII by designation; the "JP" part is really a
> misnomer, it's fully multilingual).  Presto! no more accidental Mule
> corruption in the repository.

This is a different issue: We are not discussing the encoding that the
Python sources use in the Python CVS tree, we are discussing the
encoding that Python source code uses.

> Oh, and does Python have message catalogs and stuff like that?  Do you
> really want people doing multilingual work like translation mucking
> about with random coding systems and error-prone coding cookies?
> UTF-8 detection is much easier than detecting that an iso-8859-1
> cookie should really be iso-8859-15 (a reverse Turing test).

Python supports gettext, but this is still a different issue. The
Unicode type of Python is precisely that - it is not that Python would
support different wide character implementations internally. Again,
the issue is how source code is encoded.

> So much for the alleged "backward compatibility" non-issue.  :-)
> People are abusing implementation dependencies; Just Say No.

A very radical opinion :-) but I get the feeling you might be missing
the point in question ...

>     Martin> Will you reject a source module just because it contains a
>     Martin> latin-1 comment?
> 
> That depends.  Somebody is going to run it through the converter; it's
> just a question of whether it's me, or the submitter.  

'you' in this case isn't the maintainer of a software package; it is
the Python source code parser...

> GNU Emacs supports your coding system cookies.  XEmacs currently
> doesn't, but we will, I already figured out what the change is and
> told Barry OK.  And I plan to add cookie-checking to my latin-unity
> package (which undoes the Mule screwage that says Latin-1 NO-BREAK
> SPACE != Latin-2 NO-BREAK SPACE).  Other editors can do something
> similar.

I assume you are talking about the -*- coding: foo -*- stuff here?
*This* is the issue in question. Should we allow it, or should we
mandate that all Python source code (not just the one in the Python
CVS) is UTF-8?

> So people who insist on using a national coded character set in their
> editor use cookies.  Then the python-dev crew prepares a couple of
> trivial scripts which munge sources from UTF-8 to national codeset +
> cookie, and back (note you have to strip the cookie on the way back),
> for the sake of people whose editor's Python-mode doesn't grok cookies.

Again, not the issue: Most people run Python programs without ever
submitting them to python-dev :-)

You may wonder why Python (the programming language) needs to worry
about the encoding at all. The reason is that we allow Unicode
literals, in the form

   u"text"

The question is what is the encoding of "text", on disk. In memory, it
will be 2-byte Unicode, so the interpreter needs to convert. To do
that, it must know what the encoding is, on disk. The choices are
using either UTF-8, or allowing encoding cookies.

> My apologies for the flood.  I've been thinking about exactly this
> kind of transition for XEmacs for about 5 years now, this compresses
> all of that into a few dozen lines....

I'm not sure whether a similar issue exists in XEmacs: the encoding of
ELisp would be closest, but only if the Lisp interpreter ever needs to
worry about that.

Regards,
Martin



From gh_pythonlist at gmx.de  Mon Feb 11 06:23:27 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Mon, 11 Feb 2002 12:23:27 +0100
Subject: Long int to int
In-Reply-To: <7xeljsp90d.fsf@ruckus.brouhaha.com>
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com>
Message-ID: <20020211112326.GB4618@lilith.hqd-internal>

Le 11/02/02 ? 03:12, Paul Rubin ?crivit:
> Tim Howarth <tim at worthy.demon.co.uk> writes:
> > (Or worse, I can convert with eval
> > 
> > a = str(hex(a))[2:-1]
> > a = eval('0x'+a)
> 
>   a = eval('0x%x' % a)

Wow, simpler. And I thought I knew at least the Python basics.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 9.2 ?C      Wind: 6.6 m/s



From tim.one at home.com  Mon Feb  4 15:17:34 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 4 Feb 2002 15:17:34 -0500
Subject: Backwards Compatibility of Python versions
In-Reply-To: <a3md47$h98$1@panix2.panix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGELFNJAA.tim.one@home.com>

[Paul Rubin]
> The rand module also disappeared, completely gratuitously as far as I
> can tell.  That's what broke the Redhat scripts.

[Aahz]
> The rand module was already deprecated in 1.5.2; it's in
> Lib/lib-old/rand.py; it's not even in the docs.

That was explained to Paul last time he complained about rand; I'll be
charitable and assume he skipped reading c.l.py that month <wink>.

BTW, it is in the docs:  rand is still listed in the "Obsolete" section of
the "Undocumented Modules" appendix of the Library Reference manual.  It
will probably be listed there forever, and rand.py will probably remain in
lib-old/ forever too (no module has ever been *removed* from Python).




From bldrake1 at yahoo.com  Tue Feb  5 18:21:11 2002
From: bldrake1 at yahoo.com (Barry Drake)
Date: 5 Feb 2002 15:21:11 -0800
Subject: ANN: HappyDoc 2.0.1
References: <mailman.1012913477.7851.python-list@python.org>
Message-ID: <fb072181.0202051521.1a10e7f5@posting.google.com>

Doug,
Thank you for the fast response.  HappyDoc is working great.  I'll be
using it heavily for my documentation.

Barry Drake

Doug Hellmann <doug at hellfly.net> wrote in message news:<mailman.1012913477.7851.python-list at python.org>...
> Announcing the latest version of HappyDoc, a Python documentation extraction 
> tool.
> 
>   HappyDoc is a tool for extracting documentation from Python source
>   code.  It differs from other such applications by the fact that it
>   uses the parse tree for a module to derive the information used in
>   its output, rather that importing the module directly.  This allows
>   the user to generate documentation for modules which need special
>   context to be imported.
> 
>   More details are available on the HappyDoc home page at
>   http://happydoc.sourceforge.net.
> 
>   !!! NOTICE !!!
> 
>     Before installing HappyDoc 2.0+, any version earlier than 2.0 must
>     be removed.  This includes the file 'happydoc.pth' and directory
>     'happydoc' in site-packages, as well as the executable 'happydoc'
>     in the 'bin' directory.
> 
>   Version 2.0.1 -- 
> 
>     The primary reason for this release is to resolve several bugs
>     which prevented Win32 support from working properly.  Support for
>     Win32 operating systems should now be restored.
> 
>     - **New Features**
> 
>       - Added limited support for CGI programs by expanding the docset
>         processing to look for files ending in '.cgi' as well as
>         '.py'.
> 
>     - **Bug Fixes**
> 
>       - Resolved defect #505456 and 498204, a problem with
>         'happydocwin.py' that prevented it from working properly on
>         Win32 systems.
> 
>       - Resolved defect #501240 so that path handling under Win32
>         systems works properly and HappyDoc can generate output.  
> 
>       - Resolved defect #505188 so that using the '-o' option will
>         properly send output to stdout.
> 
>     - **Other Changes**
> 
>       - Added more unit tests for formatters, especially
>         'fileformatterbase.py' and 'formatter_HTMLFile.py'.
> 
>       - Updated the StructuredTextNG docstring converter so that if an
>         exception is generated during parsing of the ST, we fall back
>         to ClassicStructuredText.  The two syntaxes appear to be
>         incompatible, and this should protect backwards compatibility
>         for most users.
> 
> 
> Download
> 
>   Download the latest version of HappyDoc from the home page on
>   SourceForge:
> 
>      http://sourceforge.net/projects/happydoc


From starx at pacbell.net  Mon Feb 11 16:19:30 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 11 Feb 2002 21:19:30 GMT
Subject: Best python way to do 'object messaging'
References: <Xns91AFB91DF83A2RASXnewsDFE1@130.133.1.4> <poV98.2470$yV3.183554@newsread2.prod.itd.earthlink.net>
Message-ID: <Xns91B2878848989RASXnewsDFE1@130.133.1.4>

Patrick K. O'Brien || Mon 11 Feb 2002 11:59:17a:

> 
> Philip,
> 
> You might find this code helpful:
> 
> http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/87056

Well, yes, that does look useful, but for a different problem. (When i 
rip things apart and seperate the UI and data stuff, i could probably 
use it :). 

What i'm doing is pretty much modling stimuli (sort of a vague cousin of 
the 'command' pattern), as one-way messages into an object that is 
simulating some entity (like an orc :). A good example would be a sound. 
A sound would be a message... actualy a lot of messages. They would be 
propigated through the envorment with some function 
(apply_message_radius or somthin'), and for each cell (assuming a grid 
map) covered by the radius (or until the volume passes below some range) 
a copy would be sent to that cell with it's volume property affected by 
1/r^2 falloff. That cell would copy the messages to each of it's 
objects, which could do whatever they want with it. For example, if 
there was a functioning ear attached to some creature, it could recieve 
the message (with a chance for garbling and a check against hearing 
thresholds), and pass it along to the AI of the creature.

It needs to be copied because various things in the depth of the passing 
can filter and modify the message, and this must not affect other things 
in the breadth of the passing. Creature A's fire ring has to reduce the 
value of a fire damage message for the creature, but must not do so to 
the message that Creature B will see.

I already have a solution to this problem, however, see the 'Is 
this...evil?' thread for more ;).

Sorry if it's hard to follow... i'm sick, muddled, and short on time. :) 
(brain->page with less filtering than usual :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From shalehperry at attbi.com  Wed Feb 20 21:08:00 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 20 Feb 2002 18:08:00 -0800 (PST)
Subject: strings and sort()
In-Reply-To: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com>
Message-ID: <XFMail.20020220180800.shalehperry@attbi.com>

On 21-Feb-2002 Jason wrote:
> How do I sort the characters in a string, so far I use this:
>         a="qwerasdfzxcv"
>         b=[x for x in a]
>         b.sort()
> 

b = list(a) # this works too

> Why doesn't sort() return the sorted list.  I would like to chain it
> to other operations:
>         b=[x for x in a].sort()
> 

efficiency.  Why return the sequence when we usually just need to sort the
list.  Sort is also a meothod of an object, why should it return a new object
instead of just affecting the one it belongs to?





From tom at 8dynamics9.com  Thu Feb  7 12:01:50 2002
From: tom at 8dynamics9.com (Tom F.)
Date: Thu, 07 Feb 2002 17:01:50 GMT
Subject: dependency hell
References: <xIw88.500$9%6.51706@bgtnsc04-news.ops.worldnet.att.net> <1013094698.15749.0.nnrp-02.3e318052@news.demon.co.uk>
Message-ID: <2qy88.716$9%6.84333@bgtnsc04-news.ops.worldnet.att.net>

gnutter wrote:
> Could it be that your LD_LIBRARY_PATH is pointing to the wrong place?
I suppose it could be, how would I check that?

-- 
Tom F.
Please remove the "9"
to reply to this message.


From mwh at python.net  Thu Feb  7 06:54:03 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 7 Feb 2002 11:54:03 GMT
Subject: array precision
References: <mailman.1012956250.23108.python-list@python.org> <fb072181.0202060723.e400eaa@posting.google.com>
Message-ID: <ug04da4k1.fsf@python.net>

bldrake1 at yahoo.com (Barry Drake) writes:

> Maybe I'm missing something, but here is what I get:

Well, I'm missing your point.

> Python 2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)] on
> win32
> Type "copyright", "credits" or "license" for more information.
> IDLE 0.8 -- press F1 for help
> >>> import math
> >>> n = 10
> >>> nf = 10.0
> >>> a = 1.0
> >>> b = 0.5
> >>> a/n
> 0.10000000000000001
> >>> b + a/n
> 0.59999999999999998
> >>> a/nf
> 0.10000000000000001
> >>> b + a/nf
> 0.59999999999999998
> >>> 
> 
> Python does double precision floats only (see Python Essential
> Reference 2nd ed. by David Beazley, p.23).

But you can have arrays (as in, objects of type array.ArrayType) that
store C single floats.  Which was Jason's point.

> Python implements IEEE 754: 17 digits of precision with exponent in
> -308 to 308.

Oh No It Doesn't.  Python uses (for the objects of type
types.FloatType) whatever the C compiler that compiled it called
"double".  This may be IEEE 754 doubles on the platforms you've used,
but there's certainly no guarantee of that (e.g. Crays, IBM boxes (I
think)).

>          float     double
> python   64 bits   64 bits
> C        32 bits   64 bits

"python double" is meaningless.

> BTW, you might want to check out the Numeric module in the NumPy
> distribution at http://numpy.sourceforge.net.

That lets you store floats or doubles too.

Cheers,
M.

-- 
  I'm about to search Google for contract assassins to go to Iomega
  and HP's programming groups and kill everyone there with some kind
  of electrically charged rusty barbed thing.
                -- http://bofhcam.org/journal/journal.html, 2002-01-08


From klaatu at evertek.net  Sat Feb 23 17:41:53 2002
From: klaatu at evertek.net (Mike Dean)
Date: Sat, 23 Feb 2002 16:41:53 -0600
Subject: Correction: Re: string substitutions
Message-ID: <20020223164153.A20097@evertek.net>

* Mike Dean <klaatu at evertek.net> [2002-23-02 14:22]:
> import re
> 
> # Replace all occurances of one or more spaces with a single space
> re.sub(' +', ' ', mystring)
> # Ditto for newlines
> re.sub('\n+', '\n', mystring)
> # And, to combine the two into one operation:
> re.sub('(\n+| +)', '\1', mystring)

This last line doesn't work as intended - sorry! (didn't proofread it
carefully enough :-()  It won't convert those chars to single characters
(in fact, it does absolutely nothing except consume processor time)...
the following (using a lambda to return the first character of the
string of whitespace) works:

re.sub('(\n+| +)', (lambda m:m.group(0)[0]), mystring)

If there's a way (there probably is) to write this without the lambda
trickery, could someone enlighten me?

will-try-to-remember-to-proofread-code-before-posting-ly y'rs
Mike



From steve at lurking.demon.co.uk  Mon Feb  4 06:20:40 2002
From: steve at lurking.demon.co.uk (Steve Horne)
Date: Mon, 04 Feb 2002 11:20:40 +0000
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com>
Message-ID: <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com>

On 4 Feb 2002 01:42:33 -0800, alanmk at hotmail.com (Alan Kennedy) wrote:

>Gerard Braad <g_braad at survion.com> wrote:
>> Any help would be appreciated. We hope this will become a standard for
>> inclusion of Python in XML documents.
>
>Gerard,
>
>I don't understand what's wrong with
>
><script language="python" xml:space="preserve">
>    def messiah():
>        print "He's NOT the Messiah."
>        print "He's a very naughty boy."
></script>
>
>or something similar.
>
>Alan.
>
>P.S. It's from "The Life of Brian", in case you're wondering.

This solves one problem - a simple embedding of Python in XML.

I haven't looked at their site, but XML (using tags for structures,
statements etc) may have considerable benefits over plain text as a
source code representation - the only problem being the need for good
XML editors...

One major advantage is the possibility that source code could be more
concerned with semantics, and the presentation could be an issue for
the editor. No more arguments about indentation against braces - tell
the dissenters to simply configure their editor to work with braces
instead of (or as well as) indentation. With a sufficiently
sophisticated editor, the syntax of the language becomes 'skinnable'.

Another advantage is that identifiers inherently occupy a different
namespace to tag names (ie keywords), so the issue of adding new
keywords becomes a complete non-issue. Odds are there would be a
simple <id name="xxx" /> tag for identifiers, editors would make the
keyword/identifier issue clear using colour or style, and conflicts
between identifiers and keywords would simply be impossible.

Another advantage - issues where the semantics of certain syntax have
changed become a non-issue. An example would be the division operators
semantic change. Whether a tag is represented in the editor using a
function or an operator, in addition to the actual symbol used, could
be configurable. Simply create additional division tags for the new
semantics (keeping the old ones) and let people reconfigure their
editors to use the syntax they want for the semantics they want.

Of course, all of this is technically possible now - you'd simply need
an editor capable of translating Python syntax to/from a user
configurable presentation - but working from an XML based language
would be much more practical.

BTW - this isn't the first time this idea has been proposed.

-- 
Steve Horne
steve at lurking.demon.co.uk


From rdacker at pacbell.net  Thu Feb  7 17:32:37 2002
From: rdacker at pacbell.net (rdack)
Date: 7 Feb 2002 14:32:37 -0800
Subject: [osx] milter problem
References: <mailman.1013023572.9880.python-list@python.org>
Message-ID: <644f6688.0202071432.7b66a43e@posting.google.com>

bob ackerman <rdacker at pacbell.net> wrote in message news:<mailman.1013023572.9880.python-list at python.org>...
> >
> > Maybe you need:  "-lpthread"  added to the cc switches. (?)
> >
> > -- Steve Majewski
> >
> >

problem was in libmilter.a. i got the milter to run after removing
references to pthread_sigmask and sigwait in
sendmail/libmilter/signal.c, rebuilding libmilter.a, and rebuilding
the milter module.
Thanks for help.


From spencer at spencerdoidge.com  Sat Feb  2 21:40:26 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sat, 2 Feb 2002 18:40:26 -0800
Subject: z Windows serial communication
Message-ID: <a3i7o7$me7$1@news.efn.org>

Hello,

Many thanks and kudos to Issac Barona for his serial communications Python
enhancements, which I downloaded from
http://balder.prohosting.com/ibarona/en/python/uspp/uspp_en.html

>From this collection, which I have been using and studying, I have been
examining SerialPort_win.py, and I cannot find where there is an input
buffer that I can access after calling self.read(n). I do not know Python
well enough yet to modify SerialPort_win.py, but I will try to learn now.

Of course, I will welcome any help from anyone willing to save me a little
time. I have emailed to Mr. Barona, but he has not replied.

I notice that self.read(n) generates an exception when Win32 API's
ReadFile(...) times out. For my own purposes, I prefer not to generate an
exception unless ReadFile() returns FALSE, denoting a communication error
such as a framing error or parity error. In particular, my understanding of
ReadFile(..) is that it treats a timeout as a normal event and returns TRUE
when there is a timeout, even if lnNumberOfBytesToRead points to a different
value than lpNumberOfBytesRead. I can detect a failed receive operation this
way:

if (lnNumberOfBytesToRead != *lpNumberOfBytesRead){
// protocol error: did not receive expected number of bytes
}

When I am receiving a packet, I decide that the sender is finished sending
when I have waited long enough after the last received character, that is,
when ReadIntervalTimeout as set in SetCommTimeouts(..) is allowed by the
sender to expire. It is after that event that I parse the received message.

I have modified SerialPort_win.py accordingly.
I will be glad to share my work with anyone interested, and I will gladly
accept anyone's help.

I have written TP3, a small multithreaded Win32 API app (VC++ project) that
sends and receives, via the serial port, binary data translated from ASCII
entered on the keyboard. If you have a 2nd PC and a null-modem cable to
connect your two PCs, this makes a nice way to test Mr. Barona's software.
Since the app is changing from day to day, I will wait for someone to
express an interest before posting the source somewhere.

Spencer Doidge





From neal at metaslash.com  Tue Feb 26 08:02:06 2002
From: neal at metaslash.com (Neal Norwitz)
Date: Tue, 26 Feb 2002 08:02:06 -0500
Subject: Problems with select
References: <pan.2002.02.26.04.17.29.538545.11917@onesite.org>
Message-ID: <3C7B874E.8BD9818@metaslash.com>

Alain Tesio wrote:
> 
> Hi, I can't manage to make select work, in both 1.52 and 2.1.2
> 
> I just want to read stderr and stdout when executing a command
> with popen3, and I must use select, read() blocks when there is
> more than a fixed amount of data (4 KB I think)
> 
> The problems is that select still gives something after the end
> of the file, or sometimes never returns anything at all.
> 
> I create the pipes with :
> 
> (pipe_out,pipe_in,pipe_err)=popen2.popen3(command)
> 
> and use select like this:
> 
> pollstatus=select.select([],[pipe_out],[pipe_err],1)
> 
> (the doc says : select(iwtd, owtd, ewtd[, timeout])
> I didn't put pipe_in as it's closed)

It looks like you may want to do something like:

	select_fds = [ pipe_out, pipe_err ]
	pollstatus=select.select(select_fds,[],[],1)

select() should return whenever there is something to read from
either pipe_out or pipe_err, or after 1 second.

Neal


From paul1brian at yahoo.com  Thu Feb 14 10:50:04 2002
From: paul1brian at yahoo.com (Paul Brian)
Date: Thu, 14 Feb 2002 15:50:04 -0000
Subject: watching the number of references to an object
Message-ID: <1013700953.2880.0.nnrp-01.c1c3e191@news.demon.co.uk>

Dear all,

Does anyone know how to monitor the references pointing to an instance of an
object?

the gc module (according to docs) seems to show those with no references
left, but I am interested in seeing how many refs a given instance has...any
ideas on back of a postcard please

-----------------------------
P Brian
"I haven't worked out how to add signatures to my newsreader yet"




From ahmebah at hotmail.com  Sat Feb 23 11:39:57 2002
From: ahmebah at hotmail.com (ameoba)
Date: 23 Feb 2002 16:39:57 GMT
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <3C747268.6C756B9D@earthlink.net> <7xvgcrfosh.fsf@ruckus.brouhaha.com> <3C747B81.367B5D50@earthlink.net> <7xofijct0y.fsf@ruckus.brouhaha.com>
Message-ID: <Xns91BE58118876ahmebahhotmailcom@198.99.146.10>

Paul Rubin <phr-n2002a at nightsong.com> wrote in 
news:7xofijct0y.fsf at ruckus.brouhaha.com:

> Hans Nowak <wurmy at earthlink.net> writes:
>    def sort2(lst):
>       return sort3(lst[:])

IANAFP, but...

sort2c = lambda X : (lambda r : (not r.sort()) and r)((lambda c : c[:])(X))



From quinn at seniti.ugcs.caltech.edu  Sat Feb 16 05:20:07 2002
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 16 Feb 2002 10:20:07 GMT
Subject: line break confusion
References: <XFMail.20020215193042.shalehperry@attbi.com> <mailman.1013835745.20583.python-list@python.org>
Message-ID: <slrna6scin.reb.quinn@seniti.ugcs.caltech.edu>

On Fri, 15 Feb 2002 21:01:18 -0800, Michael Mell <mike at nthwave.net> wrote:
>Sean 'Shaleh' Perry wrote:
>
>> Windows: \r\n
>> Linux (UNIX): \n
>> Mac: \r
>>
>> Each OS's libs transform the sequence '\n' to whatever the right thing is for
>> them.
>
>Thanks for answering such an apparently ignorant question, but the answer is not
>so simple.
>
>I want to read and write files with line breaks which are not of the native
>platform type.
>
>Problem #1 is that on a Mac, the values of Mac and Unix are /swapped/ and
>therefore wrong. Writing a file using '\n' results in a /Mac/ line break (as
>reported by BBEdit, the venerable Mac editor). Writing a file using '\r'

That's what he said above.  The Mac stdio converts \n to the right thing (\r on
a Mac).


>reported by BBEdit, the venerable Mac editor). Writing a file using '\r'
results
>in a /Unix/ file. The Mac seems unable to write Windows \r\n no matter what
>order is used.

But that seems weird.

>Problem #2 is that when reading a Windows file on Linux or Mac, a /single/ line
>('\r\n') is interpreted as /two/ lines ('\n\n'). Files written out on Linux
then >have double the correct number of lines.

That's also weird.  Doesn't happen for me when I read CRLFs on unixes.


You should probably open your files in binary mode, to get stdio to leave the
line ending alone.  Then do your own EOL translation.


From marklists at mceahern.com  Sun Feb 10 16:53:51 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 13:53:51 -0800
Subject: Are there any Python books or sites with good exercises?
In-Reply-To: <g3fd6u86ug9e7taimt40ibh7ihe4ncqo0r@4ax.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEAAKCAA.marklists@mceahern.com>

In direct response to the subject:

Have you checked out diveintopython.org?

// mark



From sabren at manifestation.com  Fri Feb  8 17:59:20 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 8 Feb 2002 17:59:20 -0500 (EST)
Subject: Pythonistas? <- Re: OT: Perl programmers?
In-Reply-To: <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4>
Message-ID: <Pine.LNX.4.44.0202081755340.11581-100000@mercury.sabren.com>

On 8 Feb 2002, Philip Swartzleonard wrote:

> Aahz Maruch || Fri 08 Feb 2002 11:10:47a:
> 
> > I know some of y'all still keep roots to the Perl community, so I was
> > wondering if there's any kind of definitive nickname for Perl
> > programmers, the way we call ourselves Pythonistas?
> 
> Hm, i don't seem to remeber seeing this Pythonistas thing till maybe two 
> days ago. Anyway, where did it come from-- based on some word? And HTF do 
> you pronouce it? :)


It's been around a while... I say "python - ee - sta"...

random factoid: I used to own pythonista.com, but gave it
to Steve holden...  not sure what he's got planned for
it... :)

Aahz -> perl programmers call themselves JAPHs... for "Just
Another Perl Hacker". They often even sign their messages
with perl oneliners that produce that string, if I recall.


Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From jgardn at alumni.washington.edu  Thu Feb 21 07:20:22 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Thu, 21 Feb 2002 21:20:22 +0900
Subject: Why this behavior? (was: Stupid Lambda Tricks)
References: <3c74c211.20012718@news.isar.de>
Message-ID: <js5d8.116$Az6.466@news.hananet.net>

< snip excellent lambda tricks >
> 
> Many list functions return None, so you cannot easily use them in a
> stacked expression inside lambda. Examples of this are [].sort() and
> [].reverse(). Instead of writing:
> 
> list.some_function_that_returns_None()
> function(list)
> 
> you can write
> 
> function((list.some_function_that_returns_None(),list)[1])
> 
> Here is an example:
> 
> import random
> 
> sort = lambda x: (x.sort(),x)[1]
> reverse  = lambda x: (x.reverse(),x)[1]
> 

I've always wondered:

>>> "abc".upper()
"ABC"
>>> [3,1,2].sort()
< nothing >

Why is that? Is it just easier to implement it this way, or is there some 
sound philosophical reason for this?

Jonathan



From yura at sunet.ru  Fri Feb  1 01:28:03 2002
From: yura at sunet.ru (Yura Ushakhow)
Date: Fri, 01 Feb 2002 09:28:03 +0300
Subject: calling upper constructor (thank you!)
References: <3C5A04E9.3040502@sunet.ru> <3C5A1C1D.10F4B14D@engcorp.com>
Message-ID: <3C5A3573.2050801@sunet.ru>

What the subject says, appreciate greatly. ;)



From eric.brunel at pragmadev.com  Wed Feb  6 06:18:18 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Wed, 6 Feb 2002 11:18:18 +0000
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary?
References: <3c5a84eb.70228013@news.online.de> <3c5d9698.182096361@news.online.de> <mailman.1012835529.12793.python-list@python.org> <3c60fb80.199334@news.online.de>
Message-ID: <a3qvmn$6o2$1@norfair.nerim.net>

<posted & mailed>

Hi Johannes,

> def __init__(self, master=None, **kw):
>       Frame.__init__(self, master)
>      ...
>       self.label = Label(self, text=kw['labelText'])
> ...
> 
> But then, how do I handle default values, and how do I ensure that the
> objects are created with the right parameters?

Do not use kw['labelText'], but kw.get('labelText', ''). The get method on 
a dictionary does just what you want. Please refer to:
http://www.python.org/doc/current/lib/typesmapping.html

HTH
 - eric -


From DeepBleu at DeepBleu.org  Fri Feb  8 20:22:18 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Sat, 09 Feb 2002 01:22:18 GMT
Subject: RELEASED - StandaloneZODB 1.0 final
References: <mailman.1013215579.9688.python-list@python.org>
Message-ID: <eR_88.122627$XZ1.5188116@typhoon.austin.rr.com>

Now this is something to dig into!
DeepBleu
"Barry A. Warsaw" <barry at zope.com> wrote in message
news:mailman.1013215579.9688.python-list at python.org...
>
> I'm please to announce the release of StandaloneZODB, the Python
> object persistency system also known as the Z Object Database.  ZODB
> is the object-oriented database underlying Zope; the StandaloneZODB
> project's goal is to provide those same facilities to non-Zope related
> Python applications.
>





From usenet at thinkspot.net  Sat Feb  9 19:43:43 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 09 Feb 2002 16:43:43 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com> <8852d33c.0202091432.452ad6f9@posting.google.com> <a43due.3vv9a5r.1@kserver.org> <7xg04ai3j2.fsf@ruckus.brouhaha.com>
Message-ID: <a43jk0.3vsk84f.1@kserver.org>

On 09 Feb 2002 16:28:17 -0800, Paul Rubin <phr-n2002a at nightsong.com>
wrote in comp.lang.python in article
<7xg04ai3j2.fsf at ruckus.brouhaha.com>:

> Sheila King <usenet at thinkspot.net> writes:
> > I have a friend I'm working with right now, that we are putting together
> > a project in Python and PHP (he's doing all of the PHP part). He has
> > only just been learning Python for a few weeks. He might have some good
> > suggestions for you, as to how he approached learning Python (as an
> > experienced programmer in other languages). If you'd like me to put you
> > in touch with him, send me an email.
> 
> Topic drift: why are you doing something in Python -and- PHP?
> Can't you do it all in Python, or maybe all in PHP?  Why use two languages?
> There's some Python template systems that do pretty much what PHP
> does.  If something is missing from them, maybe they need improvement.

Let's put it this way:
I prefer to handle the part of the project that deals with scanning in
the emails and processing them. So that part is being done in Python.

The PHP is a web interface for the user to configure the Python script.
It is really irrelevant what language the web interface is in. Anyhow,
my partner is doing the interface, and he is more skilled in PHP.

I had started out on a very similar project on my own, before I became
aware that this other person was also working on a project like mine. I
was doing the web interface in Python and proceeding quite slowly (only
because I don't have much experience at it). This guy already had a
working interface in PHP when I found out that we were doing similar
projects and might as well work together.

In other words: pragmatics. I joined a project that was already in
mid-stream and started by someone else.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_




From tdelaney at avaya.com  Wed Feb 20 18:12:16 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 21 Feb 2002 10:12:16 +1100
Subject: comp.lang.python in English? (was Re: Why Python is like BASI
	C)
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1A0@natasha.auslabs.avaya.com>

> From: Russell Nelson [mailto:nelson at crynwr.com]
> 
> Henrik Motakef <henrik.motakef at web.de> writes:
> 
> > Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> > 
> > > Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> > > > Blig matro en cordiolios, plur tanto lestanto 
> forgolumnia.  Plachero
> > > > czignat?  Classifero en bragna!
> > > Woas da Deifi wos des hoasn mog.
> > 
> > Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> 
> Even better: we could switch languages within threads, to *really*
> keep the monolinguists confused.

Como se quiser ...

Tim Delaney



From ark at research.att.com  Sat Feb  2 13:30:44 2002
From: ark at research.att.com (Andrew Koenig)
Date: Sat, 2 Feb 2002 18:30:44 GMT
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>
Message-ID: <yu99sn8j7n3f.fsf@europa.research.att.com>

Stede> I was thinking of doing an E-book (because I don't like the
Stede> high price of regular books) on an introduction to programming
Stede> and artificial intelligence in Python.

I have yet to buy my first E-book, because I am reluctant to pay for
something useful that might suddenly become unavailable if, say, I
upgrade my computer.

Bruce Eckel claims that he has had great success in making the entire
text of his books available on the web for free -- people are still
willing to pay the price for having a bound volume that they can read
at the beach, or wherever.  Moreover, that way he benefits from readers'
comments before he has to commit to print.

-- 
Andrew Koenig, ark at research.att.com, http://www.research.att.com/info/ark


From magnus at thinkware.se  Fri Feb 22 10:35:40 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 16:35:40 +0100
Subject: Python 2.1 and PIL
References: <94id8.563$s03.36009@news2.east.cox.net>
Message-ID: <3C76654C.9010002@thinkware.se>

What OS are you using? (For windows there are binaries for
the different Python versions.)

william wrote:

> Ok been playing with this python 2.1.2 and PIL.  The problem is when i
> install the PIL it installs it for python1.5.  I don't know how to get the
> PIL to install for Python2.1.
> 
> I am assuming I need to make some changes in some files before i do
> ./configure.  Any suggestions?
> 
> Thanks.
> 
> 
> 



From tanzer at swing.co.at  Fri Feb 15 09:27:49 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 15 Feb 2002 15:27:49 +0100
Subject: Convert String to Dictionary question
In-Reply-To: Your message of "14 Feb 2002 22:23:12 GMT."
             <slrna6oe6g.d0b.quinn@hurl.ugcs.caltech.edu>
Message-ID: <E16bjL3-00045W-00@tswings.swing.cluster>

quinn at hurl.ugcs.caltech.edu (Quinn Dunkan) wrote:

> On Thu, 14 Feb 2002 21:27:48 +0000, John J. Lee <jjl at pobox.com> wrote:
> >On Thu, 14 Feb 2002, MDK wrote:
> >
> >> I saved my dictionary {'hello1': [1, 0, 0]} to a file.
> >>
> >> When I do a readline the I get back a string.
> >>
> >> How do I convert this string to a dictionary?
> >
> >You probably don't want to.  Read the documentation for the pickle module.
>
> Or if you do, say because 'repr' format is more human readable than
> pickle, do 'd = eval(fp.readline())'. As always with eval(), if bad
> guys can edit your file, you're in trouble. Of course, if they can
> edit your pickle, you're in trouble too.

Not quite:

>>> import os
>>> os.system ("echo 'kilroy'")
kilroy
0
>>> eval ("""os.system ("echo 'kilroy'")""")
kilroy
0
>>> eval ("""os.system ("echo 'kilroy'")""", {}, {})
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "<string>", line 0, in ?
NameError: name 'os' is not defined

Making pickle safe is left as an exercise ;-)

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From lac at strakt.com  Mon Feb  4 10:09:59 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 04 Feb 2002 16:09:59 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary? 
In-Reply-To: Message from skywalkerpackage@hotmail.com (Johannes Eble) 
   of "Sun, 03 Feb 2002 20:07:10 GMT." <3c5d9698.182096361@news.online.de> 
References: <3c5a84eb.70228013@news.online.de>  <3c5d9698.182096361@news.online.de> 
Message-ID: <200202041509.g14F9xGo011308@ratthing-b246.strakt.com>

> Hello Laura and Eric,
> 
> thanks very much for your answer, it has helped me a lot :)

Good to hear it.

> 
> Laura, I don't think,
> 
> widget=widgetType(master, **mydict)
> 
> is necessary. I have tried
> 
> widget=widgetType(master, mydict)
> 
> with mydict={'text':'hello', 'relief':'sunken'}. It worked without the
> two stars.
> Aren't the two stars just necessary for function calls with variable
> length key parameters?
> 
> 
> ByeBye
> 
> Johannes
> 

Fredrik Lundh's Introduction to Tkinter has a warning about this:

http://www.pythonware.com/library/tkinter/introduction/x386-backwards-compatibility.htm

        Quoting Fredrik Lundh here:

	For example, if you create a custom widget which needs to pass 
        configuration options along to its parent class, you may come up
        with something like:

              def __init__(self, master, **kw):
                  Canvas.__init__(self, master, kw) # kw is a dictionary

        This works just fine with the current version of Tkinter, but it may 
        not work with future versions. A more general approach is to use
        the apply function:

              def __init__(self, master, **kw):
                   apply(Canvas.__init__, (self, master), kw)
--------

  The ** syntax is just a convenient shorthand for the apply.  It won't
  work before Python 2.0 and you will have to use the apply form.
  
  apply is documented here http://www.python.org/doc/lib/built-in-funcs.html

Laura Creighton




From pbienst at MIT.EDU  Tue Feb 26 13:43:32 2002
From: pbienst at MIT.EDU (Peter Bienstman)
Date: 26 Feb 2002 13:43:32 -0500
Subject: Precompiled regular expressions slower?
In-Reply-To: <XFMail.20020226103648.shalehperry@attbi.com>
References: <XFMail.20020226103648.shalehperry@attbi.com>
Message-ID: <1014749012.11444.18.camel@hunsmac.mit.edu>

On Tue, 2002-02-26 at 13:36, Sean 'Shaleh' Perry wrote:
> 
> On 26-Feb-2002 Peter Bienstman wrote:
> >> Perhaps Peter should post the relevant code so we can see what is actually
> >> happening for ourselves?
> > 
> > Here is the relevant portion of the code:

(this was the fastest, orginal version, i.e without precompiling)
 
> After changing things as Jon Ribbens explains did you get a sufficient speed
> boost?

Changing from precompiling to a global var to precompiling to a local
var improved performance a bit, but still the non-precompiled version
was faster.

> My only other comment would be one of style and efficiency.  In the loop you
> check the line against numerous regexs.  Would each line really match more than
> one or two of those?  Perhaps a few strategic calls to continue may help.

Interesting tip, thanks.

Does't explain the slowdown of precompiled reg ex's, though.

Peter




From loewis at informatik.hu-berlin.de  Mon Feb 25 05:30:10 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Feb 2002 11:30:10 +0100
Subject: Time to rename Stackless? (Re: Stackless Platform Independence?)
References: <mailman.1014420937.15379.python-list@python.org> <KyEd8.66437$dG.27580630@news1.rdc1.sdca.home.com> <987e7uoo5v87fsjruq0ug6bff0vha5ear0@4ax.com> <I7Qd8.66687$dG.27729291@news1.rdc1.sdca.home.com> <3C7999AE.A7786828@cosc.canterbury.ac.nz>
Message-ID: <j4d6ytn9al.fsf@informatik.hu-berlin.de>

Greg Ewing <greg at cosc.canterbury.ac.nz> writes:

> Seems to me that "Stackless" is a misnomer now. If it
> works by switching between multiple stacks, then it now
> uses *more* stacks than normal Python!

That is not really true. There is only a single C stack at any point
in time, and it is bounded in size. This is just like the old
"stackless" Python: it also had a single C stack, and it also was
bounded in size.

If you count the stack copies that are made on the heap: this is also
like the old stackless Python: there always was a Python "stack" (a
linked list of frames) on the heap.

Regards,
Martin


From jdoak at lanl.gov  Wed Feb 20 13:09:17 2002
From: jdoak at lanl.gov (Justin Doak)
Date: 20 Feb 2002 10:09:17 -0800
Subject: sys.path Problem with 2.2?
References: <3d7957d4.0202191547.1f13d03b@posting.google.com>
Message-ID: <3d7957d4.0202201009.2a59d8e6@posting.google.com>

Doh! I was trying to add $HOME/python/lib to the PYTHONPATH when what
I _meant_ to do was add $HOME/lib/python. The interpreter was smart
enough not to add a directory that didn't exist to the module search
path. This seems to be an argument for pair programming. I looked at
the PYTHONPATH a hundred times without seeing this problem. Someone
else would've seen it immediately.

JD

jdoak at lanl.gov (Justin Doak) wrote in message news:<3d7957d4.0202191547.1f13d03b at posting.google.com>...
> I add a directory to my PYTHONPATH, but when I fire up the
> interpreter, sys.path is getting the old value for PYTHONPATH. I ran
> 2.1, and it seems to be loading up the new PYTHONPATH just fine. 2.2
> also seems to know when my PYTHONPATH is set or not; when I unset it,
> nothing from the PYTHONPATH shows up in sys.path. This is decidedly
> strange behavior; is there some caching of the PYTHONPATH going on
> somewhere/somehow?
> 
> JD
> 
> # sys.path doesn't get all the directories from PYTHONPATH.
> 
> 134 doak->doak% echo $PYTHONPATH
> .:/n/toaster/u/doak/cvs/computer_vision:/n/toaster/u/doak/python/lib
> 135 doak->doak% python
> Python 2.2 (#1, Jan  4 2002, 14:14:33)
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> import sys
> >>> sys.path
> ['', '/nh/toaster/u/doak', '/n/toaster/u/doak/cvs/computer_vision',
> '/packages/lib/python2.2', '/packages/lib/python2.2/plat-linux2',
> '/packages/lib/python2.2/lib-tk',
> '/packages/lib/python2.2/lib-dynload',
> '/packages/lib/python2.2/site-packages',
> '/packages/lib/python2.2/site-packages/Numeric']
> 
> # But sys.path knows when I've unset PYTHONPATH.
> 
> 136 doak->doak% unsetenv PYTHONPATH
> 137 doak->doak% python
> Python 2.2 (#1, Jan  4 2002, 14:14:33)
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> import sys
> >>> sys.path
> ['', '/packages/lib/python2.2', '/packages/lib/python2.2/plat-linux2',
> '/packages/lib/python2.2/lib-tk',
> '/packages/lib/python2.2/lib-dynload',
> '/packages/lib/python2.2/site-packages',
> '/packages/lib/python2.2/site-packages/Numeric']
> 
> # 2.1 reads the PYTHONPATH correctly.
> 
> 141 doak->doak% source ~/.cshrc
> 142 doak->doak% echo $PYTHONPATH
> .:/n/toaster/u/doak/cvs/computer_vision:/n/toaster/u/doak/python/lib
> 143 doak->doak% /packages/.packages/python/python-2.1/bin/linux/python
> Python 2.1 (#1, Jun  8 2001, 09:50:58)
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>> import sys
> >>> sys.path
> ['', '/nh/toaster/u/doak', '/n/toaster/u/doak/cvs/computer_vision',
> '/n/toaster/u/doak/python/lib', '/packages/lib/python2.1',
> '/packages/lib/python2.1/plat-linux2',
> '/packages/lib/python2.1/lib-tk',
> '/packages/lib/python2.1/lib-dynload',
> '/packages/lib/python2.1/site-packages',
> '/packages/lib/python2.1/site-packages/Numeric']


From phr-n2002a at nightsong.com  Sun Feb 10 16:53:55 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 10 Feb 2002 13:53:55 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com> <7xlme1hv0y.fsf@ruckus.brouhaha.com> <3C66EA99.B22478E3@alcyone.com>
Message-ID: <7xd6zdhuks.fsf@ruckus.brouhaha.com>

Erik Max Francis <max at alcyone.com> writes:
> > It seems to me like the principle that programs should be able to "do
> > nothing" gracefully.  The argument is that split should be the inverse
> > of join, even when the splitting string is empty.
> > 
> > Imagine if "a+0" raised an error.  You wouldn't say "just use 'a'
> > instead".
> 
> That is simply not comparable.  The purpose of split is to divide up a
> string according to a delimiter.  Asking for it to split a string given
> a null delimiter doesn't make much sense at all; if you're calling
> string.split with the intent of dividing it up and you're passing in ''
> as a delimiter, that's almost certainly a mistake. 

That's the counter-argument.  Time for a PEP.  <wink>


From jason at jorendorff.com  Wed Feb  6 15:18:11 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 14:18:11 -0600
Subject: RFC -- Hygienic Macros in Python
In-Reply-To: <HFEKILOLEFEFMKAECNDLCEMPDAAA.jason@jorendorff.com>
Message-ID: <HFEKILOLEFEFMKAECNDLIENADAAA.jason@jorendorff.com>

I wrote:
> My intuition says: 
>   [...]
>   macros: 20%

I only meant this to apply to a theoretical "macros in Python"
feature.  In other more Lisp-like languages, the number may be
much better.

## Jason Orendorff    http://www.jorendorff.com/



From robin at jessikat.fsnet.co.uk  Thu Feb 14 09:44:08 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 14 Feb 2002 14:44:08 +0000
Subject: win32 Zope 2.5.0 vs python2.1&win32all
References: <TfOVwLAlB7a8EwLX@jessikat.demon.co.uk> <3C6BB8B5.8020903@mxm.dk>
 <jPLZkiAw77a8EwM7@jessikat.demon.co.uk>
Message-ID: <mIZoFKA408a8EwcN@jessikat.demon.co.uk>

In article <jPLZkiAw77a8EwM7 at jessikat.demon.co.uk>, Robin Becker <robin at jessikat.fsnet.co.uk> writes
.....
>well I think you just have to look at the path printed out to see that something 
>is wrong. There should
>logically be no path elements starting without the prefix c:\Python\ZOPE_2_5_0 
>as that is my Zope home. The
>other path elements creep in because of the win32all registry paths.
>
>I'm not doing anything here except running the python.exe that came with zope. 
>If the path is wrong in
>zopexxx\bin\python.exe there's not much I can do about it except to try and 
>eliminate path elements in z2.py
>which I'm not yet doing.
>
>When I run zope it's via the standard service mechanism. Using procexp I can see 
>mixed pyds if I don't
>eliminate the registry path.
......
>sys.path = filter(lambda p,swhome=swhome: not p or swhome.lower()==p[0:len(swhom
>e)].lower(),sys.path)
>but it seems hard to guarantee that we get the right sys unless we mess with imp 
>et al.
....
sys being built in we should always get the right one. The service starts up python
with the 8.3 mangled dos names as standard so then it's hard to filter out
non-standard path elements using swhome which is in long filename form.
When I put in a filter looking like

sys.path = filter(lambda p,swh=swhome[:-5]: not p or swh.lower()==p[:len(swh)].lower(),sys.path)

then magically I can start up and stop the service OK and all seems to be well.
-- 
Robin Becker


From jkraska at san.rr.com  Thu Feb  7 12:06:50 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 17:06:50 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com>
Message-ID: <dfd56u841fd12otq54grmd63548j7u8f4e@4ax.com>

>natural for that to occur).  You might be the first to make the claim
>that those principles can be represented in "nine commands".

A friend of mine did a Master's thesis on the premise of four;
it was really out there, bizarre transformational mathematics
stuff. Totally useless, too.

C//



From deathtospam43423 at altavista.com  Fri Feb 22 18:09:47 2002
From: deathtospam43423 at altavista.com (netvegetable)
Date: 22 Feb 2002 23:09:47 GMT
Subject: help: Problem with cgi form
Message-ID: <Xns91BE6246576BAAusYourStandingInIt@130.133.1.4>

I'm a beginner. How do I do this? It's probly very simple ......

I have a hidden input on a cgi form... 

    print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE=',last_byte_pos,'>'

I need the VALUE to be the  variable "last_byte_pos", but unfortunately 
when I run the script for the first time the script crashes because it's 
undeclared. 

What subroutine do I use to check if last_byte_pos is undeclared, and 
assign it a default value if it is? 



-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From michael at stroeder.com  Tue Feb 12 20:04:22 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 13 Feb 2002 02:04:22 +0100
Subject: Re-throwing an exception without loosing the stack trace
References: <mailman.1013552683.30067.python-list@python.org> <Xns91B4F52BC58cliechtigmxnet@62.2.16.82>
Message-ID: <3C69BB96.369FC648@stroeder.com>

Chris Liechti wrote:
> 
> "Clark C . Evans" <cce at clarkevans.com> wrote in
> news:mailman.1013552683.30067.python-list at python.org:
> 
> > How do I catch and then re-throw an exception without
> > loosing the information about the stack trace?  I've
> > looked around alot and don't see the "obvious" answer.
> 
> try a simple, naked "raise" ;-)

Is that thread-safe?

Ciao, Michael.


From marklists at mceahern.com  Tue Feb 12 10:36:05 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 12 Feb 2002 07:36:05 -0800
Subject: searching backwards in a string
In-Reply-To: <a4bc9e$ope$06$1@news.t-online.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEGJKCAA.marklists@mceahern.com>

[Mark Z?ltsch]
> I'm new to this python stuff and to this newsgroup too..
> but why you don't use 'rfind' instead of 'find'?
> I think that's much easier...

<sheepish>Wow, that IS way easier!

Thanks,

// mark




From mcherm at destiny.com  Wed Feb 13 15:02:47 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Wed, 13 Feb 2002 15:02:47 -0500
Subject: Proceedings CD-ROM of IPC10?
Message-ID: <3C6AC667.7030705@destiny.com>

> Hi,
> 
> I'm not sure if Foretec plans to do this or not, but
> is there already a CD-ROM with Proceedings for IPC10?
> I vagualy remember having seen something similar for 
> IPC9, or so...
> 
> Thanks,
> 
> Dinu


My copy's for sale at the right price! Gotta defray the as much of the 
conference cost as I can...  ;-)

-- Michael Chermside






From rony.steelandt at bucodi.com  Sat Feb  2 07:59:59 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 2 Feb 2002 04:59:59 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com>	<20020130073810.382$1b@news.newsreader.com>	<b34d3835.0201310005.548ee6be@posting.google.com>	<mailman.1012498757.26825.python-list@python.org>	<b34d3835.0202010038.6a907bff@posting.google.com> <mailman.1012591871.11392.python-list@python.org>
Message-ID: <b34d3835.0202020459.6ec3964d@posting.google.com>

Cliff Wells <logiplexsoftware at earthlink.net> wrote in message news:<mailman.1012591871.11392.python-list at python.org>...

> ...
> These arguments may apply somewhat less to you, since your presence on this
> list implies that you are actively pursuing technical knowledge, but in
> general managers tend to be far less knowledgable on these things.
> 

This is definitifly true Cliff, and i think this is the case for all
rather small software companys, remember we're 10 persons.
Alltough even in a small company you need structure (one of the first
lessons i got) i think that as a manager you can't afford yourself to
be the prototype general  manager and rely on technical people, even
if they capable people. This is just my opinion, but if you stick to
that, you need as that type of  manager to have the technical skills
to be able to evaluate the arguments of your technical staff. And this
explains my presents here on the forum.
Python is one of our principle software development tools and even if
my job at my company isn't programming, i think i know it as good as
my programmers... (they better not read this <wink>). But the main
reason for this is that i really like it as a language and use it quit
often for some personnal software i have. Of course it gives me allso
some technical skill, and since my engineers know this, they know that
they can't tell me storys since they know i 'll know it isn't true. So
this is a gain of time in some mather ...

Regards Rony


From brueckd at tbye.com  Wed Feb 27 18:10:25 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Wed, 27 Feb 2002 15:10:25 -0800 (PST)
Subject: PIL vs eps
In-Reply-To: <HRCwzIADVVf8EwB0@jessikat.fsnet.co.uk>
Message-ID: <Pine.LNX.4.33.0202271504480.14811-100000@ns0.tbye.com>

On Wed, 27 Feb 2002, Robin Becker wrote:

> In article <just-B1FF7B.22204827022002 at newszilla.xs4all.nl>, Just van
> Rossum <just at xs4all.nl> writes
> >In article <mailman.1014835309.1869.python-list at python.org>,
> > <brueckd at tbye.com> wrote:
> >
> >> Forgive me if I'm totally off my rocker, but I think by definition eps
> >> files should *not* have a showpage and it is technically correct that
> >> nothing gets rendered unless you have a showpage command. EPS is
> >> "encapsulated postscript" - postscript that you intend to encapsulate or
> >> embed in another document, so showpage would muck things up.
> >
> >The EPS spec says showpage is optional, and that the surrounding ps code
> >should redefine showpage as a no-op.
> >
> >  http://partners.adobe.com/asn/developer/pdfs/tn/5002.EPSF_Spec.pdf
> >
> >Just
> so I guess that EPSPlugin isn't going to be able to handle some eps
> files. Indeed the sample .eps that came with ghostscript didn't have a
> showpage.

I don't know what kind of constraints you're dealing with, but the correct
way to handle an eps file (i.e. make it display) would be to actually
embed it in what is essentially a dummy .ps file, e.g. take the eps data
and surround it with the minimal ps header and footer necessary and then
process that instead of the original. (The problem you're having is rooted
in the fact that you're trying to treat an eps file like a ps file, so the
solution is to use the eps to make a real ps file).

-Dave




From scarblac at pino.selwerd.nl  Tue Feb 12 07:59:48 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 12 Feb 2002 12:59:48 GMT
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <mailman.1013426479.20868.python-list@python.org> <f36322074b.tim@worthy.demon.co.uk>
Message-ID: <slrna6i491.56f.scarblac@pino.selwerd.nl>

Tim Howarth <tim at worthy.demon.co.uk> wrote in comp.lang.python:
> All I want is a=b ! but lose longness of b !

That's simply not possible with a=0xFFFFFFFFL, since 0xFFFFFFFF is too large
for an int, and if b=-1 then a!=b.

What you're looking for is an int that has the same bit pattern as the
unsigned int would have, and people gave several solutions for that,
although they're not very straightforward; so it goes.

Bit shifting is for C, not Python, imo.
-- 
Remco Gerlich


From jeff at ccvcorp.com  Wed Feb 27 19:47:43 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 16:47:43 -0800
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
		<ad052e5c.0202261950.72991328@posting.google.com>
		<3c7c8b74.63255807@news.la.sbcglobal.net> <mailman.1014831886.23090.python-list@python.org>
Message-ID: <3C7D7E2E.6A932E4B@ccvcorp.com>


"Christopher A. Craig" wrote:

> hankc at nospam.com (HankC) writes:
>
> > - does the del(datalines) call free the memory (not necessarily
> > immediately) for the list resource?
>
> It frees it immediately.

Note, however, that this applies only to your list item itself, not to the
contained items.  Those contained items are *also* referenced by the
dictionary -- you haven't copied the strings, you've created two different
references to them.  Deleting the list will eliminate one of those sets of
references; however, it's not *that* big of a deal, since references don't
take much space, relative to strings.  For the most part, only worry about
freeing references if you start to notice memory consumption being a
problem.  (Since you've got 10,000 items, then it might be justified, but
under most circumstances, it works just fine to leave things alone until
they go out of scope.  Using del is not often necessary.)

Jeff Shannon
Technician/Programmer
Credit International




From clpy at snakefarm.org  Tue Feb 12 14:29:32 2002
From: clpy at snakefarm.org (Carsten Gaebler)
Date: Tue, 12 Feb 2002 20:29:32 +0100
Subject: SystemError: unknown opcode
Message-ID: <3C696D1C.AE3F87BE@snakefarm.org>

Hi there,

I just got an exception from a script that is regularly run via cron.
The script is supposed to choose a random signature file, which it
does perfectly well. Just every now an then I receive the following
error. I'd be glad if anyone could explain what's happening. It's
Python 2.2, the script is run every 20 seconds.

XXX lineno: 454, opcode: 0
Traceback (most recent call last):
  File "/home/cg/bin/sigrotate.py", line 5, in ?
    sigs = glob.glob("/home/cg/.sig[0-9]*")
  File "/usr/local/python-2.2/lib/python2.2/glob.py", line 37, in glob
    sublist = glob1(dirname, basename)
  File "/usr/local/python-2.2/lib/python2.2/glob.py", line 50, in
glob1
    return fnmatch.filter(names,pattern)
  File "/usr/local/python-2.2/lib/python2.2/fnmatch.py", line 47, in
filter
    _cache[pat] = re.compile(res)
  File "/usr/local/python-2.2/lib/python2.2/sre.py", line 178, in
compile
    return _compile(pattern, flags)
  File "/usr/local/python-2.2/lib/python2.2/sre.py", line 226, in
_compile
    p = sre_compile.compile(pattern, flags)
  File "/usr/local/python-2.2/lib/python2.2/sre_compile.py", line 430,
in compile
    p = sre_parse.parse(p, flags)
  File "/usr/local/python-2.2/lib/python2.2/sre_parse.py", line 623,
in parse
    p = _parse_sub(source, pattern, 0)
  File "/usr/local/python-2.2/lib/python2.2/sre_parse.py", line 318,
in _parse_sub
    if source.match("|"):
  File "/usr/local/python-2.2/lib/python2.2/sre_parse.py", line 454,
in _parse
    min, max = 0, MAXREPEAT
SystemError: unknown opcode

And this is the script:

[cg at cg cg]$ cat bin/sigrotate.py
#! /usr/bin/python -S

import glob, whrandom, os

sigs = glob.glob("/home/cg/.sig[0-9]*")
sig = whrandom.choice(sigs)
open("/home/cg/.sigtemp", "w").write(open(sig, "r").read())
os.rename("/home/cg/.sigtemp", "/home/cg/.signature")


Regards
Carsten.
-- 
'In iteger arithetric divsion is no the oposite of multiplication.'
                                   -- scenes from comp.lang.python


From sjones16 at rochester.rr.com  Wed Feb 27 23:05:12 2002
From: sjones16 at rochester.rr.com (Seth Jones)
Date: Thu, 28 Feb 2002 04:05:12 GMT
Subject: problems with photoimage and tkinter
Message-ID: <Y%hf8.149154$dh.38770971@typhoon.nyroc.rr.com>

Hi, I'm looking for some help using images in tkinter. I've spent hours with
it, and I can't
seem to get anything working. Well, I have something. The following code is
from the Python Imaging Library 'painter.py'.. This is the minimum necessary
to display an image on the canvas.
The line

                #self.tile[(x,y)] = box, tile

must be uncommented in order for this code to work, and for the life of me,
I can't imagine why. I can't see how it has anything to do with the rest of
the code, but over and over, I've commented it and ran the code and
uncommented it and ran it. The results are consistent. (It was needed for
some code that I deleted.) The file being loaded is a 160 x 120 Windows
Bitmap. I'm using python 2.2 on windows 95. What i'd like is to be able to
just display a single color image on a canvas. I'd really appreciate some
help; I was in love with python until I spent the past few days being
bewildered by this. (^ :

Thanks to anyone who can enlighten me.

Keith Jones

----------------------------------------------
from Tkinter import *
import Image, ImageTk
import sys

class PaintCanvas(Canvas):
    def __init__(self, master, image):
        Canvas.__init__(self, master, width=image.size[0],
height=image.size[1])

        self.tile = {}
        self.tilesize = tilesize = 32
        xsize, ysize = image.size
        for x in range(0, xsize, tilesize):
            for y in range(0, ysize, tilesize):
                box = x, y, min(xsize, x+tilesize), min(ysize, y+tilesize)
                tile = ImageTk.PhotoImage(image.crop(box))
                self.create_image(x, y, image=tile, anchor=NW)
                #self.tile[(x,y)] = box, tile


root = Tk()

im = Image.open('d:\\cyberdeck\\firstattempt\\b3.bmp')

if im.mode != "RGB":
    im = im.convert("RGB")

PaintCanvas(root, im).pack()

root.mainloop()
sys.exit()
------------------------------------------




From kp87 at lycos.com  Tue Feb 26 12:41:21 2002
From: kp87 at lycos.com (kevin parks)
Date: 26 Feb 2002 09:41:21 -0800
Subject: Iterators and generators for the masses
References: <5e8bd451.0202250924.70f47df1@posting.google.com> <mailman.1014659385.24579.python-list@python.org>
Message-ID: <5e8bd451.0202260941.71c31966@posting.google.com>

> This is a pretty good article:


It's great! Thank you thank you thank you! It is really helpful *AND*
by the starngest coincidence, the function they are discussing (random
walk) is one that i am trying to implement in Python (My query on pink
and brown noise curiously went un answered (twice!), but as i am
struggling to understand 1/f^1 noise (and other fractional noises) I
haven't pushed it.

Thanks so much for the heads-up. That web site has a lot of
interesting items.

<kp>


From logiplexsoftware at earthlink.net  Fri Feb 15 18:12:21 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 15 Feb 2002 15:12:21 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <a4k2si$1skr$1@norfair.nerim.net>
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
	<a4k2si$1skr$1@norfair.nerim.net>
Message-ID: <20020215151221.7cb5124a.logiplexsoftware@earthlink.net>

On Fri, 15 Feb 2002 23:49:44 +0100
Gillou wrote:

> And when downloading another one he gets the python interpreter again,
and a
> new copy of some packages...

Not necessarily.  There could be a "full" install that gives you pretty
much everything (as it does now), and a "lite" version that includes only
the python interpreter and _required_ libraries.  Additional libraries
would be available that do _not_ include the interpreter, so you could
download the "lite" version and add the minimum required libs for your
application.

> No, Python deserves a dedicated RPMfind  like utility that manages
> dependencies (OS, python version, required modules...).
> This would be a really great extension for distutils.

Unfortunately, that's a bad example.  I've rpm'd myself right into OS
reinstalls just trying to test some new application that requires (or
thinks it does) some newer library than I have (and that library is
depended upon by some major component, say GNOME).  It may have more to do
with the rpm packagers than rpm itself, but IMHO rpm does a fairly poor job
of managing dependencies.  Frankly I'm starting to prefer tarballs over
rpms... 

Sorry to go OT, but I like to bitch about things that annoy me ;)

However, I do agree that a tool that manages library interdependencies
would be a good thing.  Apparently it's easier said than done (right)
though.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From grante at visi.com  Fri Feb 22 17:04:56 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 22 Feb 2002 22:04:56 GMT
Subject: Stackless Platform Independence?
References: <mailman.1014406958.15414.python-list@python.org> <m37kp5i5a4.fsf@chvatal.cbbrowne.com>
Message-ID: <slrna7dg48.gtf.grante@isis.visi.com>

In article <m37kp5i5a4.fsf at chvatal.cbbrowne.com>, Christopher Browne wrote:
> Centuries ago, Nostradamus foresaw when Christian Tismer <tismer at tismer.com> would write:
>> Dear Stackless Python Community,
>>
>> As you might have noticed, Stackless has changed quite
>> much:
>> - Platform independence is gone.
> 
> I hope not...  I would hope to hear that "platform indepedence has
> _ARRIVED_"

TANSTAAFL.  IIUC, platform independance has been sacrificed to make the
other gains that were listed in the post.

-- 
Grant Edwards                   grante             Yow!  Sign my PETITION.
                                  at               
                               visi.com            


From jkraska at san.rr.com  Tue Feb  5 01:29:44 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 05 Feb 2002 06:29:44 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c5f72c6$1_2@corp.newsgroups.com>
Message-ID: <qvuu5uc9dcmanb8g4d49anrf3lf02hshqt@4ax.com>

>Looking at what you proposed, I'm not sure I understand what differentiates
>your "lingo" from a common function.  A function seems to be everything
>you've defined for lingo.  Is there some problem that using function calls
>can't solve?

Did you read the examples? Python doesn't have a do/while form. Or a switch
form (not shown). Lingos are intended to address to demands of programmers,
many of whom currently use lisp, who have a need to generate domain-specific
languages as dialects of the language they are using. While lingos aren't
complete, they are a start. This is a specialized application and not really
at all Pythonic. But that's a different issue. I digress.

You are confused. Consider the following:

>>> def f(i): i = 3

>>> i = 4
>>> f(i)
>>> i
4
>>> 

Using a lingo, i would be 3.

Return or yield in a lingo might or might not (undecided) cause the
caller to return or yield. Haven't thought that through yet. If it
were a macro, it clearly would.

One possible way to get proper behavior while using a macro is to
open a first class lexical scope in the definition and then otherwise
insert the entire AST of the defining body into the caller. This would
work correctly, I think, and possibly is the proper implementation. TBD.

C//



From m.hadfield at niwa.co.nz  Sun Feb 24 22:01:41 2002
From: m.hadfield at niwa.co.nz (Mark Hadfield)
Date: Mon, 25 Feb 2002 16:01:41 +1300
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014418936.10783.python-list@python.org> <3C799833.AC08A401@cosc.canterbury.ac.nz>
Message-ID: <a5ca53$4ma$1@newsreader.mailgate.org>

"Greg Ewing" <greg at cosc.canterbury.ac.nz> wrote in message
news:3C799833.AC08A401 at cosc.canterbury.ac.nz...
>
>      ACHTUNG!
>
> Dies Komputermaschine ist nicht fur gerfingerpoken

That's "gefingerpoken". A search for this word on Google turns up 656
articles. This one...

  http://manetheren.cl.msu.edu/~vanhoose/humor/0261.html

...has the gefingerpoken quote & many others, of which I especially
like:

"Usenet is like a herd of performing elephants with diarrhea--
massive, difficult to redirect, awe-inspiring, entertaining, and a
source of mind-boggling amounts of excrement when you least expect
it."

---
Mark Hadfield
m.hadfield at niwa.co.nz  http://katipo.niwa.co.nz/~hadfield
National Institute for Water and Atmospheric Research




From schneida at seiu.org  Wed Feb 27 15:13:57 2002
From: schneida at seiu.org (Anders Schneiderman)
Date: 27 Feb 2002 12:13:57 -0800
Subject: Printing to a "log screen" in a multi-module QT app
Message-ID: <54abe19d.0202271213.5e6ef2b0@posting.google.com>

I'm writing an application using QT.  Right now when I want to use the
equivalent of "print", I call a method called "log" that's defined in
the main class in my main module:

def log(self, msg):
   try:
      self.logArea.insertline(msg, -1)
   except:
      self.logArea.insertline(str(msg), -1)

To call it in other classes that are in other modules of the app, I
pass "log" to the new class.  For ex:

class CodeEditor:
   def __init__ (self, log, ...)
       self.log = log

Is there a cleaner way to do this?  For example, is there a way to
redefine the "print" command?  Whatever solution I come up with, the
command to print has to be short;  "self.logObject.log('File
successfully updated')" is too much of a pain.  Any suggestions?

Thanks,
Anders Schneiderman
Information Manager
SEIU International


From sandskyfly at hotmail.com  Wed Feb  6 05:27:13 2002
From: sandskyfly at hotmail.com (Sandy Norton)
Date: 6 Feb 2002 02:27:13 -0800
Subject: Please stop responding to Rue
Message-ID: <b03e80d.0202060227.368f6ea7@posting.google.com>

"Whereof one cannot speak, thereof one must be silent."


From gimbo at ftech.net  Tue Feb 26 04:14:09 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 09:14:09 +0000
Subject: Editors for Python
In-Reply-To: <slrna7l7k6.74f.huaiyu@gauss.almadan.ibm.com>
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com> <slrna7l7k6.74f.huaiyu@gauss.almadan.ibm.com>
Message-ID: <20020226091409.GB18069@andy.tynant.ftech.net>

On Mon, Feb 25, 2002 at 08:28:54PM +0000, Huaiyu Zhu wrote:

> This might not help you, but it may be of interest to other Emacs users:
> 
> M-x speedbar 

Oh wow.  I love you.  You just made my week.  Thanks!

-Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From James_Althoff at i2.com  Thu Feb 28 20:06:54 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Thu, 28 Feb 2002 17:06:54 -0800
Subject: Status of PEP's?
Message-ID: <OFCD6A7C3F.5FA0DF1E-ON88256B6F.000363A1@i2.com>

[David Eppstein]
> I would argue with the "without success" opinion you express.
> I still like
>     for lower_bound <= var < upper_bound
> (or similar three-way comparisons, with middle argument of
> comparison required to be a variable), in parallelism with the current
> "for var in seq" with first argument of "in" required to be a variable.
> I guess we haven't made it into a PEP for the BDFL to rule on, if that's
> your measure of success, but the only arguments I can remember seeing
> against it are (1) some people view "in" as a keyword integrally attached
> to "for" rather than part of an expression, and don't like for loops
without
> "in", and (2) this isn't sufficiently terse for those situations where
you
> want an integer range outside of a for loop.  Am I misinterpreting the
> previous discussion?

I think this comment illustrates that a PEP would, in fact, be useful as a
way to focus discussion on this specific approach to dealing with integer
intervals.

For example, one problem I had when playing with an experimental
implementation of the "middle argument" variant mentioned above is that the
specific relational operators are tied to the values of the bounds.  If you
want to iterate from the smaller value (up) to the larger value you write:

for 1 <= var < 5:

whereas if you want to iterate from the larger value (down) to the smaller
value you have to turn things around and write:

for 5 > var >= 1:

which all looks very nice when you are dealing with literal constants.  But
suppose you have variables for the bounds, "start" and "end"?  Let's say I
have a function

def doSomethingFromTo(start,end):

The function has to test to see which of "start" and "end" is the smaller
value in order to know whether to execute a for-loop with "<" or a for-loop
with ">" (and the code specified in each separate for-loop has to be
written redundantly).  I found this to be awkward.

Another issue that I have is that the syntax does not work outside of a
for-loop header.  I want to be able to create integer intervals anywhere in
my code, not just in the header of a for-loop.

If a large-scale change (that includes new syntax) is to be made in Python
to handle general intervals, I would much rather see an approach that does
not have these drawbacks.

Other issues might surface if there were a proper PEP and discussion of
this specific suggestion.

Jim






From pedronis at bluewin.ch  Tue Feb  5 19:31:01 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Wed, 6 Feb 2002 01:31:01 +0100
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <3c6011e3$1_5@news.bluewin.ch> <tor06u0j1ce99ibgd98uflckcms89apfci@4ax.com>
Message-ID: <3c607be8$1_1@news.bluewin.ch>

Courageous <jkraska at san.rr.com> wrote in message
tor06u0j1ce99ibgd98uflckcms89apfci at 4ax.com...
>
> >3. Given the absence of closure blocks
> >local vars clashes should be handled through
> >renaming (?) (issues?)
>
> If I understand you properly, I'm considering making the the
> macro call itself _generate_ what amounts to a poor man's
> closure block. I have a few implementation details for this,
> but the trick to understanding it is to understand that no
> _first class_ form needs to exist to create the block, only
> the backend facility.
>

Yup you don't need first class forms.

Q: are you working on modifying the C Python parser
and compiler? What do you mean by poor man's closure?

Personally having some time I always  considered to start
with a pure python import hook implementation.
Error handling is poorer but working over a vanilla
interpreter can get you more testers. And you have
more prototyping gas ;) .

regards, Samuele  Pedroni.








From QnickQm at alum.mit.edu  Thu Feb 28 10:46:09 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Thu, 28 Feb 2002 15:46:09 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <slrna7rdna.5m1.QnickQm@localhost.localdomain> <3c7dcc76.261969897@mammoth.usenet-access.com>
Message-ID: <slrna7skc3.8fp.QnickQm@localhost.localdomain>

In article <3c7dcc76.261969897 at mammoth.usenet-access.com>, A. Jones wrote:
> On Thu, 28 Feb 2002 04:46:31 GMT, QnickQm at alum.mit.edu (Nick
> Mathewson) wrote:
> 
> Yeah.  I've designed dozens of ciphers before, all of which fail
> horribly the second I think about frequency analysis.  On the other
> hand, I've never tried actual encryption... I don't quite grok modular
> mathematics yet enough to do anything but monkey-see-monkey-do copies
> of other opensource ciphers.

Well, if you're going to copy, you may as well spend a while trying to
implement ciphers the crypto community already believes to be secure.  

See:
       http://www.faqs.org/faqs/cryptography-faq/part04
       http://www.faqs.org/faqs/cryptography-faq/part05
       http://www.faqs.org/faqs/cryptography-faq/part06
 
and have fun!  But see...
       http://www.faqs.org/faqs/cryptography-faq/snake-oil
   
  [...]
>>Well, I hope this that didn't dampen your entusiasm for Python!
>>You're off to a good start (esp. for two days!), and I hope we'll hear
>>more of you in the future!
> 
> Good Lord, no, it didn't dampen my enthusiasm.  Hell, it only
> heightens it knowing that there're people in this community who're
> willing to take a look at a total newbie's code and then spend the
> time not only commenting but actually *re-writing* it.  Thank you.

No trouble at all.  That's one of Python's Sneaky Advantages: because
it's so fun to write, people love to tweak it, and so they wind up
giving good advice over python-list/c.l.py.

It's-all-part-of-Guido's-master-plan-ly Yrs,

-- 
 Nick Mathewson    <Q nick Q m at alum dot mit dot edu>
                      Remove Q's to respond.  No spam.


From phr-n2002a at nightsong.com  Mon Feb  4 19:20:09 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 16:20:09 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com> <k7F78.46094$KM2.1548938@atlpnn01.usenetserver.com>
Message-ID: <7xheow3hl2.fsf@ruckus.brouhaha.com>

"Steve Holden" <sholden at holdenweb.com> writes:
> > No.  I'm taking a practical view of this.  I use HTML tables in my
> > webpages even though there's still a few ancient browsers out there
> > that don't display them.  But if 25% of browsers didn't support them,
> > I'd avoid using them when I didn't need them.  I do avoid making my
> > pages depend on Javascript, for basically that reason.
> >
> "When I didn't need them": surely "must run in old browsers" is not a
> partial requirement. Do you really specify your output format as "must
> remain compatible with Netscape Navigator 2.0 unless it's really important
> not to". No wonder you have problems with the meaning of "backward
> compatibility" <0.9 wink>.

Of course it's frequently a partial requirement, not "must run in old
browsers" but "should run in old browsers unless there's a good reason
not to".  For example,

   http://www.nightsong.com/crypto/dice.php

is a page that uses javascript to do something that can't be done
legitimately without client side scripting (an applet would also
work).  That's a page that does something worthwhile, that won't work
in a non-JS-enabled browser.  (It generates random passphrases from
user-supplied random data).  I used javascript in that page because it
was needed.  That doesn't mean I shouldn't avoid javascript in other
pages where it isn't needed.

> > The question is, what percentage of users are using what Python version?
> >
> Well, if you want to take a count, add me in to the user population of
> 1.5.2, 2.0, 2.1.2 and 2.2. And I think there are a few beta copies lying
> around from pre-2.2. <grin>

Sounds about normal.

> Well, the major problem for most would-be Python software authors right now
> is that Red Hat are still shipping 1.5.2 without any real attempt to conceal
> it from a user wanting to install some later version from the RPMs. 

I think RH 7.2 has Python 2.1 installed as "python2.1" or something
like that.  Running "python" still gives 1.5.2.

> Building their system so it's hard to upgrade Python without
> breaking Anaconda was dumb. Defensible, but dumb. And I think it
> hurts Python more than the benefit of being able to say "Red Hat
> built Python into their system maintenance software."

They read the python web page which claimed that Python was a stable
language and assumed it meant what it said and that their scripts
would keep working.

> I appreciate you are keen to distribute "nifty Python programs". If so, you
> might just have to accept the bloat of building installer versions, untilt
> someone comes up with a way for installers to determine whether Python is
> currently available in an appropriate version or not.

I don't know what I'm going to do about the installer situation.  I'd
like Windows users to be able to use my stuff so I'll probably have
to deal with the installer thing.  Unfortunately (but reasonably) the
installers can only be built on Windows systems, which I currently
don't use.

> These engineering issues are what will stand in the way of Python's
> increasing popularity, which is why I prefer to address them when I see them
> raised rather than ignore them.

I agree with this--among other things, maturity of software means
accepting having to run legacy code and rarely having to tell people
to install new versions of stuff.  

> If you insist that Python isn't backwards compatible, I have to
> think what you must think about VB.NET, where they really *have*
> broken backward compatibility, and in a bug way. Unless VB
> application authors are really careful (and many VB programmers
> aren't familiar with the concept of software engineering) we're
> going to see some really badly broken applications there.

I don't know any specifics about VB.NET or .NET.  I don't use Windows
when I can help it.  To paraphrase Barney Frank, everything about VB
probably violates the sodomy laws of most states ;-).

> By comparison, Python is a hotbed of sanity. So start working on
> 1.5.3 right now. If you ever produce it I'd like to think it would
> get a prominent placed on the website. But remember: with very minor
> exceptions, the rule for bugfix releases is "no new functionality",
> so don't expect to sneak augmented assignments in there, for example!

Would anyone actually WANT to use a 1.5.3 release if it existed?
(Not that I'm volunteering).  I assume 2.1.2 was done because someone
wanted to use it rather than switching to 2.2.  But 1.5.anything
is crufty enough that I'd use 2.1.2 or 2.2 in any new installation.


From aahz at panix.com  Wed Feb  6 09:27:24 2002
From: aahz at panix.com (Aahz Maruch)
Date: 6 Feb 2002 06:27:24 -0800
Subject: Web Hosting - one report
References: <283adf56.0202060503.6556b483@posting.google.com>
Message-ID: <a3regc$6fo$1@panix2.panix.com>

In article <283adf56.0202060503.6556b483 at posting.google.com>,
Kevin Dahlhausen <kdahlhaus at yahoo.com> wrote:
>
>I thought I had found a place at 'Hostdepartment.com'  $5/month for
>100Mb and a full shell. They have 1.6 installed.  You can compile your
>own python, but many modules do not build.  It turns out that they
>have disabled dynamic loading of modules from user accounts.  So if
>you have a swigged extension, python can load any shadow files, but
>the .so is never found.  That's a shame, as I was hoping to actually
>post an *answer* to this reoccuring question.  Personally, I need to
>be able to use extension modules - at least PIL, GD charts, etc.

Just use Panix.  $100/year for 50MB and a full shell.  It's pretty
cheap, but service/support is still excellent.  Current default version
of Python is 2.1.1, which reminds me to request an upgrade to 2.1.2 and
an install of 2.2.

You do have to add $50/year if you want your own domain name to be used.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From pereira at cis.upenn.edu  Tue Feb 26 08:25:28 2002
From: pereira at cis.upenn.edu (Fernando Pereira)
Date: Tue, 26 Feb 2002 08:25:28 -0500
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de>
Message-ID: <B8A0F6F8.3C3B%pereira@cis.upenn.edu>

On 2/26/02 4:32 AM, in article m3pu2sfv0m.fsf at mira.informatik.hu-berlin.de,
"Martin v. Loewis" <martin at v.loewis.de> wrote:

> Nick Collier <nick at src.uchicago.edu> writes:
> 
>> Anyway, I thought that the Python grammar was LL(1) compliant and
>> yet when I translate the grammar into the Coco/R LL(1) parser
>> generator I get lots of errors like:
> 
> [...]
>> Where X and Y are productions such as arglist, etc.
> 
> My guess is that your translation into Coco/R is incorrect. I cannot
> spot any ambiguity in arglist. Considering
> 
> arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**'
> test)
> 
> it may be that, given "argument "," argument ","', it finds itself
> incapable of determining whether this is the second argument belongs
> to the first subexpression or the second one.

This production cannot be disambiguated top-down with *any* lookahead, for
the reason you suggest. Thus, it doesn't seem Python's grammar is LL(1). In
general, languages with infix operators and complex nested structures like
lists are not LL(k), because a parser can only decide the role of a
subexpression after it sees the operator or terminator that follows it. But
such languages are easily parsed bottom-up (shift-reduce) with bounded
lookahead. That rule (in context) can be disambiguated bottom-up with two
tokens of lookahead: once an 'argument" is parsed and pushed onto the
shift-reduce stack and a "," is found, if the next token is ")" then the top
of the stack is the second occurrence of "argument", otherwise it's the
first.

-- F



From sholden at holdenweb.com  Fri Feb  8 08:44:13 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 08:44:13 -0500
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <3C629FF9.CF4670FF@metaslash.com> <a3u7hl$1an8dd$1@ID-98166.news.dfncis.de> <7xpu3gixt2.fsf@ruckus.brouhaha.com>
Message-ID: <BCQ88.105314$KM2.4016626@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xpu3gixt2.fsf at ruckus.brouhaha.com...
> "MDK" <mdk at mdk.com> writes:
> > > from  mymod import *
> > > mymod_dosomething = dosomething
> > >
> > > def dosomething(x):
> > >     mymod_dosomething(1000)
> > >
> > > Neal
> >
> > Neat!  Thanks, Neal.
>
> Alternatively, in Python 2.x,
>
>   from mymod import dosomething as mymod_dosomething
>
> Generally, it's best to avoid "from mymod import *", precisely
> because of collisions like this.  As other people have mentioned, it's
> better to say
>
>   import mymod
>
>   def dosomething(x):
>     mymod.dosomething(1000)

We should not overlook the third alternative (nobody expects ...) which is
to write

    from mymod import dosomething as banana

    def dosomething(x):
        banana(1000)

When I use the mxDateTime module I frequently write

    from mx.DateTime import DateTime as dt

just because (as regular readers will know) I'm a crappy typist.

still-wondering-why-dosomething-has-an-argument-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From caljason76 at yahoo.com  Wed Feb 20 17:42:12 2002
From: caljason76 at yahoo.com (Jason)
Date: 20 Feb 2002 14:42:12 -0800
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu>
Message-ID: <e9e6dd1f.0202201442.35c75d87@posting.google.com>

sort() does not return the sorted list, it returns None.  sort() does
an inplace sort and is not a fuctional construct; it does not generate
a new sorted list.

> However when I use:
> 
> for i in d.keys().sort():
>   do something
> 
> I get:
> 
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: loop over non-sequence
> 
> But I know that d.keys() returns a list to which the sort() function should 
> be applicable.
> 
> When I do:
> 
> k = d.keys()
> k.sort()
> for i in k:
>   do something
> 
> it works fine.
> Why can't I use d.keys().sort() directly? And is there any other more 
> concise (elegent?) way to loop over sorted dictionary keys?
>


From jlh at cox.net  Wed Feb 20 01:24:55 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Wed, 20 Feb 2002 06:24:55 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com>
Message-ID: <XiHc8.640$XT1.47512@news1.east.cox.net>

If you wanted to get away from the dangerous eval, you could put your
functions inside of a class and then,
class foo:
    def func(self):
        return 'hello'

y=foo()
x =getattr(y,"func")
print x()

-Jeff

"Grant Edwards" <grante at visi.com> wrote in message
news:slrna769j8.1hv.grante at tuxtop.visi.com...
> In article <a4v7s6$13tk at r02n01.cac.psu.edu>, Rajarshi Guha wrote:
> > Hi,
> >   is it possible to have a variable contain a function name, and use the
> > variable to call the function?
> > An example:
> >
> > def func:
> >    some code
> >
> > funcvar = 'func'
> >
> > Now use funcvar (somehow!) to call func
>
> How about this:
>
>     def func():
>         print "hello"
>
>     x = func
>
>     x()
>
> It doesn't use the name of the function, but if you really want
> to do it that way you can do this:
>
>     def func():
>         print "hello"
>
>     x = "func"
>
>     eval(x)()
>
> --
> Grant Edwards                   grante             Yow!  I want a
VEGETARIAN
>                                   at               BURRITO to go... with
>                                visi.com            EXTRA MSG!!




From philh at comuno.freeserve.co.uk  Mon Feb 11 16:01:45 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 11 Feb 2002 21:01:45 +0000
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net>
Message-ID: <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk>

On Mon, 11 Feb 2002 10:59:36 GMT, Fredrik Lundh <fredrik at pythonware.com> wrote:
>    "People tend to hold overly favorable views of their
>    abilities in many social and intellectual domains. The
>    authors suggest that this overestimation occurs, in
>    part, because people who are unskilled in these
>    domains suffer a dual burden: Not only do these
>    people reach erroneous conclusions and make un-
>    fortunate choices, but their incompetence robs
>    them of the metacognitive ability to realize it"
>
>    http://www.apa.org/journals/psp/psp7761121.html
>
>of course, Rue is going to argue that this doesn't apply to him, but
>to everyone else.  after all, he KNOWS he's smarter than everyone
>else, and we're just too stupid to realize how stupid we are ;-)

Either that or we're part of the Evil Programmers Conspiracy to make
programming hard.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From grante at visi.com  Tue Feb 19 23:30:15 2002
From: grante at visi.com (Grant Edwards)
Date: Wed, 20 Feb 2002 04:30:15 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu>
Message-ID: <slrna769j8.1hv.grante@tuxtop.visi.com>

In article <a4v7s6$13tk at r02n01.cac.psu.edu>, Rajarshi Guha wrote:
> Hi,
>   is it possible to have a variable contain a function name, and use the 
> variable to call the function? 
> An example:
> 
> def func:
>    some code
> 
> funcvar = 'func'
> 
> Now use funcvar (somehow!) to call func

How about this:

    def func():
        print "hello"
    
    x = func

    x()

It doesn't use the name of the function, but if you really want
to do it that way you can do this:

    def func():
        print "hello"
    
    x = "func"

    eval(x)()

-- 
Grant Edwards                   grante             Yow!  I want a VEGETARIAN
                                  at               BURRITO to go... with
                               visi.com            EXTRA MSG!!


From eppstein at ics.uci.edu  Mon Feb 25 16:14:42 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Mon, 25 Feb 2002 13:14:42 -0800
Subject: functional programming with map()
References: <mailman.1014667482.23800.python-list@python.org>
Message-ID: <eppstein-FEA312.13144225022002@news.service.uci.edu>

In article <mailman.1014667482.23800.python-list at python.org>,
 Steven Majewski <sdm7g at virginia.edu> wrote:

> List comprehensions are declarative (no loops) and they avoid side effects
> (no loop variables or temporaries)

No side effects?  Try this:

L = [x for x in range(10)]
print x
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From logiplexsoftware at earthlink.net  Fri Feb  8 19:56:29 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 16:56:29 -0800
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <a41oi1$a22$1@panix2.panix.com>
References: <mailman.1013196740.22073.python-list@python.org>
	<a41oi1$a22$1@panix2.panix.com>
Message-ID: <20020208165629.4a6ce25d.logiplexsoftware@earthlink.net>

On 8 Feb 2002 15:55:45 -0800
Aahz Maruch wrote:

> * Yes, I'm aware that prices have come down a lot, but the equivalent
> flat-panel display is still twice as expensive.  And I *won't* buy
> anything at Fry's.

What's wrong with Fry's?  As long as you don't listen to the salespeople,
that is.

 
> I have to admit that I'm seriously thinking about an iMac, though.

Yikes.  That's an expensive Linux box ;)


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From logiplexsoftware at earthlink.net  Thu Feb 21 12:46:11 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 21 Feb 2002 09:46:11 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <de3fc1ef.0202201233.2228e5b1@posting.google.com>
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
	<de3fc1ef.0202201233.2228e5b1@posting.google.com>
Message-ID: <20020221094611.5daff60c.logiplexsoftware@earthlink.net>

On 20 Feb 2002 12:33:23 -0800
William Tanksley Google wrote:

> rmunn at pobox.com (Robin Munn) wrote:
> 
> > * Batteries included. This is the big one, which makes Python more than
> 
> My friend was griping about one battery that Basic comes with but
> Python doesn't (at least not that I've seen): graphics.  Yes, I know
> there's no single way to draw in Basic, but every version has a
> platform-specific graphics library.
> 
> Does Python?  If so, I've never seen it.  I'd like to.
> 
> Personally, I need pixel graphics.  Others will want vector graphics.

There are several ways to do graphics with Python, depending upon your
needs.  Most of the GUI toolkits (wxPython, Tkinter, etc) provide ways of
drawing pixels, blitting images, etc.  Additionally there are toolkits like
pygame, OpenGL, PIL, etc for doing more complex graphics and image
manipulation.  Best of all, they're portable, which I've yet to see in any
version of BASIC.  In my opinion, nothing in any version of BASIC can come
close to the graphics support in Python (although, admittedly, it isn't
built in).

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From h.jehadalwan at student.kun.nl  Fri Feb 22 09:26:34 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Fri, 22 Feb 2002 15:26:34 +0100
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com> <3C74AAA0.D3A050EE@macrozon.net> <3C7571A2.41D4E9E2@ccvcorp.com>
Message-ID: <3C76551A.7060107@student.kun.nl>

Jeff,

the Popen2 module does not work. I'm trying to test it with these lines:



 >>> passout,passin=os.popen2("/home/husam/profit_1/profit")
  # Also tried  calling this with "popen2.popen2()" no go.

 >>> passout.write("reference /chemistry/pdb_files/Complex_1g.pdb\n")
 >>> passout.write("mobile    /chemistry/pdb_files/Complex_2g.pdb\n")
 >>> passout.write("fit\n")
 >>> passin.read()
# Here it hangs, the only way to stop it is by pressing: ctrl+c
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
KeyboardInterrupt



From nospam at sebsauvage.net  Fri Feb  1 08:19:59 2002
From: nospam at sebsauvage.net (Sébastien SAUVAGE)
Date: 01 Feb 2002 13:19:59 GMT
Subject: Python like the old-time batch files.
References: <a3bhcf$176195$1@ID-98166.news.dfncis.de> <a3bk4b$11u4$1@norfair.nerim.net> <Xns91A86D0EAA121sebsauvagedotnet@127.0.0.1> <Xns91A882B25A2BCmhueningzedatfuberli@130.133.1.4>
Message-ID: <Xns91A89267C7EC6sebsauvagedotnet@127.0.0.1>

Matthias Huening <mhuening at zedat.fu-berlin.de> wrote in 
news:Xns91A882B25A2BCmhueningzedatfuberli at 130.133.1.4:

> "S?bastien SAUVAGE" <nospam at sebsauvage.net> wrote in 
> news:Xns91A86D0EAA121sebsauvagedotnet at 127.0.0.1:
> 
>> http://effbot.org/efflib/console/
> 
> Try this one:
> http://effbot.org/downloads/
> 

Thank you !    :-)

-- 
S?bastien SAUVAGE
sebsauvage at sebsauvage dot net
http://sebsauvage.net


From magnus at thinkware.se  Tue Feb 19 18:12:47 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 19 Feb 2002 15:12:47 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com>
Message-ID: <258fd9b8.0202191512.4f4693ff@posting.google.com>

ajeru at vknn.org (Alexander Jerusalem) wrote in message news:<24c39b2c.0202190059.41874b92 at posting.google.com>...
> The argument that exhaustive testing should be done anyway is
> certainly true. I nevertheless tend to think that it's better to let
> the compiler do as much as possible in terms of error checking and
> then use testing to find the true runtime errors. It just takes much
> less time than starting a test run every time just to find all my
> typos. I don't know if there are test suites that let you run all the
> branches of all your methods and show you all typos at a time. I
> suspect that I'd have to start the test suite for each and every typo
> but I'm not knowledgable enough when it comes to Python test suites to
> be sure of that claim.

I'm trying to adopt the XP habit of writing tests first, and then
only code enough to make the test pass. It's sometimes difficult.
It requires some diciplin not to get carried away while writing the
code, and do more than I've written tests for. When I do get carried
away I end up having to write a whole bunch of boring tests afterwards,
and that feels like a drag, because then the writing of tests is not
part of the creative process. And it's difficult to be sure that I
actually have all the tests I need then.

Normally, the process goes like this:
1. "Hm, what do I want this method to do?" - write a test.
2. "How do I accomplish that?" - write the code.
3. Run all unit tests for the module.

These are small cycles, where I probably don't write more
than ten lines of application code in each cycle. It's
not a new test method for each ten lines of application code,
the test methods grow just a few lines per cycle.

In my current project, I have roughly as many lines of
test code as I have application code, but don't think that 
I spend as much time writing tests as writing running code.
The test methods are mainly written once, while the 
application code is obviously changed a bit now and then... 
Perhaps 20% of my programming is coding of tests. I haven't 
measured, it might be a bit more. It's probably not more
time than I'd spent typing declarations, braces etc in static
languages. ;-)

I can only say that I feel that this works better than any
C, C++ etc project I worked with before.

Having run my entire test suite successfully, I feel fairly confident
that I haven't messed anything up with my last bug fix or new feature.
Having compiled and linked C++ code only made me feel that it was
time to do all the tedious tests...

Running my entire test suite is certainly much faster than any
C++ (or ADA I suspect) build process would be for the same amount
of program content. Running the tests for one module is not slower
than a simple compile would be in C++.

> There's another thing that bugs me with the missing static typing and
> that is that it makes code less readable. I don't know what I can do
> with a function parameter when I see it. For example, when I see a
> function like def computeSalary(person)... I don't know at first sight
> if person is the id of person, a person object and if it's an object

Well, I don't know any programming language that forces you to write
good variable names. If it's an ID, it should be "personId", not 
"person" in my opinion. You might even claim that such a misnaming
is a bug!

Imagine that the ADA programmer had written computeSalary(Id : in Oid)
Then you would have known that it was an object id, but not for what!
Both ADA and Python give fairly good support to produce clarity in 
code, although in very different ways. But you _can_ obfuscate code 
in any language...

Unfortunately I've seen a lot of C++ code where methods have
maybe a dozen parameters, most of the class string, the rest
being integers. How much wiser do you get from that? Then someone 
manages to switch the order between two string parameters...
{And when you look at sequence diagrams from Rational Rose, you
don't see variable names, just "(string, string, string..." :-( }

If you wish, you can naturally have a coding standard saying
that there should be assertions on all method parameters. Then
you get "documentation" which won't go stale.

You can also make assertions that are more qualified than simple
type tests. I see a lot of "string" variables in C++ code where
a maximum length in a database table should really be checked,
and "short" where only 0 to 100 are good values etc. Although in
this regard, ADA is much more supportive than C++.

Ok, asserts aren't checked on compile, but surely you need to run 
enough tests to call each method in every class!

> what its exact class in a Person class hierarchy is. I have to find
> the place where the constructor is called to find that out. That's not
> a big problem for small programs but if I have many classes that use
> inheritance, it's really not easy to figure out what interface a
> parameter actually supports. This is, however, a minor problem if
> everyone on the team agrees to document parameter types in the doc
> string. The language doesn't enforce it though.

doctest might be a module for you? It's always tricky with
documentation that goes out of date. assert or doc-strings used
by doctest are really verified. That's certainly a good thing.

I see my unit tests as important documentation and examples of how my
classes and methods are to be used. In my code I don't have much more
comments that the doc string that describes the purpose of each 
module/class/method and minor notes about quirks and temporary notes
about things that should be changed etc.

> And the final argument I have for static type checking is that it
> enables method dispatching based on parameter types. In a statically
> typed language you can create two methods that have the same name but
> differ on the parameter types. The correct method will be called for
> you depending on the type of the argument you pass in your call. That
> makes for a quite flexible way of extending a program. You can just
> add another method with the same name and another type to handle a
> special case without touching the existing methods.

I don't think that's a big deal. In python you can very often use
the same method without a single special case check for all those
different types that you would have needed to write separate methods
for in ADA, Java or C++.

If you really want to do completely different things based on
parameter, even though you wan't the name of the called method
to be the same, you can do something like:

import types

class A:

    def myMethod(self, a, b):
        if (type(a), type(b)) == (types.IntType, types.StringType):
            self.__int_string(a, b)
        elif type(a) == type(b) == types.IntType:
            self.__int_int(a, b)
        else:
            raise TypeError

    def __myMethod_int_int(self, a, b):
        ...

    def __myMethod_int_string(self, a, b):
        ...

But I never ran across the need to do that. After all, most
of the time, if you want the method call to have the same name,
you basically want to do the same thing, and in a dynamically
typed language, you will probably be able to share the same lines
of code for most of your method.

> I think type checking could be added as an optional feature, like in
> Dylan, without hurting the character of Python.

Sure. At least for the interface of a class. There has been a
number of proposals on that. In the typically small python methods, 
where local variables are rarely used for more than a screenful, 
I think there is little point in declaring locals, and we don't 
use much globals, do we...
 
> I'm currently using Python to write generators that produce Java code
> from an XML representation. So the runtime system is type checked by
> the Java compiler anyway. I tried to do the generator in Java first
> but the text processing capabilities of Java are just terrible, so I
> came to use Python and I like it for many reasons. It's just that I'm
> using much more time on debugging than I used to in Java where I
> didn't have to care too much about typos because the compiler
> complained anyway...

If we're talking about getting regular expressions to do what you
want etc, I don't think static typing would help you...

But I can recommend the method of writing tests first, and code
after. The unittest module works very well. If I have a python
module named something.py, I will have a unit test module called
something_ut.py.

This will start with "import unittest, something" and end with
if __name__ == "__main__":
    unittest.main()

In between there are classes subclassed from unittest.TestCase.
They can have methods setUp and tearDown that are run before
and after each test, and the actual tests are methods with names
that start with "test". In these tests you use methods such as
self.failUnlessEqual or self.failUnlessRaises etc. It's as simple
as that. You just run something_ut.py every time you changed
something.py.

Unless you have too big methods, these trivial little tests will
nail down more bugs than you would imagine. Every feature you see
that you want, you test for. Every bug that you find, you write a
test for. This will mean that you loose a bit of the speed in writing
python code, but it will still be a lot faster than other languages,
and for larger projects, I think this investment is well worth the
effort.

If you feel that the tests get too complicated, you probably have too
complicated code, and should consider making more, smaller methods.

I've come to a point where I feel like when I learned how to prove
things in mathematics with induction etc. You have this big problem,
but you don't need to solve it. You just look at a little piece at a
time, and if a piece isn't small enough to be (fairly) trivial, you
subdivide it, and solve that. When the pieces are all solved, your
problem is solved too. With the unit tests, you always keep on track!
 
Then there are some tricks to think of. Most typos will after all
give syntax errors or name errors that you find at once. The problem
is if you assign to the wrong name, and create a new variable or
attribute where you intended to update an existing one. For attributes
in classes you can get away from this problem by using __setattr__.
Always have a __setattr__ method, and don't allow any "extra" 
attributes to be set. You can check type here as well.

Ok, it's not found until run-time, but you will clearly spot where
the problem ocurs. Tracebacks are rather good...


From quinn at baht.ugcs.caltech.edu  Sun Feb 24 06:32:58 2002
From: quinn at baht.ugcs.caltech.edu (Quinn Dunkan)
Date: 24 Feb 2002 11:32:58 GMT
Subject: giant stdlib tuples (was: Re: Help with code)
References: <a4votp$mmf$1@infocom.km.ua> <RHNc8.17137$Zp2.1765797@atlpnn01.usenetserver.com> <slrna77qo7.dg2.quinn@schilling.ugcs.caltech.edu> <uadu3m5dz.fsf@python.net> <slrna7agtq.1vk.quinn@regurgitate.ugcs.caltech.edu> <uadu1lrum.fsf@python.net>
Message-ID: <slrna7hjra.i6a.quinn@baht.ugcs.caltech.edu>

On Fri, 22 Feb 2002 10:40:21 GMT, Michael Hudson <mwh at python.net> wrote:
>> pwd and grp still don't return structs, but it's a good start.
>> 
>> Deserves more advertisement, I think:
>
>Perhaps.  Don't try and pickle the results until 2.2.1 gets released...
>
>Anyone feel like contributing code to similarly enhance pwd and grp?

I submitted some patches.


From chris.gonnerman at newcenturycomputers.net  Sun Feb 17 23:06:16 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 17 Feb 2002 22:06:16 -0600
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1> <8867.813T2365T9245607threeseas@earthlink.net> <a4p219$1o0ne$1@ID-11957.news.dfncis.de>
Message-ID: <006101c1b831$9f2e3b40$0101010a@local>

----- Original Message -----
From: "Emile van Sebille" <emile at fenx.com>


> Complete re-posts smell like spam.
>
> I have no doubt that your nine steps can be super-imposed on everything.
> However, as long as the deep magic involved in converting these steps
> into things people want done is kept as magic, let Aahz's quote reign:
>
> "This is Python.  We don't care much about theory, except where it
> intersects with useful practice."  Aahz Maruch on c.l.py

Indeed.  Like many from Missouri, I am a true believer in our state motto:

"Show me"

The nine commands (which we are told are not a language, but rather physics,
and therefore science) are not exactly backed up by much research, proof,
etc.

Many on this list ARE scientists, and would be willing to consider any
proper scientific evidence and/or well-formed theory.

I'm not a scientist, but I worked with them before.  Knew one who created
a 3D chess game which had ten each 10x10 Lucite boards, stacked; with the
attached base, the whole thing was 5 feet tall and beautiful.  He sculpted
the pieces (10 pawns and 10 special pieces per side, including two "Dukes"
added) from fist-sized chunks of plastic.  The rules were based on particle
motion dynamics taken from quantum physics; and perhaps ten people in the
whole world could understand the rules.

Probably could sell it now over the Internet (are you there, Capt. Goff?)

SO, what I am saying is, I've seen science before, and know it when I see
it.  Haven't seen any in Rue's posts.





From boblancaster at zxmail.com  Tue Feb 26 10:01:23 2002
From: boblancaster at zxmail.com (Bob Lancaster)
Date: 26 Feb 2002 07:01:23 -0800
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com> <Xns91C167290C907andy47halfcookedcom@203.109.252.31>
Message-ID: <b470cb58.0202260701.53857ce0@posting.google.com>

Andy,

Thanks for the advice.  I did check to see that the files were in the
ORACLE_HOME\oci\include directory.

I was using the pre-release version of DCOracle2, which I downloaded a
few weeks ago, and Python v. 2.1.1, which I downloaded last December.
I got rid of my old versions, and switched to DCOracle2 v.1.0, and
Python v. 2.1.2.  The release version of DCOracle2 checks for
version[0:3], just as your code did.  In the pre-release version, I
rewrote another part of the code to specifically accept version 2.1.1.
Things work a bit better, but the program crashes if I either (a)
SELECT a TIMESTAMP_WITH_TIMEZONE (I may not be interfacing properly
with the mxDateTime module) or (b) select a user-defined type. 
(Oracle is a bit nasty about arrays, unlike PostgreSQL).
INSERT, UPDATE and DELETE all work fine, even with user-defined types
and TIMESTAMP_WITH_TIMEZONE.
Not only that, but it is really hard to figure out the precise type of
a field from the description (unlike either pgdb or psycopg with
PostgreSQL).
Unfortunately, many of our customers use Oracle, and we have to
interface with a system which uses Oracle, so I probably have no other
choice but to rewrite the database in such a way as to make DCOracle2
happy.  The alternative cx_Oracle is even worse.

Thanks for your advice!
Bob

Andy Todd <andy47 at halfcooked.com> wrote in message news:<Xns91C167290C907andy47halfcookedcom at 203.109.252.31>...
> boblancaster at zxmail.com (Bob Lancaster) wrote in 
> news:b470cb58.0202250816.7839c1c4 at posting.google.com:
> 
[my (Bob L) original post snipped to save bandwidth]
> 
> Bob,
> 
> I think the DCOracle2 documentation is incorrect. All of the header and 
> library files needed to compile OCI clients for Oracle should now (Oracle8i 
> release 2 and above) be included in the %ORACLE_HOME%\oci directory on 
> Windows. For information on which components are necessary to build OCI 
> client modules have a look at the "Oracle Call Interface Getting Started 
> Release x.x.x for Windows" manual provided with your database installation. 
> Mine is for 8.1.6, ymmv. 
> 
> Something that has caught me out in the past is that I have performed a 
> server install but forgot to include the Oracle client software. This then 
> doesn't include the appropriate dlls which will result in run time errors 
> as soon as you try to call them. If you do a 'full' install of Oracle on 
> Windows this shouldn't be a problem, but if you have performed a 'custom' 
> install you may have missed out the appropriate headers and dll files and 
> that would definitely cause issues.
> 
> As far as I know you should have no problems using DCOracle2 with Python 
> 2.1.1 or 2.1.2. I'm running Python 2.1.2 as part of Zope on my Windows2000 
> machine and I have successfully managed to install DCOracle2 into that 
> environment. All I did was change line 14 of install.py from;
> 
> version = string.split(sys.version)[0]
> 
> to
> 
> version = string.split(sys.version)[0][0:3]
> 
> and everything was hunky dory. Hope this is of some help.
> 
> Regards,
> Andy


From risnes at alfanett.no  Fri Feb 15 16:42:07 2002
From: risnes at alfanett.no (André Larsen Risnes)
Date: Fri, 15 Feb 2002 21:42:07 GMT
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
Message-ID: <20020215225431.7bdef34e.risnes@alfanett.no>

Artur Skura <arturs at iidea.pl> wrote:
[...]
>
> Now can you suggest an easy, lazy way of doing this (with any
> toolkit) so things don't pile up and everything looks nice?
> 
Sure (with tkinter):


from Tkinter import *

text = "When is a %text% not like a %text%? When it's a %text%!"
root = Tk()

for part in text.split("%"):
    if part == "text":
        Text(root, width=8, height=1).pack(side=LEFT)
    else:
        Label(root, text=part).pack(side=LEFT)

root.mainloop()


This will align everything on one line, though. If you want to
"break lines" you'll have to construct sub-frames for each
line. I guess this would be easy if the text data is stored in
a line-oriented file, like:


for line in file.readlines():
    for part in text.split("%"):
        f = Frame(root)
        if part == "text"
            Text(f, width=8, height=1).pack(side=LEFT)
        else:
            Label(root, text=part).pack(side=LEFT)
        Label(f, ...)
        f.pack(side=TOP, anchor=W, fill=X, expand=YES)

(haven't tested this last part, though)

-- 
Regards, Andr? Risnes



From python at chajadan.net  Tue Feb 19 00:57:07 2002
From: python at chajadan.net (chajadan)
Date: Mon, 18 Feb 2002 21:57:07 -0800
Subject: help - inserting values into database tables
Message-ID: <5.1.0.14.0.20020218215652.00a41ec0@63.118.78.184>

Hi Joe =)

Your problem shouldn't be too hard to fix.

Using odbc, you can't open the database file directly like that, since odbc 
supports so many different database standards, it needs to know which one.

This requires a DNS. In order to set up a DNS for MS Access, just go to 
your control panel, and you should have an option called "OBDC Datasources" 
or similar. Click that, and it will show you your current DSNs. There are 
several types of DSNs, but you will want to make this file a System DSN, so 
click that tab. Select New, and choose the correct database driver, in this 
case MS Access. Then you have to name the DSN, anything you'd like (avoid 
spaces), chose a login/password or leave that blank, and finally show it 
the database file this whole DSN should be referring to.

Then, you can odbc.odbc("myDsnName")

=)

Let me know if you have any other questions and I'll try to help. I work on 
Win98, so some of what I described may look slightly different under XP, 
but it should be there.

Good luck!

--chajadan
aka charlie

At 06:35 PM 2/18/02 -0600, you wrote:

>Hi again,
>Just wondering if you can help?
>
>I installed the python ver 2.0(from CD)for windowsX on a windows98
>system.  Then I installed the windows32all stuff.
>So are the dbi and the odbc modules should now be installed
>on my system now, or do I need some other package ??????
>
>It seems I can import the dbi then the odbc with no errors...
>But my simple script below won't run; has an error that I
>don't understand ???????
>
>I am running from the command line, heres my environ stuff
>******* Do I need to put in another path to Lib ?????
>
>COMSPEC=C:\WINDOWS\COMMAND.COM
>windir=C:\WINDOWS
>PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\JDK11~1.8\BIN;C:\PYTHON20;
>
>******** Heres the error I get upon running from prompt
>
>C:\codepy>python conntest.py
>
>Traceback (most recent call last):
>   File "conntest.py", line 8, in ?
>     conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
>dbi.operation-error: [Microsoft][ODBC Driver Manager] Data source name not
>found and no default driver specified in LOGIN
>
>The Manufacturers.mdb is the ms-access database of ver office XP
>2002.  It is just a small database with a small table in it.
>Very simple kind.  The file is in that codepy directory also.
>So I don't understand ?????????????????
>What is DSN, does it allow for a ms-access file??????????
>
>Well anyways heres my test database connection script, I am
>just trying to connect and print some output from the database.
>I think I am doing this correct but not sure ??????????
>
>The script:  conntest.py listing
>#! /usr/bin/env python    #not needed for windowspython
>
>#test script to connect to a ms-access database
>
>import dbi
>import odbc
>
>conn = odbc.odbc("c:\\codepy\\Manufacturers.mdb")
>rs = conn.cursor()
>rs.execute("select * from ManufName")
>print rs.description
>
>rs.close
>conn.close
>
>Can you shine any light my way ???????????
>
>Thank you (working on a windows system today(UGH...)),
>Joe F.
>#####
>
>----- Original Message -----
>From: chajadan <python at chajadan.net>
>Newsgroups: comp.lang.python
>Sent: Monday, February 18, 2002 2:34 PM
>Subject: Re: help - inserting values into database tables
>
>
> > I always use PythonWin, part of the win32all distribution. It's a
> > color-coded ide, with a basic syntax check, and has one nifty advantage
> > over any other input editor I've tried: when you call a function, it takes
> > the document strings and actually gives you a little note about what kind
> > of input it's expecting!
> >
> > So all in all it should be the same thing as it ever was, but now in the
> > environment your particular task seems to ask for. The only "new" things
> > will be the other possible windows-only features you might want to learn
> > and use in your time on windows.
> >
> > Happy computing!
> >
> > --chajadan
> >
> > At 08:15 AM 2/18/02 -0600, mylinuxboxroot wrote:
> > >Hey thanks,
> > >This is great, and I have the project now that it will be developed on a
> > >windows
> > >server so I guess I don't need to worry about going from linux to the
> > >MS-Access
> > >database.  I will develop it right on the server using the win32all
> > >distribution that
>
>snip




From philh at comuno.freeserve.co.uk  Thu Feb  7 07:02:07 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 12:02:07 +0000
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net>
Message-ID: <slrna64r5v.ahr.philh@comuno.freeserve.co.uk>

On Thu, 07 Feb 2002 04:28:38 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:
>>>>>>> Timothy Rue writes:
>
>>> There is only the creation of a configuration of very common code
>>> functionality, carried out to it's natural conclusion in support of
>>> versatility.
>
>>That has already been done.  It's been done over and over for the last
>>50 years of computer programming.  Without a clear definition of what
>>you mean by "common code functionality" to differentiate your approach
>>versus what's already out there, I see nothing worthwhile in your
>>project.
>
>I've defined the nine commands, so show me where this configuration you
>claim has been done over and over again.

Some technologies that do it are subroutines, classes/methods, macros.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jamcguir at yahoo.com  Mon Feb  4 17:15:18 2002
From: jamcguir at yahoo.com (Jake McGuire)
Date: 4 Feb 2002 14:15:18 -0800
Subject: Python readline hooks and threading
Message-ID: <dbb82eb7.0202041415.7fdf442c@posting.google.com>

I'm a bit confused about how the readline module (Modules/readline.c)
is supposed to work, especially with regards to threading.  It *looks*
like the code is broken, and my own testing seems to bear this out,
but I think I may just be confused.

In the most recent version of readline.c [1], there is a comment in
on_hook() that says "Note that readline is called with the interpreter
lock released", followed immediately by a call to
PyThreadState_Swap().  The Python documentation says that the lock
must be held during any calls to PyThreadState_Swap, and looking at
the source shows that PyThreadState_Swap is indeed not thread-safe,
accessing a global variable without any locking.

In my application (multi-threaded, with all threads using C extension
functions and one thread using readline hooks) I was getting all kinds
of tstate mix-up errors.  Debugging showed that although readline is
indeed always called inside of a Py_BEGIN_ALLOW_THREADS block,
sometimes the call to PyThreadState_Swap() returned a non-null value. 
Commenting out the calls to PyThreadState_Swap() in on_hook seemed to
make these go away.

Is this a bug in the current Python distributions, or am I just on
crack?

-jake

[1] http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/python/dist/src/Modules/readline.c?rev=2.42&content-type=text/vnd.viewcvs-markup


From sam_collett at lycos.co.uk  Fri Feb  8 09:46:20 2002
From: sam_collett at lycos.co.uk (Sam Collett)
Date: 8 Feb 2002 06:46:20 -0800
Subject: Python ActiveX
Message-ID: <2030665d.0202080646.3126aec0@posting.google.com>

Is it possible to have python installed as a simple activex control
(like jscript.dll and vbscript.dll)?

Using code similar to the following would install an activex control
(the following code installs msxml ver 4)...
<object classid="clsid:88d969c0-f192-11d4-a65f-0040963251e5"
codebase="msxml4.cab" width="0" height="0">
</object>

Can this be done with python as well e.g.
<object classid="clsid:[python clsid code here]" codebase="python.cab"
width="0" height="0">
</object>

I ask this because the user may not need all the extra features
(Pythonwin win32all etc) - it is quite a big download.

I hope to use it as a scripting language in html code:
<script language="python">
python statements here
</script>


From threeseas at earthlink.net  Wed Feb  6 23:28:38 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:38 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com>
Message-ID: <1192.802T1446T14104708threeseas@earthlink.net>

On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> There is only the creation of a configuration of very common code
>> functionality, carried out to it's natural conclusion in support of
>> versatility.

>That has already been done.  It's been done over and over for the last
>50 years of computer programming.  Without a clear definition of what
>you mean by "common code functionality" to differentiate your approach
>versus what's already out there, I see nothing worthwhile in your
>project.

I've defined the nine commands, so show me where this configuration you
claim has been done over and over again.

>--
>David Masterson                dmaster AT synopsys DOT com
>Sr. R&D Engineer               Synopsys, Inc.
>Software Engineering           Sunnyvale, CA


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From scarblac at pino.selwerd.nl  Wed Feb  6 12:04:28 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 6 Feb 2002 17:04:28 GMT
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de> <a3p2fg$49g$1@mail1.wg.waii.com> <a3p9b6$1a7qtb$1@ID-98166.news.dfncis.de>
Message-ID: <slrna62oc8.r0v.scarblac@pino.selwerd.nl>

MDK <mdk at mdk.com> wrote in comp.lang.python:
> I made a file mylist.py
> 
> # mylist.py
> class MyList:
>     def __init__(self):
>         self.list=[]
>     def append(self, item):
>         self.list.append(item)
> 
> >>> import mylist
> >>> x = mylist()
> Traceback (most recent call last):
>   File "<interactive input>", line 1, in ?
> TypeError: object of type 'module' is not call
> 
> Why is this happening?  My list object must be a class in its own,
> importable file.

'mylist' is a module. 'mylist.MyList' is the class you defined in the module.

So x = mylist.MyList() is what you meant.

-- 
Remco Gerlich


From phr-n2002a at nightsong.com  Fri Feb  8 03:11:24 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 08 Feb 2002 00:11:24 -0800
Subject: web proxy in python?
Message-ID: <7xvgd85r6b.fsf@ruckus.brouhaha.com>

Anyone got a nice client side web proxy written in Python?  Something
along the lines of junkbuster or the j-marshall.com perl proxy, maybe.

I want to a set a proxy that does a fair amount of site-specific html
personalization.  For example, if I click a someone's link to a
nytimes.com article, I want my browser to go directly to the "print
version" of the article instead of the linked version with all the
ads.  Or on certain sites with obnoxious javascript that won't work
with javascript disabled, I want to strip out the particularly
obnoxious pop-ups.

I could write such a thing from scratch but figure it would be easier
if there's already a proxy that I can customize.

Thanks for any pointers.

Paul


From m.wetherill at salfordNOSPAM.acNOSPAM.uk  Wed Feb 20 04:29:27 2002
From: m.wetherill at salfordNOSPAM.acNOSPAM.uk (Matt)
Date: Wed, 20 Feb 2002 09:29:27 -0000
Subject: COM file references
Message-ID: <a4vq9q$1jsn$1@mimas.salford.ac.uk>

Hi group,
I'd just like to introduce myself - I'm just learning python and really like
the language, but I'm very much a newbie (to both python and programming),
so apologies if my queries seem astonishingly stupid!

Anyway to my problem (or the python related one anyway :-):

I have a requirement to convert ms word docs to raw text in order to allow
further processing of the text file.  I figured the best way to do this was
to use python to access word via COM and do a 'save-as'.  The following code
works fine from the interactive prompt (python 2.2, Office 2k)

import win32com.client
w=win32com.client.Dispatch("Word.Application")
w.Visible=1
w.Documents.Open("c:\\test.doc")
#save file as text file - number '5' is position of word to text converter
in the word 'file type' dialogue
w.ActiveDocument.SaveAs ("c:\\test.txt", 5)
w.ActiveDocument.Close(SaveChanges=0)

However, I really need to turn this into a function to which I can pass an
input file name and path, and an output file name and path.

This is troubling me - however I try, I can't work out how to pass the path
and file to the 'Documents.Open' expression (or vice-versa)

If anyone could offer any guidance I would really appreciate it. - Apologies
for the astonishingly simple nature of the query!

cheers

Matt




From jason at jorendorff.com  Thu Feb 28 22:35:56 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 28 Feb 2002 21:35:56 -0600
Subject: PEP 263 comments
In-Reply-To: <wzk7sz6vsb.fsf@nono.cs.uu.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLMECNDGAA.jason@jorendorff.com>

Piet van Oostrum wrote:
> JO> Advantage: simple, universal, easy, similar to what Java does.
>
> Java does accept iso-latin-1 files as input. In fact on my machine (Mac
> OSX) it doesn't even accept utf-8 files with the utf-8 signature. And
> strings containing utf-8 are interpreted as just 8-bit characters, meaning
> every byte is a character.

Oh!  Yes, it works this way on Windows, too.  javac assumes source
files are latin-1, and System.out.println() encodes output in latin-1.
Odd, because Java assumes UTF-8 everywhere else.  Somehow I missed
this before (I guess because UTF-8 data slips through intact).

Scratch that, then.

> JO> No confusion about embedded 0x22 bytes in strings.  Also,
> JO> stylistically I prefer not to have a document specify its own
> JO> encoding, or for comments to affect the meaning of a source
> JO> file.
>
> Which 0x22 bytes?

I'm referring to this paragraph in Martin's original post:

  The only problem with this approach is that encodings where " or '
  could be the second byte of a multi-byte character cannot be
  supported as a source encoding. Python supports no such encoding
  in the standard library at the moment, anyway, so this should not
  be a problem.

\x22 is a double-quote mark.  Martin is a little off on the last
bit, though; UTF-16 can produce \x22 bytes.

## Jason Orendorff    http://www.jorendorff.com/




From philh at comuno.freeserve.co.uk  Mon Feb  4 13:39:41 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Mon, 4 Feb 2002 18:39:41 +0000
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com>
Message-ID: <slrna5tlbd.393.philh@comuno.freeserve.co.uk>

On 04 Feb 2002 03:18:59 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:
>"Tim Peters" <tim.one at home.com> writes:
>> > That's the problem--if I don't import future division, my code will
>> > break in the future, because the // operator isn't supported in the
>> > present.
>> 
>> Not so:  introducing "//" was 100% backward-compatible, so isn't controlled
>> by __future__.  "//" works out of the box in 2.2, no __future__ required.
>
>If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
>it isn't backwards compatible.

If you are *depending* on getting this exception for your program to
work, then it doesn't deserve to work anyway.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From tim at worthy.demon.co.uk  Sun Feb 10 03:09:41 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Sun, 10 Feb 2002 08:09:41 GMT
Subject: Help! Using Boa Constructor
References: <8f8ffe67.0202071344.6673ed5d@posting.google.com> <3k276u87qkibb6p773sg318chvh0mupusj@4ax.com> <ad4c6u8ei16r9n35fssq7clrul9pt3r7qb@4ax.com>
Message-ID: <101f82064b.tim@worthy.demon.co.uk>

In message <ad4c6u8ei16r9n35fssq7clrul9pt3r7qb at 4ax.com>
          Tim Roberts <timr at probo.com> wrote:

> Wolfgang Strobl <ws at mystrobl.de> wrote:
> >
> >I successfully run it with wxPython 2.3.2.1 + Python21 (works well) and
> >wxPython 2.3.2.1 + Python22 (works, but it throws a GPF in python
> >during shutdown). 
> 
> I also get GPF on shutdown with the same setup.  It doesn't seem to affect
> anything.

Ditto, memory access errors at app shutdown, IIRC.

Can you get the listview object to work, when I try it generates errors
(PC off ATM so can't check details).

-- 
___
 |im    ---- ARM Powered ----


From jkraska at san.rr.com  Wed Feb 27 03:11:24 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 27 Feb 2002 08:11:24 GMT
Subject: External DLL won't load: resolution
References: <mailman.1014745204.15257.python-list@python.org> <7934d084.0202262258.53fbe36@posting.google.com>
Message-ID: <nj5p7ug97vr3qgh2bctqac8ujq5rj8arb8@4ax.com>

So I had fiddled around with the Python and rebuilt it in debug mode,
running python23_d.exe.

It failed to find my extension libarary because I forgot that Python
wants your libraries named using the same pattern, namely:

lramm_d.dll


C//



From jason at jorendorff.com  Tue Feb 19 12:08:05 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 19 Feb 2002 11:08:05 -0600
Subject: Adding static typing to Python
In-Reply-To: <fjs47u0n8bscvgtme1vc7j6jajlt4og2n3@4ax.com>
Message-ID: <HFEKILOLEFEFMKAECNDLAELMDDAA.jason@jorendorff.com>

Courageous wrote:
> Speaking of typedef and the C/C++ languages, I've long thought
> that in a statically-typed language, the types should never be
> aliased, and should always be distinct.

I agree, for your example; but a common use of typedefs now is
to create an alias for a type whose name is just too long to be
allowed.

  typedef list<pair<string, bool> >::iterator ItemIter;

## Jason Orendorff    http://www.jorendorff.com/



From greg at cosc.canterbury.ac.nz  Mon Feb 11 22:33:01 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 12 Feb 2002 16:33:01 +1300
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl>
Message-ID: <3C688CED.4583E7DA@cosc.canterbury.ac.nz>

Martijn Faassen wrote:
> 
> I suppose it might be a brain thing. I use my left hand for writing and
> drawing while I use my right hand for more coarse-grained tasks, like
> throwing a ball and such. I wouldn't count myself as ambidextrous though.

I'm not sure that coarse/fine is the right way to describe
the distinction even in normal unidextrous (is that the
right word?) people. Consider the fact that most right-handed
guitar players use the left hand for fingering, and the
right hand for the relatively dextrously undemanding
function of strumming. Why is that?

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


From jkraska at san.rr.com  Sat Feb  2 23:05:40 2002
From: jkraska at san.rr.com (Courageous)
Date: Sun, 03 Feb 2002 04:05:40 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org> <slrna5p5ba.g91.tim@vegeta.ath.cx>
Message-ID: <p6ep5u83kdr7qt7mkleiojl6g5rm562ae9@4ax.com>

>Imagine a "Lisp-style" indent!
>
>if ( some_condition() ) {
>    i = first_i();
>    for (i = first_i(); i <= 10; i++) {
>        initialize_some_stuff();
>        while ( some_other_condition() ) {
>            do_some_more_stuff() } } }

Yeah, one of the reasons that I really have grown
to dislike Lisp is that you basically can't write
Lisp code using standard-style Lisp formatting
unless you also have a Lisp-aware editor (namely,
Emacs). That truly irks me.

C//



From markabarclay at attbi.com  Tue Feb 26 08:27:35 2002
From: markabarclay at attbi.com (Mark Barclay)
Date: Tue, 26 Feb 2002 13:27:35 GMT
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net> <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com> <3C7AB746.1030800@skippinet.com.au> <3C7ACEB4.B590E892@mediaone.net> <3C7AD777.7050401@skippinet.com.au> <3C7B46B5.3060001@yumpee.org> <3C7B4AF2.3060105@yumpee.org>
Message-ID: <b3Me8.24959$%b6.7314448@typhoon.ne.ipsvc.net>

Thanks for all your help.
The qasid pointer looks like it's going to do the trick for me.

Mark

"Manoj Plakal" <terabaap at yumpee.org> wrote in message
news:3C7B4AF2.3060105 at yumpee.org...
> Manoj Plakal wrote:
> > Mark Hammond wrote:
> >> Mark Barclay wrote:
> >>
> >>> Is there a way of coding a COM server that can be running before it
> >>> is accessed
> >>> by a COM client? I'm not sure if this makes sense, but I'd like to
> >>> have a program
> >>> running, that can be connected to, perhaps multiple times, without it
> >>> having to
> >>> create multiple instances of teh server, just have multiple clients
> >>> using it.
> >>
> >> IIRC, it gets a little messy - you need to implement a LocalServer
> >> object, and add it to the running object table (ROT).  Searching old
> >> posts at groups.google.com is probably your best bet.
> >
> >          Take a look at the CVS tree of qasid.sourceforge.net.
> >          There's a small example of a wxPython GUI app that acts
> >          as a COM server. The COMHostServer function in qasidserver.py
> >          does the job of creating an instance of the server and
> >          registering it in the ROT.
>
>
>    Whoops, I was referring to an old version of the code.
>             You should look at file qasid_win32com.py, class
>             QasidCOMHostThread, method QasidCOMHostServer (cross fingers).
>             The sample GUI app is in testapp.py
>
>             For the curious, Qasid is an idea I had for encapsulating
>             platform-specific mechanisms for local intra-machine "desktop"
>             RPC with a portable API. This is an API can be used by
>             desktop apps to talk to each other for sending simple
messages.
>             Currently, you have to use platform-specific stuff
>             like COM, or use XML-RPC/CORBA and implement a registration
>             and lookup service, or do ad-hoc things like
>             command-line parameters (Netscape/Opera/XMMS etc).
>
>             The idea is to have a simple API wrapping
>             the existing Win32 COM, AppleEvents, KDE DCOP and
>             Gnome CORBA/ORBit. So far, there's the Win32 COM
>             instantiation with a very thin API (start, register
>             method, stop). There's a short blurb on qasid.sf.net
>             about why CORBA and XML-RPC may not be a good idea
>             for intra-machine (between desktop apps) communication,
>             though they work great on the intra/internet. Haven't
>             had too much time to work on this since the initial
>             COM effort, but I do plan to get started on the
>             Apple Events backend sometime soon.
>
>             Manoj
>




From brian at sweetapp.com  Tue Feb 12 17:18:50 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 12 Feb 2002 14:18:50 -0800
Subject: photo id ?!?!
In-Reply-To: <04a501c1b40c$7d17bf00$0201a8c0@josiah.dalkescientific.com>
Message-ID: <000a01c1b413$3f7205a0$445d4540@Dell2>

[snipped]

I can't believe that you even made it to the showing photo ID stage;
they wouldn't even consider letting me in wearing only a balaclava! And
a PythonLabs member recognized me too (I won't name names - but you know
you are <wink-il-y-ours>).

I am also willing to help organize next year's convention to prevent
future fascist abuses from occurring. Here are some of my suggestions:
- Cedar Trails Nudist Retreat (http://www.bright.net/~ctnr/)
- Twin Falls Nature Retreat (http://www.twin-falls.com/)

Cheers,
Brian




From theguru at mb.sympatico.ca  Fri Feb 22 16:11:56 2002
From: theguru at mb.sympatico.ca (Dieter Lunn)
Date: Fri, 22 Feb 2002 15:11:56 -0600
Subject: Python and Daemons
Message-ID: <02022215115601.01886@localhost.localdomain>

Hi,

I was wondering how to write a linux daemon using Python.



From gcordova at hebmex.com  Tue Feb 12 12:10:48 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 12 Feb 2002 11:10:48 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A40@SRM201002>

> >"-ist" - an individual professionally or avocationally 
> occupied with the
> >idea or activity defined by the root [1].
> 
> >Of course, in either Spanish or Esperanto it would more commonly be
> >"Pythonisto" anyway, so who knows.

Kinda wrong for Spanish though.

A follower of Zapata ("Emiliano Zapata", perhaps?) would be "Zapatista",
be it a male or female. "Zapatisto" is a gross error, and it's funny
also :-)

It's common for little children to mistakenly use "-isto" as well as
"-ista" when talking about boys or girls, that's why it's funny, because
it's an infantile mistake.



From kalath at lycos.com  Thu Feb  7 10:24:20 2002
From: kalath at lycos.com (Mark Brady)
Date: 7 Feb 2002 07:24:20 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <e840346c.0202070321.452d405d@posting.google.com> <7xr8nxy0rb.fsf@ruckus.brouhaha.com>
Message-ID: <e840346c.0202070724.13440c8d@posting.google.com>

Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7xr8nxy0rb.fsf at ruckus.brouhaha.com>...
> I don't know what would happen if someone tried a big project in
> Python (Zope is medium sized).  Maybe it would work ok, maybe not.
> 
> I do believe a native-code compiler would make Python a lot more
> interesting for many purposes.

I think this would be very useful also but how would it work with
Python's dynamic features? I know some native code comipled languages
such as Objective C and GJC (the GNU java native code compiler) come
with a runtime library for the native exe, is this possible for a
language like python?

Mark.


From missive at frontiernet.net  Sun Feb 24 17:50:31 2002
From: missive at frontiernet.net (Lee Harr)
Date: Sun, 24 Feb 2002 22:50:31 -0000
Subject: GUI (PyQt) & thread problem
References: <8b5d764c.0202240451.1c4bda3b@posting.google.com>
Message-ID: <slrna7ird3.a0.missive@whave.frontiernet.net>

> Can someone tell me what the problem is with the following program, as it 
> hangs. I'm pretty sure there's some inherent design-flaw involved, but
> I'm not familiar enough with threads or GUI-programming to know what it 
> could be.
> 
> I've also tried modyfing the program to use PYSIGNALs instead of passing the
> caller to the thread, but the results were the same.
> 

I did a google search for pyqt threading and I am not so sure that
these things are working together.

There is a qt QThread class, but it does not seem to be available
to pyqt yet.



From alanmk at hotmail.com  Tue Feb  5 05:59:34 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 5 Feb 2002 02:59:34 -0800
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <2hy78.23175$l93.4207585@newsb.telia.net>
Message-ID: <f72dac0d.0202050259.6bc6fe6f@posting.google.com>

"Fredrik Lundh" <fredrik at pythonware.com> wrote:
> Alan Kennedy wrote:
> 
> > I don't understand what's wrong with
> >
> > <script language="python" xml:space="preserve">
> >     def messiah():
> >         print "He's NOT the Messiah."
> >         print "He's a very naughty boy."
> > </script>
> 
> something tells me that you're a pragmatic programmer
> working in real-life projects, right? ;-)

I'll have you know I have a very responsible job at the Ministry ....

of Silly Walks!

http://www.mwscomp.com/mpfc/silwalk.html

Alan.

    |__
  o/   
 / )


From danscott at chocolatefudge.com  Thu Feb 21 16:08:48 2002
From: danscott at chocolatefudge.com (DanS)
Date: Thu, 21 Feb 2002 13:08:48 -0800
Subject: Record Audio (Automation)
References: <140415ad.0202211224.40d8b34b@posting.google.com>
Message-ID: <a53nl0$6ld@news.or.intel.com>

The best thing here would be to write your own Visual Basic app and use the
MS Multi Media control...

Write your program to take command line input.. and well, thats about it..

Dan





"T H" <tth_1 at yahoo.com> wrote in message
news:140415ad.0202211224.40d8b34b at posting.google.com...
> I need a simple way to record audio with a scripting language in
> Windows.  I
> think this can be done by executing sndrec.exe or sndrec32.exe (the
> sound recorder that comes with windows that can be executed by going
> to Programs->Accessories->Entertainment->Sound Recorder), however I
> can't seem to be able to make it record.  I know you can use this
> program to PLAY a WAV file by executing
> sndrec32.exe /play /close filename.wav
>
> However I was not able to make it record by executing
> sndrec32.exe /record /close filename.wav
>
> The command above I tried is just a guess, i have no idea if there is
> a /record parameter.  Any idea where I can find a list of command line
> parameters for sndrec32.exe?
>
> If anyone knows any other way of automating the task of recording
> audio using a scripting language (which executes a free audio
> recording program), please let me know.  Any help is much appreciated.




From skip at pobox.com  Tue Feb 12 22:48:05 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 21:48:05 -0600
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <7xzo2ekv5g.fsf@ruckus.brouhaha.com>
References: <mailman.1013557302.9749.python-list@python.org>
        <7xzo2ekv5g.fsf@ruckus.brouhaha.com>
Message-ID: <15465.57845.533984.489234@12-248-41-177.client.attbi.com>

    >> ... Upon my return to the shell prompt, I discover that the
    >> /usr/local/bin/python link is gone as are a number of shared
    >> libraries (readline among them).

    Paul> That should be impossible assuming you don't have write access to
    Paul> the /usr/local/bin directory.  If you can reproduce it, it's a big
    Paul> Linux security hole.

I realized that might be an issue on my way home.  This is on my development
machine.  I (user == skip) do have write access to /usr/local/*.  Sorry
about the missing info.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From david.abrahams at rcn.com  Tue Feb 26 13:01:15 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Tue, 26 Feb 2002 13:01:15 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <9EF5B844-2ADF-11D6-A367-0030657EBC3E@lanl.gov>
Message-ID: <0ebb01c1beef$a09efe00$0500a8c0@boostconsulting.com>

----- Original Message -----
From: <rasmussn at lanl.gov>

> On Monday, February 25, 2002, at 05:32 PM, David Abrahams wrote:
>
> > I've been interested in systems for automatically wrapping C++ and
> > generating documentation for a while, so I downloaded the PDT (the
parser
> > toolkit used by SILOON) and tried it on one of my source files. After
lots
> > of hacking, I was able to get the EDG front-end component to parse my
> > source
> > and generate an intermediate language (.cil) file, but the PDT
executable
> > which is supposed to interpret the .cil and produce a high-level
> > description
> > of the code crashed with an assertion:
>
> SILOON uses scripts to run the PDT and we've worked hard to make it easy
> to use.  Essentially all you have to do is edit a file (that is included
> in a Makefile) and point macro variables to the location of your source
> files (.h .cpp ...).  I would be interested to know if you have similar
> problems using the EDG front-end from SILOON directly.

Well, I'm less interested in SILOON itself than in the PDT, but I might try
it if I get some more time to waste ;-)

> > Assertion failed: ptr->assoc_template, file c:\documents and
> > settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-
> > tau\il_display_tau.
> > c, line 3047
> >
> > Since the PDT doesn't come with source, there was no way for me to
debug/
> > fix
> > the problem.
>
> The PDT comes with sources for everything (almost) but the EDG front end.

There's no "il_display_tau.c" in either the .zip or the .tar.gz.

> What PDT version were you using?

The very latest I could find at
http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/.

> Would you like to send your source files to me directly
> and I can
> look at the problem, although if it is windows-specific code, I can't help
> you
> as I don't have access to Microsoft compilers.

It's code that compiles on 7 different compilers including several different
EDG-based compilers on four OSes, so it shouldn't be a problem. If you get
the current boost cvs state (see http://sourceforge.net/cvs/?group_id=7586)
and a copy of Python-2.2, you can try for yourself. The file is
<boost-root>/boost/python/class.hpp.

> (On further examination, it looks like the PDT doesn't have source for
> taucpdisp,
> I'll ask the developers why not.)
>
> I can provide access to a prerelease version of SILOON which uses the
> latest
> version of the PDT, which may fix your problems.

If you can get this to work, I might be interested in seeing the results.

> > Furthermore, the PDT developers don't seem to want to be reached. Their
> > contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/)
> > says:
> >
> >     To contact the developers of PDT, please use the "Comments" section
in
> > the download page.
> >
> > But of course, there is no "Comments" section on the download page. :=(
>
> I've forwarded you comments on to the PDT developers so that they can fix
> their web pages and comment personally on the core dump.

Thanks,
Dave





From gcordova at hebmex.com  Wed Feb  6 14:25:31 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 6 Feb 2002 13:25:31 -0600 
Subject: What is the command to determine an object's type?
Message-ID: <19650FF08A99D511B19E00B0D0F0632301898B8F@SRM201002>

NO NO NO NO NO NO!!!

**WHY*** do you want to break beautiful OO-code with this?

> $ python
> >>> import types
> >>> l = []
> >>> if type(l) == types.ListType:
> ...     print "yeah baby!"
> ...
> yeah baby!

Or with this??
> or better:
>    if type(l) is types.ListType:
>        ...
> or even better:
>    if isinstance(l, types.ListType):
>        ...
> or, in 2.2 or later:
>    if isinstance(l, list):
>        ...

What if I pass an instance of an object which
behaves like an array? This code will surely
break, uglily [sp].

So, yes, you *can* identify when an object
is descendant of types.ListType, or or list
in case of Python 2.2, but that'll limit
your code's functionality.

Alas, I don't have an answer to your question,
besides checking if your received object has
a __getitem__ and __len__ method.

Sorry.

-gus



From quinn at hork.ugcs.caltech.edu  Sat Feb  9 21:14:28 2002
From: quinn at hork.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 02:14:28 GMT
Subject: A newbie question about class
References: <a50631f6.0202091635.1dc42d4b@posting.google.com>
Message-ID: <slrna6bls3.9tv.quinn@hork.ugcs.caltech.edu>

On 9 Feb 2002 16:35:19 -0800, newgene <newgene at bigfoot.com> wrote:
>can not get the value of "a.y" like the former. Of course I can use a
>method to set the value of "a.y", but is there a machanism to get the
>"a.y" automatically whenever I change the value of "a.x"?

Yes.  Under python2.2, it's (relatively) easy:

>>> class A:
...     def __init__(self, x=42):
...         self.x = x
...     def get_y(self):
...         return self.x**2
...     y = property(get_y, doc="'y' is computed as 'x**2'")
... 
>>> a = A()
>>> a.y
1764
>>> a.x = 2
>>> a.y
4
>>> 


Under older pythons, it's a bit more tricky.  I'd recommend defining a
'get_y()' as above, and calling that instead of doing 'a.y'.  If you really
want 'a.y', then check the refman under 'customizing objects' or something for
__getattr__.


From chris_mk at hotmail.com  Tue Feb 12 19:23:12 2002
From: chris_mk at hotmail.com (Christopher)
Date: 12 Feb 2002 16:23:12 -0800
Subject: Hidden files, selecting directories, & taxes
Message-ID: <cc14131d.0202121623.47ffaa99@posting.google.com>

Hello everybody,
    I had a couple things to ask:

1)  Is it possible to generate a hidden file (under Windows)?  I am
creating a program that saves information into a dbhash file and I
would prefer to have the dbhash file created as a hidden file.  I
'think' I have looked through the documentation but, as I have said
before, I am not a computer person.  I am a biology person who
recently discovered that he likes computers.  I have tried looking at
a file's 'stat' values, then making the file hidden and looking at the
'stat' values again.  I tried looking at some of the flags but I seem
to be stumped.

2)  Is there a simple and easy way to allow the end-user to select a
directory?  For example, let's say all files of rpt type are saved to
a certain directory (named by each user).  And we want to perform an
operation on each of the files, is there an easy way to get the
directory (when I say easy, I mean something like
tkFileDialog.askopenfilename or the like).  Right now I skirt around
the issue by asking the user to select one of the files they wish to
process and then I us os calls to get the directory path then use
listdir to get a list that I loop over.  It's a bit of a cheesy method
but I spent a significant amount of time making it feel 'natural.'

3)  (For US only) Is it just me, or is it rough that we have to pay
back the tax 'rebate' that got sent to us last year?  I plan my taxes
very well, but my wife and I owe $590 (instead of getting a refund of,
whoo-hoo, $10).  Just curious.

Thanks a bunch.

Chris


From vtail at yandex.ru  Tue Feb 12 11:02:25 2002
From: vtail at yandex.ru (vtail)
Date: 12 Feb 2002 08:02:25 -0800
Subject: Tkinter and grid: strange alignment with and without sticky option
Message-ID: <9ff6b66.0202120802.3f20bcaf@posting.google.com>

Hi, guys,

I have very odd problem with using grid() with and without sticky
option. Here is my python version:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32

I'm using Tkinter that ships with standard distribution.

Consider the following snipet:
----
from Tkinter import *

def main(is_sticky = 0):
    root = Tk()

    if is_sticky:
        Listbox(root, width = 20, relief = RIDGE, font = "Courier",
            bg = "white").grid(row = 0, sticky = W)
        Entry(root, width = 20, relief = RIDGE, font = "Courier") \
                    .grid(row = 1, sticky = W)
    else:
        Listbox(root, width = 20, relief = RIDGE, font = "Courier", 
            bg = "white").grid(row = 0)
        Entry(root, width = 20, relief = RIDGE, font = "Courier") \
                    .grid(row = 1)
    root.mainloop()

if __name__=="__main__":
    main(1)
---

The code in if..else.. block only differs in whether or not we are
using a "sticky = W" option

When running main(0), all is OK. But when running main(1), I see
Listbox moved one pixel to the right.

It all seems very strange for me. Does anybody has any ideas? I really
need your help.

Thanks a lot,

Vtail.


From sholden at holdenweb.com  Mon Feb 11 12:19:59 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 12:19:59 -0500
Subject: Question on if, while, for....  Figured it out..
References: <e959e7b.0202110901.485da0b0@posting.google.com>
Message-ID: <U2T98.152185$KM2.6837837@atlpnn01.usenetserver.com>

"Fred" <c at tech.usd253.org> wrote in message
news:e959e7b.0202110901.485da0b0 at posting.google.com...
> OK, I figured it out (one other way at least).

A few more observations interspersed with your (now much cleaner!) code...
>
> -------------------
>
> num = input("Enter the numerator: ")
> den = input("Enter the denominator: ")
> test = 0
>
> if num > den: #Test for improper fraction.
>     i = num/den #This will only return the whole part.
>     j = num%den #This will return only the remainder.
>
Rather than comment the usage of "i" and "j", why not just use "d" for
dividend and "r" for remainder? The idea behind being able to choose
variable names is to allow them to have some mnemonic significance.

>     if j != 0:

Here you know there's a remainder, and you know what the dividend is. If you
subtract the remainder from your original number then divide by the
dividend, this will be the whole part. So you can replace the loop below
(which only works for small numbers) by a simple computation. I'm leaving
this to you, since you like to work out your own solutions...

>         for i in range(9,0,-1):
>             if num%i == 0 and den%i ==0 and test != 1:
>                 a = j/i
>                 b = den/i
>                 print "You entered an improper fraction, simplified it
> is: %d %d/%d" % (i,a,b),
>                 test = 1

By the way, if you *do* want to use a loop, rather than maintaining a
variable that tells you when your loop has doen its work, you can just use
the "break" statement to terminate it. This would have saved you testing the
"test" variable above, which in turn would have saved you from having to
initialise it even further above.

>     else:
>         print "You entered an improper fraction, simplified it is: %d"
> % (i)
>
>
>
> else:
>     for i in range(9,0,-1):
>         if num%i == 0 and den%i ==0 and test != 1:
>             a = num/i
>             b = den/i
>             print "The fraction you entered is %d/%d in lowest terms."
> % (a,b),
>             test = 1
Since here you know that the remainder is zero, this would be the
appropriate place to work out the lowest common denominator.

lots-more-meat-in-this-problem-yet-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jimd at vega.starshine.org  Thu Feb 28 03:16:14 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 28 Feb 2002 08:16:14 GMT
Subject: calling a function indirectly
References: <a4v7s6$13tk@r02n01.cac.psu.edu> <slrna769j8.1hv.grante@tuxtop.visi.com> <XiHc8.640$XT1.47512@news1.east.cox.net> <a50udo$1dou@r02n01.cac.psu.edu> <Vg%c8.4179$XT1.116344@news1.east.cox.net> <a5cnvo$2mdu$1@news.idiom.com> <UTAe8.9921$jC3.362611@news1.east.cox.net>
Message-ID: <a5kp0e$2kj4$1@news.idiom.com>

In article <UTAe8.9921$jC3.362611 at news1.east.cox.net>, Jeff Hinrichs wrote:

>"Jim Dennis" <jimd at vega.starshine.org> wrote in message
> news:a5cnvo$2mdu$1 at news.idiom.com...
>> In article <Vg%c8.4179$XT1.116344 at news1.east.cox.net>, Jeff Hinrichs wrote:

>>> see:
>>> http://mail.python.org/pipermail/python-list/2001-August/060976.html
>>> for some info.   Unless you have total control over what is being eval'd
>>> you are at risk.
>>>-Jeff

>>> "Rajarshi Guha" <rxg218 at psu.edu> wrote in message
>>> news:a50udo$1dou at r02n01.cac.psu.edu...
>>>> On Wednesday 20 February 2002 01:24 in comp.lang.python Jeff Hinrichs
>>> wrote:

>>>>> If you wanted to get away from the dangerous eval, you could put your
>>>>> functions inside of a class and then,

>>>> Why is eval dangerous?

>>  In THIS case it's not a problem.  The string being eval'd is a
>>  reference to be bit of code he wrote himself.  However, if it was
>>  being combined with a string from any untrusted source (esp.
>>  user input, or data from a network connection or subprocess, even
>>  filenames from an os.listdir() or some such, IF that was the case
>>  then eval could be quite dangerous because it could be executing
>>  arbitrary Python code (which could, in turn, execute arbitrary bits
>>  of system code).

>>  Regulars on this newsgroup display a knee jerk reaction to eval().
>>  This could be construed as a healthy wariness in some cases.  One
>>  should always ask:
>> 		where did this data come from?
>> 		how did I validate it?
>>  and especially,
>> 		what are the risks of using/trusting this data in this way?

>>  ... but those questions should not be limited merely to strings
>>  that we might be passing to eval, they must be constantly applied
>>  throughout our coding if we intend to write robust code which
>>  works across security contexts (handles any sort of foreign data)
>>  and which is capable of enforcing the most basic implicit security
>>  policy (don't get subverted to executing arbitrary code or tricked
>>  into removing, or corrupting our data).

> This is true, but you should always consider that 
> modifications/enhancements in the future might expose those 
> "shielded" yet vulnerable sections of code.  With that in mind, 
> you should always code "on the defensive."

 "those 'shielded' yet vulnerable"... which are those?  I didn't
 refer to any thing as shielded or vulnerable.  I said that we
 should not curse eval() (merely the mechanism) but that the 
 questions we should be asking are:

 	where did the data come from?
	how did we validate it?
	what are the risks of using this data in this way?

 Remember that our Python code is already relying on a long
 trusted path (from our hardware and OS vendors and distributors,
 through all of the authors and contributors of the compilers, 
 linkers, archivers, and all the way to the interpreters (and 
 shells, dynamic library loaders, kernels, etc) and modules that
 we use.  
 
 That many of us use Python (open source) compiled with
 gcc (open source) on Linux or {Free,Net,Open}BSD systems (more
 open source) and processing (linked, archived, packaged, distributed,
 installed, and configured) with more open source tools, this fact
 is reassuring to some.  That there are wacky people out there who
 disassemble Python byte-compiled code, and binaries (into assembly)
 and people who run their systems with ICEs (in-circuit emulators)
 while debugging *their* software (which depends on our tool chains)
 --- that these people exist and a large number of them have ready
 access to the Internet (and places like slashdot especially)
 is also reassuring.

 People who've read Ken Thompson's essay on "Trusting Trust" (???)
 will see what I've getting at.  While the "Trusting Trust" 
 scenario might be remotely possible even with Linux and GCC, the
 chances are truly infinitesimal.


> Danger lurks not only where we expect it but where we don't.  So having a
  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> "knee jerk" reaction to eval is proper.  It has a number of uses and can be
> quite handy in one shot scripts but any code with a lifetime of more than
> one run...."Danger! Will Robinson.  Danger!"

>-Jeff

 Ummm.  This is self-contradictory.  Since danger lurks where we
 don't expect it then "knee jerk" attention or focus on eval expressions
 (where you are EXPECTING danger) is necessarily detracting from 
 the rest of your coding and data manipulation assumptions (where
 we aren't *expecting* to have problems.

 It would be folly to read Python code, looking for copies of eval(),
 scrutinizing those, and then waltz away with the misconception that
 you'd done a code audit.

 Attention is ignorance.  Ignorance is Attention.

 To attend to anything is to ignore others and vice versa.
 I hate to make Zen references when talking about coding and 
 especially when discussing computer security, but the fact is 
 that we must be holistic (looking at *everything*) which does
 require abandonning prejudices and questioning our own assumptions.

 You are right about one thing, the bit of code which was "safe" and
 "secure" at one point (given a set of assumptions which were valid,
 even *provable* at the point that it was implemented) have frequently
 been rendered dangerous and insecure by later enhancements to their
 environment.  
 
 UNIX is rife with examples of this.  Some will say it's evidence
 of a poor security design.  I will assert that it is a natural 
 byproduct; UNIX is one of the oldest and definitely is the most
 widely ported, wildly "enhanced" and modified, operating system 
 in history.  In fact UNIX is really a set of conventions regarding
 programming APIs, tool chain and utility sets, and file system
 naming and semantics.  UNIX is not "an operating system" but a term
 that describes many vastly divergent operating systems.

 So, in the thirty year history of UNIX (over half of the history
 of electronic computing) we've seen enhancements like symlinks, 
 DNS (over /etc/hosts), networking (especially NFS, and other networking 
 fileysystems), shared/dynamic libraries, longer filenames, dropped
 constraints on filename character sets (and other i18n features), 
 introduction of GUIs and countless other enhancements which have
 each resulted in new bugs --- failures in various programs
 to implement the intentions of their programmers, and many of those
 were security bugs --- failures to enforce the intended, usually
 implicit policies of their creators and/or users/owners.

 (Other OS' are also peppered with examples of how new features
 break the previously "safe" assumptions design and programming
 decisions).
 
 Telling people to avoid eval() because it's "dangerous" and might
 lead to "insecure" code is a gross oversimplification.  It is better
 to teach them about good coding practice and design theory then to
 tout empty guidelines.  (Granted, eval() might usually be the
 *wrong* approach for other reasons, such as unmaintainability, 
 lack of introspection, spurious or opaqueness to the structured
 exception handling mechanisms, etc; those are NOT hollow reasons.
 But "because it's dangerous" is much to vague and oversimplified).
 


From klaatu at evertek.net  Sun Feb 24 15:34:21 2002
From: klaatu at evertek.net (Mike Dean)
Date: Sun, 24 Feb 2002 14:34:21 -0600
Subject: Correction: Re: string substitutions
In-Reply-To: <c76ff6fc.0202232228.201f829e@posting.google.com>; from sjmachin@lexicon.net on Sat, Feb 23, 2002 at 10:28:50PM -0800
References: <mailman.1014504639.21135.python-list@python.org> <c76ff6fc.0202232228.201f829e@posting.google.com>
Message-ID: <20020224143421.A10674@evertek.net>

> Try testing instead of proofreading. Post actual examples of input and

OK - I'll remember that in the future (it'll definately save me from
embarassing blunders like this...)

And I had forgotten about the raw strings bit - one question about them
though - if I'm using raw strings in my RE, is there a way to use a
newline in an RE, or do I need to resort to regular strings for such a
situation?

Thanks (and immensly grateful for a community that doesn't chew me out
for making a mistake :-),
Mike



From tim at worthy.demon.co.uk  Mon Feb 11 08:20:15 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Mon, 11 Feb 2002 13:20:15 GMT
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <mailman.1013426479.20868.python-list@python.org>
Message-ID: <f36322074b.tim@worthy.demon.co.uk>

In message <mailman.1013426479.20868.python-list at python.org>
          Gerhard H?ring <gh_pythonlist at gmx.de> wrote:

> Le 11/02/02 ? 10:40, Tim Howarth ?crivit:
> > I need to convert some long int values to normal ints.
> > (4 byte words) [...]
> 
> >  |im    ---- ARM Powered ----
> 
> Assuming 32-bit hardware, like my Intel x86 and StrongARM processors :-)

Testing/writing done on StrongARM RiscPC/RISCOS 4, but final to be able
to run on anything. (I'm playing with zipfile module to give RISC OS
filetype support.)

>
>     print struct.unpack("i", struct.pack("I", 0x81234567L))[0]

All I want is a=b ! but lose longness of b !


-- 
___
 |im    ---- ARM Powered ----


From gerson.kurz at t-online.de  Fri Feb 15 13:46:03 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Fri, 15 Feb 2002 18:46:03 GMT
Subject: help with silly algorhythm
References: <5e8bd451.0202142155.143ca4d9@posting.google.com> <3c6ce21d.190762859@news.isar.de> <3C6D6067.F4AC05A4@ccvcorp.com>
Message-ID: <3c6d52bd.19180421@news.t-online.de>

On Fri, 15 Feb 2002 11:24:24 -0800, Jeff Shannon <jeff at ccvcorp.com>
wrote:

>> An ideal job for lambda! Lets start with the data.
>>
>> > * list of input values =[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
>> > * a list of acceptable new 'partner values' = [1, 3, 7, 10]
>>
>> input_list = range(12)
>> partner_values = [1,3,7,10]
>>
>> > so we might get (in mode 1: all closest higher value)
>> > [ [0, 1], [1, 3], [2, 3], [3, 7], [4, 7], [5, 7], [6, 7], [7, 11]....
>>
>> GetUpper = lambda i,p:filter(lambda x:x[1],[(y,(filter(lambda
>> x:x>y,p)+[None])[0]) for y in i])
>>
>> In typical Python fashion the code is obvious.
>
>Heh.  I positively cannot make heads or tails of that lambda--perhaps its a
>shortcoming of mine, but extended lambdas like that make my eyes glaze and my
>brain shut off....

OK, my post was meant as a joke. Besides, it became clear to my that I
assumed the lists were ordered in advance. Here is, in explicit
notation, what the above lambda expression does, and how. 

def GetUpper(i,p):
    result = []

    # this is the [for y in i] array
    for y in i:

        # this is the innermost lambda
        def IsGreaterThanY(x):
            return x>y

        # will return a list, where each
        # element is greater than the current y        
        temp = filter( IsGreaterThanY, p )

        # make sure the list has at least one element
        # (namely "None")
        temp = temp + [None]

        # return the lowest of that list = the
        # lowest element that is higher than y
        temp = temp[0]

        # this is done by the filter x[1]: it will
        # remove all elements, where x[1] is None
        if temp:
            result.append( (y, temp) )        

    return result        

GetLower() looks similar, but for two changes:

- IsGreaterThanY is replaced by IsLowerThanY, with obvious
implementation
- the last element of the temp list is choosen (=the highest element
that is lower than y)

GetLowerWrap() looks like normal GetLower(), only instead of adding
"None" elements and filtering them, the highest value in the list is
choosen (and not filtered).

Mix() uses zip() to create a list of result tuples, and then a simple
xrange to choose odd/even members of that list.




From Bill.Scherer at VerizonWireless.com  Wed Feb 27 11:23:18 2002
From: Bill.Scherer at VerizonWireless.com (Scherer, Bill)
Date: Wed, 27 Feb 2002 11:23:18 -0500 (EST)
Subject: Rotor in Jython?
In-Reply-To: <Pine.LNX.4.33.0202270847100.4697-100000@bagend>
Message-ID: <Pine.LNX.4.33.0202271120480.4697-100000@bagend>

My rotor code is available at:

http://jobs.verizonwireless.com/python/rotor.py

I've implemented the encryption side (which I don't currently 
need), and it can now decrypt what it has encrypted. I can not 
yet decrypt texted encrypted by the stdlib rotor module.


On Wed, 27 Feb 2002, Scherer, Bill wrote:

> 
> Hello - 
> 
> I've a need to perform rotor decryption in Jython.  A quick 
> search didn't reveal anything to help.  I've begun coding it in 
> Python from rotormodule.c. My C knowledge is not strong, so there 
> are spots that are surely not right.
> 
> What I have now creates a 'rotor' object and performs a 
> decryption without error, but the results are incorrect.
> 
> I can make the code available.
> 
> Any help is appreciated.
> 
> Thanks!
> 
> 

-- 
Bill.Scherer at Verizon Wireless
RHCE 807101044903581




From philh at comuno.freeserve.co.uk  Fri Feb 22 17:53:57 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 22 Feb 2002 22:53:57 +0000
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie> <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk> <Uayd8.5117$D6.14249@news.iol.ie>
Message-ID: <slrna7dj05.1bt.philh@comuno.freeserve.co.uk>

On Fri, 22 Feb 2002 20:51:00 GMT, Roy Madden <roy at linksheaven.com> wrote:
>On Fri, 22 Feb 2002 14:49:38 +0000, philh at comuno.freeserve.co.uk (phil
>hunt) wrote:
>
>>What do you consider the other 2 methods to be? I can think of 
>>encoding it in the URL, and the server remembering the browser's IP 
>>address. Are there others?
>
>Yeah, the generally accepted methods by which you pass session ids are:
>- cookies
>- url like you say
>- hidden form fields

If the form uses http GET, then these last 2 are the same; http POST 
is different.


-- 
### Philip Hunt === philh at comuno.freeserve.co.uk ###





From bokr at oz.net  Mon Feb 25 00:50:56 2002
From: bokr at oz.net (Bengt Richter)
Date: 25 Feb 2002 05:50:56 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu> <eppstein-69765A.19343424022002@news.service.uci.edu> <7xd6yu42j6.fsf@ruckus.brouhaha.com>
Message-ID: <a5cjc0$cb5$0@216.39.172.122>

On 24 Feb 2002 20:18:53 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:

>David Eppstein <eppstein at ics.uci.edu> writes:
>> > : But what is the functional equvalent of:
>> > 
>> > : for x in items:
>> > :     x.f()
>> > 
>> > 
>> > 
>> > Here's one way to do it:
>> > 
>> > ###
>> > map(lambda x: x.f(), items)
>> > ###
>> 
>> I'd prefer
>> [x.f() for x in items]
>> 
>> It's not functional syntax, but so what?
>
>Both of those build up a new list of the results, instead of
>discarding the values.  If the f function takes an integer and
>computes a 20-megabyte structure, you've got a problem.  This really
>calls for a generator comprehension as discussed a few weeks back.
>
>   reduce(lambda x,y:(y.f(),0)[1], items, 0)
>
>almost works, but invokes f an extra time at the end.
>
>Someone see how to fix it?

 >>> class F:
 ...     def __init__(self,msg):
 ...         self.msg = msg
 ...     def f(self):
 ...         print self.msg
 ...
 >>> a=F('this is a')
 >>> b=F('this is b')
 >>> c=F('this is c')
 >>> items = [a,b,c]
 >>> reduce(lambda x,y:y.f() and 0, items, 0)
 this is a
 this is b
 this is c

but I didn't get an extra call trying your version:

 >>> reduce(lambda x,y:(y.f(),0)[1], items, 0)
 this is a
 this is b
 this is c
 0

you can use the 'and 0' ploy to get a zero length
list comprehension also:

 >>> [ 0 for x in items if x.f() and 0]
 this is a
 this is b
 this is c
 []

Regards,
Bengt Richter



From hst at empolis.co.uk  Thu Feb 21 12:24:55 2002
From: hst at empolis.co.uk (Harvey Thomas)
Date: Thu, 21 Feb 2002 17:24:55 -0000
Subject: What values are considered false?
Message-ID: <8FC4E7C302A6A64AAD5DB1FA0E825DEB220AD5@hendrix.empolisuk.com>

The empty function is not only false, it is also None

In the interpreter,

def foo():pass

foo() is None

1

> -----Original Message-----
> From: Jeff Henry [mailto:jhenry19 at ford.com]
> Sent: 21 February 2002 16:27
> To: python-list at python.org
> Subject: Re: What values are considered false?
> 
> 
> Roy Smith wrote:
> > 
> > Roy Smith <roy at panix.com> wrote:
> > > PS -- In the lists of false things in python that have 
> been mentioned on
> > > this thread, I don't think I've seen anybody mention {}, the empty
> > > dictionary.
> > 
> > Along those same lines, I half-expected the empty function 
> to be false,
> > i.e.:
> > 
> > def foo():
> >    pass
> > 
> > if (foo):
> 
> Oops! I think you meant to say -
> 
> if (foo()):
> 
> >    print "it's true"
> > else:
> >    print "it's false"
> > 
> > should have printed "it's false".  Alas, it says it's true. 
>  Seems kind of
> > irregular to me, not that I would imagine there's much practical
> > consequence one way or the other.
> -- 
> http://mail.python.org/mailman/listinfo/python-list
> 

_____________________________________________________________________
This message has been checked for all known viruses by the MessageLabs Virus Scanning Service.



From mhuening at zedat.fu-berlin.de  Tue Feb 12 09:15:33 2002
From: mhuening at zedat.fu-berlin.de (Matthias Huening)
Date: 12 Feb 2002 14:15:33 GMT
Subject: Newbie question: Why does read() method of urllib hang?
References: <%CT98.13104$HC4.90692@NewsReader> <41Y98.143353$XZ1.5765053@typhoon.austin.rr.com> <Ry3a8.40197$bP3.337803@NewsReader> <8f070afd.0202120550.45d8873b@posting.google.com>
Message-ID: <Xns91B39B73814D6mhueningzedatfuberli@130.133.1.4>

dwfreeze at yahoo.com (dwfreeze) wrote in
news:8f070afd.0202120550.45d8873b at posting.google.com: 

> 
> Standard "Mee too" post follows:
> 
>>>> sys.version
> '2.1.2 (#31, Jan 15 2002, 17:28:11) [MSC 32 bit (Intel)]'
>>>> import urllib
>>>> f = urllib.urlopen('http://slashdot.org').read()
> 
> After I execute the urlopen the IDLE shell is definately "hung".

Hhmm, I can't reproduce this. Everything works just fine.

>>> import sys
>>> import urllib
>>> sys.version
'2.1.1 (#20, Jul 26 2001, 11:38:51) [MSC 32 bit (Intel)]'
>>> f = urllib.urlopen('http://slashdot.org').read()
>>> print f
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<HTML><HEAD><TITLE>Slashdot: News for nerds, stuff that matters</TITLE>
...
etc.etc.



From jon+usenet at unequivocal.co.uk  Sun Feb  3 16:51:30 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Sun, 3 Feb 2002 21:51:30 +0000 (UTC)
Subject: Python 2.2 Modules/Setup lies?
References: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net> <mailman.1012768207.25312.python-list@python.org>
Message-ID: <slrna5rc6q.dtp.jon+usenet@snowy.squish.net>

In article <mailman.1012768207.25312.python-list at python.org>,
  Oleg Broytmann wrote:
>Because Python 2.1+ uses Disutils, not Setup for configuring itself. This
>allows better testing current OS/libraries configuration and selection of
>modules.

Sorry, I don't understand your answer. Do you mean that the README
file is completely wrong? This would appear to be a bit of an
oversight if so. Python 2.2 most certainly is using the Setup file to
configure itself, because it changes things if I alter it. On the
other hand, it does appear to be including modules that are excluded
in Setup. Is there any *correct* documentation for all of this
somewhere? How do I configure things like where socket.c looks for SSL
libraries?


From jegan_fernando at mumbai.tcs.co.in  Wed Feb 13 13:47:02 2002
From: jegan_fernando at mumbai.tcs.co.in (jegan_fernando at mumbai.tcs.co.in)
Date: 13 Feb 2002 18:47:02 GMT
Subject: Embedding Python in C
Message-ID: <a4ecb6$b4a$1@bcarh8ab.ca.nortel.com>

class pythontest:
   def test(self,ckt, tkc, name='',site=None):


From m.bless at gmx.de  Thu Feb  7 07:44:25 2002
From: m.bless at gmx.de (Martin Bless)
Date: Thu, 07 Feb 2002 12:44:25 GMT
Subject: import subtleties - what's going on?
Message-ID: <3c6376a8.15716308@news.muenster.de>

Is there a way to overcome "this last barrier"?
It would save my day and make my life *so much* easier !-)

Here it goes:
I would so much like being to launch Python modules from the
commandline given their module names instead of those often very long
and unpythonic filenames.

The launcher module below _almost_ does the trick - but one issue
remains unsolved. Modules launched this way will itself not be able to
import sibling modules from the package they reside in unless that
package itself is on the path or the import expression is fully
qualified. Which normally it isn't and it shouldn't be.

Is there a way to really and completely DO_THE_TRICK?

See code below, and thanks

Martin


"""Launch modules by module name """
"""instead of filename.

Place this module 'do.py' somewhere in the Python path.
To launch 'mymodule.py' from package 'pack_a' you may
then use

(1)
 python -c"import do" pack_a.mymodule
        [params ...] <infile.txt >outfile.txt
        
instead of

(2)
 python c:\...\longunpythonicpath\...\pack_a.mymodule.py
        [params ...] <infile.txt >outfile.txt

'do.py' will first remove '-c' from sys.argv and
afterwards try importing sys.argv[0]. It can handle
the import of hierarchically structured modules.

If DO_THE_TRICK ist true, the last module
will be imported with "__name__"=="__main__".

STILL UNSOLVED:

If 'mymodule.py' tries to just
'import siblingmodule' referring to
'pack_a/syblingmodule.py' this will
ONLY succeed with the standard filename method
(2) is used and not with the desired new
method (1) unless 'pack_a' itself
is on the Python path.

Why?

Is there a way to overcome this (last ;-)
barrier?

See also 'Locator.py' by Alex Martelli,
comp.lang.python, 2001-04-09.

Martin Bless, m.bless at gmx.de, 2002-02-07
"""

## Windoze users:
## Add the following line to AUTOEXEC.BAT:
## SET pycmd=python -c"import do"
## Later on launch Python modules from the commandline:
## %pycmd% my.wonderful.module

DO_THE_TRICK = 1

import sys, imp

if sys.argv:
    if sys.argv[0] == '-c':
        del sys.argv[0]
if not sys.argv:
    sys.exit()

head = []
tail = sys.argv[0].split('.')
path = None
while tail:
    part = tail.pop(0)
    head.append(part)
    try:
        if path:
            file, path, desc = imp.find_module(part,[path])
        else:
            file, path, desc = imp.find_module(part)
    except ImportError,msg:
        print "ImportError:", msg
        sys.exit(1)
    try:
        if not tail and DO_THE_TRICK:
            imp.load_module('__main__', file, path, desc)
        else:
            imp.load_module(".".join(head), file, path, desc)
    finally:
        if file:
            file.close()



From hamish_lawson at yahoo.co.uk  Thu Feb 14 10:13:22 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 14 Feb 2002 07:13:22 -0800
Subject: Is there a reverse operator to the '%' operator on string ?
References: <20020214.121107.1335354340.24152@cantal.lip6.fr>
Message-ID: <915a998f.0202140713.19bf2ed@posting.google.com>

Boris Boutillier wrote:

> Now  what I'd like to do is revert the '%' operator ... something
> like the 'scanf' C function

The usual way of doing this is with regular-expression matching. They
are a bit more verbose than scanf's patterns, but more general.
Employing named groups in your RE should give you the dictionary of
results that you are looking for.

Hamish Lawson


From mjbarber at ascc.artsci.wustl.edu  Wed Feb 27 13:22:02 2002
From: mjbarber at ascc.artsci.wustl.edu (Michael James Barber)
Date: 27 Feb 2002 12:22:02 -0600
Subject: Profiling memory usage?
References: <a5iogb$laa@ascc.artsci.wustl.edu> <a5j1vb$7mq9u$1@ID-11957.news.dfncis.de>
Message-ID: <a5j84a$qtt@ascc.artsci.wustl.edu>

In article <a5j1vb$7mq9u$1 at ID-11957.news.dfncis.de>,
Emile van Sebille <emile at fenx.com> wrote:
>"Michael James Barber" <mjbarber at ascc.artsci.wustl.edu> wrote in message
>news:a5iogb$laa at ascc.artsci.wustl.edu...
>>[are there memory profiling tools?]
>
>I also looked for a tool and found none.  One of my projects requires a
>
Drat.  Ah well, at least it supports the "no such tool" explanation over 
the "I'm incompetent at using search tools" explanation.  (Either that, or 
I have some bad news for you Emile... ;->)

>there weren't good internal profiling tools, I looked at memory usage
>using pstat.exe on windows, but you could probably use top in a similar
>kludgey output parsing way.  (Now someone can chime in with the *easy*
>way to do this ;-)  
>
Using 'top' is exactly what I'm doing, and it is quite kludgey. Seems like 
there must be an easy way, so I'll second this request!  For me, 
determining the memory used by an object would let me do the needed
profiling quite easily.

>
>The kinds of changes I made included:
[selectively snipping and rearranging a list of nice changes]
>  -- not holding on to references to large structures when I was done
>with them (doh)
>
This is what's killing me.  I need to hold the references somewhere, since 
it takes too long to generate the data not to preserve it all.  Could 
store to disk, I suppose.

>  -- forcing commonly used string parts of keys to be interned
>
What I'm doing is most analogous to this, I think.  I'll present my 
solution, in case others run into the same, and - in a totally 
self-serving fashion - in the the hopes that others will suggest 
improvements.

I think the approach I used is called the "Flyweight" design pattern - but 
I would need to reread the description of the pattern to be sure.  The 
class name could be wrong.  At any rate, here's the class:

class FlyweightFactory:
    """Simple class for working with the flyweight design pattern.  Call 
as a function to use.
    
    If fwf is a FlyweightFactory instance, use as: fw = fwf(obj).  If obj 
is not hashable, fw 
    will be obj itself (i.e., fw is obj). When obj is hashable, fw will be 
equal (fw == obj) to 
    obj, but not the same object (fw is not obj)."""
    
    def __init__(self):
        self._flyweights = {}
        
    def __call__(self,  obj):
        try:
            fw = self._flyweights.setdefault(obj, obj)
        except TypeError:
            # TypeError is raised when obj is mutable
            fw = obj
        return fw

Possibly, the try should be omitted, just letting the TypeError be raised.  
It may also be a better idea to make this a new style class by subclassing 
from dict.  And the docstring isn't very good. 

I was surprised by how easy this was to do.  I certainly hadn't expected 
that identifying where to use it would be the hard part!
-- 
					Michael J. Barber


From cguardia at aldeasystems.com  Mon Feb 25 14:22:58 2002
From: cguardia at aldeasystems.com (Carlos de la Guardia)
Date: Mon, 25 Feb 2002 13:22:58 -0600 (CST)
Subject: Repost: Any Pythonistas in Mexico?
Message-ID: <200202251917.g1PJH1k07875@mail.aldeasystems.com>

Hello,

Sorry abot the repost, I felt that maybe the message wasn't read.

my company is located in Mexico and we use Python for our development
projects, I am trying to find out how many of us Python enthusiasts live in
Mexico and see if maybe we can form a user group or something.

Thanks,

Carlos de la Guardia
Aldea Systems




From magnus at thinkware.se  Wed Feb 20 20:33:57 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 17:33:57 -0800
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com>
Message-ID: <258fd9b8.0202201733.3888e711@posting.google.com>

fumingw at yahoo.com (Fuming Wang) wrote in message news:<b0fd9226.0202201104.2c6a220 at posting.google.com>...
> Hi,
> 
> I am having trouble using the arrow key to recall commands in console
> interactive mode.  I even installed Cygwin and start Python from there
> but still got the same effect.  Is there anything I can do to get
> arrow key working?

You can use something like IDLE, PythonWin or PyCrust instead.

As suggested you can also use the cygwin version of Python,
but then I guess you will have trouble with a lot of extensions
that use C or C++ such as mxDateTime, win32com etc.

Then you can use a less crappy OS... Both Windows 2000 or linux
will solve this. You might be locked in I guess, but there are
really a lot of reasons to dump Win98...

Command.com is really such a piece of junk. Even compared to
cmd.exe in win 2000. You will soon curse other problems as well
if you do anything substancial at the DOS prompt. For instance
you don't have any possibility to scroll back in the screen
output. In cmd.exe or an xterm etc, you can go back and see what
all that text that flickered by really contained.

So, I suggest you get away from command.com one way or another.
Go GUI or change OS...


From kuran42 at yahoo.com  Fri Feb  1 23:51:35 2002
From: kuran42 at yahoo.com (Jp Calderone)
Date: Sat, 02 Feb 2002 04:51:35 GMT
Subject: urllib2 bug?
Message-ID: <3C5B7056.2080305@yahoo.com>


   Should urllib2.URLError.__init__ not call exceptions.IOError.__init__?
As it is, a URLError, despite being an IOError, is missing a few of 
IOError's attributes... which seems like a bug.

   Jp

-- 
"There is no reason for any individual to have a computer in their
home."
                 -- Ken Olson, President of DEC, World Future Society
                    Convention, 1977
--
  11:50pm up 3 days, 15:52, 0 users, load average: 1.59, 1.25, 1.09



From phr-n2002a at nightsong.com  Tue Feb  5 03:32:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 00:32:36 -0800
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com> <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu> <M6M78.23304$l93.4247257@newsb.telia.net>
Message-ID: <7xu1swe3bv.fsf@ruckus.brouhaha.com>

"Fredrik Lundh" <fredrik at pythonware.com> writes:
> > Any will do, e.g., uniform, normal, etc.
> 
> does any of the "uniform" or "normal" generators in the standard
> random module do what you want?
> 
> http://www.python.org/doc/current/lib/module-random.html
> 
>     This module implements pseudo-random number generators
>     for various distributions. For integers, uniform selection from
>     a range. For sequences, uniform selection of a random element,
>     and a function to generate a random permutation of a list in-
>     place. On the real line, there are functions to compute uniform,
>     normal (Gaussian), lognormal, negative exponential, gamma,
>     and beta distributions. For generating distribution of angles,
>     the circular uniform and von Mises distributions are available.
> 
> no "etc" generator in there, though.

Skewness is a measure of the assymetry of a distribution.  I don't
remember the formula but it's similar to the formula for standard
deviation except it involves (x-mean)**3 instead of (x-mean)**2.  A
normal distribution (bell shaped curve) is symmetrical around the mean
so its skewness is zero.  Same for a uniform distribution (same
density everywhere).  Some of the other available distributions
(exponential etc.) are assymetrical and so they have nonzero skewness,
but the OP asked for a generator where you could just specify what
skewness you want.  The random module doesn't give a direct way to do
that.  You have to pick a distribution, then crunch out some integrals
and solve some equations to get the distribution parameters for the
skewness you want.

Further description:
http://www.everything2.net/index.pl?node=skewness


From bkelley at wi.mit.edu  Tue Feb 19 10:04:19 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Tue, 19 Feb 2002 10:04:19 -0500
Subject: Warning about "for line in file:"
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu> <mailman.1013799686.7054.python-list@python.org> <3C6D73DE.3090201@wi.mit.edu> <c76ff6fc.0202161443.3ba0f746@posting.google.com>
Message-ID: <3C726973.8080401@wi.mit.edu>

John Machin wrote:

> Brian Kelley <bkelley at wi.mit.edu> wrote in message news:<3C6D73DE.3090201 at wi.mit.edu>...
> [snip]
> 
>>The behavior that I would expect is exemplified in the following correct 
>>example that actually prints out all lines of a file.
>>
> 
> Correct? Does "a file" mean a particular file with 11 or fewer lines?
> It certainly doesn't work in generality for larger files:
> 
> C:\junk>cat iter_titter.py
> fsim = ["Line " + str(x) for x in range(20)]
> count = 0
> iterator = iter(fsim)
> for line in iterator:
>      if count > 10: break
>      print line
>      count = count + 1
> for line in iterator:
>      print line
> 


This is just a bug in the codingm you need to move the break statement 
after the "print line" otherwise the last iteration will not be seen.

fsim = ["Line " + str(x) for x in range(20)]
count = 0
iterator = iter(fsim)
for line in iterator:
      print line
      count = count + 1
      if count > 10: break

for line in iterator:
      print line


> C:\junk>python iter_titter.py
Line 0
Line 1
Line 2
Line 3
Line 4
Line 5
Line 6
Line 7
Line 8
Line 9
Line 10
Line 11
Line 12
Line 13
Line 14
Line 15
Line 16
Line 17
Line 18
Line 19



From sholden at pgrad.cs.usyd.edu.au  Wed Feb  6 01:52:58 2002
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 6 Feb 2002 06:52:58 GMT
Subject: Autocoding project proposal.
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net>
Message-ID: <slrna61kma.57s.sholden@pgrad.cs.usyd.edu.au>

On Wed, 06 Feb 2002 04:54:13 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 03-Feb-02 23:18:30 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>On Mon, 04 Feb 2002 00:00:49 GMT, Timothy Rue <threeseas at earthlink.net>
>>wrote:
>>>On 30-Jan-02 23:23:31 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>>
>>>>This is a discuss group (gnu,misc.discuss anyway). It is not a place to
>>>>seek coding help. If you don't want a debate don't post to a discussion
>>>>forum. If comp.lang.python is like every other comp.lang group I know then
>>>>it is also not a place to seek coding help - other than specific language
>>>>problems
>>>>(such as how do I convert a string into a number?).
>>>
>>>>You just don't understand how programmers work. And until you do you
>>>>won't find many willing to help you. In fact if someone did think your
>>>>ideas were good they would probably just fork your code and do it
>>>>themselves without your involvement - since you seem to be going out of
>>>>your way to turn
>>>> people off your project.
>>>
>>>
>>>Don't like the idea of autocoding huh?
>
>>Actually I have no opinion on it at all. I don't have time to get involved in
>>a project at the moment, and thuis I haven't tried to understand your long
>>explanations of what it is.
>
>>Since I don't know what it actually is, I don't know if I like or don't
>>like the idea.
>
>>If it is something that will help non-programmers better use the
>>computer resources they have access to without having to learn how
>>to program, then it would actually be something I would be interested
>>in - since that is an area I am interested in.
>
>>If it is that old assembler type thing then I'm not intersted since
>>that is an area I have no interest in.
>
>>If it is to do with autocoder by Smart Software Company, then I'm also
>>not interested, as automating open ended questions is not something I
>>personally find interesting.
>
>>I only posted to try and help you better communicate with programmers, and
>>if your reply is anything to go by you didn;t bother reading my advice,
>>or didn't think it was valid. Since your reply has not generated
>>any interest in me (a programmer) for your project - but that really
>>doesn't matter, as as I said, I'm busy at the moment.
>
>
>>Sam Holden
>
>
>And so is the logic failure of a programmer.
>
>
>You don't have the time to find out if I'm on to something that would
>interest you or not, yet you are arrogant enough to give advice regarding
>something you don't know about because you don't have the time to find out
>if what I'm on to is something you'd be interested in or not.

I gave advice that was independant of what you are actually doing. You
mentioned you were looking for programmers to help with a project. I am
a programmer, and I found your descriptions too hard to understand, to be worth
spending time on. I think I am pretty typical for a programmer. I am not
especially good at understanding things, and not especially poor. I suspect
other programmers will also find your project descriptions hard to understand.

If you don't care that others have trouble understanding, that's fine but it
means you have limited the potential population of those who might help you.

If you do care then I gave you some advice on what types of descriptions the
programmers I know find understandable and useful.

You don't have to take the advice, feel free to keep alientating the
people you asked for assistance from. From your more recent posts I see you
have basically given up on programmers (and their conspiracy to keep 
computers from the messes) so I guess you don't want the advice. 

If you aren't looking for programmers why are posting to these groups?

>
>Full circle!
>
>My first post had links that would have eliminated your multipul choice
>list above. But you don't have time for that, do you?

As I've said before I don't have time to work on a project at the moment. I
do have time to read a description that is of reasonable length, but I
can't understand your language. And since you have classed me as an arrogant,
elitist who is conspiring to prevent society from advancing, and as an
illogical fool, I don't have a lot of motivation to struggle through the 
text.

>
>What advice?

The huge wad of text, that described how to better communicate to the
programmers that I know (and I extrapolate to other programmers I don't know).

>
>Here some for you. Start thinking logically!

I think I already do - not perfectly of course, I'm a human I have
emotions and prejudices that get in the way of pure logical thought. I
don't understand what is illogical about offering advice on something I
have more experience than you in (in all probability anyway). I know
understand that trying to communicate with you is illogical, as you seem
to just want to rant without communicating anything.

Yet another programmer you have driven away by meaningless ranting. Then
again my thesis won't be finished until at least late this year, so I
wouldn't have been able to chip in until then anyway. 

-- 
Sam Holden



From xx758 at cam.ac.uk  Fri Feb 15 13:58:03 2002
From: xx758 at cam.ac.uk (Xiao-Qin Xia)
Date: Fri, 15 Feb 2002 18:58:03 +0000
Subject: Can Tk be connect with Copyboard of Linux
Message-ID: <a4jlhf$6uv$1@pegasus.csx.cam.ac.uk>

Hi, everybody,

Under windows, the copyboard is shared by all the applications. So I can 
copy some messages from different applications, and paste them into a Text 
or Entry widgit based on Tkinter/Tk. But I found the same operation cannot 
work under linux. Under KDE, I can not paste messages from other 
applications into a widget based on Tkinter/Tk. Is it possible to solve 
this problem?

Cheers,
Xiao-Qin


From gustav at morpheus.demon.co.uk  Fri Feb 15 17:05:31 2002
From: gustav at morpheus.demon.co.uk (Paul Moore)
Date: Fri, 15 Feb 2002 22:05:31 +0000
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82>
Message-ID: <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com>

On 14 Feb 2002 23:34:12 +0100, Chris Liechti <cliechti at gmx.net> wrote:

>Paul Moore <gustav at morpheus.demon.co.uk> wrote in
>news:efao6us1tt3ut2hp6aor19p7qfqad3k4t1 at 4ax.com: 
>...
>> Hmm. Looks more messy than I'd hoped. Needs a bootstrap module, and
>> then extra hacking to handle open/read. And I guess the fact that 
>> I'm doing this manually means that there's not much chance of 
>> something like a wxPython toolbar "just working" with icons stored
>> in the archive. 
>
>well you could pack the stuff in a zipfile and retreive the data with 
>the zipfile module. you get the data as string but you can wrap that 
>again in a cStringIO object to get a file-like object.

I looked a bit closer at this. Interesting. Using the zipfile module
makes it trivial to get the raw data from an entry in a zipfile. And as
you say, I can use cStringIO to get a Python file object from this.

For text files, that's all I need really.

For sound files (on Windows) I can use the winsound module, with the
SND_MEMORY option to play from a memory image.

>but then wxWindows Bitmap.LoadFile wants real file...
>but: """wxBitmapFromXPMData(listOfStrings)  Create a bitmap from a 
>Python list of strings whose contents are XPM data""" so you would be 
>limited to XPM images.

Looks like it's wxPython that has the problem...

Annoyingly, wxWindows has a wxImage constructor, wxImage(wxInputStream)
to load from a stream, but wxPython *doesn't* seem to wrap this
constructor. OK, so there would need to be a conversion from a Python
file object to a wxInputStream (or something).

But some way of getting a wxImage from in-memory data (of arbitrary
format) is really all I'm after - and there ought to be a way to get it.
I've copied this to the wxPython list. Is there some way I've missed -
or could a way be added?

Thanks,
Paul Moore


From spencer at efn.org  Fri Feb  1 18:49:10 2002
From: spencer at efn.org (Spencer Ernest Doidge)
Date: Fri, 1 Feb 2002 15:49:10 -0800
Subject: Windows, RS232, and Python
In-Reply-To: <3c5ada50$0$2204$7586b60c@news.frii.com>
References: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org> <3c5ada50$0$2204$7586b60c@news.frii.com>
Message-ID: <Pine.GSU.4.21.0202011548230.28552-100000@garcia.efn.org>

Definitely this helps, I think. The author mentions "the win32 extension
modules." Do you know what exactly he is referring to?

Spencer Doidge
www.spencerdoidge.com

On Fri, 1 Feb 2002, Marc Poulin wrote:

> In article <Pine.GSU.4.21.0202010910040.11090-100000 at garcia.efn.org>,
> "Spencer Ernest Doidge" <spencer at efn.org> wrote:
> 
> > I want to have control of every aspect of RS232 transmitting and
> > receiving in Win9x/NT with Python commands.
> > 
> 
> Does this help? 
> http://balder.prohosting.com/ibarona/en/python/uspp/uspp_en.html
> 



From peter at engcorp.com  Sun Feb 24 15:21:34 2002
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Feb 2002 15:21:34 -0500
Subject: execute file interactively
References: <e9e6dd1f.0202171731.6cf2a5e2@posting.google.com>
Message-ID: <3C794B4E.C20890A7@engcorp.com>

Jason wrote:
> 
> When using the Python interpreter interactively, how can I execute a
> file, so I can use the function definitions and data defined in it
> interactively.

'python -i filename.py' might be of some help.

Use 'python -h' to learn what that would do.


From logiplexsoftware at earthlink.net  Fri Feb 15 15:02:02 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 15 Feb 2002 12:02:02 -0800
Subject: Why Python is like BASIC (and why this is a good thing)
In-Reply-To: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org>
Message-ID: <20020215120202.3444a368.logiplexsoftware@earthlink.net>

On 15 Feb 2002 18:59:13 GMT
Robin Munn wrote:
[snip]

> All three of these things make for a very low barrier-to-entry for
> people who've never done any programming before. I'm sure there are
> other similarities as well, that maybe someone else will point out. Of
> course, Python goes beyond BASIC in many ways:

Well, sort of.  Of course air is like water too, except for breathing ;) 

[snip]
 
> * Batteries included. This is the big one, which makes Python more than
>   just a toy language suitable for learning. The richness of the
>   standard library makes Python incredibly useful, and I for one would
>   like to see it continue to grow. I understand the concern that people
>   developing Python-based programs for distribution would like their
>   users to be able to run a 100k script without downloading a 10M
>   distribution, but I don't want to see that cause a reduction in the
>   standard library. Instead, we could make make two Python
>   distributions, a "full" install containing everything and a "lite"
>   install containing only the interpreter and maybe -- *maybe* -- a
>   couple of essential modules like os and sys. Then people developing
>   Python programs like, say, a pygame-based game, could distribute a
>   single package containing the "Python lite" distribution, their own
>   code, and any other modules their code depended on -- all in one
>   easy-for-the-end-used-to-install package.

We had a thread on this topic a week or so ago.  I agree that there should
be a standard, "batteries-included/kitchen-sink" distro and a "lite" distro
that can then be added to as necessary.  However, as was mentioned by
someone (Tim Peters?), someone will need to do the work to make this
happen.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From daves_spam_dodging_account at yahoo.com  Tue Feb  5 13:30:05 2002
From: daves_spam_dodging_account at yahoo.com (David Brady)
Date: Tue, 5 Feb 2002 10:30:05 -0800 (PST)
Subject: RFC -- HYGIENIC MACROS IN PYTHON
In-Reply-To: <AFE3FBA300B6F04282BE07949DB2E12303F3C763@TORINO.corp.es.com>
Message-ID: <20020205183005.45824.qmail@web21107.mail.yahoo.com>

I've looked over your lingos examples, and my reaction
is twofold:  One, I'm excited that Python is powerful
and flexible enough to let you experiment with this
kind of stuff.  Two, you are dabbling in the dark arts
and are at risk of condemning your immortal soul to
the fires of hell.  ;-)

I am far from a Python guru, so I hardly speak for the
PSU; but I *am* an accomplished perl hacker who has
completely forsaken his immoral ways and converted to
the true light of Python.  What you are doing looks to
me like trying to bring perl and C syntax into Python,
rather than embracing the appropriate Python approach.

Specifically, a maxim of Python is that there should
be a right way to do everything, as opposed to there
being a hojillion-zillion different ways to do it (a
la perl).

Please understand I'm not attacking you or anything,
it's just that the examples you've shown are solutions
to problems that don't exist in Python.

Invariant: I don't see the purpose of the invariant
lingo, except as the target of a goto command.  Goto
has it's place, usually in trying to clean up
monolithic C code.  In Python, however, a better
solution is to simply design so that gotos are not
necessary.  Python replacement: def.

Unless: if not.  Having an unless keyword would make
*some* python code more clear to perl immigrants; but
it's only a matter of (a very short) time before
somebody writes "unless not" instead of "if".  Trust
me--I've seen that kind of syntax in perl, usually
written by people not paying attention.  (That's a
dirty little secret, isn't it?  That people write code
while half asleep, or they make a quick logic change
without really thinking it through?  But they do.  And
with the "one best way" maxim of Python, the code has
a greater tendency to survive these somnogenic edits.)

factor.  I'm not sure what "factor" does, but it looks
like a function call without the parens.

f(i): i=3. (Functions modifying their args.)  This is
the thing that tipped me off to your turning to the
dark side.  :-)  I struggled to get my mind around
this exact concept when I first came to Python.  The
fact is (as best I can tell) that Python deliberately
does NOT permit this.  If you want to modify a
variable, you HAVE to do it explicitly: i = f(i).

do.  Python replacement: while.  I agree that do and
while are not the same thing; do ensures that a loop
is executed at least once.  But again, with the
constraint of "just one way to loop" solutions seem to
sort themselves out.  So you have to write a
precondition; that's just a simple

p=GetFirst()
while(p):
    p = GetNext(p)

Personally, I think switch and do..while might be
worthy of a PEP, but in their absence I find myself
not just coding around their absence, but actually
*thinking* differently about the code.  For example, I
have started trying to redesign switch(x) syntax so
that x is an object instead of a condition, and
x.DoSomething() is the appropriate case/action for
that object.

This is just my $0.02.  Again, I think you may be on
to something, and I think it's cool that Python is
flexible enough to let you get there.  Good luck, and
please keep reporting on what you come up with!

Cheers,

-dB

=====
David Brady
daves_spam_dodging_account at yahoo.com
I'm feeling very surreal today... or *AM* I?

__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com



From fperez528 at yahoo.com  Sun Feb 10 11:13:44 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Sun, 10 Feb 2002 16:13:44 +0000
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org> <3C66F2CA.802A00BF@earthlink.net> <86lme1xdoc.fsf@apo.hmotakef.int>
Message-ID: <a46uun$gc7$2@peabody.colorado.edu>

Henrik Motakef wrote:

> This might be slightly off-topic, but usually pressing both buttons at
> once emulates pressing the middle one...

As long as you have 

    Option "Emulate3Buttons"
    Option "Emulate3Timeout" "50"

in your "InputDevice" section of /etc/X11/XF86Config-4

cheers,

f.


From mikael.lexen at stop.spam.volvo.com  Mon Feb 18 06:17:44 2002
From: mikael.lexen at stop.spam.volvo.com (Mikael =?iso-8859-1?Q?Lex=E9n?=)
Date: Mon, 18 Feb 2002 12:17:44 +0100
Subject: VB code -> Python
References: <3C6CDDC8.4ED3B329@stop.spam.volvo.com>
Message-ID: <3C70E2D8.BD35E149@stop.spam.volvo.com>

Thanks to everyone for the help. It works now.

/Mikael



From aahz at panix.com  Fri Feb 15 08:02:00 2002
From: aahz at panix.com (Aahz Maruch)
Date: 15 Feb 2002 05:02:00 -0800
Subject: Python License Issues
References: <mailman.1013768726.14741.python-list@python.org>
Message-ID: <a4j0s8$k1d$1@panix2.panix.com>

In article <mailman.1013768726.14741.python-list at python.org>,
 <christoforou at gaaros.com> wrote:
>
>What worries me is that the 'legal climate' of our times will hurt open 
>source acceptance and small soft shops in general. For it seems to me that 
>the recent escalation in legal actions against third party IP infringments 
>are driving coorporate clients to seek indemnity from IP violations. In 
>their (reasonable) view the cost of software is: 
>
> Cost= ActualCostOfApplication  
>+(ProbabilityOfLegalActionRelatingToApplication*SettlementAmount) 

Your equation ignores the fact that SettleAmount is hardly the only
possible cost.  No matter what the indemnity, if software you rely on is
withdrawn due to a suit, you'll have to eat the cost of using another
software package.  Heck, if the software is withdrawn without a lawsuit
you'll still have to eat the cost -- and that is a major disadvantage of
commercial software.  You picks your poison and you takes your chances.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"usenet imitates usenet"  --Darkhawk


From michael at stroeder.com  Fri Feb 22 06:34:25 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 22 Feb 2002 12:34:25 +0100
Subject: HTTP state management without cookies?
References: <mailman.1014303815.10286.python-list@python.org> <7xbseizx7c.fsf@ruckus.brouhaha.com>
Message-ID: <3C762CC1.CB0DEEF4@stroeder.com>

Paul Rubin wrote:
> 
> mod_rewrite lets you do all kinds of neat stuff.

Hmm, I'd like to rephrase that to "mod_rewrite let's you do all kind
of dirty stuff".

I consider the configuration and change managment of your server and
application to be a nightmare and security risk if you rely on
handling of session IDs implemented in the server's mod_rewrite
configuration.

Ciao, Michael.


From magnus at thinkware.se  Wed Feb 20 05:05:30 2002
From: magnus at thinkware.se (=?ISO-8859-1?Q?Magnus_Lyck=E5?=)
Date: 20 Feb 2002 02:05:30 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <3C7283E4.4050604@physics.ucsb.edu>
Message-ID: <258fd9b8.0202200205.7b0bc418@posting.google.com>

Craig Maloney <cmaloney at physics.ucsb.edu> wrote in message news:<3C7283E4.4050604 at physics.ucsb.edu>...
> I have no experience with Boost.Python, but from what I
> understand, it would not involve writing wrapper definitions
> (ala SWIG or IDL), right?  Are there other options?

There are MANY options. :-) Some are listed at:

http://www.thinkware.se/cgi-bin/thinki.cgi/UsingPythonWithOtherLanguages


From mrchameleon at hotmail.com  Sat Feb  2 19:33:37 2002
From: mrchameleon at hotmail.com (Chris Reay)
Date: 2 Feb 2002 16:33:37 -0800
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012671184.6510.python-list@python.org>
Message-ID: <7652139e.0202021633.19db946@posting.google.com>

http://andrich.net/python/selfmade.html#ncursesmodule 

This gave me 404.


From dsavitsk at e-coli.net  Fri Feb  8 00:56:57 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 08 Feb 2002 05:56:57 GMT
Subject: Delphi vs Python
References: <zBm88.333397$pa1.80603034@news3.rdc1.on.home.com>
Message-ID: <JMJ88.2389$pC1.447616@newssrv26.news.prodigy.com>

i had the experience of learning VB, getting comfy with it, then learning
python.  this, i believe, led me to appreciate python that much more than i
ever could of had it been my first. so, to ever really know what you are
gaining by using python, you should learn something sucky first.

-d


"Bo Vandenberg" <bosahv at netscapenospam.net> wrote in message
news:zBm88.333397$pa1.80603034 at news3.rdc1.on.home.com...
> I'm planning learning strategy and I thought I would ask people who use
both
> Python and Delphi their opinion.
>
> I've determined that I at least cannot speedily learn both at once. So my
> current strategy is to get up to speed in Python first and then maybe do
the
> slowest bits of code in Delphi.
>
> Since Kylix my head spinning with options but I think Python as a learning
> language (This is my first computer language) is turning the tide.
>
> Has anyone used Delphi to make python called DLL's ?
>
> Has anyone seen any Delphi 6 personal edtn. stuff show up to serve python?
>
>
> Any opinion would be appreciated.
>
> Bo
>
>




From baas at ira.uka.de  Wed Feb 27 04:31:15 2002
From: baas at ira.uka.de (Matthias Baas)
Date: Wed, 27 Feb 2002 10:31:15 +0100
Subject: controlling web browser in windows
References: <1103_1013605186@netnews.cf.ac.uk> <Zeua8.206395$KM2.8473296@atlpnn01.usenetserver.com> <3C6ACDB6.95186821@ccvcorp.com>
Message-ID: <ea9p7u85b38k4gdgf5bc4mjlovq7p4ta61@4ax.com>

On Wed, 13 Feb 2002 12:33:58 -0800, Jeff Shannon <jeff at ccvcorp.com>
wrote:
>Of course, I then discovered that much of what I *wanted* to do, was forbidden
>from scripts for security reasons, but that's another story.  ;)  

Reminds me of something.....recently I was also playing around with
COM and IE and tried to navigate to a page and save it to disk.
Navigating was no problem (using the Navigate() method) but saving the
page didn't work. I tried to save it via the ExecWB() call, but that
only raised an exception (which didn't really contain a useful error
message. The message was in German, so I think it doesn't help much if
I post it). So my question: Is that an example of what is forbidden
from Python? Or is there a way to save the content to disk
automatically?

And another thing that confused me: When I start the COM makepy
utility I don't see IE in the list ("InternetExplorer.Application"
doesn't show up). Why not?

- Matthias -



From Roman.Fischer at teleweb.at  Sun Feb 10 15:31:57 2002
From: Roman.Fischer at teleweb.at (Ing. Roman Fischer)
Date: Sun, 10 Feb 2002 20:31:57 GMT
Subject: Is Parrot dead?
Message-ID: <B88C9617.532F%Roman.Fischer@teleweb.at>

For a long time I do not find informations about Parrot (the programming
language). O Reilly still estimates July 2001 for the "Programming in Parrot
in a Nutshell"-book and even Amazon (which is not very shy to offer books
long before they are available) does not have it in its offer. I also did
not find any real complete definitions of this language.

So is this language dead?

Roman



From gerson.kurz at t-online.de  Mon Feb 25 09:17:37 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 14:17:37 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com>
Message-ID: <3c7a46c0.4540406@news.t-online.de>

And then, there is always the braindead solution to choose from

------------------------ (cut here) ------------------------
import sys, os

# create a list of unique functions printing values
functions = [lambda x=i:sys.stdout.write("f(%d) called\n" % x) for i
in range(10)]

print "The easy way:"

for function in functions:
    function()

print "The braindead way:"

callem = lambda functions,l=[tempfile.mktemp()]:map(eval,\
('sys.stdout.write(str(l))','l.append(open(l[0],"w"))',
 'map(lambda x,l=l:l[1].write("functions[%d]()\\n"%x),'\
 'xrange(len(functions)))','l[1].close()','execfile(l[0])',
 'os.remove(l[0])'))

callem(functions)
------------------------ (cut here) ------------------------
The code creates a temporary file and then uses execfile() to execute
it. This is not threadsafe ;) I cannot use "exec" because thats a
statement & not allowed in lambda.

Note that I sense a slight problem with the scoping rules in the
callem code: the lambda in line 3 has to pass "l=l", otherwise the
code won't work. I have no idea why, because I was under the
impression that that is not necessary with those new rules. Go figure
it out ;)



From joshm at taconic.net  Mon Feb 11 01:35:00 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Mon, 11 Feb 2002 01:35:00 -0500
Subject: xor on strings
References: <QbH98.3291$pC1.663612@newssrv26.news.prodigy.com> <3c674cbd_2@corp.newsgroups.com> <%hI98.3302$pC1.668524@newssrv26.news.prodigy.com> <vqI98.3303$pC1.669077@newssrv26.news.prodigy.com>
Message-ID: <3c6763c6_3@corp.newsgroups.com>

It isn't clear why you want to compare to '00', but this code should do the
trick nicely:

>>> (r,g,b) = (56,156,12)
>>> theString = '&H%02x%02x%02x' % (b,g,r) # note the order!
>>> theString
'&H0c9c38'

Is the test for '00' for anything other than the formatting of the string?
Are you trying to find the brightest color, or something?
--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From tanzer at swing.co.at  Fri Feb 15 09:15:27 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 15 Feb 2002 15:15:27 +0100
Subject: Pythonistas? <- Re: OT: Perl programmers?
In-Reply-To: Your message of "Thu, 14 Feb 2002 10:36:04 PST."
             <3C6C0393.F7C16D06@ccvcorp.com>
Message-ID: <E16bj95-000424-00@tswings.swing.cluster>

Jeff Shannon <jeff at ccvcorp.com> wrote:

> Tim Hammerquist wrote:
>
> > OTOH, exceptions in Spanish are much less frequent than in English or
> > French.  Pronunciation is nearly 100% predictable, and there are more
> > regular verbs than most languages.
> >
> > (I have no knowledge of German.)
>
> German has fewer exceptions than English.  There are a fair number of
> irregular verbs, but many of those follow similar patterns.
> Pronunciation is also extremely predictable, unlike English.

All true. The only problem in German is that there are two
incompatible standards <wink>.

For instance,

`ich bin angefangen...` vs. `ich habe angefangen...` (both mean `I
started...`). Only one is correct but which one depends on the area
where you live.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From maxm at mxm.dk  Wed Feb 27 10:45:29 2002
From: maxm at mxm.dk (Max M)
Date: Wed, 27 Feb 2002 15:45:29 +0000
Subject: string title() method does not work on Danish characters
Message-ID: <3C7CFF19.5070703@mxm.dk>

I am writing strings from a database, and am using the title() method to 
format the look of the strings.

like 'this is a name: MAX M?LLER'.title()
 >>> Max M?ller

But the Danish characters does not get the right case.

I tried:

import locale
locale.setlocale(locale.LC_ALL, 'de')

but get:

"C:\python\Python21\Lib\locale.py", line 374, in setlocale return 
_setlocale(category, locale) Error: locale setting not supported

What else can I do? Please.

regards Max M



From bokr at oz.net  Mon Feb  4 03:15:11 2002
From: bokr at oz.net (Bengt Richter)
Date: 4 Feb 2002 08:15:11 GMT
Subject: Equivalent of Perl chomp?
References: <mailman.1012626605.22240.python-list@python.org>
Message-ID: <a3lfuf$6sj$0@216.39.172.122>

On Sat, 2 Feb 2002 00:09:15 -0500, "Tim Peters" <tim.one at home.com> wrote:

>[Steve Holden]
>> 1,000,000 iterations.
>>
>> Testing for length 1:
>> Endswith 0.00000505
>> sliceIt  0.00000209
>>
>> Testing for length 2:
>> Endswith 0.00001137
>> sliceIt  0.00001148
>>
>> Length 3:
>> Endswith 0.00001055     [!]
>> sliceIt  0.00001264
>>
>> Of course, these tests are with the tests succeeding. I'm
>> presuming failure
>> would shorten the timings, but I'm not going to bother checking that
>> assumption.
>>
>> semi-believab-ly y'rs  - steve
>
>This is the kind of fun I was hoping the OP would have.  Good job!  A lesson
>to take is that timing even seemingly trivial things can turn up major
>surprises with seemingly trivial changes in exactly what it is you're
>timing.  The results are almost never accidental, but determining the true
>causes can leave you crawling on your belly for weeks begging for relief.
>
>A few things going on in this test under the covers:
>
>+ endswith does indeed pay a large price for building a bound
>  method object every time.  It's not only the dict lookup to
>  resolve "endswith", but also malloc and free to build the
>  bound method object and discard it afterwards.
>
>+ The huge increase in sliceit's time going from 1 to 2 is due to
>  that s[:-1] creates a one-character string, and Python interns
ITYM s[-1:] ;-)
>  those by magic:  every instance of, e.g., 'a', no matter how
>  computed, is exactly the same object (that's not actually true,
>  but it's close <wink>).  But 2-character computed strings are
>  not interned, so s[:-2] has to malloc a fresh string object each
...similarly
>  time to hold the two characters, and free it later.  endswith
>  doesn't have to build new string objects, so doesn't care as much.
>
>Now for some real fun:  I have no idea why endswith took a huge jump going
I don't think _it_ really did (my results don't show it ;-).

>from 1 to 2.  The saving grace for me is that it's so easy to find timing
>mysteries that I'd go mad if I felt compelled to chase them all down, so I
>can walk away without even looking back.  I hope not everyone is so
>well-adjusted <wink>, though, as I'd like to know why.
>
I tried to eliminate spurious big times to get a better measure. My goal
was to get a minimum time, hoping that spurious time-consuming effects
in the scaffolding code (loops, hidden memory allocations/deallocations,
interrupt processing, etc.) would be near equivalent in the best case.
Then I tried to make the best case more likely for the timed segment.

It is interesting how hard it is to time things on the order of a few
microseconds properly in this kind of context.

Code to try to counter some of these effects is at the end of this post.
Suggestions for improvements welcome.

Whereas I like to count cycles for an algorithm, it _is_ easy to have the
interpreter fool you. Scaffolding code has cache effects, and it's easy
to generate a lot of cache-smashing activity with interpreted code.

Putting just the thing of interest inside a loop of a million causes
other effects, and you hope they get averaged out, but there can be some
large outliers in the average.

In C/C++ timing 'nothing' takes 32 cycles on my box. Using an extension to do the
same pattern in python takes about 450 cycles. Which of those extra cycles are
invalidating cache, and which aren't? It's pretty hard to say. OTOH, if there
were a VM bytecode which logged a timestamp in an internal pre-configured
circular buffer, it could be done in very few instructions, minimizing disturbance.

Anyway, there's more to getting net time than subtracting out the _apparent_ cost
of "timing nothing."

Say the minimum cycles for executing x is X, and that sampling time takes T
cycles, with Tb cycles before the instantaneous sample point (say when
rdtsc instruction completes), and Ta cycles after. Let's call a non-optimum
time T' to show some unknown cache miss effect etc. Then simple timing of x
gives us
   dt = Ta' + X' + Tb'
I'm thinking that we can insert dummy T and X operations, on the assumption
that doing X' will assure as good a cache situation as possible for a
following X (therefore written without the [']) -- so we get
   dt  = T'*0 + Ta + X' + Tb'
(the '*0' meaning we ignore the value, just initializing cache so
we get predictable Ta at the beginning of the interval), and
   dt2 = T'*0 + Ta + X' + X + Tb'
and then subtracting,
   X = dt2-dt
We shouldn't have to know what the ['] effects are except to hope
that their _minimums_ would be closely repeatable and comparable in
the two sequences.

On this theory, I timed s.endswith and slice expression for
1- to 4-character tails. Also a "nothing" routine to subtract
the call/return time. I added gc just before each of the two
timing loops, and that seemed to narrow the variability a bit.
I also added an outer loop to average minimums found in the
inner loops and take advantage to get global minimums.

BTW, these are ticks at 300mhz on a P2 with 320MB RAM, with NT4.

[22:59] C:\pywk\crdtsc>python endsvslice.py 0 10   0 1000
10 loops, 1000 avgloops, 0 tadj
   nothing:   850, avg   856

(I.e., ten loops in inner loop finding a minimum estimated X, and 1000
of those averaging the minimums, as well as finding the global minimums
whose difference might be a better estimate of X, where X is theoretical
best time for the function call.)

(The values below are adjusted by 850 for 'nothing', which is
a function returning a constant 1, so were estimating how much
longer it takes to prepare the 1 for return using endswith or
slice ==, compared to the constant 1. Non-result lines snipped.)

[23:00] C:\pywk\crdtsc>python endsvslice.py 1 10 850 1000
10 loops, 1000 avgloops, 850 tadj
 endswith1:  1312, avg  1327
 endswith2:  1335, avg  1366
 endswith3:  1368, avg  1407
 endswith4:  1432, avg  1388
    slice1:   669, avg   740
    slice2:  1898, avg  1942
    slice3:  1937, avg  1934
    slice4:  1915, avg  1907

IOW, endswith did not take a huge jump measuring my way, but
the slice was certainly affected from 1 to 2.

The little dip for slice4 is curious. Here are some more trials of that:
    slice4:  1898, avg  1916
    slice4:  1909, avg  1913
    slice4:  1909, avg  1913
Hm, maybe slice3 above is off?
    slice3:  1945, avg  1934
    slice3:  1926, avg  1933
    slice3:  1928, avg  1942
    slice3:  1927, avg  1941
Curiouser and curiouser... anyway, no huge anomalies ;-)

Regards,
Bengt Richter

=======================================================================

Usage: python endsvslice.py id nloops adjust avgloops
  where id is:
  0: nothing
  1: endswith1
  2: endswith2
  3: endswith3
  4: endswith4
  5: slice1
  6: slice2
  7: slice3
  8: slice4

================ endsvslice.py ========================================
# endsvslice.py
import readtsc   #experimental, provides access to pentium rdtsc
import gc

def nothing(s):
    return 1

# brute force cut/paste to minimize scaffolding code
def endswith1(s):
    return s.endswith('\n')
def endswith2(s):
    return s.endswith(' \n')
def endswith3(s):
    return s.endswith('  \n')
def endswith4(s):
    return s.endswith('   \n')

def slice1(s):
    return s[-1:] == '\n'
def slice2(s):
    return s[-2:] == ' \n'
def slice3(s):
    return s[-3:] == '  \n'
def slice4(s):
    return s[-4:] == '   \n'

# Notation:
#  X is minimum time, X' may be disturbed by cache miss etc.
#  Tb is time before time sampling instant[1], Ta is time after
#  T is total time in time sample code, hence  T=Tb+Ta
#  [1] e.g., instant might be defined as when rdtsc instruction copies counter
def t(loops,avgloops,func,arg):
#    print loops,avgloops,func,len(arg)
    r = readtsc.readtsc_dt # returns delta since last lower 32 bits of 64-bit time stamp
    sum = sum2 = 0; gmin = gmin2 = 0x7fffffff
    for aloop in xrange(avgloops):
        dt = 0x7fffffff         # not likely min
        gc.collect()            # attempt to normalize state somewhat
        for i in xrange(loops):
            r()    # T'         # Possibly non-optimal
            r()    # Ta         # Tb is before, and Ta is hopefully run from cache
            func(arg) # X'      # x code not fully cached
            e=r()  # Tb'        # Tb' time before sample may have cache miss etc.
            dt = min(dt,e)      # min (Ta + X' + Tb')
        dt2 = 0x7fffffff        # not likely min
        gc.collect()
        for i in xrange(loops): # same as above except x is run cached one time
            r()    # T'
            r()    # Ta
            func(arg) # X'
            func(arg) # X       # x code hopefully cached after preceding X'
            e=r()   # Tb'
            dt2 = min(dt2,e)    # min (Ta + X' + X + Tb')
        ddt = dt2-dt            # X, more or less
        sum += ddt
        gmin = min(gmin,dt)
        gmin2 = min(gmin2,dt2)    # finding min X
    return gmin2-gmin,(sum)/avgloops # should be X time for one clean func(arg) call

if __name__=='__main__':
    import sys
    selections = [nothing,endswith1,endswith2,endswith3,endswith4,
        slice1,slice2,slice3,slice4]
    try:
        s='s'*1000000+'\n'; tadj = 0; loops = 10; avgloops = 1
        id = int(sys.argv[1])
        if len(sys.argv)>2: loops = int(sys.argv[2])
        if len(sys.argv)>3: tadj= int(sys.argv[3])
        if len(sys.argv)>4: avgloops= int(sys.argv[4])
        tm,tav =t(loops,avgloops,selections[id],s)
        print loops,'loops,',avgloops,'avgloops,',tadj,'tadj'
        print '%10s: %5d, avg %5d' % (selections[id].__name__, tm-tadj, tav-tadj)
                             
    except:
        print 'Usage: python endsvslice.py id nloops adjust avgloops'
        print '  where id is:'
        for i in xrange(len(selections)):
            print '  %d: %s' %(i,selections[i].__name__)
=======================================================================



From nospam at bigfoot.com  Wed Feb 20 15:01:40 2002
From: nospam at bigfoot.com (Gillou)
Date: Wed, 20 Feb 2002 21:01:40 +0100
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com>
Message-ID: <a50uti$ckg$1@norfair.nerim.net>

Python for Windows has no readline support.
You need to use python for cygwin (another distro) to have line recall in
interactive mode.

"Fuming Wang" <fumingw at yahoo.com> a ?crit dans le message news:
b0fd9226.0202201104.2c6a220 at posting.google.com...
> Hi,
>
> I am having trouble using the arrow key to recall commands in console
> interactive mode.  I even installed Cygwin and start Python from there
> but still got the same effect.  Is there anything I can do to get
> arrow key working?
>
> Thanks!
>
> Fuming Wang




From threeseas at earthlink.net  Fri Feb  1 06:31:11 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Fri, 01 Feb 2002 11:31:11 GMT
Subject: Autocoding project proposal.
References: <690.793T1095T13354397threeseas@earthlink.net> <un0yxkupb.fsf@synopsys.com> <1726.795T908T14258threeseas@earthlink.net> <uvgdjk7yl.fsf@synopsys.com> <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com>
Message-ID: <886.797T2741T4375803threeseas@earthlink.net>

On 31-Jan-02 13:24:19 David Masterson <dmaster at synopsys.com> wrote:
>>>>>> Timothy Rue writes:

>> #! Python

>I don't know Python (yet), but I think I get the idea.

>> # PK file array - VIC Dictionary
>> # Setting the standard variables to defaults

>For brevity, I'm going to cut out your code and simply comment on it.


12 hour work days suck. Have barely had enought time to read your message
not to mention better responding.

This is not a responce to you message, but only saying I haven't yet and
why.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jimmy at retzlaff.com  Wed Feb 20 17:47:09 2002
From: jimmy at retzlaff.com (Jimmy Retzlaff)
Date: Wed, 20 Feb 2002 14:47:09 -0800
Subject: No space left on device
Message-ID: <A7B5C22F20EE3C4E95744E2F03F1887F55B1@MAYOR.retzlaff.com>

Sue said:
>I have a program that generates a LOT of small files in a 
>Windows2000 subdirectory.  It is my understanding (often faulty) 
>that there is no limit to the number of files such a subdirectory can 
>contain.  However, I am getting an error when I hit about 21,843 
>files.

>  File "C:\Python21\Scripts\Tests\ManyFiles.py", line 14, in ?
>    fp = open(fileName, 'wb')
>IOError: [Errno 28] No space left on device: 'c:\\legacystoret\\~916-
>22850tst.pck'

I just ran your code on my Windows XP machine and it created all 100K
files. It became extremely slow by the end - it started out creating
about 600 files per second and at the end it was down to about 3 files
per second. I'm running Python 2.2 and my partition is formatted with
NTFS. FAT variants may have more trouble with this if that is what your
partition is.

A few years back I had a nightly process that created, processed, and
deleted about 30K files in a directory. It ran on Python 1.5 and later
1.5.2 on an NTFS partition under NT4. I later restructured it to use a
directory tree because directory access can become excruciatingly slow
when you get that many files - I assume filename lookups do a linear
search. If you can come up with a hierarchy for your files, directories
can help immensely. For example, if you are storing files representing
people, you could separate them by the first letter of their last name -
so Giller would go in the "G" directory and Retzlaff in the "R"
directory. If you still have tons of files in each of those directories,
you could further subdivide those - so Giller would go in the "G\I"
directory and so on. Then when you ask the file system for a file, it
doesn't have to search through 100K file names, but something like 200
(26 to find "G" plus 26 to find "I" plus around 100K/26/26 to find the
file, assuming a relatively even distribution). A similar approach
helped my process' performance significantly. And it might help you get
around whatever problem you're running up against.

Jimmy



From chris.gonnerman at newcenturycomputers.net  Mon Feb  4 08:25:00 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 4 Feb 2002 07:25:00 -0600
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <u0qs5u81eflliq3pij3s2lpuje9irps4mq@4ax.com>
Message-ID: <008d01c1ad7f$6ee77940$0101010a@local>

----- Original Message ----- 
From: "Steve Horne" <steve at lurking.demon.co.uk>

[[ hack hack hack ]]

> One major advantage is the possibility that source code could be more
> concerned with semantics, and the presentation could be an issue for
> the editor. No more arguments about indentation against braces - tell
> the dissenters to simply configure their editor to work with braces
> instead of (or as well as) indentation. With a sufficiently
> sophisticated editor, the syntax of the language becomes 'skinnable'.

Heaven help the poor fool who has to work at someone else's station or 
on someone else's account...  would be MUCH worse than just a difference
in coding style.





From dan at eevolved.com  Fri Feb  1 14:18:29 2002
From: dan at eevolved.com (Dan P)
Date: Fri, 01 Feb 2002 14:18:29 -0500
Subject: Something strange with 'object'
References: <20020201.114203.1856669179.8940@cantal.lip6.fr>
Message-ID: <l%B68.8082$dz5.671899@weber.videotron.net>

Boris Boutillier wrote:

> hi all,
> 
> I made a python program and had a bug that turn out to be a problem with
> 'object' heritage, here is a tiny script that reproduce the bug :
> Buggous version :
> 
>>class a(object):
>>        def __add__(self,other):
>>                print 'a.add'
>>class b(a):
>>        def __radd__(self,other):
>>                print 'b.radd'
>>x = b()
>>y = b()
>>x.__class__ = a
>>x+y
> b.radd
> 
> This means that a.add hasn't been called for x.
> 
> If you remove the object heritage, the buggous behaviour disappear :
>>class a:
>>       def __add__(self,other):
>>                print 'a.add'
>>class b(a):
>>        def __radd__(self,other):
>>                print 'b.radd'
>>x = b()
>>y = b()
>>x.__class__ = a
>>x+y
> a.add
> 
> which is what you want.
> 
> Is this a real bug ? or did I miss something about use of 'object' ? I
> tried to find documentation in python reference guide but found nothing
> helpfull.
> Someone's got an idea ?

My guess is that it's the __mro__. You changed the instance's class, but 
it's method resolution order (stored in the __mro__ attribute) still has 
class b as its first element.

Tell me what you find out,
Dan

 
> Boris Boutillier



From anon at anon.com  Tue Feb 12 20:29:37 2002
From: anon at anon.com (Andrew)
Date: Wed, 13 Feb 2002 01:29:37 GMT
Subject: import images (wxPython)
Message-ID: <5kja8.21380$Cg5.1177018@news1.calgary.shaw.ca>

Hello,

I was looking at some of the demo code and notice that they do an 'import
images' in some files to get some predefined images.  Now I'm writing my own
script and would like to make use of those images as well, but I get the
following error...  ImportError: No module named images.
I'm still learning python and am not 100% sure how the import statement
works.  The images file is in the demo folder of the wxPython folder.  I
tried things like import wxPython.demo.images, but that doesn't work either.
Can somebody give me a hand?
Thanks,
Andrew




From syring at email.com  Thu Feb 21 07:22:38 2002
From: syring at email.com (Karl M. Syring)
Date: Thu, 21 Feb 2002 13:22:38 +0100
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de> <7x7kp7krjl.fsf@ruckus.brouhaha.com> <a52ml7$42r9l$1@ID-7529.news.dfncis.de> <7x8z9n6nwl.fsf@ruckus.brouhaha.com>
Message-ID: <a52os1$45k2v$2@ID-7529.news.dfncis.de>

"Paul Rubin" <phr-n2002a at nightsong.com> schrieb
> "Karl M. Syring" <syring at email.com> writes:
> > > > > Not likely.
> > > >
> > > > You really should have a look at high precison libraries (e.g.
> > > > http://www.hipilib.de). They will use rational numbers internally.
> > >
> > > Yes, but who actually uses those libraries, and for what?
> >
> > The author of the Piologie library said somewhere that extended
precision
> > floating point arithmetic is completely useless, but it is fun.  Then
there
> > is the secret message coded into pi ... locate your name in pi
> > (http://pi.nersc.gov)
>
> High precision floating point and exact rational arithmetic have their
> uses, but almost all ordinary numerical computing uses ordinary
> floating point.  It's very unlikely that anyone would use exact
> rational arithmetic for robot motion control.

But you really should to it. I have floating point code that was running for
a long time without a hitch and then the horrible happened. The ugly thing
is, the code often will not crash but produce wrong results that seem to be
plausible. If you don't have any results to check the output, the worst
nightmares might come true.

Karl M. Syring




From mike at nthwave.net  Wed Feb 20 15:16:58 2002
From: mike at nthwave.net (Michael Mell)
Date: Wed, 20 Feb 2002 12:16:58 -0800
Subject: module setting file location
Message-ID: <3C74043A.A28A4CC6@nthwave.net>

I've created a module which needs to read a certain file to initialize
several user-specific variables. Currently the module only looks for the
settings file in the same directory as itself:

 settings = os.path.split(__file__)[0] # get the module's parent
directory
 settings = os.path.join(settings, "SettingsFile.txt")

I want the installation to be as easy as possible for end users while
conforming to good Python practices. In the current arrangement, the
settings file will often be located in the Python/Lib directory.

Where should the settings file be located?

thx
--
mike at nthwave.net
llemekim         YahooIM
415.455.8812     voice
419.735.1167     fax





From zeitlin at seth.lpthe.jussieu.fr  Fri Feb  8 11:12:56 2002
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: Fri, 8 Feb 2002 16:12:56 +0000 (UTC)
Subject: Which GUI toolkit to use?
References: <3c607ffd_3@corp.newsgroups.com>
Message-ID: <slrna67u87.cmp.zeitlin@seth.lpthe.jussieu.fr>

On Tue, 5 Feb 2002 20:07:41 -0500, Joshua Muskovitz <joshm at taconic.net> wrote:
> 1.  Needs to run on Windows platform.
> 2.  Needs to easily support completely custom paint (buttons, scrollers,
> listboxes, etc).
> 3.  Needs to handle simple input (editboxes, onLeftClick events).
> 
> The goal is to produce an app that runs full screen on a Win platform

 This could be achieved by using wxPython with wxUniv/MSW port of wxWindows.
However nobody has built such a beast yet to the best of my knowledge and
wxUniv itself is still in beta (i.e. has a few known bugs - but we're working
on it...)

> Tk seems like it is a pain to use, wxWindows seems like it doesn't make
> painting easy, pythonwin (win32ui) might be the way to go, ...

 Painting is as easy in wxPython/MSW as in win32ui: basicly none of them
provides a way to have completely custom-painted controls (they both use the
standard Windows controls and they can only be customized in a very limited
sense of the word). wxPython/MSW is able to do anything win32ui does so if
this is enough for your purposes, try it out. If not, see above: wxUniv paints
all controls itself so you may have whatever look you need. If you need more
information about this, please ask on comp.soft-sys.wxwindows as this is
rather OT here.

 Regards,
VZ
-- 
GCS/GM d? H+ s++:-- p2 au--- a- w+ v C+++ UBLS+++ P- L++ N++ E--- W++++ M? V--
-po+ R++ G`` !tv b+++ D--- e++++ u++ h--- f+ r++ n- y?


From ronjeffries at REMOVEacm.org  Wed Feb 27 22:45:27 2002
From: ronjeffries at REMOVEacm.org (Ron Jeffries)
Date: Wed, 27 Feb 2002 22:45:27 -0500
Subject: please do not spam other groups
Message-ID: <C572010FEA1E1B0A.DCEDC0ADFA5CC4BB.6423D10D67C165F6@lp.airnews.net>

Ron Stephens has been spamming other newsgroups with glorious statements about
Python. Python is cool, but spam is spam. Please stop.

Thanks,

Ronald E Jeffries
http://www.XProgramming.com
http://www.objectmentor.com
I'm giving the best advice I have. You get to decide whether it's true for you.


From magnus at thinkware.se  Sat Feb 23 18:16:49 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Sun, 24 Feb 2002 00:16:49 +0100
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org>
Message-ID: <3C7822E1.508@thinkware.se>

Tripp Scott wrote:

> Suppose I want to make a function that modifies its argument:
> 
>  a = -1
>  make_abs(a)
>  print a # = 1
> 
> What is the Pythonic way of doing this? Wrap it in a list?
> 
>  make_abs([a])


In the above case, a in an integer. An integer is
immutable. You can't change an immutable variable,
that's the whole point with immutable variables.

Thus make_abs([a]) is meaningless. Sure, you can
change the content of the list, but that will be lost
since you don't have any name that references that list.
The changed list will be garbage collected as soon as
make_abs() ends.

Let's think of what happens technically. (I have never
bothered to look at the Python internals, but I think
I'm right in principle...) When you do 'a = -1' the
following things will happen: The integer value -1 will
be stored at a location in memory that we call X. The
variable name 'a' will be associated with another
memory location Y. At Y there will be a pointer to X.
Python will bump up a reference counter for X to 1.

If you do 'make_abs([a])' a list will be created at
another location in memory and somewhere there we will
have another pointer to X. The ref-count for X will be
bumped up to 2 since there are two references to the
value at X (-1): The variable a, and the unnamed list.

Now you execute your function:

def make_abs(l):
     l[0] = abs(l[0])

What happens here is that the abs-function is given a
pointer to X. It can then calculate abs(-1) which is
1. It will then place the value 1 at some free location
in memory which we call Z. Now it can return a pointer to Z.
In the assignment 'l[0] = abs(l[0])', the first element
in our list will have it's content changed. Instead of
a pointer to X, there will now be a pointer to Z here.

So, the ref-count for X is decreased to 1 again, and the
ref-count to Z is increased to 1. When the function
make_abs terminates there will not be anything that
references our list any longer. The ref-count for the
list will go to zero, and the memory used by the list can
now be used for other stuff. Naturally, Python must also
decrese the ref-count for all (one in this case) elements
pointed out by the list, so Z will get its ref-count
decreased to 0, and it's "lost in space"...

The variable named 'a' never had anything to do with the
interior of make_abs...

You could make 'a' point to a mutable variable, such as
a list.

def make_abs(l):
     return map(abs,l)

a = [-1]
make_abs(a)
print a[0] # 1

But that's not very convenient. Or you can make a wrapper
class as Erik suggested, but both these solutions are fairly
contrieved, and not pythonic.

a = abs(a) is the Pythonic way to do it as others have said.
That way it's obvious that 'a' changes (since it's assigned to)
and you can use the vanilla abs-method. You are explicit and
you use the included batteries. Good!

But you need to be aware of the difference between mutable
and immutable variables on function calls. As long as you
use immutables, you can be sure that nothing happens to
your local variables in functions you call. If you use mutable
variables, you need to hand over copies in your function
calls to be sure of their integrity (unless you actually
know what happens in the functions you call... ;-)

The main reason why there is no need for out-parameters in
Python, as opposed to "out" or "&" in other languages is
that we have the lovely tuple type in Python.

In languages like C or Pascal, you wouldn't need out-parameters
if you always only wanted one variable back from the function,
but if you want several values back, you'd have to make a struct
or record to contain your values, and that is a bit clumsy.

In python you simply do

 >>> import math
 >>> m, e = math.frexp(15)
 >>> m
0.9375
 >>> e
4

Without this possibility you'd need to do either something like

nutty_math.frexp(15, m:out, e:out) # No, you don't do that in python

or

x = silly_math.frexp(15)
m = x.mantissa
e = x.exponent

So, keep it simple. <output> = function(<input>) is Pythonic.

 >>> def splitEmailAddress(email):
... 	assert type(email) == type('')
... 	assert email.count('@') == 1
... 	ix = email.index('@')
... 	assert 0 < ix < len(email) -1
... 	name = email[:ix]
... 	domain = email[ix+1:]
... 	return name, domain
...
 >>> user, host = splitEmailAddress('magnus at goblin')
 >>> print user
magnus
 >>> print host
goblin

Just compare

user, host = splitEmailAddress(email)

splitEmailAddress(email, &user, &host);

splitEmailAddress(email, out user, out host);

userStruct = splitEmailAddress(email);
user = userStruct.name
host = userStruct.domain

I think the Python way is simpler and clearer than the alternatives.

On a related note, I actually did something similar to a=[-1] once.
I.e. I used a list when I always only wanted one object stored. And
the list was really just to "trick Python". The reason was that I
wanted to supply different filter functions to various sub classes,
and a method in a super class used the method. My first attempt was
something like this

class Super:
     ...
     def doSomething(self):
         ...
         ch = self.modifier(ch)

class Sub1(Super):
     modifier = string.upper

class Sub2(Super):
     modifier = string.lower

This won't work!

If I do:
x = Sub1()
x.doSomething()

self.modifier(ch) will be translated into Sub1.modifier(self, ch)
=> string.upper(self, ch) and I will get

Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
   File "<interactive input>", line 4, in doSomething
TypeError: string.upper() takes exactly 1 argument (2 given)

My next attempt was to be more clear about this not being a
method, but just an attribute I have in the sub class... :-)

     def doSomething(self):
         ...
         ch = self.__class__.modifier(ch)

That won't work either:

Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
   File "<interactive input>", line 6, in doSomething
TypeError: unbound method string.upper() must be called with instance as 
first argument

Python insists that if the class has a variable that is a function,
it should be treated as a method... Quite logical, but not what I
wanted right now...

So I evaded this by putting the function in a list.

class Super:
     ...
     def doSomething(self):
         ...
         for function in self.modifiers:
             ch = function(ch)

import string
class Sub1(Super):
     modifiers = [string.upper]

class Sub2(Super):
     modifiers = [string.lower]

I never intended to supply more than one modifier function,
but now I can...and Python happily accepts that these are
plain functions, not class methods.

Of course I could have done:

class Sub1(Super):
     def modifier(self, ch):
         return string.upper(ch)

Would that have been more pythonic? (I guess I liked the idea of
just dumping in those simple filter functions as plain attributes. :)
And it feels like a waste with methods that don't use 'self'... ;-)




From logiplexsoftware at earthlink.net  Wed Feb 20 19:08:13 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 20 Feb 2002 16:08:13 -0800
Subject: WXWINDOW: CANCEL -button and validation
In-Reply-To: <3C740D43.E925C8F@vip.fi>
References: <3C740D43.E925C8F@vip.fi>
Message-ID: <20020220160813.4787e65f.logiplexsoftware@earthlink.net>

On Wed, 20 Feb 2002 22:55:31 +0200
Pekka Niiranen wrote:

> Hi,
> 
> In code below:
> 
> -  when user enters empty values and presses OK -button, validator
>     warns about the empty values.
> 
> -  when user enters empty values and presses CANCEL -button, validator
>     is not run and I must add "useless checks for empty values" in code
> + default values

I don't understand why you would want to validate user input when cancel is
clicked.  That's the purpose of cancel - to cancel the user action.  If you
want to set default values if the user clicks cancel, that's a different
story - I would have a separate routine that is run when the dialog is
canceled or closed that sets those values.

dlg =  EditConversionParameters(a, bunch, of, args)
retval = dlg.ShowModal()

if retval == wxCANCEL:
	dlg.set_defaults()

values = dlg.GetValues() # or whatever you're doing to get the values from
the dialog
dlg.Destroy()

I didn't bother trying to make much of your code as I don't know what your
many arguments are supposed to be, among other things.  Also, I'm not sure
what your "useless checks" are supposed to do, since they are in fact,
useless.  They get executed before the dialog is even shown onscreen.  They
have nothing to do with user input.

> 1) How can I run Validate(self, parent) -method when CANCEL -button is
> pressed ?
> 2) How can disable the close button (X -button on top right corner of
> window) from wxDialog ?

IIRC, the close button causes the dialog to return wxID_CANCEL just as if
the user clicked the cancel button, so the same logic would apply.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From nas at python.ca  Wed Feb 20 18:56:23 2002
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 20 Feb 2002 15:56:23 -0800
Subject: why does popen2 silently ignore IOError?
In-Reply-To: <a50vu1$29da$1@nntp6.u.washington.edu>; from donn@u.washington.edu on Wed, Feb 20, 2002 at 08:11:45PM +0000
References: <mailman.1014229773.31846.python-list@python.org> <a50vu1$29da$1@nntp6.u.washington.edu>
Message-ID: <20020220155623.A2359@glacier.arctrix.com>

os.system() and os.popen() are security nightmares because they use the
shell (all kinds of quoting problems).  I don't like popen2 either
because getting the exit status of the command is inconvenient, IMHO.  I
wrote a little module for executing child processes.  You can get it
from here:

    http://arctrix.com/nas/python/child_process.py

Examples:

    >>> p = child_process.execute(("ls", "/"))
    >>> p.readline()
    'bin\n'
    >>> p.readline()
    'boot\n'
    >>> p.close()
    >>>

    >>> p = child_process.execute(("cat",))
    >>> p.write("hello world\n")
    >>> p.stdin.close()
    >>> p.read()
    'hello world\n'
    >>> p.close()
    >>>

    >>> p = execute(("false",))
    >>> p.close()
    256
    >>>

I would like to get something into the standard distribution but I need
to come up with a better module name first.  Any suggestions?

  Neil



From nce0485 at flashmail.com  Tue Feb  5 23:47:48 2002
From: nce0485 at flashmail.com (noel)
Date: 5 Feb 2002 20:47:48 -0800
Subject: SOAP (Air Fare Checker) - Problems building Velocigen's getFare request
Message-ID: <7316200b.0202052047.7584b71@posting.google.com>

I am having problems building the correct request for getFare for the
following wsdl: http://services.xmltoday.com/vx_engine/wsdl.vep?flight.wsdl.

I can see that I am (at least) receiving the getFareResponse by
setting dumpSoapIn, but it tells me that I am missing all the
parameters of the request.  I am using SOAPpy097.  The following was
taken off the xmethods site (www.xmethods.com):

<namesp13:getFare xmlns:namesp13="urn:vgx-flight">
  <search>

   <depart>
     <depart_airportcode>SFO</depart_airportcode>
     <arrive_airportcode>SAN</arrive_airportcode>
     <depart_year>2001</depart_year>
     <depart_month>Mar</depart_month>
     <depart_day>17</depart_day>
   </depart>
   <return>
     <return_year>2001</return_year>
     <return_month>Mar</return_month>
     <return_day>27</return_day>
   </return>    

  </search>
 </namesp13:getFare>


Anyone have any experience with the SOAP Air Fare Checker? 
Specifically the python code to build the request?

Soapy and Fareless...
noel


From logiplexsoftware at earthlink.net  Fri Feb 22 12:58:41 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 22 Feb 2002 09:58:41 -0800
Subject: Newbie: Crazy, but Quick
In-Reply-To: <19650FF08A99D511B19E00B0D0F06323060AE1@SRM201002>
References: <19650FF08A99D511B19E00B0D0F06323060AE1@SRM201002>
Message-ID: <20020222095841.77d30a41.logiplexsoftware@earthlink.net>

On Fri, 22 Feb 2002 11:34:23 -0600
Gustavo Cordova wrote:

> Has a python wrapper for SQLite been written yet?
> 
> SQLite, for those who haven't checked it out, is a
> C library (or dll) which implements a SQL data store,
> complete with foreign keys, indexes, etc etc.
> 
> I think it's modeled after the PostgreSQL engine,
> bit I could be very mistaken. The author has done
> a few comparisons with PostgreSQL, and says that
> it's about 4 times faster.
> 
> Also, it stores the whole database in a single file,
> which is mighty convenient.

I'll have to check that out.

> There's the hassle of a db server. I KNOW, to us geeky
> people setting up a MySQL, an mSQL (ewww), or a pgsql
> engine isn't a hassle at all, and it pays back quite
> nicely by separating data storage from the application.
> 
> But to a non-developer it may be more of a hassle than
> to us.

Actually, MySQL on Win32 is very easy these days.  I have a friend who,
while not a total loss on computers, is far from an expert.  I pointed him
to the MySQL website and he managed to get MySQL running on his Win2000 PC
without any problem.  He had far more problems with Access (which he uses
as a front-end to MySQL) than he did with MySQL itself.

Probably the most difficult part of setting up a SQL server is getting
user-permissions correct.  It's very tempting for a beginner to just give
admin access to everyone (or have everyone use the admin account) to avoid
access problems.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From david.abrahams at rcn.com  Sat Feb 23 09:25:41 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Sat, 23 Feb 2002 09:25:41 -0500
Subject: Adding static typing to Python
References: <slrna7b3dk.2on.quinn@regurgitate.ugcs.caltech.edu> <a55qob$ebu$1@thorium.cix.co.uk> <slrna7dhbm.5et.quinn@regurgitate.ugcs.caltech.edu> <u7f8mhek950ec6@news.supernews.com>
Message-ID: <a588iq$mu2$1@bob.news.rcn.net>

"John Roth" <johnroth at ameritech.net> wrote in message
news:u7f8mhek950ec6 at news.supernews.com...

> I've never understood "message" as terminology either. I've
> always made the totally unsupported assumption that way back
> in the beginning, someone thought of each object instance as being
> a separate thread, so that message actually meant something other
> than 'function call with a misleading name.'

Bingo.

I was privileged to hear a talk by Niklaus Wirth, the inventor of the
original OO language, SIMULA, at OOPSLA 2001. In SIMULA, every object /is/ a
separate thread. Incidentally, he thinks that modern OO languages are all
missing something important by dropping this particular idea.

-Dave





From jeff at ccvcorp.com  Thu Feb 14 13:36:04 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 14 Feb 2002 10:36:04 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org> <slrna6kkpj.is0.tim@vegeta.ath.cx> <pu6e4a.5ti.ln@wombat.san-francisco.ca.us> <slrna6m7jd.o2v.tim@vegeta.ath.cx>
Message-ID: <3C6C0393.F7C16D06@ccvcorp.com>


Tim Hammerquist wrote:

> OTOH, exceptions in Spanish are much less frequent than in English or
> French.  Pronunciation is nearly 100% predictable, and there are more
> regular verbs than most languages.
>
> (I have no knowledge of German.)

German has fewer exceptions than English.  There are a fair number of
irregular verbs, but many of those follow similar patterns.
Pronunciation is also extremely predictable, unlike English.  (English
and German are my only two languages (well, human languages, as opposed
to computer languages), so I can't well compare German with
Spanish/French....)

Jeff Shannon
Technician/Programmer
Credit International




From store_li at sina.com  Thu Feb 28 11:23:01 2002
From: store_li at sina.com (Ricky)
Date: Fri, 1 Mar 2002 00:23:01 +0800
Subject: Can use python to play MP3 file?
Message-ID: <a5llkf$8a75f$1@ID-12869.news.dfncis.de>

I hear that PHP can online play MP3 file. Is there any method that can make
me play MP3 by using Python CGI?

Thanks for any reply!




From aleax at aleax.it  Mon Feb 18 02:33:02 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 07:33:02 GMT
Subject: I Don't Know A Thing...Compiling?
References: <3C709746.AD83377F@mchsi.com>
Message-ID: <O62c8.81993$3J1.2613014@news2.tin.it>

3[ED] (Thread) wrote:
        ...
> Python. So I was just wondering, can you compile a Python script into
> ane executable, or do you need an interpreter. I don't know if this is

Yes to both: you can pack the interpreter (which you do need) together
with the rest of your Python application into one (largish) executable.
http://www.mcmillan-inc.com/install1.html is a cross-platform solution,
http://starship.python.net/crew/theller/py2exe/ a Windows-specific one.


Alex





From bokr at oz.net  Mon Feb  4 05:39:06 2002
From: bokr at oz.net (Bengt Richter)
Date: 4 Feb 2002 10:39:06 GMT
Subject: Equivalent of Perl chomp?
References: <mailman.1012513005.17840.python-list@python.org> <%ok68.88316$8b.3281206@atlpnn01.usenetserver.com> <a3etov$hgu$0@216.39.172.122> <3C5B0EAA.F4527F5E@ccvcorp.com>
Message-ID: <a3loca$41k$0@216.39.172.122>

On Fri, 01 Feb 2002 13:54:51 -0800, Jeff Shannon <jeff at ccvcorp.com> wrote:

>
>
>Bengt Richter wrote:
>
>
>> I think slicing is probably a winner because endswith subsumes more
>> complex slicing options that have to be tested for, and the argument can be
>> multicharacter. I would be surprised if single-character string handling
>> wasn't special-cased a lot for speed, benefiting s[-1:] == '\n'.
>
>Well, one of the things that Tim Peters hinted at is more likely to be the main reason
>that endswith is slower...  every iteration, the endswith version has to resolve a name.
Oops, yup *<8^p

>If you changed the loop to something more like:
>
>func = s.endswith
>for n in xrange(iterations):
>    func('\n')
>
>you would likely see a notable difference in the results.   :)
>
Yup.
>(But I'm too lazy/otherwise busy to actually run the tests myself....)
>
I'm going to continue this in another post. The first trick is
actually timing what you think you're timing ;-)

Regards,
Bengt Richter



From deoradh at yahoo.com  Sun Feb 17 23:56:39 2002
From: deoradh at yahoo.com (Keith Farmer)
Date: 17 Feb 2002 20:56:39 -0800
Subject: PyAppBar/wxAppBar opened at SourceForge
Message-ID: <adec2554.0202172056.62b0f735@posting.google.com>

https://sourceforge.net/projects/pyappbar/

I decided to toss it out upon the unsuspecting masses.

There are currently two main issues which I'm stuck on:
* Resizing while docked doesn't work. I think there's a conflict with other
messages kidnapping the mouse.
* Moving from a docked to an undocked state -- it works, kinda. If you drag
the mouse too quickly, it'll lose its grip. If you release the mouse button,
and grab the (now undocked) window again, it's fine.

I'd certainly appreciate some help in finishing this up, if anybody's
interested.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org


From hamish_lawson at yahoo.co.uk  Fri Feb 22 13:05:51 2002
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: 22 Feb 2002 10:05:51 -0800
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie> <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk>
Message-ID: <915a998f.0202221005.345c9c73@posting.google.com>

> What do you consider the other 2 methods to be? I can think of 
> encoding it in the URL, and the server remembering the browser's IP 
> address. Are there others?

HTTP authentication can also be used as a means of identifying a session.

Hamish Lawson


From mwh at python.net  Thu Feb 14 11:02:45 2002
From: mwh at python.net (Michael Hudson)
Date: Thu, 14 Feb 2002 16:02:45 GMT
Subject: watching the number of references to an object
References: <1013700953.2880.0.nnrp-01.c1c3e191@news.demon.co.uk>
Message-ID: <uu1skf3qx.fsf@python.net>

"Paul Brian" <paul1brian at yahoo.com> writes:

> Dear all,
> 
> Does anyone know how to monitor the references pointing to an instance of an
> object?
> 
> the gc module (according to docs) seems to show those with no
> references left, but I am interested in seeing how many refs a given
> instance has...any ideas on back of a postcard please

You want gc.get_referrers (new in 2.2, I think), or maybe just
sys.getrefcount.

>>> import gc
>>> len(gc.get_referrers(gc))
2

(the referrers are going to be sys.modules & __main__.__dict__ here).

Cheers,
M.

-- 
  at any rate, I'm satisfied that not only do they know which end of
  the pointy thing to hold, but where to poke it for maximum effect.
                                  -- Eric The Read, asr, on google.com


From l0819m0v0smfm001 at sneakemail.com  Sat Feb  9 09:13:04 2002
From: l0819m0v0smfm001 at sneakemail.com (Joshua Macy)
Date: Sat, 09 Feb 2002 14:13:04 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com>
Message-ID: <3C64F51A.2070208@sneakemail.com>

wooks wrote:

> New to the language, new to OO but I do understand the basic concepts,
> not at all new to programming, trying to do the right thing by
> searching archives before asking questions.



   Have you tried following the Python for Beginners link on 
www.python.org?  http://www.python.org/doc/Newbies.html has links to all 
sorts of material for beginners, including help at python.org (a volunteer 
help desk) and tutor at python.org (a low volume mailing list).
   As for being new to OO, there's no requirement that a Python program 
be organized as an OO program; just like VB, you can write purely 
procedural code (although, again just like VB, you'll have to create and 
perform calls on external OO objects to do most interesting work).

   I don't think searching the discussion archives is all that helpful 
unless you have a specific question that you want answered and if you 
do, you haven't mentioned it.

> Yet I feel I need to be
> doing a university degree course in philosophy or mathematics to
> follow almost any discussion. If that's the way it is fair enough but
> Python is supposedly simple to learn (thus far I've had an easier time
> of it learning Perl).



   comp.lang.python is not a forum specifically for beginners; it's 
better than almost any other general language newsgroups that I've 
encountered at answering questions from beginners, even if the questions 
are covered in the FAQ or easily found in the archives, but most of the 
posters here are here to get their own practical questions answered, or 
to discuss extensions and enhancements that they're working on--which 
can be some very hairy stuff indeed.  If you dip into a random 
discussion and it sounds like it's between two PhDs in computer science, 
it's probably because it is, and they're using Python in their work.

> If someone comes and says I've had the following
> understanding of concept X since I was in grade 2 why is it not so in
> Python, the answers seem to be heavy on the philosophical
> justifications fo the Python way and light on the practical benefits.
> The answers also tend to be in a totally different intellectual
> context.
> 



   Posters on comp.lang.python tend to assume that if you ask the 
questions "Why is it thus-and-so in Python?" that you actually care 
about the answer, and that if you actually meant "How do I do 
thus-and-so in Python (I tried X, like in some other language, but it 
didn't work)" that you'd say that instead.  They're often wrong about 
that, but so it goes.
   As for concepts "understood since grade 2", that usually means "it 
worked like X in this other language Y that I know, why doesn't Python 
follow this 'natural' approach?"  If you hang around a while, though, 
you'll start to notice what the veterans of the newsgroup realized a 
long time ago, which is that the 'naturalness' of X depends entirely on 
which language Y the person is thinking of.  People will frequently 
complain something like "Python is the only language that does this 
stupid thing Z! All other languages do A!"; they are just as frequently 
shot down by posters pointing out that languages B, C, D, and F do 
things just the way Python does, E, G, and L do it the way the 
complainer is thinking about, and H, I, J do something else entirely.
   Python does very few things that no other language does, and those 
few are usually (e.g. colons at the end of statements introducing 
blocks) because there was as study that showed it was easier to learn 
that way.


> True I haven't been around here that long but it's a 1st impression
> and it counts. Perhaps it's because so many come to Python with
> baggage from another language so that issues they raise are addressed
> in that context rather than conceptually.
> 


   I think that's exactly it.  If you really want to know why something 
in Python is the way it is conceptually, you often have to be prepared 
for a somewhat complicated answer, philosophically or mathematically. 
Sometimes, though, the answer is because Guido wanted it that way.

One of the basic philosophies of Python design is that it is wrong for 
the language to guess when presented with something that's ambiguous. 
That something is "automagical" is usually a swear word in Python.  This 
can make Python seem hairier than other languages which cheerfully take 
a stab at what the designers think you may have wanted, but the Python 
belief is that those languages are leading you down the garden path, and 
sooner rather than later you'll be bitten by the subtlety that the other 
language tried to gloss over.

A good example of this is floating point math and its representation. A 
frequent question is something like:

 >>> 1 - .1
0.90000000000000002

What's up with that?  1 - .1 = .9, right?  Right, but that isn't the way 
that any computer you're likely to use represents it internally, and if 
you don't know that you're eventually going to come to grief if you try 
to use mathematics at all seriously in your programs.  I won't go into 
why this is so (there's a whole FAQ page on it), but this is the sort of 
thing that is a problem with computers in general (or at least binary 
floating point math) and one of the places where Python's approach can 
be surprising to a newbie.


> Bought Python for Win32 book and already been caught out (Part 1 was a
> breeze part 2 and I shuddered to a grinding halt). For perspective I
> am a tester who wants to learn Python because of it's excellent unit
> test framework and it's scripting glue like abilities (not interested
> in building applications).



Did you actually try any of the examples of manipulating Word or Excel 
from Python?   Python for Win32 is an excellent book, IMO, but way 
overkill unless you have some specific need to interact with COM or the 
Win32 OS APIs. It is not a book about learning Python.


> 
> The benefits of using of Python as a scripting glue like language are
> widely touted but don't seem (I am probably wrong but it's a 1st
> impression) to be well catered for. As I said I am not interested in
> building applications. Is there a forum for the dumber less demanding
> Python user?
> 


   You need to ask a more specific question, or be more specific about 
what exactly you're trying to accomplish and finding difficult.  General 
complaints that it seems difficult to understand aren't likely to 
generate much helpful advice.

   And, don't take this the wrong way, but it *is* dumb to go to a 
language newsgroup and say "Maybe I'm dumb but [vague complaints about 
the language not being easy]" if you have any expectation that posters 
will rush in saying "No, no, you're not dumb, it's obviously the 
language's fault" the way your friends and relatives might.  Usenet is 
not the place to cruise to have your ego soothed.

   Joshua





From logstx at bellatlantic.net  Mon Feb 18 20:09:35 2002
From: logstx at bellatlantic.net (logistix)
Date: Tue, 19 Feb 2002 01:09:35 GMT
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <jBhc8.6565$_45.6330@nwrddc01.gnilink.net>

Do I understand you correctly? You're worried that a dynamic language won't
throw a compiler error if you change a method name?

That's what grep and "Find in Files..." is for. ;)

Seriously though, from a testing perspective you could have a test function
like legacyMethods() that actually scans the source code old method names,
instead of trying to climb through the codebase until an error shows up.
Lint (for C, not python) is an example of a program that tests for logic
bugs at the source code level.

You could also do stuff along the lines of

def oldMethod(self):
    raise RuntimeError, "oldMethod() no longer supported, use newMethod()"

def newMethod(self):
    pass

This will at least force your code to break in a predictable way, instead of
introducting vague and unpredictable errors.


--
-

"Alexander Jerusalem" <ajeru at vknn.org> wrote in message
news:24c39b2c.0202181531.187fad4c at posting.google.com...
> Has anyone heard of plans to add static type checking to Python? I
> like Python very much but I don't use it for bigger projects because I
> fear the testing effort. For example I just refactored a rather large
> Java codebase. And when I change a method name for example, all I have
> to do is run everything thru the compiler and it will show me all
> places in other classes where this method is called, so I can go there
> and chenge the method call as well.
>
> In Python I'd have to test the whole software suite including each and
> every branch in each and every if statement to be sure that I didn't
> forget something. To build such a test suite would take almost as long
> as it takes to build the whole software and I'd have to maintain the
> test suite as well. That's just unafordable.
>
> On the other hand, Python is the best language that currently exists
> and I would like to use it on bigger projects not just for prototyping
> as some suggest.
>
> Regards,
> Alexander Jerusalem




From phr-n2002a at nightsong.com  Thu Feb 21 21:35:21 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 18:35:21 -0800
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie>
Message-ID: <7xheoafdli.fsf@ruckus.brouhaha.com>

Roy Madden <roy at linksheaven.com> writes:
> I'm going to try and dig up some stats for you, but from experience and
> research the population of those who do not accept cookies is very very
> small (albeit tend to be technically proficient and vocal online). I once
> worked for an online bank with operations in Germany, and we considered
> this issue *very* carefully before choosing cookies for our user
> interface.
> 
> A slightly larger population do not 'like' them, but of the 3 methods of
> maintaining session state they are the most secure option (waiting for the
> flames to start :) ) - if security is an issue for you, and the 'like'
> issue is a matter of user education.

I think a lot of people who don't like cookies mainly don't like
persistent cookies.  As long as your cookies are used only for the
current session, most people don't mind them as much.  Unfortunately,
browser interfaces for dealing with different kinds of cookies are
still clumsy, but MSIE has a setting to reject all persistent cookies
while accepting temporary ones.  You could suggest people use that
setting if they want to have web sessions but don't want to be tracked
over long periods.


From Python at lantech.com  Wed Feb 27 17:11:04 2002
From: Python at lantech.com (Python)
Date: Wed, 27 Feb 2002 17:11:04 -0500
Subject: Thank you Python!!!!!
Message-ID: <BC74F8F7E87DD511990C000103E2FEFA2F36@XCHG>

A big thanks to Mark Hammond/Andy Robinson (for "Python programming on
Win32), Guido van Rossum, and all those who help maintain Python.

I was able to program an script that walks through a directory and appends
many thousands of faxes together.  Lowest outside bid:  $10,000.   I did it
for the price of Mark's book, an activeX dll that cost $180 (PIL might have
worked, but the documentation was lacking - at least for my level of
experience), and a few hours (ok, maybe a few + a few:).  My previous
scripting/programming experience:  WinNT login scripts (i.e. almost none).
Ideas for future use of Python:  Many!!!



Christopher Encapera
Lantech.Com
chrise at lantech.com
502-267-4200




From peter at engcorp.com  Wed Feb 27 20:12:17 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Feb 2002 20:12:17 -0500
Subject: Time to rename Stackless?
References: <mailman.1014800988.1436.python-list@python.org>
Message-ID: <3C7D83F1.3EBD4DC3@engcorp.com>

Tim Peters wrote:
> 
> [Christian]
> > Not bad!
> > Should I start a contest?
> 
> I vote for "Christian Python".  This would give Python a huge advantage in
> the all-imporant American Bible Belt market <wink>.
> 
> suffer-the-tasklets-to-come-unto-me-ly y'rs  - tim

I thought this was the reason for the move of Python Labs
to Virginia.  Not, as is commonly thought, to be closer to
PSU headquarters in


From eric.brunel at pragmadev.com  Mon Feb  4 05:42:23 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Mon, 4 Feb 2002 11:42:23 +0100
Subject: just another Tkinter installation problem
References: <d013da35.0201290628.5f6ef54d@posting.google.com> <a36chk$v38$1@norfair.nerim.net> <3C578ECD.8D88B430@master.sm.go.dlr.de>
Message-ID: <a3loc0$6du$1@norfair.nerim.net>

Hi Heiko,

Sorry for the delay...
> > Can you please include the lines concerning Tkinter in your Setup file?
It
> > seems to me that the problem may be here.
> OK, here they are:
[snip]
> And the tcl/tk headers and libs _are_ at the places given above...

It seems the problem is not here after all. Did you try to explicitely
delete the lippython2.2.a before re-running "make"? I often had to do that
when I was using a Solaris box. Maybe the new make does not delete the old
version of the library and is looking for the new symbol in it.
HTH
 - eric -




From nobody at nobody.com  Mon Feb  4 17:54:07 2002
From: nobody at nobody.com (John)
Date: Mon, 4 Feb 2002 16:54:07 -0600
Subject: random generator
Message-ID: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu>

Hello,

Is there any random number generator in Python, that allows users to specify
"skewness"? Thanks!




From logiplexsoftware at earthlink.net  Fri Feb  1 14:35:07 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 1 Feb 2002 11:35:07 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <cb4ba455.0202010543.7d4edd1a@posting.google.com>
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<cb4ba455.0202010543.7d4edd1a@posting.google.com>
Message-ID: <20020201113507.797224f3.logiplexsoftware@earthlink.net>

On 1 Feb 2002 05:43:20 -0800
Bill Tate wrote:

> Cliff Wells <logiplexsoftware at earthlink.net> wrote in message > 
> > However, your last statement that managers make _all_ the decisions was
an
> > unfortunate one.  It's better if the technicians make the decisions and
> > then managers _approve_ them.  This may be splitting hairs, but when
it...

> I find little to disagree with what you're saying, but managers also
> need to know that their development team is being unbiased and
> objective in their analysis :>).

I think this comes from establishing a good working relationship between
management and developers.  Respecting the developer's opinions is the only
way to establish this relationship.  Not only that, but allowing developers
to take this responsibility will soon reveal those who are not reliable in
this area (and by extension, other areas as well).  There will be setbacks
as these people reveal themselves, but weeding these people out can only
have a positive effect on the overall endeavor.
Obviously this will take time, and the flip side is that developers who
haven't yet established this trust need to be patient.  However if
management isn't willing to delegate some responsibility in this area that
trust can never be established.  


-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From marklists at mceahern.com  Mon Feb 11 00:51:34 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 21:51:34 -0800
Subject: FixedPoint
In-Reply-To: <HFEKILOLEFEFMKAECNDLKEKEDBAA.jason@jorendorff.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEBKKCAA.marklists@mceahern.com>

[Jason Orendorff]
> No, instead of "self, float(other)" it should be "float(self), other".

That makes eminent sense--and it works!  Now I feel dumb.  :)  But I embrace
these moments.

Thanks!

// mark




From michels at linmpi.mpg.de  Thu Feb  7 10:03:07 2002
From: michels at linmpi.mpg.de (Helmut Michels)
Date: Thu, 07 Feb 2002 16:03:07 +0100
Subject: [ANN] Data Plotting Library DISLIN 7.6
Message-ID: <3C62972A.17A14540@linmpi.mpg.de>

I am pleased to announce version 7.6 of the data plotting software
DISLIN.

DISLIN is a high-level and easy to use plotting library for
displaying data as curves, bar graphs, pie charts, 3D-colour plots,
surfaces, contours and maps. Several output formats are supported
such as X11, VGA, PostScript, PDF, CGM, WMF, HPGL, TIFF, PNG and BMP.

The software is available for several C, Fortran 77 and Fortran 90
compilers. Plotting extensions for the interpreting languages Perl,
Python and Java are also supported for the most operating systems.

DISLIN distributions and manuals are available from the DISLIN Home Page

     http://www.dislin.de

and via FTP from the server

     ftp://ftp.gwdg.de/pub/grafik/dislin

 -------------------
  Helmut Michels
  Max-Planck-Institut fuer Aeronomie  Phone: +49 5556 979-334
  Max-Planck-Str. 2                   Fax  : +49 5556 979-240
  D-37191 Katlenburg-Lindau           Mail : michels at linmpi.mpg.de



From donnal at donnal.net  Sun Feb 24 22:02:42 2002
From: donnal at donnal.net (Donnal Walter)
Date: 24 Feb 2002 19:02:42 -0800
Subject: functional programming with map()
Message-ID: <918bc22f.0202241902.2160b75a@posting.google.com>

I know that

map(f,items)

is equivalent to:

for x in items:
    f(x)

But what is the functional equvalent of:

for x in items:
    x.f()

Thanks.


From elflord at panix.com  Thu Feb 28 19:22:16 2002
From: elflord at panix.com (Donovan Rebbechi)
Date: Fri, 1 Mar 2002 00:22:16 +0000 (UTC)
Subject: Scripting C++ -- 2 -- a more concrete example.
References: <3C7E6642.2030202@physics.ucsb.edu>
Message-ID: <slrna7tidr.8aq.elflord@panix2.panix.com>

In article <3C7E6642.2030202 at physics.ucsb.edu>, Craig Maloney wrote:
 
> If I use wrapper generators, it seems like there is a chance (e.g. using 
> SILOON/PDT or BOOST/PDT) that it would be possible to have the wrapper 
> code automatically generated.

Most of it. Using sip, I generate about 4 times as much code as I write, so
the ratio is pretty good. There are things sip doesn't do like mapping python
and C++ exceptions that I need to take care of manually.

> 2) Marshalling and Objects-by-value.
> When the "Crystal" computes its energy it should *NOT* have to marshall 
> calls to get at the atom data structures.  This is an operation that 
> will be repeated about 2.85 bijillllion times.  In CORBA, this is no 
> problem -- I would either have Crystal::addAtom take a valuetype (atoms 
> are easy enough to serialize), or I would use some "co-location" trick 
> so that the ORB knows that the atom *really* lives in the same address 
> space as the crystal.  Preferrably the former.
> 
> If using a wrapper generator (e.g. SILOON) I don't see a way around the 
> marshalling of the call to get at the atom data structure.

I don't really understand exactly what the problem is. 
> 
> Of course, this would be taken care of by a revamp of the design of the 
> class library... maybe making the crystal an atom factory that produces 
> new atoms in its own address space.  But I would like to avoid thinking 

Now I'm really confused. If you're not using CORBA, the atoms are in the
same address space, aren't they ?

-- 
Donovan


From mdk at mdk.com  Thu Feb 14 13:28:15 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 13:28:15 -0500
Subject: tcl question
References: <a4c2ge$1dognu$1@ID-98166.news.dfncis.de>
Message-ID: <a4gvk0$7ug1$1@ID-98166.news.dfncis.de>

"MDK" <mdk at mdk.com> wrote in message
news:a4c2ge$1dognu$1 at ID-98166.news.dfncis.de...
> In PythonWin I entered this script from a book:
>
> from Tkinter import Label
> widget = Label(None, text='Hello GUI world!')
> widget.pack()
> widget.mainloop()
>
> When I run it in PythonWin it runs fine.
>
> When I double click on it in Windows explorer I get a message that
tcl80.dll
> can not be found.  That file does not exist on my PC.  How then is it
> running under PythonWin if that file does not exists and how can I make it
> run when double-clicked?
>
> My path includes:
>
> PATH=C:\Python22\;c:\python22\dlls\;C:\Python22\Lib;C:\Python22\Lib\lib-tk
>
> Thanks.
>
>

Well, I just deleted everything Python related and started from scratch.
Now it works.  :-)




From lac at strakt.com  Wed Feb  6 13:12:20 2002
From: lac at strakt.com (Laura Creighton)
Date: Wed, 06 Feb 2002 19:12:20 +0100
Subject: What is the command to determine an object's type? 
In-Reply-To: Message from "Mark McEahern" <marklists@mceahern.com> 
   of "Tue, 05 Feb 2002 13:52:14 PST." <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com> 
References: <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com> 
Message-ID: <200202061812.g16ICKGo016684@ratthing-b246.strakt.com>

I think that everybody should read Design Patterns because it is the
single best way I know of to improve your code.  (And your life.
Life is more fun when your code is more maintainable.)  However,
for a Python programmer, there is one problem with the original
text:  Design Patterns.

It is written for C++.

This means that while a lot of the text is about the really good stuff,
the patterns, at least half of it could be entitled 'my struggles with
the C++ type system'.  And since you came to this discussion because
you wanted to test the type of something, which is something that
Mark McEahern and I both think that you probably ought not to be doing
at all, you may get the wrong lesson from reading that book now.  

There is another book, The Design Patterns Smalltalk Companion,
(Alpert Brown Woolf), which is meant to be read at the same time
as the Design Patterns.  Smalltalk does not have declared types
either.  As a result, patterns in Python end up looking a lot 
more like Patterns in Smalltalk.  I find the Design Patterns
Smalltalk Companion to be a lot easier to understand than its
parent, as well.  But when you get a look at the Smalltalk syntax,
you will see one of the reasons why Smalltalk has not taken over
the world by now, mores the pity.

What Mark McEahrn and I are plugging is called signature based
polymorphism.  (Well, I should only speak for myself.  I am.)

This has been discussed on the python list recently.

See:  (isinstance considered harmful)
http://groups.google.com/groups?q=group:comp.lang.python+insubject:isinstance+insubject:considered&hl=en&as_qdr=m3&selm=83n0z44t11.fsf%40panacea.canonical.org&rnum=1

which became (typetesting, adaptation, typeclasses, ... (was Re: isinstance() considered harmful) )

http://groups.google.com/groups?hl=en&threadm=a2ohsf%24mjj%241%40serv1.iunet.it&rnum=3&prev=/groups%3Fq%3Dsignature%2Bbased%2Bpolymorphism%2B-ruby%2Bgroup:comp.lang.python%26hl%3Den%26as_qdr%3Dm3%26selm%3Da2ohsf%2524mjj%25241%2540serv1.iunet.it%26rnum%3D3

Laura Creighton



From jason at jorendorff.com  Fri Feb  1 08:47:37 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 07:47:37 -0600
Subject: Why it does NOT work on Linux ?
In-Reply-To: <3C5AAA9E.18034.11AB3F0@localhost>
Message-ID: <HFEKILOLEFEFMKAECNDLKEKNCPAA.jason@jorendorff.com>

Ladislav wrote:
> This program works well under windows but on Linux it does not 
> find any number. Can you please help?

Thoughts:

You should be using a raw string for your regular expression.

Compare versions of Python.

## Jason Orendorff    http://www.jorendorff.com/



From sholden at holdenweb.com  Thu Feb  7 10:44:48 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 10:44:48 -0500
Subject: FixedPoint
References: <mailman.1013070672.23308.python-list@python.org>
Message-ID: <Dhx88.95486$KM2.3465594@atlpnn01.usenetserver.com>

"Mark McEahern" <mark at mceahern.com> wrote ...
> In the FixedPoint module found here:
>
>
>
ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoin
> t.py.Z
>
> there seems to be a bug comparing to None like this:
>
>   if f == None:
>
> Demo:
>
>     $ python
>     Python 2.2 (#1, Dec 31 2001, 15:21:18)
>     [GCC 2.95.3-5 (cygwin special)] on cygwin
>     Type "help", "copyright", "credits" or "license" for more information.
>     >>> import FixedPoint
>     >>> f = FixedPoint.FixedPoint(1.0)
>     >>> if f == None:
>     ...     print "that's not true!"
>     ...
>     Traceback (most recent call last):
>       File "<stdin>", line 1, in ?
>       File "FixedPoint.py", line 272, in __cmp__
>         xn, yn, p = _norm(self, other)
>       File "FixedPoint.py", line 415, in _norm
>         y = FixedPoint(y, x.p)
>       File "FixedPoint.py", line 216, in __init__
>         raise TypeError("can't convert to FixedPoint: " + `value`)
>     TypeError: can't convert to FixedPoint: None
>
> This doesn't happen if you use either of the following:
>
> if f is None:
> ...
>
> if not f:
> ...
>
> I ran into this because PyPgSql (v2.0) uses the first method:
>
>     File "/usr/lib/python2.2/site-packages/pyPgSQL/PgSQL.py", line 1434,
in
>   _quote
>
>       if value == None:
>
> I'm not sure what the best fix is here but one that works is to modify the
> __cmp__ method for FixedPoint to anticipate comparison to None.  Here's
the
> current __cmp__:
>
>     def __cmp__(self, other):
>         xn, yn, p = _norm(self, other)
>         return cmp(xn, yn)
>
> The fix would look like this:
>
>     def __cmp__(self, other):
>         ## To avoid crashing on if x == None and the like:
>         if other is None:
>             return 1
>         xn, yn, p = _norm(self, other)
>         return cmp(xn, yn)
>
> Or perhaps it'd be better to muck with __eq__ and __ne__?  Well, anyway,
> that mod to __cmp__ works.
>
> Cheers,
>
Mark:

Since the *intent* for None is that there should only ever be one object of
type None, the better test would be

    if something is None

which can easily be used without requiring any changes to FixedPoint.

Good general rule: if Tim made it that way, it's supposed to be that way :-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From sabren at manifestation.com  Tue Feb 12 17:46:57 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 12 Feb 2002 17:46:57 -0500 (EST)
Subject: cgi and mime type
In-Reply-To: <Xns91B3E00B94C16cliechtigmxnet@62.2.16.82>
Message-ID: <Pine.LNX.4.44.0202121743590.23520-100000@mercury.sabren.com>

On 12 Feb 2002, Chris Liechti wrote:

> aseem at neurobs.com (A M) wrote in news:ed6d1c1a.0202121225.295ca950
> @posting.google.com:
> 
> > fp = open('fp.exe')
> > print fp.read();
> 
> use open(name, 'rb') here. binary mode is important when the cgi runs 
> on win32. not sure if this fixes your other problem though...


I missed the start of this thread, but this might help
someone: when printing binary files on stdout on win32, you
also have to make stdout binary:

    if sys.platform=="win32":
        import msvcrt
        msvcrt.setmode(sys.__stdin__.fileno(), os.O_BINARY)
        msvcrt.setmode(sys.__stdout__.fileno(), os.O_BINARY)

And you'd probably also want to say:

sys.stdout.write(fp.read())

instead of "print" in the code above...

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From db3l at fitlinxx.com  Mon Feb 18 13:45:46 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Feb 2002 13:45:46 -0500
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com>
Message-ID: <u4rkezl0l.fsf@ctwd0143.fitlinxx.com>

sameer_ at email.com (sameer) writes:

> sameer_ at email.com (sameer) wrote in message news:<6bd9f01b.0202160255.77244c9e at posting.google.com>...
> > Here's the situation:
> > I have a python package (or two) to be installed on a windows machine
> > (for now) which has python installed.  I am using distutils to create
> > a windows installer which installs my python packages just fine.  Now,
> > I want to also create some shortcuts on the Desktop as well as the
> > Start Menu.  As of now, distutils doesn't allow me to do either of
> > those things.  What are my options here to do either or both of these
> > things.  Can the distutils package be modified easily to provide this
> > kind of support on windows?  Or is there another python package
> > available that would allow me to create the desktop and Start Menu
> > shortcuts?
> 
> it's a pity no one has replied.  I suppose there are no sane options for this.

It might help if you at least waited through a weekend for responses
before declaring failure :-)

Although I do expect there's no trivial solution at the moment -
whether or not that consititutes "sane" is another question.

The binary portion of the windows specific installer is hardcoded into
the distutils package (encoded in the bdist_wininst.py module), and it
doesn't appear that source is available (at least I can't seem to find
it even in the CVS tree) so changing how it behaves doesn't seem easy.
You could replace it outright, but then you'd have to duplicate its
existing functionality as well.

What you might try doing in this case is combining the distutil output
with one of the free Windows installers (like InnoSetup) which do have
plenty of capabilities for handling the additional duties.  The tricky
part of this is knowing the installation location since the distutils
installer is picking that.

You could wrap an InnoSetup install around the existing executable
from distutils (running that as one of the components in a InnoSetup
install), but then you wouldn't necessarily know where to create the
shortcut to.  Perhaps better would be using a simpler output from
distutils (say a simple tar or zip package) to gather the necessary
files, and then incorporate them directly into an InnoSetup install.

It doesn't look like distutils has a simple collect method to gather
the files into a directory tree without using an archive like tar/zip,
but worst case is you unpack the created archive to provide individual
file input to InnoSetup.  You'd need to do some work in InnoSetup to
locate the existing Python installation (if you didn't want the user
to have to manually enter it), but you can access registry keys in the
setup script (or worst case use the InnoSetup Extensions version with
scripting).

Having some additional hooks created in distutils to execute
post-installation sounds like it could be a useful enhancement.  At
least with that sort of scenario you could write your own shortcut
creating code without modifying the core distutils itself.

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


From cliechti at gmx.net  Fri Feb 15 19:46:07 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 16 Feb 2002 01:46:07 +0100
Subject: JAR equivalent in Gordon McMillan's Installer
References: <2upn6uoqotqejiji69bg8kebfacdu9oamb@4ax.com> <$7JeJeA7o$a8EwsQ@jessikat.demon.co.uk> <efao6us1tt3ut2hp6aor19p7qfqad3k4t1@4ax.com> <Xns91B5F04E81A38cliechtigmxnet@62.2.16.82> <fm0r6ugt88ada5ev9lu6fs5drdvrd3dg1u@4ax.com> <mailman.1013812227.6703.python-list@python.org>
Message-ID: <Xns91B712898988Fcliechtigmxnet@62.2.16.82>

"Mike C. Fletcher" <mcfletch at geocities.com> wrote in 
news:mailman.1013812227.6703.python-list at python.org:
> wxImages can be loaded from strings.  See the wxPython Wiki entry on 
> working with images (particularly the sections on PIL and Numeric):
ok just tried this:
------------
import zipfile
from wxPython.wx import *
wxInitAllImageHandlers()    #init wx stuff
#convert file
bmp = wxImage('img.png', wxBITMAP_TYPE_PNG)
raw = bmp.GetData()
zipf = zipfile.ZipFile('hello.zip','w')
info = zipfile.ZipInfo('hello.%d.%d' % (bmp.GetWidth(), bmp.GetHeight
()))
info.compress_type=zipfile.ZIP_DEFLATED
zipf.writestr(info, raw)
del zipf

#use it
zipf = zipfile.ZipFile('hello.zip','r')
entry = zipf.namelist()[0]
name,w,h = entry.split('.')
raw = zipf.read(entry)
img = wxEmptyImage(int(w), int(h))
img.SetData(raw)
#img = img.Rotate(45, (10,10))  #just for fun
img.SaveFile("hello.png", wxBITMAP_TYPE_PNG)
--------

you can use wxPython to convert the images to strings, save them to a 
zip using the zipfile module. the image sizes have to be saved 
separately. i've choosen to stuff them into the filename.

storing uncompressed images in a zip isn't that bad, as e.g png or gif 
also uses some similar compression algos. (in my test the zip was 
smaller than the original png)

-- 
Chris <cliechti at gmx.net>



From mwh at python.net  Fri Feb 22 05:59:57 2002
From: mwh at python.net (Michael Hudson)
Date: Fri, 22 Feb 2002 10:59:57 GMT
Subject: fork problem
References: <3c76205c$0$7111$4d4efb8e@news.be.uu.net>
Message-ID: <usn7tkcdj.fsf@python.net>

"DelPiccolo Ivano" <ivano.delpiccolo at degroof.be> writes:

> I have the following error when i try to fork :
> 
> Traceback (most recent call last):
>   File "explorer_proxy.py", line 33, in ?
>     child = os.fork()
> AttributeError: 'module' object has no attribute 'fork'
> 
> Here is my code :
> child = os.fork()
>  if child == 0 :
>         print "\nEntering in child process"
> 
> Where is my error ?????

Trying to use on Windows a function that doesn't exist on Windows?

You can use cygwin if you're desperate, but I beleive fork() has some
issues there.

Cheers,
M.

-- 
  ... so the notion that it is meaningful to pass pointers to memory
  objects into which any random function may write random values
  without having a clue where they point, has _not_ been debunked as
  the sheer idiocy it really is.        -- Erik Naggum, comp.lang.lisp


From martin.franklin at westgeo.com  Thu Feb 14 03:33:30 2002
From: martin.franklin at westgeo.com (Martin Franklin)
Date: Thu, 14 Feb 2002 08:33:30 +0000
Subject: Hidden files, selecting directories, & taxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
Message-ID: <a4fsn9$7ie$1@mail1.wg.waii.com>

Christopher wrote:

> Hello everybody,
>     I had a couple things to ask:

> 
> 2)  Is there a simple and easy way to allow the end-user to select a
> directory?  For example, let's say all files of rpt type are saved to
> a certain directory (named by each user).  And we want to perform an
> operation on each of the files, is there an easy way to get the
> directory (when I say easy, I mean something like
> tkFileDialog.askopenfilename or the like).  Right now I skirt around
> the issue by asking the user to select one of the files they wish to
> process and then I us os calls to get the directory path then use
> listdir to get a list that I loop over.  It's a bit of a cheesy method
> but I spent a significant amount of time making it feel 'natural.'

###Tk directory selection dialog:-

from tkCommonDialog import Dialog
class Chooser(Dialog):
    command = "tk_chooseDirectory"
    def _fixresult(self, widget, result):
        if result:
            self.options["initialdir"] = result
        self.directory = result # compatibility
        return result

def askdirectory(**options):
    return apply(Chooser, (), options).show()

### for some reason this crashes on Win95 but not on any other win I 
### tested...






> 
> Chris
> 



From jeff at ccvcorp.com  Mon Feb 11 15:19:24 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 11 Feb 2002 12:19:24 -0800
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net>
Message-ID: <3C68274C.F6F702F5@ccvcorp.com>


Timothy Rue wrote:

> But then group voting is like a conspiracy huh?

Only for nonstandard definitions of 'conspiracy'.

Conspiracy implies secrecy, and imposing the will of the few upon the many.
Voting is nonsecret (even if individual votes are secret, the process is not)
and implies determining the will of the many (at the possible expense of the
will of the few).

Then again, we shouldn't be surprised at you using nonstandard definitions of
words.

Jeff Shannon
Technician/Programmer
Credit International




From AdrianLeu at kelseus.com  Thu Feb 21 07:36:54 2002
From: AdrianLeu at kelseus.com (Adrian Leu)
Date: 21 Feb 2002 04:36:54 -0800
Subject: Python crash in PyImport_ImportModule
Message-ID: <993ae4ca.0202210436.21805729@posting.google.com>

Hi ! I am getting at the end of my tether with Python. 

I have written a script in Python 2.0 under Windows 2000. I call this
script from inside a C++ class method which has the Py_Initialize()
and Py_finalize() in its constructor and destructor respectively.
Everything seemed to work OK. However, recently a crash started to
appear and it cannot be repeated always. It is in the
PyImport_ImportModule(module_name) function and the stack looks like
this:

__sbh_free_block(tagHeader * 0x03ddbc14, void * 0x04a1f490) line 350 +
6 bytes
_realloc_base(void * 0x04a1f490, unsigned int 96) line 101 + 13 bytes
realloc_help(void * 0x04a1f4b0, unsigned int 60, int 1, const char *
0x00000000, int 0, int 1) line 636 + 16 bytes
_realloc_dbg(void * 0x04a1f4b0, unsigned int 60, int 1, const char *
0x00000000, int 0) line 806 + 27 bytes
realloc(void * 0x04a1f4b0, unsigned int 60) line 755 + 19 bytes
PyNode_AddChild(_node * 0x04a1d310, int 4, char * 0x023ac9b0, int 108)
line 35 + 44 bytes
shift(stack * 0x0569d228, int 4, char * 0x023ac9b0, int 3, int 108)
line 109 + 26 bytes
PyParser_AddToken(parser_state * 0x0569d228, int 4, char * 0x023ac9b0,
int 108, int * 0x0012f010) line 211 + 25 bytes
parsetok(tok_state * 0x049d0c80, grammar * 0x1e1e14f0
__PyParser_Grammar, int 257, perrdetail * 0x0012eff8) line 123 + 34
bytes
PyParser_ParseFile(_iobuf * 0x00286868, char * 0x0012f1ac, grammar *
0x1e1e14f0 __PyParser_Grammar, int 257, char * 0x00000000, char *
0x00000000, perrdetail * 0x0012eff8) line 75 + 21 bytes
PyParser_SimpleParseFile(_iobuf * 0x00286868, char * 0x0012f1ac, int
257) line 941 + 30 bytes
parse_source_module(char * 0x0012f1ac, _iobuf * 0x00286868) line 627 +
18 bytes
load_source_module(char * 0x0012f308, char * 0x0012f1ac, _iobuf *
0x00286868) line 750 + 13 bytes
load_module(char * 0x0012f308, _iobuf * 0x00286868, char * 0x0012f1ac,
int 1) line 1227 + 17 bytes
import_submodule(_object * 0x1e1e8d48 __Py_NoneStruct, char *
0x0012f308, char * 0x0012f308) line 1755 + 33 bytes
load_next(_object * 0x1e1e8d48 __Py_NoneStruct, _object * 0x1e1e8d48
__Py_NoneStruct, char * * 0x0012f41c, char * 0x0012f308, int *
0x0012f304) line 1611 + 17 bytes
import_module_ex(char * 0x00000000, _object * 0x00000000, _object *
0x00000000, _object * 0x00000000) line 1462 + 35 bytes
PyImport_ImportModuleEx(char * 0x0032f21c `string', _object *
0x00000000, _object * 0x00000000, _object * 0x00000000) line 1503 + 21
bytes
PyImport_ImportModule(char * 0x0032f21c `string') line 1436 + 20 bytes

I don't have a clue why this happens and why it is not always
repeatable. Can someone help. Thanks anyway!

Adrian.


From borcis at geneva-link.ch  Thu Feb  7 17:40:18 2002
From: borcis at geneva-link.ch (Boris^2)
Date: Thu, 07 Feb 2002 23:40:18 +0100
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de>
	 <a3sk2i$67m$1@panix2.panix.com>
	 <e840346c.0202070321.452d405d@posting.google.com>
	 <7xr8nxy0rb.fsf@ruckus.brouhaha.com>
	 <e840346c.0202070724.13440c8d@posting.google.com> <E+9AIXAiRqY8EwKc@jessikat.demon.co.uk>
Message-ID: <3C630252.6605FA49@geneva-link.ch>

Robin Becker wrote:
> 
> I think there are good compilers for lisp which can be a very dynamic
> language so in principle it's not the dynamic nature of python that
> prevents native compilation.

Wouldn't a first step be optional type declarations ?

Boris


From mcfletch at geocities.com  Fri Feb  8 17:38:59 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Fri, 08 Feb 2002 17:38:59 -0500
Subject: [ZODB] getOID( object )?
Message-ID: <3C645383.7090908@geocities.com>

I'm doing some refactoring in a little project here, and realising that 
I'm maintaining duplicate information for my objects to create unique ID 
values.  Since ZODB is maintaining the IDs anyway, I'd like to be able 
to use those instead.

In particular, I have a lot of objects that need to have unique 
identifiers across iterations of the programme (I use these to create 
indices which allow for loading only those items which are seen in the 
current visualisation).  I currently do this by storing an incremented 
index in the database root, and manually assign an ID integer to each 
new object.  This is a really hacky-feeling approach, and I'd like to 
eliminate it entirely.

I read in the developer's documentation that I can use _p_oid if the 
object has already been stored, but I'm wondering if there's a way to 
get the oid _before_ I store the object (since I want to store the index 
information at the same time as the object itself).

Alternately, is there a way to force a Persistence.Persistence-based 
object to be stored immediately (preferably without requiring that I 
commit the whole transaction, as I'm not sure the indexing will work for 
all objects, and may need to abort) so that I can rely on there being a 
_p_oid available?  Could I use a sub-transaction, commit that and then 
use the oid?  It seems a bit of overkill for something this trivial.

By the way, ZODB is _really_ fun to work with.  Automated persistence 
machinery makes visualisation programming very slick-feeling (the GUI is 
just reflecting the object namespace living in the database).

Secondary question: given an oid from the ZODB, is there a readily 
accessible getObject( oid ) method that can go directly to the object? 
Not a big problem, I can just use the oid as keys in a PersistentMapping.

Thoughts appreciated,
Mike
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From jeff at ccvcorp.com  Mon Feb 11 15:35:44 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 11 Feb 2002 12:35:44 -0800
Subject: I wish.............
References: <nTZ88.121900$XZ1.5177196@typhoon.austin.rr.com>
Message-ID: <3C682B20.C621242B@ccvcorp.com>


DeepBleu wrote:

> .....that we quit calling Python as CPython.

Or, to rephrase what others have said,

Python is to CPython as C is to gcc

(I'd say C++/Visual C++, but CPython is at least a conforming implementation ;)
)

Jeff Shannon
Technician/Programmer
Credit International





From greg at cosc.canterbury.ac.nz  Mon Feb 25 21:33:40 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 26 Feb 2002 15:33:40 +1300
Subject: Parse variable into variables >Re: Apology
References: <slrna6ie79.2so.philh@comuno.freeserve.co.uk> <2320.811T442T11913096threeseas@earthlink.net> <slrna6t8b2.nt2.philh@comuno.freeserve.co.uk> <12393.813T2860T7905870threeseas@earthlink.net> <3c717f62_3@corp.newsgroups.com> <2240.820T2407T12845765threeseas@earthlink.net> <r29j7uo1ckvhesgc7po0thcevmdg0jeukg@4ax.com>
Message-ID: <3C7AF404.6BC628F0@cosc.canterbury.ac.nz>

Courageous wrote:
> 
> >>> s = "a,b,c,d"
> >>> tple = s.split(",")
> >>> tple
> ['a', 'b', 'c', 'd']

One further tip: you can then "unpack" the tuple
into variables, if you wish, like this:

   var1, var2, var3, var4 = tple

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


From logiplexsoftware at earthlink.net  Wed Feb 13 14:21:39 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 13 Feb 2002 11:21:39 -0800
Subject: redirect console output to widget?
In-Reply-To: <de6c123c.0202131109.2708c2f4@posting.google.com>
References: <de6c123c.0202101612.72b27f8d@posting.google.com>
	<mailman.1013395757.3817.python-list@python.org>
	<de6c123c.0202131109.2708c2f4@posting.google.com>
Message-ID: <20020213112139.102255f6.logiplexsoftware@earthlink.net>

On 13 Feb 2002 11:09:19 -0800
Lei Chen wrote:

> I think you are right.  Is there an alternative to Tkinter that is thread
safe?
 
 
> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
message news:<mailman.1013395757.3817.python-list at python.org>...
> > I didn't think you were supposed to use Tkinter in parallel threads?

I haven't followed this thread, so I'm not sure what problem you are trying
to solve, but it is possible to use wxPython with threads.  The demo comes
with an example of updating a dialog from multiple threads.


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From doug at humanoid.nsacom.net  Sun Feb 10 23:48:23 2002
From: doug at humanoid.nsacom.net (Doug  Satan Pants  Shawhan)
Date: Sun, 10 Feb 2002 22:48:23 -0600
Subject: PIL on OpenBSD 3.0
Message-ID: <nei74a.oai.ln@evansville.net>

Howdy all.

I am having some difficulty getting the Python Imaging Library to compile on OpenBSD 3.0.

My setup is as follows:

Python 2.1.1 installed from the pre-built packages
tk-8.3.3 installed from the pre-built packages
tcl-8.3.3 installed from the pre-built packages
PIL Imaging-1.1.2 from tarball

Built as per instructions in README.

The problem seems to start when finding tk.h

-------------sneep sneep-------------------------------
cc -fpic -O2 -pipe  -fPIC -I/usr/local/include/python2.1 -I/usr/local/include/python2.1 -DHAVE_CONFIG_H  -IlibImaging -I/usr/local/include -c
 ././_imagingtk.c -o ./_imagingtk.o
././_imagingtk.c:20: tk.h: No such file or directory
././_imagingtk.c:23: syntax error before `*'
././_imagingtk.c:31: syntax error before `Tcl_Interp'
././_imagingtk.c: In function `_tkinit':
././_imagingtk.c:37: `Tcl_Interp' undeclared (first use in this function)
././_imagingtk.c:37: (Each undeclared identifier is reported only once
././_imagingtk.c:37: for each function it appears in.)
././_imagingtk.c:37: `interp' undeclared (first use in this function)
././_imagingtk.c:39: syntax error before `long'
././_imagingtk.c:41: `arg' undeclared (first use in this function)
././_imagingtk.c:41: `is_interp' undeclared (first use in this function)
././_imagingtk.c:45: syntax error before `)'
././_imagingtk.c:50: `app' undeclared (first use in this function)
././_imagingtk.c: At top level:
././_imagingtk.c:55: warning: parameter names (without types) in function declaration
././_imagingtk.c:55: conflicting types for `TkImaging_Init'
././_imagingtk.c:23: previous declaration of `TkImaging_Init'
././_imagingtk.c:55: warning: data definition has no type or storage class
././_imagingtk.c:57: syntax error before `&'
*** Error code 1

Stop in /home/towering_intellect/Imaging-1.1.2 (line 324 of Makefile).
--------------sneep sneep-------------------------------

I have attempted to add the path to tk.h in _imagingtk.c to no avail.

Anyone else had any luck getting this dandy package to compile?

Thanks,

d



From spencer at spencerdoidge.com  Sun Feb  3 13:47:11 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sun, 3 Feb 2002 10:47:11 -0800
Subject: Compiling from Pythonwin?
Message-ID: <a3k0cf$dko$1@news.efn.org>

Is there a way to compile .py scripts from the Pythonwin GUI screen other
than using
compileFile(...)?





From bokr at oz.net  Tue Feb 19 20:14:33 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 01:14:33 GMT
Subject: Adding static typing to Python
References: <3C726575.5090400@destiny.com> <mailman.1014133892.1456.python-list@python.org> <q3057us1qab29uvm2c6f551p4vs6j7pjvr@4ax.com>
Message-ID: <a4ut9p$ngd$0@216.39.172.122>

On Tue, 19 Feb 2002 16:37:23 GMT, Courageous <jkraska at san.rr.com> wrote:

>
>>Thanks for the link. Reading it makes me wonder if some sort of partial
>>static typing would be beneficial, e.g. if the compiler was able to
>>determine up front that X is an atomic ...
>
>That's everyone's first thought, however in the generation of
>a compiler, several other first steps are necessary to achieve
>performance; these steps can be achieved without static typing
>or even compiler hints. While optional static declarations are
>high on the list of things that can increase performance in a
>compiled python executible, there are things which are higher
>and have nevertheless not been done. That they haven't means in
>my mind that we should stop, and not add (optional) static typing
>until and unless they have. Because right now it would be mostly
>linguistic masturbation and not really help anything at all.
>
>If you're thinking that the compiler folks haven't made any
>progress because strictly there's no static type system, you'd
>be wrong.
>
I'm wondering whether at any point the interpreter has been instrumented
to count symbol bindings in terms what types they bind to at first
and how often they get rebound to a different type vs the same type?
For locals, globals, attributes, (now) slots, and maybe dictionary
key:value bindings in general?

It would be an interesting measure of how dynamic things really are
in practice.

Regards,
Bengt Richter



From mcherm at destiny.com  Tue Feb 19 09:47:17 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Tue, 19 Feb 2002 09:47:17 -0500
Subject: Adding static typing to Python
Message-ID: <3C726575.5090400@destiny.com>

No discussion of adding static typing to Python would be complete 
without a reference to Guido's thoughts on the subject:

    http://www.python.org/~guido/static-typing/index.htm

I suspect that these may be a bit dated (~2 yrs old), but a few key 
ideas should still be relevent:

    * Static typing should be optional.
    * The 'purpose' of static typing is to ensure that type errors are
      caught at compile time, not runtime (in checked code).
    * Don't expect huge performance gains -- catching certain kinds of
      errors earlier is the main gain.

Not that I fully agree, or want to tie Guido to old ideas, but those who 
come to this idea fresh should realize it has been considered before. 
The main results of the previous consideration are:

    * Performance isn't really the issue (though people often think it
      is).
    * Unit testing could (if used well) catch the same errors that static
      type checking catches.
    * Unit testing is a good thing.
    * Python is a REALLY dynamic language, and one which tends to trust
      the programmer.

-- Michael Chermside




From joshm at taconic.net  Tue Feb  5 20:07:41 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 5 Feb 2002 20:07:41 -0500
Subject: Which GUI toolkit to use?
Message-ID: <3c607ffd_3@corp.newsgroups.com>

Specifics:

1.  Needs to run on Windows platform.
2.  Needs to easily support completely custom paint (buttons, scrollers,
listboxes, etc).
3.  Needs to handle simple input (editboxes, onLeftClick events).

The goal is to produce an app that runs full screen on a Win platform, and
is interacted with via a touchscreen.  I would like to just paint the whole
damn thing myself, but a widget manager will make input handling easier.
There will be a lot of bitmaps making the whole thing pretty.

Tk seems like it is a pain to use, wxWindows seems like it doesn't make
painting easy, pythonwin (win32ui) might be the way to go, ...

Decisions, decisions.  There wouldn't be anyone around here who has an
opinion, is there?  :-)

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From jlh at cox.net  Sat Feb 23 12:01:19 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Sat, 23 Feb 2002 17:01:19 GMT
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones> <a58885$5acoc$1@ID-12869.news.dfncis.de>
Message-ID: <zVPd8.4096$jC3.163217@news1.east.cox.net>

fp is a file pointer.
open the file (for write "w") and you are handed a file pointer
pass it to the ConfigParser, it writes to the file pointer
then you close the file.

hth
"Ricky" <store_li at sina.com> wrote in message
news:a58885$5acoc$1 at ID-12869.news.dfncis.de...
> Yes, as you guess, I haven't call .write method, because I was confused by
> the usage of write method. It always required a fp object, which I totally
> don't what it means. Is it a object create by
a=ConfigParser.ConfigParser()
> or a file name or a section name?
>
> Please help me! Thank you!
>
>
> "Graham Ashton" <graz at mindless.com> wrote in message
> news:fUJd8.418$PE2.43612 at stones...
> > On Fri, 22 Feb 2002 16:36:22 +0000, Ricky wrote:
> >
> > > I intend to use ConfigParser module to manage my INI file. I can use
get
> > > to retrieve data from my ini file, and use set to change option value
in
> > > Python, just in python, after I closed python, I found that all
changes
> > > I made in python lost. There is nothing altered in my ini file!
> > >
> > > How can I make my ini file to be changed?
> >
> > As you haven't posted any code I can't be sure, but I suspect from what
> > you've said that you've not called the write() method on your
> > ConfigParser object.
> >
> > Read the module's docs.
> >
> > --
> > Graham
>
>




From spencer at spencerdoidge.com  Sat Feb 16 22:35:05 2002
From: spencer at spencerdoidge.com (Spencer Doidge)
Date: Sat, 16 Feb 2002 19:35:05 -0800
Subject: timer delays and kbhit() in Pythonwin?
Message-ID: <a4n85k$ij2$1@news.efn.org>

The following script works fine running under Windows in the console, but it
doesn't work right in Pythonwin.
The point of this script is that while we are waiting 20 seconds for the
'hello' message to be printed, we can intervene with a keystroke and cause
function doit() to terminate immediately.
I suppose it doesn't work right in Pythonwin because kbhit() is only
supported in the console. Is there a way to accomplish this same purpose
running Pythonwin?
----------------------------------------------------------
from threading import *
from msvcrt import *

def hello():
    print 'hello'

def doit():
    t=Timer(20.0,hello)
    t.start()
    done = 0
    while(done==0):
        if(kbhit()):
            done=1
            t.cancel()
            print 'cancelled'
        elif (t.finished.isSet()):
            done=1
            print 'done'
----------------------------------------------------------






From ftouch at aol.com  Wed Feb  6 16:32:06 2002
From: ftouch at aol.com (david)
Date: 6 Feb 2002 13:32:06 -0800
Subject: Looking for a python developer
Message-ID: <ef7772ea.0202061332.7f882f56@posting.google.com>

Hello

A python programmer with Object Oriented working experience is needed
in Baltimore for a full time on site job , for immediate
consideration.

Please send your updated resume to ftouch at aol.com (directly to the
company). Good opportunity.

Thank You
David Evgey


From cfelling at iae.nl  Thu Feb 28 15:25:06 2002
From: cfelling at iae.nl (Carel Fellinger)
Date: 28 Feb 2002 21:25:06 +0100
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch> <a5kcf4$qei$1@peabody.colorado.edu> <slrna7s4fc.4pv.scarblac@pino.selwerd.nl> <a5lk0a$8enfq$1@ID-11957.news.dfncis.de> <a5lltf$4ra$1@animus.fel.iae.nl> <a5lvav$8614t$1@ID-11957.news.dfncis.de> <a5m0co$8klqb$1@ID-11957.news.dfncis.de>
Message-ID: <a5m3n2$8fk$1@animus.fel.iae.nl>

Emile van Sebille <emile at fenx.com> wrote:
> Hmmm... Gerson may have given higher marks to my prior sample, but this
> code is a bit cleaner without losing any of the evilness... ;-)
...
> for ii in dexing(mylist):
>     ii *= 4 #* ii

You evil wicked man:)

The downside is that it only works for sequences.
It's a no-go for lists, iterators and generators:(

-- 
groetjes, carel


From mcherm at destiny.com  Thu Feb  7 16:24:09 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Thu, 07 Feb 2002 16:24:09 -0500
Subject: __class__ for 2.2 types
Message-ID: <3C62F079.5020906@destiny.com>

> hi, rookie here, I was hoping someone could explain to me why this happens:
> 
> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
>>>> [].__class__
> <type 'list'>
>>>> int.__class__
> <type 'type'>
> 
> why doesn't the int one return <type 'int'>? 
> 
> thanks in advance
> 


Because "int" isn't an int. "5" is.


>>> [].__class__
<type 'list'>
>>> x = 7
>>> x.__class__
<type 'int'>


>>> int.__class__
<type 'type'>
>>> str.__class__
<type 'type'>






From wookiz at hotmail.com  Sat Feb  9 17:32:35 2002
From: wookiz at hotmail.com (wooks)
Date: 9 Feb 2002 14:32:35 -0800
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <3C64F51A.2070208@sneakemail.com>
Message-ID: <8852d33c.0202091432.452ad6f9@posting.google.com>

Joshua Macy <l0819m0v0smfm001 at sneakemail.com> wrote in message news:<3C64F51A.2070208 at sneakemail.com>...
> 
>    Have you tried following the Python for Beginners link on 
> www.python.org?  http://www.python.org/doc/Newbies.html has links to all 
> sorts of material for beginners, including help at python.org (a volunteer 
> help desk) and tutor at python.org (a low volume mailing list).

Yes. See reply to Sheila.

>    I don't think searching the discussion archives is all that helpful 
> unless you have a specific question that you want answered and if you 
> do, you haven't mentioned it.
> 

I did have a specific question but I searched the archives first
before posting it. What I read led me to put up this post rather than
ask my question.


> 
>    comp.lang.python is not a forum specifically for beginners; 

wouldn't expect it to be
If you dip into a random 
> 
>    As for concepts "understood since grade 2", that usually means "it 
> worked like X in this other language Y that I know, why doesn't Python 
> follow this 'natural' approach?"  If you hang around a while, though, 
> you'll start to notice what the veterans of the newsgroup realized a 
> long time ago, which is that the 'naturalness' of X depends entirely on 
> which language Y the person is thinking of.  

because that language handles it the way Grade 2 taught him it would
be handled.  To me that should set the intellectual level of the
discussion.

> 
>    I think that's exactly it.  If you really want to know why something 
> in Python is the way it is conceptually, you often have to be prepared 
> for a somewhat complicated answer, philosophically or mathematically. 

Not found this with Perl. The Camel book does an excellent job of that
sort of thing. I've had my problems with that book but they are of a
different nature. Unfortunately the reviews have dissuaded me from
acquiring it's Python counterpart - Programming Python.

> 
> One of the basic philosophies of Python design is that it is wrong for 
> the language to guess when presented with something that's ambiguous. 

Which seems to be the opposite of the Perl philosophy - we will try to
do the right thing.

> That something is "automagical" is usually a swear word in Python.  This 
> can make Python seem hairier than other languages which cheerfully take 
> a stab at what the designers think you may have wanted, but the Python 
> belief is that those languages are leading you down the garden path, and 
> sooner rather than later you'll be bitten by the subtlety that the other 
> language tried to gloss over.
> 

Hmmm. The good thing about the Perl approch is that (if you read the
right resource) you get an explanation of the process via it tries to
do the right thing, which is very effective at getting the conceptual
message across.
 
> 
> Did you actually try any of the examples of manipulating Word or Excel 
> from Python?   Python for Win32 is an excellent book, IMO, but way 
> overkill unless you have some specific need to interact with COM or the 
> Win32 OS APIs. It is not a book about learning Python.
> 
Haven't got that far. The book is the most appropriate for my needs -
scripting glue, interacting with WSH and COM rather than building
applications. I met the prerequisites for readership but I am finding
out that the devil is in the detail.

> 
> > 
> > The benefits of using of Python as a scripting glue like language are
> > widely touted but don't seem (I am probably wrong but it's a 1st
> > impression) to be well catered for. As I said I am not interested in
> > building applications. Is there a forum for the dumber less demanding
> > Python user?
> > 
> 
> 
>    You need to ask a more specific question, or be more specific about 
> what exactly you're trying to accomplish and finding difficult.  General 
> complaints that it seems difficult to understand aren't likely to 
> generate much helpful advice.
> 

I was very specific. Is there a forum for the dumber less demanding
Python user? I got answers that I consider helpful. Thanks to you and
Sheila.

>    And, don't take this the wrong way, but it *is* dumb to go to a 
> language newsgroup and say "Maybe I'm dumb but [vague complaints about 
> the language not being easy]" if you have any expectation that posters 
> will rush in saying "No, no, you're not dumb, it's obviously the 
> language's fault" the way your friends and relatives might.  Usenet is 
> not the place to cruise to have your ego soothed.
> 

Let me put it somewhat differently so that there is no doubt about my
objective.
You guys say our language is a good one to learn first, it's really
simple. Well I recently got to grips with VB and when I've had cause
to go to a VB newsgroup have had no problem coping with the
intellectual level at which discussion is pitched. Now that to me says
something (even if it is VB is a language for idiots). I found
VBScript relatively painless to learn and it can get the job done for
me, I have been drawn to Python because of the unit testing tools
available (PyUnit and Pester) and decided to use it as a learning
opportunity.

I am not dumb. My ego is fine. I am just giving my impressions of
Python and some of the claims that advocates make of it.


From scarblac at pino.selwerd.nl  Tue Feb 12 08:47:10 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 12 Feb 2002 13:47:10 GMT
Subject: how to compare two strings?
References: <7xofivazj1.fsf@ruckus.brouhaha.com>
Message-ID: <slrna6i71s.597.scarblac@pino.selwerd.nl>

Paul Rubin <phr-n2002a at nightsong.com> wrote in comp.lang.python:
> Is there any built-in function in Python for comparing two strings
> alphabetically?  Something suitable for passing as the optional
> comparison function for the list "sort" method.  I could code up
> something like
> 
>   def scmp(a,b):
>     if a<b: return -1
>     if b<a: return 1
>     return 0
> 
> but it's pretty ugly to need to do that.  I've looked in the docs but
> don't see any obvious counterpart to perl's lozenge (<=>) operator.
> Is there one?

Uhm. scmp() is the same as the builtin cmp(), as far as I can see.

Which means that this is the default action of sort(), no need to give it
any function.

-- 
Remco Gerlich


From christophertavares at earthlink.net  Tue Feb 19 15:17:40 2002
From: christophertavares at earthlink.net (Chris Tavares)
Date: Tue, 19 Feb 2002 20:17:40 GMT
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com> <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com> <3C715962.A42F1E48@ccvcorp.com> <mailman.1014094712.4434.python-list@python.org>
Message-ID: <Epyc8.856$0C1.79684@newsread1.prod.itd.earthlink.net>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in message
news:mailman.1014094712.4434.python-list at python.org...
> ----- Original Message -----
> From: "Jeff Shannon" <jeff at ccvcorp.com>
> >
> > Michael Kelly wrote:
> > >
> > > Out of curiosity, did they give you a type library
> > > with the control?
> >
> > This is important.  As I understand it, without a type library,
pythoncom
> > has no way of knowing which params are [out] params.  Since Python
doesn't
> > have reference semantics like those of C++ or Visual Basic, it cannot
use
> > [out] params in the "normal" way.  So if you're using an interface that
> > makes use of [out] params, you *need* the type library in order to be
able
> > to use it from Python.  If you're not sure whether you've got one, try
> > Makepy, as suggested...
>
> Type libraries have a .tlb extension normally, correct?  The ZIP contains
> ReportOcx.lic, .lib, and .ocx; I'm a novice at Windows and had to be
banged
> on the head before I put the .lic and .ocx together in the SYSTEM folder.
> I did register the OCX using REGSVR32 as noted on several websites, and I
> ran makepy from the Pythonwin menu as I have in several other instances.
>
> Have I missed anything?
>

Type libraries are often compiled right into the control they're for. Try
running makepy on the OCX file.

> The source files are not available (it's a form of shareware, where you
pay
> for source access) so I can't build a type library even if I knew how.
>
> I should just go back to Linux where everything makes sense...

Linux doesn't make any more sense than Windows, it's just a different kind
of sense. :-)

-Chris





From MaximilianScherr at T-Online.de  Tue Feb 26 13:50:15 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Tue, 26 Feb 2002 18:50:15 -0000
Subject: thread question
In-Reply-To: <HFEKILOLEFEFMKAECNDLOENNDEAA.jason@jorendorff.com>
Message-ID: <a5gld7+3jal@eGroups.com>

hmm if i now just have a random threadobject, like the one returned 
from enumerate()

how can i then access the data(attributes, functions...) of that 
thread?


--- In python-list at y..., "Jason Orendorff" <jason at j...> wrote:
> max wrote:
> > can i acces attributes, or functions of a thread(like if i 
started 
> > the class with threading.Thread(None, someclass)) like i can 
access 
> > classes? and if yes how? if no, how can i do it in another way?
> 
> You probably want something like this.
> 
> class SomeClass(threading.Thread):
>     def run(self):
>         ...
>         ... Here you can access attributes and methods 
>         ... of self, which is a Thread object.
>         ...
> 
> myThread = SomeClass()
> myThread.start()
> 
> ## Jason Orendorff    http://www.jorendorff.com/
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list




From rasmussn at lanl.gov  Tue Feb 26 12:38:18 2002
From: rasmussn at lanl.gov (rasmussn at lanl.gov)
Date: Tue, 26 Feb 2002 10:38:18 -0700
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <a5ekpq$n83$1@bob.news.rcn.net>
Message-ID: <9EF5B844-2ADF-11D6-A367-0030657EBC3E@lanl.gov>

On Monday, February 25, 2002, at 05:32 PM, David Abrahams wrote:

> I've been interested in systems for automatically wrapping C++ and
> generating documentation for a while, so I downloaded the PDT (the parser
> toolkit used by SILOON) and tried it on one of my source files. After lots
> of hacking, I was able to get the EDG front-end component to parse my 
> source
> and generate an intermediate language (.cil) file, but the PDT executable
> which is supposed to interpret the .cil and produce a high-level 
> description
> of the code crashed with an assertion:

SILOON uses scripts to run the PDT and we've worked hard to make it easy
to use.  Essentially all you have to do is edit a file (that is included
in a Makefile) and point macro variables to the location of your source
files (.h .cpp ...).  I would be interested to know if you have similar
problems using the EDG front-end from SILOON directly.


> Assertion failed: ptr->assoc_template, file c:\documents and
> settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-
> tau\il_display_tau.
> c, line 3047
>
> Since the PDT doesn't come with source, there was no way for me to debug/
> fix
> the problem.

The PDT comes with sources for everything (almost) but the EDG front end.  
What PDT version
were you using?  Would you like to send your source files to me directly 
and I can
look at the problem, although if it is windows-specific code, I can't help 
you
as I don't have access to Microsoft compilers.

(On further examination, it looks like the PDT doesn't have source for 
taucpdisp,
I'll ask the developers why not.)

I can provide access to a prerelease version of SILOON which uses the 
latest
version of the PDT, which may fix your problems.

> Furthermore, the PDT developers don't seem to want to be reached. Their
> contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/) 
> says:
>
>     To contact the developers of PDT, please use the "Comments" section in
> the download page.
>
> But of course, there is no "Comments" section on the download page. :=(

I've forwarded you comments on to the PDT developers so that they can fix
their web pages and comment personally on the core dump.

Craig




From jwbaxter at spamcop.net  Thu Feb 21 21:59:14 2002
From: jwbaxter at spamcop.net (John Baxter)
Date: Thu, 21 Feb 2002 18:59:14 -0800
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3c7455a8_4@corp.newsgroups.com> <jwbaxter-CE1C5D.00553921022002@corp.supernews.com> <3C74BD08.FD8CAACE@alcyone.com>
Message-ID: <jwbaxter-E8520F.18591421022002@corp.supernews.com>

In article <3C74BD08.FD8CAACE at alcyone.com>,
 Erik Max Francis <max at alcyone.com> wrote:

> John Baxter wrote:
> 
> > I don't think strings contain characters.  I think strings contain
> > one-character strings.
> 
> Naturally, because there is no distinct concept of a character in Python
> as opposed to a one-character string.

Yes...it was knowing that which led to the "test" and the remark.  Which 
should have spoken of a string of length 1 rather than a one-character 
string.

  --John


From philh at comuno.freeserve.co.uk  Tue Feb 12 15:43:24 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Tue, 12 Feb 2002 20:43:24 +0000
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013534600.28550.python-list@python.org>
Message-ID: <slrna6ivjb.4ol.philh@comuno.freeserve.co.uk>

On Tue, 12 Feb 2002 11:10:48 -0600, Gustavo Cordova <gcordova at hebmex.com> wrote:
>> >"-ist" - an individual professionally or avocationally 
>> occupied with the
>> >idea or activity defined by the root [1].
>> 
>> >Of course, in either Spanish or Esperanto it would more commonly be
>> >"Pythonisto" anyway, so who knows.
>
>Kinda wrong for Spanish though.
>
>A follower of Zapata ("Emiliano Zapata", perhaps?) would be "Zapatista",
>be it a male or female. "Zapatisto" is a gross error, and it's funny
>also :-)
>
>It's common for little children to mistakenly use "-isto" as well as
>"-ista" when talking about boys or girls, that's why it's funny, because
>it's an infantile mistake.

Is -ista always feminine? E.g. is it always _la dentista_ not _el dentista_?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jkraska at san.rr.com  Mon Feb 11 21:15:41 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 12 Feb 2002 02:15:41 GMT
Subject: [Python-Dev] RE: RFC: Option Parsing Libraries
References: <CLEFLIBFLLLHJNOKPNGFKEDNCDAA.Gerson.Kurz@t-online.de> <mailman.1013456481.26091.python-list@python.org>
Message-ID: <45vg6u8sq3e7j7pfng2h6h7a4j6n73pd0a@4ax.com>

>Interesting idea; should be trivial given a case-insensitive dictionary.

The smallest of changes to the hashing and matching algorithms should
do it. It might be better to downcase everything pertinent as its lexed,
however. For that matter, one might simply enforce lower case and issue
a syntax error whenever an upper case letter is encountered. :)

C//



From dgrisby at uk.research.att.com  Tue Feb 19 05:52:02 2002
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 19 Feb 2002 10:52:02 GMT
Subject: Python calling Java
References: <a4rjq4$4n8$1@web1.cup.hp.com>
Message-ID: <a4taoi$fe1$2@pea.uk.research.att.com>

In article <a4rjq4$4n8$1 at web1.cup.hp.com>, jimh <jim at jim.jim> wrote:

>We have a bunch of Python (cgi) scripts producing html behind an Apache web
>server.  We also have a set of Java libraries that I want to access from
>these Python scripts, but I haven't been able to figure out how to do this.
>I have downloaded and tried out Jython, but it takes 5-6 seconds just to run
>a one-line 'print "hi"' script, and this is, of course, not acceptable for a
>web server.

Depending on exactly what you need to do, it might be possible to run
the Java code in a separate process, and talk to it from your Python
scripts using CORBA or XML-RPC. Since the Java process would be
running all the time, the start-up overhead would not be an issue.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From ivo at nospamAmaze.nl  Mon Feb  4 11:38:07 2002
From: ivo at nospamAmaze.nl (ivo at nospamAmaze.nl)
Date: 4 Feb 2002 16:38:07 GMT
Subject: Fun with lambda and map
References: <mailman.1012542242.4324.python-list@python.org>
Message-ID: <a3mddf$7do$1@news1.xs4all.nl>

Jason Orendorff <jason at jorendorff.com> wrote:
: Ivo wrote:
:> I implemented some of the pure (?) lambda calculus in python:
:> http://vanderwijk.info/lambda
:> Who needs '+' , 'if' and numbers? :)

: Wow.  It's extra cool that this should come up, because I just
: recently tore through the first chapter of SICP, and wrote this:

:   http://www.jorendorff.com/toys/truth.html

I extended the calculus a bit with mul (based on your sig :) and pred, 
hoping that I could fac in lambda, but I guess I've reached pythons lambda 
limits.

The statement:

fac = lambda q: if_then_else(is_zero(pred(q)))(n1)(mul(q, fac(pred(q))))

(check vanderwijk.info/lambda for the other definitions)

will recurse endlessly, because the else-part will always be evaluated
(but just not used as a result).

(why I do all this? Just pure academic interest - lambda calculus
used to fascinate me at the university)

	Ivo

-- 
Drs. I.R. van der Wijk                              -=-
Brouwersgracht 132                      Amaze Internet Services V.O.F.
1013 HA Amsterdam                                   -=-
Tel: +31-20-4688336                          Linux/Web/Zope/SQL
Fax: +31-20-4688337                           Network Solutions
Web:     http://www.amaze.nl/                    Consultancy
Email:   ivo at nospamAmaze.nl                         -=-



From phr-n2002a at nightsong.com  Tue Feb 19 09:26:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 19 Feb 2002 06:26:04 -0800
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <mailman.1014106650.11778.python-list@python.org> <7x7kp9g6pr.fsf@ruckus.brouhaha.com> <1014115178.2891.30.camel@matijek> <mailman.1014126336.32560.python-list@python.org>
Message-ID: <7x7kp9blab.fsf@ruckus.brouhaha.com>

Gustavo Niemeyer <niemeyer at conectiva.com> writes:
> > Sure, granted that you keep track of the middle link of your list, so
> > that you don't have to traverse up to that point. But even if keep track
> > of it you'll have to be doing pretty large lists before the linked list
> > will outperform the builtin list.
> 
> Agreed.
> 
> It's also nice to remember why we use python in the first place. If I
> had to write a linked list to keep billions of records with a very specific
> and performance dependent algorithm, I would probably write a python module
> in C. And if the whole application is dependent on performance, I wouldn't
> use any high level language to develop it.

By "middle" I didn't mean the geometric middle, but the usual case of
wanting to insert a new node after an existing node that you already
have a handle on. 

Yes, the lists have to be pretty large for implementing linked lists
in Python to beat Python's built-in vectors.  But that's only because
interpreting Python code is so abysmally slow, which is a Python
implementation artifact.  If and when faster Python implementations
(i.e. with real compilers) become available, the crossover point will
probably become a lot lower.  Even as it is, though, your measurement
only went up to 20,000 nodes; I'd be interested in the timing at a
million nodes.



From mdk at mdk.com  Fri Feb 15 08:40:34 2002
From: mdk at mdk.com (MDK)
Date: Fri, 15 Feb 2002 08:40:34 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013729004.19316.python-list@python.org>
Message-ID: <a4j34i$ltrn$1@ID-98166.news.dfncis.de>

"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1013729004.19316.python-list at python.org...
> MDK wrote:
> > Pickling seems too slow.
>
> Really?  What makes you say this?
>
> In some random tests on my box, it seems that
> repr() is a teensy bit faster than cPickle.dumps(), and
> eval() is several times slower than cPickle.loads().
>
> ## Jason Orendorff    http://www.jorendorff.com/
>

Here is a copy/paste from the documentation (see note at end):

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

3.14 pickle -- Python object serialization

The pickle module implements a basic but powerful algorithm for ``pickling''
(a.k.a. serializing, marshalling or flattening) nearly arbitrary Python
objects. This is the act of converting objects to a stream of bytes (and
back: ``unpickling''). This is a more primitive notion than persistence --
although pickle reads and writes file objects, it does not handle the issue
of naming persistent objects, nor the (even more complicated) area of
concurrent access to persistent objects. The pickle module can transform a
complex object into a byte stream and it can transform the byte stream into
an object with the same internal structure. The most obvious thing to do
with these byte streams is to write them onto a file, but it is also
conceivable to send them across a network or store them in a database. The
module shelve provides a simple interface to pickle and unpickle objects on
DBM-style database files.

Note: The pickle module is rather slow. A reimplementation of the same
algorithm in C, which is up to 1000 times faster, is available as the
cPickle module. This has the same interface except that Pickler and
Unpickler are factory functions, not classes (so they cannot be used as base
classes for inheritance).

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

My concern was that even though cPickle is supposed to be faster is does not
mean that is it as fast as it needs to be.




From sholden at holdenweb.com  Thu Feb  7 11:11:31 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 11:11:31 -0500
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <f72dac0d.0202070547.8cb548a@posting.google.com>
Message-ID: <HGx88.95599$KM2.3475019@atlpnn01.usenetserver.com>

His slides are available at

    http://www.w3.org/2002/Talks/0206-python

As for the content, he took an ec;ectic view of Python, and suggested that
his thoughts on "Webizing" weren't necessarily particularly practical (with
which I'd agree).

The Semantic Web, about which he spent much time talking, appears to be a
binary relational database for storing semantic knowledge. Interesting
twenty years ago...

I especially liked the AaronSwartz link on the last slide. :-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py

"Alan Kennedy" <alanmk at hotmail.com> wrote in message
news:f72dac0d.0202070547.8cb548a at posting.google.com...
> Greetings,
>
> Unfortunately, I wasn't able to attend Python 10, even though I'd love
> to have heard what Tim Berners Lee had to say about "Webizing Python".
>
> Does anyone have any reports that they could share with us, or better
> still, a transcript of his speech?
>
> Regards,
>
> Alan.




From James.Strater at exolink.com  Tue Feb 26 15:29:28 2002
From: James.Strater at exolink.com (James Strater)
Date: Tue, 26 Feb 2002 14:29:28 -0600
Subject: embedding a jpg in an excel spreadsheet
Message-ID: <F666B26E8EF5B84C88E3DF34E5A1E49CF07FCA@venus.exolink.com>

Hello,  

I am creating an excel spreadsheet using python win32com.client.  This all
works fine.

In the first row/cell of the spreadsheet, I want to embed a logo which is
stored in a a jpg file.  Can anyone suggest how to do this?

Also, I need to change the cell background colors on some rows.  Again, any
suggestions?

Thanks,

James



From nospam at bigfoot.com  Wed Feb 13 13:24:56 2002
From: nospam at bigfoot.com (Gillou)
Date: Wed, 13 Feb 2002 19:24:56 +0100
Subject: win32com.client.DispatchWithEvents and exceptions from callback
Message-ID: <a4eak8$1t2o$1@norfair.nerim.net>

Hi,

I got a strange behaviour of COM object made accessible through the Dispatch
This is a script that uses the latest MSXML 4.0 DomDocument.
The callback class DomDocumentEvents runs but the DomParseException
exception never raises.
Means that when using it with non well-formed or valid XML document, it
prints "Exception should raise" but the exception is NOT raised (the main
script carries on as usual) !
That said it returns a good DOM object when used with a valid XML doc, but I
don't find a *pythonic* way to handle validity errors if the exceptions is
not raised.

Any idea ? or is there something I'm missing ?

Thanks in advance.

--Gilles

======= MSXML4 wrapper =====
TRUE = 1
FALSE = 0
import exceptions
import win32com.client

class DomParseException(exceptions.Exception):
    '''Raised on well-formedness or validity error
    '''
    def __init__(self, errobj):
        # ...
        return
    def __str__(self):
        # ...
        return

class DomDocumentEvents:
    '''Handle callback events of MSXML4
    '''
    def Ononreadystatechange(self):
        '''Check for well-formedness or validity error
        '''
        if (self.readyState == 4 and self.childNodes.length == 0):
            print 'Exception should raise'
            raise DomParseException(self.parseError)
        return

    def Onondataavailable(self):
        '''May handle this for your own use
        '''
        print 'Data available'
        return

def newDomDocument():
    dd = win32com.client.DispatchWithEvents('Msxml2.DOMDocument.4.0',
DomDocumentEvents)
    dd.setProperty("SelectionLanguage", "XPath");
    dd.validateOnParse = TRUE
    dd.resolveExternals = TRUE
    dd.async = FALSE
    dd.preserveWhiteSpace = FALSE
    return dd

if __name__ == '__main__':
    domdoc = newDomDocument()
    # Loadind insane XML
    domdoc.load('text.xml')
    # Should have an exception before next line
    print domdoc.xml.encode('cp850')







From Mitch.Chapman at bioreason.com  Thu Feb  7 11:56:38 2002
From: Mitch.Chapman at bioreason.com (Mitch Chapman)
Date: Thu, 07 Feb 2002 09:56:38 -0700
Subject: Delphi vs Python
References: <zBm88.333397$pa1.80603034@news3.rdc1.on.home.com>
Message-ID: <3C62B1C6.DFFE5963@bioreason.com>

Bo Vandenberg wrote:
> Has anyone used Delphi to make python called DLL's ?

Have you heard of PythonForDelphi?  It uses Delphi's VARIANT 
types to wrap up access to Python objects.  In other words,
it lets you embed Python in Delphi and access Python objects as
if they were native Delphi objects.  And it does this without
requiring you to write any wrapper code.  It's amazing,
especially if you've tried embedding Python in C/C++.


http://www.multimania.com/marat/delphi/python.htm


Highly recommended for building GUI apps under Windows.

The documentation consists mostly of tutorials, some of which show 
the "hard" way to solve various problems.  But it's not bad. 
Besides, if I think it could be better then I ought to help improve 
it rather than whine like an ingrate :)

-- 
Mitch Chapman
Mitch.Chapman at bioreason.com


From chris.gonnerman at newcenturycomputers.net  Thu Feb  7 01:13:46 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Thu, 7 Feb 2002 00:13:46 -0600
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>
Message-ID: <00b901c1af9e$9de1d080$0101010a@local>

The time has come, my friends and colleagues, to bring the issue at hand
to a vote:

Should Timothy Rue go away?

All in favor, post AYE.  All opposed, post NAY.

(Yes, this is a waste of bandwidth, but in my opinion not as bad as Rue.)

I recommend you top-post your answers, or better yet, remove all this text
from your posting and just post the single word.





From marknenadov at winningteam.com  Mon Feb 18 11:56:25 2002
From: marknenadov at winningteam.com (Mark Nenadov)
Date: Mon, 18 Feb 2002 11:56:25 -0500
Subject: Performances of Pyhton programs
References: <Xns91B9AA94FE461bairef2ctiecpfr@138.195.8.3>
Message-ID: <20020218.115624.1096689772.1978@d57-47-79.home.cgocable.net>

> Are there any benchmark or does anoyone know what are the performances
> of a program written in python (in terms of CPU and RAM) compared with
> other languages (especially C++ and compiled languages) ?
>    I've met many articles which explain that python programs are slowler
> because it uses the interpreter, but none says how much slower it is ! I
> would like to know this for a rather big OO program which uses standard
> fonctions of the language.
> Thanks
> --
> Fran?ois Baire

You may want to check out "The Great Computer Language Shootout" 
(http://www.bagley.org/~doug/shootout/) or "An empirical comparison of 
C, C++, Java, Perl, Python, Rexx, and Tcl for a search/string-processing 
program" (http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprtTR.ps.gz). I 
found those to be interesting. They will probably have some helpful info 
for you.

I hope that helps,
~Mark Nenadov


From fredrik at pythonware.com  Fri Feb  8 03:04:27 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 08 Feb 2002 08:04:27 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com> <slrna650nr.ahr.philh@comuno.freeserve.co.uk> <a3v1fc$jmc$1@newshost.accu.uu.nl>
Message-ID: <fEL88.24196$l93.4429766@newsb.telia.net>

Martijn Faassen wrote:

> those of us who are smart enough to see it and can look it straight in
> the eye, but deliberately suppress the knowledge because we'd be out of
> a job. We'll call this the 'evil' category.

well, I'm sure lots of people agree that US Military could qualify
as "evil", but it's not exactly like they're suppressing the know-
ledge:

http://www.mcci-arl-va.com/techpape.htm

    "[the] Autocoding Toolset automates generation of
    production quality, thread-safe, multi-tasking, multi-
    threaded application code for parallel architectures.
    ...
    The Autocoding Toolset is commercially available now
    ...
    The Autocoding Toolset is the result of an effort to
    port military automatic code generation technology to
    COTS processors funded under the DARPA-sponsored
    RASSP program and commercialization under a Navy
    SBIR."

</F>




From rkashyap at sympatico.ca  Tue Feb 19 20:00:31 2002
From: rkashyap at sympatico.ca (Ramkumar Kashyap)
Date: Tue, 19 Feb 2002 20:00:31 -0500
Subject: Python and GIS
Message-ID: <3C72F52F.5050901@sympatico.ca>

Hi,

I was wondering if there are any modules for GIS, Mapping, Layering, or 
hooks to Arcinfo, Mapinfo, etc.

TIA,

Ramkumar



From dan at eevolved.com  Mon Feb 18 02:17:44 2002
From: dan at eevolved.com (Dan P)
Date: Mon, 18 Feb 2002 02:17:44 -0500
Subject: RELEASED! [OOPS v 0.9.1] (OO Embedded DB lib)
Message-ID: <e22c8.10251$dS6.602362@weber.videotron.net>

Hot on the heels of release 0.9 is the 2nd beta release of the 
object-oriented embedded database library OOPS.

Intro)
OOPS stands for object-oriented persistent storage and it allows you to 
build & organize files using python objects (well, python object 
equivalents anyways). It is small, fast, easy to learn and use and scales 
well enough to utilize in real-world apps and projects. 

See what happens when you turn a very high level language (Python) into a 
very very high level language (Python + OOPS).

News)
-Many bug fixes thanks to a lot of feedback from my first release. Thanks 
everyone!
-The web site is now up with class docs at pyobjfs.sf.net
-Added a simple benchmark suite [benchmark.py] so I could keep track of 
bottlenecks (and show off the speed). Try it out & see for yourself!

Web Site: http://pyobjfs.sourceforge.net. 
Download: http://pyobjfs.sourceforge.net/download.html
Project Page: http://sourceforge.net/projects/pyobjfs/
License: LGPL
Mailing List: pyobjfs-users at lists.sourceforge.net; subscribe at 
http://lists.sourceforge.net/lists/listinfo/pyobjfs-users


Enjoy & Happy Hacking,

Dan Parisien <dan at eevolved.com>
Software Systems Consultant
eEvolved Real-Time Technologies


From sabren at manifestation.com  Thu Feb 14 12:33:49 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 14 Feb 2002 12:33:49 -0500 (EST)
Subject: RELEASED [OOPS - The Python Datastore v0.9] 
In-Reply-To: <JmGa8.5516$MR3.311561@wagner.videotron.net>
Message-ID: <Pine.LNX.4.44.0202141223310.9629-100000@mercury.sabren.com>

On Wed, 13 Feb 2002, Dan P wrote:

> OOPS achieves transparent persistence in the Python
> environment by mirroring the python namespace over top of
> a shelve-like database.  Behavioral methods were added to
> classes to make the bind to each other and receive extra
> identification information so each object could build a
> unique identifying database key in which it stores
> persistent data.


Hey Dan! I've been working on a very similar thing for some
time, except my stuff stores data in a relational object.
I've been working on revamping it to use metaclasses the way
you do - I love your "Definition" class.

Looking through the docs, I have a suggestion: I think you
could gain a lot by decoupling your "Obj" classes from any
particular type of storage.

One way to do this is through the Observer pattern. You
create another object that can fetch and store your
Obj-style classes, and once it fetches one, can listen to it
for changes.  Obj, on the other hand, would fire off events
whenever a property changes.

The benefit is that you can then use the psuedo-static
typing scheme you've created anywhere, including storing
them to a database, ZODB, or as models in a true 
model-view-controller scheme.

[I'm actually working on this sort of scheme.. if you're
interested, maybe we could work together.]

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From tim.one at comcast.net  Mon Feb 18 01:08:35 2002
From: tim.one at comcast.net (Tim Peters)
Date: Mon, 18 Feb 2002 01:08:35 -0500
Subject: Editable strings anyone?
In-Reply-To: <HFEKILOLEFEFMKAECNDLGEFIDDAA.jason@jorendorff.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEDMNNAA.tim.one@comcast.net>

[Jason Orendorff, on the regexp engine code]
> ...
> About halfway down, you'll see a function getstring() which returns
> a pointer to the searchable data in an object.  This is where the
> requirement of a *contiguous* buffer is enforced.
>
>   if (... buffer->bf_getsegcount(string, NULL) != 1)
>           ... then raise an exception.

More, if you scour the core for calls to bf_getsegcount(), you'll discover
that nothing in the core knows what to do if it doesn't return 1:  "multi
segment buffers" are of no use in the std distribution.  I was told they
were added because "somebody insisted" they were needed.  In return for
being unused, at least they confuse people <wink>.

guido-is-much-keener-on-"you-ain't-gonna-need-it"-these-days-ly y'rs  - tim




From wurmy at earthlink.net  Sat Feb 16 12:17:06 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Sat, 16 Feb 2002 17:17:06 GMT
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <m34rkhbvw3.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3C6E943B.BC7CCF9@earthlink.net>

"Martin v. Loewis" wrote:
> 
> Hans Nowak <wurmy at earthlink.net> writes:
> 
> > I manage to store that string just fine; only when I retrieve
> > the record, all the newlines are gone.
> 
> I can hardly believe this. Can you post a small script that
> demonstrates this problem?

That will be a bit difficult, because the XML comes back from
the server, but I'll try to emulate what it's doing:

#---begin---

from xml.dom.minidom import *   # who cares, it's only a test :-)

xml = """\
<root>
  <ticket foo="1" bar="2" image="
         one
         two
         three
         four" baz="4" />
</root>
"""

parsed_xml = parseString(xml)

print repr(parsed_xml.toxml())

for node in parsed_xml.getElementsByTagName("ticket"):
    print `node.getAttribute("foo")`
    print `node.getAttribute("image")`

#---end---

The multiline XML string is in roughly the same format as the
raw XML the server returns. That is, it does contain newlines
after 'one', 'two', and 'three' (assume these were in the
original multiline string). As soon as the XML is parsed,
these newlines disappear. Maybe this is normal and to be
expected when using xml.dom.minidom.parseString; I don't know.
Either way, my question is, is there a way to parse the XML
*and* somehow preserve the newlines in the string?

I have to do a comparison of the record data before posting
and the record retrieved from the database, and therefore the
data must be exactly the same. Unfortunately I don't have much
control over how the server returns its XML. :-(

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From greg at cosc.canterbury.ac.nz  Tue Feb 26 23:36:34 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 27 Feb 2002 17:36:34 +1300
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu>
Message-ID: <3C7C6252.FC779971@cosc.canterbury.ac.nz>

Fernando Pereira wrote:
> 
> This production cannot be disambiguated top-down with *any* lookahead, for
> the reason you suggest. Thus, it doesn't seem Python's grammar is LL(1).

Python's parser is an interesting beast. I haven't looked at
the internals of the parser generator, so I can't be sure,
but it seems that, although it may be using a recursive
descent algorithm at the production level, *within* a
production it must use some sort of regular expression 
matching engine.

Another thing is that all the items it collects while
parsing a given production are simply put into a tuple,
regardless of the internal structure of the production
used to parse it. Later on, while traversing the parse
tree, the compiler performs a *second* parsing operation
on the contents of the tuple, at which point it has
everything laid out before it, and can look ahead as
much as it wants.

For these reasons, saying just "Python's grammar is
LL(1)" is oversimplifying things considerably.

As for Nick Collier's problem, it seems likely that the
target parser generator isn't smart enough to cope with
the regular-expression-within-a-production thing that
the Python parser generator does. It should be possible
to transform it into something acceptable without too
much difficulty, though.

As it happens, I'm engaged in doing exactly that myself
at the moment, as I'm writing a recursive-descent LL(1)
parser for a superset of Python. I'm not using any parser
generator, just writing it by hand. I did all of the
expression stuff the other day, including the arglist
production, without any particular difficulty. So I 
know it can be done!

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


From wolfson at uchicago.edu  Wed Feb 20 19:48:55 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Wed, 20 Feb 2002 18:48:55 -0600
Subject: How do you conver output of "popen('ls')" to a list?
References: <3C737245.BC5D7164@macrozon.net> <20020220140806.A30148@ActiveState.com> <mailman.1014244477.1738.python-list@python.org> <3C7391F6.CB6CAE01@macrozon.net>
Message-ID: <pan.2002.02.20.18.48.54.828268.2561@uchicago.edu>

On Wed, 20 Feb 2002 06:09:26 -0600, husam wrote:

> os.listdir('.') gets you all the files in the '.' dircetory. This want
> work if i want to process some selected files. In contrast to this,
> "os.popen() allows you to pass arguments like "ls *txt" to retrieve only
> *.txt files.

In that case, you may want to try glob.glob().

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From rhymes at myself.com  Sun Feb 17 09:34:55 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sun, 17 Feb 2002 15:34:55 +0100
Subject: [Tkinter] Bitmap Image class
Message-ID: <vlfv6u0g37u7hfnud9f8cr02fuht3v70gk@4ax.com>

When i try to display a simple bmp (128*128, 16bpp)
in a window widget i get this error:

---------
malformed bucket chain in Tcl_DeleteHashEntry

abnormal program termination
--------

here is the code:

------
from Tkinter import *

root = Tk()

bmp = BitmapImage(root)
bmp.config(file="image.bmp")

root.mainloop()
-------

how can i fix this problem?

----
Lawrence "Rhymes" Oluyede
rhymes at myself.com


From grante at visi.com  Fri Feb 22 10:20:13 2002
From: grante at visi.com (Grant Edwards)
Date: Fri, 22 Feb 2002 15:20:13 GMT
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4>
Message-ID: <Nktd8.1458$N7.276578@ruti.visi.com>

In article <Xns91B6DFD6A36CDRASXnewsDFE1 at 130.133.1.4>, Philip Swartzleonard wrote:

>> Personally, I would like to see octal number entry migrated from 0---
>> to something like 0o---.  This would bring things more in line with
>> the entry of hex numbers (0x---)
>> 
>> I know it's only a minor thing, but it'd be nice if my numbers behaved
>> as I expected...
> 
> Steve has mentioned that UNIX people are used to it... 

I'm a Unix person, I'm used to it, and I still think it sucks.

> i think it's a little deeper than that, b'cause as far as i
> know, _every_ programming language that has ever supported
> octal does it this way...

And I still think it sucks.  :)

-- 
Grant Edwards                   grante             Yow!  FROZEN ENTREES may
                                  at               be flung by members of
                               visi.com            opposing SWANSON SECTS...


From phd at phd.pp.ru  Wed Feb 20 03:50:00 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 20 Feb 2002 11:50:00 +0300
Subject: command-line history problems
In-Reply-To: <71319333921.20020219165350@neosynapse.net>; from stevena@neosynapse.net on Tue, Feb 19, 2002 at 04:53:50PM -0500
References: <71319333921.20020219165350@neosynapse.net>
Message-ID: <20020220115000.E15274@phd.pp.ru>

On Tue, Feb 19, 2002 at 04:53:50PM -0500, Steven D. Arnold wrote:
> I'm using Python 2.2.  On a particular machine, command history isn't
> working.

   This particular installation does not include readline. Install
libreadline and recompile Python.

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



From webmaster at prankcallsunlimited.com  Sun Feb 17 18:25:41 2002
From: webmaster at prankcallsunlimited.com (Bob Pranky)
Date: 17 Feb 2002 15:25:41 -0800
Subject: FREE SOUND EFFECTS
Message-ID: <638cbd51.0202171525.6f5345ee@posting.google.com>

Free downloads of sound effects in lots of different categories at
http://www.a1freesoundeffects.com
Professional sounds also are created for all needs
http://www.a1freesoundeffects.com/service.htm


From carifio.nospam at nospam.usys.com  Tue Feb 19 19:23:31 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Wed, 20 Feb 2002 00:23:31 GMT
Subject: newbieish question
References: <mailman.1014157771.6596.python-list@python.org>
Message-ID: <70Cc8.26235$zJ3.1880339@typhoon.maine.rr.com>

Perfect. Change add one letter to my program and it suddenly works.
I guess the lesson is to read the documentation more carefully. Thanks.




From anton at vredegoor.doge.nl  Tue Feb 26 08:42:39 2002
From: anton at vredegoor.doge.nl (Anton Vredegoor)
Date: Tue, 26 Feb 2002 14:42:39 +0100
Subject: Anagram
References: <3c4e8efc.1036790726@news.mch.sni.de> <slrna7mijs.lfm.gaul@bifrost.cs.utexas.edu>
Message-ID: <a5g3j1$rbq$1@news.hccnet.nl>

On Tue, 26 Feb 2002 08:42:37 +0000 (UTC), gaul at spam.utexas.edu (Andrew
Gaul) wrote:

>I'm suprised no one else has posted this (it's standard material in a
>university probability course).
>
>def uniq_perm(l):
>    ''' returns the number of unique permutations of a list.
>
>                      n!
>            ------------------------
>            n_1! * n_2! * ... * n_r!
>
>        computes the number of unique permutations of n objects, of which n_1
>        are alike, n_2 are alike, ... n_r are alike.  Take from _A First Course
>        in Probability_, 5th edition, Ross, pg 5.
>    '''
>
>    acc = fac(len(l))
>    l.sort()
>    i = 0
>    while i < len(l):
>        cnt = l.count(l[i])
>        acc /= fac(cnt)
>        i += cnt
>    return acc
>
>def fac(n):
>    ''' returns n! '''
>    assert n >= 0
>    if n == 0:
>        return 1
>    acc = 1
>    for i in range(1, n+1):
>        acc *= i
>    return acc
>
>
>Python 2.2 (#5, Jan 23 2002, 16:23:03)
>[GCC 3.0.3] on linux2
>Type "help", "copyright", "credits" or "license" for more information.
>>>> from perm import *
>>>> uniq_perm(list('pepper'))
>60
>>>> uniq_perm(list('aa'))
>1
>>>> uniq_perm(list('aba'))
>3

The OP confused the issue by naming the thread anagram while asking
for something else. It's only natural that after having answered the
OP's question the thread should go on according to its title. I wrote
a voluminous script to generate anagrams some time ago, and I am
looking for options to shorten the algorithm to a few lines of code as
has been done for the algorithm the OP asked a question about.

Its at:

http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.html

or:

http://home.hccnet.nl/a.vredegoor/multiperm/multiperm.py



From jlh at cox.net  Sun Feb 10 14:29:14 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Sun, 10 Feb 2002 19:29:14 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <Xns91AF9500D5165RASXnewsDFE1@130.133.1.4>
Message-ID: <eSz98.4759$zb.260473@news1.east.cox.net>

"Philip Swartzleonard" <starx at pacbell.net> wrote in message
news:Xns91AF9500D5165RASXnewsDFE1 at 130.133.1.4...
> Aahz Maruch || Fri 08 Feb 2002 11:10:47a:
>
[...]
> Hm, i don't seem to remeber seeing this Pythonistas thing till maybe two
> days ago. Anyway, where did it come from-- based on some word? And HTF do
> you pronouce it? :)
Pythonista - Pie thun ee sta
I took it as a unintended play on Sandanista, which is a group of  "freedom
fighting" rebels in Nicaragua.





From joshm at taconic.net  Sun Feb 24 22:41:55 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sun, 24 Feb 2002 22:41:55 -0500
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com>
Message-ID: <3c79b01a_4@corp.newsgroups.com>

> I know that
> map(f,items)
> is equivalent to:
> for x in items:
>     f(x)

Actually, map does more than that.  map returns the results of f(x) as a new
list, of the same length as the original items.  If f() returns no value,
then you get a list of Nones.

>>> map(string.upper, ['dfbdfg','sdfgberg','dfbsadg','dfbdfag'])
['DFBDFG', 'SDFGBERG', 'DFBSADG', 'DFBDFAG']

> But what is the functional equvalent of:
>
> for x in items:
>     x.f()

List comprehensions make easy work of this:

>>> [ x.upper() for x in ['dfbdfg','sdfgberg','dfbsadg','dfbdfag'] ]
['DFBDFG', 'SDFGBERG', 'DFBSADG', 'DFBDFAG']

so in your example, "[ x.f() for x in items ]".  Cool, eh?
--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From plipp.plopp at flipp.com  Sun Feb 17 17:51:01 2002
From: plipp.plopp at flipp.com (Mikke M.)
Date: Sun, 17 Feb 2002 23:51:01 +0100
Subject: Tkinter window on top without keyboard focus?
Message-ID: <utWb8.6425$qZ3.11822@news1.bredband.com>

Hi!

Does anyone have a method of placing a window 'always on top' without
letting it have the keyboard focus?

/mikke.




From sameer at cs.uoregon.edu  Tue Feb 26 13:29:15 2002
From: sameer at cs.uoregon.edu (Sameer Suresh Shende)
Date: Tue, 26 Feb 2002 10:29:15 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <0ebb01c1beef$a09efe00$0500a8c0@boostconsulting.com>
Message-ID: <Pine.SGI.4.33.0202261024390.1311-100000@krypton.cs.uoregon.edu>

Hi David,
	This looks like a bug in the IL Analyzer that is easily fixed.
I'll look into this. We cannot distribute the sources of the IL Analyzer
as they have routines that walk the intermediate language AST that are
EDG proprietary. Our CREDITS file in PDT lists all contact e-mail
addresses, but due to some recent security issues, we were advised to
remove any personal information from the LANL webpages.
	If the code compiles on different OSes (perhaps under Linux), I
wonder if you can parse it using PDT under Linux properly. We'll look into
this rightaway and provide you with a bug fix soon.
	Thanks for your interest in SILOON and PDT.
	- Sameer


 On Tue, 26 Feb 2002, David Abrahams wrote:

>
> ----- Original Message -----
> From: <rasmussn at lanl.gov>
>
> > On Monday, February 25, 2002, at 05:32 PM, David Abrahams wrote:
> >
> > > I've been interested in systems for automatically wrapping C++ and
> > > generating documentation for a while, so I downloaded the PDT (the
> parser
> > > toolkit used by SILOON) and tried it on one of my source files. After
> lots
> > > of hacking, I was able to get the EDG front-end component to parse my
> > > source
> > > and generate an intermediate language (.cil) file, but the PDT
> executable
> > > which is supposed to interpret the .cil and produce a high-level
> > > description
> > > of the code crashed with an assertion:
> >
> > SILOON uses scripts to run the PDT and we've worked hard to make it easy
> > to use.  Essentially all you have to do is edit a file (that is included
> > in a Makefile) and point macro variables to the location of your source
> > files (.h .cpp ...).  I would be interested to know if you have similar
> > problems using the EDG front-end from SILOON directly.
>
> Well, I'm less interested in SILOON itself than in the PDT, but I might try
> it if I get some more time to waste ;-)
>
> > > Assertion failed: ptr->assoc_template, file c:\documents and
> > > settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-
> > > tau\il_display_tau.
> > > c, line 3047
> > >
> > > Since the PDT doesn't come with source, there was no way for me to
> debug/
> > > fix
> > > the problem.
> >
> > The PDT comes with sources for everything (almost) but the EDG front end.
>
> There's no "il_display_tau.c" in either the .zip or the .tar.gz.
>
> > What PDT version were you using?
>
> The very latest I could find at
> http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/.
>
> > Would you like to send your source files to me directly
> > and I can
> > look at the problem, although if it is windows-specific code, I can't help
> > you
> > as I don't have access to Microsoft compilers.
>
> It's code that compiles on 7 different compilers including several different
> EDG-based compilers on four OSes, so it shouldn't be a problem. If you get
> the current boost cvs state (see http://sourceforge.net/cvs/?group_id=7586)
> and a copy of Python-2.2, you can try for yourself. The file is
> <boost-root>/boost/python/class.hpp.
>
> > (On further examination, it looks like the PDT doesn't have source for
> > taucpdisp,
> > I'll ask the developers why not.)
> >
> > I can provide access to a prerelease version of SILOON which uses the
> > latest
> > version of the PDT, which may fix your problems.
>
> If you can get this to work, I might be interested in seeing the results.
>
> > > Furthermore, the PDT developers don't seem to want to be reached. Their
> > > contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/)
> > > says:
> > >
> > >     To contact the developers of PDT, please use the "Comments" section
> in
> > > the download page.
> > >
> > > But of course, there is no "Comments" section on the download page. :=(
> >
> > I've forwarded you comments on to the PDT developers so that they can fix
> > their web pages and comment personally on the core dump.
>
> Thanks,
> Dave
>
>





From tim.one at home.com  Thu Feb 14 00:07:23 2002
From: tim.one at home.com (Tim Peters)
Date: Thu, 14 Feb 2002 00:07:23 -0500
Subject: Python License Issues
In-Reply-To: <20020213191640.13956.qmail@mail.yifansoft.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECINMAA.tim.one@home.com>

[christoforou at gaaros.com]
> Hello from rainy London

Rain!  What a concept <wink>.

> We have an issue with the python license which basically boils
> down to this:

You're going to have to ask your lawyer/solicitor for legal opinions.

> If we ship an application that uses python, is the user of our
> application bound by the python license agreement (PSF)?

My understanding (and I am not a lawyer) is that, with respect to the Python
portion of the application you ship, yes.  With respect to your portion, no,
they're bound by whatever license you add to the stack.  You can add
additional protections to the Python portion of your work, if you want to,
but there's nothing you can do to make the PSF magically offer protections
to your users that it explicitly refused to offer to you.

> It seems so given clause 8 of the PSF:
>
>    8. By copying, installing or otherwise using Python 2.1.2, Licensee
>       agrees to be bound by the terms and conditions of this License
>       Agreement.
>
> If such is the case we will encounter problems with clients that
> do not like clause 4 of the PSF which states:
>
>  4. PSF is making Python 2.1.2 available to Licensee on an "AS IS"
>     basis.  PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR
>     MPLIED.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND
>     DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR
>     FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON
>     2.1.2 WILL NOTINFRINGE ANY THIRD PARTY RIGHTS.
>
> Some clients demand a certain level of 'protection' from
> situations of gross misconduct, say python has stolen major chunks of
> code from other apps and thus exposing them to huge legal risks.

The Python license is par for the open-source course here:  since Python is
distributed without charge, PythonLabs has no revenue, so we either supply
the work as "use at your own risk", or not at all.  Give the PSF enough
money (say, equivalent to the GNP of a small country or two <0.9 wink>), and
I'm sure a different licensing arrangement could be worked out.

> We can consider accepting liability on all third party code we use, for
> such gross misconduct situations but is it technically possible to do so
> given the PSF license?

Actually, I don't understand your problem here.  The PSF license is a
contract between the end user and the PSF.  It (and unlike the GPL) says
nothing about additional terms you may want to add in *your* license.  If
you can consider accepting liability (and bless you, if so), fine, do so,
and that's between you and *your* users.  Clause 4 of the PSF license says
the *PSF* makes no representations etc (it does not say that you can't), and
that's the plain and unalterable truth.  It doesn't prevent you from
accepting more responsibility than the PSF is able to accept.

> Ideally we would like to negotiate a single license agreement with
> these clients (our license agreement) which provides additional
> liability over and above to that which PSF provides. However given
> clause 4 it seems that the client *must* directly be bound by the PSF
> license.

I believe that part goes much deeper than clause 4.  You're seeking to
*re*license Python, and nothing in the Python license grants you the power
to do so (clause 4 isn't really relevant to this issue).  Indeed, nothing in
the CNRI license (which the PSF inherited) grants the PSF the power to
relicense the CNRI portion of the work either; the PSF license is stacked on
top of the BeOpen license, which is stacked on top of the CNRI license,
which is stacked on top of the CWI license.  You would have to negotiate
with all these entities to let you relicense their portions of the work, if
you wanted to rid your product of their licenses.

It's hard to see a real need for that, though:  CWI, CNRI, BeOpen and the
PSF simply aren't going to accept liability no matter what (e.g., BeOpen is
out of business, and the assets of the PSF currently consist of one book on
forming a non-profit corporation, which the Treasurer paid for out of his
own pocket <wink>).  Their licenses simply say so up front, and don't
preclude you from adding whatever additional terms suit your business model.

if-someone-wants-to-sue-the-psf-for-the-book-the-psf-would-declare-
    bankruptcy-ly y'rs  - tim




From joshm at taconic.net  Tue Feb 26 18:57:02 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Tue, 26 Feb 2002 18:57:02 -0500
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net> <3c7c1fce.35698011@news.la.sbcglobal.net>
Message-ID: <3c7c1e6b_3@corp.newsgroups.com>

Perhaps it would be better to use an iterator on the file, ala:

file = open(...)
for line in file:
    # check to see if this is the proper line, etc

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From anon at anon.com  Mon Feb 11 22:59:57 2002
From: anon at anon.com (Andrew)
Date: Tue, 12 Feb 2002 03:59:57 GMT
Subject: wxPython wxListCtrl
Message-ID: <1r0a8.14824$Cg5.951049@news1.calgary.shaw.ca>

Hello,

I have a question about the wxPython wxListCtrl.  I was wondering if there
is a way to insert a combo box for the items in a column (instead of the
edit controls that are used).  If there is no easy way, what approach will I
need to accomplish this?

Thanks,
Andrew




From theller at python.net  Fri Feb 22 14:54:32 2002
From: theller at python.net (Thomas Heller)
Date: Fri, 22 Feb 2002 20:54:32 +0100
Subject: subclassing float (or not)
Message-ID: <a567mp$4vkk5$1@ID-59885.news.dfncis.de>

I'm currently porting my own rewrite of Konrad Hinsen's PhysicalValue
module to Python 2.2.
Basically a PhysicalValue instance is a float conbined with a unit.
Should I subclass float or should I not? What would I gain by doing
this (except maybe some memory savings when I use __slots__)?

Thomas




From hst at empolis.co.uk  Wed Feb 27 11:47:31 2002
From: hst at empolis.co.uk (Harvey Thomas)
Date: Wed, 27 Feb 2002 16:47:31 -0000
Subject: Equations with Regular Expressions
Message-ID: <8FC4E7C302A6A64AAD5DB1FA0E825DEB220AE9@hendrix.empolisuk.com>

try the following (extended to allow for x^n)

import re

def dosub(mg):
    if mg.group(1):
        return '%s*(%s**%s)' % (mg.group(1), mg.group(2), mg.group(3))
    else:
        return '%s**%s' % (mg.group(2), mg.group(3))

regex = re.compile('(\d*)([a-zA-Z])\^(\d+)')
print regex.sub(dosub, '2x^3-2x^2+15-x^4*6')


-----Original Message-----
From: Alves, Carlos Alberto - Coelce [mailto:calves at coelce.com.br]
Sent: 27 February 2002 10:33
To: python-list at python.org
Subject: Equations with Regular Expressions


Hi everybody, 
Could someone of you give me and example of how can I translate '2x^3-2x^2+15' into '2*(x**3)-2*(x**2)+15' using regular expressions.
Carlos Alberto 
COELCE/DPRON-Departamento de Projetos e Obras Norte 
Fone: 677- 2228 
e-mail: calves at coelce.com.br 
\|||/ 
(o o) 
--ooo0-(_)-0ooo-- 

_____________________________________________________________________
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

_____________________________________________________________________
This message has been checked for all known viruses by the MessageLabs Virus Scanning Service.



From db3l at fitlinxx.com  Tue Feb 12 22:30:36 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Feb 2002 22:30:36 -0500
Subject: Re-throwing an exception without loosing the stack trace
References: <20020212174235.A17928@doublegemini.com> <15465.39126.250327.239174@beluga.mojam.com> <mailman.1013554227.2313.python-list@python.org>
Message-ID: <ulmdym52b.fsf@ctwd0143.fitlinxx.com>

"Clark C . Evans" <cce at clarkevans.com> writes:

> Yes.  I didn't try using raise without any arguments.  Cool.
> Funny, this isn't in Learning Python... and raise all by
> itself isn't listed in the syntax in the book.  Hmm.

Not sure about the book, but it was added to the Tutorial (section
8.4) in Python 2.2.  It's also always been in the language reference
(section 6.8) at least back to the 1.5.2 copy I have, but admittedly
that's not necessarily one of the documents people might typically go
to.

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


From emile at fenx.com  Thu Feb 14 11:53:54 2002
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 14 Feb 2002 08:53:54 -0800
Subject: MemoryError using NumPy
Message-ID: <a4gqd9$68e5$1@ID-11957.news.dfncis.de>

Win2k-sp2, Zope2.4, Numeric 20.1.0, Python2.1, 1.5Gb memory.

I'm slinging some large int32 numeric arrays, and at one point do:

Numeric.sum(Numeric.take(myDataArray, indices)).astype(Numeric.Float64)

This causes a temporary jump in memory usage (from ~800Mb to ~1.2Gb)
before settling down again, but it delivers the speed I'm looking for
(~7 secs vs ~30 summing in loop).  All starts well, but soon I get
MemoryError "can't allocate memory for array" only when indices is
large.  I suspect that memory has fragmented such that a sufficiently
large contiguous block is not available to set up the array.  Is this
the likely cause, and if so is there a way to un-fragment memory?  Or
might there be an alternative to Numeric.take that gets to the same
place?

I haven't tested this on the linux side, as the target system is Win2k I
need it to work here.  If it'll help though, I can do that.

Regards,

--

Emile van Sebille
emile at fenx.com

---------



From jeff at ccvcorp.com  Mon Feb 11 20:29:43 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 11 Feb 2002 17:29:43 -0800
Subject: OOP Newbie mixing COM, wxPython, BOA
References: <u6ddfmsu6ug32@corp.supernews.com>
Message-ID: <3C687007.4ACF94EE@ccvcorp.com>


Mojo wrote:

> The subject line could easily have been "fool in way over his head..."
>
> That said, I got farther than I thought possible only to be stymied by what
> I'd guess is either an idiom I don't understand (being an OOP newbie) or a
> question of variable scope/namespace...
>
> My problem is that my app has two classes (a class encapsulating the GUI,
> and a class that implements an event handler for a COM object.)  I don't
> know how to access GUI objects in the former, from code in the latter.

....

> How do I call the "AddRoot" method of the "buddyTree" object in the VpFrame
> class, from the separate class "JabberEvents"?  Any help, pointers,
> explanations greatly appreciated.

This is pretty much a namespace issue, and has nothing to do with COM, and only
indirectly with wxPython/Boa...

What you should do, is instead of making the Jabber object be global to your
VpFrame module, you should encapsulate its creation into a function.  Have your
frame call this function in its __init__(), and then save the reference on the
frame.  At this point, you will also want to stick a reference to the frame onto
the Jabber object.  Something like this:

def InitializeJabber():
    jab = win32com.client.DispatchWithEvents("JabberCOM.JabberSession",
JabberEvents)
    return jab

##  in class VpFrame...
        ...
        def __init__(self, parent):
            self._init_ctrls(parent)
            self.jabber = InitializeJabber()
            self.jabroster = self.jabber.Roster
            self.jabber.frame = self

Note that you'll need to change all your references to 'jab' and 'jabroster' to
use this saved reference.

Now, in any of your Jabber object's methods, you can do

        self.frame.MyFrameEventHandler(mydata)

A few design tips to keep in mind--you want to have the minimum possible number
of connections between your frame and the jabber object.  Don't have the jabber
object do each step needed when you connect, for instance;  write a VpFrame
method called OnConnect() that handles everything the GUI needs to do, and have
jabber call that single method.  Similarly for any other events that you receive
from Jabber--the Jabber object handles the protocol side of things, and then
tells the frame object to handle the GUI side of things.  In fact, you might
look at defining custom wxPython command events, and having the Jabber object
notify the GUI by posting those events. (This is a somewhat advanced subject,
which essentially does the same thing--calls a specified method--but allows the
GUI side a lot more flexibility in determining where and how to handle the
event.  This could really pay off, though, if you're intending on expanding the
application considerably.)

Also, by having both objects holding references to each other, we've created a
reference cycle.  Python's garbage collector should be able to handle this okay,
but just in case, it's probably a good idea to explicitly destroy the Jabber
object in your frame's OnCloseWindow() handler (or whatever name is given to the
method that shuts down your frame).  But it's a common idiom in wxPython for
parent and child windows to each have references to the other.  (Typically, the
wxWindows framework will clean up child windows for you, but obviously wxWindows
knows nothing of your Jabber object...)

Jeff Shannon
Technician/Programmer
Credit International




From SBrunning at trisystems.co.uk  Thu Feb 14 05:08:32 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 14 Feb 2002 10:08:32 -0000
Subject: Python License Issues
Message-ID: <31575A892FF6D1118F5800600846864DCBCC4E@intrepid>

> From:	Tim Peters [SMTP:tim.one at home.com]
> > We have an issue with the python license which basically boils
> > down to this:
> 
> You're going to have to ask your lawyer/solicitor for legal opinions.
 
OK, let's forget about the legal stuff for the moment.

More importantly, good manners. Let's say I'm wrapping up a Python program
with py2exe, and I'm going to distribute it. Which files would I be expected
to include as a matter of courtesy to the Python developers (and the
developers of any additional modules that I use)? Just the licence files, or
is there anything else?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From andy47 at halfcooked.com  Thu Feb  7 18:32:27 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Thu, 7 Feb 2002 23:32:27 +0000 (UTC)
Subject: 2001 Turing Award
References: <mailman.1013114112.28909.python-list@python.org>
Message-ID: <Xns91AF6B22CC852andy47halfcookedcom@203.109.250.24>

"Mark McEahern" <marklists at mceahern.com> wrote in 
news:mailman.1013114112.28909.python-list at python.org:

> Keith Keller wrote:
>> Damn--I thought the award was for the person who best impersonated a
>> Turing machine.  :) 
> 
> Do you have any nominees?
> 
> // mark
> 

I believe there is currently only one Tim-bot.

We could turn it into a chant if you want

There's-only-one-Timmy-bot-bot,there's-only-one-Timmy-bot-bot-ly y'rs,
Andy
-- 
Contents free posts a speciality


From cbbrowne at acm.org  Sun Feb 10 17:54:22 2002
From: cbbrowne at acm.org (Christopher Browne)
Date: Sun, 10 Feb 2002 17:54:22 -0500
Subject: (Long) Re: Autocoding project proposal.
References: <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <3c66ac80.2522937@127.0.0.1> <2022.806T252T8483762threeseas@earthlink.net>
Message-ID: <m3g049j6ch.fsf@chvatal.cbbrowne.com>

"Timothy Rue" <threeseas at earthlink.net> wrote:
> On 10-Feb-02 12:19:18 (Five Fresh) Fish <root at 127.0.0.1> wrote:
>>On Sun, 10 Feb 2002 15:51:51 GMT, "Timothy Rue" <threeseas at earthlink.net>
>>wrote:
>
>>>Funny how Hitler chased Einstein and Fermi out of his territory, their
>>>homeland.....
>>
>>Does this mean this thread can be ended, now?

> Since you are not existing in the public archives...
>
> Einstien and Fermi leaving their homeland didn't end the war.

And evidently it doesn't mean that the thread gets ended...
-- 
(reverse (concatenate 'string "moc.adanac@" "enworbbc"))
http://www.ntlug.org/~cbbrowne/finances.html
DVD boycotts.  Are you doing your part?
"Now they can put you in jail if they *THINK* you're gonna commit a
crime.  Let me say that again, because it sounds vaguely important"
--george carlin


From killspam at darwinwars.com  Tue Feb 19 05:00:26 2002
From: killspam at darwinwars.com (Andrew Brown)
Date: Tue, 19 Feb 2002 10:00:26 GMT
Subject: confusion regarding os.path.walk()
References: <mailman.1014078032.30409.python-list@python.org>
Message-ID: <Xns91BA65E93C0ECandrewdarwinwarscom@194.168.222.19>

Steven Majewski <sdm7g at Virginia.EDU> wrote in 
news:mailman.1014078032.30409.python-list at python.org:

> 
> You need to os.path.join() the dirname to the filename and do your
> test on that. The filenames are relative to the directory arg,
> not to the process current directory.
> 

I had a similar problem yesterday: I wanted to rename all the files on my 
site consistently (no spaces, no uppercase, all to end .html) and then -- 
so as not to break search engines -- to symlink the new names to the old 
names. This will let me slowly fade out the old names.

Part 1 went fine, and left me with a text file recording the changes, 
which could then be reconstructed as a dictionary of file==>symlink 
needed pairs. But when I came to do the second part, with os.path.walk, I 
found that all the calls to make symlinks produced broken links with or 
without os.path.join(). I had to put into the visitor function a call to 
os.chdir(os.curdir) and then make the symlink with os.system
(linkcommand). ("linkcommand" was a previously constructed string like 
'ln-s newname oldname')

I found later that I could have used os.symlink() instead of the call to 
os.system() but I'm still puzzled that you can't apparently create 
symlinks in a directory that's a subdirectory relative to where you are. 
Is this a unix/linux feature or a python one?

The practical question I was left with was "how do you identify broken 
symlinks with python?" I'd have liked a script that rm-ed broken and only 
broken symlinks, and I can't figure out how to test for them. 
os.path.islink(file) will tell me whether it's a symlink. But is there a 
call to say what it's supposed to point at? Then I can test whether that 
exists.


From danielk at aracnet.com  Tue Feb 19 09:53:26 2002
From: danielk at aracnet.com (Daniel Klein)
Date: Tue, 19 Feb 2002 06:53:26 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
Message-ID: <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com>

Let me reverse the question: What do you think the response would be if Sun
was asked to drop static type checking (as well as checked exceptions).

When you can answer this, then you'll have _your_ answer.

Daniel Klein


On 18 Feb 2002 15:31:47 -0800, ajeru at vknn.org (Alexander Jerusalem) wrote:

>Has anyone heard of plans to add static type checking to Python? I
>like Python very much but I don't use it for bigger projects because I
>fear the testing effort. For example I just refactored a rather large
>Java codebase. And when I change a method name for example, all I have
>to do is run everything thru the compiler and it will show me all
>places in other classes where this method is called, so I can go there
>and chenge the method call as well.
>
>In Python I'd have to test the whole software suite including each and
>every branch in each and every if statement to be sure that I didn't
>forget something. To build such a test suite would take almost as long
>as it takes to build the whole software and I'd have to maintain the
>test suite as well. That's just unafordable.
>
>On the other hand, Python is the best language that currently exists
>and I would like to use it on bigger projects not just for prototyping
>as some suggest.
>
>Regards,
>Alexander Jerusalem



From SBrunning at trisystems.co.uk  Tue Feb 19 04:31:19 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 19 Feb 2002 09:31:19 -0000
Subject: comp.lang.python in English? (was Re: Why Python is like BASI
	C)
Message-ID: <31575A892FF6D1118F5800600846864DCBCC92@intrepid>

> From:	Tim Peters [SMTP:tim.one at home.com]
> Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> czignat?  Classifero en bragna!
> 
> tims-speak-their-own-language-ly y'rs  - just another tim
 
Far more comprehensible than *some* Tims we could mention...

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From fredrik at pythonware.com  Fri Feb  8 03:04:26 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 08 Feb 2002 08:04:26 GMT
Subject: Why it does NOT work on Linux ?
References: <mailman.1012871110.28776.python-list@python.org> <Vsv88.94274$KM2.3404324@atlpnn01.usenetserver.com>
Message-ID: <eEL88.24195$l93.4429576@newsb.telia.net>

Steve Holden wrote:

> > Ladislav's use of findall works, but this feature of re.findall()
> > is not documented and should be avoided, as undocumented features
> > have been known to disappear.
>
> Nonsense. The function interface continues to be a fully-documented part of
> the re module.

that's not what Jason was talking about:  he pointed out that the
original code passed a compiled pattern object, not a string, as the
first argument to re.findall.

> You are apparently confusing
>
>     http://www.python.org/doc/current/lib/node99.html
>
> (the re module contents description, which describes the functional
> interface to regular expressions) with
>
>     http://www.python.org/doc/current/lib/re-objects.html
>
> (the method descriptions for re objects). It may not be the way *you* do it,
> but that doesn't mean it's not documented <wink>.

where does the documentation say that the "pattern" argument to
re.findall can be either a string or a compiled pattern?

</F>




From pinard at iro.umontreal.ca  Tue Feb 12 14:02:55 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 12 Feb 2002 14:02:55 -0500
Subject: PyFT and PIL
In-Reply-To: <20020212173402.G8503@phd.pp.ru>
References: <mailman.987996754.9967.python-list@python.org>
	<9c11bn$sph@gap.cco.caltech.edu>
	<oqu1sm4vqr.fsf@titan.progiciels-bpi.ca>
	<20020212173402.G8503@phd.pp.ru>
Message-ID: <oqofiu4j6o.fsf@titan.progiciels-bpi.ca>

[Oleg Broytmann]

> On Tue, Feb 12, 2002 at 09:31:40AM -0500, Fran?ois Pinard wrote:

> > PyFT seems to be a wonderful thing.  But the `README' file roughly says:
> > 
> >    `Compilation of the extension is standard.'
> > 
> > The only standard I know is `Distutils'.

>    There was older Python standard using Setup.in and Makefile.pre.in
> files:

>    make -f Makefile.pre.in boot
>    make

Wow!  PyFT is now running here.  Wonderful! :-) Thanks a lot for the hint!

P.S. - There was no `Makefile.pre.in' in the distribution, but there were
a few others here and there on my disks, it was enough to copy one of them.

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

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



From gcordova at hebmex.com  Wed Feb 27 10:28:54 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 27 Feb 2002 09:28:54 -0600
Subject: Extremely Newbie wants to program and pleads to be pointed in
	 the right direction
Message-ID: <19650FF08A99D511B19E00B0D0F06323060B0F@SRM201002>

> 
> In the olden days I played around with BASIC
> just for fun, writing trivial games and making patterns move on the 
> screen.
> 

Ahhh... I remember my C64... *snif*

>
> Can I do that sort of thing with Python?
>

Sure you can, but Python doesn't have any "native" graphics
mode, it uses an external module available in many
distributions though.

> 
> I've at last got a new machine, running XP Home.
> Will this be OK?
> Or would it be wiser to use my old 200MMX?
> 

Hmmm... well... I'll not comment on the wisdom
of having/using WinXP, but yes, you can use that
without any problem.

Use the old 200MMX to play around with linux
though, and you'll have lots of fun.

>
> Before I start wrestling with Python - and with thinking - and with 
> making my brain hurt - is Python the best language for this sort of 
> play?
> Would Java be better?
> Will I be able to make a ball bounce around inside a box etc.?
> 

Ohh, definitely. Python lends itself to the "playing" mode of
learning just fine. Better than Perl, MUCH better than Java.

Since you're going to want to do some animation and
bouncy graphics stuff, look for a module called "PyGame"
(I think) which supports this stuff.

I truly don't know if PyGame is available for Windows,
I do know it's available for Linux.

It's a module which has all that bouncy graphics stuff.
Damn... I wish I'd had python back in my Amiga days...

*snif*

>
> Please be gentle!  I'm not a rocket scientist - but rather just an 
> ordinary bloke who's really quite interested in how computers 
> work and who would like to edge sidelongly towards greater technological 
> intimacy!
>

Don't worry. Thankfully, Python isn't rocket surgery,
nor brain science.

Even the "brain twisting stuff" which the oldsters like
to throw around isn't that hard, once you lose respect
for classes and types as "special objects" and start
treating them as ordinary objects.

>
> -- 
> Chris
>

Good luck Chris, you picked a nice language to play
around with.

In case you want to do a bit more than only animation
and playing around, check out a package called Blender;
it's a 3D and animation designer, with scripting and
other game stuff thrown in.

It's EXCELLENT, the site (google is your friend) has
some examples and demos, really nice, very cool.

All scripting is done with python, so you can imbue your
onscreen objects with behaviours and other stuff; you
can use Python in almost all parts of the action, really.

Good luck. :-)

-gustavo



From bkelley at wi.mit.edu  Mon Feb 25 10:40:04 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Mon, 25 Feb 2002 10:40:04 -0500
Subject: Dictionary type access to list of instances
References: <a5dcgh$b22$1@ctb-nnrp1.saix.net>
Message-ID: <3C7A5AD4.9090404@wi.mit.edu>

Grant Beasley wrote:

> Hi
> 
> If I have a list of instances, each with a property which I want to use as a
> key to access the list in dictionary-like manner, what would be the best way
> of doing this?
> 
> eg.
> func get(list, property, value):
>     for x in list:
>         if getattr(x, property) == value:
>             return x
> 

In this usage, getattr will raise an Attribute error if property doesn't 
exist which might not be what you want.

Here is an attempt using list comprehensions and the built-in list.index 
function.  I have created a NotEqual class so that getattr can return a 
default value is never equal to any other instance.  This is because 
None might be a valid value for a property.

class NotEqual:
     """instances of this class are never equal to anything"""
     def __eq__(self, other):
         return 0

def get(inputList, property, value, default=NotEqual()):
     try:
         properties = [getattr(x, property, default) for x in inputList]
         return inputList[properties.index(value)]
     except ValueError:
         raise ValueError, "attribute %s with value %s not found"%(
             `property`, `value`)

if __name__ == "__main__":
     class Foo:
         def __init__(self):
             self.dog = 1

     foo = Foo()
     print get([foo], 'dog', 1)
     try:
         get([foo], 'dog', 2)
     except ValueError, msg:
         print 'error caught:', msg

     try:
         get([foo], 'cat', 'raisin')
     except ValueError, msg:
         print 'error caught:', msg



From donn at drizzle.com  Thu Feb 21 23:42:21 2002
From: donn at drizzle.com (Donn Cave)
Date: Fri, 22 Feb 2002 04:42:21 -0000
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org> <7xit8rhebq.fsf@ruckus.brouhaha.com> <slrna79vo7.e5o.yoda@homer.interx.com>
Message-ID: <1014352937.736683@yasure>

Quoth yoda at dagobah.org (Jedi Master Yoda):
| On 20 Feb 2002 16:24:25 -0800, Paul Rubin <phr-n2002a at nightsong.com> spouted:
|> Maybe there's a principle to be found here: "design by FAQ".  If newbies
|> keep arriving and asking over and over again "how do I do X?", maybe
|> there's a good reason for having a way to do X.  We've seen that already
|> for augmented assignments.
|
| That's how Perl got to be Perl.
|
| "Perl is worse than Python, because people wanted it worse." - Larry Wall.

nail.head.hit(1)

	Donn Cave, donn at drizzle.com


From jason at jorendorff.com  Fri Feb  1 23:32:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Fri, 1 Feb 2002 22:32:57 -0600
Subject: Making a list-inhereted object return standard repr?
In-Reply-To: <Xns91A8CF7752ED0RASXnewsDFE1@130.133.1.4>
Message-ID: <HFEKILOLEFEFMKAECNDLIENPCPAA.jason@jorendorff.com>

Philip Swartzleonard wrote:
> Ok, so i decided to make my container class inherit from list, [...]
> I want to change __repr__ and __str__ to do something more like
> the first case again [...]

class Foo(list):
    def __repr__(self):
        basic = object.__repr__(self)
        values = list.__repr__(self)
        return basic[:-1] + ' containing ' + values + '>'

## Jason Orendorff    http://www.jorendorff.com/



From sjmachin at lexicon.net  Sat Feb 23 18:56:03 2002
From: sjmachin at lexicon.net (John Machin)
Date: 23 Feb 2002 15:56:03 -0800
Subject: Puzzled: Y am I ending up with extra bytes?
References: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4>
Message-ID: <c76ff6fc.0202231556.3fcea99f@posting.google.com>

"A.Newby" <deathtospam43423 at altavista.com> wrote in message news:<Xns91BF3BEAA45A2AusYourStandingInIt at 130.133.1.4>...
> Why is this happening? I read large chunks of data from a text file, 
> according to byte locations specified on another file, and for some reason, 
> this function (below), spits out a few extra bytes.
> 
> Here's the code, as entered into the Python shell...... 
> 
> 
>     	index = map(string.rstrip, open('D:\cgi-bin\indx.txt').readlines())
>     	#this opens the index file, which has precise byte locations of each 
>     	#chunk of data I want to extract from the log.txt file, and turns it 
>     	#into a list.
> 
> 
>     	def fish(end, start, deduct):
> 	    	chat = open('D:\cgi-bin\log.txt', 'r')
> 	    	g = int(index[end]) - int(index[start])

This seems to assume that chunks in the log.txt file are contiguous.
E.g. 
log.txt contains
chunk1RLlonger-chunkRLthird-chunkRL
0....v....1....v....2....v....3....v
[R = carriage return 0x0D, L = line feed 0x0A]
and indx.txt contains
0
8
22
35 ######### Better have an EOF sentinel, otherwise you are cactoid on
the last chunk.

and you expect:
fish(1, 2, 0) to give "longer-chunk\n"
and fish(0, 2, 0) to give "chunk1\nlonger-chunk\n"

Is this so? It is not difficult to see where actual might differ from
expected by 1 byte per line. seek() will position you properly, but
read(n) in text mode will pull in more than you "expect".

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> f = open("foo.txt", "rb")
>>> f.seek(0); f.read()
'01234567\r\nabcdefgh\r\nABCDEFGH\r\n\r\n'
>>> f.seek(0); f.read(20)
'01234567\r\nabcdefgh\r\n'
>>> f = open("foo.txt", "r")
>>> f.seek(0); f.read(20)
'01234567\nabcdefgh\nAB'
>>>

However an allegedly semi-constant "about 4" is a puzzler. Are most
chunks 4 lines????? I suggest that you set up a *SMALL* test set like
the above, and try that. If you still have a problem, then you can
post the contents of your test data files, and (in some precision)
what the result was.

You will probably have to open the log.txt file in binary mode, and
strip out any "\r" you find when you read from the seek position. This
would still work on *x, but go hopelessly wrong on a Mac. Better:
first change "\r\n" to "\n", then change "\r" to "\n".

> 	    	chat.seek(int(index[start]))
> 	    	print chat.read(g - deduct)
> 
> Now, if I enter the following into the command line ...
> 
> fish(205, 204, 0)
> 
> ,,, I get about four extra characters. God knows why! So that's why I try 

*ABOUT* four or *EXACTLY* four??? Are they the first ~four characters
from the next chunk?

> ...
> 
> fish(205, 204, 4) 
> 
> ... And it seems to work perfectly. I can even "fish" up to about about ten 
> "index" lines with it.

Your meaning is unclear here. Show the Python code.

> But as soon as I try and fish out any more than 
> that, I get the dreaded extra 4 bytes of code again. Why?

... and here too.

> 
> Now I know what you're thinking. You're thinking that perhaps my index file 
> is corrupted, and hasn't got an accurate account of precisely what's in the 
> log file. I suspected that might be the case myself, but ... when I try 
> fishing out each chunk of data individually, it works fine! I can even do 
> this ...
> 
> for x in range(1, 90):
> 	fish(211+x, 210 + x, 4)
> 
> ...... without ending up with that extra data I don't want. However, this 
> method is too slow for my purposes.

Not surprising, seeing fish() is opening the log.txt file each time
you call it!

> Plus, I really wanna know what it is 
> that's going wrong.
> 
> Can anyone spot it?
> 
> BTW, if you're curious, this is part of a chat script I'm putting together. 
> 
> 
> cdewin at dingoblue.net.au


From nas at python.ca  Sun Feb 10 18:49:43 2002
From: nas at python.ca (Neil Schemenauer)
Date: Sun, 10 Feb 2002 15:49:43 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
In-Reply-To: <NCBBLFCOHHDIKCAFGCFBAEAGKCAA.marklists@mceahern.com>; from marklists@mceahern.com on Sun, Feb 10, 2002 at 03:02:30PM -0800
References: <20020210145042.A25774@glacier.arctrix.com> <NCBBLFCOHHDIKCAFGCFBAEAGKCAA.marklists@mceahern.com>
Message-ID: <20020210154943.A25977@glacier.arctrix.com>

Mark McEahern wrote:
> That's what that is:
> 
> 	>>> list('abc')
> 	['a', 'b', 'c']

I should have said ['abc'] instead.

  Neil



From fperez528 at yahoo.com  Tue Feb  5 11:36:09 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Tue, 05 Feb 2002 16:36:09 +0000
Subject: ANN: HappyDoc 2.0.1
References: <mailman.1012913477.7851.python-list@python.org>
Message-ID: <a3pqbh$nru$2@peabody.colorado.edu>

Doug Hellmann wrote:

> Announcing the latest version of HappyDoc, a Python documentation extraction
> tool

An honest, non-trolling question: considering the amount of overlap that 
happydoc and pydoc seem to have, and that pydoc is included by default with 
python, what are the chances of merging the two?

It seems to make sense to me. Pydoc provides the help() function for the 
prompt and the nice 'pydoc -g' server on localhost:7464 (I have no idea if 
happydoc offers similar services). But it tends to choke on tricky imports 
and produce empty documentation pages way too often for my taste. So maybe 
merging the best of the two into something distributed by default would be a 
good idea.

I simply think that whatever comes by default with the python distro tends to 
get vastly more use/exposure, so the best tool should go there. And since 
happydoc seems to address at least one of pydoc's problems (failing imports 
in certain cases), it would be in the best open source spirit to produce a 
better tool out of the best features of these two, IMHO.

Just a thought...

f.


From frchalaoux at wanadoo.fr  Fri Feb 15 07:20:40 2002
From: frchalaoux at wanadoo.fr (FR Chalaoux)
Date: Fri, 15 Feb 2002 13:20:40 +0100
Subject: Thread and CGI
Message-ID: <a4iudk$9e4$1@wanadoo.fr>

Hi All,

The CGI I request run on a long time displaying regulary a point for the
advance. If I stop the page with the "stop button in IE", I stop the job !!
So I would like to print "Your job is in background now. Thanks for your
request."

The Thread approch is may be an alternative ??

Have you examples of thread approach if it is good one or others tips ?

FR




From st-newsgroups at thomas-guettler.de  Fri Feb  1 07:19:46 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Fri, 01 Feb 2002 13:19:46 +0100
Subject: Authentification
References: <3C5A840D.85C673A0@luc.ac.be>
Message-ID: <3C5A87E2.2010602@thomas-guettler.de>

Michel Van den Bergh wrote:

> I need to add username/password authentification to my application
> (a la apache or samba). So I need an encrypted password file to
> authentificate against, together with some utilities to add/delete users
> etc..
> 
> Is there a standard (proven) python framework I can use?
> 


Zope has a usermanagement. But I think it is overkill to use it just for
authentication.


You can encrypt the password and store it a file.
zpasswd.py of zope does it like this:

def generate_passwd(password, encoding):
     encoding=string.upper(encoding)
     if encoding == 'SHA':
         pw = '{SHA}' + binascii.b2a_base64(sha.new(password).digest())[:-1]
     elif encoding == 'CRYPT':
         pw = '{CRYPT}' + crypt(password, generate_salt())
     elif encoding == 'CLEARTEXT':
         pw = password

     return pw

HTH,

thomas



From gcordova at hebmex.com  Fri Feb 22 18:33:40 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 17:33:40 -0600
Subject: is their any limit on the size of the data passed to a progra
	      m via popen2?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AE9@SRM201002>

That doing a read() on pipes hangs, so you have to
define a timeout or something, so that you can take
control back.

But I dinnae kno how ta do such a thang.

:-/

-gus

pd: I *think* it was jeff.



From kfarmer at thuban.org  Mon Feb  4 08:35:51 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Mon, 04 Feb 2002 13:35:51 GMT
Subject: Pythonwin, DefWindowProc bug?
References: <%tL68.6127$Jp7.3608000671@newssvr14.news.prodigy.com> <3C5B90F3.3020503@skippinet.com.au> <CkN68.6145$F6.3643599033@newssvr14.news.prodigy.com> <3C5E34BA.6030702@skippinet.com.au> <5rt78.7926$Da3.4244083767@newssvr14.news.prodigy.com> <3C5E83A7.4090404@skippinet.com.au>
Message-ID: <X6w78.7937$Ke.4264630320@newssvr14.news.prodigy.com>

Actually, you know more than I how win does its message handling.

I can possibly find out from the folks at wxpython how the messages are
intercepted there.  Would PythonWin's hookmessage feature take priority over
anything defined there?

The signature for _OnWMNCHitTest is (self,params), in my source.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:3C5E83A7.4090404 at skippinet.com.au...
> Presumably _OnWMNCHitTest(self) is not the direct WndProc handler (as
> the param list is different).  Are you sure the result is getting back
> to the top-level correctly?
>
> Other than that I am out of ideas.





From bosahv at netscapenospam.net  Mon Feb 25 04:24:16 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Mon, 25 Feb 2002 09:24:16 GMT
Subject: Anybody can help on this?
References: <3C73A063.73AEB333@macrozon.net> <3C7547BF.463DD2AD@ccvcorp.com> <3C74AAA0.D3A050EE@macrozon.net> <3C7571A2.41D4E9E2@ccvcorp.com> <3C76551A.7060107@student.kun.nl> <3C768944.59BDD67E@ccvcorp.com>
Message-ID: <4pne8.456035$pa1.99989460@news3.rdc1.on.home.com>

Thanks from the lurkers :)


"Jeff Shannon" <jeff at ccvcorp.com> wrote in message
news:3C768944.59BDD67E at ccvcorp.com...
>
> husam wrote:
>
> >  >>> passout,passin=os.popen2("/home/husam/profit_1/profit")
> >   # Also tried  calling this with "popen2.popen2()" no go.
> > ....
> >  >>> passin.read()
> > # Here it hangs, the only way to stop it is by pressing: ctrl+c
>
> Yes, just to make it explicit, this will block until the pipe closes,
i.e.,
> the external program terminates.  This can be a long time, depending on
the
> external program.  :)
>
> This can be avoided by reading single lines (which blocks until a full
line
> is available), or by specifying a number of bytes to read (which blocks
> until that many bytes are available).  I think there is also an optional
> timeout, but it's been a while since I poked around with popen*() ...
>
> (I see that you already found a solution, but I wanted to post this for
the
> benefit of anyone else who might be following along...)
>
> Jeff Shannon
> Technician/Programmer
> Credit International
>
>




From brueckd at tbye.com  Sat Feb  9 13:50:20 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Sat, 9 Feb 2002 10:50:20 -0800 (PST)
Subject: Please stop responding to Rue
In-Reply-To: <a41jmf$cd2$1@newshost.accu.uu.nl>
Message-ID: <Pine.LNX.4.33.0202091047320.8686-100000@ns0.tbye.com>

On 8 Feb 2002, Martijn Faassen wrote:

> > Arguing with irrational oddballs on usenet is today's equivalent of the
> > Rubik's Cube.
>
> I never managed to solve those either. Is it still equivalent for those who
> do mananage to solve the cube?

Sure! Either way you eventually get bored and move on to something else.
Here's to hoping that happens soon...

-Dave




From chris.gonnerman at newcenturycomputers.net  Tue Feb 12 22:22:16 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 12 Feb 2002 21:22:16 -0600
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de> <mailman.1013521581.26252.python-list@python.org> <a4b8tr$1dq02n$1@ID-7529.news.dfncis.de>
Message-ID: <00d501c1b43d$aa99dea0$0101010a@local>

The thing that bugs me is, the report.zip file contains an example
in each of C++ and VB, both of which appear to work.

BTW having removed the quoted text, I'll repeat the URL for the OCX:

    http://www.qwerks.com/download/3693/report.zip

Any help would be appreciated...





From BPettersen at NAREX.com  Tue Feb 12 17:27:07 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Tue, 12 Feb 2002 15:27:07 -0700
Subject: hex to int
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192158D16@admin56.narex.com>

> From: maximilianscherr [mailto:MaximilianScherr at T-Online.de] 
> 
> how can i convert hex to int?
> 

I'm assuming you're asking how to convert a string representation of a
hexadecimal to an integer...

>>> int('0xff', 16)
255

-- bjorn



From sholden at holdenweb.com  Mon Feb  4 08:49:10 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 4 Feb 2002 08:49:10 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <20020202134649.468$Gg@news.newsreader.com>
Message-ID: <Wlw78.38517$KM2.1188216@atlpnn01.usenetserver.com>

"Hernan M. Foffani" <hfoffani at yahoo.com> wrote in message
news:20020202134649.468$Gg at news.newsreader.com...
[ ... ]
>
> I meant to say managment issues that doesn't have to do or relate to
> Python itself but, say, to business or customer policies.
> For example, if your CEO states: "We are an Oracle partner, so
> everything in PL/SQL" how much is left to say?
>
"I resign".

> Of course, I accept that turning those cases in Python succesful
> stories
> won't be offtopic.
>
> What I do saw in c.l.py from time to time and annoys me a lot is
> confusing techs issues with business ones and perhaps I was being a
> unfair with you.
>
> Worst, may be is just that *I* don't want to read hundreds of posts
> on proj-mgmnt here... Heh... So that's MY problem, aint so? :-(
>
The best managers are multidimensional, and know their strengths as well as
their weaknesses. A given manager may well be strong technically. Many
technical staff have rather less-well-developed inter-personal skills.

Ultimately the best teams (IMHO) are those which contain the right mix of
skills to be effective at their assigned tasks. The most enjoyable ones are
those where each individual can contribute the skills they are strong at
without those in a position of "power" feeling threatened by excellence.

I was taught that a good manager facilitates, and if the floor needs
sweeping when the team's busy producing, a good manager will get the broom
out. Bear in mind, however, that I'm an old fart.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Python Web Programming: http://pydish.holdenweb.com/pwp/







From idot at vt.edu  Thu Feb 14 21:38:11 2002
From: idot at vt.edu (Carl Banks)
Date: Thu, 14 Feb 2002 21:38:11 -0500
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com> <3C6C680D.B35E7719@cosc.canterbury.ac.nz>
Message-ID: <jash4a.sm1.ln@127.0.0.1>

Greg Ewing wrote:
> Courageous wrote:
>> 
>> And while vim indeed does just what I said, when you backspace over
>> these runs of spaces while in insert mode, it knows how those spaces
>> where generated and so:
>> 
>> "deletes any such spaces inserted through a tab, just as if
>> it were deleting a single tab."
> 
> This sounds ingenious, but what happens when you save the
> file and load it back? Does it remember this information
> somehow, or does it get lost?

The editor remembers this information by saving it in the text file
itself.


-- 
CARL BANKS                                http://www.aerojockey.com
"Nullum mihi placet tamquam provocatio magna.  Hoc ex eis non est."


From aa8vb at yahoo.com  Thu Feb 28 12:55:53 2002
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 28 Feb 2002 12:55:53 -0500
Subject: Python DB that supports numbers
Message-ID: <20020228125553.A2852587@vislab.epa.gov>

I'm looking for a very simple Python DB which supports:

     1) ints and floats as keys and values, 
     2) performs DB access in the client, and 
     3) stores the DB on disk.  

Just a simple DB for quick projects without the setup headache of
server-based database systems.

For example, on the following table I just want to: "select * where
column_1 is 1".  I'm using raw binary files now, but I'd prefer something
more self-describing like an SQL table.

     1       10       1.814  177.1   13.61   73.2
     1       5        1.624  182.5   -9999   -9999
     1       2        1.259  185.9   13.24   73.9
     2       10       2.672  181.4   13.59   72.5
     2       5        2.423  183.2   -9999   -9999
     2       2        1.902  190.7   13.27   73
     3       10       3.208  181.9   13.46   72.1
     3       5        2.787  186.6   -9999   -9999
     3       2        2.17   194.7   13.16   72.3

Does anything like this exist in Python?

Thanks,

Randall

-- 
Randall Hopper
aa8vb at yahoo.com



From threeseas at earthlink.net  Sun Feb 17 14:38:13 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sun, 17 Feb 2002 19:38:13 GMT
Subject: Guido wins FSF Award
References: <mailman.1013913086.401.python-list@python.org> <3c6fea5a.7618665@127.0.0.1>
Message-ID: <8867.813T2365T9245607threeseas@earthlink.net>

On 17-Feb-02 12:34:12 (Five Fresh) Fish <root at 127.0.0.1> wrote:
>On Sat, 16 Feb 2002 19:35:23 -0700 (MST), <kosh at aesaeion.com> wrote:

>>On Sun, 17 Feb 2002, Courageous wrote:
>>
>>>
>>> Really, Geedo didn't deserve this award. Since plainly Python is
>>> a system designed to obfuscate code, perpetuate complexity, and other-
>>> wise fails to recognize the simplicity of the nine command division
>>> of the entire physical universe, I redirect the committee to Sir
>>> Timothy Rue. My vehement protest to the comittee will be publically
>>> filed by 1100 GMT tomorrow, I swear. And well, if none of you see or
>>> understand why it is I have to do this, that's understandable. You are
>>> perhaps simply incapable of comprehending my rationale.
>>>
>>
>>ROFLMAO!!!
>>
>>Very well done comment and I agree completely with the point at which it
>>is aimed. :) However I think you need to state your response in terms of
>>the 9 basic commands that underly all things so as to make your point
>>better. Be sure to include about 3000 pages of explaining documentation so
>>that the simplicity of your answer can be seen.
>>
>
>I wonder if it's at all fair to make so much fun of a man who is obviously
>off his meds.  One should feel pity for Tim Rue, really.
>
>But then I start laughing and figure, heck, it's by his choice, so what the
>hell!  :-)

Hasn't anyone ever told you to not use root publicly? Or is it your
choice?

Ah, Kosh has a history of about 88 posts in Amiga newsgroups and up to
July last year. But May thru August last year he has made 30 post to c.l.p
until he added his 31st in this thread recently. I've also seen kosh
explode many time in IRC. But what Kosh says about 3000 pages.... Guess
kosh hasn't been around much enough to know better then to assume
documentation consist of about 3/7th's of my usenet posts. As well as his
putting on blinders to such simplicity as the following:

Certainly everyone does understand in reading and responding to post here,
they actually make use of all nine action. It's physics, but they may
think it's me and that somehow their reactions will cause a dismissal
of.... well physics.

Lets see now:

AI - start and stop, change interfaces - Uh start up newsreader and
     connect.

PK - keep track of where you are - Pick up where you left off on the thread..

OI - get input - read with eyes

IP - input from - internet and monitor

OP - output to - keyboard usenet postings

SF - one step at a time - damn this non-polyphony qwerty keyboard and
     mouse...

IQ - what's the meaning - whats a good non-target meaning I can attach to
     try and cause confusion to what tim is on about? ... Access the
     KE'd Thesaurus meaning of words, phrases and things Tim says.

ID - identify posters - hey there is one by tim, now I know to be
     (KE'd) constrain as defined by KE

KE - constraints to apply to IQ meanings and ID poster named tim rue -
     must be bit flipped (opposite) IQ'd, negative and anti-productive.

.
.
.
.

In the 80/20 rule, or even the 90/10, most all other industries have
productive working value of 80% or better, and problems of 20% or much
less. only about 16% of the software industries projects get out on time
and within budget. The hardware industry of computers is better than that
but also still highly incompatable.

Better yet to become master over this beast. Which even it has to use the
nine action constants of abstraction manipulation mechanics.


So how does Python address the following?



COMDEX SPRING and WINDOWS WORLD 95
Power Panel - "What's Wrong with Software Development"
** In The U.S. Only **
$81 Billion = 31% of software development gets cancelled before complete
$59 Billion = 53% of software development has cost over-runs of 189%
16% success - project success and failure ratio
61% customer requested features and functions make it in
Maintenance and repair is where most of the U.S. dollars are going,
instead of new, better, easier to use software.

---- Overall ----
Problems - all-around lack of complete documentation and weak training,
faulty user input and feed back - self contradictory user request, lack
of project leadership between developers and users, management created
problems and low quality control standards, feature creep and software
size increase, advancing technology rate of change and lack of general
standards, solutions around the corner but never arrive and our tools
are better than theirs attitude, lack of a value chain structure for
value added abilities, failure to produce a functional model before
coding and constant remodeling, etc.

Solution directions - code re-use, object oriented programming,
component-based programming, distributed components, better tools,
better programming methodologies, leaner software, a splitting of code
writer types into two catagories - architects and assemblers, better
effort to establish a working vocabulary between developers and users
so users can in some way lead development, etc.

---- A Few Comments from Panel Members ----
A culture needs to evolve that respect software engineers as
crafts-people. Writing code is not just writing code but like the field
of writing where you have technical, creative, documentary, etc., there
are different types of code writing. (Authors' note: I agree with this
but also realize end users are even more specialized in what they need
and do. Respect for the end user needs and abilities is needed even more
so. Without respect given to the end user, the software engineer will not
be given respect in return.)

A fundamental change in the programming environment needs to happen that
allows the tools to work together more. (Authors' note: the panel member
making this comment, did not specify what tools or who the tools would be
used by. It was a very general comment pointing to a fundamental
programming environment change. A lead in to the concept of componet
programming. But, there was no recognition given to the concept of
componet software or componet applications. At least not in the sense of
being outside of "plugins". Read on!)

Jokingly - one of the best ways to copy protect software is to put it in
a dll, give it an obscure name and put it in the windows system directory.
Because you'd never find it. (Authors' note: This does not make it any
easier for the end user in keeping their system organized, clean and
optimized. This attitude of constraints, though humorous, cost end users
alot.)

The meaning of "intellectual property" became questioned. Did it mean you
take the best ideas or something owned? (Authors' note: it was the panel
supporting "best ideas" but wouldn't the correct term for this use be
"intellectual value" rather than "intellectual property"? What would
happen, regarding this, in a court room? The audience member whom brought
this up, was a bit angry about the distortion. Her question was: Is it
the developers whom are creating the problems? And what are the
developers going to do about it? The responce was "that's not the
problem!")

Users shouldn't develope software but know, better than the developers,
what they want and need. (Authors' note: users don't have the time to
write code, it's not their job or duties!!! I can cut the lawn, I know
how, but if I don't have the time, I hire someone. And because I know how
to better communicate what I want done, I'll get what I want and know
I'll not be greatly overcharged.)

Analogy used to start off power panel: Getting the right software
development tools is alot like dating. And it evolved to something about
programmers not being able to get dates, while touching a nerve with a
panel member. (Authors' note: Phillip needs to get a amiga. Amiga is
mexican for - girl friend. Phillip, if you see this, you know you gotta
laugh about it.)

(Author observation from attending this gathering - alot of good points
where brought up from both the audience and members of the panel but it
became clear there was no solution being brought forward to satisfy the
majority. The audience saw this as it thinned out over the course, as
they perceived the power panel struggling for a sales pitch. There where
two on the panel not biased due their position, leaving six biased.
Microsoft, Borland, Powersoft, Oracle, Software Associates, and IBM were
the biased parties.)

Panel mix - Tools developers, Data Base Developers, Application
Developers, Application salvagers, and software consultants.

BUT NOT USERS, as in end users/consumers/clients.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From mhammond at skippinet.com.au  Mon Feb  4 02:45:41 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 04 Feb 2002 07:45:41 GMT
Subject: [repost] win32 debugging-- python21_d.dll not found (fwd)
References: <Pine.SOL.4.30.0202022147160.597-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <3C5E3C76.6080800@skippinet.com.au>

I'm afraid this seems to be a problem with mingw rather than Python. 
All I can suggest is that you start with a trivial example - such as 
"example_nt" in the Python source distribution.  Get this working under 
mingw.  Then we can work from there to isolate the problem.

Mark.

John J. Lee wrote:
> [Sorry for repost: I left a rather long gap before replying to Mark's
> suggestion, so it probably looked like the problem had been solved, and
> I'm sure someone out there must have a clue about this.]
> 
> On Fri, 25 Jan 2002, Mark Hammond wrote:
> 
> 
>>John J. Lee wrote:
>>
>>>I compiled python21_d.dll and python21_d.exe with MSVC, and stuck the
>>>_d.dll in C:\winnt\system32 along with python.dll, and the _d.exe in
>>>C:\Python21, alongside python.exe.  When I try using python_d.exe, with an
>>>extension compiled with mingw32 against libpython21_d.a, I get a message
>>>box complaining that python21_d.dll couldn't be found, yet
>>>C:\winnt\system32 is listed in the search path given in that very message
>>>box.  Are DLLs not found by filename?  Or what?
>>>
>>Check that all dependent modules can be found.  Use the free "depends"
>>tool to determine what other DLLs have been referenced by python21_d.dll
>>are available.
>>
> 
> Well, the only direct dependencies of python21_d according to depends.exe
> are (as I'm sure you know!) standard windows dlls: kernel32, user32,
> advapi32, shell32 and msvcrtd.
> 
> However:
> 
> 1. I just realised that the error only occurs on importing the extension
>    module, not on starting python_d.exe itself, which runs quite happily
>    in interactive mode until you import _recode.
> 
> 2. If, instead of looking at the python21_d DLL, I look at _recode_d.dll
>    in depends.exe, I get a dependency on PYTHON21_D with a big question
>    mark icon to the left of it.
> 
> I was under the impression that Python on Windows resided essentially in
> the DLL.  If I have that correct, then since python_d.exe works fine in
> interactive mode presumably the problem is with the _recode module rather
> than my system setup.  Not exactly a helpful error from Windows...
> 
> So, why does depends.exe complain about _recode's dependency on
> python21_d, when it seems quite happy with python21_d itself, and when,
> what's more, they both sit in the very same directory (C:\winnt\system32)?
> I guess if I knew that I'd know why I get the error on importing _recode
> with python_d.exe.  I'm guessing this is related to the fact that the
> extension was compiled with mingw, but exactly how, I'm not sure.
> 
> The extension module in question (_recode) does import ok on windows when
> compiled against the non-debugging python, though it causes a memory
> access error on actually using it (after my little test routine finishes
> without error, around when the interpreter exits, though I don't know
> exactly when this happens, obviously, since I haven't been able to debug
> it yet).  Since this is somebody else's extension, written by someone more
> competent with C than me, and given also the fact that it works fine on
> unix, I suspect some cockup by yours truly is causing this, but what it
> might be I don't know.
> 
> I've tried both a version (of the debugging DLL and .exe) I compiled
> myself with MSVC, and using the standard distribution of debugging libs.
> I'm using the standard python.org windows Python 2.1.1 distribution.  I'm
> restricted to mingw for the extension itself because I haven't got MSVC to
> compile it yet.
> 
> The error text is as follows (typed in by hand since it seems not to be
> possible to cut-and-paste the damn thing):
> 
> The dynamic link library python21_d could not be found in the specified path
> c:\Python21\DLLs;;C:\WINNT\System32;C:\WINNT\system;C\WINNT;d:\ProgramFiles\mingw\bin;
> 
> ...etc., and from Python:
> 
> $ //c/Python21/python_d.exe
> 
> Adding parser accelerators ...
> Done.
> Python 2.1.1 (#20, Jul 20 2001, 01:19:59) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> 
>>>>import _recode
>>>>
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> ImportError: DLL load failed: The specified module could not be found.
> [5430 refs]
> 
> 
> Thanks for any help!
> 
> 
> John
> 
> 




From mark at mceahern.com  Wed Feb  6 20:34:52 2002
From: mark at mceahern.com (Mark McEahern)
Date: Wed, 6 Feb 2002 17:34:52 -0800
Subject: why i love python...
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEEJKBAA.mark@mceahern.com>

[please forgive the embedded html; this is posted on my weblog, but since
that's not exclusively about python, i didn't want to bore you all with it.]

what started out as just an attempt to make debugging easier turned into an
epiphany about why i love python.  allow me to share that love...

python comes with a slew of special methods for objects that you can
leverage to do some really neat things.  consider __str__, which is called
on your objects when you <b>print</b> them.  if you don't define __str__,
that's ok, you'll get something like this:

<pre>
>>>class Simple:pass
...
>>> s = Simple()
>>> print s
<__main__.Simple instance at 0x100fd4c8>
</pre>

by defining __str__, i get to control the output generated by print (and
str()):

<pre>
>>> class Prints:
...     def __str__(self):
...             return "whatever i like"
...
>>> p = Prints()
>>> print p
whatever i like
</pre>

obviously, i should return something more meaningful than that.  for
instance, it might be nice to return an xml representation of the object.
again, with python this is unbelievably easy:

<pre>
>>> class PrintsXml:
...     def __str__(self):
...             s = "<%s>" % self.__class__.__name__
...             for key in self.__dict__:
...                     s += "<%s>%s</%s>" % (key, self.__dict__[key], key)
...             s += "</%s>" % self.__class__.__name__
...             return s
...
>>> p = PrintsXml()
>>> p.name = "foo"
>>> p.type = "bar"
>>> print p
<PrintsXml><type>bar</type><name>foo</name></PrintsXml>
</pre>

this snippet shows:

<ul>
  <li>how you can use self.__dict__ to access an object's attributes.</li>
  <li>how you can iterate over the keys of a dictionary (i believe that's
new with python 2.2).</li>
  <li>how you can use self.__class__ to access an instance's class; and
__class__.__name__ to access the name of the class as a string.</li>
</ul>

side note:  i'm doing all this in the interactive interpreter.  just fire up
python and start typing.  i like that.

i think this shows some of the power of python.  particularly, how special
methods can save you an amazing amount of time.  to me, the natural next
step for this code is to make PrintsXml a virtual base class and beef up its
handling of complex attributes, particularly:

<ul>
  <li>handle list attributes intelligently.</li>
  <li>determine whether attributes are themselves subclasses of PrintsXml,
in which case you can just let them print themselves rather than wrapping
them in xml tags manually.</li>
</ul>

of course, i haven't even touched on the most exciting aspect of all this:
__repr__.  __repr__ allows me to return a string representation of my object
that can be used to re-create it.  what i don't know is how that compares to
pickling, but i'm sure the latter is tastier.  ;-)

cheers,

// mark




From logiplexsoftware at earthlink.net  Fri Feb 22 12:23:24 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 22 Feb 2002 09:23:24 -0800
Subject: Newbie: Crazy, but Quick
In-Reply-To: <slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu>
References: <u7ag65cbc0q3e@corp.supernews.com>
	<CMfd8.5037$D6.14223@news.iol.ie>
	<u7b81629bo9o4a@corp.supernews.com>
	<slrna7b9hr.6lj.quinn@yak.ugcs.caltech.edu>
Message-ID: <20020222092324.0fefd4bf.logiplexsoftware@earthlink.net>

On 22 Feb 2002 02:00:27 GMT
Quinn Dunkan wrote:

> On Thu, 21 Feb 2002 17:35:44 -0800, Virginia O <vodine at direct.ca> wrote:
> >Thanks, but the database will have to hold sub-groups etc. It will be
easier
> >for me to enter and organize the items in a database.
> >I only gave the scenario as an example. The actual program will hold
> >hundreds of items.
> 
> Hundreds is just fine for flat text.  If you have a few thousand and
you're
> modifying it a lot, then flat text will start being clunky.
> 
> In that case, shelve should work fine, especially for complicated data
> structures since you can dump python types directly, with no clunky
converting
> to/from some format SQL likes.  And it's much simpler to have everything
in
> python instead of a whole other database program with its own language,
etc.

For that matter, you could also store the Python objects directly in the
SQL database.  My own opinion is that SQL is so easy that it's a no-brainer
to use it whenever possible.  A problem I've seen many times is people
utilize some shortcut because they feel the program will never need
anything more, only later to discover that the program has utility far
beyond what they expected, but the shortcut they've taken cripples it, and
they end up rewriting (or worse, they don't).

What if she decides to maintain historical data?  Then to keep the files
reasonably sized, she'll have to start juggling files.

Software has a tendency to grow beyond the developer's expectations, so I
consider it unwise to take shortcuts when a better long-term solution isn't
much more difficult to implement.

Not only that, but storing the data in a SQL database separates the data
storage from the program, so in the future she could use that data from
some other program, not just the Python one.  Then if she decides she wants
to access the data from, say, a web server, it's not a big problem. 


-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From db3l at fitlinxx.com  Mon Feb 11 20:08:46 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 11 Feb 2002 20:08:46 -0500
Subject: Problems using calldll library
References: <84323230.0202080309.382bf7f4@posting.google.com>
Message-ID: <u7kpjqzfl.fsf@ctwd0143.fitlinxx.com>

picodello at yahoo.it (Piero) writes:

> I need to interface with a windows DLL that is declared in C like:
> 
> getword(&ncard,&nb,&adr,tabl,&status)
> 
> where the type of parameters are :
> 
> ncard=16-bit integer
> nb=16-bit integer
> adr=32-bit integer
> tabl= array of 16-bit integer
> status=16-bit integer
> 
> where ncard,nb,addr are input parameters
> and   tabl,status are output parameters
> 
> I try to implement this function, but don't work.
> I suppose the problem is in the input parameters format.
> Something can help me ?

Do you have any API documentation for the function at hand that
describes the parameters in more detail?

>From the above, I would assume that "tabl" is a pointer to an array of
16-bit integers, but typically one of the other parameters (perhaps
ncard, or nb) would represent - on input - the size of the array, and
possibly - on output - the number of elements returned.

But in your sample code you only allocate enough memory for tabl to
hold a single 16-bit integer, and I also see you putting the supplied
parameters as strings into the memory buffers allocated to ncard and
nb, whereas I would normally expect them to be integer values.  This
could be creating an overrun condition where the function thinks you
have more room in tabl than you really do.

Also, you use "i" in the calldll.call_foreign_function() method for
your references, but I'd probably stick with "l" as a C "long" for a
32-bit pointer.  Each of the parameters is a 32-bit pointer even
though some point to 16-bit integer data.  Then again, I don't know
that any Windows platform currently uses "int" as 16-bit unless you're
using an old 16-bit compiler, so it probably works either way :-)

> 	def getword(self,ncard,nb,adr,tabl):
> 		getword =calldll.get_proc_address(self.applicomdll,"getword")
> 		p = calldll.membuf(4)
> 		px = p.address()
> 		pncard = calldll.membuf(2)
> 		pncard.write(str(ncard))
> 		pxcard = pncard.address()
> 		pnb = calldll.membuf(2)
> 		pnb.write(str(nb))
> 		pxnb = pnb.address()
> 		padr = calldll.membuf(4)
> 		padr.write(str(adr))
> 		pxadr = padr.address()
> 		tabll = calldll.membuf(2)
> 		ptabll = tabll.address()
> 		ret= calldll.call_foreign_function (getword, 'iilii', 'l',
> (pxcard,pxnb,pxadr,ptabll,px))
> 		res = p.read(0, 2)
> 		print struct.unpack('bb', res)[0]
> 		ress = tabll.read(0, 2)
> 		print struct.unpack('bb', ress)[0]

Assuming that ncard, nb, and adr really are numeric, I would at a
minimum suggest changing your function into something like:

    def getword(self,ncard,nb,adr):
	"""Returns tuple (rc, tabl, status)"""

	getword =calldll.get_proc_address(self.applicomdll,"getword")

	# Allocate space for reference parameters and status
	ncard_buf = calldll.membuf(2)
	nb_buf = calldll.membuf(2)
	adr_buf = calldll.membuf(4)
	tabl_buf = calldll.membuf(2 * ncard)     # Assume ncard is count
        status_buf = calldll.membuf(2)

	# Assign initial values
	ncard_buf.write(struct.pack('h',ncard))  # Use 'H' if unsigned
	nb_buf.write(struct.pack('h',nb))        # Use 'H' if unsigned
	adr_buf.write(struct.pack('l',adr))      # Use 'L' if unsigned

	ret = calldll.call_foreign_function (getword, 'lllll', 'l',
                                             ncard_buf.address(),
                                             nb_buf.address(),
                                             adr_buf.address(),
                                             tabl_buf.address(),
                                             status_buf.address())

        status = struct.unpack('h',
                               status_buf.read())[0]   # 'H' = unsigned
	tabl = struct.unpack('%dh'%ncard,
                             tabl_buf.read())[0]       # 'H' = unsigned

        return (ret,status,tabl)

Note that I don't bother with tabl as input since it's an output of
the function, and will be a new list created from the result.  I
suppose you could pass in an existing list to be mutated but it seems
somewhat unnecessary.  Also, I'm assuming that your integer values are
signed.  If they are unsigned make the suggested changes to the
struct.pack/unpack calls.  I also assume 'ncard' was the count of the
size of the array pointed to by tabl, and is unchanged on output.  If
it's really also an output of the function (e.g., the number of
entries returned), then you could unpack ncard_buf and use that rather
than ncard in the unpacking of tabl_buf.

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


From johnroth at ameritech.net  Sun Feb 10 13:24:39 2002
From: johnroth at ameritech.net (John Roth)
Date: Sun, 10 Feb 2002 13:24:39 -0500
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org> <afe3b146.0202100036.f221c9a@posting.google.com>
Message-ID: <u6degsscph4hd2@news.supernews.com>

"Brian Lenihan" <brian_l at yahoo.com> wrote in message
news:afe3b146.0202100036.f221c9a at posting.google.com...
> "Dr. David Mertz" <mertz at gnosis.cx> wrote in message
news:<mailman.1012934533.15278.python-list at python.org>...
> > |"Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
> > |> I suppose it might be a brain thing. I use my left hand for
writing and
> > |> drawing while I use my right hand for more coarse-grained tasks,
like
> > |> throwing a ball and such.
> >
> > "Steve Holden" <sholden at holdenweb.com> wrote
> > |Strange. I don't come across many other people who do this. I've
always
> > |thought my brain must have been differently wired for some reason
I'll never
> > |know.
> >
> > Same story for me.  I don't think it is that unusual though.
Culture is
> > sort of geared to the right-handed majority, so us folks who are
mostly
> > left-handed wind up learning the things they can do "well enough"
with
> > the right hand in the common style.  Once you've learned something,
> > propioceptic memory is more important than underlying "handedness."

I missed the start of this, so I'm jumping in the middle...

I do most things equally well with either hand. My father always
claimed that he lettered engineering drawings with whichever hand
was closest to the pencil (of course, he could have rigged where the
pencil was normally located...)

The science seems to be that there are obligatory right-handers
who can't do anything with their left, obligatory left-handers
who likewise can't do much with their right (about 2%, if I remember
the figures) and people who are more or less ambidexterous, who
have somewhat of a preference for one hand or the other, but can
use either.

The best security device I ever had was a disconnected mouse
that I left on the right side of my keyboard on the desktop. The
real mouse was on the left. I always got a kick out of watching
people try to use the mouse on my system, and failing...

John Roth






From gcordova at hebmex.com  Fri Feb 22 10:21:12 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 22 Feb 2002 09:21:12 -0600
Subject: Any Pythonistas in Mexico?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AD8@SRM201002>

Where are you at?

I'm in Monterrey.

> 
> Hello,
> 
> my company is located in Mexico and we use Python for our
> development projects, I am trying to find out how many of
> us Python enthusiasts live in Mexico and see if maybe we
> can form a user group or something.
>

Man, this would be so cool.

Once I met a "LUG" at the faculty, and they were just
a bunch of script-kiddie wannabes, very very bogus.

> 
> Thanks,
> 
> Carlos de la Guardia
> Aldea Systems
> 

You're welcome.

-gus



From starx at pacbell.net  Mon Feb 18 04:54:00 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 18 Feb 2002 09:54:00 GMT
Subject: Ascii Art
References: <mailman.1012327189.31750.python-list@python.org> <585.799T42T11732924threeseas@earthlink.net> <slrna5rqpa.19d.philh@comuno.freeserve.co.uk> <a3ltbd$3oi$1@verence.demon.co.uk> <Xns91B88FF6B244ERASXnewsDFE1@130.133.1.4> <a4pcmp$hdv$0@216.39.172.122>
Message-ID: <Xns91B9134B3E315RASXnewsDFE1@130.133.1.4>

Bengt Richter || Sun 17 Feb 2002 03:00:41p:

> On 17 Feb 2002 22:09:20 GMT, Philip Swartzleonard <starx at pacbell.net>
> wrote: 
> 
>>Paul Wright || Mon 04 Feb 2002 04:03:57a:
>>
>>>                       ______
>>>                      /      \
>>>                    .' PLEASE `.
>>>                    |  DO NOT  |      _____
>>>                    | FEED THE |    ,'.....`.
>>>                    `. LOONY  ,'  ,'........ )
>>>                      \_    _/   |........ ,'
>>>                        |  |     `. .... _/
>>>                        |  |      ,'.,'-'
>>>                        |  |     /../
>>>                        |  |   ,'.,'
>>>                        |  |  /../
>>>                      . |  | /..'
>>>                    .\_\|  |/_/,
>>>                    ___ |  | ___
>>>                      . `--' .
>>>                       .    .
>>> 
>>> It was fun at first but now it's tiresome. 
>>> 
>>
>>I like it, even if it's uncertain weather it's a sign or a Paddle of 
>>Punishment =)
>>
> The weather can't be that uncertain -- I see a shadow ;-)

Weather or not, it seems i saw a source of shadow that i now cannot. A 
sign then it must be, perhaps. %D

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From pbienst at mit.edu  Tue Feb 26 11:17:53 2002
From: pbienst at mit.edu (Peter Bienstman)
Date: Tue, 26 Feb 2002 11:17:53 -0500
Subject: Precompiled regular expressions slower?
Message-ID: <3c7bb531$0$3930$b45e6eb0@senator-bedfellow.mit.edu>

Hi,

I've a Python script a want to optimise. It basically loops over all the 
lines in a file and does some regular expression matching:

line = inputfile.readline()
while line:
  s = search(r".....", line)
  ...
  line = inputfile.readline()

There are several reg ex in the loop. Since the Python version runs 2 to 3 
times as slow as its Perl counterpart, I thought I'd try speeding up the 
script by precompiling the reg ex:

re1 = compile(r"....") 
...
line = inputfile.readline()
while line:
  s = search(re1, line)
  ...
  line = inputfile.readline()


Strangely enough this new version seems to run about 30% *slower* than the 
old version.

Am I missing something here? (I'm using Python 2.1)

Peter


From jonathan at onegoodidea.com  Fri Feb  1 12:32:53 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Fri, 01 Feb 2002 17:32:53 +0000
Subject: Running CGIs under my uid - going slowly insane.
References: <4b4c5d99.0202010136.8e6ff51@posting.google.com>
Message-ID: <B88081C5.40F9%jonathan@onegoodidea.com>

On 1/2/2002 9:37, in article 4b4c5d99.0202010136.8e6ff51 at posting.google.com,
"pawn" <NOSpawnPAM at lightspawn.org> wrote:

> I need my Python CGIs to run scripts under my own uid, since they read
> and update .html files which are core to the site.
> 
[...]
> 
> * Switch back to perl which respects 04755 permissions.

I don't think the lack of respect for setuid permissions is anything to do
with Python I'm afraid. Quick test on my machine:

-----
% cat > test.py
#! /usr/bin/python

import os

print os.geteuid(), os.getegid()
^D
% chmod 755 test.py
% ./test.py
501 20
% sudo chown 71:71 test.py
% ./test.py 
501 20
% sudo chmod 4775 test.py
% ./test.py 
71 20
% sudo chmod 6775 test.py
% ./test.py 
71 71
-----

Are you sure the scripts aren't running as you? How are you testing this?

Note that setting an executable's setuid bit will change the "effective" uid
of the resulting process, not the uid (see 'os.get*e*uid()' above). This
means that the process may use that user's permissions.

The webserver doesn't contain a Python interpreter does it? Perhaps it's
directly executing the scripts instead of 'exec'ing them?

Jonathan



From threeseas at earthlink.net  Sun Feb  3 19:01:09 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:01:09 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
Message-ID: <922.799T2387T11875593threeseas@earthlink.net>

In reviewing over many messages in the autocoding and related threads and
seeing the evidence of contridictions, hypocracy, irrational and
illogical responses......

Its damn clear that there is a need to remove the sort of people posing to
be professional who are in essence just good liars, abstraction
manipulators, from the industry.

Who knows, maybe that stench will be enough to motivate me see that it
gets done, gets removed.

Maybe I should be more appriciative of the stench if it so motivates me.

Anyone want to add anything to this?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From aleax at aleax.it  Sun Feb 17 12:34:49 2002
From: aleax at aleax.it (Alex Martelli)
Date: Sun, 17 Feb 2002 17:34:49 GMT
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <7xg040yac2.fsf@ruckus.brouhaha.com>
Message-ID: <ZQRb8.67062$om6.1831847@news1.tin.it>

Paul Rubin wrote:

> "I.J." <shiver at yubc.net> writes:
>> Is there a module for python that grants ussage of decimal numbers with
>> more than 14 decimal places, including all scientific functions to do
>> with those numbers?
> 
> Gmpy (http://gmpy.sourceforge.net) does arbitrary precision floating
> point math.  It doesn't include any scientific functions beyond
> arithmetic and square root though.  You might be able to adapt the
> "bc -l" library to compute scientific functions with gmpy.

I suspect it might be easier to adapt/wrap MPFR, see
http://www.loria.fr/projets/mpfr/ , exploiting the C-API that
gmpy exposes and the fact that MPFR is meant to work with GMP
(which gmpy wraps).  I'll have to get back to gmpy and polish
it up a bit one of these days (even though I'm currently doing
no multiprecision computations -- which is my excuse for having
let gmpy collect some dust lately...) and I may then check
back into MPFR and how hard it would be to wrap.

MPFR only supplies log, exp, sin and cos (but, to whatever
degree of precision is required).  So, it may not meet the
requirement for "all" scientific functions.


Alex



From marklists at mceahern.com  Fri Feb  8 14:36:44 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Fri, 8 Feb 2002 11:36:44 -0800
Subject: What is the most efficient way to do this?
In-Reply-To: <a418qm$1c055f$1@ID-98166.news.dfncis.de>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOELAKBAA.marklists@mceahern.com>

[MDK]
> What is the most efficient way to remove empty items from this list?
> 
> ['', '1', '2', '3', '', '456', '789', 'wow']

Try filter:

>>> l = ['', '1', '2', '3', '', '456', '789', 'wow']
>>> filter(None, l)
['1', '2', '3', '456', '789', 'wow']




From jkraska at san.rr.com  Wed Feb  6 19:43:54 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 00:43:54 GMT
Subject: Autocoding project proposal.
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net> <mailman.1012972032.30205.python-list@python.org> <gbl16u4ug6pau2l393eh50t85o031csmtm@4ax.com> <slrna62cpr.afn.philh@comuno.freeserve.co.uk>
Message-ID: <cnj36u82v6ohdcqqmi19j2fldnlctdr76j@4ax.com>

>You can sometimes get a lot into 1000 LOC. Herbrip is currently
>about 4000 LOC, but 1000 of that is the test suite. Of the rest,
>about half is comments and blank lines. So Herbrip is about
>1500 LOC of actual code.

True! The pride of my professional career -- while nothing innovative,
it was just work -- was reifying a program written by one developer
at about 35,000 lines of code to just about 5,000.

C//



From DeepBleu at DeepBleu.org  Thu Feb 14 14:25:54 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Thu, 14 Feb 2002 13:25:54 -0600
Subject: zlib.so and libz.so and Python
References: <u6niisr909gtd6@corp.supernews.com> <mailman.1013703624.30857.python-list@python.org> <u6nuqr108os0ae@corp.supernews.com> <mailman.1013710106.1576.python-list@python.org>
Message-ID: <u6o3uknfbb9pe7@corp.supernews.com>

Oleg
The machine where libz.so is not found, does not have libz.so at all.
I figured that thing out after sending the first reply.
I have to install the zlib package itself for Sun.  I found it at
http://www.gzip.org/zlib/
The package has libz.so in it.
DeepBleu

"Oleg Broytmann" <phd at phd.pp.ru> wrote in message
news:mailman.1013710106.1576.python-list at python.org...
> On Thu, Feb 14, 2002 at 11:59:40AM -0600, DeepBleu wrote:
> > % ldd zlib.so
> >          libz.so =>       /usr/openwin/lib/libz.so
> >
>
> > % ldd zlib.so
> >          libz.so =>       (file not found)
>
>    export LD_LIBRARY_PATH=/usr/openwin/lib
>
> Oleg.
> --
>      Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
>            Programmers don't die, they just GOSUB without RETURN.
>




From tdelaney at avaya.com  Mon Feb  4 17:36:24 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 5 Feb 2002 09:36:24 +1100 
Subject: Making a list-inhereted object return standard repr?
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A163@natasha.auslabs.avaya.com>

> From: Steve Holden [mailto:sholden at holdenweb.com]
> 
> "Delaney, Timothy" <tdelaney at avaya.com> wrote ...
> > This brings up the interesting question of whether there 
> should be a an
> > object method which returns the "canonical repr" of an 
> object i.e. the
> > <module.Class object at address> or possibly even a builtin 
> method (so it
> > would work with classic classes too).
> >
> > I know I've wanted to use this in classic classes many 
> times for debugging
> > purposes - I want to be easily able to distinguish 
> instances from each
> > other. Each time I end up having to code it myself.
> >
> Although, of course, even this would not necessarily be 
> unambigous if you
> had code that produced classes from factory functions, as you 
> would then
> have to identify the class by address rather than by name (each class
> returned by the factory having either the same name or no name at all,
> depending on your point of view).

I presume you mean factory functions that have an embedded class definition
...

def factory():

    class Klass:

In that case, you are of course correct. I didn't think of this as it's not
a construct I use very often - in Python I tend to keep my classes at the
module scope.

However, the current __repr__ for such classes fails to distinguish between
instances of different classes produced in this manner as well. I think this
is a case where it is impossible for the interpreter to give you any more
information without reading your mind, and it is therefore incumbent on
either the programmer giving more hints to the interpreter (e.g. writing
their own __repr__, possibly based on the default __repr__) or for Guido to
integrate the mind-reading module into the core library. And I'm afraid this
would be backwards incompatible, as programs started to do what we wanted
them to do, rather than what we told them to do.

Tim Delaney



From maxm at mxm.dk  Tue Feb 26 10:58:11 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 26 Feb 2002 15:58:11 +0000
Subject: .NET only solves problems I don't have! (Was: advocacy help ( Python vs VB ))
References: <FF1A87368802D41199CD009027DE5C7904715630@srfmsg01> <mailman.1014706963.7898.python-list@python.org>
Message-ID: <3C7BB093.3020600@mxm.dk>

Alex Martelli wrote:

> Ah, but WHAT VB are we "selling" against?  Microsoft is in the process
> of changing VB *incompatibly* -- you would need to study it again and
> rewrite just about all of your code to migrate to old VB6, which MS does
> not like any more, to new VB.NET, which means betting your company
> on MS's unproven, fascinating but risky .NET strategy.

What I find most amusing about com and .NET is that they are trying to 
solve a problem I only had when programming using MS tools.

Ie. when I coded webpages in JScript and wanted to upload a file I had 
to buy a component for it (Or write it in JSCript from scratch :-( )

If i wanted to send mail I had to buy a component for it.

Etc. etc.

And most of the modules I bought had bugs in them. Ie. they would work 
under one version of Microsoft Scripting Host (MSH) in both VBScript and 
JSCript but in another version of MSH the module would only work under 
VBScript ... So depending of you servicepack your components might stop 
working in your language of choice.

Under Python I have several excellent modules for what I want to do. If 
I need something there is usually almost a library available for it. Or 
it is easy to make my own library using other Python libraries.

And the available libraries have ALLWAYS been better than the commercial 
cr^H^H stuff I have spend my cash on in MS land.

To take an example: If i had to write something like the file-upload 
modules for IIS in JScript I would be in trouble because I would have to 
write my own MIME parser and other stuff in JSCript.
In Python I can use the MIME library to parse the multipart data and get 
it done relatively painlessly.

So often it is easier to roll your own in Python than to find the right 
COM-component, reading the docs, checking it out, write code for it, 
only to find it lacking.

I believe that the main driving force behind .NET is to make it easier 
for people to develop and sell closed source components. And I am ok 
with that. I write software for money too. But as a developer using open 
surce tools I have NO need for it at all.

All they have is a great solution to a non-issue.

So MS is developing this massive strategy and technological toolset to 
solve a problem that could be solved simply by switching to Python ;-)

As any programmer worth his salt knows, you can have too many layers of 
abstractions in your code. And I don't think that this common runtime 
solves any significant problems.

  - I can develop in any language I please
  - well I allready do

  - I can share objects written in othes languages
  - Well I allready do

  - Compiled code written in your language can run unchanged on .NET
  - Well it allready runs on most any platform. Including Windows.

I believe that mixing languages at several meta levels makes great 
sense. So a combination of Scripting, c and even assembly, as C. Tismer 
has shown in stackless, is a great toolbox. But for me to change to 
another scripting language it would have to be an order of magnitude 
better than Python. Or else it is not worth the time and energy.

So why should I care that I can write components in C#, J.NET, 
VBScript.NET, JSCript.NET etc. ?? I will bet that any language running 
under the common runtime will not be an order of a magnitude better than 
Python.NET, as those languages will have too much in common by design.

I am shure developers using other languages will think the same. Oh yeah 
should I choose to develop in Python.NET and use a class written in C# 
what do I do when it's buggy.? Will I have to learn every language under 
.NET to be shure I can debug other peoples poor classes? Using other 
peoples classes is only fun if they are good and open source. An 
preferably in a language you know. Otherwise they might as well be closed.

Beside, using classes made for all languages will allmost certainly 
lower the expressive level of my favourite language.

MS are trying to take a massive detour around problems they have created 
for themself with their closed approach.

A few good languages at different hardware abstraction levels and good 
open libraries I can read and debug is what I need.

Oh wait. I allready have that.

regards Max M



From hankc at nospam.com  Tue Feb 26 18:53:27 2002
From: hankc at nospam.com (HankC)
Date: Tue, 26 Feb 2002 23:53:27 GMT
Subject: Find an Item in a Sorted List
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
Message-ID: <3c7c1fce.35698011@news.la.sbcglobal.net>

On Tue, 26 Feb 2002 23:50:58 GMT, hankc at nospam.com (HankC) wrote:

I wrote:

dataline = filter(Match, eodflist)

should be:

dataline = filter(Match, flist)

Sorry ;)


From mcfletch at geocities.com  Mon Feb 25 20:28:58 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Mon, 25 Feb 2002 20:28:58 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu>
Message-ID: <3C7AE4DA.7090103@geocities.com>

socket.recv is letting you specify the maximum amount returned, it 
doesn't wait for new information to fill the buffer before returning, it 
just returns up to maximum from what has been received at the current 
time.  It will only wait for information if there is _nothing_ yet 
received and you've specified blocking sockets (default).


Here's a simplified timeline (caveat, I don't use TCP much, just UDP, so 
I might be getting some of the TCP semantics messed up):

packetReceivedByPort -> puts fractional message into OS-level buffer

packetReceivedByPort -> puts fractional message into OS-level buffer

OS-level buffer is now full, OS TCP stack signals server to stop sending 
packets, as it can't deal with any new data until the OS-level buffer is 
read/recv/cleared [I'm assuming TCP does this explicitly, I normally do 
with my own networking code]

[repeat these two steps]
mySocket.recv( ..., X ) -> gets up to X bytes from the OS-level buffer

OS TCP stack tells the server to continue sending, as it can down deal 
with the new data.  Buffer re-fills.


When mySocket.recv( ..., X ) -> returns "", your TCP connection is no 
longer sending information, you got the whole thing, charge onward.


Note:
	Special case for blocking sockets if there is nothing currently in the 
buffer is to wait for the next packet before returning.  Also note that 
with length-of-data information, you can stop reading when the length of 
the received data == reported length (changing X so that the max 
returned == just enough to get the last byte of the desired information).

Look at the standard library's various network-aware code, you'll see 
lots of while loops for reading from sockets :) .  I'd personally 
suggest asyncore as a good module for study (it uses non-blocking 
sockets by default, incidentally).

HTH, and enjoy,
Mike

Locke wrote:
> The argument to recv is supposed to be the buffer, acording to the
> documentation. But the doc doesnt say: buffer measured in what? Is it bytes?
> Octets? (by the way, what is the diff between a byte and an octet?)
> 
> See, I am writing a pop3 client type thing, after requesting a message from
> the server, my first s.recv tells me how long the message will be in octets.
> So then I call s.recv again passing the number of octets as the argument to
> s.recv. Shouldn't this download the entire message? What actually happens is
> that it cuts off right in the middle of the message.
> 
> What's my problem? Is there an upper limit on what the buffer argument can
> be to recv? Or does pop3 just refuse to send the whole message with one call
> to recv? I don't think it is that, because what it sent me was like 600some
> characters, which isnt a power of two or any other computer-ish number.
> 
> Thanks
> 
> 
> 


-- 
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From marklists at mceahern.com  Wed Feb 13 15:50:12 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Wed, 13 Feb 2002 12:50:12 -0800
Subject: How to generate XML
In-Reply-To: <3C6AC01B.7070404@destiny.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBKENBKCAA.marklists@mceahern.com>

[Michael Chermside]
> The following code will create an xmlPrint() method. Pass it an object,
> and it prints it out in a pseudoreadable fashion. If I were really going
> to be using this, I'd modify _xmlPrintObj() so it checks for a .toXML()
> method and invokes that instead if such a thing exists. But there's no
> reason to write boilerplate in Python, where introspection is so easy.

There seem to be two bugs in the code you posted, Michael.  For what it's
worth, here's a fully working demo of the code with the minor bug fixes:

#! /usr/bin/env python
# xmlPrint.py
#
# Code from Michael Chermside's posting to comp.lang.python
#
# http://mail.python.org/pipermail/python-list/2002-February/087283.html

import sys
import types

def xmlPrint(obj, writer=sys.stdout, indentText='   '):
     """Print an object in XML format."""
     _xmlPrintObj(obj, _IndentingWriter(writer, indentText))

class _IndentingWriter:
     def __init__(self,
                  writer=sys.stdout,
                  indentText='   ',
                  indentLevel=0):
         self.writer = writer
         self.indentText = indentText
         self.indentLevel = indentLevel
         self.indentStr = indentText * indentLevel
     def incr(self):
         return _IndentingWriter(self.writer, self.indentText,
                                self.indentLevel + 1)
     def writeline(self, text):
         self.writer.write(self.indentStr)
         self.writer.write(text)
         self.writer.write('\n')

def _xmlPrintObj(obj, writer):
     """Print an object in XML format (internal interface)."""
     # Print tags for the object itself based on class name
     tagname = obj.__class__.__name__
     writer.writeline('<%s>' % tagname)

     # for all fields not beginning with an underscore...
     for fieldName in filter(lambda x: x[0]!='_', dir(obj)):
         _xmlPrintField(getattr(obj, fieldName), fieldName,
                        writer.incr())

     # close the tags for the object itself
     writer.writeline('</%s>' % tagname)

def _xmlPrintField(value, fieldName, writer):
     """Print a field of some object, labled with its fieldName."""
     if type(value) == types.InstanceType:
         # for sub-objects we print field name then recurse
         writer.writeline('<%s>' % fieldName)
         _xmlPrintObj(value, writer.incr())
         writer.writeline('</%s>' % fieldName)
     elif type(value) in (types.ListType, types.TupleType):
         # for sequences we print all items
         writer.writeline('<%s>' % fieldName)
         for v in value:
             _xmlPrintField(v, 'value', writer.incr())
         writer.writeline('</%s>' % fieldName)
     else:
         # for all other fields just use str()
         writer.writeline('<%s>%s</%s>' % (fieldName, value, fieldName))

if __name__=="__main__":
    class Person:pass
    p = Person()
    p.firstName = "Mark"
    p.lastName = "McEahern"
    p.favoriteColors = ['blue', 'purple']
    xmlPrint(p)




From kfarmer at thuban.org  Fri Feb  1 16:26:10 2002
From: kfarmer at thuban.org (Keith Farmer)
Date: Fri, 01 Feb 2002 21:26:10 GMT
Subject: PyAppBar -- bug fix, nifty widget
References: <7OP58.4001$KC5.2845539457@newssvr14.news.prodigy.com>
Message-ID: <SJD68.5984$NY2.3490816229@newssvr14.news.prodigy.com>

I've updated it again, adding the ability to resize (imperfectly),
dock/undock by moving the window, and undocked movement.

----------
Keith J. Farmer
kfarmer at thuban.org
http://www.thuban.org





From phd at phd.pp.ru  Mon Feb 11 03:54:36 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 11 Feb 2002 11:54:36 +0300
Subject: Three primary user Interfaces
In-Reply-To: <2082.807T1534T304260threeseas@earthlink.net>; from threeseas@earthlink.net on Mon, Feb 11, 2002 at 05:40:46AM +0000
References: <a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com> <5659.806T1220T5243918threeseas@earthlink.net> <YsC98.16409$Zu6.75381@news-server.bigpond.net.au> <2082.807T1534T304260threeseas@earthlink.net>
Message-ID: <20020211115436.B23093@phd.pp.ru>

On Mon, Feb 11, 2002 at 05:40:46AM +0000, Timothy Rue wrote:
> >   There was the MPW Shell available to programmers.
> 
> <shrug> only counts if user oriented. User oriented is one of the
> requirements to being a User Interface....

   It dependes on your definition of a "user". As old words are saying,
"UNIX is user friendly. It's just very selective about who it's friends
are."

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



From only at example.com  Sun Feb 10 14:24:04 2002
From: only at example.com (only at example.com)
Date: Sun, 10 Feb 2002 19:24:04 GMT
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org>
Message-ID: <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com>

Okay... you're a little confused about where the data from your form is
going. The data will not be STORED in script.cgi ...  script.cgi is
supposed to be a program (in this case, a python program) that processes
all of the VARIABLES (e.g. name, etc) from your form. Take a look at the
tutorial Devshed has and you will begin to understand how cgi works. 

http://www.devshed.com/Server_Side/Python/CGI/page1.html

Hope this helps.


In article <mailman.1013350037.30507.python-list at python.org>, "Chris
Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote:

> ----- Original Message -----
> From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>
> 
> 
>> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
>> news:mailman.1013316023.10782.python-list at python.org:
>> 
>> > "PERL type CGI's" should be read "scripts".  You can't run Python,
>> > Perl, Ruby,
>> > Lua, etc. from this particular server.
>> > 
>> > Try Xitami.  It is quite compatible with Python.  http://xitami.com
>> > 
>> Thanx. I'm downloading it as we speak.
> 
> I have used it a little; Steve Holden wrote about it in his book which I
> just finished reading.
>  
>> So with this server, will I be able to just send info direct from my
>> browser form to a file? If I can do that, then I'm up and running.
> 
> I'm not sure what you are saying here.  You don't send info "to a file,"
> you send it to a CGI script.  Your script must handle the data from
> there.
> 
>> [snip]
>> > 
>> > The form looks good, but where is script.cgi?
>> > 
>> In the same directory as the html form, cgi-bin.
> 
> Ah... what I meant was, that you should *post* the source to script.cgi.
>  Anyway, if your form is HTML it is bad form to have it in the cgi-bin
> folder.
> 
> GENERALLY (though it's not real safe to say this) CGI scripts can be
> stored in the cgi-bin folder, or in the htdocs/html/webroot folder; this
> depends on webserver settings.  The Xitami default allows CGI scripts
> only in the cgi-bin folder, but you can define that cgi-bin is actually
> the same as webroot.  Xitami recognizes executable files by extension on
> Windows, so your script should likely be called script.py (assuming it
> is Python).
> 
> Gah.  So.  Put script.py in the Xitami cgi-bin folder, and your form in
> the Xitami webpages folder, and if you are using the default
> configuration, you should be in good shape.  Start Xitami and then surf
> to http://localhost/yourform.html (whatever you called it) and when you
> click the Submit button, the form data is sent to
> http://localhost/cgi-bin/script.py (which should be in the action=
> attribute of the form tag, perhaps abbreviated as /cgi-bin/script.py).
> The script receives the data, does whatever, and then must send a new
> page to the web browser... this is why I want to see your script.
> 
> 
> 
>


From jkraska at san.rr.com  Wed Feb  6 19:49:17 2002
From: jkraska at san.rr.com (Courageous)
Date: Thu, 07 Feb 2002 00:49:17 GMT
Subject: What is the command to determine an object's type?
References: <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com> <marklists@mceahern.com>  of "Tue, 05 Feb 2002 13:52:14 PST." <NCBBLFCOHHDIKCAFGCFBAEAJKBAA.marklists@mceahern.com> <mailman.1013019261.19443.python-list@python.org>
Message-ID: <87k36ucstd24v0v32nv73td98v9ka0n7pl@4ax.com>

[Design Patterns]
>... at least half of it could be entitled 'my struggles with
>the C++ type system'. 

Nominated for quote of the day. :)

C//



From sholden at holdenweb.com  Thu Feb  7 08:34:55 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 7 Feb 2002 08:34:55 -0500
Subject: What's the current state of the "interfaces" proposal?
References: <gnF78.18266$zJ3.1417982@typhoon.maine.rr.com>
Message-ID: <Tnv88.94185$KM2.3403823@atlpnn01.usenetserver.com>

"Mike Carifio" <carifio.nospam at nospam.usys.com> wrote in message
news:gnF78.18266$zJ3.1417982 at typhoon.maine.rr.com...
> I'm not sure if this is the right place to ask this question. Is there a
> better forum?
> I started at http://python.sourceforge.net/peps/pep-0245.html and got a
> little lost
> following some of the links. It looks like "interfaces" has been
"accepted"
> as a future
> part of python. Is that accurate? What's the time frame? Pl. advise.
Thanks.
>
No, it's listed in PEP 0 as "open (under consideration for 2.3)". As far as
I know there are no current patches tom implement it, so it's at least
*that* far off inclusion. I was baiting Alex Martelli, one of the PEP's real
fans, at the Conference about how he should write the patches if he likes
the PEP (unlike me, Alex *could* write the patches if he chose to and had
the time). But no result yet ;-)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From nas at python.ca  Wed Feb 20 17:33:07 2002
From: nas at python.ca (Neil Schemenauer)
Date: Wed, 20 Feb 2002 14:33:07 -0800
Subject: Warning about "for line in file:"
In-Reply-To: <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com>; from jason@jorendorff.com on Wed, Feb 20, 2002 at 02:01:43PM -0600
References: <just-8C3043.11330718022002@news1.xs4all.nl> <HFEKILOLEFEFMKAECNDLMEBCDEAA.jason@jorendorff.com>
Message-ID: <20020220143307.A2018@glacier.arctrix.com>

Jason Orendorff wrote:
> I agree!  This could be improved by having file.__iter__()
> always return the same object.  But that doesn't solve the problem
> in full.

Perhaps the buffer should be attached to the file object itself instead
of to the iterator.  read() and readline() could then be changed to use
the buffer if it exists.

  Neil



From claird at starbase.neosoft.com  Fri Feb 15 15:44:17 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Feb 2002 14:44:17 -0600
Subject: Tkinter and Greek Symbols
References: <3daa1aa8.0202141718.3500c462@posting.google.com> <m3heoj5fda.fsf@mira.informatik.hu-berlin.de> <3c6d3278$1@news.mt.net.mk>
Message-ID: <EADB9ECB38A1CC89.EB3E089A1460B627.42CCB2B5E254D906@lp.airnews.net>

In article <3c6d3278$1 at news.mt.net.mk>,
=?iso-8859-5?Q?=B4=D0=DC=F8=D0=DD_=B3=2E?=
  <=?iso-8859-5?Q?=DF=D5=DD=D3=E3=D8=DD=D8=E1=E2=D0?=@=?iso-8859-5?Q?=DC=D0=D8=DB?=.=?iso-8859-5?Q?=DD=D5=E2?=.=?iso-8859-5?Q?=DC=DA?=> wrote:
			.
			.
			.
>Does this mean that Tkinter support unicode?
>Does it support the iso10646-1 fonts too?
>Can I display Symbols from several alphabets n a single widget?
>And how do I get the keyboard input, in unicode od locale specific?
			.
			.
			.
Yes, I think so, yes, and it depends.

I hope Fredrik drops in soon with details.
-- 

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


From mwh at python.net  Wed Feb 13 12:03:02 2002
From: mwh at python.net (Michael Hudson)
Date: Wed, 13 Feb 2002 17:03:02 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <a4e44s$tli$1@netnews.upenn.edu>
Message-ID: <ud6z9gvmd.fsf@python.net>

Paul Hughett <hughett at mercur.uphs.upenn.edu> writes:

> It could be worse.  In one language I worked in, "10" was an octal
> integer, "10." was a decimal integer, and "10.0" was floating point.
> Nice language otherwise, though.

Maclisp?

Cheers,
M.

-- 
  "The future" has arrived but they forgot to update the docs.
                                        -- R. David Murray, 9 May 2000


From peter at engcorp.com  Tue Feb 12 19:16:14 2002
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 12 Feb 2002 19:16:14 -0500
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <Pine.LNX.4.33.0202101938190.17089-100000@ns0.tbye.com> <mailman.1013534182.27264.python-list@python.org> <e5dfe6d0.0202121435.79da2484@posting.google.com>
Message-ID: <3C69B04E.3FED0F96@engcorp.com>

George Partlow wrote:
> 
> "Afonso Fernandez Nogueira" <afonso at writeme.com> wrote in message news:<mailman.1013534182.27264.python-list at python.org>...
> 
> > Esperantists always use the -ist- suffix ("Esperantistoj") to refer to
> > themselves though. I've seen several webpages about Python written in
> > Esperanto, so I'm sure there has to be an established way to call
> > Pythonistas in this beautiful language.
> 
> How established it is I don't know, but I did find (using Google) an
> occurence of "pitonisto" at
> <http://www.geocrawler.com/archives/3/3453/2000/5/0/3808756/>.

Ha!  Even in Esperantujo newcomers to Python, trained in the "rules" of
"proper" object oriented programming, believe the lack of strict private
variables will tend to cause code to break when the implementation of a 
class changes.


From borcis at geneva-link.ch  Tue Feb 26 06:23:43 2002
From: borcis at geneva-link.ch (Boris^2)
Date: Tue, 26 Feb 2002 12:23:43 +0100
Subject: Time to rename Stackless?
References: <3C7A151F.1090709@tismer.com> <mailman.1014690284.11692.python-list@python.org> <mnvl7ukagdojifcev30gibrj1dr7c7euja@4ax.com>
Message-ID: <3C7B703F.3040002@geneva-link.ch>


Courageous wrote:

>>>It is still stackless in the sense that order of execution is not
>>>restricted by the C stack.
>>>
>>Even so, I still think it's a poor name, because it
>>focuses on an implementation detail rather than
>>telling you what it can do for you.


Imho, matters of stacks can't quite pass as a marginal detail for a 
programmer. Or if they are assumed to, then an ad-hoc parable might 
serve as interface, e.g. "stack" freed from the programmers' context 
evokes the stack of money or equivalent tokens, and hence "stackless" 
may refer to accessing the Father's Realm through the proverbial 
camel-stopping pinhole :)


> meanwhile, most of the people that care about
> Stacklessness and continuations are hardcore computer
> science junkies, folks who've been exposed to some fairly
> obscure languages (sorry, scheme guys!),


Your use of "obscure" I find obscurantist (and I am really no "scheme 
guy" by any stretch of imagination).

Boris
--
filter(lambda W : W not in "ILLITERATE","BULLSHIT")



From marklists at mceahern.com  Thu Feb  7 15:34:09 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Thu, 7 Feb 2002 12:34:09 -0800
Subject: 2001 Turing Award
In-Reply-To: <2jnu3a.4sq.ln@wombat.san-francisco.ca.us>
Message-ID: <NCBBLFCOHHDIKCAFGCFBMEHNKBAA.marklists@mceahern.com>

Keith Keller wrote:
> Damn--I thought the award was for the person who best impersonated
> a Turing machine.  :)

Do you have any nominees?

// mark



From gtalvola at nameconnector.com  Wed Feb 20 11:43:55 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Wed, 20 Feb 2002 11:43:55 -0500
Subject: Job Opening in Medford, MA
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E41240D4@MAILBOX.nameconnector.com>

Software Development Engineer

Parlance Corporation (Medford, MA)

Parlance Corporation is an exciting, visionary small company in the Boston
area that is using speech recognition technology to change the way people
use the telephone...forever! As a member of the Parlance team you'll be
joining the group that will make this happen!

When you join this growing development team, you'll be responsible for
developing Web-based reporting and administration capabilities for our
customers. Tools used include Windows 2000, SQL Server, Python, Webware for
Python, and Apache.

Ideal candidates have the following qualifications:

- Several years experience in the software industry
- Experience with Python
- Experience developing dynamic web sites
- Strong HTML and CSS design skills; familiarity with JavaScript
- Strong RDBMS and SQL knowledge
- Experience with Windows NT/2000, MS SQL Server, Apache, ADO, Visual Basic
and/or MS Access
- C++ experience a plus

Interested parties should submit their resume to Geoffrey Talvola at:
gtalvola at nameconnector.com

For more information about Parlance Corporation see:
http://www.nameconnector.com/

-- 

- Geoff Talvola
  gtalvola at NameConnector.com



From shalehperry at attbi.com  Thu Feb  7 01:36:37 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 06 Feb 2002 22:36:37 -0800 (PST)
Subject: Delphi vs Python
In-Reply-To: <3C62170D.5095436D@engcorp.com>
Message-ID: <XFMail.20020206223637.shalehperry@attbi.com>

On 07-Feb-2002 Peter Hansen wrote:
> Bo Vandenberg wrote:
>> 
>> I'm planning learning strategy and I thought I would ask people who use both
>> Python and Delphi their opinion.
> 
> I used Delphi (back in the 3 and 4 days, I think) extensively for several 
> years, and had a lot of respect for the environment and language.  Very
> productive for me, relatively few bugs (compared to C++ and C, roughly
> similar to my levels with Java).
> 

This is mostly a "me too" post.  With the addition of attributes in 2.2 python
now does almost everything syntactically Delphi does.  I do on occasion miss
the 'with' keyword, nice bit of sugar that.

For me what made Delphi nice was how much it abstracted the Window's API away
and the ease of creating GUI apps by putting things where I wanted.  I have yet
to find another environment that made good looking apps that fast.  If I could
have one of those that outputs python code AND was not proprietary I would be
quite happy.  glade is a very pale comparison.

If your goal is to write GUI apps, Delphi may be a good place.  Python is great
as an all around language and it really can churn out code fast.  On top of
that you have the interpreter available to you by just running python without a
script AND it is available to you in your running code (introspection).  Being
able to quickly test an idea by just firing up python is one of the things I
love most.  Hell, I even use it as a desk calculator (whoever made '_' return
the last result -- I love you).



From jason at jorendorff.com  Mon Feb  4 20:00:22 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 4 Feb 2002 19:00:22 -0600
Subject: Why it does NOT work on Linux ?
In-Reply-To: <Fdx78.39388$KM2.1215789@atlpnn01.usenetserver.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEFGDAAA.jason@jorendorff.com>

Steve Holden wrote:
> Markus Sch?nhaber wrote:
> > 1.) Here
> > > IDs=re.compile('.*<a href="lead.asp\?ItemID=(\d{5,10}).*')
> > > Results=re.findall(IDs,Text)
> >
> > you call re.findall with a regular expression object as a first
> > parameter which should be a string. What you want to do is
> >
> > Results = IDs.findall(Text)
> >
> > i. e. call the appropriate method on the re object you created.
> >
> While your suggestion is correct, it is incorrect to say that Ladislav's
> method is wrong.

Ladislav's use of findall works, but this feature of re.findall()
is not documented and should be avoided, as undocumented features
have been known to disappear.

## Jason Orendorff    http://www.jorendorff.com/




From sandeep182 at hotmail.com  Tue Feb 26 17:08:41 2002
From: sandeep182 at hotmail.com (Sandeep Gupta)
Date: Tue, 26 Feb 2002 22:08:41 GMT
Subject: Need advice for obfuscating Python code
Message-ID: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>

Hi,

I've written a product in Python and I wanted to find out how best to
obfuscate the code.  I want the distributed files to be moderately hard to
reverse engineer.

One option is to compile the code and distribute the .pyo files.  Are there
any better alternatives?

Thanks

Sandeep




From sholden at holdenweb.com  Tue Feb 12 09:21:04 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Feb 2002 09:21:04 -0500
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <3C688CED.4583E7DA@cosc.canterbury.ac.nz> <7xeljqq1lo.fsf@ruckus.brouhaha.com>
Message-ID: <ax9a8.173237$KM2.7621260@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xeljqq1lo.fsf at ruckus.brouhaha.com...
> Greg Ewing <greg at cosc.canterbury.ac.nz> writes:
> > I'm not sure that coarse/fine is the right way to describe
> > the distinction even in normal unidextrous (is that the
> > right word?) people. Consider the fact that most right-handed
> > guitar players use the left hand for fingering, and the
> > right hand for the relatively dextrously undemanding
> > function of strumming. Why is that?
>
> You have to precisely control the timing of hitting the strings with
> the right hand in order to not mess up the music.  Since the left
> hand's actions don't actually make sound

Unless, of course, you use hammered notes, which are made with the left hand
and typically follow a plucked note on the same string.

> , the timing requirements
> aren't so precise as long as stuff happens fast enough.

how-long-is-it-since-i-picked-up-my-guitar-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From kragen at pobox.com  Sat Feb  2 21:25:29 2002
From: kragen at pobox.com (Kragen Sitaker)
Date: 02 Feb 2002 21:25:29 -0500
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it>
Message-ID: <83bsf7z4h2.fsf@panacea.canonical.org>

Alex Martelli <aleax at aleax.it> writes:
> Kragen Sitaker wrote:
>         ...
> > and "x in somedict", which makes it impossible to have a class
> > instance faithfully emulate both dictionaries and lists.  I'm not
> 
> No "make it impossible", because it ALREADY was impossible
> before.  How "faithfully" am I emulating a dictionary if rebinding
> x[-1] changes, say, x[5]?!  Yet that is part of what I MUST do
> to "faithfully emulate" a list which has 6 items at this time.

Well, I had a filesystem mapping; indexing by strings was associative
and accessed files by name, while indexing by integers was sequential
and iterated over the files in a directory.

Faithfully emulating a dictionary doesn't mean you can't do things
dictionaries don't do --- it just means you can do everything
dictionaries can.

> We need to reach a new stable plateau reasonably fast, at least as
> badly as we need to keep backwards compatibility.

What do you mean by "stable"?  What's "unstable" about having to put
'from __future__ import division' at the top of your module to get
reasonable division behavior for the next two, three, five, or ten
years?

> Right now we're clearly smack in the middle of it -- the worst of
> times to start slowing down.  I suspect Guido has an excellent handle
> on the balance of THIS tradeoff just like he has on the huge number
> of tradeoffs that cumulatively make a language.

Well, he's certainly better at it than I am.



From emile at fenx.com  Wed Feb 27 01:51:25 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 26 Feb 2002 22:51:25 -0800
Subject: Numeric arrays - initialising with binary data
References: <cf0ad9fb.0202250446.3df07a08@posting.google.com> <a5dc4f$6dr9c$1@ID-11957.news.dfncis.de> <cf0ad9fb.0202261010.4c778108@posting.google.com>
Message-ID: <a5hvvi$7i95s$1@ID-11957.news.dfncis.de>

"Jesper Olsen" 
> The most recent I can find is this
> 
> http://www.pfdubois.com/numpy/html2/numpy-9.html#pgfId-36538
> 

That's newer than mine.  

-- 

Emile van Sebille
emile at fenx.com

---------



From thomas at gatsoft.no  Tue Feb 26 02:36:13 2002
From: thomas at gatsoft.no (Thomas Weholt)
Date: Tue, 26 Feb 2002 07:36:13 GMT
Subject: Support for resume in HTTP-server
Message-ID: <NVGe8.99$5vh.176191488@news.telia.no>

Some web-servers ( and ftp-servers too ) support resume; the ability to get
the rest of a previous partial download. How do I support this using
BaseHTTPServer?

Any clues?

Regards,
Thomas Weholt




From gh_pythonlist at gmx.de  Fri Feb 22 17:34:55 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Fri, 22 Feb 2002 23:34:55 +0100
Subject: hep needed with pipes (popen4)
In-Reply-To: <iUyd8.18492$X2.205012@nnrp1.uunet.ca>
References: <iUyd8.18492$X2.205012@nnrp1.uunet.ca>
Message-ID: <20020222223455.GA3329@lilith.hqd-internal>

Le 22/02/02 ? 16:38, Darcy Kahle ?crivit:
> I need some help with pipes in python 2.1.
> 
> I am running a comand using os.popen4().  The command produces output until
> the user hits control-C, at which time, it prints a summary, and exits.  I
> want to send that character down the pipe to it without the user doing so
> (control-c kills the script, by default).  I tried sending "\x003", which to
> the best of my knowledge, is the hex code for it, but it didn't work.

AFAIK you need to send the process a signal.  Using the signal module
and os.kill the code you need might look not entirely unlike this:

import signal, os

def sig_handler(signum, frame):
    global child_pid
    os.kill(signal.SIGINT, child_pid)

# open child process and store pid in global variable child_pid

signal.signal(signal.SIGINT, sig_handler)
while 1:
    pass
    # whatever

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.8 ?C      Wind: 5.9 m/s



From usenet at thinkspot.net  Sun Feb 24 02:25:41 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 23 Feb 2002 23:25:41 -0800
Subject: string substitutions
References: <c4e6b17d.0202231152.5f87765d@posting.google.com> <3c787c1f.2523859@news.t-online.de>
Message-ID: <a598dm.3vshvrl.1@kserver.org>

On Sun, 24 Feb 2002 05:51:46 GMT, gerson.kurz at t-online.de (Gerson Kurz)
wrote in comp.lang.python in article
<3c787c1f.2523859 at news.t-online.de>:

> So, the simple "while"-solution (test1) is actually the fastest if you
> really only want to replace spaces (and not newlines and tabs, also).
> I find that quite interesting because the other three solutions seem
> so much more sophisticated.

Very interesting, indeed. I felt so silly after I posted that one, and
it was followed by the other more "sophisticated" solutions. Actually, I
had replace my test #1 with #4, since I wanted to replace any white
spaces with a single space, as it turns out.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From philh at comuno.freeserve.co.uk  Wed Feb  6 11:56:13 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 6 Feb 2002 16:56:13 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <mailman.1013007554.26921.python-list@python.org>
Message-ID: <slrna62o1d.b33.philh@comuno.freeserve.co.uk>

On Wed, 06 Feb 2002 09:56:46 -0500, Justin Sheehy <justin at iago.org> wrote:
>philh at comuno.freeserve.co.uk (phil hunt) writes:
>
>> And what if I decide that for my particular application, that's just
>> the behaviour I want?
>
>I wonder why your reaction is so strong here.

In the past I've been fucked about by languages which mandate that
there is One True Way to write software, and all other ways are forbidden.
If I wanted that attitude I would code in Pascal not Python.

Any programming system that takes this approach is one that I simply
am not interested in learning to use.

>> Or do you think that you know better than all the users of your
>> macro system, even though you've not met them and don't know
>> anything about the software they are writing?
>
>At this point, he does know that, since he is presumably the only user.

But he won't be if his system gets popular.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From andda715 at student.liu.se  Fri Feb  1 12:49:11 2002
From: andda715 at student.liu.se (Anders Dahlberg)
Date: Fri, 1 Feb 2002 18:49:11 +0100
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012578783.8284.python-list@python.org>
Message-ID: <a3ekkn$1mv$1@news.island.liu.se>

<brueckd at tbye.com> skrev i meddelandet
news:mailman.1012578783.8284.python-list at python.org...
> On Thu, 31 Jan 2002, Bryan wrote:
>
> > > I'm still not willing to give java the benefit of the doubt yet.
> >
> > Agreed. The test suites are rather limited in nature. I would welcome
the
> > results of a more in-depth analysis. However, the trollish nature of my
> > post and resulting low number of replies leads me to believe no one can
or
> > is willing to really back up any claims to the contrary.
>
> That's why I asked whether or not the JIT was enabled in the tests and, if
> so, if you could re-run them without it. Maybe the low number of replies
> is because it's not suprising that a semi-compiled-to-native program is
> faster?

well, but why would you want to compare interpreted java with python? As
java is not supposed to be *only* interpreteded anymore (it once was in the
cold distant past - like 4-5 years ago) but compiled ("jitted") to native
code during run-time? If you were to compare interpreted java with python
should you then compare interpreted c++ with java (same ballpark)?

just some observations ;)

Would be fun to see comparisons on jython (running on for example jdk1.4)
versus python2.2!

>
> -Dave
>
/Anders




From caljason76 at yahoo.com  Tue Feb 19 19:34:12 2002
From: caljason76 at yahoo.com (Jason)
Date: 19 Feb 2002 16:34:12 -0800
Subject: add-a-gram
Message-ID: <e9e6dd1f.0202191634.21da06b9@posting.google.com>

Here is a puzzle that was floating around comp.lang.lisp, and I was
wondering what a Python solution would be.  I am still learning Python
and am very curious how Python solutions differ from other languages,
like Perl or C.  Are there some set of maxims to the Python
programmer, and are there Python idioms for problem solutions?

Appended to the message is the problem, you can find the problems and
word list http://www.itasoftware.com/careers/programmers.php

----

Add-A-Gram 
An "add-a-gram" is a sequence of words formed by starting with a
3-letter word, adding a letter and rearranging to form a 4-letter
word, and so on. For example, here are add-a-grams of the words
"CREDENTIALS" and "ANACHRONISM":
 
ail + s = 
sail + n = 
nails + e = 
aliens + t = 
salient + r = 
entrails + c = 
clarinets + e = 
interlaces + d = 
CREDENTIALS (length 11) 

mar + c = 
cram + h = 
march + s = 
charms + o = 
chromas + n = 
monarchs + i = 
harmonics + a = 
maraschino + n = 
ANACHRONISM (length 11)


From db3l at fitlinxx.com  Tue Feb 12 17:16:15 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Feb 2002 17:16:15 -0500
Subject: Just for the gurus -- Remote admin password changing with privileged user ??
References: <d7d5ebdf.0202121332.4f7d3723@posting.google.com>
Message-ID: <ur8nqmjm8.fsf@ctwd0143.fitlinxx.com>

matthias_j at lycos.com (Matthias Janes) writes:

>             BUT THE REAL PROBLEM is that I do not have admin rights on
> the other domains.

Well, you're going to have to have admin rights on those
domains/machines to make the change.  I assume what you really mean is
that your local Matthias account doesn't have those admin rights.

> I could copy the script to all the Domains and execute it logged on as
> Admin - but I would realy like to avoid this if possible.  Maybe there
> is a way to send an authetification to the domains like
> username:TechAdmin + the old passwort - (like a Su command) I'm not
> sure..... -- ANY HELP

In general, the network calls don't have their own authentication
information provided in each call.  Instead, NT establishes an IPC
link to the remote machine, and keeps cached authentication credential
information that is then used for any remote requests to that same
machine.

The simplest way I know to control the credentials used is to
establish a network use across to the remote machine in question.  You
can use any given account/password combination when doing that.  Once
you have one network link to a remote machine with authentication
information, NT will continue to use that same authentication
information for other remote requests of any type to that machine.  In
fact, in some cases that's a major pain.

It doesn't have to be any formal resource on the remote machine, just
use the automatic IPC$ share (which is going to be grabbed anyway once
you try to make the remote network request).  You just have to make
sure that you establish your own network use before doing any other
network operation, or else you'll already have an IPC$ use with
default credentials (e.g., Matthias).

You can establish the share with NetUseAdd, and remove it when done
with NetUseDel.  I haven't done it myself yet, but once you have that
share, I would think that NetUserChangePassword (or NetUserSetInfo) to
that machine might work.

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


From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 08:39:53 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 07:39:53 -0600
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <3C67C369.597E3517@optonline.net>
Message-ID: <003001c1b301$aba76ca0$0101010a@local>

----- Original Message ----- 
From: "Frederick H. Bartlett" <fbartlet-fixit at optonline.net>


> Years ago, my father -- a professor emeritus of music at the 
> University of Western Ontario -- coined the acronym ETD: Ego-Talent 
> Discrepancy.
> 
> He never published, so he can't get official credit for the 
> observation; but I try to do what I can to get the coinage into 
> common currency.

What's his name, so we'll know who to credit for this gem?





From jason at jorendorff.com  Sat Feb  2 19:39:04 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sat, 2 Feb 2002 18:39:04 -0600
Subject: hex to normal chars?
In-Reply-To: <a3hq9f+2q3f@eGroups.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMEPLCPAA.jason@jorendorff.com>

maximillianscherr writes:
> how can i transfer hex numbers to normal chars, like "41" to "A" etc.?

>>> '41'.decode('hex')
'A'
>>> 'jason'.encode('hex')
'6a61736f6e'
>>> '6a61736f6e'.decode('hex')
'jason'

## Jason Orendorff    http://www.jorendorff.com/




From tim.one at home.com  Sat Feb  2 00:09:15 2002
From: tim.one at home.com (Tim Peters)
Date: Sat, 2 Feb 2002 00:09:15 -0500
Subject: Equivalent of Perl chomp?
In-Reply-To: <ffB68.107390$8b.4046814@atlpnn01.usenetserver.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEFONJAA.tim.one@home.com>

[Steve Holden]
> 1,000,000 iterations.
>
> Testing for length 1:
> Endswith 0.00000505
> sliceIt  0.00000209
>
> Testing for length 2:
> Endswith 0.00001137
> sliceIt  0.00001148
>
> Length 3:
> Endswith 0.00001055     [!]
> sliceIt  0.00001264
>
> Of course, these tests are with the tests succeeding. I'm
> presuming failure
> would shorten the timings, but I'm not going to bother checking that
> assumption.
>
> semi-believab-ly y'rs  - steve

This is the kind of fun I was hoping the OP would have.  Good job!  A lesson
to take is that timing even seemingly trivial things can turn up major
surprises with seemingly trivial changes in exactly what it is you're
timing.  The results are almost never accidental, but determining the true
causes can leave you crawling on your belly for weeks begging for relief.

A few things going on in this test under the covers:

+ endswith does indeed pay a large price for building a bound
  method object every time.  It's not only the dict lookup to
  resolve "endswith", but also malloc and free to build the
  bound method object and discard it afterwards.

+ The huge increase in sliceit's time going from 1 to 2 is due to
  that s[:-1] creates a one-character string, and Python interns
  those by magic:  every instance of, e.g., 'a', no matter how
  computed, is exactly the same object (that's not actually true,
  but it's close <wink>).  But 2-character computed strings are
  not interned, so s[:-2] has to malloc a fresh string object each
  time to hold the two characters, and free it later.  endswith
  doesn't have to build new string objects, so doesn't care as much.

Now for some real fun:  I have no idea why endswith took a huge jump going
from 1 to 2.  The saving grace for me is that it's so easy to find timing
mysteries that I'd go mad if I felt compelled to chase them all down, so I
can walk away without even looking back.  I hope not everyone is so
well-adjusted <wink>, though, as I'd like to know why.

like-living-in-pumpkin-ly y'rs  - tim




From sjmachin at lexicon.net  Sun Feb 17 22:26:02 2002
From: sjmachin at lexicon.net (John Machin)
Date: 17 Feb 2002 19:26:02 -0800
Subject: python10 digest + articles at pythonjournal.cognizor.com  comprise PythonJournal 2(2)
References: <mailman.1013981368.9153.python-list@python.org>
Message-ID: <c76ff6fc.0202171926.36131461@posting.google.com>

editor at pythonjournal.com wrote in message news:<mailman.1013981368.9153.python-list at python.org>...
> Python Journal new issue 2(2) is out and about at pythonjournal.cognizor.com
> with articles and a fairly detailed digest of the content of the 4 very full 
> days of the 2002 Python10 conference.   Focus in the articles of this issue 
> is on deepening understanding of how python works -- recursion, PySpy.
> 
> First -- The pythonista thinks about repeating functions: Alternative
> recursion methods with iteration splitoff auxiliaries, having different oopses
> and efficiencies. Then -- Inspect your own pet python's looping and leaping.
> In realtime! Tutorial on Malcolmson's PySpy shows what documentation ;)
> looks like inside out, how basic program structures including recursion
> vs. iteration really slither along. Screenshots, congeners  (and cf.
> inspect.py). 
> 

The journal quotes an URL http://www.malcolmson.com/pyspy ... this cracks a 404.

http://www.malcolmson.com/ relates to a photographer named Matt Malcolmson.

What has happened to Tom the PySpy man??


From gerson.kurz at t-online.de  Wed Feb 20 15:17:57 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Wed, 20 Feb 2002 20:17:57 GMT
Subject: a question regarding conciseness
References: <a50umn$1dou@r02n01.cac.psu.edu> <3C740BB4.3070703@free.fr>
Message-ID: <3c7403d7.25939921@news.t-online.de>

On Wed, 20 Feb 2002 21:48:52 +0100, Christophe Delord
<christophe.delord at free.fr> wrote:

>Writting a function to sort lists may not be efficient :
>
>def my_sort(L):
>     L2 = L[:]
>     L2.sort()
>     return L2
>
>X = [...]
>for x in my_sort(X):
>     ....

Of course, you can always use quicksort:

q=lambda x,o=filter,r=lambda o:(lambda x=list(o)+[0],\
y=len(o)-1:(lambda:map(lambda i,s=x.__setitem__:s(-1,\
x[i])or s(i,x[y-i])or s(y-i,x[-1]),range((y+1)/2))and\
0)() or (lambda:"".join(x[:-1]))())(),s=')]1=<)x(nel'\
'[]"x",))"o(lambda i,j=x[0]:i"(r,"~"(ecalper.)"q(~<j'\
',x))+~==j,x)+q(~>j,x))"(r[(lave':eval(r(s))

import random
test = range(20)
random.shuffle(test)
print test, "->", q(test)



From eppstein at ics.uci.edu  Thu Feb 28 19:35:50 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 16:35:50 -0800
Subject: PEP 276 (was Re: Status of PEP's?)
References: <mailman.1014938330.6533.python-list@python.org>
Message-ID: <eppstein-CFD487.16355028022002@news.service.uci.edu>

In article <mailman.1014938330.6533.python-list at python.org>,
 James_Althoff at i2.com wrote:

> [David Eppstein]
> > Maybe it's as natural as Perl's ability to concatenate numbers?
> 
> I don't appreciate your sarcasm.  Maybe others do.

I'm sorry that my sarcastic tone seems to have obscured what I think is a 
real point: one feature of Python is that, although variables are untyped, 
their values are strongly typed: what you can do with a value depends on 
its type.  So, you can concatenate strings but not integers; you can 
multiply integers but not strings.  The credo "explicit is better than 
implicit" is in part about this: if you want to concatenate the decimal 
representation of two numbers, you should use an explicit function to 
convert the numbers into strings.  Your proposal, by blurring the 
distinction between integers and ranges of integers, decreases this level 
of explicitness in Python.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From martin at v.loewis.de  Wed Feb 27 15:56:41 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 27 Feb 2002 21:56:41 +0100
Subject: Status of PEP's?
References: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com>
Message-ID: <m3k7syiqye.fsf@mira.informatik.hu-berlin.de>

Gon?alo Rodrigues <op73418 at mail.telepac.pt> writes:

> Anyone knows how is the "acceptance status" of the following PEP's, that
> is, is there any idea if they will be accepted for the next Python
> realease:
> 
> PEP 276: Iterator for ints.
> (I really love this one - as simple and "obvious" as it gets)
> 
> PEP 279: Enhanced generators

You'll need to ask the authors. If they feel it is ready, they need to
ask for BDFL pronouncement.

Regards,
Martin



From wayne at mishre.com  Sat Feb 16 22:53:30 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Sat, 16 Feb 2002 19:53:30 -0800
Subject: Have to try
References: <004501c1b763$78bca9c0$b278c53f@pacbell.net>
Message-ID: <3C6F293A.4030705@mishre.com>

Rick,

Rick Hamilton wrote:
> OK - I know I am going to get fried here but - I am out of options. I'm 
> a programmer (Java, VB and C++) and am taking a class on Operating 
> systems. The nut teaching this course is asking us to do a ton on 
> programming in Python. OK -

   You've come to the wrong newsgroup if you want to get flamed, sorry. 
  Is the professor a nut because he's using Python?  Or does he go 
galloping around the room with two coconut halves talking about swallows?

> So I am trying to do a (what should be) simple thing. I have a string 
> stored in a variable. I want to search that string for the occurrence of 
> a sub string. I have tried
> 
> Find(s,sub)
> 
> Rfind
> 
> MyString.find(sub)

Let's say you have a variable "var" with a value of "my string"

var = "my string"

If you are using a recent version of Python you can do the following:

 >>> var.find("tr")
4

Since var is a string variable all the string functions can be accessed 
directly using dot-notation.  The documentation on the web site for the 
String module is the older way of doing it.  To use that you would need 
to type:

import string

Before you could use the functions, then to use them you would type:

 >>> string.find(var, "tr")
4

> Nothing - I have not yet figured out the appropriate way to treat a 
> string and am heading for a very weak book that we have as a reference. 
> Anyone will to toss a fellow programmer a helping had I'm willing to get 
> flamed to get this over with.

If all else fails try the following:

 >>> dir()
['__builtins__', '__doc__', '__name__', 'var']

(this is what mine shows)

This is your current namespace, if you wanted to see information about 
var you could type:

 >>> dir(var)
['capitalize', 'center', 'count', 'encode', 'endswith', 'expandtabs', 
'find', 'index', 'isalnum', 'isalpha', 'isdigit', 'islower', 'isspace', 
'istitle', 'isupper', 'join', 'ljust', 'lower', 'lstrip', 'replace', 
'rfind', 'rindex', 'rjust', 'rstrip', 'split', 'splitlines', 
'startswith', 'strip', 'swapcase', 'title', 'translate', 'upper']

To use any of these you can just type var.<function> .

TTFN,

Wayne




From phr-n2002a at nightsong.com  Wed Feb 27 05:08:21 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 27 Feb 2002 02:08:21 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com>
Message-ID: <7xk7sznsoa.fsf@ruckus.brouhaha.com>

"Terry Reedy" <tjreedy at home.com> writes:
> > Local variable declarations are for code cleanliness.  Currently, using
> > some name for a temp variable leaves detritus around after you're done
> > with it.  To figure out what the program is doing, you have to check
> > whether further references to 'temp' really wanted the value left from
> > someplace far away, or were holding just a different temporary value.
> > When the usefulness of a variable ends, its scope should also end.
> 
> An explicit 'del temp' ends the scope of the variable.  Its not much
> more to type that any sort of brackets and allows for overlapping
> scope of temporaries instead of artificially forcing nesting.  If you
> want to extend the scope of 'temp', it is probably easier to get rid
> of the explicit deletion than a pair of brackets.

"del temp" ends the inner scope by destroying the outer scope!  That's
throwing the baby out with the bathwater.  Sorry, no cigar.  It's
really a deficiency that there's no way to create a new local scope.

Also, there's no way to make sure that the "del temp" at the end of
a block actually gets run, unless you put try...finally around it,
which is a big mess.


From jeff at ccvcorp.com  Fri Feb 15 13:48:27 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 15 Feb 2002 10:48:27 -0800
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org> <3C6C3805.37BD87FB@ccvcorp.com> <tyfk7tfduo6.fsf@pb-s-513-1-26-6-1.cern.ch>
Message-ID: <3C6D57FB.1F6E10C8@ccvcorp.com>


Jacek Generowicz wrote:

> Jeff Shannon <jeff at ccvcorp.com> writes:
>
> > This works, but is generally a Bad Idea(tm).  The problem
> > with eval/exec on arbitrary strings (and any string you read
> > in from a file is arbitrary) is that it's hard to be
> > positive that what you're eval/exec-ing is what you expect.
> > And if it's *not*, then many, many very bad things can
> > happen.  (Imagine someone "accidentally" replacing that
> > textfile with one that contains the line "import
> > os;os.system('rm -s /')" -- suddenly your entire filesystem
> > is blank....)
>
> Writing python scripts works but is generally a Bad Idea(tm).
>
> Imagine somone "accidentally" replacing the textfile containing your
> script with one that contains the line "import os;os.system('rm -s
> /')" -- suddenly your entire filesystem is blank....)

Okay, so that particular example is not the best.  ;)  However, I've
seen people posting code that eval()'s interactive input, which *does*
provide all sorts of opportunity for bad things.  The point is still
true, that exec and eval() are better avoided in 99.9% of situations;
they're emergency-use-only.  (It's good to keep a fire extinguisher
around, but I wouldn't want to use it to put out a candle...)

Jeff Shannon
Technician/Programmer
Credit International




From usenet at thinkspot.net  Sat Feb 16 16:11:55 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 16 Feb 2002 13:11:55 -0800
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no>
Message-ID: <a4llqr.3vvdhqr.1@kserver.org>

On 16 Feb 2002 18:08:10 +0100, Jan-Fredrik Braseth
<janfbr at sex.ifi.uio.no> wrote in comp.lang.python in article
<xyuofipl5hh.fsf at sex.ifi.uio.no>:

> If you look closely at the traceback you see that the error with opt is in the
> telnetlib.py, that's a module for python. But it's unlikely that there is a bug
> in python itself, isn't it?
> 
> BTW: it's python 2.2

Oops! Sorry. I didn't look closely.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From chris.gonnerman at newcenturycomputers.net  Thu Feb  7 17:27:46 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Thu, 07 Feb 2002 16:27:46 CST
Subject: Apology
In-Reply-To: <Xns91AE6032817F5ahmebahhotmailcom@198.99.146.10>
Message-ID: <200202072231.QAA19882@scotland.k12.mo.us>

Consider this an apology to the list, regarding my recent Survivor-like
suggestion that we "vote off" Rue.  It was late, I wasn't feeling good, but
still, it was in bad taste, and for this I'm sorry.

I've always made it a point to admit my mistakes and accept due blame.
This isn't because I am more virtuous, but rather just to be different (since
so few people I know do it...)

My hope was not to cause an official banning but to show to Rue that no
one understood him and that no one really cared anymore.  I could have
been proven wrong just as easily, and was aware of that possibility; but
it had honestly not occurred to me that it was a "bad thing" to suggest.

In retrospect it seems obvious (it usually does).




From philh at comuno.freeserve.co.uk  Wed Feb  6 11:56:48 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 6 Feb 2002 16:56:48 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <psor8nywric.fsf@jekyll.curl.com>
Message-ID: <slrna62o2g.b33.philh@comuno.freeserve.co.uk>

On 06 Feb 2002 10:37:31 -0500, Christopher Barber <cbarber at curl.com> wrote:
>philh at comuno.freeserve.co.uk (phil hunt) writes:
>
>> On Wed, 06 Feb 2002 00:04:24 GMT, Courageous <jkraska at san.rr.com> wrote:
>> >
>> >>What if a macro needs to create local variables in its expansion?
>> >
>> >This is perfectly legitimate. What's not legitimate is...
>> >
>> >	deflingo myterm (i, codeblock):
>> >		j=3
>> >
>> >...where j=3 modifies the variable "j" in the context of the caller
>> >if this variable is preexisting in the caller. That's non hygienic
>> >behavior. Past experience with macros has shown this to be bad
>> >behavior; hence there is a design in the community for good "hygiene,"
>> >as it were.
>> 
>> And what if I decide that for my particular application, that's just
>> the behaviour I want? 
>
>There is no reason that the macro system cannot provide a means to create
>"unhygienic" references, but you would not want that to be the default.

That seems to me to be an entirely sensible position.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From wolfson at uchicago.edu  Thu Feb 21 16:14:00 2002
From: wolfson at uchicago.edu (Ben Wolfson)
Date: Thu, 21 Feb 2002 15:14:00 -0600
Subject: Stupid Lambda Tricks [LONG]
References: <mailman.1014324996.17145.python-list@python.org>
Message-ID: <pan.2002.02.21.15.13.59.643790.1064@uchicago.edu>

On Thu, 21 Feb 2002 14:52:49 -0600, Tim Peters wrote:

> Dear Lord!  The things you could accomplish if you weren't insane <wink>.

See also: http://www.p-nand-q.com/obfuscated-python.htm

-- 
BTR    
BEN WOLFSON HAS RUINED ROCK MUSIC FOR A GENERATION
 -- Crgre Jvyyneq


From skip at pobox.com  Sun Feb  3 09:03:45 2002
From: skip at pobox.com (Skip Montanaro)
Date: Sun, 3 Feb 2002 08:03:45 -0600
Subject: Expect like syntax in Python
In-Reply-To: <3c5cf3d6_2@news.nwlink.com>
References: <3c5cf3d6_2@news.nwlink.com>
Message-ID: <15453.17217.415421.27061@12-248-41-177.client.attbi.com>

    Guy> I'm trying to find a way to write something like expect in a python
    Guy> module, ...

Any particular reason you're not using one of the other Expect-like Python
thingies:

    http://www.vex.net/parnassus/apyllo.py?find=expect

?
-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From python at chajadan.net  Fri Feb 15 17:02:35 2002
From: python at chajadan.net (chajadan)
Date: Fri, 15 Feb 2002 14:02:35 -0800
Subject: Is making Python executables a reality?
In-Reply-To: <a4gkfj$3up8$1@ID-98166.news.dfncis.de>
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de>
 <Xns91B55AA48E7DCgmcmhypernetcom@199.171.54.215>
Message-ID: <5.1.0.14.0.20020215135833.00a8c360@63.118.78.184>

I'm very happy to report that I made my first standalone executable 
yesterday! Using py2exe :-)

This choice to try was inspired by a few different topics I'd been seeing 
lately about making executables, specifically the search that "took 4 
clicks" to respond to ;) So I pulled up the site, read, and voil?!

I always worried about having to make everyone download an entire 
python/win32all platform just to add 1 + 1.

--chajadan

At 10:18 AM 2/14/02 -0500, you wrote:

> >
> > > What percentage of people make exe files?  Do most people have
> > > end-users download the Python environment?
> >




From jmfbahciv at aol.com  Sat Feb 16 05:44:42 2002
From: jmfbahciv at aol.com (jmfbahciv at aol.com)
Date: Sat, 16 Feb 02 10:44:42 GMT
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net> <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net> <a4geh2$hir$12@bob.news.rcn.net> <140220020703593280%jwbaxter@spamcop.net> <a4j24g$jao$9@bob.news.rcn.net> <3C6DA497.9AB8D624@ev1.net>
Message-ID: <a4ll2i$6bt$10@bob.news.rcn.net>

In article <3C6DA497.9AB8D624 at ev1.net>,
   Charles Richmond <richmond at ev1.net> wrote:
>jmfbahciv at aol.com wrote:
>> 
>>      [snip...]     [snip...]     [snip...]
>> 
>> I do have two memoes that I found in my office boxes that were
>> addressed to PDP-1 users.  I don't know the people who wrote
>> them (Roland Silver, Edward Fredkin) nor where they were.
>> 
>>      [snip...]     [snip...]     [snip...]
>> 
>> Well, if you can give me a clue about these memoes that I have, I'd
>> appreciate it.
>> 
>Ed Fredkin is semi-famous. He was mentioned prominantly in David
>Levy's book _Hackers_. He also invented some data structures that
>are used today, notably the "trie". 
>
I don't know what a "trie" is, but I do (make that, did) have 
the _Hackers_ book.  I don't know where these memoes came from.
I guess there was a reason for making us put a logo on top of
all memoes after all (I take back all of the swears I said back
then).

Now I'll have to see if I can find my _Hackers_ book.  I neated
my library so that all of my books were in one place.  I wonder
what book I'll have to look for in order to have _Hackers_ pop
up in front of me.

/BAH

Subtract a hundred and four for e-mail.


From anton at vredegoor.doge.nl  Fri Feb  1 17:53:50 2002
From: anton at vredegoor.doge.nl (Anton Vredegoor)
Date: Fri, 01 Feb 2002 23:53:50 +0100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org>
Message-ID: <a3f6ch$c5$1@news.hccnet.nl>

On Thu, 31 Jan 2002 16:03:13 -0800, Cliff Wells
<logiplexsoftware at earthlink.net> wrote:

>True, to a certain extent.  Probably most do like I do: I see a review (in
>Python's case I saw it in Linux Journal sometime in the mid-90's) and they
>decide to evaluate it on a test application.  While this may not be the
>most thorough research, it's probably the most practical, given the
>time-crunch most of us operate under.  The unfortunate truth is that very
>few individuals can make a real evaluation of a language in a short period.
> I doubt that these studies amount to much as there is obviously more to a
>language than the language itself.  Even if a language is the best ever and
>studies show that it increases productivity 2000% (on what? developing
>quicksort routines?), it's worthless if it can't do X and we need X.  For

Sometimes this X is: 'It should be ready next friday'. So there's no
time to invest in changing programming language now, even if it would
be be a good long term choice. 

There seems to be no opportunity to introduce a change since the work
must go on.

How to handle this kind of resistance? 

Anton.


From Simon.Budig at unix-ag.org  Sat Feb  2 16:11:34 2002
From: Simon.Budig at unix-ag.org (Simon Budig)
Date: 2 Feb 2002 22:11:34 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de>
Message-ID: <3c5c5606@si-nic.hrz.uni-siegen.de>

In article <3c5b2c1f at si-nic.hrz.uni-siegen.de>, Simon Budig wrote:
> Michael Hudson <mwh at python.net> wrote:
>> Simon Budig <Simon.Budig at unix-ag.org> writes:
>>> Something is broken with the completion of readline:
>> 
>> Well, I knew about it.  I don't regard it as a huge problem.
> 
> Hmm. Imagine you are writing an interactive interpreter of a
> custom language to manipulate something non-file'ishes and
> suddenly the commandline starts completing *filenames*. I
> think this is a good showstopper...
> 
>>> Is there an evil hack to avoid this?
>> 
>> Dunno.  Doubt it, somehow.  You'll need to look at the documentation
>> for readline, though.
> 
> Well - I looked at the documentation for the readline-module, but that
> doesn't say anything about it. Maybe I should read the sources.

Ok, did that. Here is the Patch to fix this behaviour. It is against
Modules/readline.c from Python 2.1.2.

----8<-----------------------------
--- readline.c.orig	Sat Feb  2 21:44:09 2002
+++ readline.c	Sat Feb  2 22:01:16 2002
@@ -346,6 +346,9 @@
 		   lock released! */
 		save_tstate = PyThreadState_Swap(NULL);
 		PyEval_RestoreThread(tstate);
+		/* Don't use the default filename completion if we
+		   have a custom completion function... */
+		rl_attempted_completion_over = 1;
 		r = PyObject_CallFunction(completer, "si", text, state);
 		if (r == NULL)
 			goto error;
----8<-----------------------------

I really would appreciate, if this gets incorporated in the regular
Python releases. Unfortunately I don't have an account at sourceforge
to submit this the proper way.

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From cs1spw at bath.ac.uk  Tue Feb 19 18:09:38 2002
From: cs1spw at bath.ac.uk (Simon Willison)
Date: Tue, 19 Feb 2002 23:09:38 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56>
Message-ID: <3C72DA1D.7020203@bath.ac.uk>

You need to use a server side session with a unique session identifier 
passed in the URL. Basically when a user visits your page you create a 
random string (20 characters long is probably long enough) and create a 
record in a database or file recording that string. You also ensure that 
EVERY link on the page has this random string tagged on to it:

nextpage.py?s=xdf98dsyfisdof

Every time the user clicks a link the new page will "know" who they are 
by looking at the session passed in the URL. You can also send this 
unique identifier as a hidden form field whenever you ask the user to 
fill in a form.

As far as storing information goes you need some kind of server side 
storage mechanism that can "remember" things about the user who's 
session is xdf98dsyfisdof - a good way of doing this is by pickling a 
python data structure and calling the file "xdf98dsyfisdof.sess" or 
something like that.

HTH

Simon


Lutz Schroeer wrote:
> I'm creating a website which makes intensive use of cgi scripts and there 
> is the possibility for the visitor to register as a user for having the 
> ability to customize the site appearance, contents and other stuff.
> 
> Jumping from one script to the other I need to remember at least the user's 
> name. The W3C recommends using cookies (RFC 2109). Unfortunately many 
> people don't like cookies and filter them using WebWasher or a personal 
> firewall. 
> 
> Is there any other technique for keeping the user's data while he is using 
> my site without installing any large suite as Zope or the like? I've read 
> an article about utilizing soft links but I didn't quite get the point.
> 
> Lutz
> --
> (http://www.kybernetik-manufaktur.de)
> 




From phr-n2002a at nightsong.com  Sat Feb 16 23:57:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 16 Feb 2002 20:57:17 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net>
Message-ID: <7xg040yac2.fsf@ruckus.brouhaha.com>

"I.J." <shiver at yubc.net> writes:
> Is there a module for python that grants ussage of decimal numbers with more
> than 14 decimal places, including all scientific functions to do with those
> numbers?

Gmpy (http://gmpy.sourceforge.net) does arbitrary precision floating
point math.  It doesn't include any scientific functions beyond
arithmetic and square root though.  You might be able to adapt the
"bc -l" library to compute scientific functions with gmpy.

Note that normal 64-bit IEEE doubles give you 53 mantissa bits or
16-17 decimal digits.  You might be able to recompile Python to use
long doubles (80 bits on x86 computers) which will give you about 20
decimal digits.  I don't know if that's enough for your application.


From skip at pobox.com  Tue Feb 26 17:23:05 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 26 Feb 2002 16:23:05 -0600
Subject: Need advice for obfuscating Python code
In-Reply-To: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
References: <JHTe8.18749$0y2.1063258@typhoon.austin.rr.com>
Message-ID: <15484.2761.12991.934857@12-248-41-177.client.attbi.com>

    Sandeep> I've written a product in Python and I wanted to find out how
    Sandeep> best to obfuscate the code.  I want the distributed files to be
    Sandeep> moderately hard to reverse engineer.

Rewrite in Perl? ;-)

ducking-for-cover-ly, y'rs,

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From maxm at mxm.dk  Mon Feb 25 19:21:26 2002
From: maxm at mxm.dk (Max M)
Date: Tue, 26 Feb 2002 00:21:26 +0000
Subject: help: Problem with cgi form
References: <Xns91BE6246576BAAusYourStandingInIt@130.133.1.4> <3C7AAF31.8040201@mxm.dk> <slrna7lgm7.1kf.jon+usenet@snowy.squish.net>
Message-ID: <3C7AD506.3000906@mxm.dk>

Jon Ribbens wrote:

> In article <3C7AAF31.8040201 at mxm.dk>, Max M wrote:
> 
>>And then print it to the form like:
>>
>>print '<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % last_byte_pos
>>
> Beware that, in general, that is a bug. You are open to Cross Site
> Scripting attacks. Even in the absence of these, if the value happens
> to contain characters significant to HTML (such as '"', '&', '>', etc)
> then your program will go wrong.

Yeah ok ... I thought that there was a limit to how much should be 
explained in one post :-) But here goes:


from cgi import escape

'<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % escape(last_byte_pos)


Or in 2.2:


from xml.sax.saxutils import quoteattr

'<INPUT TYPE=HIDDEN NAME= "last_pos" VALUE='%s'>' % quoteattr(last_byte_pos)



regards Max M



From jeff at ccvcorp.com  Thu Feb  7 15:32:28 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 07 Feb 2002 12:32:28 -0800
Subject: encapsulation, typing & other questions
References: <3C62C5C5.7FEF0B10@daylight.com>
Message-ID: <3C62E45C.84B8506A@ccvcorp.com>


Ragu Bharadwaj wrote:

> Hi Guys,

Hi!  I'll only answer the questions that I feel qualified to say anything on...


> 2. How easy is it to use design patterns in Python? A lack of typing
> suggests it wouldn't be too easy, but I'd like to hear back from folks
> that have tried this.

It's very easy.  Many design patterns seem to almost be built-in to Python, many
others can be expressed in only a few lines of code.  The patterns that *don't*
seem to apply to Python, in fact, seem to be the ones whose whole point is
evading the type system.  (Laura Creighton recently made a comment to this
effect, about the GoF "Design Patterns" book....)


> 3. Are there plans to include method & variable encapsulation in the
> object models in python? The answer I've heard is that since it's
> possible to overcome encapsulation by long or dirty ways, it isn't worth
> it. But my comeback is that if one sticks to the rules while coding,
> this won't happen and the resulting gains are worth it.

Python *does* have encapsulation.  What it doesn't have is strictly enforced data
hiding.  However, if one sticks to the rules while coding, then having that
enforcement is not necessary.  And when in a situation that doesn't quite fit in
what the rules were intended to cover, then having that added flexibility can be
very much worth it.

(Note that Python *does* have some level of data hiding, but it's deliberately
easy to circumvent.  Instead of enforcing secrecy, it simply asks "Please don't
look at this"... )

Jeff Shannon
Technician/Programmer
Credit International




From kalafutj1 at yahoo.com  Sun Feb 24 14:51:36 2002
From: kalafutj1 at yahoo.com (Jim Kalafut)
Date: Sun, 24 Feb 2002 19:51:36 GMT
Subject: Jython & SAXException errors
Message-ID: <cvbe8.134446$Pz4.586708@rwcrnsc53>

I'm trying jython out for the first time and am running into compilation
errors.  My solely python app make use of XML/SAX support which seems to
cause a hicup with jythonc.  An edited transcript looks something like this:
.
Creating .java files:
  _exceptions module
   SAXNotSupportedException extends xml.sax._exceptions.SAXException
   SAXNotRecognizedException extends xml.sax._exceptions.SAXException
   SAXParseException extends xml.sax._exceptions.SAXException
   SAXException extends java.lang.Exception
   SAXReaderNotAvailable extends
xml.sax._exceptions.SAXNotSupportedException
...
Compiling .java to .class...
  .\jpywork\xml\sax\_exceptions.java:493: cannot resolve symbol
symbol  : constructor Exception  (java.lang.String,java.lang.Throwable)
location: class java.lang.Exception
            super(arg0, arg1);
            ^
.\jpywork\xml\sax\_exceptions.java:498: cannot resolve symbol
symbol  : constructor Exception  (java.lang.Throwable)
location: class java.lang.Exception
            super(arg0);
            ^

I checked the Java API docs and there really is no Exception constructor
that takes those combinations of arguments, so I'm not sure why jython is
trying to call super with them.

Any help would be appreciated.

Regards,
Jim




From phd at phd.pp.ru  Wed Feb  6 07:28:03 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Wed, 6 Feb 2002 15:28:03 +0300
Subject: Is anyone capable of explaining ?
In-Reply-To: <Pine.LNX.4.33.0202060712170.12958-100000@bagend.vzwcorp.com>; from Bill.Scherer@VerizonWireless.com on Wed, Feb 06, 2002 at 07:16:47AM -0500
References: <3C6069DA.18895.92952C@localhost> <Pine.LNX.4.33.0202060712170.12958-100000@bagend.vzwcorp.com>
Message-ID: <20020206152803.Q18177@phd.pp.ru>

On Wed, Feb 06, 2002 at 07:16:47AM -0500, Scherer, Bill wrote:
> > Question: How does web browser know what cookies should be 
> > sent to server as a response?
> 
> The browser determines which cookies to include with a request by 
> comparing the domain of the server it is about to contact with 
> the domain of the cookies.   I didn't see any domain in the 
> cookies you are setting; there should be one.

   This is true only partially. If you omit "domain" part of the cookie the
browaser will store current server's domain.

> The template http cookie header:
> Set-Cookie: name=value; expires=date; path=pathname; domain=domainname
                                                     ^^^^^^^^^^^^^^^^^^^
                                                     this is not neccessary

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



From BPettersen at NAREX.com  Thu Feb 28 16:51:16 2002
From: BPettersen at NAREX.com (Bjorn Pettersen)
Date: Thu, 28 Feb 2002 14:51:16 -0700
Subject: Status of PEP's?
Message-ID: <60FB8BB7F0EFC7409B75EEEC13E20192215149@admin56.narex.com>

> From: David Eppstein [mailto:eppstein at ics.uci.edu] 
> 
> In article <mailman.1014922495.31163.python-list at python.org>,
>  "Bjorn Pettersen" <BPettersen at NAREX.com> wrote:
> 
> > And exactly this notation has been rejected in PEP 204 -- basically 
> > because it's butt-ugly in a Perl'ish sort of way. Besides, 
> this goes 
> > under the section of solving the "interval" problem which PEP 276
> > *explicitly* says it is not doing (and which I personally 
> don't find 
> > very compelling, especially if PEP 276 is adopted).
> 
> You can't have it both ways -- you seem to be saying that 
> it's not ok to 
> argue against PEP 276 on the basis that you'd like a better 
> general range 
> syntax, because that's not the problem the PEP is trying to 
> solve.  But 
> then in the parentheses you seem to be saying that it's ok to 
> argue against 
> proposals for a better range syntax on the basis that PEP 276 
> covers most 
> of the need.  Which is it?

I think it's fine to argue against 276 if you think what we need is
really a general solution to the interval problem, especially if you
have something new to contribute -- none of the current proposals seem
to have universal appeal. PEP 276 explicitly states that it's not trying
to solve the universal problem however, so I think it's unfair to
criticize it for not providing a general solution without also saying
that the only thing you'll accept is a general solution.

Since many of the brightest minds on c.l.py have tried hard to come up
with a general range syntax without much success, I would rather have
PEP 276 which solves the most common problem superbly. That it doesn't
try to come up with a general solution, but doesn't prevent one from
being implemented in the future is an additional bonus.

Keep-the-general-case-trivial-and-the-hard-cases-possible 'ly y'rs
-- bjorn



From dave at pythonapocrypha.com  Wed Feb  6 16:00:55 2002
From: dave at pythonapocrypha.com (dave at pythonapocrypha.com)
Date: Wed, 6 Feb 2002 13:00:55 -0800 (PST)
Subject: ANN: Pyco 0.1 - Tiny Python Windows Distributions
Message-ID: <Pine.LNX.4.33.0202061241570.3709-100000@ns0.tbye.com>

Announcing Pyco - a tool for building tiny Python Windows executables.

What is it?
 Pyco (pronounced like "pico") is a program that creates standalone
 (single-file) Windows executables from your Python programs, with an
 emphasis on the finished executable being as small as possible (starting
 at about 350K).

Why should I use it?
 Most people probably shouldn't. ;-)
 It started as an experiment but I ended up using it for a real project,
 so I thought somebody else might like to use/play with it too. It does
 not attempt to replace the other tools (py2exe, McMillan), it is not as
 good as the other tools, and you should probably avoid Pyco if you
 haven't used the other tools before.

 I needed a very small Windows executable (for rapid downloads) to which
 I could easily append additional files after the executable had already
 been created. If you have similar needs, Pyco will be useful as is or
 as a starting point.

Where can I get it?
 http://www.pythonapocrypha.com/projects/pyco/

What is the license?
 Do whatever you want, don't sue me.

-Dave




From jeff at ccvcorp.com  Mon Feb 18 16:07:48 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 18 Feb 2002 13:07:48 -0800
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <4296.811T2599T12485989threeseas@earthlink.net>
Message-ID: <3C716D24.4C4AE2FF@ccvcorp.com>


Timothy Rue wrote:

>  There is plenty being withheld from
> the consumers, including freedom to create what they need ...

So, are you also expecting that the automotive industry allow you to build your own car?
I mean, it's all done by robots anyhow, right?  So why not let *you* direct those robots?
For that matter, you're a cabinet-maker.  Do you forge your own hammers?  Make your own
saws?  I mean, it's a simple procedure -- just make a mold and dump molten metal in it,
right?  So why is the tool-industry preventing *you* from doing that?  Sounds like a
conspiracy to me...


> and to do so
> without abuses of the over complexity generated by the computer industry,
> who continuely asks the consumer "where do you want to go today?" because
> those in the computer industry simply are to disconnected from the
> consumer/user base to know and lacking in ethics plenty enough to simply
> give the consumer/user the tools to go where they want..

While there are indeed some companies that lack ethics, and MS is indeed the biggest and
probably the worst of them, your sweeping over-generalization of the entire computer
industry as being similarly unethical, is simply not true.  It's just as accurate to say
"White people are racist" -- many are, but many are not.  (And many non-white people are
racist as well.)   Painting all of them with the same brush will not gain you anything but
resentment.


> What of the double speak computer industry dictionary? The one that has
> "we" in it for talking to the in a manner of them preceiving it as
> inclusive of them but for those using it it really means not inclusive f
> the public. And other such double meaning ........ standard
> definitions.....?????

This paragraph makes no sense.  I am unable to parse this into anything meaningful.  I
don't know if English is your first language or not, but please be aware that it is
important to use clear language if you wish to be taken seriously.  We can't discuss with
you if we can't understand you.

However, if you're referring to someone using "we" to describe what that individual
believes to be opinions shared by others, well...  this is *hardly* limited to the
computer industry, and seems to be a standard behavior throughout society.


> There is a physics to abstraction manipulation, including such abstraction
> manipulation as the double speak dictionary and programming abstractions
> that boils down to bit manipulation of physical switches.

No.  Physics deals strictly with *physical* things.  Abstractions are, by definition,
non-physical.  There may indeed be specific rules involved in the manipulation of
abstraction, but they are, by definition, not physics.  If you want to understand more
about the current state of science regarding abstraction manipulations, I suggest that you
take a few cognitive psychology courses.  Indeed, if you *were* coming from a psychology
background, then your ideas about these "nine actions" would carry a lot more weight.

Also note that even if we do accept that your "nine actions" are a valid summation of all
manipulations of abstraction (something that you've failed to demonstrate), this still
does not mean that it is straightforward to implement a system which somehow makes use of
these actions to create nontrivial, *correct* computer applications.  (Correct, in this
sense, meaning that they do what they are supposed to do, i.e., provide correct answers,
handle invalid input gracefully, etc.)


> I have no doubt that those who have made a living off of such
> manipulations of bits, abstraction and consumers certainly don't want to
> be exposed for the wrongful manipulators that are.

Hm.  You, as a cabinet-maker, have presumably required people to pay *you* money to
manipulate chunks of wood and bits of metal and plastic.  You seem to have no compunctions
about making a living off of those consumers, instead of giving them your tools and
letting them build their own cabinets.  Does this not make you, equally, a wrongful
manipulator?


> Clearly there is a priority here where the coding is not on top of the
> list but rather ...... well what do you suppose Neo ment in that last
> line of the Movie, The Matrix?

Y'know, I haven't seen that movie.  And if you consider anything from a mainstream movie,
made in Hollywood, to be an unbiased representation of reality, even in a symbolic way,
.... then I pity you.


> You don't understand the problems because you refuse to look at yourself.

Hello, Kettle.  May I introduce you to Mr. Pot?

Jeff Shannon
Technician/Programmer
Credit International




From asim_beg at hotmail.com  Wed Feb 20 14:18:15 2002
From: asim_beg at hotmail.com (Asim Beg)
Date: 20 Feb 2002 11:18:15 -0800
Subject: MySQLdb build on Solaris 7
References: <F203anSs21Ud9mTP2vZ000016c2@hotmail.com> <mailman.1014214245.28703.python-list@python.org>
Message-ID: <d1dba0f2.0202201118.506f0635@posting.google.com>

Jeff Bauer <jbauer at rubic.com> wrote in message news:<mailman.1014214245.28703.python-list at python.org>...
> Asim Beg wrote:
> > Here's an excerpt from setup.py
> > 
> > elif sys.platform == "sunos5": # Solaris 2.8 + gcc
> >     include_dirs = ['/usr/local/mysql/include']
> >     library_dirs = ['/usr/local/mysql/lib']
> >     libraries = [mysqlclient, "z"]
> >     runtime_library_dirs = ['/usr/local/lib:/usr/openwin/lib:/usr/dt/lib']
> >     extra_compile_args = ["-fPIC"]
> 
> Looks reasonable.  You've checked that the above directories
> actually have mysql include files and libraries present, right?
> 
> Here's mine:
> 
>   elif sys.platform == "sunos5": # Solaris 2.8 
>       include_dirs = ['/usr/local/include/mysql'] 
>       library_dirs = ['/usr/local/lib/mysql'] 
>       libraries = [mysqlclient, "z"] 
>       runtime_library_dirs =
> ['/usr/local/lib:/usr/openwin/lib:/usr/dt/lib'] 
>       extra_objects = [] 
> 
> And:
> 
>   LD_LIBRARY_PATH=/usr/local/lib:/usr/local/lib/mysql
> 
> Also, in your situation, your administrator may not permit you
> to install files in /usr/local/lib/python2.1/site-packages/ --
> though there isn't any real reason she shouldn't -- and it would
> make it more convenient for running cgi scripts.  But assuming
> she doesn't, you'll need those files (in the particular
> relative path structure I mentioned in my earlier post) to 
> be present in your importable path, e.g. PYTHONPATH or
> (ugh) manually appended to sys.path.
> 
> I have worked with ISP admins whom I was able to convince
> that installation into /site-packages/ was a good thing to
> do for packages such as the MySQL interface; it would not
> compromise security or maintenance and benefit other
> Python users on their system.
> 
> Good luck.
> 
> Jeff Bauer
> Rubicon Research


Yes both /usr/local/mysql/lib and /usr/local/mysql/include exist and
have 644 permissions. Frankly, I am surprised (shocked...) to find
that an important package such as MySQLdb for python would not compile
on Solaris.....In my many many years of installing software on
solaris/linux I have rarely seen such dumb error messages.....

May be I am clueless and am missing something basic here....I just
can't beleive I am the only one having this kind of a problem.


Asim


From youvebeenowned at home.com  Mon Feb 18 00:55:08 2002
From: youvebeenowned at home.com (3[ED] (Thread))
Date: Mon, 18 Feb 2002 05:55:08 GMT
Subject: I Don't Know A Thing...Compiling?
Message-ID: <3C709746.AD83377F@mchsi.com>

I don't really work with Python, although I might sometime, I don't
know. Anyway, I've just been wondering this. I always thought you had to
have a Python interpreter to run any Python program, but an online
friend of mine said she had ran exe's which were games written in
Python. So I was just wondering, can you compile a Python script into
ane executable, or do you need an interpreter. I don't know if this is
dumb or not. I don't think I'm hip to the hacking scene. :o)

--
-matt
||and she pretends that we have everything in common||




From jason at jorendorff.com  Sun Feb 17 13:11:00 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 17 Feb 2002 12:11:00 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <E16cM76-0000po-00@tswings.swing.cluster>
Message-ID: <HFEKILOLEFEFMKAECNDLGEDKDDAA.jason@jorendorff.com>

Christian Tanzer wrote:
> Jason Orendorff wrote:
> > Whereas there are no known security holes in pickle.
> 
> Not true.

I stand corrected.  Yow.  My statement above is not just wrong
but excessively wrong.

> To wrap this up, writing secure applications is hard -- and rules of
> thumb like `eval is bad, pickle is good` aren't going to make it any
> easier.

Good call.

## Jason Orendorff    http://www.jorendorff.com/



From magnus at thinkware.se  Fri Feb 22 14:37:36 2002
From: magnus at thinkware.se (Magnus =?ISO-8859-1?Q?Lyck=E5?=)
Date: Fri, 22 Feb 2002 20:37:36 +0100
Subject: Python 2.1 and PIL
References: <94id8.563$s03.36009@news2.east.cox.net> <3C76654C.9010002@thinkware.se> <cfabd740.0202221014.5b410cd6@posting.google.com>
Message-ID: <3C769E00.1090509@thinkware.se>

As far as I understand RedHat is dependant on
Python 1.5.2 for a lot of administration SW.
It's about time they upgrade! For this reason,
/usr/bin/python is python 1.5.2 here, and
python 2.1 is /usr/bin/python2. Yuck! "Upgrade"
to Mandrake? ;-) Sorry...I don't know.

Try the mailing list found at
http://www.pythonware.com/products/pil/index.htm

wwaldo wrote:

> Redhat 7.2.  It installs pythong to
> 
> /usr/lib/python1.5
> /usr/lib/python2.1
> 
> Magnus Lyck? <magnus at thinkware.se> wrote in message news:<3C76654C.9010002 at thinkware.se>...
> 
>>What OS are you using? (For windows there are binaries for
>>the different Python versions.)
>>
>>william wrote:
>>
>>
>>>Ok been playing with this python 2.1.2 and PIL.  The problem is when i
>>>install the PIL it installs it for python1.5.  I don't know how to get the
>>>PIL to install for Python2.1.
>>>
>>>I am assuming I need to make some changes in some files before i do
>>>./configure.  Any suggestions?
>>>
>>>Thanks.
>>>
>>>
>>>
>>>



From quinn at baht.ugcs.caltech.edu  Sun Feb 24 06:42:09 2002
From: quinn at baht.ugcs.caltech.edu (Quinn Dunkan)
Date: 24 Feb 2002 11:42:09 GMT
Subject: Why ConfigParser can't write to file?
References: <a56sr2$52ge1$1@ID-12869.news.dfncis.de> <fUJd8.418$PE2.43612@stones> <a58885$5acoc$1@ID-12869.news.dfncis.de> <zVPd8.4096$jC3.163217@news1.east.cox.net> <a5acho$5j86i$1@ID-12869.news.dfncis.de>
Message-ID: <slrna7hkcg.i6a.quinn@baht.ugcs.caltech.edu>

On Sun, 24 Feb 2002 17:41:17 +0800, Ricky <store_li at sina.com> wrote:
>I see now. Thank you! But I am still confuse, why I write???? method can't
>use filename as a parameter.

Many files don't have names.

E.g. sockets, pipes, StringIOs, etc.


From cliechti at gmx.net  Wed Feb 13 15:53:35 2002
From: cliechti at gmx.net (Chris Liechti)
Date: Wed, 13 Feb 2002 21:53:35 +0100
Subject: ANN: pyserial - a mulitplatform serial port library (Win32, Linux, Jython)
Message-ID: <WU62A9ZBA5633YKF07QKHD2WLLI96.3c6ad24f@rock>

seems that others too want to use the serial port from python...
here's my module (python license). this is the first public release as 
separrate package. but its already used in pybsl. however its now very 
handy with a disutils install.

from the readme/homepage:

pySerial
--------
This module capsulates the access for the serial port. It provides 
backends for stadard Python running on Windows, Linux, BSD (possibly 
any POSIX compilant system) and Jython. The module named "serial" 
automaticaly selects the appropriate backed.

(C) 2001-2002 Chris Liechti <cliechti at gmx.net>

Features
--------
- same class based interface on all supported platforms
- port numbering starts at zero, no need to know the port name in the   
  user program
- port string can be specified if access through numbering is 
  inappropriate
- support for diffrent bytesizes, stopbits, parity and flow control
  with RTS/CTS and/or xon/xoff
- working with or without receive timeout
- file like API with "read" and "write"
- The files in this package are 100% pure Python.
  They depend on non standard but common packages on Windows 
  (win32all) and Jython (JavaComm). POSIX (Linux, BSD) uses only 
  modules from the standard python distribution)
- The port is set up for binary transmission. No NULL byte stripping,   
  CR-LF translation etc. (which are many times enabled for POSIX.) 
  This makes this module universally useful.

more details on the Homepage:  http://pyserial.sourceforge.net/

chris

-- 
Chris <cliechti at gmx.net>






From skip at pobox.com  Tue Feb 12 23:33:11 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 22:33:11 -0600
Subject: searching backwards in a string
In-Reply-To: <7xvgd2589b.fsf@ruckus.brouhaha.com>
References: <mailman.1013549132.16176.python-list@python.org>
        <7x8z9y46l1.fsf@ruckus.brouhaha.com>
        <mailman.1013571921.18465.python-list@python.org>
        <7xvgd2589b.fsf@ruckus.brouhaha.com>
Message-ID: <15465.60551.130122.707197@12-248-41-177.client.attbi.com>

    Paul> here 50000 and -1 are the starting and ending locations.

    Skip> If start > end isn't direction == -1 implied?

    Paul> Nope.  I tried it and it didn't search backwards.  

That wasn't quite what I was asking.  Let me ask it a slightly different
way.  Presuming the behavior can be changed so that searching in reverse
works, shouldn't the ordering of start and end be sufficient to indicate the
direction of the search?

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From tim.one at home.com  Mon Feb 11 02:51:27 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 11 Feb 2002 02:51:27 -0500
Subject: subclassing tuple
In-Reply-To: <20020211071605.GA2528@lilith.hqd-internal>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEBLNLAA.tim.one@home.com>

The difference between lists and tuples here is that tuples are immutable.
This is a distinction that didn't exist at the class level in 2.1, but must
exist in 2.2 in order to make what used to be builtin immutable *types* into
classes instead.

__init__ is too late to influence the base contents of an object of an
immutable type:  the object's memory has already been allocated by the time
__init__ is invoked.  It must be that way, else __init__ couldn't have a
self argument (to have a self, the memory for self must already be set up
and bound to self).

That's why __new__ was introduced in 2.2:  __new__ is, in a sense, lower
level than __init__.  It does not take a "self" argument, because it exists
to do memory allocation and initialization of the object, and for immutable
objects you only get this one shot at memory initialization.

At the C API level, an even finer distinction is possible, allowing to break
__new__ into distinct "allocate raw memory" and "minimally initialize
memory" portions.  You don't have to worry about that distinction at the
Python level, though (it wouldn't make sense there, as there is no "allocate
raw memory" facility available at the Python level).




From sameer at cs.uoregon.edu  Wed Feb 27 13:21:11 2002
From: sameer at cs.uoregon.edu (Sameer Suresh Shende)
Date: Wed, 27 Feb 2002 10:21:11 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <0ebb01c1beef$a09efe00$0500a8c0@boostconsulting.com>
Message-ID: <Pine.SGI.4.33.0202271018190.1311-100000@krypton.cs.uoregon.edu>

Hi Dave,
	We've released pdtoolkit-2.1.zip on the web. It contains
the new taucpdisp.exe for Windows that fixes the bug you reported. I
believe Robert sent you this binary earlier. I hope this allows you to
parse BOOST with PDT. Let us know if we can assist you with anything.
	Thanks!
	- Sameer

On Tue, 26 Feb 2002, David Abrahams wrote:

>
> ----- Original Message -----
> From: <rasmussn at lanl.gov>
>
> > On Monday, February 25, 2002, at 05:32 PM, David Abrahams wrote:
> >
> > > I've been interested in systems for automatically wrapping C++ and
> > > generating documentation for a while, so I downloaded the PDT (the
> parser
> > > toolkit used by SILOON) and tried it on one of my source files. After
> lots
> > > of hacking, I was able to get the EDG front-end component to parse my
> > > source
> > > and generate an intermediate language (.cil) file, but the PDT
> executable
> > > which is supposed to interpret the .cil and produce a high-level
> > > description
> > > of the code crashed with an assertion:
> >
> > SILOON uses scripts to run the PDT and we've worked hard to make it easy
> > to use.  Essentially all you have to do is edit a file (that is included
> > in a Makefile) and point macro variables to the location of your source
> > files (.h .cpp ...).  I would be interested to know if you have similar
> > problems using the EDG front-end from SILOON directly.
>
> Well, I'm less interested in SILOON itself than in the PDT, but I might try
> it if I get some more time to waste ;-)
>
> > > Assertion failed: ptr->assoc_template, file c:\documents and
> > > settings\bertie\desktop\finalwinstuff\taucpdisp\src\disp-
> > > tau\il_display_tau.
> > > c, line 3047
> > >
> > > Since the PDT doesn't come with source, there was no way for me to
> debug/
> > > fix
> > > the problem.
> >
> > The PDT comes with sources for everything (almost) but the EDG front end.
>
> There's no "il_display_tau.c" in either the .zip or the .tar.gz.
>
> > What PDT version were you using?
>
> The very latest I could find at
> http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/.
>
> > Would you like to send your source files to me directly
> > and I can
> > look at the problem, although if it is windows-specific code, I can't help
> > you
> > as I don't have access to Microsoft compilers.
>
> It's code that compiles on 7 different compilers including several different
> EDG-based compilers on four OSes, so it shouldn't be a problem. If you get
> the current boost cvs state (see http://sourceforge.net/cvs/?group_id=7586)
> and a copy of Python-2.2, you can try for yourself. The file is
> <boost-root>/boost/python/class.hpp.
>
> > (On further examination, it looks like the PDT doesn't have source for
> > taucpdisp,
> > I'll ask the developers why not.)
> >
> > I can provide access to a prerelease version of SILOON which uses the
> > latest
> > version of the PDT, which may fix your problems.
>
> If you can get this to work, I might be interested in seeing the results.
>
> > > Furthermore, the PDT developers don't seem to want to be reached. Their
> > > contact page (http://www.cs.uoregon.edu/research/paracomp/pdtoolkit/)
> > > says:
> > >
> > >     To contact the developers of PDT, please use the "Comments" section
> in
> > > the download page.
> > >
> > > But of course, there is no "Comments" section on the download page. :=(
> >
> > I've forwarded you comments on to the PDT developers so that they can fix
> > their web pages and comment personally on the core dump.
>
> Thanks,
> Dave
>
>





From jkraska at san.rr.com  Wed Feb 27 10:47:42 2002
From: jkraska at san.rr.com (Courageous)
Date: Wed, 27 Feb 2002 15:47:42 GMT
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <3C7C6252.FC779971@cosc.canterbury.ac.nz> <m37koz2trp.fsf@mira.informatik.hu-berlin.de>
Message-ID: <b70q7usiglk3kb37h7de049jdlo9nbr6ol@4ax.com>

>> For these reasons, saying just "Python's grammar is
>> LL(1)" is oversimplifying things considerably.
>
>If you count the semantic analysis as part of the grammar, I believe
>there would be few programming languages that have a context-free
>grammar.

Indeed. And further, there are sometimes these things we do
when parsing and lexing (which I consider to be about the same
thing in many ways, except that lexing merely operates a character
at at time) which we defer to later. Take for example, postfix
tags on integers to make them long 3L versus 3. This isn't LL
for any finite K. The typical approach is to mutate the AST node
of the object when the final tag is encountered. One doesn't
attempt to have a lexical production for both.

C//



From tim.one at comcast.net  Fri Feb 22 01:54:47 2002
From: tim.one at comcast.net (Tim Peters)
Date: Fri, 22 Feb 2002 01:54:47 -0500
Subject: Large integers from C
In-Reply-To: <c76ff6fc.0202212105.56aa14f8@posting.google.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEHHNOAA.tim.one@comcast.net>

[John Machin]
> In this case as the OP in essence pointed out,
> PyLong_FromSomeFormatLongerThanCLongLong()would be very useful for
> punters who somehow are generating such beasties themselves.
> _PyLong_FromByteArray() exposes no implementation details that are
> private to longobject.c. Do you have any objection to removing the _
> and getting the comments in the .h file copied into the C API manual?

"It depends", on whether stronger reasons to change the C API pop up.  When
we change the C API, PYTHON_API_VERSION (in modsupport.h) gets bumped, and
then Python whines if extension modules don't get recompiled.  That can be a
real pain for the users least able to deal with it, so changes to the public
C API are resisted.  (Heh -- people on Windows have no idea that Python
sometimes *is* binary-compatible across minor releases, because a different
mechanism forces recompilation regardless on Windows; the pain I'm talking
about is a non-Windows pain.)

I don't see enough demand for this particular function alone to justify the
pain.  If it can ride in on some other change's coattails, then its inverse
(_PyLong_AsByteArray) should get blessed at the same time.




From warkid at storm.ru  Tue Feb 19 08:23:19 2002
From: warkid at storm.ru (Kerim Borchaev)
Date: Tue, 19 Feb 2002 16:23:19 +0300
Subject: execfile question
In-Reply-To: <3C724B05.E9167812@attbi.com>
References: <3C724B05.E9167812@attbi.com>
Message-ID: <18682.020219@storm.ru>

Mark Barclay answered:

> "del c" will cause your __del__() to be called.
yes

> So will any other action that causes your instance to be garbage-collected.
I don't get it.
And can you explain who holds a reference on C instance in my example
when 'execfile' completes?

Kerim.

----------------------------------
My original question:

> Can someone explain why while running this script I don't see C.__del__
> eventually called?
>
> For me the output of it's execution is:
>
> >main.py
> it's me
> init
> executed
> >
>
> ##main.py######################
> execfile('executable.py', {})
> print 'executed'
> ########################
>
> ##executable.py######################
> print "it's me"
> class C:
>     def __init__(self):
>         print 'init'
>     def __del__(self):
>         print 'deleted'
>
> c = C()
> ########################
>
> Thanks in advance.





From mertz at gnosis.cx  Fri Feb  1 21:47:14 2002
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Fri, 01 Feb 2002 21:47:14 -0500
Subject: Why has 'dir()' changed in Python 2.2?
Message-ID: <yM1W8kKkXktc092yn@bellatlantic.net>                         

I just noticed a change in Python 2.2... that turns out to break my
[xml_pickle] module.  It shouldn't be hard to fix (although I *do* have
a decision to ponder about the best fix)... but I mostly wonder what the
thoughts behind the change were (it seems like it will break a lot of
code that is moderately introspective).

Here's a test program:

    % cat dir_change.py
    import sys
    class C:
        "The docstring for C"
        def __init__(self):
            self.foo = 3
            self.bar = "bar"
        def othermethod(self): pass
    o = C()
    print sys.version
    print dir(o)

Running it (2.1):

    % python dir_change.py
    2.1 (#0, Jun 17 2001, 11:51:05) [EMX GCC 2.8.1]
    ['bar', 'foo']

compared with (2.2):

    % python dir_change.py
    2.2 (#0, Dec 24 2001, 18:42:48) [EMX GCC 2.8.1]
    ['__doc__', '__init__', '__module__', 'bar', 'foo', 'othermethod']

Here's a few others FWIW:

    % python dir_change.py
    1.5.1 (#0, Sep 21 1998, 08:28:40) [VisualAge C/C++]
    ['bar', 'foo']

    % python dir_change.py
    1.5.2 (#0, Jun 27 1999, 11:23:01) [VisualAge C/C++]
    ['bar', 'foo']

    % python dir_change.py
    2.0.42-S1.2.23 (#0, Apr 25 2001, 20:59:49) [GNU C/C++]
    ['bar', 'foo']

    % jython dir_change.py
    2.0
    ['bar', 'foo']

Or even:

    # python dir_change.py
    2.1.1 (#1, Aug 13 2001, 19:37:40)
    [GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-96)]
    ['bar', 'foo']

    # python1 dir_change.py
    1.5.2 (#1, Jul 5 2001, 03:02:19) [GCC 2.96 20000731 (Red Hat Linux 7.1 2
    ['bar', 'foo']

Or:

    > python dir_change.py
    2.2a2+ (#22, Sep  5 2001, 14:10:41) [MSC 32 bit (Intel)]
    ['__doc__', '__init__', '__module__', 'bar', 'foo', 'othermethod']

Ok, enough versions...

--
mertz@  | The specter of free information is haunting the `Net!  All the
gnosis  | powers of IP- and crypto-tyranny have entered into an unholy
.cx     | alliance...ideas have nothing to lose but their chains.  Unite
        | against "intellectual property" and anti-privacy regimes!
-------------------------------------------------------------------------




From phr-n2002a at nightsong.com  Sun Feb  3 20:31:17 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 03 Feb 2002 17:31:17 -0800
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org> <a3k9am$p0c$1@newsreader.mailgate.org> <fmbr5uc0uikf74flscdlilq7bq8q1i3tmr@4ax.com>
Message-ID: <7xofj6m3ru.fsf@ruckus.brouhaha.com>

Courageous <jkraska at san.rr.com> writes:
> And at that, I'll leave you with a movie quote for the day:
> 
> "Have you ever heard of Plato? Aristotle? Socrates?"
> "Yes."
> "Morons."

"Remember how bad Einstein's grades were in school?  Well, mine are
even worse!"  (Calvin & Hobbes).


From quinn at hurl.ugcs.caltech.edu  Mon Feb  4 20:35:31 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 5 Feb 2002 01:35:31 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <Wls68.14178$om6.357057@news1.tin.it> <mailman.1012590723.7301.python-list@python.org> <a3kb3p$se44@news1.gtech.com> <mailman.1012843209.19773.python-list@python.org> <a3n03l$tfn6@news1.gtech.com> <mailman.1012865530.1367.python-list@python.org>
Message-ID: <slrna5udn2.ihs.quinn@hurl.ugcs.caltech.edu>

On Mon, 4 Feb 2002 15:22:23 -0800, Cliff Wells <logiplexsoftware at earthlink.net>
wrote:
>We developed this approach based upon statements made by Richard Stallman
>(and common sense ;) on the GNU approach to development: 
>
>1. no arbitrary limits
>2. embed a language in your application and use that to develop parts of
>the application
>3. turn parts of the program into independent libraries  <-- repeat,
>repeat, repeat
>
>To this I would add that using an n-tier design will greatly extend the
>lifetime of a piece of software (especially since much "upgrading" often
>involves cosmetic changes or changes that are irrelevant to the function of
>the software [like supporting a different SQL backend]) and makes things
>like redundancy easier to implement.  Of course this is just an extension
>of the Unix "tools" philosophy.

To me, it seems like especially #2 and #3 contradict the "tools" philosophy (or
at least what I understand that it is).  #2 because "tools" are scripted with
the shell and pipelines, or any language which can make use of the unix
fork()/exec() and pipe() syscalls such as python.  If they become so monolithic
as to require a seperate internal language or huge rc file, they're closer to
"applications".  #3 because in the tools approach, the programs themselves are
the units of reusability, not the code within the programs.  Hence, instead of
making a libjpeg that decodes and encodes jpegs, you write [cd]jpeg programs to
do the same, and read them through a pipe when you want a jpeg (this way you
also get a nice lazily evaluated data stream).

I'm not saying that the tools philosophy is the end-all be-all or that libjpeg
should be just the tools instead of the library and the tools, or anything bad
about libraries.  Tools force a simplicity of interface (single directional
data stream started by a small number of cmdline args) which is very good in
many cases and unworkable in many other cases.  INN is a good example of a
"tools" oriented program that I can't stand... but that may have more to do
with the 500 line shell scripts than the approach.  Definately a more
reasonable language is needed for things that are out of the shell's purview...
like, say, python (struggling to appear somewhat on-topic).


From shippy at nmt.edu  Fri Feb 22 17:27:51 2002
From: shippy at nmt.edu (Jeff Shipman)
Date: Fri, 22 Feb 2002 15:27:51 -0700
Subject: re. help needed  :(
References: <a56838$qb8$1@newsg4.svr.pol.co.uk>
Message-ID: <3C76C5E7.9070703@nmt.edu>

G. Willoughby wrote:
> Hi,
>     Im trying to replace the single word "lamb" and the word lamb followed
> by any punctuation in this string:
> 
> x="Mary had a little lamb called lamb, but didn't like the lamb's mood"
> 
> i want to replace all of the single words "lamb[punctuation]" but not the
> word "lamb's" to achieve this:
> 
> x="Mary had a little pig called pig, but didn't like the lamb's mood"
> 
> any help with be greatly appreiciated!
> 
> ta
> 
> G. Willoughby
> 
> 
> 

x = re.sub('lamb(?!\')', 'pig', x)

should do the trick. This replaces all occurances
of lamb with pig only if it's not followed by an
apostrophe. If you want to look for more types
of punctuation, then you could use a [] to indicate
a set rather than the \'.

-- 
Jeff "Shippy" Shipman     E-Mail: shippy at nmt.edu
Computer Science Major    ICQ: 1786493
New Mexico Institute of Mining and Technology
Homepage: http://www.nmt.edu/~shippy



From mrstephenson at dof.se  Fri Feb 15 05:53:37 2002
From: mrstephenson at dof.se (Rod Stephenson)
Date: 15 Feb 2002 11:53:37 +0100
Subject: Problems with Tkinter based scripts under emacs...
Message-ID: <wk4rkj583y.fsf@dof.se>

I'm having problems with emacs on win95 when running a python script
that invokes Tkinter. I'm using the bash shell under emacs, and
python-mode, so that I can start up an interpreter which appears in a
comint buffer and issue commands. Theres no problem with regular
python scripts, and its a nice environment for edit-run-debugging.

I've just started using Tkinter and I find the following: The script
runs fine, but when I quit the application, the shell just hangs and I
don't get the >>> prompt back. I end up having to kill the process and
restart the interpreter.

Actually I'm not sure whether this is specifically an emacs issue. I
find a similar thing happens when I use idle - after running a Tkinter
based script the shell window no longer responds, though I can
continue to reload the scripts after editing from the editor window
using the menus or keyboard accelerators. There is some discussion in
the FAQ on not using Tkinter with pythonWin.

Everything works just fine in the DOS command line shell, but this is
so annoying to use! Any hints about setting up emacs or idle to cope
with Tkinter would be much appreciated.


From robin at jessikat.fsnet.co.uk  Thu Feb  7 10:59:30 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 7 Feb 2002 15:59:30 +0000
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de>
 <a3sk2i$67m$1@panix2.panix.com>
 <e840346c.0202070321.452d405d@posting.google.com>
 <7xr8nxy0rb.fsf@ruckus.brouhaha.com>
 <e840346c.0202070724.13440c8d@posting.google.com>
Message-ID: <E+9AIXAiRqY8EwKc@jessikat.demon.co.uk>

In article <e840346c.0202070724.13440c8d at posting.google.com>, Mark Brady
<kalath at lycos.com> writes
>Paul Rubin <phr-n2002a at nightsong.com> wrote in message news:<7xr8nxy0rb.fsf at ruck
>us.brouhaha.com>...
>> I don't know what would happen if someone tried a big project in
>> Python (Zope is medium sized).  Maybe it would work ok, maybe not.
>> 
>> I do believe a native-code compiler would make Python a lot more
>> interesting for many purposes.
>
>I think this would be very useful also but how would it work with
>Python's dynamic features? I know some native code comipled languages
>such as Objective C and GJC (the GNU java native code compiler) come
>with a runtime library for the native exe, is this possible for a
>language like python?
>
>Mark.
I think there are good compilers for lisp which can be a very dynamic
language so in principle it's not the dynamic nature of python that
prevents native compilation.
-- 
Robin Becker


From thomas.heller at ion-tof.com  Tue Feb  5 15:16:01 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Tue, 5 Feb 2002 21:16:01 +0100
Subject: distutils on win32
References: <6bd9f01b.0202051200.74b0267b@posting.google.com>
Message-ID: <a3peik$1a1or9$1@ID-59885.news.dfncis.de>

"sameer" <sameer_ at email.com> wrote in message news:6bd9f01b.0202051200.74b0267b at posting.google.com...
> i'm using distutils to make a windows installer for a python/wxpython
> app.  what's the fastest way for me to create shortcuts on the desktop
> as well as the Start menu for the file I want the users to be able to
> run.  I don't want them to have to go into the correct package, and
> run the app from there.

Not easy, I suppose. The fastest way would probably be to prepare the
distribution with distutils, and use inno setup to install everything
and create the shortcuts.

Thomas

PS: If you insist on using pure distutils, you would probably create
your own 'install' command in the setup script, use Mark Hammond's
win32all extension (which must be present on the target computer),
to create the shortcuts.
Caveat: This will only work if the user installs from source, not
from the distutils windows installer.




From jason at jorendorff.com  Tue Feb 12 02:03:12 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Tue, 12 Feb 2002 01:03:12 -0600
Subject: RFC -- Hygienic Macros in Python
In-Reply-To: <slrna6bfrc.o7n.philh@comuno.freeserve.co.uk>
Message-ID: <HFEKILOLEFEFMKAECNDLGEPHDBAA.jason@jorendorff.com>

phil hunt wrote:
> The way i would do it, if I was implementing it, would be something
> like:
> 
> defmacro synchronized(lock_expr, code_block):
>    local lock
>    lock = ^lock_expr
>    lock.acquire()
>    try:
>       ^code_block
>    finally:
>       lock.release()      
> 
> Where the "local lock" declaration means something like: in the
> current namespace create a local variable name that doesn't clash with
> anything else.

So "local" is stripped out of the list of tokens that constitutes
the "synchronized" macro?

This seems to move away, a bit, from your previous notion that
macros should be expanded without any extra processing.  If so,
then the question becomes: why shouldn't "local" be the default?

My experience is that anything other than lexical scoping always
ends up being a real pain.  Names in classes, functions, and
top-level module code is all lexically scoped.  They should be
within defmacro too, by default.



> In something like C++ you could just put {...} round it can define
> variables within that and they'd be local to it, but you can't do 
> that in Python.

As a digression, here's the best I can do in C++.


// synchronized() in C++
// totally untested, might not compile

// Locking interface, which lockable objects must implement
class ILock {
  public:
    virtual void acquire() = 0;
    virtual void release() = 0;
};

class Locker {
  public:
    Locker(ILock &lock) : _lock(lock) { _lock.acquire(); }
    ~Locker() { _lock.release(); }
    operator bool() { return true; }
  private:
    ILock & _lock;
};

// Uses the unfamiliar standard C++ "if (decl)" syntax
#define synchronized(lock_expr) \
    if (Locker _synchronized_magic_object(lock_expr))

## Jason Orendorff    http://www.jorendorff.com/




From marknenadov at winningteam.com  Mon Feb 25 13:51:47 2002
From: marknenadov at winningteam.com (Mark J Nenadov)
Date: Mon, 25 Feb 2002 12:51:47 -0600
Subject: XML RPC Woes
Message-ID: <1846720022125185147375@winningteam.com>

Brian,

>Close. What you actually want to do return/pass a xmlrpclib.Binary
>object. You would use it this way:

Yes,  that is exactly what I needed. 

However there is some odd behavior going on. 
After three or four identical calls to the XMLRPC
server,  the server does a core dump on me.

Here is a summary of what I am doing:

The client passes the mcrypt'ed string (that was
encoded with xmlrpclib.Binary()). The function 
that  is being called on the server then makes its
 own  mcrypt'ed string, encodes it with 
xmlrpclib.Binary(), and then uses == to compare 
them. I assume that the comparision will work
 because the Binary class defines a "__cmp__()" 
method.

The client (simplified version), does this:

data = mcrypt_encode( "Msg", "pass1")

try:
	print con.dosomething( Binary( data ) )

except Error, v:
	print "ERROR", v


The server (simplified version), does this:

def dosomething( data ):
         good_data = mcrypt_encode("Msg", "pass1")
         return data == Binary( good_data )

I did a quick check to confirm that
the 'data' variable and Binary( good_data)
were actually xmlrpclib.Binary instances--and
they were. Any clue why I would be getting core
 dumps on the server side?

~Mark


_____________________________________________
Free email with personality! Over 200 domains!
http://www.MyOwnEmail.com




From st-newsgroups at thomas-guettler.de  Mon Feb 25 09:49:28 2002
From: st-newsgroups at thomas-guettler.de (Thomas Guettler)
Date: Mon, 25 Feb 2002 15:49:28 +0100
Subject: MSWord and Python
Message-ID: <3C7A4EF8.4010404@thomas-guettler.de>

Hi!

I am new to Python and COM.

1. Where can I find a introduction on the net?

2. Where can I find the documentation of methods that MSWord
  exposes by the use of COM?

3. I want take a word template. Replace text in it, and save it again.
  Has someone code examples fot something like this?

  Thank you, thomas



From brueckd at tbye.com  Fri Feb  1 13:01:18 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Fri, 1 Feb 2002 10:01:18 -0800 (PST)
Subject: Java vs Python Benchmarks: Java is faster
In-Reply-To: <a3ekkn$1mv$1@news.island.liu.se>
Message-ID: <Pine.LNX.4.33.0202010952520.21238-100000@ns0.tbye.com>

On Fri, 1 Feb 2002, Anders Dahlberg wrote:

> could re-run them without it. Maybe the low number of replies > is
> because it's not suprising that a semi-compiled-to-native program is >
> faster?
>
> well, but why would you want to compare interpreted java with python?

Well for one, it'd be nice to be a little closer to comparing apples to
apples (how fast a VM can interpret and execute the bytecodes). Secondly,
I was very suprised that, with a JIT, Java was only meagerly faster than
Python.  I was expecting an order of magnitude faster for most stuff, so I
was curious to know how slow Java was without the JIT. Finally, even
though Java and Python are wildly different in many ways, knowing the
interpreted vs. JIT'ed speeds might give a little insight into the
potential benefits of a Python JIT.

If I was really really interested in the Python-Java debate I'd go do the
tests myself, but since I'm just sorta curious I was asking Bryan in the
hopes that he still had the tests setup and it'd be easy to re-run them.

> code during run-time? If you were to compare interpreted java with python
> should you then compare interpreted c++ with java (same ballpark)?

Sure, why not?

-Dave




From quinn at schilling.ugcs.caltech.edu  Fri Feb  1 16:42:30 2002
From: quinn at schilling.ugcs.caltech.edu (Quinn Dunkan)
Date: 1 Feb 2002 21:42:30 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com>
Message-ID: <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu>

On Fri, 1 Feb 2002 15:02:18 +1100, Peter Milliken <peter.milliken at gtech.com>
wrote: >
>> Usually the problem with these sort of tools is that it takes too much
>time
>> to set up and use them, and further, I suspect that the benefits are
>> somewhat negated by consistently having to "hit key sequence to generate
>> template, go back and edit template" versus occasionally "make a typo that
>> I have to fix".  Then again, I suppose taking the time to become
>proficient
>> with these tools would probably make a difference, so I won't disagree.
>>
>
>You've tried the wrong ones there mate. I agree, if the tool gets in the way
>then it isn't worth the effort. The one I am refering to is fairly
>unobtrusive to the programmer i.e. if<fn key of your choice> generates
>
>  if {expression}:
>    {statement}...
>  [elif_part]...
>  [else_part]
>
>The cursor is position within the "{expression}" placeholder automatically
>and when you start typing it is automatically deleted by the editor as you
>type the first character. Navigate backwards and forwards to the other
>"placeholders" with a couple of other function keys and select from menu
>choices when you attempt to "expand" a placeholder (such as "statement")
>i.e. while, for etc etc. Don't want a placeholder (such as the "elif_part"
>above) then hit another function key and the editor moves everything around
>(reasonably :-)) correctly :-). Of course, you only get this with Emacs
>though...... :-) But that was one of my points - select the tools with the
>most productivity features or perhaps graft the features into your tool of
>choice :-).

In my case at least, after memorizing all the various FKeys (never mind being
stuck if you like keyboards without all the extra gunk) to the point where I
could hit them without thinking (the way I can hit alphabetic keys), I'd have
spent rather more time than it would have taken to type 'elif:' and 'else:'.
And I can't even reliably hit numbers or #$%^&* characters without looking, let
alone go all the way up to wherever the FKeys happen to live on this keyboard.

When using vim, I do have one macro that I think I find useful: ';di' writes
out 'def __init__(self,):' and positions the cursor after the comma, which
occaisionally saves a microscopic amount of time, but I'm not even sure if that
really saves any time worth writing and memorizing the macro for.  I spend a
vastly greater amount of time staring at the screen muttering "hmm... no,
wait... oh, gross...".

Of course, vi itself requires loads of memorization, and I'm not convinced
it's really worth it.  For example, to move a word two words back to the end of
the line, is it really quicker to type '^[2Bhdt $pa ', or is it quicker to
select the word with the mouse, cut, click, and paste?

>I mean, you have a good point about familiarity etc but assuming a "level
>playing field" in that area, the programmer *should* be better off with the
>tool that offers the most features. I view programming tools just like any
>other trade i.e. use the one best fitted for the job. Many programmers take
>the approach of "well, I'm good with a hammer, so that's what I'll use with
>this job" - even when the job requires the use of a jigsaw! :-) I know this
>is a far fetched analogy but not too far off the mark :-).

Well, say it took a few months to really learn how to use a hammer.  And
suppose that a hammer could put in screws, not as well as a screwdriver, but
pretty well.  If you needed to put in a lot of screws, you could write a script
for the hammer for putting in screws.  Still not as easily as a screwdriver,
though.  Then your analogy would be closer to the mark, and the attitude of
many programmers easier to understand :)

Also, the first sentence in the quoted paragraph seems to contradict the rest.
I'd think that the tool with the most features is meant for *everything* (i.e.
not the "best tool for the job" but the "one tool to rule them all"
philosophy).  Is a carpenter best served by a swiss army hammer with a philips
head on one end, and a jigsaw blade in the handle?  It does offer the most
features...

Ok, I think the metaphor has been abused enough by now :)


From bosahv at netscapenospam.net  Wed Feb 13 04:47:28 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Wed, 13 Feb 2002 09:47:28 GMT
Subject: Random Module help
References: <20020210215559.13516.00000455@mb-cq.aol.com> <whK98.289$ab7.1102@news.hananet.net>
Message-ID: <QCqa8.373000$pa1.87470284@news3.rdc1.on.home.com>

Wouldnt you have to dump your initial deck into a string or tuple and then
remove cards as they were played?

That would set you up to deal elegantly with multiple hands from the same
deck.




From brueckd at tbye.com  Fri Feb 15 10:10:47 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Fri, 15 Feb 2002 07:10:47 -0800 (PST)
Subject: Java vs Python ... ???
In-Reply-To: <3C6CAFF8.4D3B5092@dotcom.ee>
Message-ID: <Pine.LNX.4.33.0202150709130.25420-100000@ns0.tbye.com>

On Fri, 15 Feb 2002, Martin Roos wrote:

> > I've been learning JAVA for a while, but now I've heard avout Python.
> > Any coments concerning these two languages?
> >
> java is powerful and quite fast with jit,
> python is more fun and slower ....
>
[snip]
>  java is for serious kinda stuff , python is for testing etc.

100% rubbish.




From phr-n2002a at nightsong.com  Thu Feb  7 20:09:45 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 07 Feb 2002 17:09:45 -0800
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <3C629FF9.CF4670FF@metaslash.com> <a3u7hl$1an8dd$1@ID-98166.news.dfncis.de>
Message-ID: <7xpu3gixt2.fsf@ruckus.brouhaha.com>

"MDK" <mdk at mdk.com> writes:
> > from  mymod import *
> > mymod_dosomething = dosomething
> >
> > def dosomething(x):
> >     mymod_dosomething(1000)
> >
> > Neal
> 
> Neat!  Thanks, Neal.

Alternatively, in Python 2.x,

  from mymod import dosomething as mymod_dosomething

Generally, it's best to avoid "from mymod import *", precisely
because of collisions like this.  As other people have mentioned, it's
better to say

  import mymod

  def dosomething(x):
    mymod.dosomething(1000)


From roy at linksheaven.com  Fri Feb 22 15:51:00 2002
From: roy at linksheaven.com (Roy Madden)
Date: Fri, 22 Feb 2002 20:51:00 GMT
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie> <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk>
Message-ID: <Uayd8.5117$D6.14249@news.iol.ie>

On Fri, 22 Feb 2002 14:49:38 +0000, philh at comuno.freeserve.co.uk (phil
hunt) wrote:

>What do you consider the other 2 methods to be? I can think of 
>encoding it in the URL, and the server remembering the browser's IP 
>address. Are there others?

Yeah, the generally accepted methods by which you pass session ids are:
- cookies
- url like you say
- hidden form fields

Remembering the browsers ip address is often combined with other methods
to increase security (e.g. some cookie id generation schemes), but not
used on it's own! Listing the advantages and disadvantages of the above
schemes from a security perspective will almost always (almost, if
implemented properly etc etc) see cookies as the best method.


Roy


 
-- 
http://www.linksheaven.com


From member at dbforums.com  Mon Feb 18 07:17:33 2002
From: member at dbforums.com (krallabandi)
Date: 18 Feb 2002 06:17:33 -0600
Subject: Can Tk be connect with Copyboard of Linux
References: <a4jlhf$6uv$1@pegasus.csx.cam.ac.uk>
Message-ID: <3c70f0dd_1@spamkiller.newsgroups.com>

e



--
Posted via dBforums
http://dbforums.com


From raphaelr at softhome.net  Sun Feb 17 10:29:25 2002
From: raphaelr at softhome.net (Raphael Ribeiro)
Date: Sun, 17 Feb 2002 12:29:25 -0300
Subject: Python.NET
Message-ID: <00b501c1b7c7$e2cac900$02c8a8c0@dsl>

I want to know about you guys what do you think of Python.NET . ( I'm thinking on learning python with .net , but i don't know if it is a good option  , so i want to hear comments!)

Any comments would be appreciated.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020217/a4a66ca6/attachment.html>

From aahz at panix.com  Wed Feb  6 20:08:34 2002
From: aahz at panix.com (Aahz Maruch)
Date: 6 Feb 2002 17:08:34 -0800
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de>
Message-ID: <a3sk2i$67m$1@panix2.panix.com>

In article <3c6164f0.50243062 at news.t-online.de>,
Gerson Kurz <gerson.kurz at t-online.de> wrote:
>
>Well, it sure doesn't look like that in their programming contest:
>http://www.google.com/programming-contest/index.html#rules
>
>Or was the Google-Python relation just an "urban legend"?

No, it's not an urban legend, as a quick perusal of
http://www.google.com/jobs/openings.html
will attest.  However, they don't AFAIK use Python for their core
high-performance code, and it appears that this contest is slated for
that role.

What does surprise me is that they think Java can do better than Python.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From aleax at aleax.it  Fri Feb  1 17:22:35 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 22:22:35 GMT
Subject: confused by isinstance()
References: <3c5717bb$0$23518$626a54ce@news.free.fr> <3c5a3352$0$4604$626a54ce@news.free.fr> <Xns91A85C660FC3Bduncanrcpcouk@127.0.0.1> <3c5b0a9f$0$23532$626a54ce@news.free.fr>
Message-ID: <LyE68.18835$3J1.584901@news2.tin.it>

Francois Petitjean wrote:
        ...
> Where in the documentation is explained what means "Importing a module
> executes each line in the file". 

Language reference manual.

> The lines are syntaxically checked, sure,
> but most of the work is not done at this stage:

No, but every top-level statement is still executed.

> def curve(x):
>     return return 10.0*math.sin(x)/x
> conversion to byte code.

def is the top-level statement, and it's executed on import.

>>Every time you execute a class statement, even if the class name is the
>>same, you create a new class.
> This is the crux of the problem, but after the first import, it seems that
> successive import of a module give the same class, due to the sys.module
> caching. Right?

Only if the imports use the same name.  In your case, you were
importing the same module as __main__ (implicitly) and foo
(explicitly).  sys.modules is checked with the modulename as
the key, after all.  (You may also choose to import with various
modulenames -- check standard module imp for that, although
you rarely need this).  Note that "import x as y" is not an example:
the module is imported (and set/checked in sys.modules) with
a name 'x', it's only the local binding that uses 'y' here.


> So, if I want to be on the safe side and to test a module Foo as it
> appears from outside, I can begin the __main__ part with import Foo. In
> this case, instanciation is of the form x = Foo.Foo(), all is good except
> for the doubling of the memory to store the class.

If all the top-level statements in the modules are OK to execute
twice, yes.  But if the top-level does e.g. an append to a file, you'll
find two appends using this approach.


Alex



From jason at jorendorff.com  Wed Feb 20 14:44:12 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 20 Feb 2002 13:44:12 -0600
Subject: Embedding: How to set globals for PyObject_Call
In-Reply-To: <79b2b0bb.0202180959.16ae249c@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLKEAPDEAA.jason@jorendorff.com>

Scott Gilbert wrote:
> I dove through the code, and it appears that I need to setup a
> PyFrameObject to be associated with my PyThreadState.  I wasn't able
> to figure out how to do that though.
> 
> As you can see in my example below, I have the globals sitting right
> there.  I just don't know how to tell Python that those are the
> appropriate globals while invoking PyObject_Call(...).  A similar
> problem occurs when evaluating "globals", "locals", "dir", or
> functions that call those.

Really?  The part about "functions that call those" sounds a bit off.
I thought Python functions that call globals() would work.
Each Python function contains a pointer to its globals namespace
(the fn.func_globals attribute), and calling a Python function
creates a PyFrameObject.  It seems as though it should work.

So, what are you really trying to do?  :)  I have a feeling that
calling vars() directly from C is a mistake; there's probably an
easier way to do what you need.

For example, if you just want to look at module's namespace,
that's easy:

  result = PyModule_GetDict(module);

## Jason Orendorff    http://www.jorendorff.com/



From lucio at movilogic.com  Tue Feb  5 10:51:42 2002
From: lucio at movilogic.com (Lucio Torre)
Date: Tue, 05 Feb 2002 12:51:42 -0300
Subject: ANNOUNCE: autochino release 1
Message-ID: <3C5FFF8E.8090004@movilogic.com>

autochino release 1 is now available

wee it working at:
http://vampira.dyndns.org/autochino

download it from:
http://vampira.dyndns.org/autochino_rel1.zip

whats autochino?
autochino is a set of web tools i built to help me while studing 
chinese. of course it was made with python.

the features:
- complete pronunciation audio
- most of the bopomofo pronunciation with tables and practice
- tones practice
- vocabulary and practice (character, pinyin and sound)
- caligraphy sheets
- its expandable
- included dictionary (cedict)
- includes gif characters, so no extra software is needed
- downloadable. you can run its own tiny web server in you desktop if 
you dont want to run it from my machine, or to make mirrors

help is sought for the following tasks (time is scarce!):
- extension of the dictionary: unihan.txt / cedict /others merge, better 
structure and tools
- internationalization and localization to english (its in spanish now)
- more lessons
- more features

if you find it useful, please, please, please, drop me a line. any other 
comments are also welcome!

sayonara! :)





From bokr at oz.net  Sat Feb 23 15:50:19 2002
From: bokr at oz.net (Bengt Richter)
Date: 23 Feb 2002 20:50:19 GMT
Subject: Puzzled: Y am I ending up with extra bytes?
References: <Xns91BF3BEAA45A2AusYourStandingInIt@130.133.1.4>
Message-ID: <a58vab$9hv$0@216.39.172.122>

On 23 Feb 2002 19:23:44 GMT, "A.Newby" <deathtospam43423 at altavista.com> wrote:

>Why is this happening? I read large chunks of data from a text file, 
>according to byte locations specified on another file, and for some reason, 
>this function (below), spits out a few extra bytes.
>
>Here's the code, as entered into the Python shell...... 
>
>
>    	index = map(string.rstrip, open('D:\cgi-bin\indx.txt').readlines())
>    	#this opens the index file, which has precise byte locations of each 
>    	#chunk of data I want to extract from the log.txt file, and turns it 
>    	#into a list.
>
>
>    	def fish(end, start, deduct):
>	    	chat = open('D:\cgi-bin\log.txt', 'r')
>	    	g = int(index[end]) - int(index[start])
>	    	chat.seek(int(index[start]))
>	    	print chat.read(g - deduct)
>
>
>
>Now, if I enter the following into the command line ...
>
>fish(205, 204, 0)
>
>,,, I get about four extra characters. God knows why! So that's why I try 
>...
>
>fish(205, 204, 4) 
>
>... And it seems to work perfectly. I can even "fish" up to about about ten 
>"index" lines with it. But as soon as I try and fish out any more than 
>that, I get the dreaded extra 4 bytes of code again. Why?
>
>Now I know what you're thinking. You're thinking that perhaps my index file 
>is corrupted, and hasn't got an accurate account of precisely what's in the 
>log file. I suspected that might be the case myself, but ... when I try 
>fishing out each chunk of data individually, it works fine! I can even do 
>this ...
>
>for x in range(1, 90):
>	fish(211+x, 210 + x, 4)
>
>...... without ending up with that extra data I don't want. However, this 
>method is too slow for my purposes. Plus, I really wanna know what it is 
>that's going wrong.
>
>Can anyone spot it?
>
If your index was created by finding something prefixed to the substring
you want, that gives a start position, but what about the end? If there is
a prefix for the next item, the difference between two indices will include
a prefix and possibly extra stuff, like CR/LF etc:
...<prefix>what you want<extra><prefix>what you want next<extra2><prefix ...
           ^-index[n]                  ^-index[n+1]
           |<----- wrong length ------>|

Just a guess. Try printing enough raw data to see what's there,
e.g., print `fish(206,204,0)` and some other examples (note backquotes).

BTW, you might be able to map int instead of string.rstrip and get a list
of integers that you can use directly instead of converting them later.

HTH
Regards,
Bengt Richter



From niemeyer at conectiva.com  Sun Feb 10 02:04:26 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Sun, 10 Feb 2002 05:04:26 -0200
Subject: ANN: python-mhash 1.0
Message-ID: <20020210050426.A4360@ibook.distro.conectiva>

Hello everyone!

Just a side note before introducing this interface to mhash. Even if
you're not interested in the hash algorithms by theirselves, this
extension is a nice example of usage of some of the new features
Python 2.2 offers at the C level, including subclassing of types.

http://mhash.sourceforge.net

Description
-----------

The python mhash module provides an interface for the mhash library.
The mhash library provides an easy way to access strong hashes such as
MD5, SHA1 and other algorithms.

Features
--------

The python-mhash module provides the following features:

o  Free software (GNU Lesser General Public License)
o  Reentrancy support (*)
o  Support for known strong algorithms (*)
o  Support for less-known algorithms such as GOST (*)
o  Support for HMAC hashing (*)
o  Support for Python newstyle classes, allowing subclassing
o  Unittests testing functionality provided by the module

(*) mhash library features

Disclaimer
----------

Please read the file LICENSE before using python-mhash. It's
licensed under LGPL.

New Versions
------------

Information on how to retrieve new versions of mhash and python-mhash
can be found on http://mhash.sourceforge.net.

Credits
-------

Please read the file AUTHORS.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From sabren at manifestation.com  Sat Feb  9 17:56:20 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 9 Feb 2002 17:56:20 -0500 (EST)
Subject: Is this..... evil?
In-Reply-To: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4>
Message-ID: <Pine.LNX.4.44.0202091754450.22239-100000@mercury.sabren.com>

On 9 Feb 2002, Philip Swartzleonard wrote:

> class Reciever:
>     def message( self, message ):
>         methname = 'rec_' + message.type
>         if hasattr(self, methname):
>             if callable(getattr(self, methname)):
>                 getattr(self, methname)(message)

[snip]

> Anyway, i was just wondering, since this seems so great, is there 
> anything wrong with doing things this way that i'm not seeing?

I do the exact same thing with web apps. I have an "Actor"
class, which, when you pass it an "action" on the command
line, dispatches it to the approprirate act_xxxx() routine.
It's one of the cleanest ways I've seen to write web
applications.

Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From mhammond at skippinet.com.au  Mon Feb 11 17:59:56 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 11 Feb 2002 22:59:56 GMT
Subject: I wish.............
References: <mailman.1013443881.3294.python-list@python.org>
Message-ID: <3C684D2C.6040307@skippinet.com.au>

Michael Chermside wrote:
>  > .....that we quit calling Python as CPython.
> 
> However, most of the time when people say "Python" meaning "CPython" 
> there isn't any confusion, so I don't really mind. But using the terms 
> properly helps people keep in mind that things like Jython (and Pippy, 
> and a few other implementations out there) 

Just to be clear, I believe that Pippy is a port rather than an 
implementation.  It is a port that has ripped the guts out of the 
implementation (for size purposes) but that is still a port :)

The only implementations I am aware of are CPython, Jython, Vyper and 
Python for .NET.  I also understand only the first 2 of them are 
currently viable.

Mark.



From SBrunning at trisystems.co.uk  Mon Feb 25 07:44:12 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 25 Feb 2002 12:44:12 -0000
Subject: Remote debugging ?
Message-ID: <31575A892FF6D1118F5800600846864DCBCCFB@intrepid>

> From:	F. GEIGER [SMTP:fgeiger at datec.at]
> > > It must not be freeware.
> >
> > What does that mean?
> 
> In German it reads "Es muss nicht Freeware sein". In English this should
> read "It does not need to be freeware".
 
Yes, but does he mean *freeware*, or *free software*?

Take it, Oleg!

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From martin at v.loewis.de  Mon Feb 25 18:31:40 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 26 Feb 2002 00:31:40 +0100
Subject: Tkinter.createfilehandler analog for Windows?
References: <mailman.1014678341.1704.python-list@python.org>
Message-ID: <m3wux13zqb.fsf@mira.informatik.hu-berlin.de>

Andrew Kuchling <akuchlin at mems-exchange.org> writes:

> The Tkinter.createfilehandler() method isn't supported on Windows,
> only Unix and Mac.  So, when confronted with moving a Python program
> that uses createfilehandler() to Windows, what do people do?  Use
> threads?  Use Tcl's channel C API (but that seems inaccessible from
> Python)?  Something completely different?

Use threads. I believe using channels will lead nowhere. I assume this
is for sockets? You *could* use the Tcl socket(n) library, which has a
different programming model than Python sockets (which follow C
sockets closely). I'm not aware of anybody who has taken this route,
but I think it could work.

Regards,
Martin


From philh at comuno.freeserve.co.uk  Fri Feb 22 09:49:38 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 22 Feb 2002 14:49:38 +0000
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <6Ufd8.5038$D6.14223@news.iol.ie>
Message-ID: <slrna7cmk2.v6u.philh@comuno.freeserve.co.uk>

On Fri, 22 Feb 2002 00:02:10 GMT, Roy Madden <roy at linksheaven.com> wrote:
>On 19 Feb 2002 21:31:00 GMT, Lutz Schroeer
><Lutz.Schroeer at kybernetik-manufaktur.de> wrote:
>
>>I'm creating a website which makes intensive use of cgi scripts and there 
>>is the possibility for the visitor to register as a user for having the 
>>ability to customize the site appearance, contents and other stuff.
>>
>>Jumping from one script to the other I need to remember at least the user's 
>>name. The W3C recommends using cookies (RFC 2109). Unfortunately many 
>>people don't like cookies and filter them using WebWasher or a personal 
>>firewall. 
>>
>I'm going to try and dig up some stats for you, but from experience and
>research the population of those who do not accept cookies is very very
>small (albeit tend to be technically proficient and vocal online). I once
>worked for an online bank with operations in Germany, and we considered
>this issue *very* carefully before choosing cookies for our user
>interface.
>
>A slightly larger population do not 'like' them, but of the 3 methods of
>maintaining session state they are the most secure option (waiting for the
>flames to start :) ) - if security is an issue for you, and the 'like'
>issue is a matter of user education.

What do you consider the other 2 methods to be? I can think of 
encoding it in the URL, and the server remembering the browser's IP 
address. Are there others?

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From peter at engcorp.com  Wed Feb  6 23:29:51 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 06 Feb 2002 23:29:51 -0500
Subject: Quixotic Thread:  Author and Intent Revealed....
References: <mailman.1013035874.11097.python-list@python.org>
Message-ID: <3C6202BF.2599BF7@engcorp.com>

mo wrote:
> 
> YKW* is the alter ego of the BFDL.  He and the other Paters of Python
> are preparing this community for the  momentous release of p_VIC
> scheduled for release 4/1/02.
> 
> PS:  I realize that this premature disclosure will result in an order
> from the Directorate of the PSU launching a squadron of Norwegian
> Blues to deliver a massive quantity of Norwegian Blue quano in my
> direction.;)
> 
> *YKW = You Know Who
> 
> Mo

So they would be the ... (wait for it) ... "Mo batter Blues" ?

<ducking>


From david.abrahams at rcn.com  Wed Feb 27 09:29:56 2002
From: david.abrahams at rcn.com (David Abrahams)
Date: Wed, 27 Feb 2002 09:29:56 -0500
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
References: <mailman.1014748405.28236.python-list@python.org>
Message-ID: <a5iq5k$j42$1@bob.news.rcn.net>

I feel I should report that I was mistaken about the PDT developers'
commitment to support. I got a quick response (below) to my public complaint
and even a phone call yesterday! I'm still waiting for them to release a
fixed executable, but I am optimistic.

-Dave



"Sameer Suresh Shende" <sameer at cs.uoregon.edu> wrote in message
news:mailman.1014748405.28236.python-list at python.org...
> Hi David,
> This looks like a bug in the IL Analyzer that is easily fixed.
> I'll look into this. We cannot distribute the sources of the IL Analyzer
> as they have routines that walk the intermediate language AST that are
> EDG proprietary. Our CREDITS file in PDT lists all contact e-mail
> addresses, but due to some recent security issues, we were advised to
> remove any personal information from the LANL webpages.
> If the code compiles on different OSes (perhaps under Linux), I
> wonder if you can parse it using PDT under Linux properly. We'll look into
> this rightaway and provide you with a bug fix soon.
> Thanks for your interest in SILOON and PDT.
> - Sameer





From member at dbforums.com  Mon Feb 18 07:17:33 2002
From: member at dbforums.com (krallabandi)
Date: 18 Feb 2002 06:17:33 -0600
Subject: ODBC on unix w/o mxODBC
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>
Message-ID: <3c70f0dd$2_1@spamkiller.newsgroups.com>

p



--
Posted via dBforums
http://dbforums.com


From jimd at vega.starshine.org  Wed Feb 20 06:37:49 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 20 Feb 2002 11:37:49 GMT
Subject: Mystery Guest from Number Theory
Message-ID: <a501qd$1c3k$1@news.idiom.com>

 Years ago I know I read about a function which exhibits 
 very simple chaotic behavior.  I don't remember where I read
 about it (my first guess was Gleick's Chaos, Penguin Books, 1987,
 but I couldn't find it there).  So it was probably in some other
 tome on Number Theory.

 So, as odd as it sounds, I'd like to post the code that implements
 this here and see if anyone can name our mystery guest:

def next_it(n): 
        """Core function: what is this called in number theory?"""
        if not n % 2:   n=n/2
        else:           n=n*3+1
        return n

def do_it(n): 
        """How many iterations through next_it() to reach 1?"""
        it=0
        while n > 1: 
                it += 1
                n = next_it(n)
        return it

 This should be pretty readable, for any whole number, if it's even 
 cut it in half, otherwise multiply by three and add one.  Iterate until 
 you get to one.  (So far all whole numbers seem to reach 1, but the 
 number of iterations it takes to so is non-trivial.  I guess no one
 back then knew of a formula that would short-circuit the algorithm.

 Here's one more function to go wit those:

def max_it(l,u=None,f=None):
        """Return tuple of tuples: for each new max return the 
           n that generated it and the number of iterations it took to reach
           1; alternatively call function f for each new max found"""
        if u is None: # only one number given
                u=l # set upper bound
                l=2 # set lower bound
        if l < 2 or u <= l:  # check upper and lower bounds
                raise IndexError
                ## is that really the right exception for this case?
        max=0
        if f is None:   result = []
        else:                   result = None
        for i in xrange(l,u):
                x = do_it(i)
                if x > max: 
                        max=x
                        if f is None:   result.append((i,x))
                        else:           f(i,x)
        return result

 ... this one simply iterates over the mystery guest function 
 for all the numbers up to our argument (or between our ordered
 integer arguments and finds successive  "maximal" values for the 
 "do_it" function. 

 (Ooops, I forgot to check for integer and negative values in these; 
 undefined results for those).  

 Not surprisingly the number of iterations returned by do_it does
 get bigger for larger numbers.  However the iterations grow relatively
 slowly, so we have to work every harder to find a new n which returns
 a new maximal do_it().   So in the range 2:10000 (10e5) the last couple
 of tuples is: (3711, 237) and (6171,261), at n=156,159 do_it() reaches
 one in only 382 steps, and it goes from 410011 (@ 448) to 511935 (@ 469)
 to 626331 (@ 508 interations).  So we're taking 100K strides to find
 a mystery function increase of less than 100 iterations.  There's only
 one more n (837799) less than a million, and it reaches zero in only
 524 iterations.

 Whoever our mystery guest is, he won't ever attain the fame, fortune,
 and perrennial intrigue of prime or even perfect numbers, but I'd 
 still like to know his name.



From wayne at mishre.com  Mon Feb 11 20:18:00 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Mon, 11 Feb 2002 17:18:00 -0800
Subject: just a moron trying to get python working on win98 and personal web  server
References: <3C685DDE.AA8AFD10@kc.rr.com>
Message-ID: <3C686D48.6090103@mishre.com>

> RE: just a moron trying to get python working on win98 and personal web
> server

If you say so...

> i am trying to get python to work on my MS personal web server ( which
> is supposed to support python in a web url ).
> 
> what do i need to do?

Usually you need to make sure your Python scripts (.py files) are in the 
/cgi-bin/ directory and you have permissions to execute them.  Then at 
the top of the file put:

#! c:/python/python.exe -u

Where /python/ is the directory you installed Python at; this should be 
the very first line in the program, with nothing before the #! .

> i have set the PATH in autoexec.bat.

When you call the script from the web server it usually looks in the 
/cgi-bin/ directory, so the PATH isn't used.  Unless PWS does something 
completely non-standard, which wouldn't be too surprising.

> is there an apache alternative.

Apache does run on Windows, you might also want to look at Xitami (there 
were a few emails today about this same question on Xitami).  Apache 
might be a bit of overkill for Win98, but Xitami should work well.
-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From fperez528 at yahoo.com  Mon Feb 18 09:18:55 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 18 Feb 2002 14:18:55 +0000
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014062915.17675.python-list@python.org>
Message-ID: <a4rr7h$81c$1@peabody.colorado.edu>

>> Exactement.  Bravo!  Bien dit! :-)
> 
> I agree with Fran?ois and I'm english speaking.

Muy bien. Como parece que todos est?n de acuerdo en que es mejor mantener las 
discusiones en m?ltiples idiomas, dor?navant on changera de langue toutes les 
deux lignes. Apr?s les r?centes interventions de Mr. Rue, le quotient 
signal/bruit i?i ?tait d?ja en chute libre, alors il n'y a pas beaucoup plus 
qu'on puisse y faire.

Und da sind wir alle gl?cklich!

Cheerfully yours,

f.


From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 19:23:02 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 18:23:02 -0600
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <mailman.1013352256.7990.python-list@python.org> <3C66F2CA.802A00BF@earthlink.net>
Message-ID: <00ae01c1b292$463da920$0101010a@local>

----- Original Message ----- 
From: "Ron Stephens" <rdsteph at earthlink.net>


> Since id dint get any mail despite putting your insert into the top 
> pf mygale.py 
> 
> import sys
>     sys.stdout = open("/home/ron/.error-log","a")
>     sys.stderr = sys.stdout
>     print "\n(job name) Starting..."
> 
> I wonder if my mail is broken or if it is record somewhere else???

Exactly.  You misunderstood, I think... if you DIDN'T have the 
lines described, cron should send you mail.  With those lines, 
all your output from the script(s) should be in the file

/home/ron/.error-log

and you can view them with less, more, vi, etc.





From frederic.giacometti at arakne.com  Mon Feb 18 18:40:25 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Mon, 18 Feb 2002 23:40:25 GMT
Subject: Performances of Pyhton programs
References: <mailman.1014062970.17792.python-list@python.org>
Message-ID: <Jhgc8.63659$dG.26055877@news1.rdc1.sdca.home.com>

"Christian Tanzer" <tanzer at swing.co.at> wrote in message
news:mailman.1014062970.17792.python-list at python.org...

Roy Smith <roy at panix.com> wrote:

> bairef2 at cti.ecp.fr (Fran?ois Baire) wrote:
> > Are there any benchmark or does anoyone know what are the performances
of a
> > program written in python (in terms of CPU and RAM) compared with other
> > languages (especially C++ and compiled languages) ?
> >    I've met many articles which explain that python programs are slowler
> > because it uses the interpreter, but none says how much slower it is ! I
> > would like to know this for a rather big OO program which uses standard
> > fonctions of the language.
>
> There is little doubt that a program written in C/C++ will be faster than
> the same algorithm implemented in Python (or any of the common "scripting"
> languages such as Perl or TCL).  How much slower depends to a very large
> degree on exactly what you're doing.

+True, but misleading. A Python solution might use very different
+algorithms than a C/C++ solution. As the first working version of a
+Python solution is achieved much earlier than a C/C++ solution, there
+is a lot of time left to optimize if and where necessary.

+In my experience, a Python solution might be anywhere between 100
+times slower to 1000 times faster than a C/C++ solution.

But in practice, your python program does not work alone, but calls existin
C/C++ libraries, which will actually be doing significant computations or
database crunching, and Python only links to these (for instance, who would
even think of rewriting a Mesh solver in Python...).
This will be especially true in the 'large OO program' approach Francois
refers too.
In such situations, the overhead due to the Python solution to introduce
will be less or in the order of 10-30%, in a typical applicative context.

FG









From db3l at fitlinxx.com  Mon Feb 25 23:05:27 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 23:05:27 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu>
Message-ID: <u8z9gx4zc.fsf@ctwd0143.fitlinxx.com>

"Locke" <brown.2053 at osu.edu> writes:

> The argument to recv is supposed to be the buffer, acording to the
> documentation. But the doc doesnt say: buffer measured in what? Is it bytes?
> Octets? (by the way, what is the diff between a byte and an octet?)

An octet is formally an 8-bit value.  A byte was traditionally defined
as the unit of data used to represent a character (thus the term DBCS
- double byte character set for those sets that needed two bytes to
represent a character), and less than a word.  On most typical systems
nowadays they tend to be the same thing (8-bits), but historically
that wasn't always true.

Network protocols are often defined in terms of octets because it
doesn't have the potential to be platform specific.

> See, I am writing a pop3 client type thing, after requesting a message from
> the server, my first s.recv tells me how long the message will be in octets.
> So then I call s.recv again passing the number of octets as the argument to
> s.recv. Shouldn't this download the entire message? What actually happens is
> that it cuts off right in the middle of the message.

This is standard behavior for a stream socket (like TCP, which is what
you are using to communicate with the POP server).

Between the two endpoints of a socket, and then within the network in
the middle, data traversing the connection is being subjected to
various latencies and buffering.  So there need not be - and generally
is not - any direct relationship to how data is written at one end
(e.g., chunks or number of I/O operations) and how data is received at
the other end.

In other words, if the POP server tells you there are 1000 octets
coming, and even if it issues a single write() call for all 1000
octets, you may end up receiving them all at once, in 10 chunks of
100, 5 chunks of 200, or 1000 chunks of 1.  And this is even if you
tell recv() to accept up to 1000 octets in each call.  You have no way
of knowing, and its crucial that you should avoid creating any
dependencies on such behavior in your code.

> What's my problem? Is there an upper limit on what the buffer argument can
> be to recv? Or does pop3 just refuse to send the whole message with one call
> to recv? I don't think it is that, because what it sent me was like 600some
> characters, which isnt a power of two or any other computer-ish number.

How much you happen to receive is going to be dictated by a
combination of the socket buffering on the POP server, the latency
between the server and your client (and any data loss or retries that
occur along the way which TCP takes care of transparently), and then
socket buffering in your protocol stack.  If you were to, for example,
delay a few seconds before issuing your recv() call you might happen
to let your stack receive all the data and give it to you in one
chunk, but it would be a bad design.  And if the total amount of data
exceeded the socket buffers and TCP window, then it would get held up
at the sending end until you started draining some out at your client.

As you note in a later post, asking recv() for data if there is none
available on the socket can create blocking (unless you mark the
socket as non-blocking).  So the way you'd get around that in this
case is build a loop that worked to receive any remaining data (out of
the total expected) until it was done, and then stopped calling
receive.  Something like:

   # Assume msg_octets tells you how many octets you expect on socket s

   # Keep receiving until we have it all.  Append to list for efficiency
   data = []
   while msg_octets:
      data.append(s.recv(msg_octets))
      msg_octets = msg_octets - len(data[-1])

   # Merge list contents into a single data buffer   
   data = string.join(data,'')

This will keep receiving until it runs out of the total message size,
decreasing the possible maximum size based on the data already
received.  Note that it destroys msg_octets and assumes that you
aren't limiting your maximum size up front (e.g., that you're going to
hold the entire message in memory), but it should give you the idea.
It ends up with the received contents buffer referenced by 'data'.

In this case, you're also depending on TCP (as a stream protocol) to
guarantee that data written in at one end of the connection will
arrive intact, and in sequence, at the other.  So as long as the
server has properly told you how many octets it is sending, the recv()
calls will be safe even on a blocking mode socket.

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


From akuchlin at crystal.mems-exchange.org  Wed Feb 27 18:17:24 2002
From: akuchlin at crystal.mems-exchange.org (A.M. Kuchling)
Date: 27 Feb 2002 23:17:24 GMT
Subject: Tkinter.createfilehandler analog for Windows?
References: <mailman.1014678341.1704.python-list@python.org> <m3wux13zqb.fsf@mira.informatik.hu-berlin.de>
Message-ID: <slrna7qq83.isd.akuchlin@crystal.mems-exchange.org>

In article <m3wux13zqb.fsf at mira.informatik.hu-berlin.de>, 
   Martin v. Loewis wrote:
> Use threads. I believe using channels will lead nowhere. I assume this

Oh, well; thanks, Martin!  Changing the code to use a thread was
simple enough, but it seems to be a bit destabilizing; every few
executions Python will crash and crash hard, getting terminated by
Windows and sometimes generating an error log (which I can't actually
find).  "Now you have two problems", I guess.

--amk                                                  (www.amk.ca)
I had it in me to be the Pierce Brosnan of my generation.
    -- Vincent Me's past career plans in EGYPT #1



From peter.milliken at gtech.com  Tue Feb  5 16:54:01 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Wed, 6 Feb 2002 08:54:01 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com><Wls68.14178$om6.357057@news1.tin.it><mailman.1012590723.7301.python-list@python.org><a3kb3p$se44@news1.gtech.com><3C5EE7C2.4DF4A885@ccvcorp.com><a3n03m$tfn7@news1.gtech.com> <mailman.1012863309.22323.python-list@python.org>
Message-ID: <a3pkgq$2tq1@news1.gtech.com>

G'Day Cliff, I won't reply to the earlier email in the thread - that one is
finished IMO - thanks for the insight into how you work, it gives me
something to "chew" over :-)

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote in message
news:mailman.1012863309.22323.python-list at python.org...
> On Tue, 5 Feb 2002 08:53:27 +1100
> Peter Milliken wrote:
>

[snip]

> > code. If the individuals involved approach the problem with the correct
> > frame of mind then I have never, ever seen inspections/reviews fail! It
> is
> > only when it is not tackled with the right sort of commitment that it
can
> > appear not to work.
>
> Exactly.  I still venture that they won't ever be "tackled with the right
> sort of commitment" because it's no fun having to read other people's
code,
> and lack of enthusiasm always seems to win over policy and "requirements"
> in the end...
>

Not so, in my experience, if you get the code reviewed before it is compiled
or (at very least) before it is unit tested (which is my preference) then
there is a real incentive for reviewers to look at it because they *know*
there are bugs to find. Incentive in code reviews dwindles towards zero when
you "know" that the author has already (supposedly) "tested it to death"
:-). Sometimes if you know the programmer is bad enough there is still
incentive, but then you have to battle the frustration of reviewing
absolutely gross code :-)

My experience ( :-) ) is that if reviews (or inspections, whatever you want
to call them) are carried out with diligence then the project is successful,
the unsucessful projects had the (one of) symptoms of either no reviews or
reviews that were paper only i..e. they satisfied a QA audit and that is it
:-). Of course, this was from a background of SRS, SDD (Preliminary and
Detailed), Test Plan, Test Documentation etc etc etc ad nauseum. In the
"commercial" world of requirements discussion with the customer and then
informal talks between programmers and then code, there aren't too many
steps to review! :-)

>
> > I get somewhat frustrated sometimes when I look at yet more attempts to
> > solve the same problem when I believe there are already workable
> solutions
> > (this will, I am sure be disputed :-) but is only a statement based upon
> > personal experience :-)). I think it is most likely the lack of
> > self-discipline (in programmers, not themselves) that caused the authors
> to
> > look for alternative methods - also, the fact that they needed something
> new
> > to write about otherwise they wouldn't be able to publish a book! :-) I
> must
> > admit that in a weak moment, I purchased a copy of Extreme Programming,
I
> > read it on the train one evening (not the whole book! :-)) and saw
enough
> to
> > make me put it where it is now, on my shelf gathering dust :-).
>
> I haven't read much beyond magazine articles regarding XP myself, but I
> suspect they realize the futility of traditional approaches (not that the
> approaches wouldn't work... people just won't use them which amounts to
the
> same thing) and tried to develop new approaches that mirror what
successful
> developers do already.  For example, the "make one to throw away" approach
> is something that most of us do already (or should, I've had _very_ few
> projects where the spec was so perfect and my understanding of the tools
so
> complete that I've gotten it perfect the first time), they just made it
> into a "formal" design phase.
>

Interesting that there is this "acceptance" that the traditional approaches
are futile because people just won't use them. I don't agree with that
philosphy, a similar argument would be "lets decriminalise burglaries
because people persist in doing them" :-) BTW, I appreciate your
qualification on this one, many others would have just said that
"traditional methods" don't work :-). Of course, if you can get people to do
them then they work extremely well - else why would we have so many books
advocating them? (plus I have seen them work extremely well :-)). I remember
my introduction to reviews/inspections - the project manager said do them or
find a new job - gave us great incentive to learn how and do them with
enthusiasm :-).

So, Rony, assuming we haven't bored you to death by now - what are managers
going to do about the above situation? Cliff and I (seem) to be in agreement
that traditional methods would work if they were accepted by the
programmers - so how do you plan to accomplish this? :-)

So we get back to the basic fact that our industry is undisciplined and
hardly "professional" :-) Broad sweeping statements, I know and there are
obviously many exceptions, but as a generic, general statement it seems to
be true.

Thanks for the dialog Cliff,

Peter

> --
> Cliff Wells, Software Engineer
> Logiplex Corporation (www.logiplex.net)
> (503) 978-6726 x308  (800) 735-0555 x308
>
> "Then with your new power you'll accomplish all sorts of cool stuff
>  in no time, and We'll All Be Sorry.  At that point you can either
>  gloat a bit, and then relent, or go ahead and send the robot army
>  after us." - Quinn Dunkan
>




From llchen223 at hotmail.com  Sat Feb  9 23:47:15 2002
From: llchen223 at hotmail.com (Lei Chen)
Date: 9 Feb 2002 20:47:15 -0800
Subject: signal through popen2
Message-ID: <de6c123c.0202092047.424c8091@posting.google.com>

Hi,

I would like to send a CTRL-C on a process opened by popen2.  Is this possible?


From carifio.nospam at nospam.usys.com  Wed Feb  6 14:09:53 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Wed, 06 Feb 2002 19:09:53 GMT
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <5cf88.18880$zJ3.1464485@typhoon.maine.rr.com>

>  ... The Python steering committee rightly
> worries themselves with the idea that macros can seriously
> worsen the maintainability and comprehensibility of code when
> misused; further Guido has historically been adamantly opposed
> to macros in Python. I am not addressing any of those political
> issues at this moment in time.

Can someone fill in the newbies on this topic? Not all macro systems
are just "blind text substitution". Scheme and LISP both have reasonable
macro facilities which, when used judiciously, can actually enhance the
comprehensibility of code (in my opinion, of course). So, while
understanding
that a "white-space sensitive" language introduces some challenges, what's
at issue here? I'm missing the context...




From benjamin.sauthier at worldonline.fr  Thu Feb 28 12:55:38 2002
From: benjamin.sauthier at worldonline.fr (Benjamin SAUTHIER)
Date: Thu, 28 Feb 2002 18:55:38 +0100
Subject: Getting running subprocess CPU Use
Message-ID: <3c7e6e12$0$14915$626a54ce@news.free.fr>

I launch a child process using Popen4('command'). This is on UNIX SGI
environment

How can I get the current  USER CPU Use of the subprocess (like what is
exactly done with the command 'top' or 'gr_top'). Is there any python
builtin command? How is it done in 'top'?

os.times() and resource.getrusage() don't work (they wait for the child to
complete).

No way to query the kernel?


Benjamin





From mikael.lexen at stop.spam.volvo.com  Fri Feb 15 05:07:04 2002
From: mikael.lexen at stop.spam.volvo.com (Mikael =?iso-8859-1?Q?Lex=E9n?=)
Date: Fri, 15 Feb 2002 11:07:04 +0100
Subject: VB code -> Python
Message-ID: <3C6CDDC8.4ED3B329@stop.spam.volvo.com>

Hi

I have the following VB code snippet:

    Set oContext = GetObjectContext
    Err.Clear
    Set oUser = GetObject("WinNT://" & CStr(Domain) & "/" &
CStr(UserID))
    If Err = 0 Then
        For Each oGroup In oUser.Groups
            If UCase(Groupname) = UCase(oGroup.Name) Then
                MemberofGroup = True
                'Exit Function
                Exit For
            End If
        Next
        oContext.SetComplete
    Else
        oContext.SetAbort
        MemberofGroup = False
    End If

So my questin is how do I translate the GetObject(...) thing in to
Python code. I have looked in the win32 ref doc but I can't figure out
out how to do it and a search in Google didn't help me much either. So
any help is appreciated.

--
/Mikael




From mhammond at skippinet.com.au  Fri Feb 15 08:08:38 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 15 Feb 2002 13:08:38 GMT
Subject: PythonCOM arbritrary vtable based interfaces
References: <9d4c7023.0202150041.34dc5e6f@posting.google.com>
Message-ID: <3C6D08C1.2000307@skippinet.com.au>

Sanjay wrote:
> I saw a thread last september stating that this limitation may soon be
> lifted.  Does the latest release of PythonCOM support this feature? 

It supports implementing such interfaces, but not calling them.

> If not, are there any workarounds?  I also saw the suggestion that
> creating your own pyd to wrap the custom interface would be a good
> alternative.  Is this still the best way to go?

For calling interfaces, using SWIG is still the best option - though I 
would much prefer to bite the bullet and see the universal gateway stuff 
extended so that we can call those interfaces. :-)

Mark.



From James_Althoff at i2.com  Fri Feb  8 13:58:50 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 8 Feb 2002 10:58:50 -0800
Subject: 2001 Turing Award
Message-ID: <OFFD21B9C4.A2D3403B-ON88256B5A.006716A9@i2.com>

[James_Althoff]
>
>
http://arc.cs.odu.edu:8080/dp9/getrecord/oai_rfc1807/0018.mit.theses/1963-10

>
> Happy reading,
>

[joost_jacob]
>That link has a bib. record but no thesis, the thesis itself is not
>available online?

It is available online as scanned pages from the thesis.  If you click on
the first link in the "Available Service Providers" box the resultant page
shows a link to the thesis, namely:

http://theses.mit.edu/Dienst/UI/2.0/Describe/0018.mit.theses/1963-10

Click the link "Overview of thumbnail pages" and start reading!

Jim




From gtalvola at nameconnector.com  Wed Feb 20 11:15:12 2002
From: gtalvola at nameconnector.com (Geoffrey Talvola)
Date: Wed, 20 Feb 2002 11:15:12 -0500
Subject: HTTP state management without cookies?
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E41240D0@MAILBOX.nameconnector.com>

Steve Holden wrote:
> ... That's why a recent
> thread concluded you EITHER pass session id through all URLs 
> OR use a cookie
> to identify the session.

Webware's WebKit app server has a different approach that doesn't require
cookies and also doesn't require adding the session ID to all URL's.  It has
an option to use "automatic path sessions".  If you turn on this option,
then WebKit will redirect a request without a session ID like:

http://foo.bar/cgi-bin/WebKit.cgi/MyServlet

to:

http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet

Since the session ID comes _before_ the name of the servlet, all relative
links work just fine.  WebKit knows how to parse the URL and extract the
session ID.

What makes this work is that WebKit passes all requests through a single CGI
script (or uses the special mod_webkit Apache module) and uses its own logic
to determine how to map the URL to a servlet file.  It would be hard to make
this work with plain CGI.  Perhaps you could use some mod_rewrite magic to
transform:

http://foo.bar/_SID_=2374820394/myscript.cgi/

into:

http://foo.bar/myscript.cgi?_SID_=2374820394

so the same trick will work for CGI?

- Geoff




From cmaloney at physics.ucsb.edu  Tue Feb 19 11:57:08 2002
From: cmaloney at physics.ucsb.edu (Craig Maloney)
Date: Tue, 19 Feb 2002 08:57:08 -0800
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
Message-ID: <3C7283E4.4050604@physics.ucsb.edu>

Hi all.

I'm doing some scientific computing and have
amassed a moderate set of classes.  Due to the
heavy computational demands, I don't think I
can justify rewriting everything in an interperated
language; nice as that would be now that I'm
playing around with what I've created.

I'm in a situation now where I've written about
a dozen of quickie, throw away, utility routines
that I call from the command line, and in essence,
BASH has become my scripting environment.

Unfortunately, the number of throw-away routines
is exploding, and I feel like I'm loosing the advantage
of having constructed a nice set of classes with
which to play.

I would, of course, like to have some sort of scripted
environment (e.g. Python) in which to "play"
with my objects.

That said, I'd like to solicit opinions as to the relative
merits of the various possible approaches.

I'm making essential use of templates and polymorphism.

I have no experience with Boost.Python, but from what I
understand, it would not involve writing wrapper definitions
(ala SWIG or IDL), right?  Are there other options?


Or should I just wait for .NET/mono to come out? ;)



Cheers,
Craig



From philh at comuno.freeserve.co.uk  Thu Feb  7 18:48:43 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:48:43 +0000
Subject: [Slightly-OT] Wasn't Google supposed to be "avid Python-Users" ?
References: <3c6164f0.50243062@news.t-online.de> <a3sk2i$67m$1@panix2.panix.com> <e840346c.0202070321.452d405d@posting.google.com> <7xr8nxy0rb.fsf@ruckus.brouhaha.com>
Message-ID: <slrna664ir.d87.philh@comuno.freeserve.co.uk>

On 07 Feb 2002 03:44:40 -0800, Paul Rubin <phr-n2002a at nightsong.com> wrote:
>kalath at lycos.com (Mark Brady) writes:
>> As python lover trapped developing in Java I see this a lot. Sun's
>> marketing has convinced a lot middle/upper management that Java/J2EE
>> is the only viable option to C/C++ for large server side systems.
>> Actually the endorsement and support from IBM/Oracle/HP/etc. for Java
>> is what seems to have given Java the 'respectability' it currently
>> enjoys as a C/C++ alternative. On a personal note Java/J2EE while
>> keeping me employed is driving me insane. Thank god for Pygame
>> (www.pygame.org). Java feeds the body but Python feeds the soul!
>
>As much as I like Python, I don't see it working so well in the kind
>of large, real-world projects that Java works pretty well in:
>
>  - big codebase (>200K LOC) being developed/maintained by lots (say
>    20 or more) of people

But 200 KLOC of Java is probably only about 80 KLOC of Python.

>  - Program stays around for years but developers come and go.
>    Average turnover is maybe 1 programmer per month.  So there
>    are always people hacking on the codebase who aren't that familiar
>    with it.

The resultant architecture will be a Big Ball Of Mud.



-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From kdahlhaus at yahoo.com  Wed Feb  6 13:02:01 2002
From: kdahlhaus at yahoo.com (Kevin Dahlhausen)
Date: 6 Feb 2002 10:02:01 -0800
Subject: Web Hosting - one report
References: <283adf56.0202060503.6556b483@posting.google.com> <7x7kpqoip9.fsf@ruckus.brouhaha.com>
Message-ID: <283adf56.0202061002.352ad3c8@posting.google.com>

> If you can compile your own python, can't you statically link your
> extension into it?

That's a good point that I haven't tried.


From phr-n2002a at nightsong.com  Mon Feb 11 17:42:23 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 11 Feb 2002 14:42:23 -0800
Subject: Question on if, while, for....  Figured it out..
References: <e959e7b.0202110901.485da0b0@posting.google.com>
Message-ID: <7xg047prn4.fsf@ruckus.brouhaha.com>

c at tech.usd253.org (Fred) writes:
> if num > den:		#Test for improper fraction.
>     i = num/den		#This will only return the whole part. 
>     j = num%den		#This will return only the remainder.
> 
>     if j != 0:
>         for i in range(9,0,-1):
>             if num%i == 0 and den%i ==0 and test != 1:
>                 a = j/i
>                 b = den/i
>                 print "You entered an improper fraction, simplified it
> is: %d %d/%d" % (i,a,b),
>                 test = 1
>     else:...

Does this do what you really want, if the fraction entered is 8/12?


From tjreedy at home.com  Tue Feb 19 12:22:36 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 19 Feb 2002 17:22:36 GMT
Subject: newbie: Some file object tips?
References: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4> <mailman.1014120512.7783.python-list@python.org> <Xns91BAF4112B7BEAusYourStandingInIt@130.133.1.4>
Message-ID: <wRvc8.54299$X64.19388131@news1.rdc2.pa.home.com>

> So does this command ....
>
> > lines = open("log.txt").readlines()
>
> ... actually load the whole log into RAM.

Yes





From gcordova at hebmex.com  Fri Feb 15 15:01:12 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 14:01:12 -0600
Subject: MemoryError using NumPy
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A8D@SRM201002>

> > I suspect that memory has fragmented such that a sufficiently
> > large contiguous block is not available to set up the array.
> > Is this the likely cause, and if so is there a way to un-fragment 
> > memory? Or might there be an alternative to Numeric.take that
> > gets to the same place?

Er... This isn't an Amiga, you know. :-)

Really. Under UNIX, or anything "Unixy", where there's
a memory management unit on the CPU and a virtual memory
subsystem in the OS, memory fragmentation really doesn't
mean much, because the OS remaps the memory layout
so that each and every process receives (virtually) the
same memory space, in a contiguous manner.

I remember having to reboot my Ami because I didn't have
enought ram left, when in fact it was all fragmented.
Virtual memory would have been good then.  :-)

(regardless, they rocked)

> I think you have two functions in one line that create a copy (take()
> and astype()). I have never gotten an answer as to when CPython frees
> the memory for temporary variables in one statement, but if you can
> delete myDataArray before calling astype(), you might save something
> there.

The problem is the C library, seems like it doesn't ever return
allocated-and-then-freed memory to the OS, so, CPython never
returns unused memory to the OS neither.

> 
> If not, I'm interested in finding oput what the options are for memory
> defragmenting...
> 

No need, don't you worry about this.

>
> I'd also be interested in what happens under Linux, I still have a much
> better feeling about the robustness of Linux over Win2k, I'd like to see
> if they are different in this respect.
>

Different how?

WinNT/2000 has virtual memory management also, so they do the
remapping dance also. They're pretty robust, both of them.

BUT, if you wanna, you can rip out Linux's VM system and
install your own if you feel like it, but you can't do that
on windows. :-/

>
> -Chris
> 

TGIF!!!

-gustavo



From pedronis at bluewin.ch  Tue Feb  5 11:58:43 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Tue, 5 Feb 2002 17:58:43 +0100
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com>
Message-ID: <3c6011e3$1_5@news.bluewin.ch>

Hi.

I have toyed with similar ideas, but never
moved to implementation. Sadly I don't know
if I have time left to invest on this :(.

Anyway here is a list of
open issues to consider, probably
you also know them all but for the
moment I cannot say much more :(
[Sorry if some sounds peremptory,
is just hurry]

1. Python does not have
block closures with non-local returns
like Smalltalk, they would be
probably a more natural extension
than macros but probably
not enough pythonic anyway.
They are probably a useful missing tool
for implemeting macros and
one should cope with that.

2. Python has a very late binding semantics
and compilation is fully  trasparent - everything:
definitions, retrivieving values from other
modules etc happen at runtime. So the
most natural macros would operate
as if expanded at runtime, given 1.
this seems difficult to implement.
Conclusion: macros implementation
probably must break compilation transparence,
and require inter modules references
at compilation time.  Some model ideas:

[assuming the implementation
uses import hooks]
a) having separated macro-defs-only
  files
b) a macro definition (at its compilation-time) is transformed
in code that at runtime produces some kind
of first-class macro object, probably normal imports
and lookup for macro defs should be
separated (?), basically the macro handling mechanism
uses at compilation time the information in the macro
objects in already imported (in classic sense)
modules ( or force import).

3. Given the absence of closure blocks
local vars clashes should be handled through
renaming (?) (issues?)

4. A problem is how to refer efficiently
to definition context values

* macro def  "module" *

def mytest(): ...

macro def of myguard: BLOCK as
if mytest(): BLOCK


*  macro usage "module" *

macro from ... import myguard

myguard:
  ...

how to refer efficiently and
"safely" to mytest in macro expansion?
[ differently from CL in Python symbols
are not first-class :(, the problem is
that python offer only the aforementioned
very late binding]

5. error handling and reporting

regards, Samuele Pedroni.




From phd at phd.pp.ru  Fri Feb  1 04:55:46 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Fri, 1 Feb 2002 12:55:46 +0300
Subject: os.fork()
In-Reply-To: <cf0ad9fb.0202010141.7d9425bd@posting.google.com>; from jolsen@mailme.dk on Fri, Feb 01, 2002 at 01:41:23AM -0800
References: <cf0ad9fb.0202010141.7d9425bd@posting.google.com>
Message-ID: <20020201125546.I10316@phd.pp.ru>

On Fri, Feb 01, 2002 at 01:41:23AM -0800, Jesper Olsen wrote:
> os.fork() is an alternative to thread's
> 
> Is there a way for the parent to find out when the child 
> has terminated?

   os.wait* functions. There are blocking and non-blocking variants.

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



From chris.gonnerman at newcenturycomputers.net  Tue Feb 12 00:37:56 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 23:37:56 -0600
Subject: Possible dumb query
References: <000801c1b385$97747aa0$0100000a@ahimsa>
Message-ID: <000a01c1b387$6f19c040$0101010a@local>

----- Original Message -----
From: "ahimsa consulting" <ahimsa at inext.co.za>


> Greeting Pythonites
>
> This might qualify as one of those "really dumb questions", but:
>
> Under Linux RH 7.2 when doing a search for which python ("$ which python")
I
> have discovered that I have one entry for python as /usr/bin/python and
one
> entry for python2.2 as /usr/bin/python2.2. My question is this: if I were
to
> delete the directory /usr/bin/python what impact - if any - will it have
on
> any of the libraries/ etc. that I would require to use python2.2

/usr/bin/python should not be a directory, it should be an executable
program
or a symbolic link.

Do this:

    $ python

and look at the version number.  If it's 1.5.2 then you have two versions,
but if it's 2.2 then that's the only version installed.

Likely you have both, as RedHat uses Python in their admin toolset, and
usually install 1.5.2 as the "approved" version.  So, deleting 1.5.2 might
be a horrible mistake.

> Once again, my apologies if this is a dumb query, but gotta learn somehow.

No problem here.





From bokr at oz.net  Tue Feb 19 20:45:46 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 01:45:46 GMT
Subject: Adding static typing to Python
References: <mailman.1014138876.23048.python-list@python.org>
Message-ID: <a4uv4a$t7b$0@216.39.172.122>

On Tue, 19 Feb 2002 11:08:05 -0600, "Jason Orendorff" <jason at jorendorff.com> wrote:

>Courageous wrote:
>> Speaking of typedef and the C/C++ languages, I've long thought
>> that in a statically-typed language, the types should never be
>> aliased, and should always be distinct.
>
>I agree, for your example; but a common use of typedefs now is
>to create an alias for a type whose name is just too long to be
>allowed.
>
>  typedef list<pair<string, bool> >::iterator ItemIter;
>
I've wondered why they don't just automatically hash those monster unwieldy
names with md5 and convert the 128-bit digest to 24 characters all alpha as
an automatic alias. The originals could be kept in a special dictionary keyed by
the latter for whatever need. Or is that already being done in newer versions?

Regards,
Bengt Richter


From cgp at le.ac.uk  Fri Feb 22 07:06:42 2002
From: cgp at le.ac.uk (Clive Page)
Date: 22 Feb 2002 04:06:42 -0800
Subject: Python Gotcha with Octal Numbers
References: <a4dt3d$6us$1@sirius.dur.ac.uk> <Xns91B6DFD6A36CDRASXnewsDFE1@130.133.1.4>
Message-ID: <d8cbfc61.0202220406.271bcb18@posting.google.com>

Philip Swartzleonard <starx at pacbell.net> wrote in message news:<Xns91B6DFD6A36CDRASXnewsDFE1 at 130.133.1.4>...
> 
> Steve has mentioned that UNIX people are used to it... i think it's a 
> little deeper than that, b'cause as far as i know, _every_ programming 
> language that has ever supported octal does it this way...

Not quite all of them.  Fortran (at least since Fortran90) allows the
forms O"01234" or O'1234' i.e. an upper-case O and a quoted string of
digits (but they are only allowed in data statements, unfortunately).

Prior to Fortran90 there was no _standard_ way of expressing octal
constants, but lots of vendors had their own extensions.  But I don't
recall ever coming across anything as error-prone as the C notation
with a leading zero.  Fortran is a much safer language than C; I
recommend it as the nearest thing to a compiled form of Python.  :-)


From SBrunning at trisystems.co.uk  Tue Feb 26 11:31:27 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 26 Feb 2002 16:31:27 -0000
Subject: Precompiled regular expressions slower?
Message-ID: <31575A892FF6D1118F5800600846864DCBCD08@intrepid>

> From:	Peter Bienstman [SMTP:pbienst at mit.edu]
> I've a Python script a want to optimise. It basically loops over all the 
> lines in a file and does some regular expression matching:
> 
> line = inputfile.readline()
> while line:
>   s = search(r".....", line)
>   ...
>   line = inputfile.readline()
 
You might find something like this faster:

for line in inputfile.xreadlines():
    s = search(r".....", line)
    ...
 
> There are several reg ex in the loop. Since the Python version runs 2 to 3
> 
> times as slow as its Perl counterpart, I thought I'd try speeding up the 
> script by precompiling the reg ex:
> 
> re1 = compile(r"....") 
> ...
> line = inputfile.readline()
> while line:
>   s = search(re1, line)
>   ...
>   line = inputfile.readline()
> 
> 
> Strangely enough this new version seems to run about 30% *slower* than the
> 
> old version.
 
Try putting the whole lot in a function - global access is slower than local
access.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From gbreed at cix.compulink.co.uk  Wed Feb 20 10:40:07 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Wed, 20 Feb 2002 15:40:07 +0000 (UTC)
Subject: Adding static typing to Python
References: <mailman.1014147452.26214.python-list@python.org>
Message-ID: <a50g0n$gt4$1@thorium.cix.co.uk>

Justin Sheehy wrote:

> Python has strong typing.
> 
> It also has dynamic typing, as opposed to static typing.  This seems
> to be the thing that you're objecting to.  
> 
> Sorry to nitpick, but if you're going to talk about these sorts of
> details, getting the terminology right is a good start.

Do you have a reference for this terminology?  I haven't seen it outside 
of comp.lang.python.  The only definition I have to hand contradicts it:

<http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=strong+typing>

"Strict enforcement of type rules with no exceptions. All types are known 
at compile time, i.e. are statically bound. With variables that can store 
values of more than one type, incorrect type usage can be detected at run 
time."

Taking "static typing" literally would probably mean that an object 
couldn't change it's type after creation.  That's almost always the case 
in Python.  If you mean identifiers have to stay bound to the same type, 
it'd mean

x = 1
x *= 5L

would be illegal, because x begins as an integer and then becomes a long.  
So long as function arguments remain generic, I don't think I'd miss that 
kind of dynamic typing either.  All you have to do is rewrite it

x = 1L
x *= 5L

or forget about the int/long distinction altogether.  The arguments for 
"static typing" seem to be for explicit declarations, like

long x = 1L
x *= 5L

which is redundant -- of course x is a long.  If you want to catch 
mis-spellings, all you need is

var x = 1
x = x * 5L

and I wish you luck of it.  The proposal for adding "optional static 
typing" to Python is about typing of function arguments.  Which makes some 
sense, but if a compiler's smart enough to infer the types and point out 
where they disagree, shouldn't it also be smart enough to recognize 
patterns like

def func(arg1, arg2):
  assert isinstance(arg1, type1)
  assert isinstance(arg2, type2)

?

For optimization of libraries, you can assume the types supplied will be 
those that get supplied in your test suite.  If different types get 
supplied on deployment, default to less efficient code.

So, um, what actually is the problem?


                    Graham

         (http://www.microtonal.co.uk/)


From andrew-pythonlist at puzzling.org  Thu Feb 28 22:48:57 2002
From: andrew-pythonlist at puzzling.org (Andrew Bennetts)
Date: Fri, 1 Mar 2002 14:48:57 +1100
Subject: Wierd traceback from threading module
Message-ID: <20020301034857.GA14419@ritsuko.xware.cx>

Here is the entire traceback:
"""
Unhandled exception in thread:
Traceback (most recent call last):
  File "D:\Python22\lib\threading.py", line 423, in __bootstrap
    self.__stop()
  File "D:\Python22\lib\threading.py", line 429, in __stop
    self.__block.notifyAll()
  File "D:\Python22\lib\threading.py", line 242, in notifyAll
    self.notify(len(self.__waiters))
  File "D:\Python22\lib\threading.py", line 224, in notify
    me = currentThread()
TypeError: 'NoneType' object is not callable
"""

I'm not doing *anything* with the threading module except starting a daemonic
thread.  This traceback occurred as the program was shutting down -- the main
thread would probably have already finished.

Is this a bug or am I doing something silly?

Please CC replies to me, as I'm not currently subscribed to the list/newsgroup.

-Andrew.




From aahz at panix.com  Sat Feb  9 11:29:43 2002
From: aahz at panix.com (Aahz Maruch)
Date: 9 Feb 2002 08:29:43 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org> <B88AE186.48CD%jonathan@onegoodidea.com>
Message-ID: <a43ipn$s37$1@panix2.panix.com>

In article <B88AE186.48CD%jonathan at onegoodidea.com>,
Jonathan Hogg  <jonathan at onegoodidea.com> wrote:
>On 9/2/2002 0:56, in article
>mailman.1013216835.14351.python-list at python.org, "Cliff Wells"
><logiplexsoftware at earthlink.net> wrote:
>> Aahz Maruch wrote:
>>> 
>>> I have to admit that I'm seriously thinking about an iMac, though.
>> 
>> Yikes.  That's an expensive Linux box ;)
>
>Hah! all smileys aside, you'd have to be mad to run Linux on it when it
>comes with Mac OS X and all of Apple's free iApps. I get to run commercial
>applications including the Adobe and Microsoft range, it's got full Java
>support, runs all the UNIX apps I need including Apache, Python and
>PostgreSQL, and is fun to use.
>
>I'd whole-heartedly recommend getting an iBook with Airport though unless
>you need the extra speed of the new iMac or the DVD burner. It's cheaper,
>has the same size screen, works on the sofa and my local coffee shop, gets 4
>hours of battery life, and sleeps and wakes in an instant.

My Apple-pushing partner told me that MacAddict says that G3 isn't
sufficient for running OS X.  I guess you feel differently?  The other
thing is that Apple's notebooks have traditionally been no more reliable
than PC notebooks, but the first generation iBooks were built like
tanks.  How true is it of the current generation?

Also, the screen sizes aren't exactly the same, but I'd say that 14.1"
is probably better for 1024x768 than 15" (I'd want 1152x864 for the
latter).  Looking at the Apple web site, actually, I think I'd just get
the 12" screen and use one of my existing external monitors at home.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From mdk at mdk.com  Thu Feb 14 16:12:51 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 16:12:51 -0500
Subject: Mutex a file?
Message-ID: <a4h98k$ao2f$1@ID-98166.news.dfncis.de>

I have a file that is to be read to and written from by many Python scripts
running at the same time.

Is there a way to put a Mutex on it so that the scripts wait their turn
before tyring to open it?

Thanks




From kjones at docstrategies.com  Thu Feb 28 11:27:36 2002
From: kjones at docstrategies.com (Keith Jones)
Date: Thu, 28 Feb 2002 11:27:36 -0500
Subject: problems with photoimage and tkinter
References: <Y%hf8.149154$dh.38770971@typhoon.nyroc.rr.com> <a5ks4b$61a$1@norfair.nerim.net>
Message-ID: <3c7e5639$1@buckaroo.cs.rit.edu>

Eric,

Thanks for the info; I'm glad I'm not just losing my mind. Btw, I'm Keith, I
was using my brother's account to access the newsgroup last night. Again,
thanks. I can't wait to get home to try to get this sucker to work. :)



Keith






From chad.maine at broadlink.com  Tue Feb 26 18:16:00 2002
From: chad.maine at broadlink.com (Chad Maine)
Date: Tue, 26 Feb 2002 23:16:00 GMT
Subject: Class functions
References: <mailman.1014764563.7542.python-list@python.org>
Message-ID: <QGUe8.213$en5.1614@typhoon.sonic.net>

Yeah, my function runs a loop looking for updates and posting them to a
StringIO file object.  I want access to that file object in real time in my
client instead of waiting for the function to "return".


"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1014764563.7542.python-list at python.org...
> Chat Maine wrote:
> > I'd like to access the data updated in real time via a class' function.
>
> Chad,
>
> Your source code doesn't provide quite enough information for
> us to give you good advice.  (It actually sounds like you want
> Python 2.2's generators, but I'm not sure I believe that.)
>
> Here is one easy way to do this...
>
> # theclass.py
> class MyClass:
>     ...
>     ...
>     def func(self):
>         return UPDATE
>
> # mainprogram.py
> x = theclass.MyClass()
> data = x.func()
> while data != '':
>     print data
>     data = x.func()
>
> I may be missing the point, though...
>
> ## Jason Orendorff    http://www.jorendorff.com/
>




From aleax at aleax.it  Fri Feb  1 17:18:10 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 22:18:10 GMT
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu>
Message-ID: <CuE68.17883$om6.426379@news1.tin.it>

Quinn Dunkan wrote:
        ...
> Of course, vi itself requires loads of memorization, and I'm not convinced
> it's really worth it.  For example, to move a word two words back to the
> end of the line, is it really quicker to type '^[2Bhdt $pa ', or is it
> quicker to select the word with the mouse, cut, click, and paste?

If you have to *think* about it, as "memorization" would suggest, then
you'll probably be happier going with the rodent.  The power of vi, for
me (and most other vi-enthusiasts I've known) is that the *fingers*
learn the patterns -- sort of like in needlepoint; your conscious brain
can focus on higher-abstraction stuff.  Which is no doubt why, while
I've been actually using vim/gvim since almost forever, most of the
basic stuff I still do vi-like... 'cause that's what my fingers learned back
in true-vi times.  Now, if you type and edit fast, having your fingers
stay close to the keyboard's home row has indisputable advantages
wrt having them fly all over the place, to cursor keys or function keys
or even worse separate devices like mice -- total travel distance and
times for the fingers are greatly reduced.  After comparable long fast
sessions with vi (or vim/gvim) versus typical visual editors, my *hands*
can tell the difference... they ache in the latter case, but not in the
former (not an issue of age -- I noticed the same effect back when
I first learned vi, and I _was_ reasonably young then:-).  To be honest
I haven't been able to measure productivity differences either way (not
that I've tried all that seriously!), but having un-aching hands is nice
in itself, and I suspect it reduces distraction and thus should benefit
speed and accuracy of whatever's being written and edited.

People who are unable to let go and let their fingers "drive" seem to me to
keep fighting vi's low-level mechanisms forever -- I've always advised
such people to give vi up in favour of other editors.  Few type and edit
so fast as to make the difference noticeable, after all.


Alex



From aahz at panix.com  Fri Feb  8 10:45:04 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 07:45:04 -0800
Subject: threads killing other threads
References: <1013105970.135322@cswreg.cos.agilent.com> <a40h4g$kl5$1@newsreader.mailgate.org>
Message-ID: <a40rq0$ajl$1@panix2.panix.com>

In article <a40h4g$kl5$1 at newsreader.mailgate.org>,
Les Smithson  <lsmithso at NOhare.SPAM.demon.co.uk> wrote:
>
>I'm sure there's a good reason for Python not implementing this. Anyone
>know what it is?

Yes.  It doesn't work.  

Oh, sure, there's a long technical explanation of how arbitrarily
killing threads leaves processes in various kinds of weird states, but
it all boils down to, "It doesn't work."  That goes double for any kind
of cross-platform facility.  Note that Java has pretty much abandoned any
pretence that it can provide this capability.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From ahmebah at hotmail.com  Thu Feb  7 12:22:21 2002
From: ahmebah at hotmail.com (ameoba)
Date: 7 Feb 2002 17:22:21 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <jb246uci04pj6q2l0aqu18ch17kmv8ujof@4ax.com>
Message-ID: <Xns91AE5FFAFDC37ahmebahhotmailcom@198.99.146.10>

Courageous <jkraska at san.rr.com> wrote in 
news:jb246uci04pj6q2l0aqu18ch17kmv8ujof at 4ax.com:

> 
>> With that in mind, it's not possible for me to convince
>> you of anything beyond your own limited two deminsional perceptions.
> 
> If you ever became delusional, how could you be sure that
> it was or wasn't so?
> 
> C//
> 

Well, I'm sure somebody would tell him he's gone over the edge.


Oh, wait.  We tried that.


I guess he's just retarded or something...


From zak at mysql.com  Tue Feb 26 15:53:37 2002
From: zak at mysql.com (Zak Greant)
Date: 26 Feb 2002 13:53:37 -0700
Subject: Building 2.2 with the Intel icc compiler
In-Reply-To: <9gSe8.18435$ZC3.1526281@newsread2.prod.itd.earthlink.net>
References: <9gSe8.18435$ZC3.1526281@newsread2.prod.itd.earthlink.net>
Message-ID: <1014756817.743.686.camel@linux>

On Tue, 2002-02-26 at 13:31, Ted Drain wrote:
> Has anyone built Python 2.2 with the new Intel C/C++ compiler icc?  I
> configured Python and then edited the Makefile by hand to use icc and icc
> options.  Python builds but the modules all report errors (and then remove
> themselves).
> 
> This compiler generates code that is roughly 30% faster than gcc.  Check
> out:
> http://www.open-mag.com/754088105111.htm
> 
> Any thoughts?

  Hey Ted, Greetings All,

  AFAIK it generates code 30% faster than GCC for mathematical, heavy
  floating point applications. The performance increase in other cases
  is not as great.

  Also, occasional mystery core-dumps have been reported in applications
  heavily optimized for GCC that have been compiled with ICC.

-- 
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /    Zak Greant <zak at mysql.com>   
 / /|_/ / // /\ \/ /_/ / /__   MySQL AB,  Advocate
/_/  /_/\_, /___/\___\_\___/   Calgary, Canada 
       <___/   www.mysql.com   403.244.7213




From skywalkerpackage at hotmail.com  Sun Feb  3 15:08:44 2002
From: skywalkerpackage at hotmail.com (Johannes Eble)
Date: Sun, 03 Feb 2002 20:08:44 GMT
Subject: Tkinter widgets: Write options *and* option tags dynamically from   dictionary?
References: <3c5a84eb.70228013@news.online.de> <3C5A9D81.3E99C36F@westerngeco.com>
Message-ID: <3c5d9879.182577461@news.online.de>

Thanks very much, Martin :)

On Fri, 01 Feb 2002 13:52:01 +0000, Martin Franklin
<martin.franklin at westerngeco.com> wrote:

...

>HTH 

Indeed :)

Johannes



From threeseas at earthlink.net  Sun Feb  3 19:00:53 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:00:53 GMT
Subject: Autocoding evolves from........
References: <4314.793T504T6883966threeseas@earthlink.net> <3c5742e8$0$79557$3c090ad1@news.plethora.net> <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk>
Message-ID: <972.799T919T11694325threeseas@earthlink.net>

On 30-Jan-02 06:52:48 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On 30 Jan 2002 00:48:40 GMT, Peter Seebach <seebs at plethora.net> wrote:

>>Anyway, it looks to me like Python is the wrong language for what you want
>>to do.  C# is almost certainly the best choice for this; it's designed for
>>the kinds of tasks you're looking at,

>Is Tim trying to lock people in to the Microsoft monopoly, then?

Why do you make up things by taking things out of context?

Are you naturally dishonest or do you work at it?

Autocoding will be a nice thing to have, it'll get rid of all the arrogant
jerks.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From gerson.kurz at t-online.de  Sat Feb  2 12:23:18 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Sat, 02 Feb 2002 17:23:18 GMT
Subject: How can I write this C code in Python?
References: <mailman.1012663266.30558.python-list@python.org>
Message-ID: <3c5c204e.11589328@news.t-online.de>

On Sat, 2 Feb 2002 09:16:32 -0600, "Jason Orendorff"
<jason at jorendorff.com> wrote:

>Fred wrote:
>> I think if I could see the below C code done in Python it
>> would get me over the beginners hump.
>
>for i in range(1, 13):
>    for j in range(1, 13):
>        print "%4d" % (i*j),
>    print
>
>## Jason Orendorff    http://www.jorendorff.com/
>
Alternatively, you can always write

for n in range(12):
    exec("abcdefghijkl"[n]+"=lambda x=0,y=0: "+filter(
        lambda x:x not in "\n$\r","""(x*y#x/x!range(x,
y#x+y!b(1,1#d(e~,e~#d(f~,f~#c(e~,e~+d(g~,d(g~,g~))#"%4
d" % a(x,y#map(lambda y:i(x,y),h~#" ".join(j(x)#"\\n".
join(map(k,h~))""".replace("~","()").replace("#",")!")
        ).split("!")[n])
print l()




From phr-n2002a at nightsong.com  Fri Feb 15 01:44:05 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 14 Feb 2002 22:44:05 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net> <mailman.1013718563.21631.python-list@python.org> <5e8bd451.0202142238.580f15c1@posting.google.com>
Message-ID: <7xzo2bnt1m.fsf@ruckus.brouhaha.com>

kp87 at lycos.com (kevin parks) writes:
> How would you sort something based on 2 criteria. Say you have a list
> of lists (i hate tuples). first thing you want is everything sorted by
> the first value and then everything that has the same value for [0]
> then sorted by [1]

# try this:

yourlist =  [['i3', 24.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5], \
             ['i8', 82.25, 0.75, 101, 7, 6, 5, 0, 0, 0, 0, 0.02, 2], \
             ['i9', 9, 0, 255], \
             ['i8', 83.0, 1.0, 101, 1, 1, 5, 0, 0, 0, 0, 0.02, 2]]

def compare_lists(x,y):
  a = cmp(x[0],y[0])
  if a: return a
  return cmp(x[1],y[1])

yourlist.sort(compare_lists)


From max at alcyone.com  Sun Feb 10 18:36:20 2002
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 10 Feb 2002 15:36:20 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <mailman.1013381417.32346.python-list@python.org>
Message-ID: <3C6703F4.8F25BE12@alcyone.com>

Neil Schemenauer wrote:

> Bengt Richter wrote:
>
> > Wouldn't it make sense to return list('abc') ?
> 
> It would also make sense to return ['a', 'b', 'c'].  Since it's not
> obvious what you want Python raises an error.

That's what list('abc') would return.  But in actuality, for it to be
consistent with the rest of string.split's behavior, 'abc'.split('')
would have to return ['', 'a', 'b', 'c', ''], not ['a', 'b', 'c'].

Another beautiful example of why string.split('') should raise an error
instead of trying to guess what you mean.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.


From oma at gncz.cz  Tue Feb 19 09:20:28 2002
From: oma at gncz.cz (Ondrej Martinek)
Date: Tue, 19 Feb 2002 15:20:28 +0100
Subject: module for accessing dBase (.dbf) files?
Message-ID: <3C725F2C.2020406@gncz.cz>

Hi,

anyone knows a python module for accessing dBase .dbf files.  All 
modules I've found so far didn't support index files.

Regards
--ondra



From mal at lemburg.com  Thu Feb 21 04:30:28 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 21 Feb 2002 10:30:28 +0100
Subject: ANN: eGenix.com mxCGIPython Version 0.4.0
References: <3C717A94.2C9ED2C5@lemburg.com> <3C72A3D4.96BE28E1@lemburg.com>
Message-ID: <3C74BE34.6DF41F5B@lemburg.com>

> Update:
> 
> Oleg Broytmann kindly sent me six files for Linux, Solaris and
> FreeBSD, Python 2.1.2 and 2.2:
> 
>         http://www.egenix.com/files/python/mxCGIPython.html
> 
> (The setup seems to work with Python 2.2 as well.)

Sorry, the permissions on the server were not correctly setup.
Downloads of Oleg's versions resulted in a 403 Permission Denied.
This is fixed now.

Any other volunteers for more platforms ? (e.g. Cobalt cubes
and other hosting machines might be interesting to have
as supported platform)

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From tim.one at comcast.net  Thu Feb  7 21:51:30 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 07 Feb 2002 21:51:30 -0500
Subject: Please stop responding to Rue
In-Reply-To: <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEGFNKAA.tim.one@comcast.net>

[Donn Cave]
> So is sci.math choked with posts from people who normally seem
> to have their heads screwed on, endlessly replying this guy as
> if they sincerely think there's some purpose in it?  People who
> take time several times every day to post about how useless his
> ideas are and how they have no interest in helping?  Or are
> they saner there than we are?

On whole, no; neither in sci.physics, or even sci.logic (in fact, if you
think about it, sci.logic attracts more than its share for a reason, and on
both sides <wink>).

indulging-an-urge-to-cruelty-is-anti-pythonic-ly y'rs  - tim




From phr-n2002a at nightsong.com  Thu Feb 21 00:18:05 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 20 Feb 2002 21:18:05 -0800
Subject: strings and sort()
References: <e9e6dd1f.0202201735.2eaa4e9f@posting.google.com> <3C747268.6C756B9D@earthlink.net> <7xvgcrfosh.fsf@ruckus.brouhaha.com> <3C747B81.367B5D50@earthlink.net>
Message-ID: <7xofijct0y.fsf@ruckus.brouhaha.com>

Hans Nowak <wurmy at earthlink.net> writes:
> > > >>> def sort2(lst):
> > >       z = lst[:]
> > >       z.sort()
> > >       return z
> > > ...
> > Just write it like this:
> > 
> > def sort3(lst):
> >   lst.sort()
> >   return lst
> > 
> > The list comprehension already makes a temporary list.  You don't need
> > to copy it around again.
> 
> True... in this case. He may not always use it with a
> listcomp though, and then sort3 will sort the list in
> place and return it.

That's what sort2 is for ;-).  It's reasonable to provide both versions.
Actually, sort2 could then be written:

   def sort2(lst):
      return sort3(lst[:])


From SBrunning at trisystems.co.uk  Tue Feb 12 11:15:19 2002
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 12 Feb 2002 16:15:19 -0000
Subject: Python IDEs
Message-ID: <31575A892FF6D1118F5800600846864DCBCC31@intrepid>

> From:	Henrik Motakef [SMTP:henrik.motakef at web.de]
> BTW: Anybody tried Eclipse yet? Is there Python support already?
 
Yes and no, respectively.

It's a pretty nice IDE for Java development. The refactoring tools are
particularly nice. It does have a tendency to grind to a halt every now and
again, though. I suspect GC.

I can't see any reason who you couldn't have a Python plug in - there is a
C/C++ one already. A lot of work, though, I suspect.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From aahz at panix.com  Tue Feb 12 16:16:39 2002
From: aahz at panix.com (Aahz Maruch)
Date: 12 Feb 2002 13:16:39 -0800
Subject: Defending the Python lanuage...
References: <3C62EE4A.5090305@destiny.com> <mailman.1013420657.25827.python-list@python.org> <a49g99$bt4$1@panix2.panix.com> <mailman.1013540788.23598.python-list@python.org>
Message-ID: <a4c0nn$1lk$1@panix2.panix.com>

[Laura, please make sure to include attributions for quotes (quoting the
.sig is not sufficient.]

In article <mailman.1013540788.23598.python-list at python.org>,
Laura Creighton  <lac at strakt.com> wrote:
>Aahz:
>>Laura:
>>>
>>>Have you read 'Engineering and The Minds Eye'?  I just finished it.
>>>Wonderful.  
>> 
>> How about _Things That Make Us Smart_?
>
>Do you mean Donald Norman's book?  I haven't.  I have read 'The Design
>of Everyday Things' though.

_Things That Make Us Smart_ is more important for programmers, IMO.  You
can see my comments in RISKS at
http://catless.ncl.ac.uk/Risks/18.45.html#subj4
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"Twenty million slave girls can't be wrong."  --SJM


From deltapigz at telocity.com  Thu Feb 21 22:19:32 2002
From: deltapigz at telocity.com (Adonis Vargas)
Date: Thu, 21 Feb 2002 22:19:32 -0500
Subject: Embedding Python in C#
Message-ID: <3c75b88c$1_5@news1>

has this been done? is it even possible?
if it has where can i obtain information on doing so?

thanks in advance,

Adonis




From s.keim at laposte.net  Fri Feb  8 08:27:52 2002
From: s.keim at laposte.net (sebastien)
Date: 8 Feb 2002 05:27:52 -0800
Subject: encapsulation, typing & other questions
References: <mailman.1013107933.30563.python-list@python.org> <3C62FD4D.2B908BA9@daylight.com>
Message-ID: <7da8d8f8.0202080527.30c2e92f@posting.google.com>

Ragu Bharadwaj <ragu at daylight.com> wrote in message news:<3C62FD4D.2B908BA9 at daylight.com>...
> brueckd at tbye.com wrote 
(about static type checking)
> > I don't know of any plans to do this, and the benefits are probably less
> > than you'd imagine. This issue tends to come up as a theoretical problem
> > much more often (if ever) than it does in practice.
> 
> Beg to disagree completely. I've seen too many folks encounter problems
> to believe this.
> -Ragu

If this is your feeling. I don't think Python will be a good language
for you.
My opinion is that Python paradigm is heavily depending on dynamic
type checking.
If you try to constrain Python to static type checking you will lose
almost every benefits of the language.
So you'd probably better to look for another language that feet better
with your requirements.

But are you really sure you need static type checking?
Take a look at pyunit and you will discover that it's really easy to
write unitests in Python.


From philh at comuno.freeserve.co.uk  Thu Feb  7 18:35:23 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Feb 2002 23:35:23 +0000
Subject: 2001 Turing Award
References: <mailman.1013109679.5469.python-list@python.org> <2jnu3a.4sq.ln@wombat.san-francisco.ca.us>
Message-ID: <slrna663pr.d87.philh@comuno.freeserve.co.uk>

On Thu, 7 Feb 2002 12:20:50 -0800, Keith Keller <kkeller at speakeasy.net> wrote:
>In article <mailman.1013109679.5469.python-list at python.org>,
>	James_Althoff at i2.com writes:
>
>> ----- Message from Gary Chapman <gary.chapman at mail.utexas.edu> on Thu, 7
>> Feb 2002 08:41:00 -0600 -----
>> 
>> I'm happy to pass on the news that the 2001 Turing Award, the highest
>> award in computer science
>
>Damn--I thought the award was for the person who best impersonated
>a Turing machine.  :)

But only using 9 commands.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From yoda at dagobah.org  Thu Feb 21 20:14:21 2002
From: yoda at dagobah.org (Jedi Master Yoda)
Date: 22 Feb 2002 01:14:21 GMT
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014248633.16973.python-list@python.org> <7xit8rhebq.fsf@ruckus.brouhaha.com>
Message-ID: <slrna79vo7.e5o.yoda@homer.interx.com>

On 20 Feb 2002 16:24:25 -0800, Paul Rubin <phr-n2002a at nightsong.com> spouted:
> Maybe there's a principle to be found here: "design by FAQ".  If newbies
> keep arriving and asking over and over again "how do I do X?", maybe
> there's a good reason for having a way to do X.  We've seen that already
> for augmented assignments.

That's how Perl got to be Perl.

"Perl is worse than Python, because people wanted it worse." - Larry Wall.

--
"Opportunity is missed by most people because it is dressed
in overalls and looks like work."            - Thomas Edison


From dyoo at hkn.eecs.berkeley.edu  Sun Feb 24 22:30:44 2002
From: dyoo at hkn.eecs.berkeley.edu (Daniel Yoo)
Date: Mon, 25 Feb 2002 03:30:44 +0000 (UTC)
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com>
Message-ID: <a5cb54$27c2$1@agate.berkeley.edu>

Donnal Walter <donnal at donnal.net> wrote:
: I know that

: map(f,items)

: is equivalent to:

: for x in items:
:     f(x)


map() collects the results of calling f() over all the x in items, so a
closer translation would be:

###
results = []
for x in items:
    results.append(f(x))
###



: But what is the functional equvalent of:

: for x in items:
:     x.f()



Here's one way to do it:

###
map(lambda x: x.f(), items)
###


We need to feed map() a function that does the action 'x.f()'.  If we
don't want to use lambda, we can do something like this:

###
def call_function_f_from_x(x):
    return x.f()

results = map(call_function_f_from_x, items)
###


If you have more questions, please feel free to ask.  Good luck!


From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 10:10:57 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 09:10:57 -0600
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <pan.2002.02.10.15.23.55.53798.1174@club-internet.fr> <3C668B73.9E6F5BF8@earthlink.net>
Message-ID: <006801c1b245$25cd8cc0$0101010a@local>

----- Original Message ----- 
From: "Ron Stephens" <rdsteph at earthlink.net>


> Hmmm, thanks for your comments. I just noticed that in my syslogs 
> there was an attempt to open /etc/news/inn.conf that failed. So I 
> just went in and chown 'ed that file to be owned by me rather than 
> root.

That's not the right solution; /etc/news/inn.conf should not be 
writable by "world".  You need to find out why/what tried to open it.

> I don't know if that is part of the problem. But in general, I 
> am finding that almost nothing I try to do with respect to running 
> Python scripts using crontab works at first, simply because of all 
> the permission stuff. So I change ownership and rights a lot. 

Bad sign.  Your crontab runs with your user credentials, so anything
you can do, it can do.  The problem here is usually environment
variables which are set differently.

> But I didn't even know that /etc/news/inn.conf was needed, and maybe 
> it isn't!

Shouldn't be, if you aren't calling inn as an external program.

> I wonder what else is needed but not owned or allowed by my 
> user? How does one ever get any real work done when unknown 
> permission and file ownership problems lurk under every hidden log 
> ;-))))

In general anything that works at your command line can be made to
work in your crontab.  Implement the logging as described in my 
previous post and see what happens.






From dale at riverhall.NOTHANKS.co.uk  Sun Feb 17 07:18:47 2002
From: dale at riverhall.NOTHANKS.co.uk (Dale Strickland-Clark)
Date: Sun, 17 Feb 2002 12:18:47 +0000
Subject: Redefining "\n"
Message-ID: <nl7v6ugpaa2skti66dlocutacafljsnlh0@4ax.com>

Is there some way to redefine the "\n" escape sequence?

When preparing strings for the Windows clipboard, I need \x0d\x0a for
a newline which is a bit cumbersome to insert all over the place.

It would be neat if I could make "\n" insert some other sequence.

I know I can do a translation before sending the string to the
clipboard but this would be nicer.

--
Dale Strickland-Clark
Riverhall Systems Ltd


From cliechti at gmx.net  Sat Feb 16 21:02:45 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 17 Feb 2002 03:02:45 +0100
Subject: Newbie: Problems with File IO
References: <Xns91B873A49D077AusYourStandingInIt@210.49.20.254>
Message-ID: <Xns91B81F8828230cliechtigmxnet@62.2.16.82>

Flavian Hardcastle <deathtospam43423 at altavista.com> wrote in 
news:Xns91B873A49D077AusYourStandingInIt at 210.49.20.254:

> out_file = open("log.txt", "w")
mode 'a' does append (thats the one you want), 'w' starts a new file 
for writing, adding a plus 'a+' or 'w+' opens for read/write

-- 
Chris <cliechti at gmx.net>



From mertz at gnosis.cx  Fri Feb 15 14:01:49 2002
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Fri, 15 Feb 2002 14:01:49 -0500
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>
 <a4j5hd$l3j2$1@ID-11957.news.dfncis.de>
Message-ID: <dsVb8kKkXgXB092yn@bellatlantic.net>                         

|"Trung Hoang" <thoa0025 at mail.usyd.edu.au> wrote in message
|> i was wondering if i can get certified with python from an
|> internationally recognised company?

"Emile van Sebille" <emile at fenx.com> wrote previously:
|I don't know if they're still doing it, but at one point Brainbench.com
|was working on competency testing, and IIRC, gearing that towards
|employers and prospective employees. http://www.brainbench.com/

Let me chime in slightly, given the thread.  I happen to have *written*
the Brainbench Python exam.

Unfortunately, my experience in working with Brainbench so was not
positive.  Aside from being a real PITA to work with, and not paying me
nearly enough for my time, I really did not feel that their guidelines
allowed production of nearly as good a test as was possible.  I have/had
some background in test design, and it was painful to me to work through
constraining and poorly conceived Brainbench test creation guidelines.
Even after creating the exam as best I could within their framework (bad
creation software also), they still required me to go back and
rephrase/restructure many question for the worse.  In the end, I felt
bad about creating something that is seriously flawed as a
skills-validation tool.

In other words, I don't believe that Brainbench's Python exam is a
*good* testing tool.  I can't speak specifically for other topics; I did
not write them, and never took one (except some initial playing around
in a topic I didn't really know, just to get a feel for the mechanics of
the exams).

That said, the original question was from someone who wanted a
certificate.  You can certainly take the Brainbench exam, and get a
piece of paper out of the process.  There's not necessarily a need to
forward my criticisms to your potential employer :-).

Yours, David...

--
    _/_/_/ THIS MESSAGE WAS BROUGHT TO YOU BY: Postmodern Enterprises _/_/_/
   _/_/    ~~~~~~~~~~~~~~~~~~~~[mertz at gnosis.cx]~~~~~~~~~~~~~~~~~~~~~  _/_/
  _/_/  The opinions expressed here must be those of my employer...   _/_/
 _/_/_/_/_/_/_/_/_/_/ Surely you don't think that *I* believe them!  _/_/




From gmcm at hypernet.com  Mon Feb 11 09:24:17 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 11 Feb 2002 14:24:17 GMT
Subject: Help with Gordon McMillan Installer
References: <d7d5ebdf.0202091152.40a6e0af@posting.google.com> <Xns91B167C6A6F08gmcmhypernetcom@199.171.54.216> <d7d5ebdf.0202101122.19931f2a@posting.google.com>
Message-ID: <Xns91B25FCC1C7C3gmcmhypernetcom@199.171.54.214>

Matthias Janes wrote:

> 2. Now to my real question : How about when the Data file is in a
> subfolder I tried this:
> 
> a = Analysis(['C:\\_Matthias\\_APPS\\TEST3\\MyApp.py'],
>              pathex=[])
> pyz = PYZ(a.pure)
> exe = EXE( pyz,
>           a.scripts,
>           a.binaries,
> 
>           [('Data.cfg', 'c:/_Matthias/_APPS/TEST3/Data/Data.cfg',
> 'BINARY')],
> 
>           name='MyApp.exe',
>           debug=0,
>           console=1)

The C code doesn't do mkdir, so it can only extract to the exe's
directory.

You can look at support/unpackTK.py as an example of how to 
extract a file in the archive (the code is more complex than
you need, because it's opening an embedded CArchive). Combine
code derived from that with:

  [('Data/Data.cfg', 'c:/_Matthias/_APPS/TEST3/Data/Data.cfg', 'DATA')]

You might want to think seriously about what benefit --onefile
is giving you with an app this complex. A --onedir will be much 
easier to set up, and much faster to start.

-- Gordon
http://www.mcmillan-inc.com/


From niemeyer at conectiva.com  Wed Feb 20 13:36:52 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Wed, 20 Feb 2002 15:36:52 -0300
Subject: Large integers from C
In-Reply-To: <20020220.163039.1504569917.30753@ic.ac.uk>
References: <20020220.163039.1504569917.30753@ic.ac.uk>
Message-ID: <20020220153652.A4115@ibook.distro.conectiva>

[...]
> I need to create Large (>300 bit) python Long's from C. I don't need (and
> in fact, don't want) to use the features of the (apparently obselete?)
> mpz module, or gmpy, or whatever. I simply want to say:
> 
> val = PyLong_FromOctets(char *v, int vlen);

You're probably looking for PyLong_FromString().

[...]
> intval = reduce(lambda x,y: (x<<8)+ord(y), strval)
> 
> I've tried, and this was too slow (the whole point of the C module being
> to speed this up).

Please, have in mind that this is mostly what long(str, base) does, so
you're not going to have great performance improvements by using it
instead of the python equivalent.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From op73418 at mail.telepac.pt  Mon Feb 25 12:20:46 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Mon, 25 Feb 2002 17:20:46 +0000
Subject: Question on Python path
References: <3qnk7ug977hk4peorufobtlo38nh8j9v87@4ax.com> <Xns91C0B74CA73EEcliechtigmxnet@62.2.16.82>
Message-ID: <3gsk7uo3d2l0aiggenr34kar0itn0f7jk6@4ax.com>

On 25 Feb 2002 17:57:44 +0100, Chris Liechti <cliechti at gmx.net> wrote:

>Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote in 
>news:3qnk7ug977hk4peorufobtlo38nh8j9v87 at 4ax.com:
>
>> I have put all my own modules in a subdirectory of site-packages. 
>> How can I instruct Python (I'm using Python 2.2 on win2k) to also 
>> look up this directory when a
>> 
>> import <whatever>
>> 
>> is found?
>
>if you put it in site packeages it should be a package... this means 
>the directory should have an __init__.py in it. then you can use it 
>with
>>>> import yourdir.yourmodule

Hmmm... Now that you tell me it does seem obvious... *rolls eyes*

>
>or you can put your modules anywhere and create a ".pth" file with 
>the path in it. this is appended on sys.path on startup.
>
>and then there is the PYTHONPATH environment variable.
>
>or you can add you path to sys.path in sitecustomize.py

This last one seems to be the best.

Thanks,
Gon?alo Rodrigues


From tim at vegeta.ath.cx  Wed Feb 13 21:16:45 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Thu, 14 Feb 2002 02:16:45 GMT
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013559740.27667.python-list@python.org> <slrna6kkpj.is0.tim@vegeta.ath.cx> <pu6e4a.5ti.ln@wombat.san-francisco.ca.us>
Message-ID: <slrna6m7jd.o2v.tim@vegeta.ath.cx>

Keith Keller <kkeller at speakeasy.net> graced us by uttering:
> Tim Hammerquist <tim at vegeta.ath.cx> writes:
>> IIRC, "agua" _is_ feminine, but takes the masculine article when
>> singular. Mi diccionario dice que "agua" es feminina, tambi?n.
>> 
>> OTOH, la palabra "mar" es masculina _y_ feminina. Por ejemplo:
>> 
>>     "el agua en la mar"     # Correct
>>     "las aguas del mar"     # Correct
> 
> Es mas malo que Perl!  ;-)

Possibly, as you seem to define "worse". At least in Perl5, if you
see $palabra, you can be sure it's a scalar, and @lista is an array.

OTOH, exceptions in Spanish are much less frequent than in English or
French.  Pronunciation is nearly 100% predictable, and there are more
regular verbs than most languages.

(I have no knowledge of German.)

> (yes, it's been a decade since I spoke Spanish, why do you ask?)

No es importante. Es importante que Ud. continue aprender. Puede
aprender Espa?ol, Ingles, Japon?s, Perl, Python, o Ruby.

> --keith

Cheers,
Tim Hammerquist
-- 
: Do you want perl everywhere or not?

No, sometimes I just want it nearby.
    -- Larry Wall in <199904091930.MAA06625 at kiev.wall.org>


From lienard.bruno at free.fr  Sat Feb  2 15:19:34 2002
From: lienard.bruno at free.fr (Bruno Liénard)
Date: Sat, 2 Feb 2002 21:19:34 +0100
Subject: Xmi parser
References: <3c5bb698$0$4609$626a54ce@news.free.fr> <m31yg49lyn.fsf@mira.informatik.hu-berlin.de>
Message-ID: <3c5c49b3$0$4617$626a54ce@news.free.fr>

Yes I know, but given the complexity of the DTD and the number of tags and
attributes, I think it is quite a lot of work, so if anybody had already
done it... I need a parser that can read a XMI file and create the
corresponding objects.  I have tried with xml_objectify, but it doesn't
work. I didn't spend a lot of time to understand why..

Regards,

Bruno




"Martin v. Loewis" <martin at v.loewis.de> a ?crit dans le message news:
m31yg49lyn.fsf at mira.informatik.hu-berlin.de...
> "Bruno Li?nard" <lienard.bruno at free.fr> writes:
>
> > I am looking for a xmi parser. Does anybody here have already
> > written one ?
>
> Given that XMI is an XML application, any XML parser should be able to
> parse XMI. I recommend xml.parsers.expat (or the SAX and DOM wrappers
> around it).
>
> Regards,
> Martin
>




From emile at fenx.com  Mon Feb 25 07:07:46 2002
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 25 Feb 2002 04:07:46 -0800
Subject: Remote debugging ?
References: <g7qj7uollai605vb158ahl11bbdnehfnam@4ax.com> <7xadtxajuo.fsf@ruckus.brouhaha.com> <1014637646.677472@news.kkf.net>
Message-ID: <a5d9oe$64uvi$1@ID-11957.news.dfncis.de>

"Timo Rainio"
> a slight difference in logic, than can cause big missunderstandings as
we
> see here ... ;-)

More like needn't vs mustn't?  ;-)

I took it to mean they required support for whatever solution was
suggested.

Anyway, IIRC, Wing and Komodo both provide some level of remote
debugging from Windows to Linux, although perhaps in a limited way.
I've seen them both mentioned specifically to remote debugging of a Zope
server, if that's the problem.

HTH,

--

Emile van Sebille
emile at fenx.com

---------



From starx at pacbell.net  Mon Feb  4 16:54:47 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 4 Feb 2002 21:54:47 GMT
Subject: Backwards Compatibility of Python versions
References: <a1rn4a.3vsgbmv.1@kserver.org> <m3elku0w20.fsf@mira.informatik.hu-berlin.de> <83sn8peybr.fsf@panacea.canonical.org> <kWE58.6100$om6.141694@news1.tin.it> <83bsf7z4h2.fsf@panacea.canonical.org> <7x4rkzw9yz.fsf@ruckus.brouhaha.com> <3C5C973E.6080607@sneakemail.com> <7xbsf7dqj9.fsf@ruckus.brouhaha.com> <B883FC7B.4309%jonathan@onegoodidea.com> <7xu1sxzkfo.fsf@ruckus.brouhaha.com>
Message-ID: <Xns91AB8D874477ARASXnewsDFE1@130.133.1.4>

Paul Rubin || Mon 04 Feb 2002 01:05:15a:

> Jonathan Hogg <jonathan at onegoodidea.com> writes:
>> > So much for compatibility then.
>> 
>> I think I missed the beginning of this thread so I'm confused. Exactly
>> which new backwards incompatible feature of the recent Python versions
>> is it that is breaking people's old scripts?
>> 
>> So far I've seen complaints of 'x in somedict' which only affects
>> writing new code for the new Python interpreter, future division which
>> is not enabled yet by default, and the import __future__ mechanism,
>> which again is only meant to be called by new code on the new
>> interpreter. 
> 
> That's the problem--if I don't import future division, my code will
> break in the future, because the // operator isn't supported in the
> present.  But if I -DO- import future division, my code will break in
> the present, because current (2.1) and older (1.5) interpreters don't
> recognize the import statement.  (The most recently released version
> of Python is a 2.1 series release, so 2.1 is still current).
> 
> I don't see any reasonable way I can write code right now that uses
> division that works in both the current 2.1 (which doesn't support
> importing future division) and in forthcoming versions (where the
> semantics of the / operator will be incompatibly different).

Specifically add a decimal to a number or float() something if you're only 
using variables, and you will have the same behavior as the new division, 
which won't change between new and old division. Do this, and int() the 
result, and you will have the same effect as the new //, even when it is 
not available. 


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From quinn at hurl.ugcs.caltech.edu  Fri Feb  8 19:52:10 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 9 Feb 2002 00:52:10 GMT
Subject: encapsulation, typing & other questions
References: <3C62C5C5.7FEF0B10@daylight.com> <a3uk4b$q4i$2@news.wrc.xerox.com>
Message-ID: <slrna68slp.qvm.quinn@hurl.ugcs.caltech.edu>

>Ragu Bharadwaj <ragu at daylight.com> writes:
>
>> 3. Are there plans to include method & variable encapsulation in the
>> object models in python? The answer I've heard is that since it's
>> possible to overcome encapsulation by long or dirty ways, it isn't worth
>> it. But my comeback is that if one sticks to the rules while coding,
>> this won't happen and the resulting gains are worth it.

Your 'comeback' is more applicable than you think, and is the answer you should
have heard (the answer you did hear is pretty silly).

It also applies to the dynamic vs. static thing.

"sticks to the rules" might be better phrased as "follows good coding
practice".


From simon at ping-pong.budig  Mon Feb  4 11:12:45 2002
From: simon at ping-pong.budig (Simon Budig)
Date: 4 Feb 2002 17:12:45 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net>
Message-ID: <3c5eb2fd$1@si-nic.hrz.uni-siegen.de>

Michael Hudson wrote:
> Simon Budig <Simon.Budig at unix-ag.org> writes:
>> In article <3c5b2c1f at si-nic.hrz.uni-siegen.de>, Simon Budig wrote:
>
> [complaint as subject]
> 
>> > Well - I looked at the documentation for the readline-module, but that
>> > doesn't say anything about it. Maybe I should read the sources.
>> 
>> Ok, did that. Here is the Patch to fix this behaviour. It is against
>> Modules/readline.c from Python 2.1.2.

[patch]

>> I really would appreciate, if this gets incorporated in the regular
>> Python releases. 
> 
> First, an awkward question: do you know which versions of readline
> support the rl_attempted_completion_over variable?  If it's not been
> present since 2.2, some nasty autoconf hackery may be needed (autoconf
> can join readline and ncurses on that list we were talking about...).

I just downloaded readline-2.2.tar.gz. The documention does not
mention this variable but readline.h has:

/* Non-zero means to suppress normal filename completion after the
   user-specified completion function has been called. */
extern int rl_attempted_completion_over;

I guess, this counts as a "yes" - missing documentation is nothing
unusual...   :-)

>> Unfortunately I don't have an account at sourceforge to submit this
>> the proper way.
> 
> Why not?

It is way too much hazzle to maintain *another* account/password to
be able to give three or four bugreports to one or two (admittedly
important) projects.

Ok, you can also say that I am too lazy.   ;-)

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From krnylund at msn.com  Sun Feb 24 07:51:40 2002
From: krnylund at msn.com (Kristian Nylund)
Date: 24 Feb 2002 04:51:40 -0800
Subject: GUI (PyQt) & thread problem
Message-ID: <8b5d764c.0202240451.1c4bda3b@posting.google.com>

Hello

Can someone tell me what the problem is with the following program, as it 
hangs. I'm pretty sure there's some inherent design-flaw involved, but
I'm not familiar enough with threads or GUI-programming to know what it 
could be.

I've also tried modyfing the program to use PYSIGNALs instead of passing the
caller to the thread, but the results were the same.

tia,

Kristian

#!/usr/bin/env python
from qt import *
from threading import *
from time import sleep

class Worker(Thread):

    def __init__(self, mw):
        Thread.__init__(self)
        # Take calling object as attribute
        self.mw = mw
        self.counter = 0

    def run(self):
        # Add 5 objects the the ListView
        while self.counter < 5:
            # Invoke method from calling object
            self.mw.addRow('val%d' % self.counter)
            self.counter = self.counter + 1
            sleep(1)
        # Remove the 5 objects from the ListView in order
        while self.counter > 0:
            # Invoke method from calling object
            self.mw.delRow('val%d' % self.counter)
            self.counter = self.counter - 1
            sleep(1)
        return

class MainWin(QMainWindow):

    def __init__(self,parent = None,name = None,fl = 0):
        QMainWindow.__init__(self,parent,name,fl)
        self.setName("Form1")
        self.resize(200,250)
        self.setCentralWidget(QWidget(self,"qt_central_widget"))

        self.ListView1 = QListView(self.centralWidget(),"ListView1")
        self.ListView1.addColumn("Column 1")
        self.ListView1.setGeometry(QRect(20,10,150,200))

        self.worker = Worker(self)
        self.worker.start()
        # If the following line is omitted, the program will
        # hang when trying to add the first object to the ListView
        QListViewItem(self.ListView1, ' ')

    def addRow(self, val):
        QListViewItem(self.ListView1, val)

    def delRow(self, val):
        # This is where it hangs
        print "Hang..."
        item = self.ListView1.findItem(val, 0)
        print "This won't be printed."
        self.ListView1.takeItem(item)
        del(item)

if __name__ == "__main__":
    a = QApplication(sys.argv)
    w = MainWin()
    a.setMainWidget(w)
    w.show()
    a.exec_loop()


From jason at jorendorff.com  Sun Feb  3 01:26:42 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 3 Feb 2002 00:26:42 -0600
Subject: The object named 'type', in 2.2
In-Reply-To: <Xns91A9D1FE73881RASXnewsDFE1@130.133.1.4>
Message-ID: <HFEKILOLEFEFMKAECNDLMEAFDAAA.jason@jorendorff.com>

Philip Swartzleonard writes:
> 3. Type is it's own parent:
>
> >>> type.__class__ is type
> 1
>
> It is this last one particularly that I do not understand.

Well, the type of TypeType has always been itself.  For a long
time, it's been like this:

  type(3) is types.IntType
  type(types.IntType) is types.TypeType
  type(types.TypeType) is types.TypeType

It's the same today, except for two things:

  1.  type (like str, int, list, etc.) used to be just a function.
      Now it is a type.  But (like str, int, list, etc.) it still
      works if you call it as a function, just like the old way.

  2.  Now a bunch more objects have a nice x.__class__ attribute
      that is the same thing as type(x).

> I ran into it
> today trying to fix a recrusion-death problem in PyCrust under 2.2 -- it
> recursivly accesses thing's __class__es in it's attribute-finding
> routines. I've tried to find information on this, but it has eluded me so
> far...

I don't know where you can find more information, sorry.
It might be that recursively looking at __class__es is not what
you want after all; maybe you should walk x.__class__.__bases__
instead.

## Jason Orendorff    http://www.jorendorff.com/




From duncan at NOSPAMrcp.co.uk  Thu Feb 21 06:45:50 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Thu, 21 Feb 2002 11:45:50 +0000 (UTC)
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com> <Xns91BC6120D2B5duncanrcpcouk@127.0.0.1> <7xbsejkrl1.fsf@ruckus.brouhaha.com>
Message-ID: <Xns91BC769387361duncanrcpcouk@127.0.0.1>

Paul Rubin <phr-n2002a at nightsong.com> wrote in
news:7xbsejkrl1.fsf at ruckus.brouhaha.com: 

>> Python's generators can only yield directly from the generator as
>> they do not attempt to maintain any kind of stack.
> 
> Is that right?  I thought the classic coroutine example is supposed to
> also work with generators:
> 
>   def traverse(tree):
>      if tree.left: traverse(tree.left)
>      yield tree.value
>      if tree(right) traverse(tree.right)
> 
> Am I missing something?
> 
> 

Look in lib/test/test_generators for Guido's binary tree example:

    >>> def inorder(t):
    ...     if t:
    ...         for x in inorder(t.left):
    ...             yield x
    ...         yield t.label
    ...         for x in inorder(t.right):
    ...             yield x



-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From gmcm at hypernet.com  Sun Feb 10 10:11:11 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 10 Feb 2002 15:11:11 GMT
Subject: Help with Gordon McMillan Installer
References: <d7d5ebdf.0202091152.40a6e0af@posting.google.com>
Message-ID: <Xns91B167C6A6F08gmcmhypernetcom@199.171.54.216>

Matthias Janes wrote:

[snip]

> I would like to create a onefile exe but I can't figure out how to
> write the right Makespec.file.

[snip]

> a = Analysis(['C:\\_Matthias\\_APPS\\Installer\\support\\useUnicode.py',
> 'C:\\_Matthias\\_APPS\\TEST3\\MyApp.py'],
>              pathex=[])
> pyz = PYZ(a.pure)
> exe = EXE( pyz,
>           a.scripts,
>           a.binaries,

            [('MyData.cfg', 'c:/_Matthias/_APPS/TEST3/Data.cfg', 'BINARY')],

>           name='MyApp.exe',
>           debug=0,
>           console=1)
> 
> 
> I tried different ways to add the Data.cfg file but this is never
> included in the ONEFILE.EXE instead it is copied extra in a folder.

Inline above. As BINARY, it will be extracted on each run. Of course,
that makes it effectively read-only.

-- Gordon
http://www.mcmillan-inc.com/


From paul1brian at yahoo.com  Tue Feb 26 05:15:22 2002
From: paul1brian at yahoo.com (Paul Brian)
Date: Tue, 26 Feb 2002 10:15:22 -0000
Subject: Help using ODBC.pyd to retrieve ntext data from SQL server
References: <24420294.0202251419.82d3373@posting.google.com>
Message-ID: <1014718489.14788.0.nnrp-13.c1c3e191@news.demon.co.uk>

if this is what I think it is, the trouble is handling large amounts of blob
anywhere other than at the end of the select stmt.

"Select HugeField, Name, Amount from tbltest"

will get you a failed select or a blank hugefield depending on dbase and
odbc.

Try

"Select  Name, Amount, HugeField from tbltest"

Which might work. Good luck.

I think the official explanation is in MSDN somewhere but I cannot find it
now

---------------
Paul Brian
"still no sig"



Stephen <stephen.tubbs at paretopartners.com> wrote in message
news:24420294.0202251419.82d3373 at posting.google.com...
> I need help with a work around to read data from an ntext field on
> Microsoft SQL Server v7.0.  Python only returns None.  I'm able to get
> at the data in the field using a variety of other applications
> (Access, Query analyzer, etc...)  I have been through the odbc.pyd
> source and don't find any obvious exclusions for ntext data types.
> Any help would be greatly appreciated
>
> Stephen




From Gareth.McCaughan at pobox.com  Thu Feb 28 19:29:56 2002
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 1 Mar 2002 00:29:56 +0000
Subject: Status of PEP's?
References: <mailman.1014854027.3045.python-list@python.org> <3c7d8f26_3@news.bluewin.ch>
Message-ID: <slrna7tis4.1m1j.Gareth.McCaughan@g.local>

Samuele Pedroni wrote:
> Maybe this is a useful insight:
> 
> in mathematetics it is not uncommon to
> identify the natural numbers with the set
> of their predecessors
[etc]
> so also the notation   1 in 5  is natural.

I am a mathematician. The areas of mathematics I always
liked best were set theory and number theory (though I
ended up doing my PhD on something else). I am very
familiar with the von Neumann ordinals, which Samuele
describes.

Nevertheless, I think this would be a terrible thing
to put into Python. "Explicit is better than implicit",
remember? It would confuse newbies severely. It would
confuse non-newbies severely too, actually, though I
expect they'd get used to it. (Heck, people get used
to Perl and C++, too.) The "for x in 5" notation seems
to me to have no advantage whatever other than brevity.

von Neumann didn't invent his construction of the ordinals
because it was obviously the right thing, or because it
was natural. He did it because it was simple. When what
you're doing is laying foundations for mathematics,
that counts for a lot. But that's not what Python is about.

Go and look at a set theory textbook that describes that
construction of the natural numbers. It's a good guess that
within a few pages they will also have defined integers
as equivalence classes of ordered pairs of natural numbers,
rationals as equivalence classes of ordered pairs of integers,
and real numbers as Dedekind sections or equivalence classes
of Cauchy sequences or something. Should we take inspiration
from those too? Then, for instance, we could do

    >>> for x in -3:
    ...   print x
    ... 
    (3, 0)
    (4, 1)
    (5, 2)
    [etc, etc, until interrupted]

It doesn't get much more natural than that. :-)
I suppose we might also get

    >>> for x in math.pi:
    ...   print x
    ... 
    Traceback (most recent call last):
      File "<stdin>", line 1, in ?
    TypeError: iteration over uncountable set

(My own preferences, not that anyone asked for them:
I always rather liked the range literal proposal, but
GvR rejected that. I also like the "for 3 <= i < 10"
one, which I think is *beautiful* but maybe slightly
too cute.)

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


From arbalt at yahoo.com  Wed Feb  6 17:17:29 2002
From: arbalt at yahoo.com (arbalt at yahoo.com)
Date: 6 Feb 2002 14:17:29 -0800
Subject: multi-threaded *nix program needs to spawn an external program/script
Message-ID: <cd45264b.0202061417.590d8bf2@posting.google.com>

Would it make sense to add a function to python for *nix systems
    that would do the "start an executable file as a new process"?

If you have a multi-threaded program and some of the threads sometimes
need to spawn off new external programs or script.

You can do the following under win32, which starts the command
in a new process without forking():

    win32api.WinExec(command, ...)

It works great.

However, *nix systems don't have the "Start an executable file as a
new process" system call (just fork/exec).

I've tried various alternatives, which aren't working, such as:
    * os.system() works for a while, but:
          Eventually, at least on solaris, extra copies of the
          parent process are left hanging around and never go
          away (they are not zombies - the zombies are being
          reaped).

    * python code doing the os.fork() + os.exec().

      This is the least reliable of all:  It leaves around lots
      of extra forked() but not defunct/zombie processes.

      With threads, if you're going to fork + exec, you're not supposed to
      do anything between the fork + exec or you're asking for trouble.

      Trying the os.fork() + os.exec() in an interpreted program is
      probably asking for trouble (and sure enough, trouble happens).

>From what I've noticed, read, and heard, fork() inside threads() is
basically a bad idea.  Different OSes have different ideas,
bugs, etc.

But people are bound to need to have threaded programs spawn off separate
processes to do something in *nix.  (We are in that situation.)

What is the solution?  Would it make sense to add a function to python
for *nix systems that would do the "start an executable file as a
new process" that would work on at least *nix systems?

Or is there another solution?

Thanks!


From gh_pythonlist at gmx.de  Tue Feb  5 15:14:12 2002
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: Tue, 5 Feb 2002 21:14:12 +0100
Subject: how can i check if a variable is set?
In-Reply-To: <a3pd1j+joko@eGroups.com>
References: <a3pd1j+joko@eGroups.com>
Message-ID: <20020205201411.GB6560@lilith.hqd-internal>

Le 05/02/02 ? 19:50, maximilianscherr ?crivit:
> how cna i check if a variable is set?
> like:
> 
> if variable a doesn't exist then set it to 0
> if it exists do nothing...

You were already told how to do this, but ...

... that's not exactly the best style ;-)

Chances are what you really want is a dictionary, which maps keys to
values, suppose you want to do something like count the number of
occurences of various words:

text = "this text contains text"
words = {}
for word in text.split():
    if not words.has_key(word):
        # First occurence
        words[word] = 1
    else:
        # We found another occurence of this word
        words[word] = words[word] + 1


This can be optimized with the get() method of the dictionary object.
The first parameter of the get() method is the key, the second is a
default value that is used if there is no entry under the given key.

text = "this text contains text"
words = {}
for word in text.split():
    words[word] = words.get(word, 0) + 1

no-everything-that's-possible-is-a-good-solution-ly yours,

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 10.3 ?C      Wind: 2.0 m/s



From h.jehadalwan at student.kun.nl  Fri Feb 22 18:30:05 2002
From: h.jehadalwan at student.kun.nl (husam)
Date: Sat, 23 Feb 2002 00:30:05 +0100
Subject: is their any limit on the size of the data passed to a progra      m via popen2?
References: <mailman.1014418718.9469.python-list@python.org>
Message-ID: <3C76D47D.5020303@student.kun.nl>

Gustavo Cordova wrote:

> Yeah, Jeff Shannon wrote something about
> pipes hanging in a .read() operation. :-/
> 
> So... any other clues yet?
> 
> -gus

sorry, I don't get it! What did Jeff wrote? and where?



From me at mikerobin.com  Thu Feb 14 13:45:39 2002
From: me at mikerobin.com (Michael Robin)
Date: 14 Feb 2002 10:45:39 -0800
Subject: OT Annoying Dir Boxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com> <b16e4ef7.0202130950.67f4004d@posting.google.com> <Xns91B4EC515D335RASXnewsDFE1@130.133.1.4>
Message-ID: <52e5ab5f.0202141045.5867bfbb@posting.google.com>

Philip Swartzleonard <starx at pacbell.net> wrote in message news:<Xns91B4EC515D335RASXnewsDFE1 at 130.133.1.4>...
> Stephen Ferg || Wed 13 Feb 2002 09:50:00a:
> 
> This doesn't really have anything to do with this thread, but i'd like to 
> make this comment: =PLEASE= don't use the (apparently) standard directory 
> selcetion dialog in windows if you can avoid it. It can be truly 
> aggrivating when you know exactly how to type your way there, but -have- to 
> use the box... and on a reasonably crowded file system, this means 'open 
> node, scroll down, open node, scroll down' repeat-until-insane. 

You can type the path into the FileOpen dialog at any time, with or
without mousing first. I would say the opposite, please don't invent
your own file open dialog unless you have to - you'll lose multiple
views, property access, drag & drop, filename completion and standard
keyboard shortcuts - not to mention those who need to hook into
accessibility features.


From gerson.kurz at t-online.de  Thu Feb 21 13:11:43 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Thu, 21 Feb 2002 18:11:43 GMT
Subject: Hosting
References: <283adf56.0202201551.7026d548@posting.google.com>
Message-ID: <3c753779.18591703@news.t-online.de>

On 20 Feb 2002 15:51:24 -0800, kdahlhaus at yahoo.com (Kevin Dahlhausen)
wrote:

>Does anyone know of an inexpensive web host that allows compilation
>and use of Python extensions from  your own directory?

For those reading from Germany I can recommend www.domainfactory.de.
For 16 Euro you get both Python 1.5 and 2.0 (and perl, and php, and a
mysql database, everything running on linux). The installed python
modules include mysqldb, but you are not allowed to add compiled
extensions (sorry), and the service is generally really good. 


From skip at pobox.com  Wed Feb 13 12:03:43 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 13 Feb 2002 11:03:43 -0600
Subject: Disabling creation of .pyc files?
In-Reply-To: <E16b1mB-0001Ls-00@mail.python.org>
References: <E16b1mB-0001Ls-00@mail.python.org>
Message-ID: <15466.40047.754661.201615@beluga.mojam.com>

    Dominic> ... I'm a bit annoyed by the fact that for every .py file I'm
    Dominic> importing, Python automatically creates a corresponding .pyc
    Dominic> file, which persists after my program exits.

It's there for your benefit (imports are faster when up-to-date .pyc files
are present because the compilation step is avoided), but yes, you can
disable it by simply making the directory containing the .py files
read-only.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From wurmy at earthlink.net  Wed Feb 20 23:42:30 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Thu, 21 Feb 2002 04:42:30 GMT
Subject: A bundle of questions from a Python newbie
References: <pk687ugc5s0uhs5h7v3bf3r2q8v4tl7bhi@4ax.com>
Message-ID: <3C747AE3.FD16CB33@earthlink.net>

"Gon?alo Rodrigues" wrote:

> 1. The way I thought that an assignment like
> 
> var += 1
> 
> worked was that the interpreter would go to the place pointed by var and
> add 1. In particular, id(var) should remain constant. But this is not
> what happens as a simple test shows it. So what is wrong in my
> perception of what += does? Is this an implemetation issue?

There is no "place pointed by var". Python is a bit different than most
languages in this respect. An "assignment" like

  x = 1

creates an integer object with value 1, then attaches the name
'x' to it in the current namespace. In, say, a Pascal tutorial,
you will learn that a variable refers to a reserved place in
memory, and various values can be written to that place. Not
so in Python. When you do

  x = 2

later, the same thing happens: integer object with value 2
is created, and the name 'x' is associated with it. Since
this is a new integer, id(x) will have a different value
than before. (The original integer with value 1 may, or
may not, still exist, but 'x' isn't associated with it
anymore.)

Now += is a bit of a different issue, because in the
case of mutable objects, it's capable of changing the
object in-place. Take a list, for example:

>>> a = [1, 2, 3]
>>> id(a)
9586800
>>> a += [4]
>>> a
[1, 2, 3, 4]
>>> id(a)
9586800

a has the same id before and after the +=. This is
different from the = operator, that always (re)binds
to a new name:

>>> a = a + [5]
>>> a
[1, 2, 3, 4, 5]
>>> id(a)
9902208

+= has the same behavior for immutable objects, like
integers and strings:

>>> x = 1
>>> id(x)
7626688
>>> x += 2
>>> x
3
>>> id(x)
7626496

By now you probably think, what the hell is the guy
talking about?! :-) See this document for a clear and
concise description of how Python's object and
name binding mechanism works:

  http://www.effbot.org/guides/python-objects.htm  

> 2. What is the idea of having in the loop
> 
> for var in range(10):
>     pass
> 
> the var variable still available after it? Wasn't it supposed to die (as
> in being out of scope) when the loop ends? 

This is not uncommon in other languages either, if I recall
correctly. Either way, no, it's not out of scope. It only 
goes out of scope if the current function/method/class body 
ends.

> 3. This is not a question just a small complaint. I really miss the
> do-until control flow in Python. I always have to waste some extra brain
> cycles to convert it to the while 1, etc. idiom. Hey, we all have a
> right to our complaints, right? ;-)

This one comes up every once in a while. The usual reply is that a
do-while construction can easily be mimicked using a while. Your
mileage may vary, but it's probably pointless to argue for this
feature on c.l.py, unless you like long flamewars. ;-)

> 6. Is there a reason why 3.__abs__() does not work? After all 3 is an
> object (in the wider sense of the word). And it does work if we assign
> it to a variable, eg,
> 
> var = 3
> var.__abs__()
> 
> I know that code like the above (methods applied to number literals) is
> probably rare, but consistence is also a good thing.

I believe this is something of a parsing issue. Not sure, though.
Either way, I don't think there are many people that want to write
3.__abs__...

> 9. What would be better in terms of speed?
> 
> for i in range(<whatever>):
>     <whatever>
> 
> Or to code an iterator like
> 
> from __future__ import generators
> 
> class Range:
>     def __init__(self, end, start = 0, step = 1):
>         self.end = end
>         self.start = start
>         self.step = step
> 
>     def __iter__(self):
>         i = self.start
>         while i < self.end:
>             yield i
>             i += self.step
> 
> and then use it in the for loop.

I'd say that normally it's fastest to loop over the
list. Iterators and generators are useful when you have
a possibly really long list that you don't want to have
in memory all at once. AFAIK, they're not faster than
looping over a sequence. I didn't test this though.
c.l.py has some incorrigible benchmarkers who could
prove me wrong... ;-)

> 10. I have a class that works like a list in the sense that supports the
> method __getitem__ and its relatives. But I'm at a loss as to what I
> have to do to support the slice notation.

__getitem__(self, x) will be called with a so-called slice
object if you call your instance with the slice notation.
Example:

>>> import types
>>> class Foo:
	def __init__(self, x):
		self.data = x
	def __getitem__(self, x):
		if type(x) == types.IntType:
			return self.data[x]
		elif type(x) == types.SliceType:
			return self.data[x.start:x.stop]
		else:
			raise IndexError

		
>>> f = Foo(range(10))
>>> f[1]
1
>>> f[:2]
[0, 1]

HTH,

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From duncan at NOSPAMrcp.co.uk  Thu Feb 21 05:34:22 2002
From: duncan at NOSPAMrcp.co.uk (Duncan Booth)
Date: Thu, 21 Feb 2002 10:34:22 +0000 (UTC)
Subject: two ideoms at one blow: line-reading and regexp-matching
References: <mailman.1014233014.12475.python-list@python.org>
Message-ID: <Xns91BC6A85E544Dduncanrcpcouk@127.0.0.1>

"Jason Orendorff" <jason at jorendorff.com> wrote in 
news:mailman.1014233014.12475.python-list at python.org:

> Gustavo Cordova wrote:
>> >>> class WrapRX:
>> ... [...]
> 
> This is standard in Python 2.2.
> 
>  >>> import re
>  >>> rx_vowels = re.compile(r"[aeiou]+", re.S)
>  >>> some_string = "This is a beautiful string."
>  >>> for match in rx_vowels.finditer(some_string):
>   ...     print "Found vowel: %r" % match.group(0)
>   ...
> 
Pity it isn't documented. Neither is the scanner method.
I have reported this on sourceforge.

-- 
Duncan Booth                                             duncan at rcp.co.uk
int month(char *p){return(124864/((p[0]+p[1]-p[2]&0x1f)+1)%12)["\5\x8\3"
"\6\7\xb\1\x9\xa\2\0\4"];} // Who said my code was obscure?


From tim.one at home.com  Mon Feb  4 17:51:45 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 4 Feb 2002 17:51:45 -0500
Subject: Backwards Compatibility of Python versions
In-Reply-To: <7xy9i942bm.fsf@ruckus.brouhaha.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOELPNJAA.tim.one@home.com>

[Paul Rubin]
> ...
> What does "nothing to stop anyone doing it" mean?  Maybe Tim can explain
> that.  If I make some patches to 1.5.2 and submit it to the maintainers
> as 1.5.3, will they release it on the python.org website?  What if
> I make patches to 1.2 and submit it as 1.2.1?  (Or 1.2.2 or whatever).

The "Bug Fix Releases" PEP covers this, as has been pointed out before:

    http://python.sourceforge.net/peps/pep-0006.html

Nobody has time to look at "some patches", whether they're for a 1.5.2 or a
2.2 bugfix release.  Someone has to volunteer to do what the PEP says,
namely create a maintenance branch under CVS and get everything ready except
for final packaging.  Assuming it's you, if Guido can be persuaded there's
enough interest in the maintenance release you've created to justify the
work (meaning enough interest to justify dropping or delaying other current
work), he'll direct the PythonLabs crew to finish the release (including
publishing on python.org, the whole bit).  Else you'll have to.  Note that
during the time you've announced you're responsible for a given bugfix
release, it's your job to review patches people may submit for it, and to
either test them yourself across all platforms, or find enough other
interested volunteers to do it for you.  It's real work.

> ...
> I wonder what the demographics really are (what percentage of users
> are using what Python version).

Nobody knows.

> ...
> I don't see at all why some mechanism can't be provided to turn off the
> change ("import from past") just like there's now a mechanism to turn on
> the change ("import from future").
> ...
> The subject has come up before?

Is there a subject that hasn't <wink>?  If Guido had any plan to do this, it
would be mentioned in the PEP.  FWIW, I predict with high confidence he
won't go for it.  He's trying to change the language forever, not provide
for incompatible dialects based on goofy pseudo-imports; __future__ was only
designed to ease one-way transitions, and it's part of the design that every
__future__ statement eventually has no semantic effect; __past__ didn't even
get off the ground.




From dude at deletia.com  Mon Feb  4 20:29:03 2002
From: dude at deletia.com (dude at deletia.com)
Date: Mon, 4 Feb 2002 20:29:03 -0500 (EST)
Subject: Scopes
Message-ID: <Pine.BSO.4.33.0202042027290.7309-100000@resin.csoft.net>

Hi all,

I've been learning C++ and have been turned onto Python

but the quesiton i have is

What is the advantage of pythons nested scope rules over c++ rules.

If I read correctly

If func B() has Func A() as its parent, then Func A() has access (can see)
all of func B()'s variable.

I am not sure of the advantage here, and would appreciate a little help

Thanks,

G






From skip at pobox.com  Wed Feb 13 00:22:23 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 12 Feb 2002 23:22:23 -0600
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <3C69ED33.CE0100DF@pizero.colorado.edu>
References: <mailman.1013557302.9749.python-list@python.org>
        <7xzo2ekv5g.fsf@ruckus.brouhaha.com>
        <mailman.1013572042.18876.python-list@python.org>
        <3C69ED33.CE0100DF@pizero.colorado.edu>
Message-ID: <15465.63503.142011.246239@12-248-41-177.client.attbi.com>

    >> I realized that might be an issue on my way home.  This is on my
    >> development machine.  I (user == skip) do have write access to
    >> /usr/local/*.  Sorry about the missing info.

    Fernando> - what filesystem are you using?

ext2.

    Fernando> - are all the files which get corrupted files that get
    Fernando> explicitly open by your crashing process?

It appears so.  As far as I could tell, all the .so files that were writable
by me and dlopened by the Python interpreter got zapped.  That included the
gcc 3.0.1 .so file, readline.so, etc, but not libc.

    Fernando> [ tale of reboot woe elided ].  I give you this background in
    Fernando> case you can spot some similarities with your situation.

Thanks, but that doesn't ring any bells.

Oh well, I'll launch myself back into the fray tomorrow and see if I can
figure anything else out.

Skip



From Scirious at Scirious.com  Fri Feb 15 15:00:07 2002
From: Scirious at Scirious.com (Scirious)
Date: Fri, 15 Feb 2002 20:00:07 GMT
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131> <Xns91B6B55E421A2Scirious@140.99.99.132>
Message-ID: <Xns91B6B77EC89ECScirious@140.99.99.130>

Scirious <Scirious at Scirious.com> wrote in 
news:Xns91B6B55E421A2Scirious at 140.99.99.132:

>Ok! Once again thaks for your rsponse.
>Scirious.
>

Sorru for this multiple postage, it was a problema with the server.
Scirious.


From phd at phd.pp.ru  Mon Feb 18 04:52:57 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Mon, 18 Feb 2002 12:52:57 +0300
Subject: ODBC on unix w/o mxODBC
In-Reply-To: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>; from vvainio@karhu.tp.spt.fi on Mon, Feb 18, 2002 at 11:31:10AM +0200
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>
Message-ID: <20020218125257.I24919@phd.pp.ru>

On Mon, Feb 18, 2002 at 11:31:10AM +0200, Ville Vainio wrote:
> Briefly put: can it be done? I'm mainly concerned with the price tag
> on mxODBC...

   http://www.iodbc.org/
   http://www.unixodbc.org/
   http://www.jepstone.net/FreeODBC/

   I have no idea how to install and/or use them, and what they have with
Python...

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



From herve.guyot at thales-avionics.com  Tue Feb 26 07:50:47 2002
From: herve.guyot at thales-avionics.com (GUYOT =?iso-8859-1?Q?Herv=E9?=)
Date: Tue, 26 Feb 2002 13:50:47 +0100
Subject: Simple translation from python to C++
Message-ID: <3C7B84A7.9AF7DF95@thales-avionics.com>

Hello everybody,

I've been reading comp.lang.python for some time, and really enjoy it.
Now, I have an "easy" problem to submit (I'm using the mailing list
since I don't have write access to the newsgroup)

In our team, we've been using Python for some years now, and some of our
older code is *really* poorly written (I mean *really* :-)

About one year ago, our company choosed Rhapsody (from Ilogix) an UML
modeling tool.

And naturally, I've been asked to "model" this poorly designed code into
Rhapsody in order to try and have a better understanding of what goes
on,
an if possible, refactor it. My first step is to define the classes
(with methods and attributes declarations) and inheritance
relationships. Not difficult.

But instead of dooing it by hand, I've thought of extracting from the
python code C++ class, methods and attributes declarations, and then,
feed the retro-engeneering tool of Rhapsody with this "skeletton"
C++ code.

Has anybody already done this ? (it's not difficult, but if some one can
save me some time :-)

Thanks in advance for your answers

Sincerly

--  
Herv? Guyot
Thal?s Avionics - DSI/AD/DOT
Valence - France



From jeff at ccvcorp.com  Fri Feb  8 15:00:48 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 12:00:48 -0800
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <a3u9sk$890$1@news.mathworks.com> <3C62D5A5.4B57F487@ccvcorp.com> <3c63f1e3@buckaroo.cs.rit.edu>
Message-ID: <3C642E70.D6D6C9A2@ccvcorp.com>

Keith Jones wrote:

> > > from mymod import *
> > >
> > > def dosomething(x):
> > >     import mymod
> > >     mymod.dosomething(1000)
> >
> > Nope,this does not work.  As the O.P. already noted....
>
> did you miss the fact that he's importing mymod in the function?  'Course,
> based on the example, then, the 'from mymod import *' is unnecessary, I
> would think.

Oh.  Duh.  Yeah, I did miss that.  :)   You're right, though, that does seem to
make the wildcard import unnecessary...  and much better avoided anyhow.

Jeff Shannon
Technician/Programmer
Credit International




From jbauer at rubic.com  Wed Feb 13 10:52:31 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 13 Feb 2002 09:52:31 -0600
Subject: Python 2.2 <-> Solaris 2.6 installation errors
Message-ID: <3C6A8BBF.EE10D42B@rubic.com>

Chris Wysocki writes:
>> Any ideas, things to try?

Martin v. Loewis writes:
> It is understandable why it fails crashes the second time. What is
> surprising is why it did not manage to remove the faulty readline
> extension the first time (when it really did build it). Unfortunately,
> Jeff did not report the output from the first make invocation, so this
> is somewhat hard to tell.

Easily enough resolved <0.2 wink>.  I've logged output from 
the configuration, make, and make install runs.  They are
too long to post, but I'll send them as email attachments
later.

I think the problems Chris and I are having may be similar.

The only thing I can find on the SourceForge Tracker is
a make test failure on sunos5 (#512007), to which Martin has
already responded, if not actually resolved.

Should I submit this as a build/installation bug?

Jeff Bauer
Rubicon Research



From robin at jessikat.fsnet.co.uk  Mon Feb  4 12:38:59 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Feb 2002 17:38:59 +0000
Subject: Pythonservice stuck
References: <ot678KAD6pX8Ewmt@jessikat.demon.co.uk>
 <3C5EBC4D.DFF874E7@noaa.gov>
Message-ID: <Jomj9VAzcsX8Ewy7@jessikat.demon.co.uk>

In article <3C5EBC4D.DFF874E7 at noaa.gov>, jim.vickroy
<jim.vickroy at noaa.gov> writes
>I'm not an experienced MS Windows programmer and I have no experience with
>Zope, in particular, but I have always been able to remove services by
>using the registry editor (regedit).  If memory serves, on some Windows
>platforms, the entries remain visible until a reboot is performed.
>
>Robin Becker wrote:
>
>> Win2k SP2
I hand deleted the service entry (which worked) and now after a second
reboot the legacy entries are gone.
-- 
Robin Becker


From jason at jorendorff.com  Mon Feb 18 17:53:36 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: 18 Feb 2002 16:53:36 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb 18)
Message-ID: <386B3C980CD992FC.BBF73A99F9407A77.B86573627A805FB4@lp.airnews.net>

QOTW:  "I am always amazed at how many things are built in to Python."
-Kevin Parks, upon being reminded about the bisect module

"Using pyperl to access Oracle is like taking a bus from New York to
San Francisco so you can fly to Los Angeles."  -Steve Holden

"In some ways, it's an like an RPN version of Python, without the
class machinery." -Roy Smith in praise of Postscript


    Charleroi, 40 km south of Brussels, will host the first European
    Python/Zope Conference on June 26-28, 2002.
	http://www.europython.org/wiki
	mailto:europython at python.org

    xml_pickle converts Python objects and data structures to XML and
    back.  Version 0.51 sports a dozen new features, including a
    fine-grained security model.
        http://groups.google.com/groups?th=a05a64ce89c893a7
    
    PEP 282 proposes a standard logging system for Python.
        http://groups.google.com/groups?th=b849e15006ee969b
    
    O'ReillyNet covers the new incarnation of Stackless Python.
        http://www.onlamp.com/pub/a/python/2002/02/14/pythonnews.html
    
    AnyCons is a common API for interactive console-mode graphics that
    works on Unix and Windows.
        http://newcenturycomputers.net/projects/anycons.html
    
    Sheila King implements file locking on Unix and Windows.
        http://groups.google.com/groups?selm=a4ghg4.3vv79at.1%40kserver.org
    
    EasyGui provides some convenient functions for simple GUI dialogs,
    allowing mostly-procedural code to pop up the occasional message box.
        http://home.att.net/~stephen_ferg/easygui/index.html
    
    Several approaches for drawing graphics "in slow motion" with Tkinter:
        http://groups.google.com/groups?th=15bf9d6a120fa5bd
    
    Several approaches for dynamically dropping input fields into a
    paragraph of text in a GUI.
        http://groups.google.com/groups?th=705793c315c8d274
    
    Just van Rossum explains how to sort a list of lists by their first
    element, using the second element to break ties.  (Hint:  There is
    a default sort order for lists.)
      http://groups.google.com/groups?selm=just-26D236.09311915022002%40news1.xs4all.nl
    
    On comp.lang.python: An innocent question about the leading-zero
    notation for octal numbers leads to a historical quest for the origin
    of this confusing syntax.  No conclusion is reached, but some Sperry
    UNIVAC nostalgia is shared.
        http://groups.google.com/groups?th=68307c927083e7d0
        http://www.fourmilab.ch/documents/univac/fastrand.html
    
    Guido van Rossum wins the fourth annual FSF Award for the
    Advancement of Free Software.
        http://www.fsf.org/press/2002-02-16-FSF-Award.html
    

========================================================================

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

    Python.org's Python Language Website is the traditional
    center of Pythonia
        http://www.python.org
    Notice especially the master FAQ
        http://www.python.org/doc/FAQ.html

    PythonWare complements the digest you're reading with the
    daily python url
         http://www.pythonware.com/daily  
    Mygale is a news-gathering webcrawler that specializes in (new)
    World-Wide Web articles related to Python.
	 http://www.awaretek.com/nowak/mygale.html 
    While cosmetically similar, Mygale and the Daily Python-URL
    are utterly different in their technologies and generally in
    their results.

    comp.lang.python.announce announces new Python software.  Be
    sure to scan this newly-revitalized newsgroup at least weekly.
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

    Michael Hudson continued Andrew Kuchling's marvelous tradition
    of summarizing action on the python-dev mailing list once every
    other week, into July 2001.  Any volunteers to re-start this
    valuable series?
        http://starship.python.net/crew/mwh/summaries/
        http://www.amk.ca/python/dev

    The Vaults of Parnassus ambitiously collect Python resources
        http://www.vex.net/~x/parnassus/   

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

    The Python Software Foundation has replaced the Python Consortium
    as an independent nexus of activity
        http://www.python.org/psf/

    Cetus does much of the same
	http://www.cetus-links.org/oo_python.html

    Python FAQTS
        http://python.faqts.com/

    The old Python "To-Do List" now lives principally in a
    SourceForge reincarnation.
	http://sourceforge.net/tracker/?atid=355470&group_id=5470&func=browse
	http://python.sourceforge.net/peps/pep-0042.html

    *Py: the Journal of the Python Language*
        http://www.pyzine.com
     
    Links2Go is a new semi-automated link collection; it's impressive
    what AI can generate
        http://www.links2go.com/search?search=python

    Tenth International Python Conference 
        http://www.python10.org            

    Archive probing tricks of the trade:
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100
        http://groups.google.com/groups?meta=site%3Dgroups%26group%3Dcomp.lang.python.*

Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://www.ddj.com/topics/pythonurl/         
  http://purl.org/thecliff/python/url.html (dormant)
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From store_li at sina.com  Tue Feb 26 03:27:44 2002
From: store_li at sina.com (Ricky)
Date: Tue, 26 Feb 2002 16:27:44 +0800
Subject: MSWord and Python
References: <3C7A4EF8.4010404@thomas-guettler.de> <a5f00p$6udv4$1@ID-12869.news.dfncis.de> <3C7B4477.2040909@thomas-guettler.de>
Message-ID: <a5fgvp$6v232$1@ID-12869.news.dfncis.de>

I am sorry, may be I didn't make myself clear. As I know you couldn't simply
create a MSWORD documents without COM.
"Thomas Guettler" <st-newsgroups at thomas-guettler.de> wrote in message
news:3C7B4477.2040909 at thomas-guettler.de...
> Ricky wrote:
>
> > The client side document is include in Python online documents.
>
>
> Sorry I didn't understand you. Is there are library for creating word
> documts without COM? This would be nice since I could create the
> documents on unix, too. What's its's name?
>
> thomas
>




From philh at comuno.freeserve.co.uk  Fri Feb  8 08:18:22 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 8 Feb 2002 13:18:22 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com>
Message-ID: <slrna67k0u.for.philh@comuno.freeserve.co.uk>

On Fri, 08 Feb 2002 01:42:51 GMT, Courageous <jkraska at san.rr.com> wrote:
>
>>It's just that being forced -- "for my own good" -- not to use certain
>>features because someone has decreed them intrinsically bad, doesn't
>>appeal to me at all.
>
>Sure. You realize that a hygienic macro doesn't prevent you from
>modifying things by consequence, right? All it does is require for
>you to mean it; this caters to Python's phenonomenon of least
>surprise. Imagine the following
>
>>>> i = 3
>>>> j = 4
>>>> mymacro i
>>>> print j
>5
>
>That would vex some folks.

It would vex me in someone else's code unless there was a good reason for
it and it was well documented.

> But that's exactly what a non hygienic
>macro can do.

And it's exactly the behaviour I want a macro to be able to do.

Python macros should, if they exist, have a superset of the capabilities of
C macros. Including the ability to be replaced by arbitrary code in the
program. Why arbitrary? firstly, because it is conceptually simple;
secondly because it doesn't restrict the programmer.

To take your example above, the ability to increment a global variable is 
useful because it can be used to determine how many times a macro has 
been run.

> Granted, in some cases non hygeniec macros can be
>quite useful. And the design I'm targeting is only _pseudo_ hygienic
>in any case. It's hygienic for the local namespace. 

In which case, it will behave complicatedly. Consider this code:

defmacro incj:
   j = j + 1

j = 4
print j
incj
print j

def myfun(j):
   print j
   incj
   print j

myfun(100)


In a non-hygenic macro system it is easy to see what underlying Python
code gets produced. In your system it isn't -- I've no idea what you
would want this to do -- therefore it is more complex. As a general rule,
whenever a program has more complexity than necessary, it is too complex. 
Complexity is bad: it leads to poor understanding and extra code 
therefore extra bugs. I would estimate over 90% of program bugs are
caused by not following this rule.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From matthias_j at lycos.com  Sat Feb 16 13:23:07 2002
From: matthias_j at lycos.com (Matthias Janes)
Date: 16 Feb 2002 10:23:07 -0800
Subject: WNetCancelConnection2 -- How Can I cancel all open connection to a Server at once
References: <d7d5ebdf.0202142247.1f7d98e1@posting.google.com>
Message-ID: <d7d5ebdf.0202161023.3e3ea179@posting.google.com>

Hi anyone a Idea how to go about it.


From waldo at saclant.nato.int  Fri Feb 22 13:14:57 2002
From: waldo at saclant.nato.int (wwaldo)
Date: 22 Feb 2002 10:14:57 -0800
Subject: Python 2.1 and PIL
References: <94id8.563$s03.36009@news2.east.cox.net> <3C76654C.9010002@thinkware.se>
Message-ID: <cfabd740.0202221014.5b410cd6@posting.google.com>

Redhat 7.2.  It installs pythong to

/usr/lib/python1.5
/usr/lib/python2.1

Magnus Lyck? <magnus at thinkware.se> wrote in message news:<3C76654C.9010002 at thinkware.se>...
> What OS are you using? (For windows there are binaries for
> the different Python versions.)
> 
> william wrote:
> 
> > Ok been playing with this python 2.1.2 and PIL.  The problem is when i
> > install the PIL it installs it for python1.5.  I don't know how to get the
> > PIL to install for Python2.1.
> > 
> > I am assuming I need to make some changes in some files before i do
> > ./configure.  Any suggestions?
> > 
> > Thanks.
> > 
> > 
> >


From andrew.gregory at npl.co.uk  Fri Feb 15 09:29:31 2002
From: andrew.gregory at npl.co.uk (Andrew Gregory)
Date: 15 Feb 2002 06:29:31 -0800
Subject: COFF2OMF C++ Builder 4
Message-ID: <2830c89c.0202150629.5a6b711f@posting.google.com>

Trying to create BCB4 DLLs which can be imported into Python 2.2
COFF2OMF python22.lib new.lib gave me error message "Invalid COFF file".
What is the easiest way to fix this?

Thanks, Andrew.


From bokr at oz.net  Sun Feb 17 18:00:41 2002
From: bokr at oz.net (Bengt Richter)
Date: 17 Feb 2002 23:00:41 GMT
Subject: Ascii Art
References: <mailman.1012327189.31750.python-list@python.org> <585.799T42T11732924threeseas@earthlink.net> <slrna5rqpa.19d.philh@comuno.freeserve.co.uk> <a3ltbd$3oi$1@verence.demon.co.uk> <Xns91B88FF6B244ERASXnewsDFE1@130.133.1.4>
Message-ID: <a4pcmp$hdv$0@216.39.172.122>

On 17 Feb 2002 22:09:20 GMT, Philip Swartzleonard <starx at pacbell.net> wrote:

>Paul Wright || Mon 04 Feb 2002 04:03:57a:
>
>>                       ______
>>                      /      \
>>                    .' PLEASE `.
>>                    |  DO NOT  |      _____
>>                    | FEED THE |    ,'.....`.
>>                    `. LOONY  ,'  ,'........ )
>>                      \_    _/   |........ ,'
>>                        |  |     `. .... _/
>>                        |  |      ,'.,'-'
>>                        |  |     /../
>>                        |  |   ,'.,'
>>                        |  |  /../
>>                      . |  | /..'
>>                    .\_\|  |/_/,
>>                    ___ |  | ___
>>                      . `--' .
>>                       .    .
>> 
>> It was fun at first but now it's tiresome. 
>> 
>
>I like it, even if it's uncertain weather it's a sign or a Paddle of 
>Punishment =)
>
The weather can't be that uncertain -- I see a shadow ;-)

Regards,
Bengt Richter



From maxm at mxm.dk  Fri Feb 22 13:36:05 2002
From: maxm at mxm.dk (Max M)
Date: Fri, 22 Feb 2002 18:36:05 +0000
Subject: Stackless: 1,000,000 Switches in 0.6 seconds
References: <mailman.1014390936.27277.python-list@python.org>
Message-ID: <3C768F95.5050107@mxm.dk>

Christian Tismer wrote:
 > Sorry, I can't hold it back.
 >
 > Not only that tasklets are smaller than expected, they
 > are also really fast.

No worry ... just keep it comming ...

I'll bet that no one will mind if they become even smaller and faster.

regards Max M



From piet at cs.uu.nl  Fri Feb  8 16:14:41 2002
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 08 Feb 2002 22:14:41 +0100
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <%vw88.94898$KM2.3445519@atlpnn01.usenetserver.com> <slrna6649e.d87.philh@comuno.freeserve.co.uk> <NkQ88.105036$KM2.4008630@atlpnn01.usenetserver.com>
Message-ID: <wzeljvy8u6.fsf@nono.cs.uu.nl>

>>>>> "Steve Holden" <sholden at holdenweb.com> (SH) writes:
 
SH> Extremely iffy when you realise that many X terminal users may run
SH> their web clients on the same CPU, not to mention Lynx users on
SH> timesharing terminals.

Or people behind NAT routers.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From justin at iago.org  Sat Feb 23 12:17:42 2002
From: justin at iago.org (Justin Sheehy)
Date: Sat, 23 Feb 2002 12:17:42 -0500
Subject: Python and Daemons
In-Reply-To: <7652139e.0202222105.1220442c@posting.google.com> (mrchameleon@hotmail.com's
 message of "22 Feb 2002 21:05:35 -0800")
References: <02022215115601.01886@localhost.localdomain>
	<mailman.1014416559.28231.python-list@python.org>
	<7652139e.0202222105.1220442c@posting.google.com>
Message-ID: <x7sn7syv61.fsf@ra.iago.org>

mrchameleon at hotmail.com (Chris Reay) writes:

>> This is what I use:
>> 
> ... and an attribution to W. Richard Stevens is also considered good mojo.

True enough.  Everyone who is going to do any serious network
programming or much of anything on UNIX should read his books.

A few of his idioms have become so standard that I sometimes
forget to make the attribution explicit.

Thank you, Richard.  You are missed.

-Justin

 




From ae98096 at aero.iitm.ernet.in  Sat Feb 16 13:31:13 2002
From: ae98096 at aero.iitm.ernet.in (decay)
Date: 16 Feb 2002 10:31:13 -0800
Subject: Embedding python with lcc-win32
Message-ID: <335764c1.0202161031.2b05e236@posting.google.com>

Hello.

I've been trying to embed python in a C program. I use the lcc-win32
compiler system. I made an import library from python21.dll using
implib. But when I link a simple program, I get "Undefined reference
to __imp__Py_Initialize" in test.c. I suppose its got to do with
making the import library. Can anyone please enlighten me?

Thanks,
Ashok.


From phr-n2002a at nightsong.com  Mon Feb  4 18:40:58 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 15:40:58 -0800
Subject: random generator
References: <lfE78.14336$tg4.165131@vixen.cso.uiuc.edu> <7xg04gsvgh.fsf@ruckus.brouhaha.com> <LvE78.14342$tg4.165028@vixen.cso.uiuc.edu>
Message-ID: <7xu1sw3jed.fsf@ruckus.brouhaha.com>

"John" <nobody at nobody.com> writes:
> Any will do, e.g., uniform, normal, etc.

Those are symmetric distributions.  If I understand what skew is,
they don't have any skew.


From com-nospam at ccraig.org  Wed Feb 27 12:43:22 2002
From: com-nospam at ccraig.org (Christopher A. Craig)
Date: 27 Feb 2002 12:43:22 -0500
Subject: Find an Item in a Sorted List
In-Reply-To: <3c7c8b74.63255807@news.la.sbcglobal.net>
References: <3c7c1b5b.34558051@news.la.sbcglobal.net>
	<ad052e5c.0202261950.72991328@posting.google.com>
	<3c7c8b74.63255807@news.la.sbcglobal.net>
Message-ID: <t1wk7syzupx.fsf@kermit.wreck.org>

hankc at nospam.com (HankC) writes:

> Thanks all, for the help.  I have one more question regarding resource
> management.  My altered version of the above code:
> 
> dict = {}
> datafile = open(..., "r")
> datalines = datafile.readlines()
> for line in datalines:
>    name = line.split(",")[0]
>    dict[name] = line[:-1] # strip trailing \n
> ###
> datafile.close()
> del(datalines)
> 
> At the ### mark, I looks like have have two versions of the data
> resource - one list and one dictionary.  My questions:

If you have a sufficiently recent Python (2.0 should work) you can use

datafile = open(..., "r")
for line in datafile.xreadlines():
  .
  .
  .

Which will do the same thing, but without the memory consumption.
This new code makes a sort of iterator by creating an object where
each successive list access returns the next line.

> - does the del(datalines) call free the memory (not necessarily
> immediately) for the list resource?

It frees it immediately.

> - does the original version of the function, by combining readlines()
> with the for statement, negate the need for freeing the list?  In
> other words, will the datafile.close() statement in itself do the
> trick?

Yes, and no.

The list is read into a temporary variable, which will be freed when
it goes out of scope (I believe that is at the end of the for loop,
but I'm not sure.)  The datafile.close() call has no effect on said
temporary.  

> I appreciate the patience!  Things like garbage collection and for
> loops iterating through lists take a little getting used to.

Python only does garbage collection on cyclic objects most things are
collected by reference counts.  When something is deleted or goes out
of scope the reference count on it is reduced, when the count reaches
zero it is freed immediately.

The problem with this, and the reason that Python does garbage
collection, is that something like

a = []
a.append(a)
del(a)

has increased the reference count on 'a' twice, but only subtracted
one, so you now have a free object which garbage collection will
eventually catch (if it is turned on).

-- 
Christopher A. Craig <com-nospam at ccraig.org>
"Microprocessors cost $700 -- far too much for a tiny slice of refined,
 impurity-laced beach sand." Steve Gibson



From chris.gonnerman at newcenturycomputers.net  Sun Feb  3 16:34:39 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 3 Feb 2002 15:34:39 -0600
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <mailman.1012678565.11760.python-list@python.org> <a3k9am$p0c$1@newsreader.mailgate.org>
Message-ID: <001a01c1acfa$96c6d8a0$0101010a@local>

----- Original Message ----- 
From: "Mark Hadfield" <m.hadfield at niwa.co.nz>
> 
> Who *are* the more intelligent of you?
> 

Looking for a flamewar?  :-)

Anyone graced by being accused of being a robot would have to qualify;
other than that, who can say?





From tim at worthy.demon.co.uk  Mon Feb 11 05:40:32 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Mon, 11 Feb 2002 10:40:32 GMT
Subject: Long int to int
Message-ID: <dec413074b.tim@worthy.demon.co.uk>

I need to convert some long int values to normal ints.
(4 byte words)

e.g. a=0x81234567L

(masked from a longer int e.g. 0xAA81234567L hence as a L)

b=int(a) works if a is less than 0x7ff..

If greater (i.e. 2's complement negative int) an error occurs.

I can convert by shifting bits, converting to int and shifting back.

a=int(a & 0xFFFF) | int(a>>16 & 0xFFFF)<<16

Is there not a neater, more correct way to go from

0x81234567L
to
0x81234567    ?


(Or worse, I can convert with eval

a = str(hex(a))[2:-1]
a = eval('0x'+a)

works but

int (str(hex(a))[2:-1],16) fails.)


-- 
___
 |im    ---- ARM Powered ----


From usenet at wojas.vvtp.tudelft.nl  Sun Feb 24 21:42:11 2002
From: usenet at wojas.vvtp.tudelft.nl (K. Wojas)
Date: Mon, 25 Feb 2002 03:42:11 +0100
Subject: GUI (PyQt) & thread problem
References: <8b5d764c.0202240451.1c4bda3b@posting.google.com>
Message-ID: <a5c8a2$r5e$1@news.tudelft.nl>

Kristian Nylund wrote:
> Can someone tell me what the problem is with the following program, as it
> hangs. I'm pretty sure there's some inherent design-flaw involved, but
> I'm not familiar enough with threads or GUI-programming to know what it
> could be.
> 
> I've also tried modyfing the program to use PYSIGNALs instead of passing
> the caller to the thread, but the results were the same.

Never use the Qt toolkit from 2 threads. You can only use the worker thread 
for non-GUI stuff. The Qt toolkit isn't threadsafe.

Regards,

Konrad Wojas


From chris_mk at hotmail.com  Fri Feb  8 17:51:55 2002
From: chris_mk at hotmail.com (Christopher)
Date: 8 Feb 2002 14:51:55 -0800
Subject: cgi POST question
References: <cc14131d.0202071230.415f2447@posting.google.com> <M6C88.96519$KM2.3594308@atlpnn01.usenetserver.com>
Message-ID: <cc14131d.0202081451.36e8d54e@posting.google.com>

Thanks for the great information.  I must admit, I felt like
embarrassed that I didn't see the POST information in urllib.  Then I
realized why:

     I looked through the urllib module BEFORE i realized it was a
POST vs. GET issue.  Had I gone back, it would have been crystal
clear, but, in my mind, I had looked inside and out at the urllib
module so there was no point.  I probably remembered the dictionary
sorting method of urlencode, which is how I thought up making a
dictionary (it seemed a little to easy for me to come up with, now I
know why).  Anyway, for the most part, it works now, there are just a
few little details I need to work out.  Thanks again.

Chris


"Steve Holden" <sholden at holdenweb.com> wrote in message news:<M6C88.96519$KM2.3594308 at atlpnn01.usenetserver.com>...


> You are writing client-side programs, and wanting to retrieve data from the
> intranet web server, correct? The urllib library *is* the one you want ...
> see below.
> 
> As far as the intranet search engine guy goes, his behavior betrays a
> fragile ego and low self-esteem typical of people who aren't really
> comfortable doing what they are doing. The good news is you can ignore hom
> from here on in.
> 
> It's likely he *could* have written his code to work with *either* POST or
> GET, but many people don't know the appropriate techniques. Ironically, in
> ASP and Cold Fusion it's actually *easier* to write CGI scripts this way!
> 
> You are so close you clearly understand Python quite well. See below ...
> 
> >      I did have my boss approve the purchases of a couple networking
> > books and "Python Web Programming" by Steve Holden (<= very good so
> 
> ... and you a clearly a person of discrimination :-) Glad you're enjoying
> the book ...
> 
> Here's an exceprt from the urllib docs:
> 
> """
> urlopen (url[, data])
> ...
> If the url uses the http: scheme identifier, the optional data argument may
> be given to specify a POST request (normally the request type is GET). The
> data argument must in standard application/x-www-form-urlencoded format; see
> the urlencode() function below.
> """
> 
> In other words, as you surmised, you can fill a dictionary with the data you
> want to POST. If you pass the dictionary as a second argument to
> urllib.urlopen(), you do a POST.
> 
> I'd write you some sample code, except you sound like you're quite capable
> of taking it from here. Let us know how it works out.
> 
> regards
>  Steve


From justin at iago.org  Wed Feb 20 12:12:10 2002
From: justin at iago.org (Justin Sheehy)
Date: Wed, 20 Feb 2002 12:12:10 -0500
Subject: Adding static typing to Python
In-Reply-To: <a50g0n$gt4$1@thorium.cix.co.uk> (gbreed@cix.compulink.co.uk's
 message of "Wed, 20 Feb 2002 15:40:07 +0000 (UTC)")
References: <mailman.1014147452.26214.python-list@python.org>
	<a50g0n$gt4$1@thorium.cix.co.uk>
Message-ID: <x7n0y45b85.fsf@ra.iago.org>

gbreed at cix.compulink.co.uk writes:

>> It also has dynamic typing, as opposed to static typing.  This seems
>> to be the thing that you're objecting to.  

> Do you have a reference for this terminology?  I haven't seen it outside 
> of comp.lang.python. 

Those terms are used fairly often in programming language
discussions.  Many languages use the static/dynamic distinction to
describe their own typing system.

Most university courses on the principles of programming languages
(and decent compilers courses that cover typechecking) talk about the
static/dynamic distinction.

>From the canonical or otherwise respectable web pages about a few
programming languages (not Python-related):

"Eiffel is statically typed to ensure that errors are caught at compile
time, not run time."

"A Java compiler enforces a syntactic discipline on program text
called static typing."

"Many languages provide dynamic typing: Smalltalk, Self, Objective-C,
and etc.  A limited dynamic typing scheme, called RTTI (Run Time Type
Identification), is even being considered for the C++ standard."

"Smalltalk is an example of a dynamically-typed language."

> The only definition I have to hand contradicts it:
> <http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=strong+typing>

So, if we're just going to pick definitions off the web, here's one
that appeared rather quickly on google:

(http://www-lp.doc.ic.ac.uk/UserPages/staff/ft/alp/net/typing/strong.html)

  Strongly typed: A strongly typed program cannot 'go wrong' (in
  Milner's words) - that is, dump core, produce a wrong answer or
  similar stuff. Note that runtime errors are still possible, but are
  clearly marked as such. (Such as accessing arrays out of the declared
  bounds, which is seldom statically checked.) (Examples: Lisp, SML,
  Goedel, Prolog.)

  Weakly typed: No such guarantee is given. (Example: C.)

  Statically typed: Type checking is done at compile-time. (SML, C, Goedel)

  Dynamically typed: Type checking is done at runtime. (Lisp, Prolog)
  Someone should probably define polymorphism (esp. parametric versus ad
  hoc in all its varieties) as well.

A common interpretation/explanation of this is that dynamic typing
assigns types to values, while static typing assigns types to
variables.  This is in line with my statement about Python's strong
yet dynamic typing.

> Taking "static typing" literally would probably mean that an object 
> couldn't change it's type after creation.  That's almost always the case 
> in Python.  If you mean identifiers have to stay bound to the same type, 
> it'd mean
>
> x = 1
> x *= 5L
>
> would be illegal, because x begins as an integer and then becomes a long.  

This shows that you don't yet understand Python's assignment rules.
In the above, no object is changing type.  The following is what happens:

 - An object of type integer is created.
 - The name "x" is bound to that object.
 - An object of type long is created.
 - The name "x" is bound to that object.

The binding of names to objects is completely independent from the
values of the objects.  Objects have types in dynamically-typed
languages, but names may not.

> So, um, what actually is the problem?

I have no idea.  I'm fairly comfortable with dynamically-typed
lanaguages, so I haven't seen a problem here yet.  <wink>

-Justin

 




From michael at stroeder.com  Sat Feb 16 07:37:04 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sat, 16 Feb 2002 13:37:04 +0100
Subject: [Python-Dev] PEP 282: A Logging System  --  comments please
References: <20020215164333.A31903@ActiveState.com> <mailman.1013828010.19562.python-list@python.org>
Message-ID: <3C6E5270.D434E56A@stroeder.com>

Michael McLay wrote:
> 
> Marshal Rose submitted RFC3195[1] to the IETF for a syslog protocol.  The
> specification is defined as a profile on top of the BEEP framework.  The
> format of the messages are encoded in XML. Here is an example of an "entry"
> element.
> 
>      C: <entry facility='8' severity='6'
>       C:   hostname='pipeworks'
>       C:   timestamp='Oct 31 23:59:59'
>       C:  >&lt;.....eeeek!</entry>

I really wonder why the timestamp is of such an unstructured format
although the log entry seems to comply to a specific XML schema. Why
the hell don't people simply start using ISO 8601 for date and time?

I also wonder which character set is used or if that's defined by a
XML attribute.

Ciao, Michael.


From bokr at oz.net  Wed Feb 20 00:10:19 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 05:10:19 GMT
Subject: execfile question
References: <mailman.1014155372.25311.python-list@python.org>
Message-ID: <a4vb3r$7hd$0@216.39.172.122>

On Tue, 19 Feb 2002 16:48:39 -0500 (EST), Steven Majewski <sdm7g at Virginia.EDU> wrote:

>
>PS:
> You can get the behaviour you were expecting if you pass an
>unbound anonymous dict to execfile -- the namespace is
>garbage that's released after execfile returns:
>
>Python 2.2 (#36, Jan 30 2002, 17:50:25)
>[GCC 2.95.2 19991024 (release)] on darwin
>Type "help", "copyright", "credits" or "license" for more information.
>>>> execfile( 'executable.py', {}, {} )
>it's me
>init
>deleted
>>>>
>
Hm. I didn't try it with two {}'s. Why shouldn't one work? Shouldn't it
just use the same for globals and locals?

This "works"
 >>> xdict={}
 >>> execfile('executable.py',xdict)
 it's me
 init
 >>> xdict.clear()
 deleted

but xdict going out of scope doesn't do it
and neither does del xdict. Shouldn't they?

Regards,
Bengt Richter
We should really stay in one thread on this;-/




From jeff at ccvcorp.com  Wed Feb 27 15:06:40 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 12:06:40 -0800
Subject: VPython--close causes crash!
References: <mailman.1014702162.21756.python-list@python.org>
Message-ID: <3C7D3C50.2CCDB902@ccvcorp.com>


chajadan wrote:

> It's working fine with VPython's Idle, though I haven't tried just good
> plain Idle (don't use it much). I wonder what pythonwin's prob is?
>
> --chajadan
>
> At 12:18 AM 2/26/02 -0800, chajadan wrote:
> >I downloaded VPython, and PyGeo, but whenever I close the VPython
> >generated windows, it crashes all of Pythonwin.
> >
> >Maybe this is a pythonwin problem, I'll try Idle--but anyone out there
> >have any ideas?

I'm not specifically familiar with VPython, but it's known that most GUI
toolkits have trouble cooperating with each other.  Tkinter and wxPython
code will not run properly inside of *either* IDLE or PythonWin, under
normal circumstances.  (It's possible to coerce Tkinter code to work inside
of IDLE, since IDLE *is* Tkinter code, but it takes some effort.)  I
suspect that it's this sort of toolkit-incompatibility that's biting you,
here, and that VPython's version of IDLE is modified to allow VPython
windows to work properly.

Jeff Shannon
Technician/Programmer
Credit International




From quinn at hork.ugcs.caltech.edu  Sat Feb  9 22:11:44 2002
From: quinn at hork.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 03:11:44 GMT
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com>	<3c62a989.3448437947@news>	<a3ucsk$12h8$1@nntp6.u.washington.edu>	<a41c4m$bf4$1@newshost.accu.uu.nl> <mailman.1013203955.3594.python-list@python.org> <2df3d89d.0202091355.222422d1@posting.google.com>
Message-ID: <slrna6bp7g.9tv.quinn@hork.ugcs.caltech.edu>

On 9 Feb 2002 13:55:04 -0800, Jeff Grimmett <grimmtoothtoo at yahoo.com> wrote:
>Cliff Wells <logiplexsoftware at earthlink.net> wrote in message
news:<mailman.1013203955.3594.python-list at python.org>...
>
>> Arguing with irrational oddballs on usenet is today's equivalent of the
>> Rubik's Cube.
>
>Only thing is, there is an actual solution to Rubik's Cube.

Yeah, you pull off all the little nodules and stick 'em back on in the right
place.  Don't see why everyone got so excited about those.

Unfortunately the last time I tried that with a person a lot of busybodies got
in a snit and said something about having to go to some kind of school just
to do that stuff.  Whatever.


From martin at v.loewis.de  Sat Feb 16 04:47:56 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 10:47:56 +0100
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net>
Message-ID: <m34rkhbvw3.fsf@mira.informatik.hu-berlin.de>

Hans Nowak <wurmy at earthlink.net> writes:

> I manage to store that string just fine; only when I retrieve
> the record, all the newlines are gone.

I can hardly believe this. Can you post a small script that
demonstrates this problem?

Regards,
Martin



From tripps81 at yahoo.com  Mon Feb 25 11:56:54 2002
From: tripps81 at yahoo.com (Tripp Scott)
Date: Mon, 25 Feb 2002 23:56:54 +0700
Subject: Implementing interfaces with inheritance, good or bad?
Message-ID: <5.1.0.14.2.20020225234807.0407db58@202.95.131.154>

I just saw:

  http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/68445

in which interfaces are implemented using class, and a class 
implementing one or more interfaces subclasses the interfaces 
class(es). Now suppose class C implements two Interfaces, I1 and 
I2. Both interfaces define an operation foo(). In one case, I 
want to instantiate C because of I1, and in other case, because 
I need a class that implements I2. How does C implement both 
foo() operations (since foo() will be implemented as a method)? 
And how do I tell C that I want I1, or I2? With an argument to 
__init__? Any other way?

I prefer providing Interface lookup via other mechanism, like 
with the __implements__ class attribute as done in Zope's 
Interface module. But are there any advantages of interfaces 
being done with inheritance?

t




From gerhard at bigfoot.de  Mon Feb 18 13:33:44 2002
From: gerhard at bigfoot.de (Gerhard =?iso-8859-15?Q?H=E4ring?=)
Date: 18 Feb 2002 18:33:44 GMT
Subject: PyGres fetchone problem
References: <695dd93e.0202181020.3c1cb4c5@posting.google.com>
Message-ID: <slrna72i89.7u6.gerhard@lilith.my-fqdn.de>

Richard Papworth wrote in comp.lang.python:
> Playing around with the PyGres module I came across the following
> problem when using fetchone to iterate over a cursor, an idiom that
> I've seen used in many examples:
>
> [cursor.fetchone() doesn't work]
> 
> Am I doing something wrong here?

Yes.

1) Using PyGreSQL ;-)

2) Using an old version of PyGreSQL

Get PyGreSQL out of PostgreSQL [1] 7.1.3 or, even better, 7.2. I fixed
this exact bug a few months ago.

[1] Yes, that's correct. The right place to get PyGreSQL is out of the
PostgreSQL sources.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))


From mdk at mdk.com  Tue Feb  5 13:46:58 2002
From: mdk at mdk.com (MDK)
Date: Tue, 5 Feb 2002 13:46:58 -0500
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de> <a3p2fg$49g$1@mail1.wg.waii.com>
Message-ID: <a3p9b6$1a7qtb$1@ID-98166.news.dfncis.de>

"Martin Franklin" <martin.franklin at westgeo.com> wrote in message
news:a3p2fg$49g$1 at mail1.wg.waii.com...
> MDK wrote:
>
> >
> > "Emile van Sebille" <emile at fenx.com> wrote in message
> > news:a3p0ss$19ob7s$1 at ID-11957.news.dfncis.de...
> >>
> >> "MDK" <mdk at mdk.com> wrote in message
> >> news:a3p01b$186eq8$1 at ID-98166.news.dfncis.de...
> >> > Hello,
> >> >
> >> > I am trying to create my own special type of list object.
> >> >
> >> > However, when I do x.append('yo') it uses Python's append instead of
> >> the one
> >> > from my class.
> >> >
> >> > How can I get it to use my append method?  I have tried def
__append__
> >> and
> >> > then called my own class' function but that did not work.  I've
played
> >> with
> >> > __getattr__ but that did not help.
> >> >
> >>
> >> Call the method append and not __append__?
> >>
> >>
> >> --
> >>
> >> Emile van Sebille
> >> emile at fenx.com
> >
> > Yes, I figured that to be the case because it should be transparant to
the
> > caller.  However, that is not the solution.
> >
> > Thanks though.
> >
> >
> >
>
> Post some code.  perhaps the answer is obvious...
>
>
> for example
>
>
> >>> class MyList:
> ...     def __init__(self):
> ...             self.list=[]
> ...     def append(self, item):
> ...             self.list.append(item)
> ...
> >>> ml=MyList()
> >>> ml.append(1)
> >>> ml.list
> [1]
> >>>
>
> would apear to work?

I made a file mylist.py

# mylist.py
class MyList:
    def __init__(self):
        self.list=[]
    def append(self, item):
        self.list.append(item)

>>> import mylist
>>> x = mylist()
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
TypeError: object of type 'module' is not call

Why is this happening?  My list object must be a class in its own,
importable file.




From troels at fast.no  Thu Feb  7 06:03:57 2002
From: troels at fast.no (Troels Walsted Hansen)
Date: Thu, 7 Feb 2002 12:03:57 +0100
Subject: Python method reference kills garbage collection?
Message-ID: <a3tmuv$qm1$1@news.ost.eltele.no>

Consider the following program. What is so special about references to the
object's own methods that cause the object to never be garbage collected? Is
this a cyclic reference problem, and if so why doesn't the cycle detector
(--with-cycle-gc) cure the problem?

class a:
    def __init__(self):
        self.f = self.f1
    def __del__(self):
        print "this is never printed"
    def f1(self):
        pass
b=a()
b=a()

--
Troels Walsted Hansen M.Eng.       Mail:  troels at fast.no
Software Engineer                  Web:   http://www.fast.no/
Fast Search & Transfer ASA

Try FAST Search: http://alltheweb.com





From infinitystwin.SPAM at IS.BAD.yahoo.com  Fri Feb  8 03:48:44 2002
From: infinitystwin.SPAM at IS.BAD.yahoo.com (Greg Krohn)
Date: Fri, 8 Feb 2002 02:48:44 -0600
Subject: Webizing Python
Message-ID: <a4032v01prd@enews3.newsguy.com>

I don't remember any threads about this, so please forgive me if there has
been.

In Guido's "The State of the Python Union"
(http://www.python.org/doc/essays/ppt/python10/py10keynote.ppt) he mentions
webizing Python. In particular adding http:/www.python.org/python2.2 to
sys.path. I would think this would probably be P3K, but does anyone have any
info on this yet? It sounds interesting.

And seeing as I have very little information, I think it would be a good
time to start criticizing. I wonder what would happen if something occurred
at python.org like hijackers or some honest mistake or accident. I would be
putting a lot of faith in the online modules' integrity. I can be relatively
certain that a program I run right now will run the same in the morning (if
no one touches my computer, of course). This wouldn't be the case if Python
were webized. For me, at least.

greg




From aseem at neurobs.com  Tue Feb 12 15:25:47 2002
From: aseem at neurobs.com (A M)
Date: 12 Feb 2002 12:25:47 -0800
Subject: cgi and mime type
Message-ID: <ed6d1c1a.0202121225.295ca950@posting.google.com>

Hi,
I have a simple cgi script that sends back an exe when a button is
pressed..

I am using the following code to do that : 

print 'Content-type: application/octet-stream'
print 'Content-disposition: application; filename="app.exe";
print;

after which I open the file and print it:

fp = open('fp.exe')
print fp.read();

Now when testing it in the browsers NS and IE, because of the content
disposition header IE prompts you to download it with the name in the
header.

However in NS it refuses to recognize the MIME type and prompts that
you are attempting to download "text/plain" script.py

However if i save it to disk: bacuase of the content disposition it
saves the file as app.exe

why doesnt NS recognize my mime type. its a predefined one as far as I
know. Am I doing something wrong in python.

The test URL is : http://66.121.5.70/links.html

Thanks in advance,
A


From bobgreaves at windowsblueyonder.co.uk  Thu Feb  7 10:11:46 2002
From: bobgreaves at windowsblueyonder.co.uk (gnutter)
Date: Thu, 07 Feb 2002 15:11:46 +0000
Subject: dependency hell
References: <xIw88.500$9%6.51706@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <1013094698.15749.0.nnrp-02.3e318052@news.demon.co.uk>

In article <xIw88.500$9%6.51706 at bgtnsc04-news.ops.worldnet.att.net>, "Tom
F." <tom at 8dynamics9.com> wrote:

> I'm trying to install python2-2.2-2. I apparently need the following
> files:
> 
> libcrypto.so.2, libreadline.so.4, libssl.so.2
> 
> As far as I can search, libopenssIO and readline are supposed to supply
> them.
>  I can even find libreadline.so.4  specificly on my system (but not the
> others).
> 
> I've done many RPM searches and more, to no avail. could someone perhaps
> point me inthe right direction?
> 


Could it be that your LD_LIBRARY_PATH is pointing to the wrong place?


-- 
Please excuse my English, I'm from the UK.

~~~~ Delete Windows to reply ~~~~~


From gh_pythonlist at gmx.de  Tue Feb 19 08:14:19 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 14:14:19 +0100
Subject: ODBC on unix w/o mxODBC
In-Reply-To: <7x7kp9zkuw.fsf@ruckus.brouhaha.com>
References: <yoxadu7xhk1.fsf@karhu.tp.spt.fi> <a4r1ch$ghd$1@norfair.nerim.net> <yoxr8niyiyf.fsf@karhu.tp.spt.fi> <23891c90.0202190434.147ee6d1@posting.google.com> <7x7kp9zkuw.fsf@ruckus.brouhaha.com>
Message-ID: <20020219131418.GA4035@lilith.hqd-internal>

Le 19/02/02 ? 05:01, Paul Rubin ?crivit:
> paul at boddie.net (Paul Boddie) writes:
> > You could always make a SWIG wrapper for iODBC - I started one to see
> > how hard it might be - but in the end it's probably cheaper (time is
> > money, so they say) to license mxODBC.
> 
> I hadn't heard of iODBC before, but just found the home page www.iodbc.org
> and it mentions that a Python bridge (whatever that is) already exists.

SQLRelay?

http://www.firstworks.com/sqlrelay.html

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 5.9 ?C      Wind: 9.8 m/s



From alife at narod.ru  Mon Feb 18 05:51:02 2002
From: alife at narod.ru (Serge Terekhoff)
Date: 18 Feb 2002 02:51:02 -0800
Subject: win32com and MS Excel
References: <mailman.1013967115.15656.python-list@python.org>
Message-ID: <98322540.0202180251.2ddf81f4@posting.google.com>

Dear all!

on
from win32com.client import Dispatch

My Python2.2 replies:

Traceback (most recent call last):
  File "F:\Users\Serge\MyPy\to_excel.py", line 1, in ?
    from win32com.client import Dispatch
ImportError: No module named win32com.client


What i did wrong?

Thank you
Serge





"Jimmy Retzlaff" <jimmy at retzlaff.com> wrote in message news:<mailman.1013967115.15656.python-list at python.org>...
> > One more question: what is going to happen when I build a standalone 
> > (using py2exe)? Will the generated file be included and affect the
>  speed 
> > on another machine too?
> 
> There are a couple of options. There is a py2exe option to include
> everything makepy has ever generated on the build machine, see:
> 
> http://groups.google.com/groups?hl=en&selm=9e1a14%24ilo5%241%40ID-598
> 85.
> news.dfncis.de
> 
> The approach I've used is as follows:
> 
> - Find the file generated by makepy (it's in win32com\gen py)
> - Copy that to your application directory and give it a readable name
> (e.g., Excel.py)
> - Change these two lines:
> 
> from win32com.client import Dispatch
> xlApp = Dispatch("Excel.Application")
> 
> to:
> 
> import Excel
> xlApp = Excel.Application()
> 
> 
> Jimmy


From carlca at dircon.co.uk  Sun Feb  3 10:00:03 2002
From: carlca at dircon.co.uk (Carl Caulkett)
Date: Sun, 3 Feb 2002 15:00:03 -0000
Subject: nice to meet you
References: <aR978.5215$3E5.412091@newsread2.prod.itd.earthlink.net>
Message-ID: <MPG.16c760cdaa8ee3639896d5@newnews.dircon.co.uk>

In article <aR978.5215$3E5.412091 at newsread2.prod.itd.earthlink.net>, 
whoRunBartertown at masterBlaster.net says...
> Hi and hello all, I'm the latest Python convert... 2 months and counting.
> 
> I LOVE IT.
> 
> You'll see me around.

Cool! Out of interest, what language(s) did you use before?

-- 
Cheers,
Carl


From maxm at mxm.dk  Wed Feb  6 15:33:50 2002
From: maxm at mxm.dk (Max M)
Date: Wed, 06 Feb 2002 20:33:50 +0000
Subject: What is the raison d'etre for Special/Magic Methods
References: <8852d33c.0202061038.3b53b455@posting.google.com>
Message-ID: <3C61932E.2070906@mxm.dk>

wooks wrote:
> I know there is a section on this at www.diveintopython.org but the
> site has been down all afternoon.

Usually it is to create a "hook" into Python syntax.

ie. if you want to get content from a dictionary object you do:

result = someDict['someKey']

The somedict then will have a "__getitem__" method that returns "result".

regards Max M



From bokr at oz.net  Mon Feb 11 00:33:55 2002
From: bokr at oz.net (Bengt Richter)
Date: 11 Feb 2002 05:33:55 GMT
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <mailman.1013382318.2151.python-list@python.org>
Message-ID: <a47l43$mj8$0@216.39.172.122>

On Sun, 10 Feb 2002 18:05:01 -0500, "Tim Peters" <tim.one at home.com> wrote:

>[Bengt Richter]
>>  >>> 'abc'.split('')
>>  Traceback (most recent call last):
>>    File "<stdin>", line 1, in ?
>>  ValueError: empty separator
>>
>> Wouldn't it make sense to return list('abc') ?
>
>[Neil Schemenauer]
>> It would also make sense to return ['a', 'b', 'c'].
>
>Well, that's what list('abc') does return, so you're in violent agreement.
>However, this still stands:
>
>> Since it's not obvious what you want Python raises an error.
>
>Indeed, my first thought was "OK, if it has to return *something*, then
>since we're asking it to split on nothing, it shouldn't split at all:
>
>    ['abc']
>
>is what it should return."  That's also compatible with
>
>    ''.join[['abc']] == 'abc'
Of course you mean
 >>> ''.join(['abc'])
 'abc'

This option is also compatible with what re does:
 >>> import re
 >>> re.split('','abc')
 ['abc']
 
(Note absence of complaint ;-)

>
>People who think '' should split "everywhere" instead of "nowhere" should
>really be arguing for 'abc'.split('') to return
>
>    ['', 'a', 'b', 'c', '']
>
>instead, since in any sense that '' could be said to "match", it matches at
>4 slice positions in 'abc', not 2.
>
Which is what I was trying to get at with:
--
or, ugly, but consistent with *find results
(finding the separator front, back, and between all):

 >>> ';a;b;c;'.split(';')
 ['', 'a', 'b', 'c', '']
--
I.e., return the latter value, as you say.
Which also joins ok-ly:

 >>> ''.join(['', 'a', 'b', 'c', ''])
 'abc'


OTOH:

 [21:33] C:\pywk>perl "-e print join('-', split(//, 'abc'))"
 a-b-c

Not sure which way that'll throw things ;-)

Regards,
Bengt Richter



From deathtospam43423 at altavista.com  Tue Feb 19 06:00:19 2002
From: deathtospam43423 at altavista.com (Flavian Hardcastle)
Date: 19 Feb 2002 11:00:19 GMT
Subject: newbie: Some file object tips?
Message-ID: <Xns91BADAC5A5CEFAusYourStandingInIt@130.133.1.4>

I'm writing a crude cgi chat script. I've already got it up and working 
with Xitami server, and have chatted with a few net buddies.

How it works is very rough and simple... it basically just takes data from 
people's browsers (using the FieldStorage function) and stores it in a file 
called log.txt, and then displays the contents of that file back to them.

My problem is ... 

The longer you chat, the bigger the log.txt file gets, and consequently the 
web page the chatters are viewing gets bigger and bigger, and becomes hard 
to download.

So I want to include a "scroll back" feature ... which will enable the user 
to control how many previous posts (s)he views. It will probably take the 
form a simple input box on the form ... if you want to see the last 8 
posts, you put 8 in the box, if you want to see the last 50 posts, you put 
50 in the box and so on.

So, I need the script to be able to search the log.txt file. Any tips as to 
how I might go about this? 

cdewin at dingoblue.net.au




From Gareth.McCaughan at pobox.com  Wed Feb 20 15:58:50 2002
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Wed, 20 Feb 2002 20:58:50 +0000
Subject: [OT]: Jabberwocky (was: Re: entries between labels dynamically)
References: <mailman.1014094772.4564.python-list@python.org>
Message-ID: <slrna783ga.abq.Gareth.McCaughan@g.local>

Jason Orendorff wrote:

> Yep, for some reason I was imagining "momeraths" as one word.

Perhaps ypu were thinking of the French version?

    Il brilgue: les toves lubricilleux
    Se gyre en vrillant dans le guave.
    Enmimes sont les gougebosqueux,
    Et le momerade horsgrave.

There are probably lots of mistakes there, and
of course all the accents are missing.

(To whoever it was: "borogoves" is correct and
"borogroves" is wrong, although for some reason
it seems to be difficult to remember this.)

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


From zombor_666 at hotmail.com  Mon Feb  4 14:58:17 2002
From: zombor_666 at hotmail.com (Jeremy Bush)
Date: Mon, 4 Feb 2002 13:58:17 -0600
Subject: heres a dumb question....
Message-ID: <a3mp6v$u04$1@wiscnews.wiscnet.net>

Hello, ive just started fooling around with python, I know C++ pretty well,
and it looks familiar.  The only problem im having is a cant seem to figure
out how to get keyboard input...ive read the whole readme that came with it,
the tutorial, everything.  I just want to know how to do a simple "cin >>
X;"  Can anyone help?




From mertz at gnosis.cx  Fri Feb 15 15:59:16 2002
From: mertz at gnosis.cx (Dr. David Mertz)
Date: Fri, 15 Feb 2002 15:59:16 -0500
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au>
 <a4j5hd$l3j2$1@ID-11957.news.dfncis.de>
 <mailman.1013800706.14636.python-list@python.org> <a4jsk2$q4e$0@216.39.172.122>
Message-ID: <kaXb8kKkX8RF092yn@bellatlantic.net>                         

|>In other words, I don't believe that Brainbench's Python exam
|>is a *good* testing tool.

|What would your thoughts be re an automated exam as a PSF - licensed free
|collaborative project? Maybe even the exam definition itself could somehow
|feed off a wiki-like process, if you constrained formats appropriately
|for defining test elements?

I don't necessarily have anything against testing in general.  It is
possible--at least to a first approximation--to learn something about
people's skill levels by means of tests.  And I wouldn't even go so far
as to suppose that a complete novice would get the same result as Alex
Martelli on the Brainbench exam.  It's not as good as it *should* be, but
it's not completely random either.

Then again, I don't have so much faith in testing to really think making
an exam is the best direction for PSF resources (just think what they
could accomplish if Tim Peters were to sell his second hand copy of _How
to Form a Non-Profit_, and spend the proceeds on something important).
Nonetheless, if someone wants to pay me more money than Brainbench did,
I would be happy to prepare a better exam than Brainbench let me do :-).

Btw. in regard to Cliff's question on this thread:  I can buy a ream of
nice paper at my local office store for about $8.  Assume my printer
costs about $0.03/page to operate.  Rounding up, I would advocate that
most programming-skill certificates are worth EVEN MORE than US$0.05!

Yours, David...

--
---[ to our friends at TLAs (spread the word) ]--------------------------
Echelon North Korea Nazi cracking spy smuggle Columbia fissionable Stego
White Water strategic Clinton Delta Force militia TEMPEST Libya Mossad
---[ Postmodern Enterprises <mertz at gnosis.cx> ]--------------------------




From James_Althoff at i2.com  Mon Feb  4 15:41:02 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Mon, 4 Feb 2002 12:41:02 -0800
Subject: Defending the rodent (was Re: Defending the Python lanuage...)
Message-ID: <OF5A9E8F4A.CE6F11AF-ON88256B56.00711E59@i2.com>

[Jim]
> Of course, if Douglas Englebart had had his way, this would all be moot.
> Englebart -- who invented the computer mouse -- didn't like moving his
hand
> off the keyboard either.  So for his famous NLS system he devised a
> one-handed "chording" keypad on which he typed (all characters) using his
> left hand while manipulating the mouse with his right.  Decades later he
> still maintained that this is *the* way to do it.  :-)

[Jacob]
>Have you ever tried a chording keyboard?  It's absolutely amazing; I can
>type about twice as fast on a chording keyboard as I can on a QWERTY one
>(although I probably couldn't now -- I haven't used one for years).

Yes. We had some at Xerox PARC.

[Jacob]
>Jacob (who now wants to get his circa 1990 serial chording keyboard to
>work with his palm for taking notes in class)

Rock on!  (If nothing else, it should get you some amused looks in the
classroom <wink>).

Jim




From jkraska at san.rr.com  Thu Feb 14 23:10:03 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 15 Feb 2002 04:10:03 GMT
Subject: Using tabs instead of spaces in IDLE?
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com> <mailman.1013655704.3770.python-list@python.org> <0ubm6ugubq49vegd0fhfeogkkcloq827vg@4ax.com> <3C6C680D.B35E7719@cosc.canterbury.ac.nz> <jash4a.sm1.ln@127.0.0.1>
Message-ID: <8m2p6u40evl9k04e8jq14tfvpvsov96df0@4ax.com>

>> This sounds ingenious, but what happens when you save the
>> file and load it back? Does it remember this information
>> somehow, or does it get lost?
>
>The editor remembers this information by saving it in the text file
>itself.

I believe that you're answering a different question than what
he asked. The ability of vim to "backspace" over a current
line and treat those spaces entered as if it were a single
tab is possible on the current *line* being introduced in insert
mode only.

Still, this mode works fairly well, and is about as compatible-
with-Python as you can get. For those who sometimes have to
switch back and forth between modes in different files, they
might like this feature:

## vim: et ts=8 sts=4

Vim reads the "vim:" command in your file and interprets it.
You typically put this at the head of your file, somewhere
near where you do the $Id: $ thingy for CVS.

C//



From ivano.delpiccolo at degroof.be  Mon Feb 11 06:10:20 2002
From: ivano.delpiccolo at degroof.be (DelPiccolo Ivano)
Date: Mon, 11 Feb 2002 12:10:20 +0100
Subject: Beginner problem.
Message-ID: <3c67a6a3$0$12223$4d4efb8e@news.be.uu.net>

How can i know if i have parameters in the command line.

Here's my script, where is the error ?

import sys

if sys.argv[??????] == "" :
    print "Usage : sys.argv[0] logname string_to_search"
else :
   ...........
    ..........
    ..........




From joost_jacob at hotmail.com  Fri Feb  8 17:06:47 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 8 Feb 2002 14:06:47 -0800
Subject: 2001 Turing Award
References: <mailman.1013194813.5888.python-list@python.org>
Message-ID: <13285ea2.0202081406.5032f318@posting.google.com>

> [James_Althoff]
> >
> > Happy reading,
> >
> 
> Click the link "Overview of thumbnail pages" and start reading!
> 

Thank you, happy indeed, it is very interesting and keep in mind it is from 1963 !


From jason at jorendorff.com  Thu Feb 28 12:29:18 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 28 Feb 2002 11:29:18 -0600
Subject: sorting
In-Reply-To: <a5ll4a$4ml$1@animus.fel.iae.nl>
Message-ID: <HFEKILOLEFEFMKAECNDLEEALDGAA.jason@jorendorff.com>

Really cool explanation by Carel, but a few things should
be fixed:

>    >>> numbers = [ real(x) for x in number_strings ]

float(x), not real(x).

>    >>> print "the number %r occured %d time(s)" % (78, count[78]))

Instead of %r, you want %s.  There are lots of different
possible formats (you can specify exactly how many decimal places
you want, etc.) but I think %s is the most friendly.

## Jason Orendorff    http://www.jorendorff.com/



From jkraska at san.rr.com  Tue Feb 26 12:27:59 2002
From: jkraska at san.rr.com (Courageous)
Date: Tue, 26 Feb 2002 17:27:59 GMT
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org> <m3pu2sancq.fsf@chvatal.cbbrowne.com> <eppstein-42B711.20584025022002@news.service.uci.edu> <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547@4ax.com> <a5fp9u$dkd@gap.cco.caltech.edu>
Message-ID: <mshn7uofbug0335eef9cjs279dkra03s69@4ax.com>

>> Clever Linguistic Integration Tricks without Stacks?
>
>Wrong body part.
>
>Tightly Integrated Tricks without Stacks.

He said tightly. Heh heh heh heh heh heh.

Beavis and Butthead meat comp.lang.python.

I declare the world destroyed now. :-)

C//



From pete at shinners.org  Wed Feb  6 11:19:31 2002
From: pete at shinners.org (Pete Shinners)
Date: Wed, 06 Feb 2002 16:19:31 GMT
Subject: Remaining nimble in the face of evolution
References: <3C5EBABF.40555C22@noaa.gov>
Message-ID: <3C6158A2.9060906@shinners.org>

jim.vickroy wrote:
> Now a reality check occurs and the requirement evolves such that the
> archival behavior is to be optional.
> 
> What is the "best" strategy to handle these types of evolution.

first thing that jumps to my mind, use the "getattr" / "hasattr" 
functions to determine if the objects have the 'archiving' methods you need.

if hasattr(myinstance, "archive_method"):
	myinstance.archive_method(file_or_something)


this should work pretty well for you. now you can define new classes 
which don't have the "archive_method". if all your classes are inherited 
from a base class that _does_ have the archive _method, perhaps you can 
just set that method to "None" for the classes that should not be written.


class new_nonsavable(master_base_class):
	archive_method = None

now you can just check if archive_method is a true value and call it

if myinstance.archive_method:
	myinstance.archive_method(file_or_something)



From tim.one at comcast.net  Sun Feb 24 01:00:59 2002
From: tim.one at comcast.net (Tim Peters)
Date: Sun, 24 Feb 2002 01:00:59 -0500
Subject: [Tutor] unrelated bug issue
In-Reply-To: <3C78529D.F4E7D754@mail.verizon.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEOBNOAA.tim.one@comcast.net>

[Lloyd Hugh Allen]
> So I tried Ruby for a bit, got frustrated by their ide and figured that
> I'd try it later in a proper linux environment since the windows version
> seems to be (I think, anyway) built on CygWin. When I uninstalled Ruby,
> IDLE stopped working (no message--just didn't start. Command line python
> worked fine). Turned out that this was because Ruby had put a few lines
> in the autoexec.bat, one of which said "please use the Tcl library in
> the Ruby directory", which no longer existed. Cleared out the
> autoexec.bat, now Python works again. Is this a Python bug (ie, should
> Python tell the user that the Tcl library doesn't exist or at least give
> some sort of feedback), a Ruby bug (that the uninstaller didn't clean up
> after itself), or both?

I got a better one:  suppose you shoved a banana down your throat, then a
carrot in each nostril.  Would it then be the banana's or the carrots' fault
that you couldn't breathe anymore?  That's right:  it would be the carrots',
because you didn't have any trouble breathing before they wound up in your
nose.  Or maybe it was the banana's fault after all, since if it weren't
there, the carrots' presence wouldn't matter.  It's a puzzler.  Just so long
as it's clear that it's never ever your fault <wink>.




From jason at jorendorff.com  Mon Feb 25 00:30:38 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 24 Feb 2002 23:30:38 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <7x8z9m665n.fsf@ruckus.brouhaha.com>
Message-ID: <HFEKILOLEFEFMKAECNDLEENPDEAA.jason@jorendorff.com>

Paul Rubin wrote:
> I don't see your idiomatic 6-line while loop as being neater or
> more orderly as the straightforward 2-line version that uses :=
> in the control expression.

Your version was:

   while (m := re.match(pat,x)):  blah()

Isn't that an endless loop?

## Jason Orendorff    http://www.jorendorff.com/



From jeff at ccvcorp.com  Mon Feb 25 20:48:43 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Mon, 25 Feb 2002 17:48:43 -0800
Subject: newbie: another problem with file reading
References: <Xns91BEBD577BCA4AusYourStandingInIt@130.133.1.4>
Message-ID: <3C7AE97B.C62377A@ccvcorp.com>


Flavian Hardcastle wrote:

> In this instance, _before_ = 8221 , _after_ = 8305.
>
> What this function should do, is read only up to position 8221 in _chat_,
> print a .gif, then read the rest of the _chat_ file. Why isn't it doing
> that?

It's hard to tell, for sure, since we know nothing of the format of your index
file (or your chat_log file, either), and you haven't told us what it *is*
doing.  Is each line of your index file an offset to the start of a post in
the log file?  A few thoughts, based on that assumption:


> def get_posts(num, before):
>     index = map(string.rstrip, open(chat_index).readlines())
>     after = int(str(index[-1]))
>     g = int(str(index[-(num+1)])) #for some reason, need to add +1
>                                   #maybe cuz of the way chat_log and
>                                             #chat_index are set up

As far as why you need to add 1, that really depends on how you're getting num
and what it's supposed to mean (it's not particularly clear from the code
here).  What I'm really puzzled by, here, is the int(str(index[...])) calls.
Your index should already be a list of strings, so I don't see what you're
accomplishing by calling str() on it -- it seems to me that int(index[...])
should work just as well.


>     print int(before)
>     chat = open(chat_log, 'r')
>     chat.seek(g) #seek to where the post the user wants is
>
>     print chat.read(before) #should mean read up to previous position

This reads before number of bytes from chat, starting at position g -- in
other words, you're getting the contents of chat from position g to position g
+ before.  Since you don't say anything about how you've calculated before,
either, it's impossible to tell if this is the behavior you expect, but it
seems like a likely point of misunderstanding.  (Your comment seems to
indicate a mistaken impression, to me.)


>     print "<img src = pointer.gif></br>"
>     print chat.read() #read rest of file.
>     print after

It's not clear what purpose after is supposed to serve, here.  If my guess as
to your index file layout is correct, then it's the offset of the most recent
post in the chat log, but I don't know why you'd want to print it (except
maybe for debugging purposes).

Jeff Shannon
Technician/Programmer
Credit International




From chris.gonnerman at newcenturycomputers.net  Mon Feb 11 19:23:03 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Mon, 11 Feb 2002 18:23:03 -0600
Subject: Help: stuck on first example!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4> <mailman.1013316023.10782.python-list@python.org> <Xns91B1A9ED71A93AusYourStandingInIt@130.133.1.4> <mailman.1013350037.30507.python-list@python.org> <oNz98.2509$q95.1624800457@newssvr21.news.prodigy.com> <Xns91B3637BF6C59AusYourStandingInIt@210.49.20.254> <3C6854F0.9040003@mishre.com>
Message-ID: <003001c1b35b$736fff00$0101010a@local>

----- Original Message -----
From: "Wayne Pierce" <wayne at mishre.com>


> I haven't used Xitami in a while, but IIRC I had some problems with it
> when I had the first line on a Windows based system.  Try remving the
> first line and testing it.  I don't think it should matter, technically,
> since it's a comment...but it used to.  Mind you it's been about 2 years
> since I've worked with Xitami...

Assuming the OP is using Python 2.2, in a default install, this should work:

    #! c:\python22\python.exe -u

This is more or less what I have done.  Note the blank space between the
"shebang" and the path.  Note also that the leading spaces should not be
in your file.

Last, that -u is for unbuffered I/O, which may be required in some cases.
Revise the path if you are using a different version of Python.





From jason at jorendorff.com  Mon Feb 11 00:16:08 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 10 Feb 2002 23:16:08 -0600
Subject: No explanation for weird behavior in re module!
In-Reply-To: <87pu3c7osj.fsf@uol.com.br>
Message-ID: <HFEKILOLEFEFMKAECNDLAEKHDBAA.jason@jorendorff.com>

synthespian writes:
> 	The problem is that I can't make Python read anything with 
> non-ASCII character set.

import re
import codecs

pattern = re.compile(ur'^(der|die|das)\s+(\w+)', re.UNICODE)

f = codecs.open('article.txt', 'r', 'iso-8859-1')
lines = f.readlines()
f.close()

f = codecs.open('article.out.txt', 'w', 'iso-8859-1')
for line in lines:
    match = pattern.match(line)
    article = match.group(1)
    noun = match.group(2)
    f.write(u"article: %s ... noun: %s\n" % (article, noun))
f.close()


I've got Python 2.2, but I think it should work for you too.

## Jason Orendorff    http://www.jorendorff.com/





From pinard at iro.umontreal.ca  Sat Feb  9 08:41:26 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 09 Feb 2002 08:41:26 -0500
Subject: Attention: Martin von =?iso-8859-1?q?L=F6wis?=
Message-ID: <oqpu3ercvt.fsf@titan.progiciels-bpi.ca>

Sorry for disturbing the list.  Martin, I'm getting quite a few:

 <martin.v.loewis at t-online.de> ... 550 Mailbox quota exceeded / Mailbox voll.

from the TP.  I wish you'll see this message and correct this.

P.S. - But I mainly hope you are well and wealthy! :-)

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



From mhammond at skippinet.com.au  Sat Feb 23 01:20:06 2002
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 23 Feb 2002 06:20:06 GMT
Subject: Python ActiveX Scripting Engine error
References: <508cb8ae.0202210917.6d97e157@posting.google.com> <3C752D64.9090100@mxm.dk> <508cb8ae.0202220033.1c3e7cd7@posting.google.com>
Message-ID: <3C773512.9050002@skippinet.com.au>

B wrote:
> Max M <maxm at mxm.dk> wrote in message news:<3C752D64.9090100 at mxm.dk>...
> 
>>Hmm I sometimes experience those. I think it is when I have PythonWin 
>>open at the same time an I try to run a script with errors. Then it sort 
>>of "gets stuck" on the iis.
>>
>>A restart of IIS solves it for me.
>>
>>
> I dont have PythonWin open and restarting IIS makes no difference, the
> error just keeps happening randomingly it seems.
> I did read something about redirecting before a html page is build can
> cause this but if anyone has any other ideas I'd like to hear them.

This is a known bug, but it has no reproducible test case :(  If you can 
make one, many people will love you :)

Mark.



From peter.milliken at gtech.com  Mon Feb 18 15:31:58 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Tue, 19 Feb 2002 07:31:58 +1100
Subject: Synchronous signals vs. robust code
References: <mailman.1013998768.20770.python-list@python.org> <a4q0f4$akj4@news1.gtech.com> <7xy9hridd1.fsf@ruckus.brouhaha.com>
Message-ID: <a4rodr$akj5@news1.gtech.com>

I interrupted (rightly or wrongly :-)) the following as desiring some form
of timely responsiveness to signals:

"However, it is still unsatisfactory in that it results in the signal
being delayed for an arbitrary amount of time; if a time-critical
signal happens during the try-block, we will not process it until
much later."

Perhaps Mark didn't mean what he says here? So what is his definition of
"time-critical"? :-) You don't get my definition of "time-critical"
satisfied by any interpreted language - but perhaps we operate in different
"ball parks". Certainly if he means "time-critical" to some user pounding on
a keyboard then Python is certainly sufficient for the job! :-)

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xy9hridd1.fsf at ruckus.brouhaha.com...
> "Peter Milliken" <peter.milliken at gtech.com> writes:
> > Why not use a language that provides your requirements of signal safety
etc
> > and do a Python API for it? (assuming one doesn't exist) i.e. I find it
> > somewhat difficult to reconcile your statement about responding to time
> > critical signals when you propose use of an interpretive language.....
:-)
> > Sounds like you would ideally like real-time language features in a
language
> > that isn't designed for it (well, I'm not sure Guido had visions of
Python
> > being used in real-time applications :-)).
>
> I don't see anything real-time specific in that post.  It's a generic
> problem.  I don't see a clean way to make sure your 'finally' clauses
> execute if the user hits ^C several times while the program is
> running.  Early versions of Unix didn't have reliable signals either,
> and it led to flaky programs, til BSD Unix fixed it.




From rhymes at myself.com  Sat Feb  2 10:59:17 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sat, 02 Feb 2002 16:59:17 +0100
Subject: curses module
Message-ID: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com>

Why in the std distribution of Python 2.2 there isn't the curses
module?

Where i can find it?

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From pedronis at bluewin.ch  Mon Feb 25 15:18:36 2002
From: pedronis at bluewin.ch (Samuele Pedroni)
Date: Mon, 25 Feb 2002 21:18:36 +0100
Subject: Jython bugs or features?
References: <gherman-035CF3.16342425022002@news.t-online.com> <3c7a70d2_4@news.bluewin.ch> <gherman-A86A1B.18413225022002@news.t-online.com>
Message-ID: <3c7a9f02_5@news.bluewin.ch>

 [Dinu Gherman]
>
> I'm pretty surprised! I knew the Python test suite is *very* far
> from complete, but list() is an *extremely* crucial function,
> isn't it?

Given how things concretely are and not considering
some abstract point of view, I would say YMMV.

I have skimmed over CPython 2.1 std lib,
honestly there are not many places relevant to Jython where list for
copying lists is used and where the bug can show through.
Consider things like:

a += list(b) OR b = ... + list(b)

and all the places where l = list(sq) and sq is not a list,
those works in Jython.

Btw here simply Jython was never up to speed
(that means always implemented the wrong semantics)
and nobody noticed this or to be precise ever
reported this.

OTOH we try to be responsive to bug reports.

Yes the test suites are *far* from optimal
and CPython sometimes regresses too.

>
> My impression is that Jython claims to be an implementation of
> Python in Java.

Yes but your PS (see below) shows that this can mean
a lot of things, Jython is more about writing new programs
and Java integration (which is a big and non-easy part and of its
codebase) than for example supporting completely CPython os
and shell-like programming and allowing effortless porting of
all kinds of CPython programs (clearly list bug is another kind of issue
given that the right behaviour is well documented).

Java integration has a higher priority than those thing.
Jython is more about embracing  the Java platform
(philosophy) than work-arounding it.
For example anything that would require writing
JNI glue and native C code is simply discarded.

There are things for which CPython is simply
better suited than Java/Jython and the other way
around.

They are not fully equivalent substitutes.

> Everybody understands the existence of bugs, but
> if functions are missing I'm not sure there is sufficient quality
> control, leave alone a useful test suite.
>
> My bold guess is that
> it should be very easy to check automatically for each module in
> the std.lib. at least if the same classes and methods do exist
> in CPython and JPython.
>
> Honestly, I don't think it makes much sense to maintain two
> code bases without some degree of automatic testing...

JPython is not born with such a test, and Jython until
now has never grown one. But I have taken note of this
adding a feature request for such a test but is more a matter of resources
and priorities than easiness.

Btw after some checking of the CPython CVS

http://groups.google.com/groups?q=g:thl2649249624d&hl=en&selm=mailman.100742
5027.30019.python-list%40python.org

(look at the change in the module-level __doc__ )

it seems that reset is vestigial and one should use seek(0)
instead.

reset is not supported by StringIO or by files
and not documented (not after 1.5 for sure).

You made me think that it was added in 2.1,
so at least IMO it is an option for Jython to have decided
not to support it.

>
> When seeing such bugs, my immediate reaction (like that of most
> others) is to think that not many people can be using this serious-
> ly.
>

See e.g.:
http://groups.google.com/groups?q=g:thl2649249624d&hl=en&selm=mailman.100742
5027.30019.python-list%40python.org
and thread.

> PS: BTW, how about this one:
>
> [localhost:~] dinu% jython
> Jython 2.1 on java1.3.1 (JIT: null)
> Type "copyright", "credits" or "license" for more information.
> >>> import os
> >>> os.chdir('..')
> Traceback (innermost last):
>   File "<console>", line 1, in ?
>   File ".../Jython-2.1/Lib/javaos.py", line 56, in chdir
> OSError: [Errno 0] chdir not supported in Java: ..

We don't support this one, sorry.




From kosh at aesaeion.com  Tue Feb 19 04:35:39 2002
From: kosh at aesaeion.com (kosh at aesaeion.com)
Date: Tue, 19 Feb 2002 02:35:39 -0700 (MST)
Subject: Adding static typing to Python
In-Reply-To: <24c39b2c.0202190059.41874b92@posting.google.com>
Message-ID: <Pine.LNX.4.32.0202190219280.1394-100000@aesaeion.com>

On 19 Feb 2002, Alexander Jerusalem wrote:

> There's another thing that bugs me with the missing static typing and
> that is that it makes code less readable. I don't know what I can do
> with a function parameter when I see it. For example, when I see a
> function like def computeSalary(person)... I don't know at first sight
> if person is the id of person, a person object and if it's an object
> what its exact class in a Person class hierarchy is. I have to find
> the place where the constructor is called to find that out. That's not
> a big problem for small programs but if I have many classes that use
> inheritance, it's really not easy to figure out what interface a
> parameter actually supports. This is, however, a minor problem if
> everyone on the team agrees to document parameter types in the doc
> string. The language doesn't enforce it though.
>

Overall I have found that the longer I use python the less I care about
types overall. I have been writing stuff that cares more about the
capability of an object to act in a certain way. For example I don't care
if something is actually a list object I might just need to check if it
can be sliced. Or in other cases I have a function that will call a method
of an object passed in if it has that method. Overall I find myself doing
a fair bit of

if hasattr(self, 'methodname'):
 self.methodname()

This makes the code easy to extend and I think it makes the intent more
clear. Since in the function you see what feature is being used and no
longer care about what kind of object it is. I also use refactoring pretty
heavily and my functions tend to be about 3-5 lines on average. The
functions encapsulate a purpose. I name the function for the purpose,
provide a docstring to further cover that purpose then implement just
that. If I find I have several purposes in one function I break it into
more functions even if some of the functions are only one line since it
makes the intent far more clear.

Overall I find this style makes debugging far cleaner and easier to do. I
also use a simple logging function so that at any time I can stick it in
somewhere and get the state of what is going on which works very well.
This also eliminates most temporary variables which tend to be the items
that are mistyped the most. Overall I think static typing is good for one
style of programming however other styles work very well aso. Dynamic type
checking works very well with refactoring and using items based on
capabilities.

> And the final argument I have for static type checking is that it
> enables method dispatching based on parameter types. In a statically
> typed language you can create two methods that have the same name but
> differ on the parameter types. The correct method will be called for
> you depending on the type of the argument you pass in your call. That
> makes for a quite flexible way of extending a program. You can just
> add another method with the same name and another type to handle a
> special case without touching the existing methods.
>

Overall I used to think that was useful when I first started using python
from a fair bit of C++ experience. However as time went on I found that
less and less useful until I just don't write things that need that
anymore. I think you feel a need for a feature not because the feature is
a good one or needed but because that is what you are used to. Same way
that many python newbies make large strings by adding up lots of little
strings. Depending on how many items you are adding together this can be
thousands of times slower then putting the strings in a list and doing a
join when you have all of them.

> I'm currently using Python to write generators that produce Java code
> from an XML representation. So the runtime system is type checked by
> the Java compiler anyway. I tried to do the generator in Java first
> but the text processing capabilities of Java are just terrible, so I
> came to use Python and I like it for many reasons. It's just that I'm
> using much more time on debugging than I used to in Java where I
> didn't have to care too much about typos because the compiler
> complained anyway...

It seems to me that the ammount of time you spend debugging a language is
directly related to how much you fight it compared to how much you work
with it. When you treat every tool like a hammer then you notice that some
tools are better hammers then others. Overall a wrench might make a decent
hammer but that does not make it a hammer. What I see a fair bit of is
your interest in using python like a hammer becuase you know hammers from
java and C++ instead of using it like a screwdriver. Not to say that one
is different from the other just that they are different and need to be
used differently. The part where this analogy breaks down is that they can
both do the same things just in a different way.




From drew at level3.net  Wed Feb 27 18:38:32 2002
From: drew at level3.net (Drew Fisher)
Date: 27 Feb 2002 16:38:32 -0700
Subject: regular expression help
Message-ID: <1014853112.982.15.camel@beast.l3.com>

Hello,

I'm having some issues doing a small but irritating regular expression
match.

Here's an example of text that I want to search through:

str = """
Location:Home
	555-1212
	555-3434
	555-5656

Location:Work
	555-9999
"""

Just for reference, there is newline characters at the end of each line
and the phone numbers have a single tab character (\t) in front of them.

I'm trying to get the location of the record and the associated phone
number(s) that go along with it.

I've tried things like:

import re
test1 = re.compile ('Location:(\w+)\s+(\d+\-\d+)', re.DOTALL)
print test1.findall (str)
[('Home', '555-1212'), ('Work', '555-9999')]

But this only gets the location and the first phone number in the
record.  How do I get all of the phone numbers for each location without
writing a second regular expression?

Thanks for your help,

Drew




From mkeijzer at cs.vu.nl  Mon Feb 18 10:01:51 2002
From: mkeijzer at cs.vu.nl (Maarten Keijzer)
Date: 18 Feb 2002 07:01:51 -0800
Subject: There is more than one way to do it - and for no apparent reason.
References: <3C70ECB1.2090709@mxm.dk>
Message-ID: <17cabb0.0202180701.68ef7a8d@posting.google.com>

Hi,

if it's syntactic sugar you want, create a base class that uses __getitem__ and 
__setitem__. For instance like this:

class MyBase:
    def __getitem__(self, item):
        return self.__dict__[item]

    def __setitem__(self, item, value):
        self.__dict__[item] = value

Then you can do:

class MyDerived(MyBase):
    pass

base = MyDerived()
base.name = 'hi'
print base['name']
base['name'] = 'ho'
print base.name

If you have other uses for the [] notation, you're still free to use it.


From ron at ire.org  Thu Feb 28 18:25:06 2002
From: ron at ire.org (ron nixon)
Date: 28 Feb 2002 15:25:06 -0800
Subject: ASV module, CVS modules
Message-ID: <61ff32e3.0202281525.3e9bb102@posting.google.com>

Hello, I'm trying to use either of these two modules ASV by Laurence
Tratt at http://tratt.net/laurie/python/asv/ and CSV by Dave Cole at
http://object-craft.com.au/projects/csv/. I tried both of these and
get this error:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> import ASV
>>> my_data = ASV.ASV()
>>> my_data.input_from_file("c:\test_data.csv", ASV.CSV(),
has_field_names =1)
Traceback (most recent call last):
  File "<pyshell#2>", line 1, in ?
    my_data.input_from_file("c:\test_data.csv", ASV.CSV(),
has_field_names =1)
  File "C:\PYTHON22\ASV.py", line 222, in input_from_file
    file = open(input_file, "rb")
IOError: invalid argument: rb
>>> 

I get a similar message from CSV module. Anyone use either of these
and can tell me what's wrong?

Ron Nixon


From owen at astrono.junkwashington.emu  Fri Feb  8 11:26:41 2002
From: owen at astrono.junkwashington.emu (Russell E. Owen)
Date: Fri, 08 Feb 2002 08:26:41 -0800
Subject: Fine Tkinter Question
References: <mailman.1013104932.12965.python-list@python.org>
Message-ID: <a40u82$mfs$1@nntp6.u.washington.edu>

In article <mailman.1013104932.12965.python-list at python.org>,
 Mariano Corsunsky <core.lists.python at core-sdi.com> wrote:

>from Tkinter import *
>def show child ():
>    win=Toplevel()
...
>    win.bind("<Return>",lambda event,myself=win:myself.destroy)
(and <Return> has no effect)

I'm not sure, but I suspect that windows never receive key events. You 
could easily test this. If it's true, try putting a frame in the window 
and putting your widget(s) into the frame, then bind the <Return> key 
event to the frame.

-- Russell


From jeff at ccvcorp.com  Wed Feb 27 18:19:48 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Wed, 27 Feb 2002 15:19:48 -0800
Subject: Proposal: local variable declarations and type advice
References: <mailman.1014706483.6455.python-list@python.org> <7x3czobs0x.fsf@ruckus.brouhaha.com> <VvNe8.68907$X64.25084093@news1.rdc2.pa.home.com> <7xk7sznsoa.fsf@ruckus.brouhaha.com> <3C7D3B2C.80A97BA2@ccvcorp.com> <7x664i8vb6.fsf@ruckus.brouhaha.com>
Message-ID: <3C7D6994.B2D43BFA@ccvcorp.com>


Paul Rubin wrote:

>
> What I mean is if you say
>
>   if x:
>     temp = frob()
>     mumble(temp)
>     foo()
>     bar()
>     del temp
>
> and the foo() or one of the other calls raises an exception that's caught
> by some outer block in the same function, then temp hasn't been deleted,
> and in fact the code could even depend on it not being deleted.  So del
> doesn't give you the same effect as local scope.

Ah, yes, got it.  That does make sense.  I don't see using try/finally as
being too onerous, in cases where you *need* to have certain names cleaned up
(which would seem a moderately rare occurence, to me), but I can see that
being able to declare a local scope would be a bit cleaner.  I'm inclinded to
think that the need for local-scope-inside-function-scope is not frequent
enough to justify a language change to be cleaner than try/finally, but maybe
it's just because I've been working in relatively straightforward application
areas.

Jeff Shannon
Technician/Programmer
Credit International





From gs at styrax.com  Thu Feb 28 14:54:57 2002
From: gs at styrax.com (Garry Steedman)
Date: Thu, 28 Feb 2002 19:54:57 +0000
Subject: Python DB that supports numbers
In-Reply-To: <20020228133655.A2853090@vislab.epa.gov>
References: <3C7E80DD.3655.6F5638C@localhost>; from gs@styrax.com on Thu, Feb 28, 2002 at 07:11:25PM +0000
Message-ID: <3C7E8B11.22795.71D3FFC@localhost>

randall,

youre not missing anything: theres nothing there im afraid... 8-(

hmmm... you could unbundle it from a zope source tarball at 
zope.org...

if thats too much of a pain, try:

http://www.mcmillan-inc.com/mksqlintro.html

but i've never used it myself...

hth,

garry

On 28 Feb 2002, at 13:36, Randall Hopper wrote:

From:           	Randall Hopper <aa8vb at yahoo.com>
To:             	Garry Steedman <gs at styrax.com>
Copies to:      	python-list at python.org
Subject:        	Re: Python DB that supports numbers
Date sent:      	Thu, 28 Feb 2002 13:36:55 -0500

> Garry Steedman:
>  |>  http://www.chordate.com/gadfly.html
>  |>            Connection Failed 
>  |
>  |http://sourceforge.net/projects/gadfly/
>  |
>  |should work...
> 
> Ok, but there are no docs, no project files, and the CVS repository
> appears empty.  
> 
> What am I missing?
> 
> Randall
> 
> -- 
> Randall Hopper
> aa8vb at yahoo.com
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list


+-------------------------------------------+
  Garry Steedman      mailto:gs at styrax.com  
  Styrax Associates  http://www.styrax.com/ 

  "The Good Man has no shape."            
+-------------------------------------------+



From marklists at mceahern.com  Fri Feb  8 12:30:23 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Fri, 8 Feb 2002 09:30:23 -0800
Subject: OT: (ambi)dexterity of Pythonistas
In-Reply-To: <a40rtu$b0r$1@panix2.panix.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBGEKJKBAA.marklists@mceahern.com>

[Aahz Maruch]
> Haven't figured out a good/cheap way to put more than two monitors on a
> desk, particularly when both are 17" or larger.  It's hard enough
> stuffing *one* monitor on most crapola office desks.

flat panel?

// mark



From carifio.nospam at nospam.usys.com  Tue Feb 19 17:09:12 2002
From: carifio.nospam at nospam.usys.com (Mike Carifio)
Date: Tue, 19 Feb 2002 22:09:12 GMT
Subject: newbieish question
Message-ID: <c2Ac8.26231$zJ3.1876517@typhoon.maine.rr.com>

Is there a way to turn off the interpretation of "\n" on sys.platform ==
"win32"
when writing to a file?

Here's an example:

fh = open("a_file.text", "w") # say
fw.write("\n")  # writes \r\n on win32
fw.write(chr(10))  # also appears to write \r\n on win32; can I get just
line feed?






From brueckd at tbye.com  Wed Feb 27 14:16:13 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Wed, 27 Feb 2002 11:16:13 -0800 (PST)
Subject: PIL vs eps
In-Reply-To: <JFkzWtAaSRf8Ewgj@jessikat.fsnet.co.uk>
Message-ID: <Pine.LNX.4.33.0202271110290.14586-100000@ns0.tbye.com>

On Wed, 27 Feb 2002, Robin Becker wrote:

> In article <mailman.1014824145.17587.python-list at python.org>, Gustavo
> Cordova <gcordova at hebmex.com> writes
> >>
> >> I think this is a pure ghostscript problem, but PIL fails to
> >> convert eps to other formats and in fact it seems that gs
> >> won't create any output without a showpage at least somewhere.
> >>
> >
> >Isn't "showpage" the postscript command to render
> >all the code that's been processed?
> >
> >-gustavo
> >
> yes, but most eps files don't seem to have a show page and thus
> ghostscript doesn't seem to render anything. The EPSPlugin seems to pass
> the data to ghostscript, but a client reports errors with .eps. When I
> do the gs command handomatically I get no file. If I add in a showpage
> just before the final restore I at least get the converted image/ppm
> etc.

Forgive me if I'm totally off my rocker, but I think by definition eps
files should *not* have a showpage and it is technically correct that
nothing gets rendered unless you have a showpage command. EPS is
"encapsulated postscript" - postscript that you intend to encapsulate or
embed in another document, so showpage would muck things up.

-Dave




From achim.domma at syynx.de  Tue Feb 12 10:15:00 2002
From: achim.domma at syynx.de (Achim Domma)
Date: Tue, 12 Feb 2002 16:15:00 +0100
Subject: read and write binary data (newbie)
References: <3c69359f.22621453@news.easynews.net>
Message-ID: <a4bb79$5jm$03$1@news.t-online.com>

look at the functions 'pack'/'unpack' in module 'struct' from the standard
library

greetings
Achim




From gbreed at cix.compulink.co.uk  Mon Feb  4 12:51:55 2002
From: gbreed at cix.compulink.co.uk (gbreed at cix.compulink.co.uk)
Date: Mon, 4 Feb 2002 17:51:55 +0000 (UTC)
Subject: Fun with lambda and map
References: <3c5eaf65.22609703@news.t-online.de>
Message-ID: <a3mhnr$r2o$1@thorium.cix.co.uk>

Gerson Kurz wrote:

> If you're in a masochistic mood, you can write this in one line:
> 
> fac = lambda n:map(lambda x:(map(lambda a:a and 1,[x[0]()])[0] and
> x[1]()) or x[2](x[3]),[[lambda:not n,lambda:n==n,lambda
> n:fac(n-1)*n,n]])[0]

If you're allowing booleans and subtraction, how about

fac = lambda x: not x or x*fac(x-1)

?


                Graham

    (<http://www.microtonal.co.uk/>)


From amuys at shortech.com.au  Tue Feb  5 20:22:48 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 5 Feb 2002 17:22:48 -0800
Subject: when does python free memory?
References: <10b3f29a.0202050840.7641923a@posting.google.com>
Message-ID: <7934d084.0202051722.4367951@posting.google.com>

reanes at banta-im.com (Robert Eanes) wrote in message news:<10b3f29a.0202050840.7641923a at posting.google.com>...
> Hi-
> Reading the docs and searching the list archives, it seemed that
> python should release memory as soon as the last reference to an
> object is deleted.  However, I can't seem to prove this in practice. 
> Look at this example:

This is true for 1.5.2, however later versions of python may use a
garbage collected store for some/all variables.  I don't know enough
about python's gc to discuss this.  (ps. anyone got a good reference
on a discussion of the current CPython (de-)allocator?)

<SNIP sample code demonstrating retained heap>

> So apparently python knows that the memory formerly allocated to z is
> available, but it doesn't tell the OS.  Is this the case, and if so,
> is there any way around it?  I have an application that needs to run
> 80-150 python processes that can stay around for 15min to an hour, so
> I'd like them not to hold onto the maximum amount of memory that they
> use throughout the whole life of the process.

This is a common optimisation of malloc that reduces the number of
brk/mmap calls required to service malloc/free/malloc/free/... call's.
 It can provide a signifigant performance advantage for the common
case, but can cause problems for certain unusual cases (such as yours
above).  Most platforms malloc libraries allow you to tune them on a
per-process level (normally by setting appropriate environment
variables) if required, check with your sysadmin/vendor.

To start getting a handle on this I suggest you read up on malloc/free
and their relationship with the brk(and friends), and mmap/munmap
syscall's.

  - A relatively easy read on how a performance problem successfully
diagnosed as inefficient malloc/free's, and the tuning parameters used
to solve it.  The paper discusses Tru64 and Linux, but the concepts
are generally applicable.
http://www.linuxshowcase.org/full_papers/ezolt/ezolt.pdf

  - A paper discussing various heap smashing attacks against glibc's
malloc/free algorithm.  Useful here because of it contains a
reasonably detailed discussion of Doug Lea's allocator (see sections
3.1-3.5)
http://www.phrack.org/show.php?p=57&a=8

  - Doug Lea's own paper on the design of his allocator. (I don't
personally know of any papers on the design of Solaris' allocator, but
I'm interested in knowing of one if it exists)
http://gee.cs.oswego.edu/dl/html/malloc.html

  - An online, HTMLised version of the GNU info page on tuning
parameters for glibc's malloc/free implementations.
http://docsrv.caldera.com:8457/cgi-bin/info2html?(libc)Malloc%2520Tunable%2520Parameters

Good luck

Andrae Muys


From niemeyer at conectiva.com  Wed Feb 27 10:25:31 2002
From: niemeyer at conectiva.com (Gustavo Niemeyer)
Date: Wed, 27 Feb 2002 12:25:31 -0300
Subject: ANN: python-mcrypt 1.0
Message-ID: <20020227122531.A2801@ibook>

Description
-----------

The python mcrypt module provides a comprehensive interface for the
mcrypt library, which provides a uniform interface to several symmetric
encryption algorithms. It is intended to have a simple interface to access
encryption algorithms in ofb, cbc, cfb, ecb and stream modes. The algorithms it
supports are DES, 3DES, RIJNDAEL, Twofish, IDEA, GOST, CAST-256, ARCFOUR,
SERPENT, SAFER+, and more.

Features
--------

The python-mcrypt module provides the following features:

o  Free software (GNU Lesser General Public License).
o  Support for Python newstyle classes, allowing subclassing.
o  Comprehensive unitttests.
o  Support for many well known and less known algorithms and modes. (*)
o  Support for fixlength algorithm, adjusting encrypted data length
   even for block algorithms and modes.
o  Support for direct file encryption.
o  Complete inline documentation.
o  Check parameters before passing to mcrypt library, protecting
   agains segmentation faults.

(*) mcrypt library features

Disclaimer
----------

Please read the file LICENSE before using python-mcrypt.

New Versions
------------

Information on how to retrieve new versions of python-mcrypt can be
found on http://sourceforge.net/projects/python-mcrypt.

-- 
Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5  60E2 2253 B29A 6664 3A0C ]



From feldspar at ix.netcom.com  Mon Feb 11 01:44:06 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Mon, 11 Feb 2002 02:44:06 -0400
Subject: defining functions
References: <20020210214159.13516.00000450@mb-cq.aol.com>
Message-ID: <3C676836.846FEDE3@ix.netcom.com>

Patio87 wrote:
> 
> I am pretty new to python, and I dont understand the parameters when defining
> functions. Whenever I see a function definition with a argument It seems like
> the argument already is a variable, but it goes threw the function and it makes
> no sense. I dont know if what I just said made any sence but if you can help me
> please reply
> 
> def factorial(n):
>     if n == 1:
>         return 1
>     else:
>         return n * factorial(n-1)
> What the hell does 'n' have assingned to it?

I think you must be very new to programming in general, not just to
Python, but that's quite all right.

A function definition tells you how to perform that function in the
abstract.  For instance, if I asked you, "How do you average two
numbers?" you would probably say something to me like the following: 
"Add the first number and the second number together.  Divide their sum
by 2.  There's your answer."  There you go, you've just defined the
averaging of two numbers -- without having to define the numbers
themselves.

If we turned your definition into Python code, it might look like this:

def average_two(first_num, second_num):
  sum = first_num + second_num
  average = sum / 2.0  # Dividing by 2 would abandon any remainder
  return average

There, you've just written a function that the interpreter can use any
time it wants to average *any* two numbers.  If you define average_two
in your interpreter in interactive mode, and then type in

>>> average_two(5, 7)

The interpreter will execute the function you wrote, with a value of 5
for "first_num" and a value of 7 for "second_num".

That's the point of functions:  you define some operation in the
abstract, with abstract inputs that you give variable names to.  Once
you do that, the interpreter then knows how to perform that operation in
the concrete, substituting the actual values.

	-jc


From mario at bio58.uni-bielefeld.de  Tue Feb  5 09:17:35 2002
From: mario at bio58.uni-bielefeld.de (Mario Wehbrink)
Date: 5 Feb 2002 09:17:35 MET
Subject: CPython vs. Jython for Web applications?
References: <c775c8db.0202041126.5177e606@posting.google.com>
Message-ID: <3c5f951f$1_2@news.uni-bielefeld.de>

Am  4 Feb 2002 11:26:27 -0800  schrieb/wrote/a ?crit  
John E. Barham  < jbarham at octopuswireless.com > in comp.lang.python : 
 
> I appreciate that Jython's performance is dependent on the JVM, so
> advice would be appreciated there as well, especially regarding the
> performance of JVMs for the free Unices.

I read some days ago in german linux-magazin that the performance of
Jython is about 80% of CPython when using Microsofts VM. When using one
of the Java ports of linux (blackdown or IBM) the performance is only
20 or 30% of CPython, but i have not done any benchmarks to support that
statement.

Bye

Mario

-- 
Mario Wehbrink
Abteilung 4 Kybernetik
University of Bielefeld


From logiplexsoftware at earthlink.net  Wed Feb 13 21:50:58 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 13 Feb 2002 18:50:58 -0800
Subject: Using tabs instead of spaces in IDLE?
In-Reply-To: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com>
References: <jaFa8.138426$a07.38775395@typhoon.southeast.rr.com>
Message-ID: <20020213185058.6f329938.logiplexsoftware@earthlink.net>

On Thu, 14 Feb 2002 02:21:03 GMT
Tom Bryan wrote:

> Is there a way to use tabs exclusively for indentation in IDLE? 
Basically, 
> I'd like tabwidth to equal identwidth while permitting developers to
adjust 
> the indentwidth.  I see that AutoIndent.py says

I don't use IDLE, so I can't answer your question (I use XEmacs).  However,
I've long been a hardcore tabber but recently decided to uses spaces
instead.  I've found through experience that this is probably the best way
to go when coding in Python.  Further, since Emacs (and I've heard, vim)
support treating groups of spaces as a single tab (tab inserts n number of
spaces, backspace takes you back n spaces), there really isn't much
functional difference between the two when editing.

So why change?  Because I've found that most Python code you will find uses
spaces.  Certainly all the code in the standard library does. 
Cutting/pasting will go much easier in the typical case than if you use
tabs.  If you decide to edit a python file you've gotten from the standard
library or from a contributed source, you will have to remember to indent
with spaces and not automatically reach for the tab key.  If you forget,
you'll have an invisible error in your file (although the traceback will
usually lead you quickly to it).  I've had this happen enough that I
finally capitulated and made the switch.  End result: editing my code is
about the same and editing other's code is far easier.

Regards,

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From shriek at gmx.co.uk  Wed Feb 27 23:13:45 2002
From: shriek at gmx.co.uk (Stephen)
Date: 27 Feb 2002 20:13:45 -0800
Subject: How to have multi-threaded server log to one file ?
References: <97ae44ee.0202261855.15c97708@posting.google.com> <f72dac0d.0202270140.97b925f@posting.google.com>
Message-ID: <97ae44ee.0202272013.2094b2b@posting.google.com>

alanmk at hotmail.com (Alan Kennedy) wrote in message news:<f72dac0d.0202270140.97b925f at posting.google.com>...
> shriek at gmx.co.uk (Stephen) wrote in message 
> > When the server is run, each request from a client is handled by
> > MyHandler() in a separate thread. If multiple requests are being run
> > at the same time, how should I write my code so that they cal all
> > write to the same log file without conflict ?  
> 
> How about representing every log message as a simple object which is
> put() on a Queue.Queue? So each server thread produces messages to go
> on the Queue, and a single logging thread consumes those messages and
> does the actual writing into the log files. Since there is only ever
> one thread writing to the files, there are no contention problems.
> 
> That Queue.Queue class is mightily useful!

Thanks Alan (and Aahz) for pointing this out to me. 

I had a look at the effbot queue-example and think I understand the
Queue class but adapting my program to share the queue instance in the
first place is making my head spin.  Namely, how do I get the same
Queue object into both MyHandler and MyServer in the first place ?

Pseudo code showing where queue will be used ~
--------------------------------------------
 
class MyHandler(SocketServer.StreamRequestHandler):
    def handle(self):
        # Do something interesting.
        # Log an error : 
        self.logerror("2002-02-28 12:00:01 Invalid argument") 
    
    def logerror(self, s):
        self.__queue.put(s)

class MyServer(SocketServer.ThreadingTCPServer):
    allow_reuse_address = 1
    queue = Queue.Queue(0) 
    self.__queue = queue
    
def start(host, port):
    server = MyServer((host, port), MyHandler)
    server.serve_forever() # serve forever


Or is this somewhere that a global variable Queue object should be
used ?

Stephen.


From westernsam at hotmail.com  Fri Feb  1 11:19:35 2002
From: westernsam at hotmail.com (sam)
Date: 1 Feb 2002 08:19:35 -0800
Subject: While loop error
References: <32c382ad.0201312208.166d5de3@posting.google.com>
Message-ID: <292c8da4.0202010819.16b5f41c@posting.google.com>

ignore last message, just re-read your post and realsied that it
should work (and does work with my python implemetation which is 1.52
on win32)

sorry 

sam


From brian at sweetapp.com  Sat Feb 23 12:14:49 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Sat, 23 Feb 2002 09:14:49 -0800
Subject: XML RPC Woes
In-Reply-To: <FPJd8.417$PE2.43414@stones>
Message-ID: <003001c1bc8d$99c0e210$445d4540@Dell2>

Graham Ashton wrote:
> > However, all of a sudden I have not been
> > able to get it to work at all. The server just keeps on returning
"500
> > Internal Error" now.
> 
> I usually find that's caused by a bug in my code that knackers things
up.
> Why don't you post an example?

The problem turned out to be that he didn't have a return statement in
one of his server's functions. So it implicitly returned None, which has
no equivalent XML-RPC type.

Cheers,
Brian




From joshm at taconic.net  Thu Feb  7 00:12:04 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Thu, 7 Feb 2002 00:12:04 -0500
Subject: money class?
References: <mailman.1013053272.13460.python-list@python.org>
Message-ID: <3c620a48_1@corp.newsgroups.com>

I have a Money class, but for some reason, it seems to leak.  At least
that's what my bank keeps telling me.  :-)

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From emile at fenx.com  Tue Feb  5 11:17:23 2002
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 5 Feb 2002 08:17:23 -0800
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de>
Message-ID: <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de>

"MDK" <mdk at mdk.com> wrote in message
news:a3p01b$186eq8$1 at ID-98166.news.dfncis.de...
> Hello,
>
> I am trying to create my own special type of list object.
>
> However, when I do x.append('yo') it uses Python's append instead of
the one
> from my class.
>
> How can I get it to use my append method?  I have tried def __append__
and
> then called my own class' function but that did not work.  I've played
with
> __getattr__ but that did not help.
>

Call the method append and not __append__?


--

Emile van Sebille
emile at fenx.com

---------



From jim.vickroy at noaa.gov  Mon Feb  4 11:45:51 2002
From: jim.vickroy at noaa.gov (jim.vickroy)
Date: Mon, 04 Feb 2002 09:45:51 -0700
Subject: Remaining nimble in the face of evolution
Message-ID: <3C5EBABF.40555C22@noaa.gov>


Given:
   A *class* of Python objects that, among other attributes, have the
   capability to archive themselves.

   This capability was implicit and never assumed to be optional.  There
is
   considerable legacy code based on this assumption (and many classes
that
   extended this base "archivable").  Starting to sound familiar?


Now a reality check occurs and the requirement evolves such that the
archival
behavior is to be optional.

What is the "best" strategy to handle these types of evolution.

A possible solution is to retro-fit the numerous class initializers
(constructors) with an archival switch parameter that defaults to the
legacy
behavior (i.e., archive=yes).  This approach "feels" wrong.  There are
many
other behaviors in the legacy classes that will **never** change.  Using

flags to affect behavior seems messy.

Another possible solution is to provide an alternate implementation of
the base
class (having the same name but necessarily in a different module) that
does
not provide the archive capability.  The desired module could be
imported in
response to some start-up parameter.

What are pitfalls of the above approaches and what are other strategies
for
managing evolution?

Thanks in advance for your comments.

... I'm not dead yet (hopefully) !



From gerhard at bigfoot.de  Fri Feb 22 15:40:41 2002
From: gerhard at bigfoot.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Fri, 22 Feb 2002 21:40:41 +0100
Subject: ncurses
In-Reply-To: <02022215300500.03018@magneto>
References: <02022215300500.03018@magneto>
Message-ID: <20020222204041.GA2806@lilith.hqd-internal>

Le 22/02/02 ? 15:30, Charles ?crivit:
> Anybody know where can i find a GOOD tutorial or explantion
> about ncurse (curses) python module ?

http://py-howto.sourceforge.net/curses/curses.html

The online version unfortunately contains a few typos. The CVS version
is better.

Before you start coding after reading the first page, wait for the
explanation of curses.wrapper - it's much easier with that.

Your application could look similar to


import curses, time

def mymain(stdscr):
    #stdscr = MyWindow(stdscr)
    
    while 1:
        stdscr.addstr(10, 10, str(time.asctime()))
        stdscr.refresh()
        time.sleep(1)
    
curses.wrapper(mymain)

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))



From steve at ferg.org  Thu Feb  7 19:12:06 2002
From: steve at ferg.org (Stephen Ferg)
Date: 7 Feb 2002 16:12:06 -0800
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <f72dac0d.0202070547.8cb548a@posting.google.com>
Message-ID: <b16e4ef7.0202071612.25dd8417@posting.google.com>

The best take on it = "visionary".  A less charitable take = vague,
incoherent, and disorganized.  I left after half an hour.

He did talk about the "semantic net", which is the hot buzzword du
jour.

The semantic net is not a relational database, but a computerized kind
of Topic Map.  Major supporting technology for the semantic web is
RDF, which is an XML format for storing subject-verb-object
statements.

I learned this stuff about the semantic web and RDF from the talk by
the two guys from 4suite.com, who are clearly very sharp and on top of
all of this stuff.


From pjdurai at hotmail.com  Tue Feb 26 13:24:39 2002
From: pjdurai at hotmail.com (pj)
Date: 26 Feb 2002 10:24:39 -0800
Subject: PythonWin repeatedly crashing on trying to create COM object
Message-ID: <e8f1f186.0202261024.6370a155@posting.google.com>

Greetings

I use python to do some COM scripting on my windows 2000 (SR2) box. It
was working well.. but it once crashed once.

Then it just keeps crashing on

    session = win32com.client.dynamic.Dispatch("Lotus.NotesSession.1")

I just cant seem to get past this point.

Is there something I can do ?( Like clearing out some temp cache files
or something ..) to get my PythonWin back to shape ?

I am using ActivePython 2.1.212

Appreciate your time.

cheers
pj


From rdsteph at earthlink.net  Sun Feb  3 00:40:11 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 03 Feb 2002 05:40:11 GMT
Subject: Automatic daily updates on Pythonic web articles works-finally!
Message-ID: <3C5CCE13.42EB95BF@earthlink.net>

http://www.awaretek.com/plf.html now has a link that gives a new list
each day at 7:30 A.M. of the last 200 web articles about Python. 

The full archive of all of the thousands of such articles since the dawn
of time is also still available, but it takes a much longer time to
download ;-)))

It's not perfect; it is still a work in progress, but at least now (if
it all continues to work !) we can expect a new list each morning of
interesting articles. Perhaps we are dealing with a case of nascent
artificial intelligence, since the resulting list is just a wee bit
indeterminate; so at least you know you'll see new articles each day,
although a few might be actually 'old" articles that didn't show up
before for some reason (right now IBM's Developer Works has a string of
such articles at the top of the list, but hey they're great articles and
we should all read Dr. David Mertz' stuff again anyway;-))). Also, a few
non-Python but nevertheless interesting programming related articles
show up from time to time.

I can't guarantee it will continue to work each day (I am out of town
again all week and the computer could crash) but I think it will and it
is interesting to see fresh Pythonic articles each day, thanks to Hans
Nowak for the mygale web spider program!

Also, you might want to check out the link to the "Tech News Web
Robot-updated each 15 minutes". This is a more general tech news web
spider that updates more or less continuously all the time and grabs
links to breaking news stories about technical subjects (mostly
computer-related) and scrolls them across the screen, with older stories
dropping off the bottom, each story might be up for 2-3 hours on a week
day before it disappears into the ether.

Also yet again ;-))), I have entered the first installments of my
experiences with Linux under the Linux Chronicles. I switched cold
turkey three weeks ago, and although I have been gone from home most of
the time since then, I am finding Linux to be mostly great, and
certainly a great environment for Python, but of course I have
encountered some difficulties. 

And more new stuff is there also...

http://www.awaretek.com/plf.html

Ron Stephens

P.S. Thanks again to Hans Nowak...


From mdk at mdk.com  Thu Feb 14 13:13:55 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 13:13:55 -0500
Subject: ActivePython 2.2.0 Alpha
Message-ID: <a4gupa$7g5e$1@ID-98166.news.dfncis.de>

Is anyone using the above?  Does it seem "stable enough"?




From usenet at thinkspot.net  Sat Feb 16 17:14:25 2002
From: usenet at thinkspot.net (Sheila King)
Date: Sat, 16 Feb 2002 14:14:25 -0800
Subject: How to get CNAME with sockets ?
Message-ID: <a4lpg1.3vvdg2v.1@kserver.org>

I'm working on using Vipul's Razor (http://razor.sourceforge.net) in
conjunction with a Python mail script that I'm working on.

I'm studying some of the Perl code (and I don't really know Perl...makes
it a bit tough). Anyhow, I've gathered that in the sub discover what I
basically need to do, is query each subdomain of razor.vipul.net where
the subdomains are a..z,1..9999 until I find one whose CNAME (or
Canonical Name) is last.terminator.

At this time, it appears that there are only two razor servers in the
zone razor.vipul.net (so far as I can tell).
a.razor.vipul.net = ubik.vipul.net = 194.109.217.74
-AND-
b.razor.vipul.net = adept.darkridge.com = 64.90.164.74

By using Sam Spade, I was able to see that c.razor.vipul.net has the
CNAME last.terminator, which is supposed to signal that I should stop
trying to "discover" new razor servers in the zone razor.vipul.net.

Now using the socket module for Python, I was able to find out all the
facts about a.razor.vipul.net and about b.razor.vipul.net by doing
something like this:

>>> socket.getfqdn('a.razor.vipul.net')
'ubik.vipul.net'
>>> socket.gethostbyaddr('ubik.vipul.net')
('ubik.vipul.net', ['a.razor.vipul.net'], ['194.109.217.74'])


But it appears that one cannot get the CNAME for c.razor.vipul.net using
the socket module. This requires a 3rd party application? I did find
this page
http://c0re.jp/c0de/dnsmodule/
which points to some python DNS modules. However, I only need this for
one line in my program, so I don't really want to import a module for
that. (I'm sharing this with other people and don't want to require them
to install additional packages.)

The machine I'm running this on does offer the command
dnsquery

and if I give the following instruction on the command line:
dnsquery -t CNAME c.razor.vipul.net

I get these results:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 45010
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0
;;      c.razor.vipul.net, type = CNAME, class = IN
c.razor.vipul.net.      14m52s IN CNAME  list.terminator.

I can extract the string "list.terminator" from the reply.

I guess the best way to handle this is with a popen command? Like:

pipe = os.popen('dnsquery -t CNAME c.razor.vipul.net')
data = pipe.read()

If anyone has remarks, comments or suggestions, I'd be happy to hear
them. I wish I didn't have to rely on 3rd party apps for this, but it
appears Python doesn't have the built-in ability to get CNAME records in
the standard library?

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From thm at duke.edu  Wed Feb 27 11:49:59 2002
From: thm at duke.edu (Hunter Matthews)
Date: 27 Feb 2002 08:49:59 -0800
Subject: Python and Daemons
References: <02022215115601.01886@localhost.localdomain> <mailman.1014416559.28231.python-list@python.org>
Message-ID: <8b3354da.0202270849.6b994141@posting.google.com>

Justin Sheehy <justin at iago.org> wrote in message news:<mailman.1014416559.28231.python-list at python.org>...
> Dieter Lunn <theguru at mb.sympatico.ca> writes:
> 
> > I was wondering how to write a linux daemon using Python.
> 
> This is what I use:
> 
> class Log_out:
>     '''file-like dummy device, used for daemonization
>     writes are redirected into our logging object'''
>     def write(self, s):
>         debugmsg(s)
> 
> def daemonize():
>     "cause this process to becaome a daemon"
>     if os.fork() != 0:
>         os._exit(0)
>     os.setsid()
>     if os.fork() != 0:      # 2nd fork is widely recommended
>         os._exit(0)         # though not always needed
>     debugmsg("daemonized")
>     os.chdir("/")   # not strictly necessary, but a good idea
>     os.umask(0)
>     sys.stdin.close()
>     sys.stdout = Log_out()
>     sys.stderr = Log_out()
> 
> debugmsg is a local function which calls a function in my logging library.
> 
> -Justin

Can someone say why the 2nd fork is widely recommneded?


From a.baan at fokkerspace.nl  Fri Feb  1 07:29:21 2002
From: a.baan at fokkerspace.nl (Arno Baan)
Date: 01 Feb 2002 12:29:21 GMT
Subject: importing in Dictionairy
Message-ID: <01c1ab1c$13174290$528e4991@pc1490>

I really don't understand this, please help me.
I've got the following problem, illustrated by the example code below.
When I embed the python interpreter and I want to compile and run some
simple python code. 
It stops when I try to import a module.
When I run the same code using PyRun_SimpleString it works the way I 
expect it to. But because I need to retrieve computed values I need 
the dict object.
Can someone explain to me why I can't import modules when I use the 
dict object??

#include<Python.h>
#include<compile.h>
#include<eval.h>
#include<string>

void main(int argc, char * argv[]){
PyObject * pcode, * pdict, * pmod;

string code1 = "print 'importing string'\nimport string\nprint 'done'";
Py_Initialize();

pdict = PyDict_New();
pcode = Py_CompileString((char *)code1.c_str(), "<embed>", Py_file_input);

(void) PyEval_EvalCode((PyCodeObject *)pcode, pdict, pdict);
}

Output:
importing string

Thanks a lot,

arno


From mal at lemburg.com  Fri Feb  8 16:56:29 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 08 Feb 2002 22:56:29 +0100
Subject: mxODBC, stored procedure syntax
References: <e746a5a8.0202080441.4bd2c023@posting.google.com> <a40s9m$aif$1@suaar1aa.prod.compuserve.com>
Message-ID: <3C64498D.1256C99E@lemburg.com>

"Fran?ois Lepoutrre" wrote:
> 
> As far as i understand it there is currently
> no support for returning params from
> storedprocedures at the moment.

There is: let the stored procedure generate a result
set (or a temporary table) and then fetch the data
using the standard .fetchxxx() methods.
 
> I understand it is on their "to-do" list.

There will be direct support for .callproc() in
mxODBC 2.1 and support for output parameters is
planned for version 2.2.
 
> I'll like the support for odbc to be extended
> to quite a lot of other things...

Please post these to the egenix-users list ... we'll
see what we can do.
 
> But odbc works - simple and stable.
> 
> That makes python useful of us.
> Many thanks to M.A Lemburg :)

Thanks.
 
> Fran?ois
> 
> cb921 wrote in message ...
> >Hi,
> >
> >I have been trying to connect to an MS SQL Server via ODBC using mxODBC.
> >It's all working, but calling of a stored procedure is causing problems.
> >Without any syntax worries, it looks like:
> >
> >cr.execute("{call storedprocedure('cellnr','passwd')}")
> >
> >and it should return me two output parameters, in the example (not Py :)
> code
> >I've seen we call this @DATABASENAME and @ISNOM.  How do I retrieve those
> >values?  So far all I've managed is a flock of errors.
> >
> >Anyone who can help me, even an insignificant bit, thank you...
> >
> >Cheerio,
> 
> --
> http://mail.python.org/mailman/listinfo/python-list

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From starx at pacbell.net  Sun Feb 10 15:16:09 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 10 Feb 2002 20:16:09 GMT
Subject: lists, uppercase?
References: <mailman.1012427451.9143.python-list@python.org>
Message-ID: <Xns91B17CC9E7B40RASXnewsDFE1@130.133.1.4>

maximilianscherr || Wed 30 Jan 2002 01:49:28p:

> just two questions this time:
> 
> how can i change a list like this
> [1, 2, 3, 4]
> to
> [4, 3, 2, 1]
> 
> how can i get to know if a string is totally uppercase?

Well, you can't really have upper or lower case with just numbers, but i 
think you want a list that's sorted highest-to-lowest. To do this, just run 
both sort and reverse on the list:

Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.8 -- press F1 for help
>>> x = [4,1,3,2]
>>> x.sort()
>>> x
[1, 2, 3, 4]
>>> x.reverse()
>>> x
[4, 3, 2, 1]
 
After these two steps, you can be sure that the list is in decending order, 
if you want to check that without re-sorting it, you could use a function 
like this (basically 'if any item is less than (i.e. not >=) the next item, 
return false, else return true'):

>>> def is_descending( in_list ):
	for i in range(len(in_list) - 1):
		if in_list[i] < in_list[i+1]:
			return 0
	return 1

>>> is_descending(x)
1
>>> is_descending([1,2,3,4])
0
>>> is_descending([4,3,1,2])
0


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From gcordova at hebmex.com  Wed Feb 20 19:31:02 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Wed, 20 Feb 2002 18:31:02 -0600
Subject: two ideoms at one blow: line-reading and regexp-matching
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AC9@SRM201002>

> 
> =/== confusion is really a weak reason to not allow assignments in
> expressions.  Do you have some other reasons?  If not, prohibiting
> them is just imposing a subjective opinion of which method makes
> code more readable.  I certainly find
> 

It's not such a weak reason: it happens, a lot. And not only
among newbies, it's only a small typo, but it's hideous
in it's consequences. I like not being able to use "="
in an expression.

Now *THIS*:

>
>    while (m := re.match(pat,x)):  blah()
> 

Indeed, this is very readable. The ":=" asignment operator
is used in Pascal, if I remember correctly; another language
where you can't assign inside an expression.

>
> more readable than some multi-statement contortion for doing the same
> thing, or worse, using some kludgy special-purpose class that's
> defined far away in the code and maybe done differently from one
> program to the next.  It's possible that some other people find the
> contorted versions more readable, but they're not the ones who have to
> maintain my code.
> 

But it only feels contorted, to you. To me it feels quite
natural:

	1. Set up the loop conditions and variables.
	2. Do the loop.
	3. Exit the loop.

Yes, your ":=" suggestion is valid, and it removes the "=/=="
readability / typo reason. But if it's not there, and if
the people maintaining your code are going to get all huffy
because it's "spelled out" instead of being in a compact
form, then the problem is another, and not the language.

> Maybe there's a principle to be found here: "design by FAQ".  
> If newbies
> keep arriving and asking over and over again "how do I do X?", 
> maybe there's a good reason for having a way to do X.
> We've seen that already for augmented assignments.

Ahh, but augmented assignments are not assignments
inside expressions, but expressions inside assignments.

:-)

Then again, maybe all the newbies used other languages
which feature the assignment / expression duality. Also,
this "limitation" (of not allowing assignments in expressions)
is clearly expressed in the docs, which means that newbies
nowadays are simply NOT reading the documentation before
launching into new waters. That severely downgrades the
quality of the questions they might pose, no?

Just count how many times the "static vs dynamic typing"
thread has reappeared, it just won't die. And it's always
*another* newbie which comes in and asks the same question,
instead of learning, slowly, breathing calmly, how to best 
use this shiny new tool.

BUT, they keep coming to Python, why?

Is it because it lacks something?

Or is it because it has something beyond those other tools?

I'm sure that it'll keep evolving, it needs to. But there's
something inherently *nice* which comes into play when
the code's all spelled out, doing the clearest thing
possible, which just isn't there in Lisp, C, C++, etc.

Anyway, it's just my opinion. :-)

-gustavo



From randytalbot at home.com  Sat Feb  9 21:35:43 2002
From: randytalbot at home.com (Randy Talbot)
Date: Sun, 10 Feb 2002 02:35:43 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com>
Message-ID: <3c65d8a7.25247159@news>

On 9 Feb 2002 02:22:44 -0800, wookiz at hotmail.com (wooks) wrote:

>New to the language, new to OO but I do understand the basic concepts,
>not at all new to programming, trying to do the right thing by
>searching archives before asking questions. Yet I feel I need to be
>doing a university degree course in philosophy or mathematics to
>follow almost any discussion. If that's the way it is fair enough but
>Python is supposedly simple to learn (thus far I've had an easier time
>of it learning Perl). If someone comes and says I've had the following
>understanding of concept X since I was in grade 2 why is it not so in
>Python, the answers seem to be heavy on the philosophical
>justifications fo the Python way and light on the practical benefits.
>The answers also tend to be in a totally different intellectual
>context.
Have you looked at The Quick Python Book by Daryl Harms & Kenneth
McDonald http://www.manning.com/Harms/index.html  published by Manning
Publishing Co.? I am a newbie to programming and have found this book
to be the best help for me in understanding the python language in a
clear and concise way. With this book by your side you will be able to
follow majority of the discussions. I highly recommend this book. 
Randy Talbot


From fredrik at pythonware.com  Tue Feb  5 17:38:48 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 05 Feb 2002 22:38:48 GMT
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com> <2hy78.23175$l93.4207585@newsb.telia.net> <f72dac0d.0202050259.6bc6fe6f@posting.google.com>
Message-ID: <Y9Z78.23593$l93.4286859@newsb.telia.net>

Alan Kennedy wrote:
> I'll have you know I have a very responsible job at the Ministry ....
> of Silly Walks!

but of course!

when exposed to "I think <braindead use of buzzword-compliant
technology> would be a great idea" arguments, both experienced
programmers and successful silly walkers tend to reply with "why
not just keep it simple, stupid".

(both groups also know when to use brute force, of course)

</F>




From rhymes at myself.com  Sun Feb  3 05:10:45 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sun, 03 Feb 2002 11:10:45 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012671184.6510.python-list@python.org> <kjho5uc7ut6tt12898b62sthsq5hdofvqu@4ax.com> <Pine.SOL.4.30.0202022205420.597-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <n43q5u4d6s3lmj88p57654d0047fft9o4p@4ax.com>

On Sat, 2 Feb 2002 22:06:41 +0000, "John J. Lee" <jjl at pobox.com>
wrote:

>You probably didn't compile it in (or the setup.py didn't do it for you).
>Check Modules/Setup.

i have the binary distribution...

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From mal at lemburg.com  Thu Feb 21 12:12:32 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 21 Feb 2002 18:12:32 +0100
Subject: mxDateTime in Python2.2 on Tru64
References: <695dd93e.0202210758.3b64f79a@posting.google.com>
Message-ID: <3C752A80.31CA0A2A@lemburg.com>

Richard Papworth wrote:
> 
> Hi
> 
> After upgrading from Python 1.52 I'm having trouble getting mx
> Extensions to work with Python 2.2.
> 
> Compilation went OK (although config.h was missing from the Python
> includes directory). However, when I try to load up the DateTime
> module I run into problems:
> 
> Python 2.2 (#2, Feb 20 2002, 11:44:40)
> [GCC 2.95 19990728 (release)] on osf1V4
> Type "help", "copyright", "credits" or "license" for more information.
> >>> from mx import DateTime
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/nfs3/users/postgres/usr/local/lib/python2.2/site-packages/mx/DateTime/__init__.py",
> line 8, in ?
>     from DateTime import *
>   File "/nfs3/users/postgres/usr/local/lib/python2.2/site-packages/mx/DateTime/DateTime.py",
> line 441, in ?
>     MaxDateTime = DateTime(sys.maxint / 366 - 1,12,31)
> OverflowError: signed integer is greater than maximum
> >>>
> 
> The other mx extensions seem to be OK. Does anyone know what's going
> wrong?

Are you using the latest version ?

If not, here's the fix... 

DateTime.py:
# Limits (see also the range checks in mxDateTime.c)
MaxDateTime = DateTime(5867440,12,31)
MinDateTime = DateTime(-5851455,1,1)
MaxDateTimeDelta = DateTimeDeltaFromSeconds(2147483647 * 86400.0)
MinDateTimeDelta = -MaxDateTimeDelta

Note that there is an eGenix User Mailing list for these sort
of things:

	http://lists.egenix.com/mailman/listinfo/
 
-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From bokr at oz.net  Sat Feb 16 17:40:20 2002
From: bokr at oz.net (Bengt Richter)
Date: 16 Feb 2002 22:40:20 GMT
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4kv8g.3vva8ld.1@kserver.org> <xyuofipl5hh.fsf@sex.ifi.uio.no> <a4me60$brk$0@216.39.172.122> <xyuk7tdku0w.fsf@sex.ifi.uio.no>
Message-ID: <a4mn4k$gd2$0@216.39.172.122>

On 16 Feb 2002 22:15:43 +0100, Jan-Fredrik Braseth <janfbr at sex.ifi.uio.no> wrote:

>bokr at oz.net (Bengt Richter) writes:
>
>[snip]
>
>> # I would change opt to c in the following line
>>                     self.msg('IAC %d not recognized' % ord(opt))
>
>But if there really is something wrong with that, shouldn't someone
>send a bug report or something to the creators of Python?
>
No, not to the creators of Python per se. This is not a bug
in Python itself. Telnetlib is just a piece of the distribution.

But yes, the telnetlib authors might want to double check this.
In the source it says "Based on RFC 854: TELNET Protocol Specification,
by J. Postel and J. Reynolds," but the latter are the RFC 854 authors,
not (probably) the authors of telnetlib.py, so who is the maintainer?

... Hm, looking at CVS I recognize the loewis name. I will CC this to him
and see if we get a reaction ;-)
(Hi Martin - that was line 424 and opt is not set in that path. See prev post ;-).

Meanwhile you could just make a safekeeping backup copy of telnetlib.py
and then try the change and see what happens, though I'm not making any
warranties ;-)

Regards,
Bengt Richter



From phr-n2002a at nightsong.com  Wed Feb  6 08:43:07 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 06 Feb 2002 05:43:07 -0800
Subject: Saving Browser State without Cookies
References: <mailman.1012938609.3651.python-list@python.org> <3c603127.18127765@news.t-online.de>
Message-ID: <7x7kpqiv4k.fsf@ruckus.brouhaha.com>

gerson.kurz at t-online.de (Gerson Kurz) writes:
> Here is a very simple example:...
> # decode "cmd" arg
> def GetArguments(form):
>     result = { 'value':1 }
>     if form.has_key('cmd'):
>         try:
>             new_dict = eval(binascii.a2b_hex(form['cmd'].value))
>             ... 

NOOOOO!!!!  You can't let the client upload arbitrary stuff and eval
it in your Python app.  That allows a malicious client to totally
take over your server.

You also can't put the server side database password in that client
side structure, unless you want the client to be able to read it,
via "View Source" or by intercepting the http stream.

Please be careful about security when writing stuff like this!!!


From mcfletch at geocities.com  Sun Feb 10 14:23:32 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Sun, 10 Feb 2002 14:23:32 -0500
Subject: Standalone ZODB + BTree question
References: <cc498.2133$0c.92062@news2.east.cox.net>
Message-ID: <3C66C8B4.4020503@geocities.com>

If I'm understanding the question correctly...

You need to put your BTree into the database, that is, it needs to be 
made accessible from the root object set.

For example consider creating a folder class, whose items are stored in 
a BTree.  Your BTree would be stored as something like F.items, and you 
would put F into your folder hierarchy, and your folder hierarchy into 
the root of the database:


ZODB
   root
     ['myObjects'] = Folder1
       .items[1] = Folder2
         .items[23] = Folder4
           .items[2] = someObject

Hope that helps,
Mike

Van Gale wrote:
> Can anyone point me in the right direction for using OOBtree with ZODB?
> 
> I've been able to figure out FileStorage and DbmStorage easy enough from the
> docs, and I can create and use an OOBTree... I just can't seem to tie it
> into persistent storage.
> 
> The BTree unittest *seem* to just be running tests on in memory btrees, so I
> haven't been able to use them as an example.
> 
> --
> Van Gale
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From tim_one at email.msn.com  Sun Feb  3 17:54:40 2002
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 3 Feb 2002 17:54:40 -0500
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
In-Reply-To: <XPg78.22996$l93.4173649@newsb.telia.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEBKMJAA.tim_one@email.msn.com>

[/F]
> really?  I thought the Python C code used 4-space hard-tab
> indents...

Guido actually uses 7-space hard-tab indents, but his comments actually
align better if you use 8 <0.9 wink>.




From jeremy at alum.mit.edu  Tue Feb  5 00:05:02 2002
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: 4 Feb 2002 21:05:02 -0800
Subject: RELEASED - StandaloneZODB 1.0 release candidate 1
References: <mailman.1011990406.22651.python-list@python.org> <3C52A2E5.6020709@mxm.dk> <mailman.1012258202.16406.python-list@python.org> <3C5F1AAD.A16F6356@astro.cornell.edu>
Message-ID: <b0f083db.0202042105.6deec6f9@posting.google.com>

Tom Loredo <loredo at astro.cornell.edu> wrote in message news:<3C5F1AAD.A16F6356 at astro.cornell.edu>...
> Hi-
> 
> I am having difficulties building StandaloneZODB on a *Mac*
> (Classic, not OS X), even though I was able to build ZODB
> from stuff I threw together from the last official Zope release
> fine.  I tried it a couple weeks ago before the official
> release, and got farther than I was able to get with this
> new release candidate (which appears to have a heavily
> revised setup.py, and to have ignored some changes that
> Guido checked in specifically to help the code clear the
> Codewarrior compiler).

I'm not sure I follow you here.  We did make substantial changes to
the setup.py.  The changes allow a single binary install to be created
from the setup script; the old version would create several different
installers for the individual components of ZODB.

But Guido never made any changes to support Codewarrior.  He did make
a couple of changes to make bsddb3Storage optional, but those changes
have been preserved in the current version.  (I'm sure here; I checked
the CVS logs.)

Are you thinking of some other setup.py or some other project?
 
> I'm really pressed for time currently and have had to give
> up on this for the time being (esp. since I already have
> something that works).  If Mac compatibility is at all a
> priority for you folks, do let me know and I'll try to work
> more on it if I can.  In the past I have not found much
> enthusiasm for Mac compatibility in Zope land....

Mac compatibility would be nice, but we need to rely on third-party
contributions for it.  None of us have or use Macs, so we've got zero
chance of getting it right.  If you could at least explain what didn't
work, some other Mac user might be able to solve the problem.

Jeremy


From sholden at holdenweb.com  Wed Feb 20 08:32:52 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 08:32:52 -0500
Subject: HTTP state management without cookies?
References: <Xns91BAE535689F7Latzikatz@139.174.2.56> <G5Bc8.1$Az6.364@news.hananet.net> <3C7368A0.5050702@bath.ac.uk>
Message-ID: <KzNc8.17085$Zp2.1759804@atlpnn01.usenetserver.com>

"Simon Willison" <cs1spw at bath.ac.uk> wrote in message
news:3C7368A0.5050702 at bath.ac.uk...
> Jonathan Gardner wrote:
>
> > There are security problems. I tried to force the guy to be coming from
one
> > unique IP address, but I found that too many places have firewalls or
> > something that change the IP address of the user each time they access
the
> > page, so I had to abandon that.
>
> I hadn't realised some places had dynamically changing IP addresses. My
> security method was to save the User Agent, IP address and the time the
> session was created (so I can expire it after X minutes). I suppose you
> could deal with IP addresses by ensuring that the first part of the IP
> address remains the same i.e
>
> User logs on initially as 172.21.16.189
> Session then allows access from 172.21.16.* or 172.21.*.*
>
That would be pretty gash: all members of a subnet share the same session?

Even in your original scheme you arre relying on not getting two
simultaneous sessions fomr the same IP address. This is easily possible --
for example, when different X terminal users whose browsers run on the same
machine access your site.

Run, do not walk, to fix your code!

> It's pretty important to provide some kind of checking mechanism like
> that to avoid people copying/pasting the URL of the page they are on
> into an e-mail / instant message and inadvertantly giving their session
> to someone else.
>
Indeed, but it needs to be a little more rigorous. That's why a recent
thread concluded you EITHER pass session id through all URLs OR use a cookie
to identify the session.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py



--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From Lutz.Schroeer at kybernetik-manufaktur.de  Tue Feb 19 16:31:00 2002
From: Lutz.Schroeer at kybernetik-manufaktur.de (Lutz Schroeer)
Date: 19 Feb 2002 21:31:00 GMT
Subject: HTTP state management without cookies?
Message-ID: <Xns91BAE535689F7Latzikatz@139.174.2.56>

I'm creating a website which makes intensive use of cgi scripts and there 
is the possibility for the visitor to register as a user for having the 
ability to customize the site appearance, contents and other stuff.

Jumping from one script to the other I need to remember at least the user's 
name. The W3C recommends using cookies (RFC 2109). Unfortunately many 
people don't like cookies and filter them using WebWasher or a personal 
firewall. 

Is there any other technique for keeping the user's data while he is using 
my site without installing any large suite as Zope or the like? I've read 
an article about utilizing soft links but I didn't quite get the point.

Lutz
--
(http://www.kybernetik-manufaktur.de)


From clarence at netlojix.com  Sun Feb 10 03:37:18 2002
From: clarence at netlojix.com (Clarence Gardner)
Date: Sun, 10 Feb 2002 08:37:18 GMT
Subject: Confusion with __del__
Message-ID: <2jq98.54$Mk5.2196@newsfeed.avtel.net>

Here is a simple class that uses gdbm. The man page for gdbm says that
it is vital that the file be closed, so I add a __del__ to be sure:

Currently, the initialization of the instance always raises an exception,
so the call to create the instance also deletes it. If I run it interactively,
the write of "in close" happens in an infinite loop, which I don't
understand. 

If I run it under mod_python (which is where it will actually be used but
may be out of scope for this newsgroup), it writes "in close" twice and
then dumps core.  And the cherry on top is that if I remove the write of
"in close", mod_python then goes into an infinite loop also.

It seems to me that what should happen is:
    __init__ raises an exception
    __del__ gets called; it checks whether the "db" attribute exists;
         it doesn't; object is now deleted
Why does it call __del__ (or close) a second (or millionth) time? 

    class ExpirationFile:
        import gdbm
        def __init__(self, name, mode):
            wait = 0.1
            for attempt in range(20):
                try:
                    self.db = gdbm.open(name, mode)
                except:
                    time.sleep(wait)
                else:
                    break
            else:
                open('/tmp/xx','a').write('Unable to open expiration file\n')
                raise KeyError, 'Unable to open expiration file'
        def __setattr__(self, key, value):
            self.db[key] = value
        def __getattr__(self, key):
            return self.db[key]
        def close(self):
            open('/tmp/xx','a').write('in close\n')
            if hasattr(self, 'db'):
                open('/tmp/xx','a').write('attempting to close %s\n' % `self.db`)
                self.db.close()
                del self.db
        __del__ = close



-- 
Clarence Gardner
Software Engineer
NetLojix Communications
clarence at netlojix.com


From marklists at mceahern.com  Tue Feb 12 09:33:19 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Tue, 12 Feb 2002 06:33:19 -0800
Subject: isNumberType? a bug or a feature?
In-Reply-To: <20020212004024.035d005e.xunningy@ece.arizona.edu>
Message-ID: <NCBBLFCOHHDIKCAFGCFBAEGFKCAA.marklists@mceahern.com>

[Xunning Yue]
> Here is my situation with Python2.2 in Linux:
> >>> from operator import *
> >>> class dummy:
> 	pass
> 
> >>> t = dummy()
> >>> type(t)
> <type 'instance'>
> >>> isNumberType(t)
> 1
> 
> Is this a bug or a feature of Python? Can anyone answer this?

This post would be more readable if your sample code were:

>>> from operator import isNumberType
>>> class dummy:
	pass

>>> t = dummy()
>>> type(t)
<type 'instance'>
>>> isNumberType(t)
1

Cheers,

// mark



From sholden at holdenweb.com  Wed Feb 13 16:00:50 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 16:00:50 -0500
Subject: Help! IIS and Python
References: <8f8ffe67.0202131245.375fe55d@posting.google.com>
Message-ID: <ZtAa8.211294$KM2.8800436@atlpnn01.usenetserver.com>

"Sunit Joshi" <sjoshi at ingr.com> wrote in message
news:8f8ffe67.0202131245.375fe55d at posting.google.com...
> Hello
> I'm trying to use IIS5.0 with Python 2.1 and whenever I try to import
> modules I get an error. The test.asp and error are below.
> thanks
> Sunit
> sjoshi at ingr.com
>
> <%@ Language=Python %>
> <%
> import string
> server = Request.ServerVariables("SERVER_NAME")
> server = string.upper(server)
> clientIP = Request.ServerVariables("REMOTE_ADDR")
> %>
> <HTML><HEAD><TITLE>Python Test Page</TITLE></HEAD>
> <BODY>
> <p>
> <h3><font color="#990033">You are coming from <b><font
> color="#669900"><%=clientIP%></font></b>
> to <%=server%></font></h3>
>
> <hr>
> PythonAspTest
> </BODY>
> </HTML>
> Error Type:
> Python ActiveX Scripting Engine (0x80020009)
> Traceback (innermost last): File "<Script Block >", line 3, in ?
> server = string.upper(server) File "C:\Python21\lib\string.py", line
> 60, in upper return s.upper() File
> "C:\Python21\win32com\client\dynamic.py", line 438, in __getattr__
> raise AttributeError, "%s.%s" % (self._username_, attr)
> AttributeError: <unknown>.upper

The problem here appears to be that you are probably getting a COM object of
some sort back as a result of your call to Request.ServerVariables(), and
when you try to apply the upper() function from the string module it
complains that you  aren't apssing the right kind of object as an argument.

If you replace

    server = Request.ServerVariables("SERVER_NAME")

with

    server = str(Request.ServerVariables("SERVER_NAME"))

for example, you will get a Python string back and all should be copacetic.

I personally I find it so tiresome to interact with the Scripting
environment in this way I have now moved away form IIS completely, and use
Apache or Xitami as a web server. I recently had to go back to coding
VBScript for IIS, and it was painful.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From jpm at pessoal.org  Tue Feb 12 15:46:30 2002
From: jpm at pessoal.org (Joao Prado Maia)
Date: Tue, 12 Feb 2002 15:46:30 -0500 (EST)
Subject: [ANN] Papercut 0.7.2 - an usenet server written in python
Message-ID: <Pine.GSO.4.44.0202121530540.6047-100000@mercury.alloy.net>

Papercut is a multi-threaded news server written in Python. Its main
objective is to integrate existing web based message board software
(Phorum [http://phorum.org] on this case) with a Usenet front-end.
However, its extensibility enables developers to write their own container
for the storage of the Usenet articles (messages). That means that the
code is extensible enough that _you_ could write new containers to
integrate the news server with other web message board projects or even
other ways to store the messages.

The source code is available freely under the BSD license, which means you
can do whatever you like with it. I do welcome any and all contributions
to the project :)

Project' Site (Not much information right now):
http://pessoal.org/papercut/index.php

There is very little documentation right now, but it is something I have
an eye on working on in the next few weeks. However, there are basic
installation procedures and such.

The server should be pretty much 100% compliant with the NNTP protocol,
that is, where the protocol made sense and was needed for the project
objective. In any case, I would love to hear comments on the project.

Cheers,
Joao Prado Maia


p.s.: I would like to thank Steve Holden and Gerson Kurz for helping me
while trying to debug some issues with this server :)

--
Jo?o Prado Maia <jpm at phpbrasil.com>
http://phpbrasil.com - php com um jeitinho brasileiro
--
Precisando de consultoria em desenvolvimento para a Internet ?
Impleo.net - http://impleo.net/?lang=br





From carlca at dircon.co.uk  Thu Feb 21 18:33:18 2002
From: carlca at dircon.co.uk (Carl Caulkett)
Date: Thu, 21 Feb 2002 23:33:18 -0000
Subject: importing dlls with python
References: <e6bacef5.0202192029.2d095cc8@posting.google.com>
Message-ID: <MPG.16df93f9a542e9539896d8@newnews.dircon.co.uk>

In article <e6bacef5.0202192029.2d095cc8 at posting.google.com>, tkeon0
@netscape.net says...
> Hi, I am new to Python, and was wondering if it was possible to import
> dlls written in Delphi 5?  I have tried a number of methods to make it
> work, and I either get:

Hi Trevor,

First of all I would recommend getting "Python For Delphi" from:

http://www.multimania.com/marat/

Amongst other things, it contains a full Delphi translation of the 
Python API, as well as sample code for a DLL that can be imported into 
Python. This should be enough to get you started.

I am working on a Delphi project which builds upon "Python For Delphi" 
and which will allow you to write classes which are able to expose 
published properties and methods to Python, with virtually no extra 
coding. As soon as it is ready for human consumption, I'll post here and 
on the Borland newsgroups.

-- 
Carl



From khaghdad28 at yahoo.com  Tue Feb 19 21:29:40 2002
From: khaghdad28 at yahoo.com (Kian Haghdad)
Date: Tue, 19 Feb 2002 21:29:40 -0500
Subject: Software engineer
Message-ID: <20020220022938.UIER27257.tomts11-srv.bellnexxia.net@Babak>

To Whom It May Concern:

I have obtained your email from the Web and I am very interested in working for your company 
as a contractor.
I have 7++ years of experience with software development, Internet development and 
hardware. I have worked for many companies both in Canada and in the United States using 
different development environments.
I have a BS in Electrical Engineering and computer science and worked on my Master degree 
in Telecommunication.
I am looking for contract work only, please!

I have attached my resume and sincerely appreciate your consideration, thank you.

Sincerely,
Kian Haghdad

Here is my resume:

Kian Haghdad
8 Kingsbridge Crt., Apt. 508 
Toronto, Ontario M2R 1L5
Tel.: (416) 630-9989
Email : khaghdad28 at yahoo.com

I have more than 7+ years of experience in application development, Database Management, 
Internet development. I have also been involved in hardware design. My software 
development experience is mainly in Visual Basic, Visual C++, Visual J++, Visual InterDev, 
COM, DCOM, MFC, SDKs, MS-Access, MS-SQL, FoxPro, Java script and VB scrip, HTML and 
DHTML. Over the years I have been responsible for the Development of many applications 
from design to commercial release.

EDUCATION:

Sharif University OF Technology, Tehran (The best technical school in Iran)	9/87-6/92
B.S. Degree: Electrical Engineering and computer science
Worked on M.S. degree in Telecommunication engineering

EXPERIENCE:

Unisearch Associates, Concord, Ontario	3/01-Present
* Developed a server client application (GUI) in Visual C++6.0 using ATL, COM, DCOM MFC. 
The application controls the laser beam in UV instruments and communicates to LasIR using 
The Parallel port. The result (NO/CO2 and TOL/CO2 ratio) is outputted to the client's computer 
via RS232. For this purpose we also developed proprietary string based protocol 

* Developed a DCOM application using visual C++6.0 and embedded .Java script. The 
application is a data acquisition software which communicates with a spectroscopy software 
and gets the Data such as Temperature, pressure and gas concentration from a laser 
instrument (FTIR). Also the application gets other data from a whether station connected to the 
serial port at real time and using an I/O port draws different charts driving a plotter. User can 
control the data acquisition parameters by modifying the script file.
* Hardware and software design for a current loop with 6 channels to run a plotter using an I/O 
port.
* Transfer the windows based application to the micro controllers using PICMicro.

American SkySat, Walnut Creek, California	6/00-3/01
* Was involved in the development of a web application in Visual InterDev 6.0 using Active 
Server Pages (ASP), Microsoft E-Commerce, SQL 7.0, XML, Visual C++ 6.0 and Visual Basic 
6.0, Visual J++ 6.0, COM (ATL), DCOM, JavaScript and VB Script. The web server was 
Microsoft Internet Information Server (IIS), Microsoft site server 3.0, with Microsoft E-Commerce 
edition 3.0 and FrontPage extension running under the NT Server The web application 
provides NT related services online. The user can register and buy services and also allows 
the user to troubleshoot and configure a system at real time online. The web site also provides 
the capability of chatting online, customer support online, statistical analysis online and 
transaction online.

* Developed a database management application for a medical health care center, using 
Visual Basic 6.0, SQL 7.0, Crystal report and Access 2000. (Windows 98 and NT)

American Computech, Pleasanton, California	3/98-6/00
Developed different applications using Visual C++, Visual J++, and Access.

* Involved in the development of a CRM (Customer Relation Management)web application in 
Visual InterDev 6.0 using Active Server Pages (ASP), Microsoft E-Commerce, SQL 7.0, XML, 
Visual Basic 6.0, Visual J++ 6.0, COM, DCOM, JavaScript and VB Script. (Windows 98 and 
NT)

Developed an application in Visual Basic 5.0, Access 97 and Crystal reports (using ODBC) for 
financial analyses department of FHP (Concord, California). The application is capable of 
importing data from other systems, analyzing and processing the data and generating hundreds 
of reports base on the original data.

*Developed an MDI application in Visual C++5,0 and Access 97 for windows called ID maker. 
The application is designed to drive a digital camera, capture a picture and generate different 
ID cards using OLE (marketing for schools and clubs). The Cannon digital camera is controlled 
via DLL functions calls to the camera's driver. The camera is connected via a parallel port. I 
have also developed several OCXs for this project.

Developed an application in Visual Basic 5.0, Access 97 and Crystal reports (using ODBC). 
The application is a very large information management system for CBI (City Building Inc. in 
San Francisco). The application contains 28 relational tables large number of queries forms 
and reports.

Tavanir Co., Tehran, Iran	9/95-3/98
*Developed an application in Visual C++ 5.0 intended for network control managements. The 
application is able to get information such as voltage, active and reactive power from power 
plants and transmission stations and specifies transmission state using power flow method after 
a fault occurs on the transmission line. 

* Developed an application for voltage drop on the net. The application was developed in 
Visual Basic 5.0 using graphical OCXs.

* Developed hardware for the high voltage network.
*Developed an application in Visual C++ 4.0. The application checks the transmitting power on 
the lines and controls the system's steady state by alarm system.

*Developed an application in Visual C++ 4.0 set the protection relays in substations to protect 
the network's reliability.

Pishro Co., Tehran, Iran	11/94-6/95
Designed, developed and implemented electrical systems using computer aided software. The 
electrical systems were intended for both indoors (Buildings) and outdoors (Airports and 
Terminals).
The systems were designed and based on the standard and regulation of electrical system.

Pardis Tower Co, Tehran, Iran	1/93-11/94
Designed simulators for training purpose.
*Developed an application in  C++ for  cycloconverter simulator. The  
cycloconverter is used in the starters of synchronous motors.
*Developed application using C++  for control speed of  DC motors. The application
 gets input voltage to the motor, speed of motor and rotor's current. It can control
 the motor speed based on requested speed.
*Developed application using C++ for step motor's control speed. In order to change the angle 
of dish antenna, the application orders to step motor to change the rotor's situation.

SKILLS:
Software development
Visual C++, Visual Basic, Visual InterDev, Visual J++, MFC, SDKs, COM, DCOM, ATL ActiveXs 
, C++, Fortran, Intel  8085 Assembler Access.
Internet development
Visual InterDev 6.0, Active Server Pages (ASP), SQL 7.0, COM, DCOM, Java script. 
Database development
MS-SQL, Access, FoxPro
Operating System Used
NT, MS-Win, UNIX, VAX, MS-DOS

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kian_Resume.doc
Type: application/msword
Size: 17331 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20020219/708b82f5/attachment.doc>

From jason at jorendorff.com  Wed Feb  6 17:34:13 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Wed, 6 Feb 2002 16:34:13 -0600
Subject: multi-threaded *nix program needs to spawn an external program/script
In-Reply-To: <cd45264b.0202061417.590d8bf2@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLMENGDAAA.jason@jorendorff.com>

arbalt at yahoo.com wrote:
> Would it make sense to add a function to python for *nix systems
>     that would do the "start an executable file as a new process"?

Try os.spawnv(); it should work for both Unix and Windows.
  http://www.python.org/doc/current/lib/os-process.html

## Jason Orendorff    http://www.jorendorff.com/



From p.mayers at ic.ac.uk  Tue Feb 19 06:23:00 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Tue, 19 Feb 2002 11:23:00 +0000
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <a4s7mq$u07$1@news.idiom.com>
Message-ID: <20020219.112259.1884661237.30753@ic.ac.uk>

In article <a4s7mq$u07$1 at news.idiom.com>, "Jim Dennis"
<jimd at vega.starshine.org> wrote:

>  Of course there are limitations to this technique.  For example I see
>  that it's common for Pythonistas to use raise NotImplementedError for
>  abstract classes.  It would be nice of PyChecker could detect any time
>  a concrete subclass fails to implement any of the requisite abstract
>  functions.  Of course it's difficult to see how this would be
>  accomplished given current language features: how would we tell that a
>  given abstract function is "abstract" (is defined to raise
>  NotImplementedError)? how could we tell that a given subclass was
>  "concrete" (final)?  (Obviously we can see that any class that is used
>  to instantiate an object must be concrete, though not necessarily
>  "final", but that only works if we're given a whole project, otherwise
>  there might be classes which are intended to be concrete but not used
>  within the scope of our hypothetical scanning engine)
>  

Interfaces rock. "implements" is one of the only features I would like
from Java (well, proper static methods, and *maybe* synchronous methods,
but that's it).

Possibly some kind of equivalent to "use strict;" from Perl (ptooi!)

Cheers,
Phil


From akuchlin at mems-exchange.org  Mon Feb 25 18:04:04 2002
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: Mon, 25 Feb 2002 18:04:04 -0500
Subject: Tkinter.createfilehandler analog for Windows?
Message-ID: <20020225230404.GA27356@dust.mems-exchange.org>

The Tkinter.createfilehandler() method isn't supported on Windows,
only Unix and Mac.  So, when confronted with moving a Python program
that uses createfilehandler() to Windows, what do people do?  Use
threads?  Use Tcl's channel C API (but that seems inaccessible from
Python)?  Something completely different?

--amk                                                  (www.amk.ca)
Jenkins! Chap with the wings there -- five rounds rapid.
    -- The Brigadier, in "The Daemons"



From zero at example.com  Sun Feb 24 22:43:03 2002
From: zero at example.com (Zero)
Date: Mon, 25 Feb 2002 03:43:03 GMT
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com> <3C751FEA.24CEFBBC@ford.com> <slrna7i5v6.7kb.scarblac@pino.selwerd.nl> <3C7953CE.8D430F71@alcyone.com>
Message-ID: <bpie8.42812$TG1.1731146956@newssvr14.news.prodigy.com>

In the strictest of definitions, a "function" that does not provide a
return value is a "method"

In article <3C7953CE.8D430F71 at alcyone.com>, "Erik Max Francis"
<max at alcyone.com> wrote:

> Remco Gerlich wrote:
> 
>> No, he didn't. He thought the *function* should be false, not the
>> function's
>> result. An interesting idea, but possibly hard to implement (write code
>> to
>> decide if a function does nothing). Also, I haven't been able to come
>> up
>> with any situation where it would be useful yet :-)
> 
> It brings up the question of where you'd draw the line, as well.  Would
> it just be functions whose body is a pass?  How about ones which just
> explicitly return None (which have precisely the same effect when
> called)?  What about those that do something local without side effect
> and then return None or let control flow out?  Same thing again, but
> you're starting to wander into unclear territory.
> 
> But, as you say, the biggest problem with the idea is its total lack of
> utility :-).
>


From xeno at eskimo.com  Mon Feb 18 14:14:56 2002
From: xeno at eskimo.com (Xeno Campanoli)
Date: 18 Feb 2002 19:14:56 GMT
Subject: chomp equivalents
References: <a4p08q$2bi$1@eskinews.eskimo.com>
Message-ID: <a4rjrg$tfj$1@eskinews.eskimo.com>

Xeno Campanoli (xeno at eskimo.com) wrote:
: Is there a chomp equivalent in the "string" object?  rstrip doesn't seem to
: take off the newline.

: Also, I looked up chomp on the Python website search engine, and it said there
: is a chomp.  Am I reading this wrong?  I don't show any chomp procedure in my
: python environment.

Okay, rstip works fine.  I was braindead.  My apologies.


From narnett at mccmedia.com  Mon Feb 25 14:06:58 2002
From: narnett at mccmedia.com (Nick Arnett)
Date: Mon, 25 Feb 2002 11:06:58 -0800
Subject: All-on, all-off button for checkbuttons?
In-Reply-To: <m3y9hh8kuy.fsf@oak.ramandgita.com>
Message-ID: <NDBBLAEMOMEHKALAMKKJKEANJHAA.narnett@mccmedia.com>

I'm just bringing myself up to speed with Python, so forgive what I think is
an simple question.  I'm building a UI that has a bunch of checkbuttons,
which also includes buttons to turn all the checkbuttons on or off.  I see
how to turn them off individually, but it seems to me that there must be a
more elegant way.

The checkbuttons are created in __init__.  I'm thinking that there must be a
way to filter or iterate through the objects in __init__, and for each one,
if it's a button, select or deselect as appropriate.  I see how to get the
namespace as a dictionary, but that hasn't helped.

I've searched the archives of this group and found lots of hints, but not
the answer I'm looking for.  I suspect it lies in parts of the language I
haven't yet explored... I started with the UI, since that's where I'm
weakest.

Thanks,

Nick Arnett
narnett at mccmedia.com




From stojek at part-gmbh.de  Tue Feb 12 10:45:45 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Tue, 12 Feb 2002 15:45:45 GMT
Subject: read and write binary data (newbie)
Message-ID: <3c69359f.22621453@news.easynews.net>

Hi,

I am using Python quite a while for all kind of things. Now i have to
read a file (Windows) in binary format that contains strings (\0
terminated), int and floats (4 byte). and I am stunned because 
I don't know how to. 

to read strings I have written a function:
-----------------------------
def getstring(dat,maxlen=8):
    dummy=""
    spos=0
    while spos <= maxlen:
        c=dat.read(1)
        spos += 1
        if c!="\0" :
        	dummy += c
        else:
            return dummy
    return -1
----------------------------------

reading ints and floats is easy: 
x= dat.read(4)

but how can i convert x into a number?
and how can I write a float or int into
a binary file?

Could anybody please tell me where to look?

thanks
marcus


From eric.brunel at pragmadev.com  Thu Feb 21 14:42:03 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Thu, 21 Feb 2002 19:42:03 +0000
Subject: While loop and time.sleep function
References: <mailman.1014306035.24844.python-list@python.org>
Message-ID: <a53eqm$8tb$1@norfair.nerim.net>

Hi,

Alves, Carlos Alberto - Coelce wrote:
> Hi all,
> I tray to make a simple clock program ( see the code below ).
> Can someone explain to me why code 01 works while code 02 doesn't
[snip]

Eeer, in fact both codes didn't work... At least on my Linux box. The first 
problem is that there is no mainloop, so Tk never actually starts. The 
second problem is that since the control never returns to Tk (because of 
the "while 1"), Tk never gets a chance to refresh the display. So you must 
explicitely refresh it manually:

-----------------------
from Tkinter import *
import time

class Clock:
  def __init__(self,root):
    self.lb=Label(root,padx=5,pady=5,fg='blue',font=('Times',20))
    self.lb.pack()
    self.ck()
    self.update()
  def ck(self):
    self.lb.configure(text=time.asctime()[11:19])
    # Refresh the display; otherwise, nothing happens...
    self.lb.update_idletasks()
  def update(self):
    while 1:
      self.ck()
      time.sleep(1)

if __name__=='__main__':
  root=Tk()
  root.title('Clock')
  Clock(root)
  # Run Tk main loop
  root.mainloop()
-------------------------------

HTH
 - eric -




From peter at engcorp.com  Wed Feb 27 20:41:24 2002
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Feb 2002 20:41:24 -0500
Subject: Seeking Python book authors
References: <mailman.1014843285.2398.python-list@python.org>
Message-ID: <3C7D8AC4.3F9A74D0@engcorp.com>

Susan Capparelle wrote:
> 
> We're a publisher of high quality computer books currently seeking
> authors for a number of Python related books.  A couple of titles or
> topics would be; 'Enterprise system development with Python,'
> 'Practical Python' and 'Effective Python.'  If you can recommend
> anyone with the necessary experience and skills to undertake any of
> these books, please feel free to contact us.

Although I'm all warm and fuzzy about Python's continually
expanding domain of influence, I'm a little uncertain what
advantage more books would bring in the category of 
"general Python programming".  What is it about "Practical
Python" and "Effective Python" which existing books don't
cover adequately.

And perhaps more interestingly, what is it about "practical"
Python which makes it ineffective, and what about "effective"
Python programs make them impractical?

(Posted and emailed, in case you don't read the newsgroup.)

-Peter


From tim at worthy.demon.co.uk  Sat Feb  2 08:08:49 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Sat, 02 Feb 2002 13:08:49 GMT
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net> <313g3a.d22.ln@wombat.san-francisco.ca.us> <6vTfsIAio7W8EwNp@jessikat.fsnet.co.uk>
Message-ID: <37d27e024b.tim@worthy.demon.co.uk>

In message <6vTfsIAio7W8EwNp at jessikat.fsnet.co.uk>
          Robin Becker <robin at jessikat.fsnet.co.uk> wrote:

> well if I could get the win32com extensions to change passwords on
> different machines for me I would really like that

Why win32com rather than win32net (where it's easy)?

-- 
___
 |im    ---- ARM Powered ----


From starx at pacbell.net  Sat Feb 16 06:18:27 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 16 Feb 2002 11:18:27 GMT
Subject: Disabling creation of .pyc files?
References: <E16b1mB-0001Ls-00@mail.python.org> <mailman.1013619835.26532.python-list@python.org> <Xns91B51A11D5957RASXnewsDFE1@130.133.1.4> <mailman.1013694263.8903.python-list@python.org>
Message-ID: <Xns91B721A13B0AFRASXnewsDFE1@130.133.1.4>

Skip Montanaro || Thu 14 Feb 2002 05:45:16a:

> 
>     >> It's there for your benefit (imports are faster when up-to-date
>     >> .pyc files are present because the compilation step is
>     >> avoided), but yes, you can disable it by simply making the
>     >> directory containing the .py files read-only.
> 
>     Philip> Does this work under windows? I just set a folder
>     read-only (it Philip> is an option, but FIAK the 'properties'
>     window might lie or Philip> cheat regarging the underling attrib
>     calls, but...), and it Philip> didn't seem to prevent me from
>     adding files or changing names... 
> 
> Can I finesse the issue and just say, "get a real OS"? ;-) Sorry, I
> don't recall the original posted indicating the operating system.  On
> Unix-oriented systems, making a directory read-only is sufficient.

Heh, well I just decided to try it, and it didn't work, even though the 
option _was_ there. I figured you were talking about a nixish, but if 
anyone tried this under windows and got confused, well, it's not really 
their fault ;). 

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From eppstein at ics.uci.edu  Tue Feb  5 14:32:33 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Tue, 05 Feb 2002 11:32:33 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org>
Message-ID: <eppstein-A1F53B.11323305022002@news.service.uci.edu>

In article <mailman.1012934533.15278.python-list at python.org>,
 "Dr. David Mertz" <mertz at gnosis.cx> wrote:

> |"Martijn Faassen" <m.faassen at vet.uu.nl> wrote ...
> |> I suppose it might be a brain thing. I use my left hand for writing and
> |> drawing while I use my right hand for more coarse-grained tasks, like
> |> throwing a ball and such.
> 
> "Steve Holden" <sholden at holdenweb.com> wrote
> |Strange. I don't come across many other people who do this. I've always
> |thought my brain must have been differently wired for some reason I'll never
> |know.
> 
> Same story for me.  I don't think it is that unusual though.  Culture is
> sort of geared to the right-handed majority, so us folks who are mostly
> left-handed wind up learning the things they can do "well enough" with
> the right hand in the common style.

In my case it's not a matter of coping with the off-hand, and the 
subdivision of tasks seems to be more random, not based on coarse or fine 
grain.  I write and throw with my right hand (my left hand writing is not 
bad and probably would improve with practice) but shave primarily with my 
left hand, can spin frisbees only on my left hand, can use chopsticks or 
eat spaghetti with either hand, and can catch equally well with either 
hand.  But about the only hand skill relevant to Python is typing, which I 
(and probably most of us) do both-handed...
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From lac at strakt.com  Fri Feb  1 09:26:47 2002
From: lac at strakt.com (Laura Creighton)
Date: Fri, 01 Feb 2002 15:26:47 +0100
Subject: Tkinter widgets: Write options *and* option tags dynamically from dictionary? 
In-Reply-To: Your message of "Fri, 01 Feb 2002 13:20:26 GMT."
             <3c5a84eb.70228013@news.online.de> 
References: <3c5a84eb.70228013@news.online.de> 
Message-ID: <200202011426.g11EQlGo003932@ratthing-b246.strakt.com>

After you make your dictionary, you need to invoke it like this:

	widget=widgetType(master, **mydict)

Laura Creighton



From cliechti at gmx.net  Tue Feb 12 19:27:13 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 13 Feb 2002 01:27:13 +0100
Subject: Re-throwing an exception without loosing the stack trace
References: <mailman.1013552683.30067.python-list@python.org>
Message-ID: <Xns91B4F52BC58cliechtigmxnet@62.2.16.82>

"Clark C . Evans" <cce at clarkevans.com> wrote in 
news:mailman.1013552683.30067.python-list at python.org:

> How do I catch and then re-throw an exception without
> loosing the information about the stack trace?  I've
> looked around alot and don't see the "obvious" answer.

try a simple, naked "raise" ;-)

> Thanks!
> 
> Clark
> 
> P.S. I need to catch exceptions to clean up some
>      resources (database locked table), but
>      then I want to re-throw the error... in
>      it's original stack trace so that I can
>      find it... loosing the stack trace makes
>      for brutal debugging.

look at the traceback module too.
you can print out an exception with
"traceback.print_exc(file=sys.stderr)" and still continue your prog.
you can also save a copy of the stack trace with
"sys.exc_info()" and evaluate it later.

chris

-- 
Chris <cliechti at gmx.net>



From ut at example.com  Sun Feb 10 15:28:06 2002
From: ut at example.com (Unintelligent Troll)
Date: Sun, 10 Feb 2002 20:28:06 GMT
Subject: ip address?
References: <mailman.1012006124.8671.python-list@python.org>
Message-ID: <qJA98.711$Fa3.335863991@newssvr14.news.prodigy.com>

As written, this returns '127.0.0.1' on most non-windows platforms. This
is not likely to be the result the original poster wants.

In article <mailman.1012006124.8671.python-list at python.org>, "Tim Peters"
<tim.one at home.com> wrote:

> [maximilianscherr]
>> how can i get the internet ip address of my connected pc?
> 
> Give this a try; your money back if it blows up <wink>:
> 
> import socket
> myip = socket.gethostbyname(socket.gethostname())
> 
>


From cliechti at gmx.net  Fri Feb 15 15:01:29 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 15 Feb 2002 21:01:29 +0100
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net> <Xns91B6CF519E265cliechtigmxnet@62.2.16.82> <slrna6qsm2.bob.arturs@aph.waw.pdi.net>
Message-ID: <Xns91B6D66B8F9BFcliechtigmxnet@62.2.16.82>

Artur Skura <arturs at iidea.pl> wrote in 
news:slrna6qsm2.bob.arturs at aph.waw.pdi.net:
> Mmm, I don't know what's wrong, but after adding longer entries I
> got a result like this:
> http://7bulls.com/~askura/awt.png
> 
> It's not what it's supposed to be, the second part got lost.
> What is more, I will need to compose dialogs with much more text
> (whole paragraphs). The software is for kids, they will have to
> insert correct word into the text entries. 
> Seems I will have to calculate everything anyway...

lines can be built with a vertical layouter either by writing
one yourself (i could send you some code for that) or by using a 
GridLayout of panels. (a vertical layouter can be found in JStruct
on my page: http://geocities.com/chrisliechti/en/java/index.html.
that program was an experiment and is unfineshed, but you can easily 
try out swing and dynamic layouting...)

too long lines get wrapped around with FlowLayout the missing data is 
maybe below? your window look a bit wide, have you resized it?

chris

-- 
Chris <cliechti at gmx.net>



From deathtospam43423 at altavista.com  Sun Feb 24 17:37:01 2002
From: deathtospam43423 at altavista.com (A.Newby)
Date: 24 Feb 2002 22:37:01 GMT
Subject: Python friendly webspace?
Message-ID: <Xns91C05CA891090AusYourStandingInIt@130.133.1.4>

If you have Python cgi scripts, what's the easiest way to get them 
installed on a server on the web? 

Do u need to install the Python interpreter on the server? If so, will most 
ISPs be happy to let you do that?

Or is it possible/easy to freeze your scripts into standalone executables 
that'll function on their own?




-- 
cdewin at dingoblue.net.au


From cliechti at gmx.net  Wed Feb 20 17:39:23 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 20 Feb 2002 23:39:23 +0100
Subject: No space left on device
References: <mailman.1014239691.10707.python-list@python.org>
Message-ID: <Xns91BBF134B416Ccliechtigmxnet@62.2.16.82>

sag at hydrosphere.com (Sue Giller) wrote in
news:mailman.1014239691.10707.python-list at python.org: 
> I am hoping someone here can give me some insight.
> 
> I have a program that generates a LOT of small files in a 
> Windows2000 subdirectory.  It is my understanding (often faulty) 
> that there is no limit to the number of files such a subdirectory 
> can  contain.  However, I am getting an error when I hit about 
> 21,843  files.
> 
> IOError: [Errno 28] No space left on device: 'c:\\legacystoret\\~

see below in code

> I have over 5 G of space on the drive, and the test files are 0 len 
> anyway.  (It still happens if the files are not 0 len)

nono.. there is a cluster size. this space is minimaly allocated for a 
file and all other sizes are rounded up to a multiply of that size.
on a 5GB drive such clusters can be 32k or 64k. you can fill your 
drive with files with no real content....

> I can then create additional files in that directory from within a
> DOS window, but I cannot rerun my file generator to make even one 
> more file. So I am wondering if the limit comes from Windows, from 
> Python 'open...' or some other innards of Python.

ok so you have space left on that drive..

> I'm hoping someone can shed some light as to where this error is 
> coming from.  I am hoping I don't have to totally change my file-
> generating system to lessen the number of files, since 22,000 is 
> just a start on the count I may need.

maybe you should consider a database?

note that filling up your system drive can cause many problems such as 
causing strange malfunctions of windows. use a separate partition to 
fill like that.

> # ----- CODE -----
> # manyFiles - generate many files with temp names to see if we 
> reach a limit
> 
> import tempfile
> 
> maxFileCount = 100000           # fails at about 21,843 files
> seed = 'tst'                              # base for generated
> filenames base = r'c:\\storetlegacy'          # dir that gets the 
files
> 
> for ii in xrange(maxFileCount):
>     tempfile.tempdir = base
>     temp = tempfile.mktemp(seed)
>     fileName = "%s.pck" % (temp)    

is the mktemp module save for such many files? i think it runs out of 
random numbers....
try adding a distinct number like your loopcounter
  fileName = "%d.pck" % (ii) 

>     fp = open(fileName, 'wb')     # error comes here
>     fp.close()
>     if ii % 50 == 0:
>         print "filecount", ii               # kekep track of how far 
we
>         are 
> along the count
> 
> 



-- 
Chris <cliechti at gmx.net>



From mats at laplaza.org  Fri Feb 22 16:34:51 2002
From: mats at laplaza.org (Mats Wichmann)
Date: Fri, 22 Feb 2002 21:34:51 GMT
Subject: certification (Brainbench)
References: <2n8b8.29223$Zu6.141297@news-server.bigpond.net.au> <mailman.1013800706.14636.python-list@python.org> <a4jsk2$q4e$0@216.39.172.122> <mailman.1013807725.17084.python-list@python.org> <XS5c8.68610$om6.1895071@news1.tin.it>
Message-ID: <3c76b29c.19147011@news.laplaza.org>

:A worse issue I have with most tests (supervised or not) is their
:"quiz" nature.  Some people are very effective at taking quizzes (I,
:for one), and the correlation between that and the ability to
:perform useful work, while not null, isn't necessarily all that
:strong either.  For Python, at least, I think a few simple,
:well-specified programming tasks (one also requiring design,
:one working to a pre-specified design, one about fixing a bug
:in a given piece of code, another about adding functionality
:to another given piece of code) might well be more meaningful.
:
:Quizes are great for fun -- and some technologies don't really
:lend themselves well to directly testing a candidate's performance
:on somewhat meaningful tasks -- but Python does, so why not take
:advantage of that?

Because they're a lot harder to grade "automatically"?

When I had to write a final exam for a course I felt quite hamstrung
because the vendor uses only "read the bubbles on the form"
technology.  I know that testing technology is far more advanced than
that, but it's still easier to determin the results if the answers are
absolute, and not affected by, say, coding style.


Mats Wichmann



From jeff at ccvcorp.com  Fri Feb 22 14:01:05 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 22 Feb 2002 11:01:05 -0800
Subject: is their any limit on the size of the data passed to a program via 
 popen2?
References: <3C768631.9050803@student.kun.nl>
Message-ID: <3C769571.D99AE432@ccvcorp.com>


husam wrote:

>
>   pdbdir = os.popen('ls /home/data/pdbs') # if this directory contains
> 10000 files, the code does not work.
>
> files=pdbdir.readlines()
> pdbdir.close()

What you should do, in this case, is not use os.popen().  :)

Use os.listdir('/home/data/pdbs') instead.  It's faster, simpler, and much
more straightforward.  Or, if you need to limit filetypes even further, you
can use glob.glob().  Either of these is far preferable to using your
popen() call (which is almost equivalent to catching the subway, in order to
cross the street...)

Jeff Shannon
Technician/Programmer
Credit International




From mdk at mdk.com  Thu Feb 14 10:18:10 2002
From: mdk at mdk.com (MDK)
Date: Thu, 14 Feb 2002 10:18:10 -0500
Subject: Is making Python executables a reality?
References: <a4endc$1ehli3$1@ID-98166.news.dfncis.de> <Xns91B55AA48E7DCgmcmhypernetcom@199.171.54.215>
Message-ID: <a4gkfj$3up8$1@ID-98166.news.dfncis.de>

"Gordon McMillan" <gmcm at hypernet.com> wrote in message
news:Xns91B55AA48E7DCgmcmhypernetcom at 199.171.54.215...
> MDK wrote:
>
> > What percentage of people make exe files?  Do most people have
> > end-users download the Python environment?
>
> I can't even guess about percentages. Lately, Installer has
> been downloading at a rate of around 200 per week (about
> 50 or those for Linux / Unix). The mailling list has a few
> dozen subscribers. I have over 1,000 private emails on
> Installer[1]. Posts on c.l.py for py2exe seem about on par
> with Installer posts. So, as a WAG, I'd say 10K people
> making exe's isn't unreasonable (it's certainly thousands).
>
> -- Gordon
> http://www.mcmillan-inc.com/
>
> [1] Another project of mine downloads at about 15 / week,
> and I have never received an email from a user.

Okay, I am going to try this.  If I have any problems with it I'll post
here.

Thanks.




From phr-n2002a at nightsong.com  Sat Feb  2 06:02:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 02 Feb 2002 03:02:36 -0800
Subject: Enhanced Generators - two-way generators
References: <a3at9p$2he$1@bob.news.rcn.net> <20020201144050.GA28252@hishome.net> <004101c1ab5b$aa2e01c0$2df6a4d8@othello> <mailman.1012635904.10159.python-list@python.org>
Message-ID: <7xr8o4uoxf.fsf@ruckus.brouhaha.com>

Oren Tirosh <oren-py-l at hishome.net> writes:
> This year is the 30th anniversary of Unix pipes, invented by Doug McIlroy.  
> I think this is enough experience to build on.  In practice, pipe 
> programming is used for pipelines of processing stages, not for dialogue. 
> Dialogues are used for terminals with human operators, but are very rare as
> a building block inside a program.  They are mostly used when a tool 
> designed for human interaction is used in an automated manner, not as a
> first choice.

Two-way pipes in Unix are called sockets, and they are very important.


From jwilhelm at outsourcefinancial.com  Fri Feb 22 15:08:57 2002
From: jwilhelm at outsourcefinancial.com (Joseph Wilhelm)
Date: 22 Feb 2002 13:08:57 -0700
Subject: Python 2.1 and PIL
In-Reply-To: <94id8.563$s03.36009@news2.east.cox.net>
References: <94id8.563$s03.36009@news2.east.cox.net>
Message-ID: <1014408538.2043.10.camel@jwilhelm.ofsloans.com>

On Thu, 2002-02-21 at 19:31, william wrote:
> Ok been playing with this python 2.1.2 and PIL.  The problem is when i
> install the PIL it installs it for python1.5.  I don't know how to get the
> PIL to install for Python2.1.
> 
> I am assuming I need to make some changes in some files before i do
> ./configure.  Any suggestions?
> 
> Thanks.
> 
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list

Try this, before you run the configure, you should set your 'PYTHON'
environment variable to point to python2

export PYTHON=/usr/bin/python2

That 'should' fix it, I believe.
Also, in your configure, you may want to specify:

./configure --prefix=/usr --sysconfdir=/etc

Just to make sure everything goes in standard locations.

--Joseph Wilhelm




From phr-n2002a at nightsong.com  Tue Feb  5 04:35:34 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 05 Feb 2002 01:35:34 -0800
Subject: Size of integers (2**32-1)
References: <3C5FA2F4.8020905@thomas-guettler.de>
Message-ID: <7x665cz2xl.fsf@ruckus.brouhaha.com>

Thomas Guettler <st-newsgroups at thomas-guettler.de> writes:

> I have two python version installed:
> with the cygwin version "print 2**32-1" works
> but with the zope version I get OverFlowError.
> 
> Background:
> I want to get a random 32-Bit integer like this randrange(0, 2*32-1)

Values of sys.maxint must be different.  Try: randrange(0,2147483647).


From phr-n2002a at nightsong.com  Tue Feb 12 23:13:36 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 20:13:36 -0800
Subject: searching backwards in a string
References: <mailman.1013549132.16176.python-list@python.org> <7x8z9y46l1.fsf@ruckus.brouhaha.com> <mailman.1013571921.18465.python-list@python.org>
Message-ID: <7xvgd2589b.fsf@ruckus.brouhaha.com>

Skip Montanaro <skip at pobox.com> writes:
>     Paul> here 50000 and -1 are the starting and ending locations.
> 
> If start > end isn't direction == -1 implied?

Nope.  I tried it and it didn't search backwards.  The behavior
was actually a little bit peculiar and may not have been consistent
with the docs.  I'll try it again later and report a doc bug if 
it's inconsistent.


From jkraska at san.rr.com  Sat Feb  2 11:52:33 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 02 Feb 2002 16:52:33 GMT
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>
Message-ID: <ep6o5ucb1l79uqtpiv9que0ipsn0l9dsk1@4ax.com>

>I was wondering if there would be interest in such a e-book and if it would
>sell?  I would love to here everyone's opinion.

I think E-book is an idea which really is before it's time.

C//



From notjonathan at yahoo.com  Wed Feb 20 15:12:41 2002
From: notjonathan at yahoo.com (Jonathan Miller)
Date: Wed, 20 Feb 2002 12:12:41 -0800 (PST)
Subject: PIL jpeg problems
Message-ID: <20020220201241.24820.qmail@web10903.mail.yahoo.com>

Hello--

I'm having trouble using jpegs with PIL on my ISP's server--FreeBSD
4.1.1-STABLE, Python 2.0 (#4, Nov 22 2000), GCC 2.95.2 19991024 (release), at
pair.com--a good box.

- I've compiled PIL from sources successfully
- No errors in gcc output (-Wall)
- latest version of libjpeg (6b), libz, etc.
- all the tested functions (almost!) work normally within Python

However, jpegs will not decode.

Here's a paste:

>>> import Image
>>> im = Image.open("wee_kirke.jpg")
>>> im.mode
'RGB'
>>> im.size
(400, 300)
>>> im.save("wee_kirke.png")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/home/peasandc/pil/PIL/Image.py", line 709, in save
    self.load()
  File "/usr/home/peasandc/pil/PIL/ImageFile.py", line 140, in load
    d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
  File "/usr/home/peasandc/pil/PIL/Image.py", line 243, in _getdecoder
    raise IOError, "decoder %s not available" % decoder_name
IOError: decoder jpeg not available
>>>

I've not had trouble with prior compilations here, hence my question.

- Other images cross-save just fine, both paletted and unpaletted.

- All jpegs I've tried (progressive and not, small and large, etc.) have failed
similarly.

Any help would be mightily appreciated.

Thanks,
-Jonathan Miller

__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From gmfawcett at operamail.com  Thu Feb 14 00:05:40 2002
From: gmfawcett at operamail.com (Graham Fawcett)
Date: 13 Feb 2002 21:05:40 -0800
Subject: controlling web browser in windows
References: <1103_1013605186@netnews.cf.ac.uk>
Message-ID: <3d799735.0202132105.7257a410@posting.google.com>

Alan,

Automated browser control might be the best-suited solution to your
problem, but there are other ways to "interfere" with a person
browsing the Web. For example, an HTTP proxy server application could
sit as a mediator between your subject and the Web, logging actions
and redirecting the Web browsers as needed. (Python could do that
too!) It depends on what you're trying to accomplish.

Would you be willing to share what you mean by "completing a task" on
a web page, and why the "back button" control is important?

Regards,

-- Graham Fawcett
-- Centre for Flexible Learning
-- University of Windsor, Canada


From joshm at taconic.net  Wed Feb 20 21:01:16 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Wed, 20 Feb 2002 21:01:16 -0500
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com>
Message-ID: <3c745276_1@corp.newsgroups.com>

What about OpenGL?  It has a Python wrapper, I believe.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From tdelaney at avaya.com  Mon Feb 18 18:51:02 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 19 Feb 2002 10:51:02 +1100
Subject: Adding static typing to Python
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A195@natasha.auslabs.avaya.com>

> From: ajeru at vknn.org [mailto:ajeru at vknn.org]
> 
> In Python I'd have to test the whole software suite including each and
> every branch in each and every if statement to be sure that I didn't
> forget something. To build such a test suite would take almost as long
> as it takes to build the whole software and I'd have to maintain the
> test suite as well. That's just unafordable.

This is precisely what you should be doing anyway. Changing a method
signature is a *minor* thing in any language compared to changing the
semantics of a method call. Your unit tests should have as close to 100%
coverage as possible (GUI work is one of the few cases where such coverage
often can't be met).

I can tell you that normally my unit tests take at least as long to write as
the supporting code - but it tells me if I change something fundamental.
Now, if it is a design change it may necessitate changing the unit tests.

I am currently looking at doing a study at work into using Python for unit
tests for Java code (using Jython and wrapping JUnit). There are a number of
advantages I see in it:

1. The unit test code will be shorter and more readable (incredibly
important). In particular, the ease of creating collections (tuples, lists,
dictionaries) are a major advantage here.

2. You don't need to entire framework in place to develop the test suite.
Currently, in order to write the unit test, every method that will be tested
needs to have at least a minimal implementation simply in order to be able
to compile the test. Using a python test suite, I could write the unit tests
ahead of time and get import errors, attribute errors, etc for missing
classes, methods, etc. This is a Good Thing(TM) - it allows for full
specification of the tests ahead of any actual coding.

Writing comprehensive test suites is not unaffordable - on the contrary, it
tends to mean that subtle errors are caught much earlier.

Tim Delaney



From DeepBleu at DeepBleu.org  Fri Feb  8 19:11:02 2002
From: DeepBleu at DeepBleu.org (DeepBleu)
Date: Sat, 09 Feb 2002 00:11:02 GMT
Subject: Python ActiveX
References: <2030665d.0202080646.3126aec0@posting.google.com>
Message-ID: <qOZ88.121898$XZ1.5175731@typhoon.austin.rr.com>

This is quite interesting.  I know there is one (although outdated I
believe) for TCL.  For Python?  This could be nice.
DeepBleu

"Sam Collett" <sam_collett at lycos.co.uk> wrote in message
news:2030665d.0202080646.3126aec0 at posting.google.com...
> Is it possible to have python installed as a simple activex control
> (like jscript.dll and vbscript.dll)?
>
> Using code similar to the following would install an activex control
> (the following code installs msxml ver 4)...
> <object classid="clsid:88d969c0-f192-11d4-a65f-0040963251e5"
> codebase="msxml4.cab" width="0" height="0">
> </object>
>
> Can this be done with python as well e.g.
> <object classid="clsid:[python clsid code here]" codebase="python.cab"
> width="0" height="0">
> </object>
>
> I ask this because the user may not need all the extra features
> (Pythonwin win32all etc) - it is quite a big download.
>
> I hope to use it as a scripting language in html code:
> <script language="python">
> python statements here
> </script>




From jason at jorendorff.com  Mon Feb 18 14:41:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 18 Feb 2002 13:41:57 -0600
Subject: Object copying itself
In-Reply-To: <Ck5c8.68519$om6.1892014@news1.tin.it>
Message-ID: <HFEKILOLEFEFMKAECNDLGEHPDDAA.jason@jorendorff.com>

Alex Martelli wrote:
> No, but following that with a specific and explicit deepcopying of
> the lists looks like a good, clear way to signal this extremely
> peculiar design intent that the OP has asserted.

I'm pretty sure the OP intended:

    I need a deepcopy of __dict__, except don't copy instances
    at all (instead, have the copy refer to the original instance
    directly).

That's not what was said, so some sketchy inference is involved.
Eventually the OP will followup, I think, and straighten things out.

## Jason Orendorff    http://www.jorendorff.com/




From eric.brunel at pragmadev.com  Thu Feb 14 07:43:21 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Thu, 14 Feb 2002 12:43:21 +0000
Subject: Layout management in tkinter.
References: <3C6B9001.E5568D2C@lucent.com>
Message-ID: <a4g7lr$oqb$1@norfair.nerim.net>

Hi Lukasz,

Lukasz Pawelczyk wrote:
> Hello,
> 
> I try to change the geometry manager pack into grid in the code
> and I have problem.
> Why it does not work?
> 
> prz1=Button(main,text="Otworz",cursor="question_arrow",command=otworz)
> prz1.grid(row=0)
> 
> prz2=Button(main,text="Zachowaj",cursor="dot",command=zachowaj)
> prz2.grid(row=0,column=1)
> 
> prz3=Button(main,text="Zamknij",cursor="pirate",command=wyjscie)
> prz3.grid(row=0,column=2)
> 
> plik=ScrolledText(main,background="grey")
> plik.grid(row=1,column=0,columnspan=3)                - there the code stop

You didn't mention where you got the ScrolledText widget, but apparently 
you did a "from ScrolledText import ScrolledText", isn't it? If it is, I do 
have the same problem: apparently, the creation or gridding of the 
ScrolledText tries to pack something in the "main" widget, which results in 
program hanging.

I would advise not to use the ScrolledText widget from the ScrolledText 
module since it is obviously not so well designed... Either use the Text 
and Scrollbar widgets from Tkinter (you'll have to bind them to each other 
explicitely, but it's quite easy), or use the ScrolledText widget 
from PMW (Python MegaWidgets) that you'll find at:
http://pmw.sourceforge.net/
You'll also find in PMW a great deal of other very useful widgets.

HTH
 - eric -


From graz at mindless.com  Wed Feb 27 08:03:22 2002
From: graz at mindless.com (Graham Ashton)
Date: Wed, 27 Feb 2002 13:03:22 +0000
Subject: Python GUI
References: <uZ_e8.24022$dj3.835926@typhoon.austin.rr.com> <MPG.16e6b1eec8fe2c9989684@news.chello.at>
Message-ID: <rO4f8.1436$51.31621@wards>

On Wed, 27 Feb 2002 08:06:06 +0000, Markus Banfi wrote:

> *)
> Glade (http://glade.gnome.org/) with
> Glade Python Code Generator (http://glc.sourceforge.net/) and PyGTK
> (http://www.daa.com.au/~james/pygtk/) and gladepyc
> (http://www.fcoutant.freesurf.fr/gladepyc.html)

Check out gladebase too:

  http://www.bioreason.com/~brian/GladeBase.html

It makes it very easy to use the MVC design pattern with libglade, and I
heartily recommend it. To find out how to use it read the powerpoint
tutorial on the site.

-- 
Graham


From jbauer at rubic.com  Mon Feb 18 09:56:04 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 18 Feb 2002 08:56:04 -0600
Subject: Python 2.2 <-> Solaris 2.6 installation errors
References: <mailman.1013616443.8156.python-list@python.org> <3cv07uk543872s04lqqupeu2q4ejl2elu8@4ax.com>
Message-ID: <3C711604.86F7B187@rubic.com>

Chris,

Thanks for your note.

> I checked the log you posted in sourceforge bug log and 
> compared it to mine.  You are clearly getting a lot further
> along the 'make install' step than I am hence I must be
> doing something wrong.

> Basically I disabled the exception handling around the
> module import in the setup.py to get a more meaningful
> error message and got something equivalent to the following:

I've considered something similar to see what distutils 
may be doing.  I had already tweaked LD_LIBRARY_PATH.

I'm a little frustrated (though not much) that I've been
able to install Python 1.5.x through 2.1.x on this machine
using the simple boilerplate (configure, make, make install), 
but now have to dig into the installation scripts to
figure out what's breaking.  Also the errors don't
show up until the end of 'make install' -- a 10 minute
wait for each iteration.

Jeff Bauer
Rubicon Research



From nas at python.ca  Fri Feb 15 14:01:59 2002
From: nas at python.ca (Neil Schemenauer)
Date: Fri, 15 Feb 2002 11:01:59 -0800
Subject: Warning about "for line in file:"
In-Reply-To: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu>; from owen@xastrox.xwashingtonx.edu.invalid on Fri, Feb 15, 2002 at 10:23:46AM -0800
References: <owen-0B3ECB.10234615022002@nntp2.u.washington.edu>
Message-ID: <20020215110159.B12454@glacier.arctrix.com>

Russell E. Owen wrote:
> The readline or xreadline file methods work fine, of course.

Why "of course"?  iter(file) does the same thing as file.xreadlines().
Have you tested xreadlines?

  Neil



From kern at taliesen.caltech.edu  Tue Feb 26 05:50:38 2002
From: kern at taliesen.caltech.edu (Robert Kern)
Date: 26 Feb 2002 10:50:38 GMT
Subject: Time to rename Stackless?
References: <mailman.1014690284.11692.python-list@python.org>
    <m3pu2sancq.fsf@chvatal.cbbrowne.com>
    <eppstein-42B711.20584025022002@news.service.uci.edu>
    <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547@4ax.com>
Message-ID: <a5fp9u$dkd@gap.cco.caltech.edu>

In article <3f6m7ukfo7lvj9t4lr7hfktiu9uf9u1547 at 4ax.com>,
	Courageous <jkraska at san.rr.com> writes:
> 
>>>  - Pulchritudinous Python?
>>>  - Busty Python?
> 
> Clever Linguistic Integration Tricks without Stacks?

Wrong body part.

Tightly Integrated Tricks without Stacks.

> C//

-- 
Robert Kern
Ruddock House President
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From pobrien at orbtech.com  Wed Feb 13 09:06:02 2002
From: pobrien at orbtech.com (Patrick K. O'Brien)
Date: Wed, 13 Feb 2002 08:06:02 -0600
Subject: The joys of weakrefs... references to instance methods (callbacks)
In-Reply-To: <3C68D073.3080609@geocities.com>
Message-ID: <NBBBIOJPGKJEKIECEMCBMEDBMCAA.pobrien@orbtech.com>

Thanks. Is your new watcher class available? I couldn't tell if this one
(http://members.rogers.com/mcfletch/programming/watchable.py) was the newly
rewritten version or not.

The dispatcher module is part of a project that is very alpha, called
Bulldozer. You can find the files in CVS on SourceForge:

http://sourceforge.net/projects/bdoz/

The license is the same as the Python license. As Magnus mentioned, there
were several implementations developed as part of the anygui project. We all
learned from each other, but still came up with different approaches.

---
Patrick K. O'Brien
Orbtech

> -----Original Message-----
> From: Mike C. Fletcher [mailto:mcfletch at geocities.com]
> Sent: Tuesday, February 12, 2002 2:21 AM
> To: Patrick K. O'Brien
> Cc: Python List
> Subject: Re: The joys of weakrefs... references to instance methods
> (callbacks)
>
>
> Very nice module.  I just did a re-write of my watcher classes using
> dispatcher.  Why did I keep the watcher pattern?  Well, it makes
> creating pre-event processing possible (my one class needs to do some
> processing to determine conflicts, and until 2.2 object-based-classes
> (with properties) can live in ZODB, I'm stuck with that).
>
> Would be nice if there was a simple download somewhere (had to re-format
> the code after Mozilla mangled it with cut-and-paste).  Should put an
> explicit license in there (as ActiveState's legal policy says that the
> license included with the contribution applies.
>
> Also looked at the References.py module in anygui.  Didn't spend much
> time on it, as dispatcher is much closer to what I need.  Was one of
> these modules based on the other (your comments on the ASPN site seem to
> suggest there's some relationship)?
>
> Good job all around. Thanks,
> Mike




From dsavitsk at e-coli.net  Mon Feb 25 16:19:56 2002
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 25 Feb 2002 21:19:56 GMT
Subject: running a COM server in Python
References: <3C7A8D72.9089BC7A@mediaone.net>
Message-ID: <0Uxe8.9774$pC1.1703125@newssrv26.news.prodigy.com>

just for the sake of giving credit where it is due, and not giving it where
it is not, Mark Hammond is responsible for the excellent COM groundwork.
ActiveState is responsible for laying him off.  While he did work at
ActiveState, the COM work was largely done before then.

-d



"Mark Barclay" <markabarclay at mediaone.net> wrote in message
news:3C7A8D72.9089BC7A at mediaone.net...
> I have succeded in getting a COM server implemented in Python, so that
> it can be instantiated by a COM client., thanks to ActivePython's
> excellent groundwork.
>
> I'm wondering if it is also possible to create an application in Python
> that acts as a COM server but can be already running before teh COM
> client attempts to attach to it. This is how Excel behaves. If Excel is
> already running when a COM client attempts to attach to it. This is the
> bahaviour I want.
>
> Can anyone give me a pointer towasr how to get similar functionality?
>
>




From mwh at python.net  Wed Feb 13 08:21:54 2002
From: mwh at python.net (Michael Hudson)
Date: Wed, 13 Feb 2002 13:21:54 GMT
Subject: Reading and writing C structs from Python
References: <3C696FC7.1090600@erols.com>
Message-ID: <ulmdxh5uw.fsf@python.net>

"Edward C. Jones" <edcjones at erols.com> writes:

> Suppose I have a C struct for example
> 
> typedef struct {
> float x;
> int n[4];
> } test;
> 
> and a pointer to an instance of the struct
> 
> test* t = (test*) malloc(sizeof(test));
> 
> I would like to wrap this in a python object "py_t" so that (in
> Python) "x = py_t.x" reads "t.x" and "py_t.x = 13.0" writes to
> "t.x". Ditto for "py_t.n[i]". Ideally I would like to do this using a
> C function something like:

You know about the functions in Python/structmember?  Used by, e.g.,
complex objects:

static PyMemberDef complex_members[] = {
        {"real", T_DOUBLE, offsetof(PyComplexObject, cval.real), READONLY,
         "the real part of a complex number"},
        {"imag", T_DOUBLE, offsetof(PyComplexObject, cval.imag), READONLY,
         "the imaginary part of a complex number"},
        {0},
};

you can stick this in the tp_members slot of the type object or do
your work in your tp_getattro function.

OTOH, this doesn't really handle things like "int n[4]".

> PyObject* PyObject_FromCStruct(void* t, char* format)

What type would the returned object have?

> void* PyObject_AsCStruct(PyObject* pyo, char* format)

How would this know how to get the data out of the PyObject?

> The format strings used in module "struct" would be fine.

I think you need to supply more info than that (like field names!).

> My guess is that this is a job for the notoriously confusing "buffer
> interface". 

I doubt it.

> Does anyone have any sample code for problems like this?

Well, when confronted with this kind of problem, I've generally coded
it by hand.  Not much fun, but it doesn't really require many
braincells.

Cheers,
M.

-- 
42. You can measure a programmer's perspective by noting his
    attitude on the continuing vitality of FORTRAN.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From python at chajadan.net  Mon Feb 18 15:34:59 2002
From: python at chajadan.net (chajadan)
Date: Mon, 18 Feb 2002 12:34:59 -0800
Subject: help - inserting values into database tables
In-Reply-To: <3c710b01$0$79561$6536d21b@news.bitstream.net>
References: <mailman.1013981787.10266.python-list@python.org>
 <mailman.1014010317.7276.python-list@python.org>
Message-ID: <5.1.0.14.0.20020218123207.00a63680@63.118.78.184>

I always use PythonWin, part of the win32all distribution. It's a 
color-coded ide, with a basic syntax check, and has one nifty advantage 
over any other input editor I've tried: when you call a function, it takes 
the document strings and actually gives you a little note about what kind 
of input it's expecting!

So all in all it should be the same thing as it ever was, but now in the 
environment your particular task seems to ask for. The only "new" things 
will be the other possible windows-only features you might want to learn 
and use in your time on windows.

Happy computing!

--chajadan

At 08:15 AM 2/18/02 -0600, mylinuxboxroot wrote:
>Hey thanks,
>This is great, and I have the project now that it will be developed on a
>windows
>server so I guess I don't need to worry about going from linux to the
>MS-Access
>database.  I will develop it right on the server using the win32all
>distribution that
>you have suggested.  It will be somewhat easier now I think. I should be
>able to use
>the modules dbi and odbc windows stuff.
>     I have not tried python on windows yet though, do you get an output
>screen
>     terminal like using it on linux ???
>
>Well I will play with it too.
>thanks much,
>Joe F.
>
>
>chajadan <python at chajadan.net> wrote in message
>news:mailman.1014010317.7276.python-list at python.org...
> > My apologies, odbc is part of the win32all distribution, and is therefore
> > available only under Windows.
> >
> > However, the module you found, or other similar modules, are bound to be a
> > viable solution for you, database interaction being so widespread as it
>is.
> >
> > I know of another possible solution, by the creators of mxTime and
>whatnot,
> > available in this general location:
> > http://www.lemburg.com/files/python/eGenix-mx-Extensions.html
> >
> > --chajadan
> >
> > At 04:15 PM 2/17/02 -0800, Paul Rubin wrote:
> > >chajadan <python at chajadan.net> writes:
> > >
> > > > Yes, this is altogether possible.
> > > >
> > > > There is a module include in the main distribution called odbc, and
> > > > one called dbi.
> > >
> > >Whaaa?  That's the first I've heard of this.  In Python 2.2 under Linux,
> > >"import odbc" and "import dbi" both throw ImportError.
> > >
> > >There's an ODBC module in Vaults of Parnassus that looks pretty good, but
> > >I haven't
> > >tried it yet.  Is that what you meant?
> > >--
> > >http://mail.python.org/mailman/listinfo/python-list
> >
> >
>
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From peter.milliken at gtech.com  Sun Feb  3 16:52:41 2002
From: peter.milliken at gtech.com (Peter Milliken)
Date: Mon, 4 Feb 2002 08:52:41 +1100
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com><20020130073810.382$1b@news.newsreader.com><b34d3835.0201310005.548ee6be@posting.google.com><mailman.1012498757.26825.python-list@python.org><a3cf7s$se42@news1.gtech.com><mailman.1012522381.16207.python-list@python.org><a3d46d$tfn2@news1.gtech.com> <mailman.1012581067.14621.python-list@python.org>
Message-ID: <a3kbm0$tfn5@news1.gtech.com>


"Justin Sheehy" <justin at iago.org> wrote in message
news:mailman.1012581067.14621.python-list at python.org...
> "Peter Milliken" <peter.milliken at gtech.com> writes:
>
> > Actually, if you are *good* at what you do, then code review, whilst
still
> > beneficial, is not nearly *that* beneficial.
>
> I deeply disagree.

Sorry Justin, I knew as soon as I pressed "send" on this one it would spark
some controversy :-)

Please note the wording, I intended my statement in a very heavily qualified
manner (obviously this hasn't come through :-)) i.e. I do not deny that code
review will *always* be beneficial, even for the best of programmers
because, as you say, nobody is perfect. My statement is based on the fact
that it is sometimes almost impossible to get people to approach a reveiw in
the correct spirit i.e. you have the review and nothing comes from it
because nobody really put in the necessary effort - mainly because they were
behind in their own schedule and just had no time or incentive to look at
someone else's code :-).

I once (not that long ago) was the "chief" programmer for a very small
effort (4 other programmers). I wrote a SDP (Software Development Plan) that
mandated code reviews - I couldn't motivate the individuals working on the
project to do them properly, so when Phase 2 of the job swung around I
removed the mandatory code reviews from the SDP - practicality at work :-).
I was determined *not* to have plans and procedures that were not being
followed - sure I had plenty of pieces of paper that said the reviews took
place but because there was no practical benefit being gained I took the
logical approach and deleted them from the 2nd phase.

So I won't argue against anything you say below, I agree whole heartedly
with everything I see in a quick "skim" of it :-) My aplogies if my wording
didn't convey what I meant.

Peter

>
> Some of the most valuable code reviews that I've seen at my current
> employer have been of programs written by very competent programmers.
>
> First off, no one is perfect.  A few more pairs of eyes on the code
> that are explicitly looking for odd things can be very useful in
> finding subtle problems.
>
> Also, it is very possible that a requirement or interrelationship has
> been ignored or mishandled.  Even if the code is good in a relative
> vacuum, having someone with a different day-to-day focus look at it
> can help to expose important issues.
>
> There is another major benefit of code review that may not occur to
> you, since it is not a direct benefit to the code being reviewed.
> Code reviews are an excellent way for programmers of all skill levels
> to learn from each other.  This is one of the major reasons that I
> like them, even if they take a bunch of time out of my schedule.
>
> -Justin
>
>
>
>




From marklists at mceahern.com  Sun Feb 10 12:47:12 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 09:47:12 -0800
Subject: A newbie question about class
In-Reply-To: <a50631f6.0202100903.1455abc0@posting.google.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBGEPGKBAA.marklists@mceahern.com>

[newgene]
> Could you explain a little more why "is not" is better than "!="?

The short answer is that I don't think comparing something to None by
identity will ever raise an error.  Comparing it by magnitude might.

Python allows you to override magnitude comparison methods:  ==, !=, <, >,
<=, >=.  If you are using code written by somebody else, you don't know
whether they've accounted for comparison to None in their comparison
methods.  If you compare to None by identity, you bypass that whole issue
because comparison by identity ("is" and "is not") happens via id() and
that's not something you can override (as far as I know).

I recently ran into precisely this problem when I tried to feed FixedPoint
instances to PostgreSQL via PyPgSql.  PyPgSql was comparing to None by
magnitude.  FixedPoint's comparison method raised an error when other is
None.

So there you go.

Did that make sense?

For what it's worth, here's a simplification of FixedPoint that demonstrates
the problem.

#! /usr/bin/env python

class MyNumber:

    def __init__(self, value=0):
        try:
            self.value = float(value)
        except:
            raise TypeError("Can't convert to MyNumber: " + `value`)

    def __cmp__(self, other):
        x, y = normalize(self, other)
        return cmp(x, y)

def normalize(x, y):
    assert isinstance(x, MyNumber)
    if not isinstance(y, MyNumber):
        y = MyNumber(y)
    return x.value, y.value

m = MyNumber(2.0)

if m is not None:
    print "Hey, imagine that, we can compare to None by identity."
    print

# Now, try comparing m to None via the equality operator...

try:
    if m != None:
        print "Hey, I didn't think we could compare to None by magnitude!"
        print
except TypeError, e:
    print "Comparing to None by magnitude raises a TypeError:"
    print
    print e

## end

Cheers,

// mark




From hfoffani at yahoo.com  Fri Feb  8 18:00:43 2002
From: hfoffani at yahoo.com (Hernan M. Foffani)
Date: Sat, 9 Feb 2002 00:00:43 +0100
Subject: Defending the Python lanuage...
References: <3C62EE4A.5090305@destiny.com><20020207140755.364d5d77.logiplexsoftware@earthlink.net><200202081311.g18DBQGo020260@ratthing-b246.strakt.com><mailman.1013194035.3597.python-list@python.org><0BV88.108473$KM2.4226261@atlpnn01.usenetserver.com> <mailman.1013198294.30291.python-list@python.org>
Message-ID: <20020208180054.409$L6@news.newsreader.com>

> > The idea behind pair programming is exactly that such rotations
should
> > take place. You seem to think that two prograqmmers are joined at
the
> > hip when teams are formed, but in fact the pairings arise fairly
> > spontaneously in most cases.
>
> Well, I obviously didn't know much about pair-programming at the
start of
> this thread (and I have learned quite a bit so far), so I was under
a
> mistaken impression regarding rotation.  Seen in this new light I
would
> have to reconsider my earlier position (although I still retain the
right
> to be skeptical ;).

I've been doing pair-prog for several years. But we didn't called like
that back then. :-) I bet most of you have had similar experiences
before.
Haven't you been sat side by side with a coworker for hours helping
him
find that elusive bug?
Didn't you been talking and discussing and typing and timing and
arguing
with your coleage trying to optimize that stuff that kept your
customer
whinning for days?
And what about when you had to redo that API like mad because the
specs
where wrong and you asked for an extra couple of eyes because you
needed
to be sure that your modifications break the minimun code as possible?

Maybe because of the challenge I had in that times or because of the
close relation you grow within team members or was it just me, I don't
know, but the sweetest moments of programming I remember more are
those
when I was doing some-kind-of "pair programming".

It's futil arguing about pair-programming per se. We've been doing it
for years!
But I concede that what is hard to accept, and so to adopt, is what
Cliff (I think he was) mentioned before: the "90% of the time" thing.

IMHO, I guess that is all about XP. Take a practice that works fine
and apply it in an extreme manner. In other words: all the time.

Regards,
-Hernan







From sameer_ at email.com  Sun Feb 10 02:52:00 2002
From: sameer_ at email.com (sameer)
Date: 9 Feb 2002 23:52:00 -0800
Subject: wxpython and images
Message-ID: <6bd9f01b.0202092351.3d73391f@posting.google.com>

probably something silly i'm missing here, but why is that when i run
the wxpython demo (on win2k), wxImage shows jpg, gif, png, and bmp
images, but when i run that same file from another directory, only bmp
images work, and wxpython gives a  No image handler found type 17
error.


From nelson at crynwr.com  Wed Feb 20 09:47:57 2002
From: nelson at crynwr.com (Russell Nelson)
Date: 20 Feb 2002 09:47:57 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014136412.9849.python-list@python.org>
Message-ID: <m2y9ho5hwi.fsf@desk.crynwr.com>

Michael Chermside <mcherm at destiny.com> writes:

> Thank you Aahz for stating it so nicely. Most conversation in c.l.py 
> goes on in English, and if it's in anything else *I* am not likely to be 
> able to follow it, but I certainly don't begrudge others the use of 
> their own languages here!!!

print "Frankly, I don't know why comp.lang.python discussion is not in Python"

-- 
-russ nelson              http://russnelson.com | Crypto without a threat
Crynwr sells support for free software  | PGPok | model is like cookies
521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | 


From justin at iago.org  Fri Feb  1 11:29:10 2002
From: justin at iago.org (Justin Sheehy)
Date: Fri, 01 Feb 2002 11:29:10 -0500
Subject: Defending the Python lanuage...
In-Reply-To: <a3d46d$tfn2@news1.gtech.com> ("Peter Milliken"'s message of
 "Fri, 1 Feb 2002 15:02:18 +1100")
References: <b34d3835.0201300026.28e51376@posting.google.com>
	<20020130073810.382$1b@news.newsreader.com>
	<b34d3835.0201310005.548ee6be@posting.google.com>
	<mailman.1012498757.26825.python-list@python.org>
	<a3cf7s$se42@news1.gtech.com>
	<mailman.1012522381.16207.python-list@python.org>
	<a3d46d$tfn2@news1.gtech.com>
Message-ID: <x7zo2t4149.fsf@ra.iago.org>

"Peter Milliken" <peter.milliken at gtech.com> writes:

> Actually, if you are *good* at what you do, then code review, whilst still
> beneficial, is not nearly *that* beneficial.

I deeply disagree.

Some of the most valuable code reviews that I've seen at my current
employer have been of programs written by very competent programmers.

First off, no one is perfect.  A few more pairs of eyes on the code
that are explicitly looking for odd things can be very useful in
finding subtle problems.

Also, it is very possible that a requirement or interrelationship has
been ignored or mishandled.  Even if the code is good in a relative
vacuum, having someone with a different day-to-day focus look at it
can help to expose important issues.

There is another major benefit of code review that may not occur to
you, since it is not a direct benefit to the code being reviewed.
Code reviews are an excellent way for programmers of all skill levels
to learn from each other.  This is one of the major reasons that I
like them, even if they take a bunch of time out of my schedule.

-Justin

 




From jason at jorendorff.com  Mon Feb 25 20:11:18 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 19:11:18 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <a4op0l$cjp$1@nntp9.atl.mindspring.net>
Message-ID: <HFEKILOLEFEFMKAECNDLIEDGDFAA.jason@jorendorff.com>

Fredrik Lundh wrote:
> it's fairly trivial to construct a pickle string that calls
> eval or os.system with arbitrary arguments.

Andrew Dalke wrote:
> In http://groups.google.com/groups?selm=9slgqh%24ffm%241%
> 40slb2.atl.mindspring.net&output=gplain
> I show how to delete an arbitrary file using pickle (note:
> doesn't work with cPickle, but there are any other attacks I
> didn't try).

For what it's worth, *this* particular hole seems to have been
patched.  But pickle can still call class constructors and
__setstate__ methods and so forth, and it seems to me that
plenty of standard lib constructors do at least a little disk
access and socket stuff.  So it's still not safe.

(in Python 2.2)
>>> t = "(S'filename.txt'\012p1\012ios\012unlink\012p2\012(dp3\012b."
>>> pickle.loads(t)
pickle.UnpicklingError:
    <built-in function unlink> is not safe for unpickling
>>> cPickle.loads(t)
cPickle.UnpicklingError:
    <built-in function unlink> is not safe for unpickling

## Jason Orendorff    http://www.jorendorff.com/



From db3l at fitlinxx.com  Fri Feb 15 21:53:35 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 15 Feb 2002 21:53:35 -0500
Subject: win32 DeleteObject ??
References: <K3dTKjAtSTb8Ewis@jessikat.demon.co.uk> <uit8yuwhk.fsf@ctwd0143.fitlinxx.com> <a+fHoXAlPab8EwQo@jessikat.fsnet.co.uk>
Message-ID: <ug042uogg.fsf@ctwd0143.fitlinxx.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> yes I know they're in there, but to use them I need to include a 650Kb
> .pyd. At present I'm using anygui and it has quite a small foot print.

Not sure how it might play with anygui, but you could try Sam
Rushing's "dynwin" package which is just Python code around the
calldll module against the raw GDI objects, and is more granular so
you could probably get a much more fine grained control over overhead.

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


From andy47 at halfcooked.com  Fri Feb  1 05:19:45 2002
From: andy47 at halfcooked.com (Andy Todd)
Date: Fri, 01 Feb 2002 21:19:45 +1100
Subject: locale on cygwin
References: <Xns91A1A678DEEE3andy47halfcookedcom@203.109.250.24> <20020125140805.GC1252@dothill.com> <3C5A607F.6060107@halfcooked.com> <20020201123241.H10316@phd.pp.ru>
Message-ID: <3C5A6BC1.6080009@halfcooked.com>

Oleg Broytmann wrote:

> On Fri, Feb 01, 2002 at 08:31:43PM +1100, Andy Todd wrote:
> 
>> >>> locale.setlocale(locale.LC_ALL, '')
>>Traceback (most recent call last):
>>   File "<stdin>", line 1, in ?
>>   File "/usr/lib/python2.1/locale.py", line 374, in setlocale
>>     return _setlocale(category, locale)
>>locale.Error: locale setting not supported
>> >>>
>>angua:/home/andy47# echo $LANG
>>en_AU
>>angua:/home/andy47#
>>
> 
>    On lInux this means your system lacks files for en_AU locale. It is not
> Python fault. Use different locale, en_EN or any that exists on the system.
> 
> Oleg.
> 

Thanks for that, but this is what happens if I don't force it;

"""
$ echo $LANG
C
$ python
Python 2.1.1+ (#1, Jan  8 2002, 00:37:12)
[GCC 2.95.4 20011006 (Debian prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
 >>> import locale
 >>> locale.getdefaultlocale()
Traceback (most recent call last):
   File "<stdin>", line1, in ?
   File "/usr/lib/python2.1/locale.py", line 339, in getdefaultlocale
     return _parse_localename(localename)
   File "/usr/lib/python2.1/locale.py", line 272, in _parse_localename
     raise ValueError, 'unknown locale: %s' % localename
ValueError: unknown locale:
"""

Regards,
Andy
-- 
-----------------------------------------------------------------------
 From the desk of Andrew J Todd esq.
"Another year older, still no wiser." - Me, on my birthday




From rhymes at myself.com  Sun Feb  3 05:11:44 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sun, 03 Feb 2002 11:11:44 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org>
Message-ID: <e63q5ukd116tkglrbp92k160bjl2jihvu5@4ax.com>

On Sun, 3 Feb 2002 02:44:56 +0100, Gerhard H?ring
<gh_pythonlist at gmx.de> wrote:

>If you're using Windows, that explains it ;-) There is indeed no curses
>module included in the Python binary distribution for Windows. IIRC the
>reason is that there is no ncurses library readily available on win32 to
>build the curses support against.

You're right, i use Win2k and the binary distribution..

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From donn at u.washington.edu  Mon Feb 11 13:29:58 2002
From: donn at u.washington.edu (Donn Cave)
Date: 11 Feb 2002 18:29:58 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1192.802T1446T14104708threeseas@earthlink.net> <u3d0dcklp.fsf@synopsys.com> <8791.806T2435T6975373threeseas@earthlink.net> <uk7tkylre.fsf@synopsys.com>
Message-ID: <a492j6$18om$1@nntp6.u.washington.edu>

Quoth David Masterson <dmaster at synopsys.com>:
...
| Or you haven't given us anything we can deal with.  Note, if we wanted
| to remain ignorant, we could simply just stop talking to you.  You
| have an audience and your "10 minutes of fame", but, if you cannot
| express your problem in understandable fashion, we will go on to other
| things that we do understand.

Who's "we" - you have a mouse in your pocket?  Are you sure you speak
for the mouse, or is he wondering what's the point, too?  I count 8
pairs of Rue, Masterson, Rue, Masterson exchanges in the "References"
in your post.  Please take it to email.  Why not?

	Donn Cave, donn at u.washington.edu


From rony.steelandt at bucodi.com  Tue Feb 19 03:41:47 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 19 Feb 2002 00:41:47 -0800
Subject: Operating system requirements
References: <3C71A556.3E7BF27E@xatt.netx> <3C71CD62.23BF11AC@erols.com>
Message-ID: <b34d3835.0202190041.40b22cc4@posting.google.com>

Jerry Gitomer <jgitomer at erols.com> wrote in message 
> Windows 95 or newer.

Well this is a very restricted answer ;)

I would say go to www.python.org to see the different distributions.
You have them for LINUX,WINDOWS,MAC,.....

Rony


From brueckd at tbye.com  Tue Feb 19 10:39:24 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 19 Feb 2002 07:39:24 -0800 (PST)
Subject: Adding static typing to Python
In-Reply-To: <3C726575.5090400@destiny.com>
Message-ID: <Pine.LNX.4.33.0202190734240.5675-100000@ns0.tbye.com>

On Tue, 19 Feb 2002, Michael Chermside wrote:

> No discussion of adding static typing to Python would be complete
> without a reference to Guido's thoughts on the subject:
>
>     http://www.python.org/~guido/static-typing/index.htm

Thanks for the link. Reading it makes me wonder if some sort of partial
static typing would be beneficial, e.g. if the compiler was able to
determine up front that X is an atomic numerical type (int, long, or
float), Y is a string type (string or unicode), Z is a simple container
type (list, tuple, dict), and W is a user-defined object. Just thinking
out loud, and I don't know if it would help or not, but maybe instead of a
full static type system you could get many of the same benefits by
classifying objects into some broad categories, defaulting to the
classification of a generic Python object if unsure. Any thoughts?

-Dave




From quinn at hork.ugcs.caltech.edu  Sat Feb  9 22:05:18 2002
From: quinn at hork.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Feb 2002 03:05:18 GMT
Subject: Perhaps I am just dumb
References: <8852d33c.0202090222.34be7da2@posting.google.com> <a42c8e.3vvbc1l.1@kserver.org> <8852d33c.0202091336.1822b76a@posting.google.com>
Message-ID: <slrna6bord.9tv.quinn@hork.ugcs.caltech.edu>

On 9 Feb 2002 13:37:00 -0800, wooks <wookiz at hotmail.com> wrote:
>Never got far enough into the book to look at win32. I couldn't
>understand why there were these 2 classes that were variants of a
>dictionary class and neither subclassed userdict and then the book
>says now I'm using magic methods go look up in the language reference
>(I did  - couldn't understand the raison d'etre). I think I have now
>sussed answers to both issues - the userdict class (I am guessing) is
>a recent addition and the book was at Python 1.5.2. The answer to the
>second issue (I think) is that this is how (or one of the ways)
>polymorphism is implemented. If I am indeed right on that then a

No, you're being too CS-PhD-y.  "poly-what?"  Yes, python is polymorphic, but
there's no need to get all fancy-pants about it.  I suspect that neither the 2
classes subclassed UserDict because their author didn't feel like it.  UserDict
defines a few methods for you, and can be convenient, but you don't hafta use
it.  I make dict-likes all the time, but rarely use UserDict, because it
insists I have an appropriate 'data' attribute.

The magic methods are methods that are magically called by non-method-looking
syntax.  So len(o) actually calls o.__len__() (for historical reasons, IMO),
and x[y] actually calls x.__getitem__(y) (since 'x[y]' is easier to type), and
x[y] = z calls x.__setitem__(y, z) (since it's more natural looking to people
used to infix algol-esques).

They don't have anything to do with polymorphism per se, but of course they
make it easier, as any time you define a general-purpose widely-used method
for your object (i.e. x[y] aka __getitem__ is widely-used so if you define it
your object can work with any __getitem__-using code).


If you want to hear the incomprehensible CS PhDs, go to c.l.functional and
listen to them talk about the latest extension to the haskell type system
(I still haven't been able to figure out exactly what existential
quantification or rank-2 polymorphism is, let alone 'arrows'... I get off when
the talk gets all ring/field-y).


From cgale1 at cox.net  Sat Feb  9 02:28:08 2002
From: cgale1 at cox.net (Van Gale)
Date: Sat, 09 Feb 2002 07:28:08 GMT
Subject: Standalone ZODB + BTree question
Message-ID: <cc498.2133$0c.92062@news2.east.cox.net>

Can anyone point me in the right direction for using OOBtree with ZODB?

I've been able to figure out FileStorage and DbmStorage easy enough from the
docs, and I can create and use an OOBTree... I just can't seem to tie it
into persistent storage.

The BTree unittest *seem* to just be running tests on in memory btrees, so I
haven't been able to use them as an example.

--
Van Gale





From p.vrijlandt at aig.azn.nl  Wed Feb 20 09:04:29 2002
From: p.vrijlandt at aig.azn.nl (Patrick Vrijlandt)
Date: Wed, 20 Feb 2002 15:04:29 +0100
Subject: Problems with telnetlib
References: <xyuvgcxlh4s.fsf@sex.ifi.uio.no> <a4r5ei$pkm$1@odysseus.uci.kun.nl> <xyuvgcujsyl.fsf@sex.ifi.uio.no> <a4shr2$49s$0@216.39.172.122> <a4t9fr$500$1@odysseus.uci.kun.nl> <jwbaxter-020177.08052619022002@corp.supernews.com>
Message-ID: <a50adf$laq$1@odysseus.uci.kun.nl>

> Well, telnetlib works for the limited purposes we use it for here.  Or
> does it?  I found the cause (in my code) of fairly frequent false alarms
> in a monitor I wrote using it, but we still have infrequent false alarms
> which I haven't yet understood.
>
>   --John

As I said, protocol implementation is incomplete in telnetlib.
If that is the problem, logging of protocolmessages could help you.
(These are the messages starting with IAC).

--
Patrick Vrijlandt




From bokr at oz.net  Tue Feb 19 22:20:45 2002
From: bokr at oz.net (Bengt Richter)
Date: 20 Feb 2002 03:20:45 GMT
Subject: execfile question [+Q: why doesn't del dict imply dict.clear() ??]
References: <mailman.1014117330.24204.python-list@python.org> <3C724B05.E9167812@attbi.com>
Message-ID: <a4v4md$ecm$0@216.39.172.122>

On Tue, 19 Feb 2002 12:51:10 GMT, Mark Barclay <markabarclay at attbi.com> wrote:
>Kerim Borchaev wrote:
>
>> Can someone explain why while running this script I don't see C.__del__
>> eventually called?
>>
>> For me the output of it's execution is:
>>
>> >main.py
>> it's me
>> init
>> executed
>> >
>>
>> ##main.py######################
>> execfile('executable.py', {})
>> print 'executed'
>> ########################
>>
>> ##executable.py######################
>> print "it's me"
>> class C:
>>     def __init__(self):
>>         print 'init'
>>     def __del__(self):
>>         print 'deleted'
>>
>> c = C()
>> ########################
>>
>> Thanks in advance.
>
[changing Mark's top-post to bottom]

>"del c" will cause your __del__() to be called.
>So will any other action that causes your instance to be garbage-collected.
>
[...example snipped...]

I think he's asking why finishing execfile is not one of those
'other actions'.

I think the contents of the anonymous dictionary he passed are not
automatically deleted (why not?) when it goes away, so a reference
in it is keeping the c instance from being collected.

Try not passing a dictionary, or pass one with a name and empty it
explicitly when you get control back. E.g.,
 >> ##main.py######################
    d={}
 >> execfile('executable.py', d)
    d.clear()
 >> print 'executed'
 >> ########################


BTW, Just deleting the dictionary (say del d) does not seem
to imply d.clear(), and hence just del d won't trigger the c destructor.

Is that correct python behavior??

Regards,
Bengt Richter



From tanzer at swing.co.at  Mon Feb 25 02:46:26 2002
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 25 Feb 2002 08:46:26 +0100
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
In-Reply-To: Your message of "Sun, 24 Feb 2002 23:46:47 EST."
             <3C79C1B7.AB9287D0@engcorp.com>
Message-ID: <E16fFq6-0002G4-00@tswings.swing.cluster>

Peter Hansen <peter at engcorp.com> wrote:

( lots of `gefingerpoken` schnapped)

> Apparently German, as Perl, has many ways of saying the same thing.
> And it's equally readable.  <grin>

To a german speaker, this actually looks like dutch. Must be Pythonic,
then. <wink>

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92





From donn at u.washington.edu  Wed Feb  6 19:49:19 2002
From: donn at u.washington.edu (Donn Cave)
Date: 7 Feb 2002 00:49:19 GMT
Subject: multi-threaded *nix program needs to spawn an external program/script
References: <mailman.1013035212.9230.python-list@python.org>
Message-ID: <a3siuf$26cm$1@nntp6.u.washington.edu>

Quoth "Jason Orendorff" <jason at jorendorff.com>:
| arbalt at yahoo.com wrote:
| > Would it make sense to add a function to python for *nix systems
| >     that would do the "start an executable file as a new process"?
|
| Try os.spawnv(); it should work for both Unix and Windows.
|   http://www.python.org/doc/current/lib/os-process.html

Yes, though according to him, that shouldn't work - he says, on Solaris,
in a threaded program, you mustn't "do anything" between fork and exec,
and fork and exec called separately from Python don't work for him for
that reason.  os.spawnv is implemented in Python.

I think it would be worth a try.  I think os.spawnv will work as well
as os.system.

The next step would be to recode os.spawnve in a C module.  I don't know
anything about this Solaris problem, but if he's right, that would help.
(But then what's wrong with os.system?)

Of course if there's a system or thread library function that addresses
this need, that would obviously be the way to go.

	Donn Cave, donn at u.washington.edu


From jeff at ccvcorp.com  Tue Feb 26 19:32:39 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 26 Feb 2002 16:32:39 -0800
Subject: subclassing float (or not)
References: <mailman.1014757996.10871.python-list@python.org>
Message-ID: <3C7C2927.A6DAF83F@ccvcorp.com>


Steven Majewski wrote:

>  If you
> subclass float instead, you don't have to wait around to see
> what else breaks because you forgot to implement a method to
> support it!

No, but you *do* have to watch carefully for things that "work", but
provide the wrong answer.  In the example of PhysicalValue, I presume
that subclassing float would get it to "automatically" work with, say,
math.sqrt()... except that it'd only work for your scalar, not your
units.  The square root of 100.0(cm**2) is 10.0cm, but the automatic
coercion would give you either 10.0 (a straight float), or 10.0(cm**2).
This particular problem is easy to spot, but who knows that *other*
things you'll be assuming will "automatically" work, but which produce
subtly wrong answers?

(Not that I'm saying that it definately *shouldn't* subclass float --
I'm just saying, don't assume that "doesn't break" == "correct".)

Jeff Shannon
Technician/Programmer
Credit International




From grayshade at libero.it  Sat Feb 23 19:59:13 2002
From: grayshade at libero.it (Domenico)
Date: 23 Feb 2002 16:59:13 -0800
Subject: problems with pythoncom and VB-generated OLE/ActiveX DLL
Message-ID: <e34e8ce6.0202231659.609787bd@posting.google.com>

Here are the facts: I need to link a Python app with a small VB
frontend (I hadn't got enough time to deal with Tkinter or WxPython,
and the anygui version of that frontend I did was way too slow to be
of any use). I compiled the frontend as an ActiveX dll providing just
one class. Its constructor (class_initialize) shows a form and two
methods from this class are used to interact with it. I registered the
DLL, used it in another VB project and it worked as it was supposed to
do. Then I built a Python wrapper to the class via the script that
comes with win32all. I imported the script in an interactive session.

>>> import display
>>> d = display.Class1()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "C:\compilers\python\display.py", line 40, in __init__
    if oobj is None: oobj = pythoncom.new(self.CLSID)
pywintypes.com_error: (-2147221164, 'Interfaccia non registrata.',
None, None)

'Interfaccia non registrata' (in Italian) means something like
'Unregistered interface'. So I did:

>>> import pythoncom
>>> d = pythoncom.new("{***CLSID of the DLL***}")

getting the same error.
I'm not very much into the whole VB-ActiveX-COM stuff (I hardly recall
having programmed in VB for a couple years before I realized what a
messy language it was) so I actually don't know at all what to check.
Been lurking for a lot of time, I hoped I could come out with
something more clever as my first post -- well, next time. TIA.

import random
sigs = [ "why the hell do I think I can write intelligent stuff",
         "sorry I'm stupid",
         "well I guess I should have solved this by myself" ]
print "-".join(random.choice(sigs).split()) + "-ly y'rs"

Domenico

"If you don't know where you're going, turn back and look at where
you're coming from." (M. Ovadia)


From phr-n2002a at nightsong.com  Mon Feb  4 06:18:59 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 04 Feb 2002 03:18:59 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org>
Message-ID: <7xsn8hjxzw.fsf@ruckus.brouhaha.com>

"Tim Peters" <tim.one at home.com> writes:
> > That's the problem--if I don't import future division, my code will
> > break in the future, because the // operator isn't supported in the
> > present.
> 
> Not so:  introducing "//" was 100% backward-compatible, so isn't controlled
> by __future__.  "//" works out of the box in 2.2, no __future__ required.

If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
it isn't backwards compatible.

> > (The most recently released version of Python is a 2.1 series release,
> > so 2.1 is still current).
> 
> I don't know where you're looking, but 2.2 final was released.  

I know this, but 2.1.1 was released more recently than 2.2, so the most
recent release was in the 2.1 series.

> Bugfix releases for past versions may (or may not) get released from
> time to time, but if you decide to build a 1.5.3 someday, that won't
> make 1.5 "current", it will just make 1.5.3 the most recent release
> of the 1.5 line (as 2.1.2 is the most recent release of the 2.1
> line, and contains none of the features new in the 2.2 line --
> active development of 2.1 ended the day 2.1 final was released, so
> calling it "current" is rhetorical gimmickry).

If the official Python maintainers think they might release a 1.5.3
someday, then it's fair to say that the 1.5 series is still current.
If they're unlikely to ever release another 1.2.x or 0.9.x, then
those versions are not current, even if I release a 1.2.x myself.

More importantly, in the real world, a lot of people are still using
2.1 and 1.5, and few are using 2.2.  So coding anything in a way that
requires 2.2 inconveniences a lot of users.  That might be worth it
for serious features like metaclasses or generators, but is silly to
do over this division nonsense (think of those obnoxious web sites
that tell you to upgrade your browser before you can view some page).

I'd non-rhetorically say that for most users, 1.5.2 or 2.1 is current
and 2.2 is bleeding edge.  (I'm still using 2.1 and plan to get around
to upgrading to 2.1.1).

> > I don't see any reasonable way I can write code right now that uses
> > division that works in both the current 2.1 (which doesn't support
> > importing future division) and in forthcoming versions (where the
> > semantics of the / operator will be incompatibly different).
> 
> The semantics of "/" didn't change in 2.2, and won't in 2.3, 2.4, 2.5, ...
> either (read PEP 238). 

Yes, but we're talking about backwards compatibility (see the thread
header).  "Backwards" from 2.2 means versions like 2.1 and earlier.

Given that 2.2 can import 3.0 division from the "future", will 3.0
have a way to import 2.x division from the "past"?  If I can put

   try: from __backward_compatibility__ import use_old_division
   except ImportError: pass

into all my scripts, and have it maintain the [012].x "/" semantics in
3.0 and later while doing nothing in 2.x and earlier, then that at
least gives me a way to write stable scripts, so I hope 3.0 can
include that.

> If you need a given static instance of "/" to do
> radically different things depending on whether you feed it ints or floats
> dynamically, you're the first person to claim such a need.

I want my programs to run the same way across the different Python 
versions that people are using now and will be using in the near future.
Am I really the first person to want that?

> Else you have a specific flavor of division in mind when you type
> "/", and it's easy to write it in such a way that it will deliver
> the same results in 3.0 as in 0.9.8.

By using divmod(x,y)[0] instead of x/y everywhere I want to divide
ints, per the PEP?  Ugh!


From gcordova at hebmex.com  Tue Feb 19 14:13:40 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Tue, 19 Feb 2002 13:13:40 -0600
Subject: Finding nonprintable characters?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060AB5@SRM201002>

> 
> Hello,
> 

Gweepnings.

>
> I have a function
> 
> isBinary(filehandle)
> 

Shades of Perl's "if(-b $filename)" !!!!

>
> that I'm not sure how to implement.
>

damn. :-(

> I've decided to define binary as 
> containing characters above \x80.  But  what is the best way 
> to do this?
> 
> 1. iterate through xreadline, so the whole thing doesn't get 
> loaded into 
> memory?
>

def isBinary(filehandle):
  # Save current position.
  lastPos = filehandle.tell()
  # Search for binary chars.
  line = filehandle.readline()
  while line:
    if .... (find a char > \x7F)  how??

> 
> 2. String searching?  If so, for what string?  Searching for anything 
> greater  than \x7f?
> 

But where??

In the first line?

Char by char?

In the first n Kb of text?

>
> 3. Re searching?  for what class?
> 

class like [\x7F-\xFF] I'd think.

> 
> Thanks in advance,
> 
> Van
> 

My suggestion:
	1. read a block of the file, say, the first 2Kb.
	2. Scan with a regex like r'[\x80-\xFF]'.
	3. If no chars found, the it's text, else it's binary.


import sre
def isBinary(filehandle, blockSize=2048):
  start = filehandle.read(blockSize)
  filehandle.seek(0)
  
  # Check for "binary" chars.
  if sre.search(r'[\x80-\xFF]',start,sre.S):
    # Sure enough, it's one of them dastardly BINARY files!
    return 1
  
  # Wait! Is there at least ONE \n in the text?
  if not sre.search(r'\n', start, sre.S):
    # Shuks, seem'd decent enough.
    return None
  
  # OK, you're good, I guess.
  return 1


More or less. I added the '\n' requirement for "textyness",
because 2Kb withough a single new-line doesn't seem quite
texty to me. Of course, you might think diferent.

Good luck :-)

-gustavo



-gus



From chappa at math.washington.edu  Tue Feb  5 02:42:54 2002
From: chappa at math.washington.edu (Eduardo Chappa)
Date: Mon, 4 Feb 2002 23:42:54 -0800
Subject: OT again... is there an IMAP expert in the house?
In-Reply-To: <mailman.1012892049.28494.python-list@python.org>
References: <01f201c1a7b7$34ebb1c0$0101010a@local> <336AF14CE0@kserver.org>
 <022601c1a7c1$f5367c80$0101010a@local> <2FF2F00E52@kserver.org>
 <mailman.1012224450.11377.python-list@python.org>
 <Pine.OSF.4.44.0202041742380.163260-100000@goedel2.math.washington.edu>
 <mailman.1012892049.28494.python-list@python.org>
Message-ID: <Pine.OSF.4.44.0202042332410.184841-100000@goedel2.math.washington.edu>

*** Chris Gonnerman (chris.gonnerman at newcenturycomputers.net) wrote in...:

:) > This is not a problem about the protocol you are using to read your
:) > folder. IMAP is the way that you communicate to the server. The way
:) > that the mail is stored in the server is the relevant point here. For
:) > example, the default for the UW server is unix, and unix (format)
:) > does not allow concurrent access to a mailbox. If you use the UW
:) > server, and you use mbx format, then you can have concurrent access
:) > to your folders. Convert your folders to mbx format and you'll be
:) > done.
:)
:) Tried that, didn't work.  I was using Netscape 4.75 as one client,
:) Outlook Express as the other, and concurrent deletion (even with
:) expunges) just didn't work.  Outlook Express never seemed to notice
:) Netscape doing deletions, and I couldn't get Outlook Express to
:) properly expunge without closing it.

Just to be sure you did that. What are the first characters of the first
line of the folder that you want to share?

-- 
Eduardo
http://www.math.washington.edu/~chappa/pine/





From db3l at fitlinxx.com  Mon Feb 25 18:14:41 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 18:14:41 -0500
Subject: HELP! Obscure Python Reentrancy Question
References: <rf3e7uk7cei0g516n0ptlstcl64vuu4m0o@4ax.com> <ur8ncpxbs.fsf@ctwd0143.fitlinxx.com> <hdde7ukvfch6jpfovghvm92abaiq26q2i2@4ax.com>
Message-ID: <uheo5w3vi.fsf@ctwd0143.fitlinxx.com>

Courageous <jkraska at san.rr.com> writes:

> I haven't considered how and when to grab the GIL, and what it
> means from the perspective of two different cooperative threads,
> each of which may (or may not!) be executing a Python code
> fragment, at some partial state of completion. Let me help:

>From later in this thread it looks like you resolved your specific
issue, but just to clarify - in the below:

> def f(task):
> 	do()
> 	some()
> 	task.Step(1)
> 	other()
> 	stuff()
> 	task.Succeed()

it seems reasonable that what's happening is that your scheduler is
controlling the Python execution through your synchronization points,
and the way in which the scheduler "dispatches" one of your bits of
Python code is to return from the Step() function call.

So the only thing that I think is critical is that you ensure that
when the scheduler returns back to the Python function that it holds
the GIL (e.g., grab the GIL before letting the Python call to
task.Step(1) return.  That's assuming that you've ever released the
GIL in your extension - if you haven't, then you're fine too as long
as you ensure only one thread/fiber in the extension lets control go
back to the Python interpreter at one time.

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


From sholden at holdenweb.com  Mon Feb 11 07:11:01 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Feb 2002 07:11:01 -0500
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013422700.5800.python-list@python.org>
Message-ID: <fxO98.148800$KM2.6667567@atlpnn01.usenetserver.com>

"Michal Wallace" <sabren at manifestation.com> wrote ...
> On 8 Feb 2002, Philip Swartzleonard wrote:
>
[ ... ]
> >
> > Hm, i don't seem to remeber seeing this Pythonistas thing till maybe two
> > days ago. Anyway, where did it come from-- based on some word? And HTF
do
> > you pronouce it? :)
>
>
> It's been around a while... I say "python - ee - sta"...
>
> random factoid: I used to own pythonista.com, but gave it
> to Steve holden...  not sure what he's got planned for
> it... :)
>
Me neither, but I'm hoping it will involve huge electronic money transfers
into my accounts, and the eventual admission by Microsoft of Python's total
world domination.

more-realistically-it'll-be-jaws-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py



--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From theller at python.net  Mon Feb 18 16:38:46 2002
From: theller at python.net (Thomas Heller)
Date: Mon, 18 Feb 2002 22:38:46 +0100
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com>
Message-ID: <a4rsa2$2grrk$1@ID-59885.news.dfncis.de>

"David Bolen" <db3l at fitlinxx.com> wrote in message news:u4rkezl0l.fsf at ctwd0143.fitlinxx.com...
> The binary portion of the windows specific installer is hardcoded into
> the distutils package (encoded in the bdist_wininst.py module), and it
> doesn't appear that source is available (at least I can't seem to find
> it even in the CVS tree) so changing how it behaves doesn't seem easy.
> You could replace it outright, but then you'd have to duplicate its
> existing functionality as well.

The source code IS available:
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/distutils/
in the misc directory. Guess I'll have to add a note about this
to the source... There is a tricky symbolic link somewhere on SF
which links parts of the distutils tree into the Python source tree.

> Having some additional hooks created in distutils to execute
> post-installation sounds like it could be a useful enhancement.  At
> least with that sort of scenario you could write your own shortcut
> creating code without modifying the core distutils itself.

I'm currently working on an bdist_wininst enhancement which
takes an additional command line option (--install-script <script.py>).
This script would then be run (with sys.argv set to [script_name, '--installed'])
after all the files have been copied, and it would again be run
(with sys.argv set to [script_name, '--removed']) just before
the deinstaller is about to remove all the files and directories.

Thomas




From sav at ulmen.mv.ru  Sun Feb 24 06:09:04 2002
From: sav at ulmen.mv.ru (Alexander Semenov)
Date: Sun, 24 Feb 2002 14:09:04 +0300
Subject: current default string encoding
References: <3c749db4.11089435@news.cis.dfn.de> <mailman.1014284016.4796.python-list@python.org>
Message-ID: <3c78c8e8@gwspb.ulj.menatepspb.com>

"Oleg Broytmann" <phd at phd.pp.ru> ???????/???????? ? ???????? ?????????:
news:mailman.1014284016.4796.python-list at python.org...
> On Thu, Feb 21, 2002 at 07:15:03AM +0000, Luc Saffre wrote:
> > How or where is this current default string encoding defined? Can i
>    http://www.python.org/doc/FAQ.html#4.102
> > change it during run-time?
>    No.

You can do

reload(sys)
sys.setdefaultencoding('Whatever your want')

but Guido dont want you do so.

Alexander.





From debl2nonospammywham at bellatlantic.net  Mon Feb 25 19:44:38 2002
From: debl2nonospammywham at bellatlantic.net (David Lees)
Date: Tue, 26 Feb 2002 00:44:38 GMT
Subject: How do I get stdout directly into a Python variable
Message-ID: <3C7ADA7E.1FFB5A4F@bellatlantic.net>

I am playing with a Python script that issues system commands on linux
and I want the stdout to go directly into a Python variable.  Right now
I am writing to a temp file and then opening it up.  This seems like a
hack.

I am doing something like this:

os.system('diff file1 file2 | wc > temp')
fh = open('temp','r')
line = fh.read()
fh.close()

I see examples of redirecting stdout in texts, but this still creates a
file that needs to be opened.

Thanks in advance.

David Lees


From mkelly2002NOSPAM at earthlink.net  Mon Feb 18 13:27:49 2002
From: mkelly2002NOSPAM at earthlink.net (Michael Kelly)
Date: Mon, 18 Feb 2002 18:27:49 GMT
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <b34d3835.0202160857.55bf6193@posting.google.com> <rfrv6ucpn8osftj7e2mbj34e1o46271v3q@4ax.com> <b34d3835.0202180007.78d987fd@posting.google.com> <b34d3835.0202180604.7a7605ab@posting.google.com>
Message-ID: <nlh27ukfu8u2ga4g2n876b211idb024rkt@4ax.com>

On 18 Feb 2002 06:04:03 -0800, rony.steelandt at bucodi.com (Rony) wrote:

>Well i did try it with the same error !
>
>Anyone has an idea ??

Out of curiosity, did they give you a type library
with the control?  Also you might try running
it through Makepy with -o option to get a module?
Perhaps if you posted the CoClass part of the
ActiveX module produced by Makepy somebody 
could see if something is weird.


--

"I don't want to belong to any club that would have me as a member."
    -- Groucho Marx


From eppstein at ics.uci.edu  Thu Feb 28 16:13:17 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 13:13:17 -0800
Subject: Status of PEP's?
References: <mailman.1014922495.31163.python-list@python.org>
Message-ID: <eppstein-774A76.13131728022002@news.service.uci.edu>

In article <mailman.1014922495.31163.python-list at python.org>,
 "Bjorn Pettersen" <BPettersen at NAREX.com> wrote:

> And exactly this notation has been rejected in PEP 204 -- basically
> because it's butt-ugly in a Perl'ish sort of way. Besides, this goes
> under the section of solving the "interval" problem which PEP 276
> *explicitly* says it is not doing (and which I personally don't find
> very compelling, especially if PEP 276 is adopted).

You can't have it both ways -- you seem to be saying that it's not ok to 
argue against PEP 276 on the basis that you'd like a better general range 
syntax, because that's not the problem the PEP is trying to solve.  But 
then in the parentheses you seem to be saying that it's ok to argue against 
proposals for a better range syntax on the basis that PEP 276 covers most 
of the need.  Which is it?
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From mcherm at destiny.com  Tue Feb 26 16:04:15 2002
From: mcherm at destiny.com (mcherm)
Date: Tue, 26 Feb 2002 16:04:15 -0500
Subject: Oddity question
Message-ID: <3C7BF84F.1090306@destiny.com>

> I was fooling around with sockets last night, using a thread to deal =
> with the read part and another method to do the print.
> 
> like this:
> 
> def thread_read (self):
>   while 1:
>     self.buf=3Dself.buf+self.my_socket.recv (8)
> 
> def print_buf(self):
>   if buf:
>     print "bufeer-->",buf
>     buf=3D'""
> 
> But, what happened to me was that the -buf =3D ""- line seemed to be not =
> working     [...]

> So, I dont know why I changed thread_read function to:
> 
>   while 1:
>     aux=3Dself.my_socket.recv (8)
>     self.buf=3Dself.buf+aux
> 
> And this version worked properly..... but I still do not know why.

Well, I can't tell exactly what you were doing here (try sending plain 
text instead of html... it doesn't get mangled as much), but I do see 
one thing. In the example that doesn't work you have a line 'buf=""' at 
the end of a function or method, while the example that DOES work has 
'self.buf=self.buf+aux' inside a loop.

If buf is a local variable, then 'buf=""' at the end of a function will 
only modify the local variable and do nothing to the actual buffer 
(strings are immutable in Python). Modifying self.buf in a loop will not 
be lost in this fashion.

Does that help any?

-- Michael Chermside







From brian at sweetapp.com  Tue Feb 12 13:14:16 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Tue, 12 Feb 2002 10:14:16 -0800
Subject: ANN: SilverCity 0.5.0
In-Reply-To: <000b01c1b389$8509fbc0$0100000a@ahimsa>
Message-ID: <001c01c1b3f1$1536d5d0$445d4540@Dell2>

SilverCity is a lexing package, based on Scintilla, which can provide
lexical analysis for over 20 programming and markup languages. 

There are two ways to access its functionality: through a C++ library
and through a Python extension package.

The Python extension package also contains modules that can generate
syntax-styled HTML documents for C++, HTML, Python, XML and XSLT.

The code was been tested reasonable well but I haven't invested much
effort in writing documentation at this point. For a simple interactive
session see:
http://silvercity.sourceforge.net/examples/

You can download the Python package from:
https://sourceforge.net/project/showfiles.php?group_id=45693

I have installers for Windows and have tested the source distribution on
RH 7.2.  Actually, I would suggest downloading the source distribution
in any case because it contains a useful example script called
"source2html.py".

Cheers,
Brian




From gcordova at hebmex.com  Fri Feb 15 15:04:38 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 15 Feb 2002 14:04:38 -0600
Subject: Pythonistas? <- Re: OT: Perl programmers?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A8E@SRM201002>

> ... also, gender inversion is not uncommon
> in the south of Spain...
> 

Son mas malos que Perl!!!

;-D

-gustavo

pd: TGIF!!



From jeff at nmt.edu  Tue Feb 26 14:08:42 2002
From: jeff at nmt.edu (Jeff Shipman - SysProg)
Date: Tue, 26 Feb 2002 12:08:42 -0700
Subject: python stripping LD_LIBRARY_PATH?
Message-ID: <3C7BDD3A.9050905@nmt.edu>

I never got a response to this and I really don't
understand why it's happening. I would appreciate
any help anyone can give me. If there's another place
I should be posting this, please let me know.

-------- Original Message --------

I've got a wrapper so I can run a SUID python script which
looks like the following:

#include <unistd.h>

int main(int argc, char **argv)
{
    char *env[] = { "TERM=linux",
                    "LD_LIBRARY_PATH=" ORACLE_HOME "/lib",
                    "SHELL=/bin/sh",
                    "PATH=/usr/local/bin:/bin:" ORACLE_HOME "/bin",
                    "ORACLE_HOME=" ORACLE_HOME,
                    "ORACLE_SID=" ORACLE_SID,
                    '\0' };
//  execve("/u/buch/bin/late_books.py", argv, env);
    execve("./late_books.py", argv, env);
    return 0;
}

I have the line commented out so I can test for the
moment.

The problem I'm having is that the LD_LIBRARY_PATH
is not showing up in os.environ. Everything else shows
up just fine. It prints out as:

{'ORACLE_SID': 'tcc', 'PATH':
'/usr/local/bin:/bin:/fs/oracle/linux/app/oracle/product/8.0.5/bin',
'ORACLE_HOME': '/fs/oracle/linux/app/oracle/product/8.0.5', 'SHELL':
'/bin/sh', 'TERM': 'linux'}

This is a problem because I really must set LD_LIBRARY_PATH
before starting this script. I've tried setting the LD_LIBRARY_PATH
inside the script before importing the Oracle module, and that
doesn't work (which is odd).

Why is python stripping out this environment variable?

I'm using:

ActivePython 2.1, build 211 (ActiveState)
based on Python 2.1 (#1, Jun 15 2001, 02:52:41)
[GCC 2.95.2 20000220 (Debian GNU/Linux)] on linux2

Any help would be extremely appreciated.

-- 
Jeff Shipman           E-Mail: jeff at nmt.edu
Systems Programmer     Phone: (505) 835-5748
NMIMT Computer Center  http://www.nmt.edu/~jeff



From jason at jorendorff.com  Sun Feb 24 23:56:39 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Sun, 24 Feb 2002 22:56:39 -0600
Subject: thread question
In-Reply-To: <a5ab54+p15f@eGroups.com>
Message-ID: <HFEKILOLEFEFMKAECNDLOENNDEAA.jason@jorendorff.com>

max wrote:
> can i acces attributes, or functions of a thread(like if i started 
> the class with threading.Thread(None, someclass)) like i can access 
> classes? and if yes how? if no, how can i do it in another way?

You probably want something like this.

class SomeClass(threading.Thread):
    def run(self):
        ...
        ... Here you can access attributes and methods 
        ... of self, which is a Thread object.
        ...

myThread = SomeClass()
myThread.start()

## Jason Orendorff    http://www.jorendorff.com/



From m.faassen at vet.uu.nl  Wed Feb  6 17:47:03 2002
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 6 Feb 2002 22:47:03 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <m3n0yqt5ib.fsf@chvatal.cbbrowne.com> <1462.801T1396T14014714threeseas@earthlink.net>
Message-ID: <a3sbp7$ih2$1@newshost.accu.uu.nl>

In comp.lang.python Timothy Rue <threeseas at earthlink.net> wrote:
[snip]
> Tell me how prestegious that email address is of your as you will be
> forces to pull your head out of your ass and admit that auto-coding is
> possible and should have been done a decade or so ago.

In fact, auto-coding has been done, in Python Secret Underground
headquarters, and in fact more than a decade ago. It was then
discovered it would seriously decrease the PSU's income. After all, many
of its 'donators' work as semi-competent programmers and would surely
be out of a job as soon as auto-coding was made available to the general
public.

The PSU then proceeded on a ruthless suppression campaign. All people who
knew about the implementation of auto-coding were either removed entirely,
or brainwashed into becoming Perl programmers.

So all ended well -- the PSU gets its income and the pretentious semi-competent
programmers retain their jobs until perpetuity. Occasionally of course a
person of exceptional insight breaks into the secret, but the PSU has
made sure they're booed away before they can do any harm.



From justin at iago.org  Wed Feb 13 11:53:52 2002
From: justin at iago.org (Justin Sheehy)
Date: Wed, 13 Feb 2002 11:53:52 -0500
Subject: python to lisp translation
In-Reply-To: <3C6A7825.714DF612@setf.de> (james anderson's message of "Wed,
 13 Feb 2002 15:30:13 +0100")
References: <3C6A7825.714DF612@setf.de>
Message-ID: <x74rkltja7.fsf@ra.iago.org>

james anderson <james.anderson at setf.de> writes:

> is there a python to lisp translator out there any where?

Not that I know of.

If there really was a good one (to either CL or Scheme) that would
have the possibility of mollifying many of the performance-obsessed
folks, since both of those languages have some good optimizing compilers.

However, it would be a seriously hard thing to do correctly and well.

-Justin

 




From dmaster at synopsys.com  Thu Feb  7 11:53:05 2002
From: dmaster at synopsys.com (David Masterson)
Date: 07 Feb 2002 08:53:05 -0800
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com> <1731.803T2524T145894threeseas@earthlink.net>
Message-ID: <uy9i5b5e6.fsf@synopsys.com>

>>>>> Timothy Rue writes:

> On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:

>> That is putting the cart in front of the horse.  We cannot create a
>> tool to solve a problem until we know what the problem is.

> That's not true.  Not only has the problem been identified but it
> has been pointed out that it is not a simple one time
> problem/solution matter.

You *may* understand the problem -- we *DON'T* (at least not your
explanation of it).

> I have pointed out that programming is the act of automating
> complexity that is made up of simpler things, in order to make use
> of complexity simple for the users of that complexity.

Yes.  There are *MANY* computer related tools that do this (to one
degree or another) already.  We haven't figured out what makes your
approach any better or even any different.

> The problems is that there is no general automation tool oriented to
> all users. Yet the essence of programming boils down to two
> deminsional bit flipping.

Yes, but every time you add another "bit" to the problem, the problem
becomes an order of magnitude (base 2) harder.  Thus far, nobody has
determined the number of "bits" needed to represent all problems.

> I've identified and defined the tool and it's natural components.

> The problem is in getting an industry that has been programmed to
> depend on some level of manual control over what the users can and
> can't do, to let go of such control and make available such a
> natural automation tool as identified, so that the user can do
> things for themselves, rather than being dependant on the
> programmers and whether or not they have the resources to get a
> programmer to do what they need done.

Thus far, I see nothing "natural" in your approach.

> The problem is perhaps to prevasive for you to see?

Possibly.  I may have been in the "forest" for so long that all I can
see are "trees".  On the reverse side, perhaps you've been looking at
the "forest" for so long that you don't see how each "tree" has an
impact on it.

-- 
David Masterson                dmaster AT synopsys DOT com
Sr. R&D Engineer               Synopsys, Inc.
Software Engineering           Sunnyvale, CA


From eppstein at ics.uci.edu  Wed Feb 27 14:46:57 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Wed, 27 Feb 2002 11:46:57 -0800
Subject: Status of PEP's?
References: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com>
Message-ID: <eppstein-546E75.11465727022002@news.service.uci.edu>

In article <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p at 4ax.com>,
 Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote:

> Anyone knows how is the "acceptance status" of the following PEP's, that
> is, is there any idea if they will be accepted for the next Python
> realease:
> 
> PEP 276: Iterator for ints.
> (I really love this one - as simple and "obvious" as it gets)

I don't know an answer to your question, but in contrast to your opinion, 
I really dislike this one.  It gives a very nonintuitive meaning for
"x in y" when both are integers (equivalent to "x <= 0 < y"), and that 
meaning would hold in all contexts, not just for-loops -- I would much 
prefer that such expressions get flagged as likely coding error and that 
more explicit syntax be required to create integer ranges.  I think there 
should be a better syntax for for-loops over integers, but this isn't it.

(See earlier discussion of PEP276 in this group for proposed alternatives, 
I don't want to repeat the whole same discussion again here.)
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From aleax at aleax.it  Mon Feb 18 11:49:59 2002
From: aleax at aleax.it (Alex Martelli)
Date: Mon, 18 Feb 2002 16:49:59 GMT
Subject: List unchanged in function--why?
References: <fa7108b0.0202180838.6557992a@posting.google.com>
Message-ID: <Xgac8.83678$3J1.2676280@news2.tin.it>

Chris wrote:

> This must have something to do with references etc, but I'm not sure
> what's going on.
> 
> def changeList(theList):
>   theList = ['a']

You're re-binding the local name 'theList' so that it now refers
to a different object to whatever it previously referred to.

Rebinding a name never affects whatever object the name was
previously bound to.

Maybe what you want to do is instead, as the body of changeList:

    theList[:] = ['a']

this rebinds the *whole-list slice*, i.e. the CONTENTS of the
object currently bound to local name 'theList', rather than
the name itself.  Apparently you do want to rebind the contents
of the object, rather than the name, right?


Alex



From peter at engcorp.com  Thu Feb 14 18:22:22 2002
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 14 Feb 2002 18:22:22 -0500
Subject: Java vs Python ... ???
References: <Xns91B5CBEF0A103Scirious@140.99.99.131>
Message-ID: <3C6C46AE.72FAF999@engcorp.com>

Scirious wrote:
> 
> I've been learning JAVA for a while, but now I've heard avout Python.
> Any coments concerning these two languages?

It's Java, not JAVA, and have you tried Google yet?  Try a 
search with "comparison" and "Python" and "Java" and see what
you get before you post here...


From sabren at manifestation.com  Sat Feb 23 09:13:56 2002
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 23 Feb 2002 09:13:56 -0500 (EST)
Subject: HELP! Obscure Python Reentrancy Question
In-Reply-To: <hdde7ukvfch6jpfovghvm92abaiq26q2i2@4ax.com>
Message-ID: <Pine.LNX.4.44.0202230904310.23509-100000@mercury.sabren.com>

On Sat, 23 Feb 2002, Courageous wrote:

> I didn't think that Python was structured to allow
> multiple interpreters in a single process?

I know python can do this because mod_python uses this
feature.

Also, this may be completely irrelevant, but I had problems
connecting a callback in a C extension to python. I was
hooking up a MIDI keyboard to python. I could play notes
just fine from python to the keyboard, but when I tried to
go the other way (hitting a key on the keyboard), it
triggered a callback into my extension module, and that
screwed up the python interpreter.

The solution had something to do with the python interpreter
lock. I forget how it worked. I stole the code from someone
else. But if it helps, the code is online at:

  http://sabren.net/code/python/midi/winmidi/winmidi.cpp


Cheers,

- Michal   http://www.sabren.net/   sabren at manifestation.com 
------------------------------------------------------------
Give your ideas the perfect home: http://www.cornerhost.com/
 cvs - weblogs - php - linux shell - perl/python/cgi - java
------------------------------------------------------------




From mal at lemburg.com  Thu Feb 21 05:27:56 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 21 Feb 2002 11:27:56 +0100
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com> <258fd9b8.0202201733.3888e711@posting.google.com>
Message-ID: <3C74CBAC.9AF1C17C@lemburg.com>

Magnus Lyck? wrote:
> 
> fumingw at yahoo.com (Fuming Wang) wrote in message news:<b0fd9226.0202201104.2c6a220 at posting.google.com>...
> > Hi,
> >
> > I am having trouble using the arrow key to recall commands in console
> > interactive mode.  I even installed Cygwin and start Python from there
> > but still got the same effect.  Is there anything I can do to get
> > arrow key working?
> 
> You can use something like IDLE, PythonWin or PyCrust instead.
> 
> As suggested you can also use the cygwin version of Python,
> but then I guess you will have trouble with a lot of extensions
> that use C or C++ such as mxDateTime, win32com etc.

The egenix-mx-* packages should compile out of the box
using cygwin, not sure about the other extensions.

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From jonathan at onegoodidea.com  Sun Feb 10 10:54:03 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Sun, 10 Feb 2002 15:54:03 +0000
Subject: OT: (ambi)dexterity of Pythonistas
References: <mailman.1013196740.22073.python-list@python.org> <a41oi1$a22$1@panix2.panix.com> <mailman.1013216835.14351.python-list@python.org> <B88AE186.48CD%jonathan@onegoodidea.com> <a43ipn$s37$1@panix2.panix.com> <B88B46C1.48F3%jonathan@onegoodidea.com> <FIi98.24853$l93.4540249@newsb.telia.net>
Message-ID: <B88C481B.495E%jonathan@onegoodidea.com>

On 9/2/2002 23:58, in article FIi98.24853$l93.4540249 at newsb.telia.net,
"Fredrik Lundh" <fredrik at pythonware.com> wrote:

> build times???  where are the pystones?
> 
> (I run python programs a lot more than I compile the inter-
> preter -- and I'm sure I compile the interpreter more often
> than the average comp.lang.python reader ;-)

Actually, it was just the first large lump of code that came to hand that I
could compile up from clean and get the other two to follow suit. Neither
use Python.

We were just curious to see how fast the new 2x1GHz Macs are. My friend just
took delivery of one. Lucky bugger.

For the curious though, my iBook 600MHz seems to do about 8300 pystones/s.
I'm not sure if that's good or bad. That's with the Mac OS X framework
version of the interpreter, which is not prebound unfortunately and so takes
a dynamic linking hit.

Jonathan



From rony.steelandt at bucodi.com  Fri Feb  8 03:10:59 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 8 Feb 2002 00:10:59 -0800
Subject: Fine Tkinter Question
References: <mailman.1013104932.12965.python-list@python.org>
Message-ID: <b34d3835.0202080010.4136671b@posting.google.com>

Mariano Corsunsky <core.lists.python at core-sdi.com> wrote in message news:<mailman.1013104932.12965.python-list at python.org>...
> Hi there!
> 
> I have a Tkinter window which opens a 2nd one in which I want to close 
> it
> when you hit the <"Return"> key upon the "Close" button. But..... isn't
> working.
> ...

What you're saying in your text is you want to hit the 'Return' key on
the BUTTON, so IMHO you have to bind the return-key event on the
BUTTON...

Rony


From phd at phd.pp.ru  Thu Feb 14 10:50:18 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 14 Feb 2002 18:50:18 +0300
Subject: ftp binary mode
In-Reply-To: <717c3e15.0202140738.111f9a71@posting.google.com>; from tick@tut.by on Thu, Feb 14, 2002 at 07:38:34AM -0800
References: <717c3e15.0202140738.111f9a71@posting.google.com>
Message-ID: <20020214185018.F7398@phd.pp.ru>

On Thu, Feb 14, 2002 at 07:38:34AM -0800, tick wrote:
> I want to use python function ftplib.retrbinary() for transfering the
> text data to a file.  After transfering every 512 bytes <NL>
> separates.
> 
> How to increase this limit?

   There is no such limit. Where is the problem depends on the program and
data.

PS. Why not to ask Russian-speaking Python and Zope User Group?
   http://zope.net.ru/About/mail-list/

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



From joekm at earthlink.net  Thu Feb 28 21:03:07 2002
From: joekm at earthlink.net (Joe)
Date: Fri, 01 Mar 2002 02:03:07 GMT
Subject: problems with photoimage and tkinter
References: <Y%hf8.149154$dh.38770971@typhoon.nyroc.rr.com> <a5ks4b$61a$1@norfair.nerim.net> <3c7e5639$1@buckaroo.cs.rit.edu>
Message-ID: <3c7ee06c.621416@news.earthlink.net>

On Thu, 28 Feb 2002 11:27:36 -0500, "Keith Jones"
<kjones at docstrategies.com> wrote:

>Eric,
>
>Thanks for the info; I'm glad I'm not just losing my mind. Btw, I'm Keith, I
>was using my brother's account to access the newsgroup last night. Again,
>thanks. I can't wait to get home to try to get this sucker to work. :)
>
>
>
>Keith
>
>
>
>

BTW - just went through a similar PhotoImage thing myself.  The FAQ at
www.python.org (#4.69 I think) mentions something similar as well -
although the author (Frederick Lundh (?) ) says it is not a bug.  It
is frustrating though.

If you search Google (I tried Tkinter PhotoImage bug), you find some
stuff as well.


From jmsun at bioeng.ucsd.edu  Thu Feb 14 16:29:52 2002
From: jmsun at bioeng.ucsd.edu (Jeffrey)
Date: 14 Feb 2002 13:29:52 -0800
Subject: Tkinter Resize
Message-ID: <3daa1aa8.0202141329.63fdc057@posting.google.com>

Hi Everyone,
  I'm working with a Tkinter Text Widget.  How do you bind a function
to the Text Widget whenever it resizes?  Plus the function I want to
bind needs to detect the new size of the Text Widget.  Is it possible
to get the new width by doing:  text_widget.cget( 'width' )?

                                                       Thanks,
                                                             Jeff


From pinard at iro.umontreal.ca  Sun Feb 17 19:21:05 2002
From: pinard at iro.umontreal.ca (=?iso-8859-1?q?Fran=E7ois?= Pinard)
Date: 17 Feb 2002 19:21:05 -0500
Subject: Problems with telnetlib
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIECINNAA.tim.one@comcast.net>
References: <LNBBLJKPBEHFEDALKOLCIECINNAA.tim.one@comcast.net>
Message-ID: <oqg03zwsge.fsf@titan.progiciels-bpi.ca>

[Tim Peters]

> [Bengt Richter]

> > BTW, why are context diffs preferred over unified?

> Because Guido says so.

So, Guido and Richard (Stallman) agree on this one! :-)

I became a bit tired of having to remember the preference of each maintainer.
In my experience, many maintainers, if not most of them, prefer unified.

> OTOH, the opposite seems true more often to me when looking at simple
> patches.

Probably most maintainers prefer simpler patches then, or write code for
which patches are simpler on average! :-)

> Someone should write a tool to generate both patch flavors, compare them,
> and objectively determine which is clearer.  Indeed, I demand someone
> do this at once <wink>.

Someone contributed `unify.[1c]', which I planned to integrate in `wdiff'
as at the time, as the `diffutils' maintainer was just not receptive to
this contribution.  Richard told me that `wdiff' should get integrated into
`diffutils'.  I do not know exactly what will happen nor how it will happen,
but surely, nothing happened in years.

I bent `unify' to GNU standards and made it part of some `wdiff' pretest,
a long while ago.  It should be easy to transmit separately to whoever
asks, I guess.  This tool converts between unidiffs and context diffs.
This should help maintainers having preferences at staying happy, without
having to fight or burden all contributors with their own little tastes
and distastes.

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



From tdelaney at avaya.com  Thu Feb 21 18:34:56 2002
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 22 Feb 2002 10:34:56 +1100
Subject: strings and sort()
Message-ID: <B43D149A9AB2D411971300B0D03D7E8BF0A1A8@natasha.auslabs.avaya.com>

> From: caljason76 at yahoo.com [mailto:caljason76 at yahoo.com]
>
> I still think that the convenience of having sort() return the list
> far outweighs the cost.  People would only get bit once, maybe twice,
> before they learned the behavior of sort.

Actually, this is the type of subtle bug which is likely to hang around for
a while without being noticed (or be the cause of hour-long bug-hunts).

OTOH, because sort() returns None, it is impossible for someone to then try
to use the result as a list without an exception being thrown, revealing
exactly where the error occurred. No subtle bugs here.

People only get bitten once, maybe twice, before they learn the behaviour of
sort(). And they are guaranteed to *know* that they have been bitten.

> I like Python, but the biggest problem I have with the language right
> now is what low code density it has at times.

Huh? Low code density? Huh?

Tim Delaney



From xscottgjunk at yahoo.com  Thu Feb 21 11:41:09 2002
From: xscottgjunk at yahoo.com (Scott Gilbert)
Date: 21 Feb 2002 08:41:09 -0800
Subject: Embedding: How to set globals for PyObject_Call
References: <mailman.1014234635.18641.python-list@python.org>
Message-ID: <79b2b0bb.0202210841.457c0b81@posting.google.com>

"Jason Orendorff" <jason at jorendorff.com> wrote:
> Scott Gilbert wrote:
> > I dove through the code, and it appears that I need to setup a
> > PyFrameObject to be associated with my PyThreadState.  I wasn't able
> > to figure out how to do that though.
> > 
> > As you can see in my example below, I have the globals sitting right
> > there.  I just don't know how to tell Python that those are the
> > appropriate globals while invoking PyObject_Call(...).  A similar
> > problem occurs when evaluating "globals", "locals", "dir", or
> > functions that call those.
> 
> Really?  The part about "functions that call those" sounds a bit off.
> I thought Python functions that call globals() would work.
> Each Python function contains a pointer to its globals namespace
> (the fn.func_globals attribute), and calling a Python function
> creates a PyFrameObject.  It seems as though it should work.
> 

You're right, it's not a problem with functions that call those.  My
bad, I should have tried before making that assumption.  So I guess
it's not really that big of a problem.  Thank you for pointing this
out.

> So, what are you really trying to do?  :)  I have a feeling that
> calling vars() directly from C is a mistake; there's probably an
> easier way to do what you need.

I have a C++ object class "Var" (short for arbitrary Python Variable)
that represents a PyObject.  Var also has a ton of overloads
(including operator ()) to make it fairly transparent to work with. 
Here are three examples:

  // This one breaks
  Var g = VarEval("globals");
  g(); // This causes a segfault

  // This one works
  Var g = VarEval("globals()");

  // As you pointed out, This one works too
  VarExec("def GLOBALS(): return globals()");
  Var g = VarEval("GLOBALS");
  g();

Many other functions in the __builtin__ module work just fine.  For
instance:

  Var hex = VarEval("hex");
  VarPrint(hex(1234));

So I think it's a little odd that some builtins are different in this
regard than others.

> 
> For example, if you just want to look at module's namespace,
> that's easy:
> 
>   result = PyModule_GetDict(module);
> 

Actually, what I want is to allow the users of my library to do
anything from C++ that they could do from Python.  I know this isn't
possible completely, but I'm trying to fill in the edges where I can. 
The builtin functions "dir", "vars", "globals", "locals" will have to
go on the list of things that have quirks...

Thanks again for your response.


From fperez528 at yahoo.com  Wed Feb  6 09:48:34 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Wed, 06 Feb 2002 14:48:34 +0000
Subject: ANN: HappyDoc 2.0.1
References: <mailman.1012913477.7851.python-list@python.org> <a3pqbh$nru$2@peabody.colorado.edu> <Xns91ADF0977CADRASXnewsDFE1@130.133.1.4>
Message-ID: <a3s8di$5r4$1@peabody.colorado.edu>

Philip Swartzleonard wrote:

> Hmm.. i can't really remeber what pydoc's html output looked like (it does
> do that, right? :), but I liked happy's a lot better when i compared them,
> and i think it had a better customization framework...

It's quite good, though I can't compare to happydoc's. The problem with pydoc 
is that it tries to do a full import of modules and fails more often than I 
care. It would also be nice if it showed html-ized (color-coded) source code, 
which it doesn't. I'm so used to syntax highlighting that I always end up 
opening the files in emacs just to read anything non-trivial.

> Well, pydoc provides internal, interactive help, and happy seems to be
> geared toward producing static documents (of which i think it does a better
> job)... but I don't see any moral reason why happydoc's html output system
> and better introspection (and whatever else it does good) couldn't be
> merged into pydoc. :\

My point exactly. We have two open projects with vast amounts of overlap and 
certain individual strengths each. *Exactly* the problem calling for a 
judicious integration of the better parts of each, all for Python's eternal 
glory :)

Cheers,

f.


From deathtospam43423 at altavista.com  Fri Feb 22 10:30:50 2002
From: deathtospam43423 at altavista.com (netvegetable)
Date: 22 Feb 2002 15:30:50 GMT
Subject: newbie: Thnx to Gustavo and Gerhard!
Message-ID: <Xns91BE1487158E7AusYourStandingInIt@130.133.1.4>

Some time ago I posted a thread asking for some File IO tips. These two 
posters provided some tips, and some practical examples, that were 
immensely useful in getting my chat script up and running. I learned a 
great deal.


I took a while in replying, so incase they don't remember the thread 
"newbie: Some file object tips?", I just wanted to thank them for their 
input. 


Thankyou to Gustavo Cordova and Gerhard =?unknown-8bit?Q?H=E4ring?=  !




-- 
"You can tell whether a man is clever by his answers. You can tell whether 
a man is wise by his questions."  Naguib Mahfouz

netvegetable at dingoblue.net.au


From claird at starbase.neosoft.com  Mon Feb 18 17:16:31 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Feb 2002 16:16:31 -0600
Subject: SOAP (Air Fare Checker) - Problems building Velocigen's getFare request
References: <7316200b.0202052047.7584b71@posting.google.com>
Message-ID: <A87943F0965223D3.893224AD2474B1B1.FD0E1A16A31D3F70@lp.airnews.net>

In article <7316200b.0202052047.7584b71 at posting.google.com>,
noel <nce0485 at flashmail.com> wrote:
>I am having problems building the correct request for getFare for the
>following wsdl: http://services.xmltoday.com/vx_engine/wsdl.vep?flight.wsdl.
>
>I can see that I am (at least) receiving the getFareResponse by
>setting dumpSoapIn, but it tells me that I am missing all the
>parameters of the request.  I am using SOAPpy097.  The following was
>taken off the xmethods site (www.xmethods.com):
			.
			.
			.
Here's a small sample program.  I started with one Noel
was working on, and Cayce Ullman made a quick correction.

  import SOAP
  
  def getFare(data = None):
      proxy = 'http://thor.velocigen.com:80/vx_engine/soap-trigger.pperl'
      flights = SOAP.SOAPProxy(proxy)
  
      flights.config.dumpSOAPIn = 1
      flights.namespace = 'urn:vgx-flight'
      flights.soapaction = 'urn:vgx-flight'
  
      return flights.getFare(search = data)
  
  
  if __name__ == '__main__':
      depart_data = {'depart_airportcode': 'SFO',
                'arrive_airportcode': 'LAX',
                'depart_year': 2002,
                'depart_month': 'Mar',
                'depart_day': 22}
      return_data = {'return_year': 2002,
                'return_month': 'Mar',
                'return_day': 26}
      print getFare({'depart': depart_data, 'return': return_data})
  
As on Monday, 18 February 2002, the Fare Checker's telling me
United Airlines has a flight for which it charges $126.50.
-- 

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


From emile at fenx.com  Sat Feb  2 18:09:32 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 2 Feb 2002 15:09:32 -0800
Subject: hex to normal chars?
References: <mailman.1012690087.19312.python-list@python.org>
Message-ID: <a3hrt7$18qgnb$1@ID-11957.news.dfncis.de>

"maximilianscherr" <MaximilianScherr at T-Online.de> wrote in message
news:mailman.1012690087.19312.python-list at python.org...
> how can i transfer hex numbers to normal chars, like "41" to "A" etc.?
>

>>> chr(int("41",16))

--

Emile van Sebille
emile at fenx.com

---------



From tim.one at home.com  Mon Feb  4 16:25:36 2002
From: tim.one at home.com (Tim Peters)
Date: Mon, 4 Feb 2002 16:25:36 -0500
Subject: Backwards Compatibility of Python versions
In-Reply-To: <7xsn8hjxzw.fsf@ruckus.brouhaha.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMELJNJAA.tim.one@home.com>

[Paul Rubin]
> ...
> If the official Python maintainers think they might release a 1.5.3
> someday, then it's fair to say that the 1.5 series is still current.
> If they're unlikely to ever release another 1.2.x or 0.9.x, then
> those versions are not current, even if I release a 1.2.x myself.

You misunderstand the bugfix release process:  "the official Python
maintainers" don't do bugfix releases, period.  We haven't had the bandwidth
for it, and it doesn't look like we will.   All bugfix releases have been
produced by community volunteers (2.0.1 was an exception, but was an
artificial release just to get a GPL-compatible according-to-FSF license on
it).  Fred Drake, Barry Warsaw and I have so far been able to put a little
bit of time into doing the last steps of bugfix releases (packaging and
file-uploading and announcing), but that's all.

The odds of getting a 1.5.2 bugfix release are, as for any other bugfix
release, identical to the odds of someone volunteering to produce one.
Guido asked several times for a volunteer to step up, starting about 6
months after 1.5.2 was released.  Nobody bit.  People *have* cared enough to
volunteer bugfix releases for the 2.1 and 2.2 series, which makes claims
about the ongoing vital importance of 1.5.2 to the community suspect.  It's
not that it's old that makes me judge the odds of a 1.5.2 bugfix release
close to zero, it's the years of historical evidence that nobody cares about
it enough to bother.

> ...
> (I'm still using 2.1 and plan to get around to upgrading to 2.1.1).

You should move to 2.1.2 next, then (2.1.1 isn't even "current" by your
definition <wink>).

> ...
> Given that 2.2 can import 3.0 division from the "future", will 3.0
> have a way to import 2.x division from the "past"?

That's a couple years away, so who knows.  My stong guess is no.

...

>> If you need a given static instance of "/" to do
>> radically different things depending on whether you feed it
>> ints or floats dynamically, you're the first person to claim such
>> a need.

> I want my programs to run the same way across the different Python
> versions that people are using now and will be using in the near future.
> Am I really the first person to want that?

You're the first in a long time to work so hard at misreading <wink>.

>> Else you have a specific flavor of division in mind when you type
>> "/", and it's easy to write it in such a way that it will deliver
>> the same results in 3.0 as in 0.9.8.

> By using divmod(x,y)[0] instead of x/y everywhere I want to divide
> ints, per the PEP?  Ugh!

I don't know:  only you know what your code is trying to achieve.  The only
instances of potential problems we found in the standard library had two
things in common:  (a) it was certain that ints (not longs) were involved,
and (b) it was certain the numerator and denominator were non-negative.  All
such cases can be made version-independent by changing, e.g.,

    h = i / 2

to

    h = int(i / 2)

In practice we *usually* recoded as

    h = i // 2

instead, because there's usually no promise that you can drop a standard
library module from one release into an older release.

If you have messier problems, and you're determined to write code that works
unchanged back through the first Python ever released, then sure -- slicing
divmod is clear and works.  If the clumsiness makes you determined to move
to a more recent release, that's good:  so long as nobody is willing to
maintain older releases, people using them are on their own, and so their
lives are going to get harder as time goes by (e.g., glibc changes in
incompatible ways too over time, so if they ever upgrade their libc, their
old Python may not even compile or link anymore).

OTOH, people "desperately wedded" to an old release are usually stuck there
because of a specific app they rely on, and whether new modules work for
them doesn't really matter.  If "works under 1.52!" were a marketing bullet
point, your market research team would probably decide to devote the ad
space to repetitions of "quantum leap" and "best of breed" instead <wink>.




From gh_pythonlist at gmx.de  Fri Feb 22 15:59:10 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Fri, 22 Feb 2002 21:59:10 +0100
Subject: Newbie: Crazy, but Quick
In-Reply-To: <19650FF08A99D511B19E00B0D0F06323060AE1@SRM201002>
References: <19650FF08A99D511B19E00B0D0F06323060AE1@SRM201002>
Message-ID: <20020222205909.GC2806@lilith.hqd-internal>

Le 22/02/02 ? 11:34, Gustavo Cordova ?crivit:
> Has a python wrapper for SQLite been written yet?

I've started to work on one in October, but didn't pursue this. If you
want to continue the work, I'd send you the files. 

Just wrapping sqlite is pretty easy - implementing a Python DB-API 2.0
wrapper around that would be a lot more effort, though I'd try to rip
the DB-API support from another module.
 
Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.9 ?C      Wind: 6.4 m/s



From jon+usenet at unequivocal.co.uk  Mon Feb  4 09:16:59 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Mon, 4 Feb 2002 14:16:59 +0000 (UTC)
Subject: Python 2.2 Modules/Setup lies?
References: <slrna5r6u4.6e9.jon+usenet@snowy.squish.net> <mailman.1012768207.25312.python-list@python.org> <slrna5rc6q.dtp.jon+usenet@snowy.squish.net> <m3sn8hkbqh.fsf@mira.informatik.hu-berlin.de>
Message-ID: <slrna5t5up.mr.jon+usenet@snowy.squish.net>

In article <m3sn8hkbqh.fsf at mira.informatik.hu-berlin.de>,
  Martin v. Loewis wrote:
>This suggests that editing Modules/Setup is optional, and indeed, it
>is.

There is nothing to tell you this though. Normally if a README says to
run ./configure, then maybe edit another file, you go take a brief
look at that file to see if you do indeed need to make any changes. If
when you get there you find that in fact things you need have not been
enabled, then you edit it. Since that is always going to be the case
with a Python 2.2 install, it means that if you're following the
instructions as given then editing Modules/Setup is mandatory and very
confusing.

The problem appears to be that the install procedure has changed
radically, without the install instructions changing at all.
Modules/Setup is no longer a file you edit to enable the standard
modules, it's a file you edit only if you need to change the compile
options (such as where to find libraries, etc). It might be a good
idea to mention this somewhere in the README!


From thomas.heller at ion-tof.com  Tue Feb  5 15:25:09 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Tue, 5 Feb 2002 21:25:09 +0100
Subject: distutils on win32
References: <6bd9f01b.0202051200.74b0267b@posting.google.com> <a3peik$1a1or9$1@ID-59885.news.dfncis.de>
Message-ID: <a3pf40$1a0o3o$1@ID-59885.news.dfncis.de>

"Thomas Heller" <thomas.heller at ion-tof.com> wrote in message news:a3peik$1a1or9$1 at ID-59885.news.dfncis.de...
> "sameer" <sameer_ at email.com> wrote in message news:6bd9f01b.0202051200.74b0267b at posting.google.com...
> > i'm using distutils to make a windows installer for a python/wxpython
> > app.  what's the fastest way for me to create shortcuts on the desktop
> > as well as the Start menu for the file I want the users to be able to
> > run.  I don't want them to have to go into the correct package, and
> > run the app from there.
>
> Not easy, I suppose. The fastest way would probably be to prepare the
> distribution with distutils, and use inno setup to install everything
> and create the shortcuts.
>
> Thomas
>
> PS: If you insist on using pure distutils, you would probably create
> your own 'install' command in the setup script, use Mark Hammond's
> win32all extension (which must be present on the target computer),
> to create the shortcuts.
> Caveat: This will only work if the user installs from source, not
> from the distutils windows installer.
>
PPS: I hope to finish a new version of bdist_wininst soon, which
allows to run an install-script after the files are copied.
So there *is* hope for a pure distutils solution.




From threeseas at earthlink.net  Wed Feb  6 23:28:50 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:50 GMT
Subject: Autocoding project proposal.
References: <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net> <slrna61kma.57s.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <1873.802T1682T14394596threeseas@earthlink.net>

On 06-Feb-02 01:52:58 Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:

[snip - some reference to thinking logically]

>I think I already do - not perfectly of course, I'm a human I have
>emotions and prejudices that get in the way of pure logical thought. I
>don't understand what is illogical about offering advice on something I
>have more experience than you in (in all probability anyway). I know
>understand that trying to communicate with you is illogical, as you seem
>to just want to rant without communicating anything.

I'm one person, dealing with many others. Some who think a three
deminsional data array is the same as 3 deminsional reality.

<shrug>

Perhaps that party was simply weak minded enough to seek a negative within
what I had said in my last group of posts, and found my use of deminsional
things to be reference to the same type of things?

Fact is, I have had more trouble communicating the VIC to programmers than
to non-programmers. In fact non-programmers often pick up quite quickly
when I'm communicating to them in person, where I can more easily show the
action set in motion, then in describing the action set in text (as here).

There continues to be this persistance to preceive the VIC as some sort of
"language" no matter how many times I say it's not. And this follows wit
the idea that it's an idea, which it is not.

Newton lied about gravity, because of why? People wouldn't understand.

Only I cannot do that here. It really just doesn't work that way.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From joshm at taconic.net  Sat Feb  9 12:52:27 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sat, 9 Feb 2002 12:52:27 -0500
Subject: Apology
References: <mailman.1013120898.32043.python-list@python.org> <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net>
Message-ID: <3c655f8a$1_1@corp.newsgroups.com>

> I kinda like the one on ethics, amoung some of the others.
> But then group voting is like a conspiracy huh?

I'll bring it up at the next meeting.  Sorry, no smiley.

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From phr-n2002a at nightsong.com  Sun Feb 10 16:44:13 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 10 Feb 2002 13:44:13 -0800
Subject: Bug or feature? 'abc'.split('') rejects empty separator
References: <a46heu$8f2$0@216.39.172.122> <3C66CDA4.8E38F4CA@alcyone.com>
Message-ID: <7xlme1hv0y.fsf@ruckus.brouhaha.com>

Erik Max Francis <max at alcyone.com> writes:
> >  >>> 'abc'.split('')
> >  Traceback (most recent call last):
> >    File "<stdin>", line 1, in ?
> >  ValueError: empty separator
> > 
> > Wouldn't it make sense to return list('abc') ?
> 
> If that's what you really wanted, it would make far more sense to just
> use list directly rather than calling string.split on the string.  The
> error seems reasonable to me -- you shouldn't be using string.split if
> that's what you really want.

It seems to me like the principle that programs should be able to "do
nothing" gracefully.  The argument is that split should be the inverse
of join, even when the splitting string is empty.

Imagine if "a+0" raised an error.  You wouldn't say "just use 'a' instead".


From mpoulin at verinet.com  Mon Feb  4 13:39:52 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Mon, 04 Feb 2002 13:39:52 -0500
Subject: directly executing a python script in Unix
References: <60018f02.0202041208.67d3100e@posting.google.com>
Message-ID: <3c5ef07d$0$2201$7586b60c@news.frii.com>

In article <60018f02.0202041208.67d3100e at posting.google.com>, "Marc
Laymon" <laymon at crd.ge.com> wrote:

> I am just learning python.  I have written a few simple scripts which
> run fine when invoked using "python myscript.py".  The Python manual
> says you can execute a Python script directly In Unix by making the
> first line "#!/usr/bin/python", then giving the command "./myscript.py".
>  However, when I try this, I get an error "Command not found.".  I get
> the same error under both Linux and Solaris. Using the same format for
> perl works fine.  Why doesn't the "#!/usr/bin/python" line work for
> Python ?  (perl and python are both in /usr/bin on my Linux system.)
> 
> Thanks.
> 
> Marc Laymon

First, use the "which" command to make sure the interpreter is in your
path:

     which python

You should see

     /usr/bin/python

Next, make your script executable. Type

     chmod +x myscript.py

at the command line.


From sholden at holdenweb.com  Fri Feb 22 07:45:14 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Feb 2002 07:45:14 -0500
Subject: Reloading modules
References: <mailman.1014332376.19219.python-list@python.org>
Message-ID: <Z2rd8.77777$n73.1505518@atlpnn01.usenetserver.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:mailman.1014332376.19219.python-list at python.org...
> > From: Oleg Broytmann [mailto:phd at phd.pp.ru]
> >
> > On Thu, Feb 21, 2002 at 12:02:08PM +0000, Michael Hudson wrote:
> > >     if file.endswith("c"):
> > >         file = file[:-1]
> >
> > if file.endswith("c"):
> >    file = file[:-1]
> > elif file.endswith("o"):
> >    file = file[:-1]
>
> if file[-1:] in ('c', 'o',):
>     file = file[:-1]
>
if file[-1] in "co":
    file = file[:-1]

not-that-there's-more-than-one-way-to-do-it-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From nelson at crynwr.com  Wed Feb 20 09:49:14 2002
From: nelson at crynwr.com (Russell Nelson)
Date: 20 Feb 2002 09:49:14 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int>
Message-ID: <m2wux85hud.fsf@desk.crynwr.com>

Henrik Motakef <henrik.motakef at web.de> writes:

> Gerhard H?ring <gh_pythonlist at gmx.de> writes:
> 
> > Le 18/02/02 ? 15:28, Tim Peters ?crivit:
> > > Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> > > czignat?  Classifero en bragna!
> > Woas da Deifi wos des hoasn mog.
> 
> Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...

Even better: we could switch languages within threads, to *really*
keep the monolinguists confused.

-- 
-russ nelson              http://russnelson.com | Crypto without a threat
Crynwr sells support for free software  | PGPok | model is like cookies
521 Pleasant Valley Rd. | +1 315 268 1925 voice | without milk.
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | 


From danielk at aracnet.com  Mon Feb 25 23:35:27 2002
From: danielk at aracnet.com (Daniel Klein)
Date: Mon, 25 Feb 2002 20:35:27 -0800
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <ar3m7ukitr0rutarcq22usg75is8vvktjl@4ax.com>

On 25 Feb 2002 08:18:39 -0800, mtaylor at lowebroadway.com (Moray Taylor) wrote:

>Can anyone reccommend a good python editor for UNIX/linux/IRIX. I
>don't mind paying for it, but $300 for SlickEdit is a little steep.

Visual SlickEdit is not the answer either. When you view the function
headings, it doesn't show any of the 'class' statements, which makes finding
things quite confusing, if your module contains more than just a few classes,
all of which have __init__() methods, etc... I submitted this as an
enhancement request but you know how that goes. :-(

That being said, I just use IDLE. :-)

Dan


From pereira at cis.upenn.edu  Tue Feb 26 22:09:12 2002
From: pereira at cis.upenn.edu (Fernando Pereira)
Date: Tue, 26 Feb 2002 22:09:12 -0500
Subject: grammar question
References: <pan.2002.02.25.23.08.10.246103.1935@src.uchicago.edu> <m3pu2sfv0m.fsf@mira.informatik.hu-berlin.de> <B8A0F6F8.3C3B%pereira@cis.upenn.edu> <mailman.1014749745.3940.python-list@python.org>
Message-ID: <B8A1B808.3DD8%pereira@cis.upenn.edu>

On 2/26/02 1:54 PM, in article
mailman.1014749745.3940.python-list at python.org, "Justin Sheehy"
<justin at iago.org> wrote:

> Fernando Pereira <pereira at cis.upenn.edu> writes:
> 
>> In general, languages with infix operators and complex nested
>> structures like lists are not LL(k), because a parser can only
>> decide the role of a subexpression after it sees the operator or
>> terminator that follows it.
> 
> One can usually (but not always) create an LL(k) grammar for such
> languages.  It just requires some interesting manipulation of the
> grammar productions such as left factoring and removal of left
> recursion.
> 
What do you mean to "usually"? It can be done for the LC(k) class by
expanding the set of nonterminals and productions considerably, but not all
deterministic languages are LC. (Daniel J. Rosenkrantz, Philip M. Lewis II:
Deterministic Left Corner Parsing (Extended Abstract). FOCS 1970: 139-152)
The LC->LL transformation is rather pretty and can be optimized in
interesting ways; unfortunately I know of no widely avaliable tools that
perform it. The problem with LC is that you need to commit to a reduction
after seeing the first RHS nonterminal and the lookahead at that point,
which may not be good enough for your favorite language.

> This is a pain to do, but once you have done it you can use a nice
> recursive-descent LL(1) parser.

Exactly. Doing the LC transformation by hand is no fun. Also, accurate error
reporting would seem to be harder than with a shift-reduce parser.

-- F



From gerson.kurz at t-online.de  Sun Feb 24 23:50:52 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 04:50:52 GMT
Subject: functional programming with map()
References: <918bc22f.0202241902.2160b75a@posting.google.com> <a5cb54$27c2$1@agate.berkeley.edu>
Message-ID: <3c79bdc9.346597562@news.isar.de>

On Mon, 25 Feb 2002 03:30:44 +0000 (UTC), Daniel Yoo
<dyoo at hkn.eecs.berkeley.edu> wrote:

>Donnal Walter <donnal at donnal.net> wrote:
>: I know that
>
>: map(f,items)
>
>: is equivalent to:
>
>: for x in items:
>:     f(x)
>
>
>map() collects the results of calling f() over all the x in items, so a
>closer translation would be:
>
>###
>results = []
>for x in items:
>    results.append(f(x))
>###

I would think that a block of statements as such has no return value
at all. That is, while every function called has a return value, it
always gets thrown away. It *can* return data, but then thats a
special statement. In a lambda expression, everything has to have a
return value, and be it "None". So, in a way, this comment applys to
all statements-mapped-to-lambda (IF, WHILE, PRINT etc.). 

The only problem I can see with this is a possible memory thing: If
you call n functions, a lambda expression attempts to generate a list
of n return values, while n statements would just throw away their
return values. If n is very large, this might be a limit. (But then,
tail-recursion in lambdaesque-while will probably run into stack
problems earlier). 


From starx at pacbell.net  Fri Feb  8 05:03:31 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 8 Feb 2002 10:03:31 GMT
Subject: __class__ for 2.2 types
References: <mailman.1013116936.12831.python-list@python.org>
Message-ID: <Xns91AF14F1DDFC4RASXnewsDFE1@130.133.1.4>

Michael Chermside || Thu 07 Feb 2002 01:24:09p:

>> hi, rookie here, I was hoping someone could explain to me why this
>> happens: 
>> 
>> Python 2.2 (#28, Dec 21 2001, 12:21:22) [MSC 32 bit (Intel)] on win32
>> Type "help", "copyright", "credits" or "license" for more information.
>>>>> [].__class__
>> <type 'list'>
>>>>> int.__class__
>> <type 'type'>
>> 
>> why doesn't the int one return <type 'int'>? 
>> 
>> thanks in advance
> 
> Because "int" isn't an int. "5" is.
> 
> 
>>>> [].__class__
> <type 'list'>
>>>> x = 7
>>>> x.__class__
> <type 'int'>
> 
>>>> int.__class__
> <type 'type'>
>>>> str.__class__
> <type 'type'>

You will also notice that 
>>> [].__class__.__class__
<type 'type'>
>>> x.__class__.__class__
<type 'type'>

Also, there is a type object laying in the standard builtins (along with 
object and int and all that).

>>>[].__class__.__class__ is type
1

And now, if you want to go insane:

>>>type.__class__ is type
1
>>>type.__class__.__class__ is type
1

and so on. :)


-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From sholden at holdenweb.com  Fri Feb  8 14:23:50 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 14:23:50 -0500
Subject: Defending the Python lanuage...
References: <3C62EE4A.5090305@destiny.com><20020207140755.364d5d77.logiplexsoftware@earthlink.net><200202081311.g18DBQGo020260@ratthing-b246.strakt.com> <mailman.1013194035.3597.python-list@python.org>
Message-ID: <0BV88.108473$KM2.4226261@atlpnn01.usenetserver.com>

"Cliff Wells" <logiplexsoftware at earthlink.net> wrote ...
> On Fri, 08 Feb 2002 14:11:26 +0100
> Laura Creighton wrote:
>
[ ... ]
> >
> > This is incorrect.
>
> How dare you.   ;)
>
> > It works great with people with radically different
> > skill levels.  Of course, they start to converge, so they do not end
> > up with different skill levels.  thats the point.  For example, people
> > who pair program with me and are not aware of design patterns and
> signature
> > based polymorphism will get that fixed in a hurry.  But what happens
> > when Alex Martelli and I are pair programming is I learn so much stuff
> > I never even heard about before.
>
Alex will be glad he isn't going to get denser :-) -- "convergence" in my
mind tends to imply both ends moving towards a midpoint, and reading that
last paragraph I immediately thought "Damn, so whoever sits with me will get
dumber" :-)

> No doubt.  This is in fact how I have trained people in the past - they
sit
> and watch me program with some toolkit and we discuss what I'm doing (I
had
> great success teaching someone with no programming experience whatsoever
> how to integrate Python, wxPython and MySQL - in less than 3 months he was
> writing on his own [and damn arrogant too, I might add]).

In the UK this is known as "sitting with Nellie" (Nellie typically being
someone who has been doing the job you are supposed to learn since before
you were a twinkle in your parents' eye). It's not terribly highly regarded
as an isolated technique, but mostly because Nellie only knows how to *do*
the job, and may not have much theoretical insight into its fundamentals.

>  However, I think
> a key point in his progress a a programmer was when he finally sat at his
> own computer and worked independently of me.  I'm reminded of a Chinese
> proverb I read in a calculus book:
>
> I hear and I forget
> I see and I remember
> I do and I understand
>
Hence the value of the "Hands-On" elements in the classes I teach, for
example.

> Now you know Laura, that your posts are so rare and insightful that I am
> loathe to disagree with you, -but- ;) my point is that while working
> together is extremely beneficial for sharing ideas and information, I
don't
> see sharing the same console with another programmer for 90% of the time
to
> be as beneficial to productivity.  Sitting side-by-side at separate
> consoles, yes, but not at the same console (except when there are
> interesting discoveries/techniques to be shared or design questions to be
> pondered).
>
Well, this merely compounds the mis-empression you mention below ...

> However, this discussion does give me the idea that pair-programming could
> be very beneficial if programmers were rotated on some schedule, so that
> one day they watched someone else and on another someone else watched
them,
> especially given a group of several programmers that could be rotated
> through.  This way, programmers would get enough time doing development on
> their own but still have a chance to share ideas with other programmers in
> their group.  Too bad I'm not in a position to give it a try :P
>
The idea behind pair programming is exactly that such rotations should take
place. You seem to think that two prograqmmers are joined at the hip when
teams are formed, but in fact the pairings arise fairly spontaneously in
most cases.

> > It is such a joy.
>
> I'm glad to hear it =)
>
> > > 2.  like and respect each other
> > > 3.  have compatible programming philosophies
> >
> > (you are spot on with those, however.)
> >
> > Cliff, again you are up against 'American management is impossible.'
> > Seriously consider visiting Sweden, for a vacation.  I think you would
> > like it here.
>
> You'd better be careful what you ask for - you're starting to tempt me -
> you could end up with some guy with a strange accent and stranger ideas
> hanging around ;)  I expect I could learn quite a bit from you.
>
>
You should try it. I spent six months in Stockholm thirty years ago, and
would still love to go back to Sweden. Also AB Strakt are serious about
Python, so you wouldn't be forced to usae something ghastly like C++.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From logiplexsoftware at earthlink.net  Fri Feb  8 13:46:33 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 10:46:33 -0800
Subject: Voting people out of newsgroups (was Re: Autocoding evolves from........)
In-Reply-To: <slrna662kc.d87.philh@comuno.freeserve.co.uk>
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk>
	<972.799T919T11694325threeseas@earthlink.net>
	<slrna5rpmb.19d.philh@comuno.freeserve.co.uk>
	<3943.801T1063T14354065threeseas@earthlink.net>
	<3C616C5A.4D1BBE1C@ccvcorp.com>
	<2300.802T2206T14055804threeseas@earthlink.net>
	<3C62190A.A54E9A57@engcorp.com>
	<00b901c1af9e$9de1d080$0101010a@local>
	<mailman.1013103792.9086.python-list@python.org>
	<slrna662kc.d87.philh@comuno.freeserve.co.uk>
Message-ID: <20020208104633.51ee1a93.logiplexsoftware@earthlink.net>

On Thu, 7 Feb 2002 23:15:24 +0000
phil hunt wrote:

> On Thu, 07 Feb 2002 18:40:33 +0100, Laura Creighton <lac at strakt.com>
wrote:
> >Voting people out of a newsgroup, as a concept, disgusts me.
> 
> I suspect you might not have heard of Matt Giwer.

Google turns up quite a bit on this guy.  Fascinating.  However, I hardly
feel Timothy Rue falls in this category.  Besides, as mentioned before,
that's what killfiles are for.

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From vvainio at karhu.tp.spt.fi  Mon Feb 18 04:31:10 2002
From: vvainio at karhu.tp.spt.fi (Ville Vainio)
Date: 18 Feb 2002 11:31:10 +0200
Subject: ODBC on unix w/o mxODBC
Message-ID: <yoxadu7xhk1.fsf@karhu.tp.spt.fi>

Briefly put: can it be done? I'm mainly concerned with the price tag
on mxODBC...

-- 
Ville Vainio - http://www.tp.spt.fi/~vvainio - ICQ #115524762
 Wild geese have no intention to cast a reflection
 Water has no mind to assume their form


From threeseas at earthlink.net  Wed Feb  6 23:28:35 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:35 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com>
Message-ID: <2300.802T2206T14055804threeseas@earthlink.net>

On 06-Feb-02 12:48:10 Jeff Shannon <jeff at ccvcorp.com> wrote:
>Timothy Rue wrote:


>> Programming is two deminsional, not three.

>This, sadly, is where you're mistaken, Tim.  Actually, programming is often
>4-dimensional, and can sometimes even be higher-dimensional than that.  I
>routinely deal with the issues involved in a 3-dimensional database changing
>over time (giving a 4th dimension) -- and I consider myself to be a novice.
>(Actually, I think describing the database as being 3-dimensional is
>underestimating its complexity, but...)  We won't even get into how often
>programming is used to solve problems of fractal dimension...

>We understand that you want to make programming easier, so that anyone can do
>it. Almost everyone here, professional programmer or no, thinks that that is
>a great goal;  if we didn't want basic tasks to be automated and simplified,
>we'd be using C and Assembler, not Python.

>The problem you're running into, is that you've failed to convince anyone
>that
>*you* understand the real problems involved, enough to be able to solve them.
> You don't need to convince us of the benefits of autocoding; you need to
>convince us of *your* competence.  So far, you're failing miserably.

>Jeff Shannon
>Technician/Programmer
>Credit International


Since you have taken it upon yourself to be the spokes person for
programmers, I'll have to assume that all of you are so lost that you
can't tell the difference between three deminsional reality and three
deminsional data arrays in the two deminsional world of programming.

So I guess when ever you get hungry, cold and wet, you can just program
yourself up some food, clothing and shelter.

Seems to me that your two deminsional mindset has trouble comprehending
what three deminsional reality really is. With that in mind, it's not
possible for me to convince you of anything beyond your own limited
two deminsional perceptions.

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From scarblac at pino.selwerd.nl  Tue Feb  5 04:33:17 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 5 Feb 2002 09:33:17 GMT
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com>
Message-ID: <slrna5v9io.oqf.scarblac@pino.selwerd.nl>

Paul Rubin <phr-n2002a at nightsong.com> wrote in comp.lang.python:
> "Tim Peters" <tim.one at home.com> writes:
> > > That's the problem--if I don't import future division, my code will
> > > break in the future, because the // operator isn't supported in the
> > > present.
> > 
> > Not so:  introducing "//" was 100% backward-compatible, so isn't controlled
> > by __future__.  "//" works out of the box in 2.2, no __future__ required.
> 
> If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
> it isn't backwards compatible.

It's a new feature. It's impossible that new features work in old versions,
basically by definition. Your idea of backwards compatibility is strange.

However, 1.5.2 programs that use "/" will still work in 2.2 (apart from
other possible incompatible changes since then).

> > I don't know where you're looking, but 2.2 final was released.  
> 
> I know this, but 2.1.1 was released more recently than 2.2, so the most
> recent release was in the 2.1 series.

I'm going to release a new version of 0.9 then, tomorrow. Then you'll have
to claim that's the most recent.

> If the official Python maintainers think they might release a 1.5.3
> someday, then it's fair to say that the 1.5 series is still current.
> If they're unlikely to ever release another 1.2.x or 0.9.x, then
> those versions are not current, even if I release a 1.2.x myself.

The official Python maintainers are working on the next version after 2.2,
probably 3.0. Other people have released a few bugfix
releases for people who really need to continue running old versions.

-- 
Remco Gerlich


From peter at engcorp.com  Fri Feb  1 21:35:50 2002
From: peter at engcorp.com (Peter Hansen)
Date: Fri, 01 Feb 2002 21:35:50 -0500
Subject: Enhanced Generators - the difference between sources and iterators
References: <a3at9p$2he$1@bob.news.rcn.net> <mailman.1012574463.26232.python-list@python.org> <83g04kzoy2.fsf@panacea.canonical.org>
Message-ID: <3C5B5086.9C2110B5@engcorp.com>

Kragen Sitaker wrote:
> 
> same thing.  Are there some situations where structuring a piece of
> code with some hypothetical "accept" expression would yield (pardon
> the pun!) something simpler than just doing the same thing with
> generators?

Wouldn't "accept" be a lot like Intercal's "COMEFROM" statement?

(I thought that would be humorous, until I did a search and noticed,
Cameron Laird's quoting Steven Majewski about how continuations are 
like goto's plus C's longjmp with a little of Intercal's COMEFROM 
thrown in... it's damn hard to be original in the world today. :( )

-Peter


From James_Althoff at i2.com  Fri Feb 15 13:49:00 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 15 Feb 2002 10:49:00 -0800
Subject: Java vs Python ... ???
Message-ID: <OFF143E271.0BADE329-ON88256B61.00674FC1@i2.com>

<Steve Orr>
For instance, with the Oracle Application Server (not the database) you
supposedly can have all your Java apps load balanced "for free" without
having to write load balancing code. So I was wondering the other day,
since
Jython compiles to bytecode, will it run under applications servers (such
as
Oracle) and enjoy the same Java features those servers provide?
</Steve Orr>

Yes.

Jim




From dsh8290 at rit.edu  Wed Feb  6 00:07:27 2002
From: dsh8290 at rit.edu (dman)
Date: Wed, 6 Feb 2002 00:07:27 -0500
Subject: Autocoding project proposal.
In-Reply-To: <3062.801T2750T14076324threeseas@earthlink.net>
References: <h2ve5u4484ejdbqhrj2oqsg6mdoocn31rt@4ax.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <slrna5s2sl.dht.sholden@pgrad.cs.usyd.edu.au> <3062.801T2750T14076324threeseas@earthlink.net>
Message-ID: <20020206050727.GA12156@dman.ddts.net>

On Wed, Feb 06, 2002 at 04:54:13AM +0000, Timothy Rue wrote:
... 
| you are arrogant enough to give advice regarding something you don't
| know about

Figure it out already!!!!  He was giving advice on __communication__,
something that is sorely lacking in all the posts I've read of yours.
I'd say he certianly knows more about communication than you.

-D

-- 

 "Piracy is not a technological issue. It's a behavior issue."   
                                                       --Steve Jobs




From fredrik at pythonware.com  Tue Feb  5 17:38:49 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 05 Feb 2002 22:38:49 GMT
Subject: What is the command to determine an object's type?
References: <mailman.1012941850.17781.python-list@python.org>
Message-ID: <Z9Z78.23594$l93.4286654@newsb.telia.net>

Mark McEahern wrote:
> $ python
> >>> import types
> >>> l = []
> >>> if type(l) == types.ListType:
> ...     print "yeah baby!"
> ...
> yeah baby!

or better:

    if type(l) is types.ListType:
        ...

or even better:

    if isinstance(l, types.ListType):
        ...

or, in 2.2 or later:

    if isinstance(l, list):
        ...

</F>




From eppstein at ics.uci.edu  Thu Feb 28 19:22:47 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 28 Feb 2002 16:22:47 -0800
Subject: algorithm snippets available
Message-ID: <eppstein-270732.16224728022002@news.service.uci.edu>

I've been teaching the undergraduate algorithms course here at UCI this 
quarter using Python syntax for my pseudocode (or, often, actual code) as 
an experiment instead of the C/C++/Java syntax I'd used before.
Anyway, to help the students experiment with the algorithms, I've been 
posting some implementations to my web site, and thought maybe others on 
this list might find them interesting or useful.  They're at
<http://www.ics.uci.edu/~eppstein/161/python/>.

Today's addition: kmp.py, a Knuth-Morris-Pratt string matcher.
Compared to using string.find or regular expressions, it allows the input 
to be any kind of list or sequence (not just a string), and allows the 
longer of the two inputs to be any kind of iterator rather than having to 
be all in memory at once.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From sholden at holdenweb.com  Wed Feb 13 12:23:24 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Feb 2002 12:23:24 -0500
Subject: cgi binary stdout
References: <xNp9EKB5zpa8EwvQ@jessikat.demon.co.uk>
Message-ID: <5ixa8.208958$KM2.8636289@atlpnn01.usenetserver.com>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote ...
> Does anyone know how stdout should be configured in cgi scripts. It
> suddenly occurred to me that IIS/Apache running on win32 might not pass
> stdout configured as binary. Does the CGI script have to ensure that it
> is writing to a binary stream or can it assume the server does that?

Of course, this will depend to an extend on your server. However, I seem to
remember that if you want to do it it's relatively simple (ferkles in
codestash...) aah, yes, try this snippet I stole from MoinMoin:

    # force input/output to binary
    if sys.platform == "win32":
        import msvcrt
        msvcrt.setmode(sys.stdin.fileno(), os.O_BINARY)
        msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From fperez at pizero.colorado.edu  Wed Feb 13 00:36:28 2002
From: fperez at pizero.colorado.edu (=?ISO-8859-1?Q?Fernando_P=E9rez?=)
Date: Tue, 12 Feb 2002 22:36:28 -0700
Subject: Major league corruption - anyone seen anything like this?
In-Reply-To: <15465.63503.142011.246239@12-248-41-177.client.attbi.com>
References: <mailman.1013557302.9749.python-list@python.org>        <7xzo2ekv5g.fsf@ruckus.brouhaha.com>        <mailman.1013572042.18876.python-list@python.org>        <3C69ED33.CE0100DF@pizero.colorado.edu> <15465.63503.142011.246239@12-248-41-177.client.attbi.com>
Message-ID: <Pine.LNX.4.21.0202122224100.12912-100000@higgs.colorado.edu>

On Tue, 12 Feb 2002, Skip Montanaro wrote:

>     Fernando> - what filesystem are you using?
> 
> ext2.

One idea: why don't you remount it as ext3 and try to trigger the condition
again? ext3 is supposed to have a lot of internal safeguards that go off when
'wrong' things happen to the filesystem. Of course, it might be that as far as
the filesystem is concerned nothing 'wrong' is happening (that is, your
process is somehow writing junk but well within the boundaries of its
permissions). But it might help, or provide some extra info.

> [snipped..] 
> Oh well, I'll launch myself back into the fray tomorrow and see if I can
> figure anything else out.

Well, I'm curious. Let us know what the final solution is if/when you find
one. Good luck!

f.





From amuys at shortech.com.au  Mon Feb 25 19:00:19 2002
From: amuys at shortech.com.au (Andrae Muys)
Date: 25 Feb 2002 16:00:19 -0800
Subject: Implementing interfaces with inheritance, good or bad?
References: <mailman.1014655721.10070.python-list@python.org>
Message-ID: <7934d084.0202251600.116cf32d@posting.google.com>

Tripp Scott <tripps81 at yahoo.com> wrote in message news:<mailman.1014655721.10070.python-list at python.org>...
> I just saw:
> 
>   http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/68445
> 
> in which interfaces are implemented using class, and a class 
> implementing one or more interfaces subclasses the interfaces 
> class(es). Now suppose class C implements two Interfaces, I1 and 
> I2. Both interfaces define an operation foo(). In one case, I 
> want to instantiate C because of I1, and in other case, because 
> I need a class that implements I2. How does C implement both 
> foo() operations (since foo() will be implemented as a method)? 
> And how do I tell C that I want I1, or I2? With an argument to 
> __init__? Any other way?
> 
> I prefer providing Interface lookup via other mechanism, like 
> with the __implements__ class attribute as done in Zope's 
> Interface module. But are there any advantages of interfaces 
> being done with inheritance?
> 

Put simply, either the semantics of one of I1.foo() or I2.foo() are a
subset of the other, or your design is broken regardless of how you
implement interfaces.  If the above condition holds, then simply
implement the stricter semantic and you're fine.  If it dosn't hold
then you have to redesign/refactor/rethink your class/interface design
'cause it's borken :(.

The interface proposals I've seen for python do provide a way around
this, but these don't follow the common definition of 'interface' you
are thinking of.  Rather they propose to provide a standardised means
of requesting an interface-adaptor from a given class to a given
contract.

OTOH I'm far less interested in this then in seeing generator
comprehensions and the new xmethods.... built in support for
lazy-evaluated lists, I personally can't think of too many things
cooler! :)

Andrae


From fredrik at pythonware.com  Mon Feb  4 13:36:33 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 04 Feb 2002 18:36:33 GMT
Subject: working with files with the same extenstion
References: <3C5EAF50.80004@rmb.co.za>
Message-ID: <RwA78.23199$l93.4217623@newsb.telia.net>

Ulrich Pfisterer wrote:
> Can anyone please tell me how I would go about doing the following.
> I would like to write a script that would take in a command line
> parameter "*.txt". The script would then loop through every file with a
> "txt" extension and something with it.

import glob
for file in glob.glob("*.txt"):
    print "do something with", file

</F>




From andda715 at student.liu.se  Fri Feb  1 14:40:51 2002
From: andda715 at student.liu.se (Anders Dahlberg)
Date: Fri, 1 Feb 2002 20:40:51 +0100
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012587303.30347.python-list@python.org>
Message-ID: <a3er63$7e3$1@news.island.liu.se>

<brueckd at tbye.com> skrev i meddelandet
news:mailman.1012587303.30347.python-list at python.org...
> On Fri, 1 Feb 2002, Anders Dahlberg wrote:
>
> > could re-run them without it. Maybe the low number of replies > is
> > because it's not suprising that a semi-compiled-to-native program is >
> > faster?
> >
> > well, but why would you want to compare interpreted java with python?
>
> Well for one, it'd be nice to be a little closer to comparing apples to
> apples (how fast a VM can interpret and execute the bytecodes).

Somewhat valid point (IMO that is)

>Secondly,
> I was very suprised that, with a JIT, Java was only meagerly faster than
> Python.

Well, python is written in c or something equivalent right?

>I was expecting an order of magnitude faster for most stuff,

object allocation (java need it alot)...

>so I
> was curious to know how slow Java was without the JIT. Finally, even
> though Java and Python are wildly different in many ways, knowing the
> interpreted vs. JIT'ed speeds might give a little insight into the
> potential benefits of a Python JIT.

I don't really know if python needs a jit, it would probably not harm
python, but is it really worth the effort?

>
> If I was really really interested in the Python-Java debate I'd go do the
> tests myself, but since I'm just sorta curious I was asking Bryan in the
> hopes that he still had the tests setup and it'd be easy to re-run them.
>
> > code during run-time? If you were to compare interpreted java with
python
> > should you then compare interpreted c++ with java (same ballpark)?
>
> Sure, why not?

well probably because you IMO should compare tools "as they are intended to
be used" (hope you understand what I'm getting at here).

> -Dave

/Anders




From chris.gonnerman at newcenturycomputers.net  Sat Feb  9 23:38:34 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sat, 9 Feb 2002 22:38:34 -0600
Subject: Help: I want a chat script!
References: <Xns91B06F8651AusYourStandingInIt@130.133.1.4> <Xns91AFAD9B27944RASXnewsDFE1@130.133.1.4> <Xns91B17A597AFBAusYourStandingInIt@130.133.1.4> <mailman.1013306296.2115.python-list@python.org> <Xns91B188F5332CCAusYourStandingInIt@130.133.1.4>
Message-ID: <000a01c1b1ec$cf346440$0101010a@local>

----- Original Message -----
From: "Flavian Hardcastle" <deathtospam43423 at altavista.com>


> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote in
> news:mailman.1013306296.2115.python-list at python.org:
>
> > Let's go over this step-by-step.  First, what web server did you
> > download? Most docs assume Apache.
> >
> It's called Lil' HTTP server by Summit Communications Networks
> http://www.summitcn.com/lilhttp/lildocs.html

Let me quote from their home page:

! CGI Directory ? This is the ONLY directory that Win32 CGI's can be run
! from!  CGI's placed in any other directory will be transferred to the
client
! rather than executed on the server.   Lil HTTP will only execute Win32
CGI's.
! These CGI's are actually Windows 32-bit programs with either the
! extension .CGI or .EXE.   PERL type CGI's cannot be executed on Lil' HTTP
at
! this time.

"PERL type CGI's" should be read "scripts".  You can't run Python, Perl,
Ruby,
Lua, etc. from this particular server.

Try Xitami.  It is quite compatible with Python.  http://xitami.com

> > You have a form (stored in a .html file or produced by a cgi script,
> > doesn't matter) which contains something like this:
> >
> >     <form method='POST' action='http://url.to.the/cgi-bin/script'>
> >
> > "script" must be a script file in the cgi-bin directory.  You probably
> > can shortcut it like this:
> >
> >     <form method='POST' action='/cgi-bin/script'>
> >
> > and this is probably better.
> >
> The actual text on my html form is
>
>     <FORM method="post" Action="script.cgi">
>
> I have tried changing the file the file extension to script.text,
> script.dat, and even as you seem to recommend above, just script. I press
> the SEND button on the form and a blank form comes up, with the filename
> "script.cgi".
>
> > The 11.2 docs then describe how you decode the data sent by the form.
> >
> This is the problem. I write data on the form, but no data turns up in the
> file(?) Isn't there supposed to be?
>
> > If you are still having problems, please post code (and tell us what
> > webserver software you are using).
> >
> Thanx. Here's the form (minus the head and title)...

<< form omitted >>

The form looks good, but where is script.cgi?





From rdacker at pacbell.net  Wed Feb  6 14:41:53 2002
From: rdacker at pacbell.net (rdack)
Date: 6 Feb 2002 11:41:53 -0800
Subject: [OSX] pthread problem
Message-ID: <644f6688.0202061141.25ac2ef2@posting.google.com>

osx 10.1.2, sendmail-8.12.2, milter-0.3.9, python2.1

when i try to run the sample.py milter i get:
dyld: python Undefined symbols:
_pthread_sigmask
_sigwait

i see postings saying pthread support is built-in to osx, but they say
support for those two symbols is missing. it looks like python itself
is referencing those symbols. i don't those symbols referenced
anywhere in the module code.

i don't think rebuilding with -lpthread will help. pthreads exist in
general, just not those two symbols. i don't know what to change to
get it to run.


From patois at calvix.org  Wed Feb 27 15:51:28 2002
From: patois at calvix.org (Yannick Patois)
Date: Wed, 27 Feb 2002 20:51:28 +0000 (UTC)
Subject: Calling baseclass method from a derivated class
References: <slrna7n2u6.f53.patois@ns.sicfa.org> <mailman.1014730722.26809.python-list@python.org>
Message-ID: <slrna7qhmg.v67.patois@ns.sicfa.org>

Le Tue, 26 Feb 2002 13:36:36 +0000, Andy Gimblett a ecrit :
>> I would like to know how I can call a method defined in a base class 
>> from a derivated class that overloaded it.
>http://www.python.org/doc/FAQ.html#4.23


Thanks for all people who kindly replyed to my question, and sorry
to have asked an FAQ.

	Yannick

-- 
 _/ Yannick Patois \___________________________________________________
| web :http://feelingsurfer.net/garp/ | CALVIX ! Le LUG de CAEN        |
| email : patois at calvix.org           | web : http://www.calvix.org/   |
| Petit Portail en Bois, Modeste et G?nial : http://www.rezo.net/      |


From cliechti at gmx.net  Wed Feb 20 17:41:15 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 20 Feb 2002 23:41:15 +0100
Subject: Is there a bug in CGI-POST
References: <8f8ffe67.0202201345.66fa13ed@posting.google.com>
Message-ID: <Xns91BBF185EEF0Dcliechtigmxnet@62.2.16.82>

sjoshi at ingr.com (Sunit Joshi) wrote in 
news:8f8ffe67.0202201345.66fa13ed at posting.google.com:
> Hello All
> I'm trying use the cgi module and I have seen that if I use the POST
> method in the form then I donot get any form fields. This however
> works with the GET method. Is this known, or am'i doing something
> wrong here. I'm using Xitami Server on Win2000

with POST the fields get transfered in sys.stdin and not trough 
environment variables.

-- 
Chris <cliechti at gmx.net>



From aahz at panix.com  Mon Feb  4 18:00:25 2002
From: aahz at panix.com (Aahz Maruch)
Date: 4 Feb 2002 15:00:25 -0800
Subject: Backwards Compatibility of Python versions
References: <mailman.1012857968.25728.python-list@python.org> <7x8za8dgle.fsf@ruckus.brouhaha.com>
Message-ID: <a3n3q9$n1i$1@panix2.panix.com>

In article <7x8za8dgle.fsf at ruckus.brouhaha.com>,
Paul Rubin  <phr-n2002a at nightsong.com> wrote:
>
>  a) If I put together a 1.5.2 bugfix release and submit it, will it go
>     up on Python.org with a link on the front page like the current
>     1.5.2 link?

In theory, yes.  In practice, you'll have document your work so
carefully that it probably won't be worth it.  I don't think anyone will
commit to making such a release available in the absence of existing
work to be submitted.  Have you read PEP 6 yet?
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From donn at u.washington.edu  Thu Feb 21 13:57:20 2002
From: donn at u.washington.edu (Donn Cave)
Date: 21 Feb 2002 18:57:20 GMT
Subject: why does popen2 silently ignore IOError?
References: <mailman.1014252874.28734.python-list@python.org>
Message-ID: <a53fug$24js$1@nntp6.u.washington.edu>

Quoth Graham Guttocks <graham_guttocks at yahoo.co.nz>:
| "Donn Cave" <donn at u.washington.edu> wrote:
| > in the general sense that it does not raise an exception in the case
| > of a non-zero exit from the child process.
|
| [...]
|
| > I personally prefer to get an exception, with value from stderr
| > output
|
| This was what I was looking for.  It sounds like it isn't possible.
|
| I thought what I wanted was simple.  I'm sending mail by opening a
| pipe to /usr/sbin/sendmail -- if that fails for whatever reason (e.g,
| sendmail program is missing) I want an exception raised with some
| indication of the problem, instead of the current silent failure.
|
| Is there any way to do this in Python?  Any example code to get me
| started?

It can be done in Python.
- Look at popen2.  Use the Popen3 class directly, and trap stderr.
- Write email to tochild file object.
- Read from the childerr object.  Your program will block until exit.
- Get exit status from the wait() member function.
- For extra credit, process as os.spawnv does to get the integer exit
  status value.
- If status is non-zero, raise exception with data from childerr.

But I didn't write that out as sample code, because there's a big
problem with it as an algorithm:  The read and write steps are in
principle concurrent.  How you solve this is up to you.  Maybe it
isn't even a problem - if you do it in the above order, you may
be delivered a SIGPIPE, or who knows depending on platform, but I'm
pretty sure it wouldn't just block forever.  If there might be input
on the fromchild object, you have another similar problem - if that
fills up because you're looking at childerr, you have a deadlock.

	Donn Cave, donn at u.washington.edu


From jw9 at cse.buffalo.edu  Sun Feb  3 21:49:59 2002
From: jw9 at cse.buffalo.edu (J. Wang)
Date: Sun, 3 Feb 2002 21:49:59 -0500
Subject: Problem while uploading a binary file using cgi module
Message-ID: <Pine.SOL.4.30.0202032135250.25346-100000@hadar.cse.Buffalo.EDU>

I test the script from
http://www.python.org/doc/essays/ppt/sd99east/sld058.htm.

My problem is that it works fine with a text file, but failed everytime
with a binary file such as jpeg file, even a very small (5k) binary file.
I tried read(), readline(), readlines(). It just doesn't work. Every time
the FieldStorage() only read a very small portion of the binary file and
then stop.

what have I done wrong? By the way, i need to use it in a webpage, so
ftplib module is not a alternative option.

Thanks!

J. Wang

the code is like this:
#!e:/Python22/python.exe
import cgi

print 'Content-type: text/html\n'
form = cgi.FieldStorage()
if not form:
	print """
	<form action="test.py" method="POST" enctype="multipart/form-data">
	<input type = "file" name = "filename">
	<input type ="submit">
	</form>
	"""

elif form.has_key("filename"):
	item = form["filename"]
	if item.file:
		while 1:
			data = item.file.readline()
			fp = file('1.jpg','wb')
			if data:
				fp.write(data)
			else:
				break
			fp.close()





From kkeller at speakeasy.net  Fri Feb  8 14:30:44 2002
From: kkeller at speakeasy.net (Keith Keller)
Date: Fri, 8 Feb 2002 11:30:44 -0800
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com>
Message-ID: <41914a.537.ln@wombat.san-francisco.ca.us>

In article <a417rn$ge8$1 at panix2.panix.com>,
	aahz at panix.com (Aahz Maruch) writes:
> I know some of y'all still keep roots to the Perl community, so I was
> wondering if there's any kind of definitive nickname for Perl
> programmers, the way we call ourselves Pythonistas?

I don't think so--having one nickname would contradict TMTOWTDI.  :)

--keith

-- 
kkeller at speakeasy.net
public key:  http://wombat.san-francisco.ca.us/kkeller/public_key
alt.os.linux.slackware FAQ:  http://wombat.san-francisco.ca.us/perl/fom



From emile at fenx.com  Wed Feb 13 12:25:31 2002
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Feb 2002 09:25:31 -0800
Subject: searching a list of tuples
References: <3c6aabeb.6302984@news.easynews.net>
Message-ID: <a4e7tr$1el2sm$1@ID-11957.news.dfncis.de>

"Marcus Stojek" <stojek at part-gmbh.de> wrote in message
news:3c6aabeb.6302984 at news.easynews.net...
> Hi,
>
> I have a (huge) list of tuples :
> list=(float 1,float 2,float 3,int).
>
> The list has been sorted with list.sort()
> Now I have to slice the list in a way
> that:
>
>   a <= float 1 <= b
>
> As I have to do this very often for different
> values of a and b and the list is quite large
> (100.000 tuples) I would like to do the slicing
> as fast as possible.
>
> Can anybody help me?
>

>>> import bisect
>>> bisect.bisect(range(1,1000,3),500)
167
>>>


--

Emile van Sebille
emile at fenx.com

---------



From martin at v.loewis.de  Tue Feb 12 19:42:28 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 13 Feb 2002 01:42:28 +0100
Subject: Major league corruption - anyone seen anything like this?
References: <mailman.1013557302.9749.python-list@python.org>
Message-ID: <m3pu3aw6tn.fsf@mira.informatik.hu-berlin.de>

Skip Montanaro <skip at pobox.com> writes:

> Obviously I have some serious problems in what was supposed to be a simple
> wrapper (perhaps the problems are with the underlying library which has only
> been lightly tested), but I've never seen a segfault take files with it like
> this.  Has anybody else encountered this before?

I sure did; it turned out to be a memory error. Replacing the faulty
chip brought everything to normal again.

Regards,
Martin



From rdacker at pacbell.net  Sun Feb  3 20:15:11 2002
From: rdacker at pacbell.net (bob ackerman)
Date: Sun, 03 Feb 2002 17:15:11 -0800
Subject: [OSX] milter won't build
In-Reply-To:  <Pine.OSX.4.43.0202031826200.4385-100000@d-128-61-180.bootp.virginia.edu>
Message-ID: <A2D50AF4-190C-11D6-994D-003065428126@pacbell.net>

On Sunday, February 3, 2002, at 03:43  PM, Steven Majewski wrote:

>
>
> On 3 Feb 2002, rdack wrote:
>
>> trying to build milter module 0.3.9 on OSX 10.1.2 i get this error:
>> cc -bundle -undefined suppress build/temp.darwin-5.2-Power
>> Macintosh-2.1/miltermodule.o -lmilter -lsm -o
>> build/lib.darwin-5.2-Power Macintosh-2.1/milter.so
>> /usr/bin/ld: -undefined error must be used when -twolevel_namespace is
>> in effect
>> error: command 'cc' failed with exit status 1
>>
>> is it telling me i need  an '-undefined error' option? or that it
>> doesn't like the 'undefined suppress' option?
>
> I don't know what the milter module is, but:
>
> OSX gcc/ld doesn't like that combination. You can use either of:
> 	-twolevel_namespace -undefined error
> 	-flat_namespace -undefined suppress
> But I think the defaults are:
> 	-undefined error
> 	-twolevel_namespace
> So, you get generate that error with only one of those switches
> with the wrong default for the other.
>
> Python modules on OSX is usually built with:
> 	-bundle -flat_namespace -undefined suppress
>
>
>> where is the 'cc -bundle' command set up? can i edit it?
>> is it possible that it finished and the module is ready to go?
>
> Is milter using distutils ? i.e. is it using a 'setup.py' file to
> build and install, or does it use a Makefile ?
>
> What command did you use to build it?
>
> (Distutils should get the build command right, but if you have more
> than one version of Python installed, you need to make sure that
> you're running the version that you're going to link to. )
>

i used setup.py. so how do i change the default or override it? i am 
pretty sure only python2.1 is installed.

>> where would the milter module end up after install?
>> what will its name be?
>> are python modules '.pyo' files? do they end up in the
>> /usr/local/lib/python2.1 directory?
>
> If it's a dynamic shared lib, it'll usually be an .so file, and it
> should usually install in the site-packages directory, likely in
> /usr/local/lib/python[v.n]/ . ( It might also go in plat-darwin*/
> or somewhere else in that tree. )
>

i see the _pg.so module i installed there, so that's fine. and i believe 
there is no milter there since 'setup.py build' had that error.

> Are you building 2.1.2 ? or just 2.1 ?
> You probably ought to stick to 2.1.2 or 2.2 on OSX.
>

i guess it has been awhile since i installed the latest version of python.
  i'll try 2.2.

Thanks for the reply.




From michael at stroeder.com  Fri Feb 22 19:13:38 2002
From: michael at stroeder.com (Michael =?ISO-8859-1?Q?Str=F6der?=)
Date: Sat, 23 Feb 2002 01:13:38 +0100
Subject: hep needed with pipes (popen4)
References: <iUyd8.18492$X2.205012@nnrp1.uunet.ca>
Message-ID: <3C76DEB2.4050809@stroeder.com>

Darcy Kahle wrote:
> (control-c kills the script, by default).  I tried sending "\x003", which to
> the best of my knowledge, is the hex code for it, but it didn't work.

"\x003" are two chars. A null-byte followed by an
ASCII char for "3".

You probably meant "\003" or "\x03".

Ciao, Michael.



From theller at python.net  Tue Feb 26 15:47:21 2002
From: theller at python.net (Thomas Heller)
Date: Tue, 26 Feb 2002 21:47:21 +0100
Subject: subclassing float (or not)
References: <mailman.1014672051.7833.python-list@python.org>
Message-ID: <a5gsa4$7b5gj$1@ID-59885.news.dfncis.de>

"Steven Majewski" <sdm7g at virginia.edu> wrote in message news:mailman.1014672051.7833.python-list at python.org...
>
> On Fri, 22 Feb 2002, Sean 'Shaleh' Perry wrote:
>
> >
> > On 22-Feb-2002 Thomas Heller wrote:
> > > I'm currently porting my own rewrite of Konrad Hinsen's PhysicalValue
> > > module to Python 2.2.
> > > Basically a PhysicalValue instance is a float conbined with a unit.
> > > Should I subclass float or should I not? What would I gain by doing
> > > this (except maybe some memory savings when I use __slots__)?
> > >
> >
> > Generally when you subclass you are saying "I am a special type of foo" where
> > foo is your parent.  I think in this instance it would make more sense to make
> > a new class which contains a float and a unit.  Give it methods so it acts how
> > you expect.
> >
>
> Except that how he expects it to act is exactly like a float except for
> carrying and propagating a little extra bit of state at each operation.
>
> Sounds like a good application for subclassing floats to me!
That's also what I thought first.
>
> You can do it (and could have always done it) with a container class,
That's what I was doing until now.
> but there are always holes where you need the operation to be
> coerced on the classes value and not the class itself. In some
> cases that still may be what you really want (because you also want
> to coerce the result to my_float rather than float), but at least
> if you miss something, a subclassed float will do something closer
> to the right thing.
>
> An example of what I mean is 'sqrt(MyFloat)' -- if MyFloat is a
> class that contains a float and has methods for __add__, __mul__,etc.
> sqrt(MyFloat) will raise an exception if passed a container class --
> you need to call 'sqrt(MyFloat.value)' , while:
>
> >>> class F(float):
> ...     pass
> ...
> >>> from math import *
> >>> sqrt( F(1.0))
> 1.0
>
> Will return a normal float where you may really want a custom F instance.
> ( Which is the better way to go wrong here may vary. )
>
> Two other float-like classes that I've had use for that are easier to
> do now with float subclasses:
>
> Values that carry and propagate standard errors or standard deviations
> with them.
>
> Values that carry an "audit trail" ( for taking chained ratios of
> experimental measures and trying to track down the inconsistent
> data. )

It seems you always end up with implementing something like

class F(float):
  # float subclass
  def __mul__(self, other):
    return F(float(self) * other, ...)

or

class F:
  # container class with a float instance variable
  def __mul__(self, other):
    return F(self.float * other, ...)

The only *real* difference seems that you can pass float
subclass instances to round(), log(), and these kind of functions.
As you already noted, normally they return the wrong results ;-(

Thomas




From tim-dated-1013908624.48bcde at catseye.net  Sat Feb  9 20:17:03 2002
From: tim-dated-1013908624.48bcde at catseye.net (Tim Legant)
Date: 09 Feb 2002 19:17:03 -0600
Subject: A newbie question about class
In-Reply-To: <a50631f6.0202091635.1dc42d4b@posting.google.com>
References: <a50631f6.0202091635.1dc42d4b@posting.google.com>
Message-ID: <86heoqb0fk.fsf@skitty.catseye.net>

newgene at bigfoot.com (newgene) writes:

> Hi, group,
>      I have a newbie question about building a class:
> 
> eg.
> 
> class A:
>    def __init__(self,x=None):
>        self.x=x
>        if self.x != None:
>             self.y=x**2
>        else:
>             self.y=x**2
> 
> If I initialize A like "a=A(2)", then I can get "a.y=4" automatically;
> but if I initialize A like "a=A()", then when I give "a.x=2" later, I
> can not get the value of "a.y" like the former. Of course I can use a
> method to set the value of "a.y", but is there a machanism to get the
> "a.y" automatically whenever I change the value of "a.x"?

A comment and a suggested answer....  First, your else does exactly
the same thing as your if.  Unfortunately it will blow up, since you
can't raise None to any power!  I will assume it's a cut-and-paste-o.

Instead of twiddling your class's bits from outside the class

a.x = 2

consider creating a set_x function:

class A:
    def __init__(self, x=None):
        self.set_x(x)

    def set_x(self, x)
        self.x = x
        if self.x:
            self.y = x**2

Now a = A(2) still sets a.x to 2 and a.y to 4.  a = A() sets a.x to
None and doesn't set a.y.  Finally, from outside the class you just
call a.set_x(3).  a.x will be set to 3 and a.y will be set to 9.


Tim



From bellman at lysator.liu.se  Sat Feb  2 11:20:19 2002
From: bellman at lysator.liu.se (Thomas Bellman)
Date: Sat, 2 Feb 2002 16:20:19 +0000 (UTC)
Subject: Equivalent of Perl chomp?
References: <mailman.1012403361.2370.python-list@python.org>  <u5i8b8jtee2r13@corp.supernews.com> <7xbsfag023.fsf@ruckus.brouhaha.com>  <3C59D2EE.3937ABBB@ndh.net> <a3f7dh$lq9$1@news.island.liu.se> <isF68.111010$8b.4218974@atlpnn01.usenetserver.com>
Message-ID: <a3h3k3$mo4$1@news.island.liu.se>

"Steve Holden" <sholden at holdenweb.com> wrote:

> Perhaps I'm just lazy, but why not process

>     for line in file(filename).read().split("\n"):

> under such circumstances?

Because

a) I don't like reading entire files into memory before processing,
   when not needed.  I'm not very fond of wasting memory.  Too many
   programs do that already, and I don't feel the need to contribute
   to that.

b) It is more complicated to read and understand.  'for line in
   fp.xreadlines()' is immediately understandable, while your
   variant takes a couple of seconds to understand the full
   implications of.


-- 
Thomas Bellman,   Lysator Computer Club,   Link?ping University,  Sweden
"Don't tell me I'm burning the candle at both ! bellman @ lysator.liu.se
 ends -- tell me where to get more wax!!"     ! Make Love -- Nicht Wahr!


From skip at pobox.com  Wed Feb 13 08:34:45 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 13 Feb 2002 07:34:45 -0600
Subject: photo id ?!?!
In-Reply-To: <a4daoi$np9$1@slb4.atl.mindspring.net>
References: <000c01c1b416$b85c9540$445d4540@Dell2>
        <mailman.1013563639.20235.python-list@python.org>
        <a4daoi$np9$1@slb4.atl.mindspring.net>
Message-ID: <15466.27509.492661.103010@12-248-41-177.client.attbi.com>

    Andrew> ... at the O'Reilly Bioinformatics conference ...  I didn't need
    Andrew> to present ID ....  During registration we talked about those
    Andrew> emails, which gave us another way to verify identity much more
    Andrew> than a piece of plastic or paper could.  I was also at the
    Andrew> biohackathon beforehand (where I knew almost everyone), which
    Andrew> provided yet another way to verify identity.

While I'm sure there were much less than the standard "six degrees of
separation" between any one of us and all the attendees at the conference,
that doesn't make it a reasonable way to confirm identity at registration.
I'm not trying to justify Foretec's use of photos to identify people, just
suggesting that web of trust may not be the best replacement.

I wonder if it had something to do with all those generals in the next room?

Skip




From clpy at snakefarm.org  Tue Feb  5 08:56:30 2002
From: clpy at snakefarm.org (Carsten Gaebler)
Date: Tue, 05 Feb 2002 14:56:30 +0100
Subject: python and xmlrpc
References: <3C5F9F42.9040508@wanadoo.fr>
Message-ID: <3C5FE48E.E692D00F@snakefarm.org>

kondor wrote:

> I'm looking for a document explained how to write xmlrpc servers and clients.
> Can you help me ?

I found this article extremely useful:

http://www.onlamp.com/pub/a/python/2001/01/17/xmlrpcserver.html

You might also want to take a look at the related articles at the
bottom of the page.

cg.
-- 
The looneys are out there - join SETI at home.


From jkraska at san.rr.com  Sun Feb  3 19:54:25 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 04 Feb 2002 00:54:25 GMT
Subject: Xython - XML-Formed Python
References: <1012751401.4431.13.camel@acropolis> <mailman.1012780267.21706.python-list@python.org>
Message-ID: <ocnr5u4bo7rmcpeg869kl73vae3u6jo4rp@4ax.com>

>There appear to be a number of ways to send Python code across a
>network connection.

Indeed. I did mobile code using Pyro as a simple exercise something
like a year ago.

C//



From jkraska at san.rr.com  Sat Feb 16 12:36:18 2002
From: jkraska at san.rr.com (Courageous)
Date: Sat, 16 Feb 2002 17:36:18 GMT
Subject: Fear of autocoding!
References: <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <slrna6gc9p.v8s.philh@comuno.freeserve.co.uk> <1476.811T2040T12515414threeseas@earthlink.net> <slrna6rk5j.2pn.quinn@cruzeiro.ugcs.caltech.edu> <3422.812T515T6683481threeseas@earthlink.net>
Message-ID: <3l6t6us5gpgkklelav01q0tlu6be8a7c9n@4ax.com>

>You see, I and the majority of people in this world do not preform the
>act of software programming 40 plus hours a week.

This much is obvious.

C//



From jp at NOSPAMdemonseed.net  Fri Feb  8 14:13:07 2002
From: jp at NOSPAMdemonseed.net (jason petrone)
Date: Fri, 8 Feb 2002 19:13:07 +0000
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org>
Message-ID: <30814a.l3p.ln@demonseed.net>

Skip Montanaro <skip at pobox.com> wrote:
> What non-MS tools are out there
> (open source or otherwise) for reading, displaying or converting PPT slides
> to other, more accessible, formats?  

StarOffice: http://www.sun.com/software/star/staroffice/
OpenOffice: http://www.openoffice.org/
Kpresenter: http://www.koffice.org/kpresenter/
Achtung(though not yet): http://www.gnome.org/gnome-office/achtung.shtml

I wish I knew of one that did conversions to PostScript, PDF, or SVG.

jlp


From f91-men at nada.kth.se  Thu Feb 28 08:34:07 2002
From: f91-men at nada.kth.se (Mattias Engdegård)
Date: 28 Feb 02 13:34:07 GMT
Subject: full closures
References: <slrna7q9m0.4pb.msoulier@pmerd071.ca.nortel.com> <a5jbra$7ok00$1@ID-11957.news.dfncis.de> <m3ofiair04.fsf@mira.informatik.hu-berlin.de>
Message-ID: <f91-men.1014903247@nada.kth.se>

martin at v.loewis.de (Martin v. Loewis) writes:

>As Zenin points out, Python uses objects to emulate closures, whereas
>Lisp uses closures to emulate objects.

An elegant and symmetric statement, but not true. Lisp has both
classes and closures natively. Neither is necessarily "emulated" in
terms of the other.

Python only has read-only closures, so you either have to use classes
or simulate mutable bindings by using a mutable object as middle-man
instead (typically an array of length 1).

This is not a statement of the relative merits of Python and Lisp;
they have different purposes and differ more than most people think
after reading Norvig's article.



From fredrik at pythonware.com  Mon Feb  4 10:56:17 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 04 Feb 2002 15:56:17 GMT
Subject: module from BLOB
References: <a3lk3g$fcs$1@news.tpi.pl>
Message-ID: <Bay78.23174$l93.4207876@newsb.telia.net>

marcin andrzejewski wrote:
> is it possible to import module directly from database BLOB/CLOB field ?

something like this could work:

import imp, marshal
from my_database_api import fetch_blob_from_database

def import_from_blob(name):
    # fetch code object
    code = marshal.loads(fetch_blob_from_database(name))
    # create a new module object
    module = imp.new_module(name)
    # execute the code in the module's namespace
    exec code in vars(module)
    return module

mymod = import_from_blob("mymod")

to create the blobs, do something like:

    module = "mymod"
    text = open(module + ".py").read()
    code = compile(text, module + ".py", "exec")
    data = marshal.dumps(code)
    store_blob_in_database(module, data)

for details, see the library reference.

hope this helps!

</F>

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




From dan at danshafer.com  Sun Feb 17 01:57:26 2002
From: dan at danshafer.com (Dan Shafer)
Date: 16 Feb 2002 22:57:26 -0800
Subject: Why Python is like BASIC ...  java "vs" python
References: <mailman.1013881166.27493.python-list@python.org> <a4mhvt$ga8$1@panix2.panix.com>
Message-ID: <39abba5c.0202162257.2e9f8379@posting.google.com>

The original poster's point is no less valid because of the very
accurate statements made here about the enormous differences between
Python and BASIC. (My gosh, I'd never go back to Basic after using
Python for even a little while!)

But Python does have some of the features that make Basic a great
beginning programming language. In fact, I think it's a much *better*
language for first-timers than Basic because the syntax is much, much
cleaner and more consistent and because the built-in stuff is so
darned powerful and easily accessible.


From lac at strakt.com  Thu Feb 14 12:56:03 2002
From: lac at strakt.com (Laura Creighton)
Date: Thu, 14 Feb 2002 18:56:03 +0100
Subject: Color database in Tkinter 
In-Reply-To: Message from Lawrence Oluyede <rhymes@myself.com> 
   of "Thu, 14 Feb 2002 17:47:56 +0100." <veqn6ucitupfrt3jaujtd7rhrq0sl39r2g@4ax.com> 
References: <veqn6ucitupfrt3jaujtd7rhrq0sl39r2g@4ax.com> 
Message-ID: <200202141756.g1EHu3Go003334@ratthing-b246.strakt.com>

> 
> Where i could find the whole database for color names
> built in Tkinter? I need a list of colors
> 
> -- 
> Lawrence "Rhymes" Oluyede (rhymes at myself.com)
> "Per cambiare, per diventare un'altra
> cosa dobbiamo prima sapere cosa siamo"
> (Bruce Lee)

Get out pynche in the Tools directory.

Laura Creighton



From buzzard at urubu.freeserve.co.uk  Sat Feb  9 14:04:01 2002
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Sat, 9 Feb 2002 19:04:01 -0000
Subject: Animated graphs / wxPython
Message-ID: <a43rur$31f$1@newsg1.svr.pol.co.uk>

I need to frequently redraw a graph (ie. nodes and edges) to a wxPython
panel.  ie. I remove and/or add edges to the graph and then redraw it.  I am
currently using wxPython ellipses and lines for drawing the nodes and edges.
The problems are the speed of redrawing and the flickering as the window is
redrawn.  SetClippingRegion() doesn't help a great deal as the clipping
region is usually nearly as large as the full graph.  I am currently using a
workaround whereby I draw edges in the background colour in order to remove
them.  But after a number of redraws the graph starts to get noticeably
whited out by drawing over existing nodes and edges.

Does anyone know of a better approach?  Ideally I would like to be able to
simply remove and add edges and send them to the back of the nodes rather
than redrawing the relevant nodes each time I add an edge.  I have looked at
a number of packages, but none seem to be able to do what I want.
Essentially I want an animated graph that will show a user what the graph
algorithms are up to.  (Gato won't do the job.)  Thanks in advance.

Duncan




From pedro_rodriguez at club-internet.fr  Sun Feb 10 09:23:57 2002
From: pedro_rodriguez at club-internet.fr (Pedro Rodriguez)
Date: Sun, 10 Feb 2002 15:23:57 +0100
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net>
Message-ID: <pan.2002.02.10.15.23.55.53798.1174@club-internet.fr>

"Ron Stephens" <rdsteph at earthlink.net> wrote:

> So this morning at 4:30 AM I sat up and ran a "top" process from the
> Bash shell, that shows all
> running processes. Sure enough, at exactly 4:30 AM, a new Python process
> was spawned. The new
> Python process was matched with a new entry in the cron log file of the
> mygale.py program being
> called.
> 
> 
Seems we have another 'Courageous' here ;) At least, it proves that your
cron setup is OK.

> I watched carefully, and the mygale.py process lasted for about 30
> seconds and then disappeared from
> the "top" screen. I assume it terminated normally. Except that when I
> run the mygale.py get all script
> manually  in the BASH shell, it takes about two minutes to complete.
> This is the only sign of any
> problem, but still no error messages are generated anywhere I can find.
> 
> 
Call me stupid, but if I had to do this on my machine, I will need to
startup my dialup connexion first, and ensure that the connection is ok
prior starting anything. But maybe you have a permanent connection.

Some hints :
- did the same script runs fine when runned manualy at the same hour ?
  (thinking that the sites you access may be down for backup for example)
- do you set up something special when you login manualy (like
  environment variables required for your scripts) ? If you set up
  variables in .bash_profile they won't be available to your scripts.

> By the way, shouldn't it be possible to cut and paste from the BASH
> shell to a text editor? It would save me a lot of typing ;-))))
> 
> The clipboard in general is a frustration; I can't yet make it work
> properly in many situations. Is it me or does the KDE clipboard lack a
> lot of functionality compared to Windows clipboard? I can't even copy
> and paste from K-EDIT, nor K-WORD, to Netscape Composer. Oh, well.
> 
> 
Works fine to me. But, I did notice the following thing : 
- run netscape composer
- run kedit
- type something in kedit
- select it and copy it (Menu:Edit/Copy) 
- go back in netscape composer
- paste is not available in Edit menu... 
- ... but middle button click works fine (X11 Paste)

> One last frustration: Linux does hang, at least to the point that this
> newbie can't recover from it. My
> box hung when I tried to open a web presentation by Guido van Rossum
> about Python Conference 10
> using K Presenter.  I could not even open a console nor close the
> offending KDE shell to get back to a
> command line. The box also hangs when I try to print using cups on my HP
> T45 printer. (I wish I could
> use my printer.) My downloaded Open Office also hangs my computer just
> by trying to open it.
> 
Same problem with kpresenter. The system start swapping and its very
difficult (well almost impossible) to take control back. Even an old
powerpoint viewer runned with wine wasn't able to open it.

Regards,
-- 

Pedro


From sarnold at arnolds.dhs.org  Tue Feb  5 22:30:51 2002
From: sarnold at arnolds.dhs.org (Stephen L Arnold)
Date: Tue, 5 Feb 2002 19:30:51 -0800
Subject: To all Higher Level....A True Beginner
In-Reply-To: <ba59f11e.0202051821.5c41221e@posting.google.com>
Message-ID: <20020206033030.9483B1FB00@shiva.arnolds.bogus>

On 5 Feb 02, at 18:21, David Pham wrote:

> Hey! Everyone!, I like to know if any of you programmres
> recommend any books, online tutourial for myself, or anyone else
> that is a beginner. I am a beginner programmre. Chose Python
> cause heard it can do a lot of things. But never have programme
> before. So I home you guys can put out some help. The only
> language I have tried out is probably Qbasic and VBdos thats all,
> but nothing came out. --

Not surprising...  But this topic rightly belongs on the python-
tutor list, rather than the regular python list.

However, if you head to the main python.org site and go to the 
Newbies documentation page, there are links to several good 
tutorials and even freely downloadable books.  I'd recommend 
starting with the book "How to Think Like a Computer Scientist" in 
Python.  There are several other sources of example code, etc, but 
I can't think of a better way to start (from where you're at) than 
with the above book.  You were made for each other...

Steve

*************************************************************
Steve Arnold                      http://arnolds.dhs.org:8080

Java is for staying up late while you program in Python...



From robin at jessikat.fsnet.co.uk  Thu Feb 28 12:06:05 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 28 Feb 2002 17:06:05 +0000
Subject: struct_time type testing 2.1-2.2
Message-ID: <VCtEHjA9Nmf8Ewzm@jessikat.demon.co.uk>

Jeff Bauer kindly tells me that Python 2.2 has a different return type
for time.localtime(...). 

In his module he had a test on an argument type of the form

    _SeqTypes = (ListType,TupleType)


    tn = type(arg)
    if tn in _SeqTypes:
      self.normalDate = int("%04d%02d%02d" % normalDate[:3])

and in 2.2 we seem to need 

    if tn in _SeqTypes or tn is time.struct_time:

is there a cleverer way than

    if tn in _SeqTypes or tn is getatr(time,'struct_time',None):

to get a cross version test?

Presumably we could extend _SeqTypes as that needs to be done only once,
but ideally a simple check like

        arg.supports_index(3)

would be more informative than defining these rather arbitrary named
type lists.
-- 
Robin Becker


From aahz at panix.com  Fri Feb  8 00:01:40 2002
From: aahz at panix.com (Aahz Maruch)
Date: 7 Feb 2002 21:01:40 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver <3C6045CC.A0A64191@optonline.net>
Message-ID: <a3vm3k$rse$1@panix2.panix.com>

In article <3C6045CC.A0A64191 at optonline.net>,
Frederick H. Bartlett <fbartlet at optonline.net> wrote:
>
>While I am mostly right-handed, I often use the left-hand settings on
>the mouse (so as to spread the load -- and annoy folks who try to borrow
>my computer). I occasionally pick up a pen to write with my left hand
>before I remember that I can't write with that hand.

Heh.  My preferred working style is to use two computers, one with each
hand when I'm not doing anything that requires actual typing.  And, of
course, I use one mouse in each hand.
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From rasmussn at lanl.gov  Tue Feb 26 13:33:34 2002
From: rasmussn at lanl.gov (rasmussn at lanl.gov)
Date: Tue, 26 Feb 2002 11:33:34 -0700
Subject: Scripting C++ -- Boost.Python vs CORBA vs ???
In-Reply-To: <3C7A751D.1080703@physics.ucsb.edu>
Message-ID: <5729323C-2AE7-11D6-A367-0030657EBC3E@lanl.gov>

The latest Siloon version (3.0beta) passes it's C++ test suite using
gcc-3.0.2.  If anyone is interested in being a beta tester, let me
know and I can provide access to the latest sources.

Alternatively, we are planning on putting the project out on sourceforge
(with a BSD-style license).  Maybe now is the time.  Has anyone
had experience doing this?  How big a process is it?

Craig


On Monday, February 25, 2002, at 10:32 AM, Craig Maloney wrote:

> Since your site specifically lists g++ 2.95 and NOT 3.x, am I to infer
> that there will be problems with the new ABI?  -- Or any problems that 
> will, in practice or principle, cause conflicts with g++ 3.0?
>
> *Holds breath... and crosses fingers...*




From aahz at panix.com  Fri Feb  1 16:51:11 2002
From: aahz at panix.com (Aahz Maruch)
Date: 1 Feb 2002 13:51:11 -0800
Subject: reading sendmail alaises
References: <m34rl2jh7m.fsf@moroni.pp.asu.edu> <864rl2p09z.fsf@k0dez.scriptkiddie.net> <m31yg4diol.fsf@moroni.pp.asu.edu>
Message-ID: <a3f2kf$ohm$1@panix2.panix.com>

In article <m31yg4diol.fsf at moroni.pp.asu.edu>,
David Bear  <iddwb at moroni.pp.asu.edu> wrote:
>
>what is the 'praliases' program? 

man praliases
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"The more you drive, the less intelligent you are."  --_Repo Man_


From jeff at ccvcorp.com  Fri Feb  8 15:23:28 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Fri, 08 Feb 2002 12:23:28 -0800
Subject: ActivePython and wxPython
References: <RiH88.37389$Jq.1973082@news2.calgary.shaw.ca> <a3vl00014vc@enews3.newsguy.com>
Message-ID: <3C6433C0.F1BE020F@ccvcorp.com>

Greg Krohn wrote:

> "Andrew" <anon at anon.com> wrote in message
> news:RiH88.37389$Jq.1973082 at news2.calgary.shaw.ca...
> > Hello,
> >
> > Sorry if this question is inappropriate for this newsgroup, but I was
> > wondering if there are known problems using wxPython with ActivePython?  I
> > try to run the first example (from www.wxpython.org) in the wxPython
> > tutorial but get a crash when I attempt to create a wxFrame object.  I
> > stepped into the code and it's failing on an 'apply' call.
>
> Yeah, that happens. If you try it from the command line you should be fine
> (i.e. 'c:\>python myscript.py'). I _think_ this has something to do with
> Pythonwin's GUI getting in a fight with wxPython's GUI. Again, I _think_
> they both use MFC.

It is indeed a problem with the GUIs not cooperating.  This is a known problem
with running two different GUIs in the same process--just doesn't work too
well.  (The same applies to IDLE as well.)  The solution is to either use an IDE
that's capable of running scripts out-of-process (I think IDLE-fork does, and
probably most/all of the commercial Python IDEs), or else editing in the IDE and
then running from a DOS box (I do this all the time).

Jeff Shannon
Technician/Programmer
Credit International




From andersaa at stud.cs.uit.no  Wed Feb 27 03:06:25 2002
From: andersaa at stud.cs.uit.no (Anders Åberg)
Date: Wed, 27 Feb 2002 09:06:25 +0100
Subject: Thread support in NetBSD
Message-ID: <a5i40k$1k59$1@news.uit.no>

I have problems getting the thread module to work in Python on NetBSD.
I was told that there were some problems with this on NetBSD, but that it
was possible to get it to work with some effort, by using a pthread library
called mit-pthread. So I installed this and tried to recompile Python with
the following parameter to the configure script:

./configure --with-threads=/usr/pkg/pthread

This parameter is explained like this in the help page:

--with(out)-threads[=DIRECTORY] disable/enable thread support"

In the pthread directory, there are three subdirectories: bin, include and
lib. Should I have passed one of these as parameter to configure, instead
of the parent directory?

Anyway, it doesn't work and I am stuck now. Does anybody know how
to fix this?

Any help appreciated!


Anders.




From steve at ferg.org  Wed Feb 13 12:50:00 2002
From: steve at ferg.org (Stephen Ferg)
Date: 13 Feb 2002 09:50:00 -0800
Subject: Hidden files, selecting directories, & taxes
References: <cc14131d.0202121623.47ffaa99@posting.google.com>
Message-ID: <b16e4ef7.0202130950.67f4004d@posting.google.com>

> Is there a simple and easy way to allow the end-user to select a
> directory?  

I'm working on a package that I'd like to polish some more, called
easygui.  Go to www.ferg.org/easygui.  The function you want is called
dirbox().

If you try it, let me know how you get on... :-)
-- Steve Ferg


From jkraska at san.rr.com  Thu Feb 21 21:30:58 2002
From: jkraska at san.rr.com (Courageous)
Date: Fri, 22 Feb 2002 02:30:58 GMT
Subject: Unintelligible Lambda Tricks
References: <3c74c211.20012718@news.isar.de> <mailman.1014338076.13501.python-list@python.org>
Message-ID: <qnbb7u0hoqpi5s7vg8hg9k9fh0ufijvts3@4ax.com>

>Basically, I have no freakin' clue how python (or any other language) runs 
>through that lambda. I guess I better go read...

Really? I was thinking of the person who wrote it.

I vote for getting some of your homies and going midieval on him.

>:-\

C//



From justin at iago.org  Mon Feb  4 22:57:21 2002
From: justin at iago.org (Justin Sheehy)
Date: Mon, 04 Feb 2002 22:57:21 -0500
Subject: CPython vs. Jython for Web applications?
In-Reply-To: <a3n3a9$km7$1@panix2.panix.com> (aahz@panix.com's message of "4
 Feb 2002 14:51:53 -0800")
References: <c775c8db.0202041126.5177e606@posting.google.com>
	<a3n3a9$km7$1@panix2.panix.com>
Message-ID: <x7pu3kzila.fsf@ra.iago.org>

aahz at panix.com (Aahz Maruch) writes:

>> - app must run on both Win32 and some free Unix, preferably FreeBSD
>
> The big problem I see is that you're likely to want threading, and my
> understanding is that threading doesn't work too well on *BSD.

Far better than on Linux, in my experience.

(Though not as good as some of the non-free unices)

-Justin

 




From logiplexsoftware at earthlink.net  Wed Feb 27 19:14:11 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Wed, 27 Feb 2002 16:14:11 -0800
Subject: Oddity question
In-Reply-To: <7934d084.0202271600.499b2b74@posting.google.com>
References: <mailman.1014745204.15257.python-list@python.org>
	<7934d084.0202262258.53fbe36@posting.google.com>
	<7934d084.0202271600.499b2b74@posting.google.com>
Message-ID: <20020227161411.62a227d1.logiplexsoftware@earthlink.net>

On 27 Feb 2002 16:00:43 -0800
Andrae Muys wrote:

> amuys at shortech.com.au (Andrae Muys) wrote in message
news:<7934d084.0202262258.53fbe36 at posting.google.com>...
> > <aside>
> <SNIP>
> > </aside>
> 
> Cliff has suggested in a very polite private email that this aside was
> a cheap shot.  On reflection I agree with him, so wish to offer a
> public appology for it.  While it may have been appropriate in a FTF
> discussion, such a comment divorced of all non-verbal cues is
> inappropriate, in bad grace and showed poor judgement on my part.
> 
> Sorry Cliff.  I will do better next time.

Aw, hey.  I was joking.  I didn't really take offense (hence the ;) after
my exclamation of "cheap shot").  Please don't feel the need to be too
politically correct.  I emailed you offlist with my arguments on the topic
because I didn't want to offend /you/, and maybe that's where the miscue
arose from.

BTW, I very much respect your opinion on the matter under discussion, it's
simply a polite disagreement.

Regards,

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From jim.vickroy at noaa.gov  Tue Feb 26 16:43:22 2002
From: jim.vickroy at noaa.gov (jim.vickroy)
Date: Tue, 26 Feb 2002 14:43:22 -0700
Subject: shelve problem
Message-ID: <3C7C0179.375E620C@noaa.gov>

I'm experiencing a problem with obtaining the list of keys in a
previously-created, shelve archive.  Note that I can ask if the archive
contains a particular key, but an exception is raised when I ask for the
list of all keys.

>>> import shelve
>>> db = shelve.open('d:/logs/products_journal')
>>> db.keys()
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
  File "C:\Python22\lib\shelve.py", line 56, in keys
    return self.dict.keys()
error: (22, 'Invalid argument')
>>> db.has_key('XPDIAG_20010514_172119758_AA_12.FTS')
1

I'm using Python 2.2 on a Win 2k platform.  Any ideas would be greatly
appreciated.



From phr-n2002a at nightsong.com  Tue Feb 12 08:19:31 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 12 Feb 2002 05:19:31 -0800
Subject: Defending the Python lanuage...
References: <b34d3835.0201300026.28e51376@posting.google.com> <20020130073810.382$1b@news.newsreader.com> <b34d3835.0201310005.548ee6be@posting.google.com> <mailman.1012498757.26825.python-list@python.org> <a3cf7s$se42@news1.gtech.com> <mailman.1012522381.16207.python-list@python.org> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <3C688CED.4583E7DA@cosc.canterbury.ac.nz>
Message-ID: <7xeljqq1lo.fsf@ruckus.brouhaha.com>

Greg Ewing <greg at cosc.canterbury.ac.nz> writes:
> I'm not sure that coarse/fine is the right way to describe
> the distinction even in normal unidextrous (is that the
> right word?) people. Consider the fact that most right-handed
> guitar players use the left hand for fingering, and the
> right hand for the relatively dextrously undemanding
> function of strumming. Why is that?

You have to precisely control the timing of hitting the strings with
the right hand in order to not mess up the music.  Since the left
hand's actions don't actually make sound, the timing requirements
aren't so precise as long as stuff happens fast enough.


From brueckd at tbye.com  Fri Feb  1 10:38:38 2002
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Fri, 1 Feb 2002 07:38:38 -0800 (PST)
Subject: Java vs Python Benchmarks: Java is faster
In-Reply-To: <C0n68.1465$8u4.318040@wagner.videotron.net>
Message-ID: <Pine.LNX.4.33.0202010733170.21044-100000@ns0.tbye.com>

On Thu, 31 Jan 2002, Bryan wrote:

> > I'm still not willing to give java the benefit of the doubt yet.
>
> Agreed. The test suites are rather limited in nature. I would welcome the
> results of a more in-depth analysis. However, the trollish nature of my
> post and resulting low number of replies leads me to believe no one can or
> is willing to really back up any claims to the contrary.

That's why I asked whether or not the JIT was enabled in the tests and, if
so, if you could re-run them without it. Maybe the low number of replies
is because it's not suprising that a semi-compiled-to-native program is
faster?

-Dave




From maxm at mxm.dk  Thu Feb 28 10:40:25 2002
From: maxm at mxm.dk (Max M)
Date: Thu, 28 Feb 2002 15:40:25 +0000
Subject: string title() method does not work on Danish characters
References: <mailman.1014849825.23495.python-list@python.org>
Message-ID: <3C7E4F69.5080806@mxm.dk>

Jason Orendorff wrote:

> First: Unicode strings will solve your problem.

Hi Jason

Thanks for the suggestion, but it does not solve my problem. I am 
pulling the data from an older MS SQL server that returns standard ascii.

The Danish characters in question (??????) is defined in the 128-255 
range, so when I try to do either .unicode() or .decode() i get "ordinal 
not in range(128)"

Se my misserable attempts below :-(

I will hit "The Essential Reference" again and see what I can dig up.

regards Max M



 >>> name.encode('latin-1')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('utf-8')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('unicode-escape')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('unicode', 'replace')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('utf-8', 'replace')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('latin-1', 'ignore')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)

 >>> name.encode('latin-1', 'replace')
Traceback (most recent call last):
   File "<interactive input>", line 1, in ?
UnicodeError: ASCII decoding error: ordinal not in range(128)



From stojek at part-gmbh.de  Tue Feb 12 11:19:13 2002
From: stojek at part-gmbh.de (Marcus Stojek)
Date: Tue, 12 Feb 2002 16:19:13 GMT
Subject: read and write binary data (newbie)
References: <3c69359f.22621453@news.easynews.net> <mailman.1013525479.14390.python-list@python.org>
Message-ID: <3c694068.25382546@news.easynews.net>

>Try the struct module.  
>
>
>Rich
>

yes! just found it. thanks


From bebert at worldonline.dk  Mon Feb 11 13:39:25 2002
From: bebert at worldonline.dk (Bjarke Dahl Ebert)
Date: Mon, 11 Feb 2002 19:39:25 +0100
Subject: Beginner problem.
References: <3c67a6a3$0$12223$4d4efb8e@news.be.uu.net>
Message-ID: <zeU98.9340$B43.434971@news000.worldonline.dk>

> Here's my script, where is the error ?
>
> import sys
>
> if sys.argv[??????] == "" :
>     print "Usage : sys.argv[0] logname string_to_search"
> else :

Six question marks in a row is not valid Python Syntax?

;-)






From gerson.kurz at t-online.de  Mon Feb 25 14:19:16 2002
From: gerson.kurz at t-online.de (Gerson Kurz)
Date: Mon, 25 Feb 2002 19:19:16 GMT
Subject: ipc pickle
References: <qb6k7ukvjcjg779e3ejrl6jf1e2m5fa367@4ax.com> <3c7a278a.13899093@news.isar.de> <d0380df9.0202251041.2d9ca447@posting.google.com>
Message-ID: <3c7a8b8c.22152046@news.t-online.de>

On 25 Feb 2002 10:41:53 -0800, hugomartires at hotmail.com (Hugo
Martires) wrote:

>This only receive twice.
>If i need to receive more than twice that is a problem.
>See this:
...
>I want to send a big vector it's size could be more or less than 8000.
>The client to receive the entire vector must receive several times
>since BUFFER SIZE it's not so big. (do you know it's maximum value ?)
>So the server send :

ok, you asked for it, you got it. Download

http://p-nand-q.com/python/rpcsys.zip

Its *slow* (well, compared to the ipc-test), but it works. Its
stripped down from something bigger, so it might contain a "legacy"
thing or two; it might even have a bug. Its not yet very well
documented, but here are some hints:

- your problem is that the socket blocks when its buffer is full.
Since you have a single-threaded app, the data is never read, and thus
you block forever. Because it was single-threaded, no actual IPC was
ever involved, so the time measured is not like "real" values.

- a real life IPC system consists of at least 2 procs, so, in the
archive you have test-server.py and test-client.py

- edit test-server.py's portnum to suit your needs (defaults to 8000
because many PFWs allow that). 

- class RpcServer handles connections from multiple clients and starts
a dedicated thread for each connection

- class RpcServerConnection handles a connection

- class RpcServerInterface is a simple python class that implements
the interface you want to make available by RPC. In this example: 

# this is the servers' RPC interface
class RpcServerInterface(RpcServerConnection):
    def test(self,argument):
        return len(argument)

it exposes just one function, "test", that has one argument, a list.
This is the only class you need to care about!

- start test-server.py

- edit test-client.py's portnum to suit your needs (defaults to 8000
because many PFWs allow that). 

- RpcClientThread implements a connection to a server (each client is
connected to one server only in this modell)

- this class has to implement stubs for functions exposed by the
server. For our test function, the stub reads

    def test(self,arguments):
        return self.RpcDispatch('test',(arguments,))

How about a more complicated example? Say, our RPC server exports this
function:

    def GetCashUnits(self,a,b,c):
        print "GetCashUnits() called", a,b,c
        return [1,2,3,4]

A function that gets three parameters and returns a list. The stub
would have to read

    def GetCashUnits(self,a,b,c):
        return self.RpcDispatch('GetCashUnits',(a,b,c))

You get the idea. Once you connect to the server you simply call these
functions, and get a sync return value, too.

LIMITATIONS: Return Values must be < 1024 bytes as of now, but that
would be easy to change. (My return values are much smaller).



From sholden at holdenweb.com  Fri Feb  8 18:20:09 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 18:20:09 -0500
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org> <00W88.108659$KM2.4246695@atlpnn01.usenetserver.com> <3C645B58.7274695A@bioreason.com>
Message-ID: <w2Z88.110795$KM2.4350238@atlpnn01.usenetserver.com>

I've already replied to the mail, but the mail gateway will probably take a
while to reflect that reply in the group. For those of you who can't wait,
the good news is I have translated all four of Guido's most recent
presentations into PDF, and Guido has already loaded them on to

    http://www.python.org/doc/essays/ppt/

I deleted my own web copy to encourage people to use the correct source.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py

"Mitch Chapman" <Mitch.Chapman at bioreason.com> wrote in message
news:3C645B58.7274695A at bioreason.com...
> Steve Holden wrote:
> >     http://www.holdenweb.com/python/py10keynote.pdf
>
> Dangit.  This URL gives me 'HTTP/1.1 404 Object Not Found'.
>
> --
> Mitch Chapman
> Mitch.Chapman at bioreason.com
>
> Stuck on NT, without PPT




From quinn at hurl.ugcs.caltech.edu  Thu Feb  7 13:15:36 2002
From: quinn at hurl.ugcs.caltech.edu (Quinn Dunkan)
Date: 7 Feb 2002 18:15:36 GMT
Subject: Autocoding evolves from........
References: <slrna5fnkg.l4j.philh@comuno.freeserve.co.uk> <972.799T919T11694325threeseas@earthlink.net> <slrna5rpmb.19d.philh@comuno.freeserve.co.uk> <3943.801T1063T14354065threeseas@earthlink.net> <3C616C5A.4D1BBE1C@ccvcorp.com> <2300.802T2206T14055804threeseas@earthlink.net> <3C62190A.A54E9A57@engcorp.com>
Message-ID: <slrna65h28.omr.quinn@hurl.ugcs.caltech.edu>

On Thu, 07 Feb 2002 01:04:58 -0500, Peter Hansen <peter at engcorp.com> wrote:
>Now my logic here may be weak, because I am *really* stupid, but
>would it not be the case that if *all* of us in this newsgroup are
>so pathetically, mind-numbingly obtuse that you cannot convince us 
>of anything beyond our own limited two-dimensional perceptions, then
>you might as well leave?

If you really wanted that you would have stopped replying.


From rdsteph at earthlink.net  Sun Feb 10 09:58:28 2002
From: rdsteph at earthlink.net (Ron Stephens)
Date: Sun, 10 Feb 2002 14:58:28 GMT
Subject: Linux and Python scripts, help needed
References: <3C665762.550F7605@earthlink.net> <pan.2002.02.10.15.23.55.53798.1174@club-internet.fr>
Message-ID: <3C668B73.9E6F5BF8@earthlink.net>

Hmmm, thanks for your comments. I just noticed that in my syslogs there
was an attempt to open /etc/news/inn.conf that failed. So I just went in
and chown 'ed that file to be owned by me rather than root. I don't know
if that is part of the problem. But in general, I am finding that almost
nothing I try to do with respect to running Python scripts using crontab
works at first, simply because of all the permission stuff. So I change
ownership and rights a lot. But I didn't even know that
/etc/news/inn.conf was needed, and maybe it isn't! I wonder what else is
needed but not owned or allowed by my user? How does one ever get any
real work done when unknown permission and file ownership problems lurk
under every hidden log ;-))))

Pedro Rodriguez wrote:
> 
> "Ron Stephens" <rdsteph at earthlink.net> wrote:
> 
> > So this morning at 4:30 AM I sat up and ran a "top" process from the
> > Bash shell, that shows all
> > running processes. Sure enough, at exactly 4:30 AM, a new Python process
> > was spawned. The new
> > Python process was matched with a new entry in the cron log file of the
> > mygale.py program being
> > called.
> >
> >
> Seems we have another 'Courageous' here ;) At least, it proves that your
> cron setup is OK.
> 
> > I watched carefully, and the mygale.py process lasted for about 30
> > seconds and then disappeared from
> > the "top" screen. I assume it terminated normally. Except that when I
> > run the mygale.py get all script
> > manually  in the BASH shell, it takes about two minutes to complete.
> > This is the only sign of any
> > problem, but still no error messages are generated anywhere I can find.
> >
> >
> Call me stupid, but if I had to do this on my machine, I will need to
> startup my dialup connexion first, and ensure that the connection is ok
> prior starting anything. But maybe you have a permanent connection.
> 
> Some hints :
> - did the same script runs fine when runned manualy at the same hour ?
>   (thinking that the sites you access may be down for backup for example)
> - do you set up something special when you login manualy (like
>   environment variables required for your scripts) ? If you set up
>   variables in .bash_profile they won't be available to your scripts.
> 
> > By the way, shouldn't it be possible to cut and paste from the BASH
> > shell to a text editor? It would save me a lot of typing ;-))))
> >
> > The clipboard in general is a frustration; I can't yet make it work
> > properly in many situations. Is it me or does the KDE clipboard lack a
> > lot of functionality compared to Windows clipboard? I can't even copy
> > and paste from K-EDIT, nor K-WORD, to Netscape Composer. Oh, well.
> >
> >
> Works fine to me. But, I did notice the following thing :
> - run netscape composer
> - run kedit
> - type something in kedit
> - select it and copy it (Menu:Edit/Copy)
> - go back in netscape composer
> - paste is not available in Edit menu...
> - ... but middle button click works fine (X11 Paste)
> 
> > One last frustration: Linux does hang, at least to the point that this
> > newbie can't recover from it. My
> > box hung when I tried to open a web presentation by Guido van Rossum
> > about Python Conference 10
> > using K Presenter.  I could not even open a console nor close the
> > offending KDE shell to get back to a
> > command line. The box also hangs when I try to print using cups on my HP
> > T45 printer. (I wish I could
> > use my printer.) My downloaded Open Office also hangs my computer just
> > by trying to open it.
> >
> Same problem with kpresenter. The system start swapping and its very
> difficult (well almost impossible) to take control back. Even an old
> powerpoint viewer runned with wine wasn't able to open it.
> 
> Regards,
> --
> 
> Pedro


From bas.vangils at home.nl  Sun Feb  3 06:52:48 2002
From: bas.vangils at home.nl (Bas van Gils)
Date: Sun, 3 Feb 2002 12:52:48 +0100
Subject: lambda expressions
Message-ID: <20020203115247.GA15877@kub.nl>

Pythonians,

I've been wondering, lately, about lambda-expressions. Most programming
languages support and I wondered why (mostly because I find them hard to
read, valueing _readable_ code). Collegues at work couldn't give me a
satisfying answer so ... can anyone explain to me why lambda-expressions
are so useful?

thankx in advance

    Bas

-- 
Bas van Gils <bas.vangils at home.nl>   -    http://members.home.nl/bas.vangils
Build a system that even a fool can use, and only a fool will want to use it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20020203/567ad6dd/attachment.sig>

From tim.one at comcast.net  Thu Feb 28 18:03:47 2002
From: tim.one at comcast.net (Tim Peters)
Date: Thu, 28 Feb 2002 18:03:47 -0500
Subject: Contributing patches (was Re: urllib and proxy)
In-Reply-To: <3c7d2377$1_6@news.newsgroups.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEFLOAAA.tim.one@comcast.net>

[Brad Clements]
> I think the Python keepers are swamped. I've had a patch pending for a
> little over a month, made 3 rounds of changes as requested but now it's
> stalled.

Yup, it sucks.  On the bright side, some patches sit over a year <wink>.

> There are a lot more patches older than mine, so I'll just wait. Don't
> get discouraged if your patch sits for a while.

I don't know what can be done about this.  Outside of Martin v. Loewis and
Michael Hudson, and even less so most days some of the PythonLabs crew, it
seems that nobody routinely looks at the bugs, patch or feature request
trackers.  It's particularly discouraging for patches because, against
initial hopes, nobody "out there" *tries* them and reports back on their
experience.

So, in practice, nothing much happens on these fronts day to day except work
on critical bugs.  Then again, that's probably a good thing to people who
fear change <wink>.

this-kind-of-bottleneck-needs-a-bigger-bottle-ly y'rs  - tim




From huaiyu at gauss.almadan.ibm.com  Fri Feb  1 14:03:15 2002
From: huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu)
Date: Fri, 1 Feb 2002 19:03:15 +0000 (UTC)
Subject: Timing and aspect oriented programming (Was Re: Equivalent of Perl chomp?)
References: <mailman.1012509053.25355.python-list@python.org>
Message-ID: <slrna5lpiu.o0.huaiyu@gauss.almadan.ibm.com>

On Thu, 31 Jan 2002 12:29:06 -0800, Mark McEahern <marklists at mceahern.com>
wrote: 
>
>I timed them with the following code and it occurred to me here's a possibly
>good case where aspect-oriented programming would be nice. 

I tried the example with aspect oriented programming.  In this case the cost
is too high, but it might be useful in some other cases.  Following are the
test program, the results, and the aspect module.


#-------------- test_ends.py -----------------------

from time import time

a = "das af dsa fdsaa fdsaf as das as fdas as fsa as dsa fadasfdsa "
x = "adsfasdf";
n = 10000; 

#------------------------------------------------------------------
# plain old test

def slice(a, x, n):
    _time = time()
    for i in xrange(n):
        #a[-1:] == x
        a[-len(x):] == x
    return time() - _time

def endswith(a, x, n):
    _time = time()
    for i in xrange(n):
        a.endswith(x)
    return time() - _time

print ; print "slice endswith"
for i in range(5):  print slice(a, x, n), endswith(a, x, n)
print ; print "endswith slice"
for i in range(5):  print endswith(a, x, n), slice(a, x, n)

#------------------------------------------------------------------
# aspect test

class Test:
    def slice(self, a, x):
        a[len(x):] == x
    
    def endswith(self, a, x):
        a.endswith(x)

from aspect1 import timeit
timeit(Test, 'slice', n)
timeit(Test, 'endswith', n)
t = Test()

print ; print "slice endswith"
for i in range(5):  print t.slice(a, x), t.endswith(a, x)
print ; print "endswith slice"
for i in range(5):  print t.endswith(a, x), t.slice(a, x)

# --------- the results
When slice uses [-1:]

slice endswith
0.226076006889 0.30391895771
0.21536898613 0.293772935867
0.214552998543 0.296599984169
0.215177059174 0.301518917084
0.214749097824 0.296058058739

endswith slice
0.287701964378 0.2228089571
0.288692951202 0.231354951859
0.28769505024 0.222257971764
0.287680029869 0.22406899929
0.287147045135 0.228953003883

When slice uses [-len(x):]

slice endswith
0.0638729333878 0.0270719528198
0.0566090345383 0.0265839099884
0.0572431087494 0.0265699625015
0.0570579767227 0.0334140062332
0.0567760467529 0.0265719890594

endswith slice
0.0266000032425 0.0571140050888
0.0265990495682 0.0568480491638
0.0266120433807 0.0567020177841
0.026643037796 0.056519985199
0.0268899202347 0.0658220052719


When slice uses [-len(x):], using "aspect-oriented" timing:

slice endswith
0.122717022896 0.0866810083389
0.122745990753 0.0862590074539
0.13433098793 0.0857419967651
0.121955990791 0.0860439538956
0.129957079887 0.0868830680847

endswith slice
0.0857659578323 0.121948957443
0.085440993309 0.12923002243
0.0860010385513 0.122326016426
0.0858659744263 0.123429894447
0.0925229787827 0.121724009514


Here's the aspect module used:

# --------------------------------- aspect1.py ----------------------

#! /usr/bin/env python
# --------------------------------------------------------------------------
# aspect.py
# 
# From: Pedro Rodriguez (pedro_rodriguez at club-internet.fr)
# Newsgroups: comp.lang.python
# Date: 2002-01-13 02:27:51 PST
# Date: Tue, 22 Jan 2002 16:29:18 +0100
#
# Modified: Huaiyu Zhu 2002-01-27, 2002-01-30

# --------------------------------------------------------------------------
# The core functions

from __future__ import nested_scopes

def wrap(aspect, wrapper, cls, *method_names):
    #print "wrap", aspect, wrapper, cls, method_names
    for method_name in method_names:
        orig_method = getattr(cls, method_name)
        wrapped_method = aspect(wrapper, orig_method, method_name)
        setattr(cls, method_name, wrapped_method)

def around(wrapper, orig_method, method_name):
    def wrapped_method(inst, *args, **kwargs):
        return wrapper(inst, orig_method, method_name, *args, **kwargs)
    return wrapped_method

def before(wrapper, orig_method, method_name):
    def wrapped_method(inst, *args, **kwargs):
        wrapper(inst, orig_method, method_name, *args, **kwargs)
        result = orig_method(inst, *args, **kwargs)
        return result
    return wrapped_method

def after(wrapper, orig_method, method_name):
    def wrapped_method(inst, *args, **kwargs):
        result = orig_method(inst, *args, **kwargs)
        wrapper(inst, orig_method, method_name, *args, **kwargs)
        return result
    return wrapped_method 

# ------------------------------------------------------------------
# Example aspects

def log_around(cls, method_name, file=None):
    def wrapper(inst, method, method_name, *args, **kwargs):
        print >> file, "starting:", inst, method_name, args, kwargs
        result =  method(inst, *args, **kwargs)
        print >> file, "stopping:", inst, method_name
        return result
    wrap(around, wrapper, cls, method_name)

def log_before(cls, method_name, file=None):
    def wrapper(inst, method, method_name, *args, **kwargs):
        print >> file, "before:", inst, method_name, args, kwargs
    wrap(before, wrapper, cls, method_name)

def log_after(cls, method_name, file=None):
    def wrapper(inst, method, method_name, *args, **kwargs):
        print >> file, "after:", inst, method_name, args, kwargs
    wrap(after, wrapper, cls, method_name)

def err_abort(cls, method_name, file=None):
    def wrapper(inst, method, method_name, *args, **kwargs):
        try:
            return method(inst, *args, **kwargs)
        except Exception, e:
            print >> file, "Error abort: %s.%s(*%s, **%s)" % (
                inst, method_name, args, kwargs), e
            raise SystemExit
    wrap(around, wrapper, cls, method_name)

def err_ignore(cls, method_name, file=None):
    def wrapper(inst, method, method_name, *args, **kwargs):
        try:
            return method(inst, *args, **kwargs)
        except Exception, e:
            print >> file, "Error ignore: %s.%s(*%s, **%s)" % (
                inst, method_name, args, kwargs), e
            return None
    wrap(around, wrapper, cls, method_name)

def timeit(cls, method_name, num=100):
    from time import time
    def wrapper(inst, method, method_name, *args, **kwargs):
        _time = time()
        for i in xrange(num):
            method(inst, *args, **kwargs)
        return time() - _time
    wrap(around, wrapper, cls, method_name)
    
# ------------------------------------------------------------------
# Example applications

if __name__ == "__main__":
    
    class A:
        def f1(self, x, y):
            s =  "f %s %s" % (x, y)
            print " - inside", s; return s
    
        def g1(self, x):
            s = "g %s" % x
            print " - inside", s; return s

        def h1(self, x):
            s = "h %s" % (1/x)
            print " - inside", s; return s

        def i1(self, x):
            s = "i %s" % (1/x)
            print " - inside", s; return s
    
    class B:
        def z(self, *args, **kwargs):
            s = "z %s %s" % (str(args), str(kwargs))
            print " - inside", s; return s
                    
    log_around(A, "h1")
    log_before(A, "g1")
    err_abort(A, "h1")
    err_ignore(A, "i1")
    log_after(B, "z")
    log_around(B, "z")

    class C(A): pass
    
    a = C()
    print a.f1(1,2)
    print a.g1(3)
    
    print
    
    b = B()
    print b.z(1,2,3, x=42)
    print b.z(4)

    print 
    print a.h1(3)
    print a.i1(0)
    print a.h1(0)
    print "Should not get here"




From cliechti at gmx.net  Fri Feb  1 13:56:02 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 1 Feb 2002 19:56:02 +0100
Subject: importing in Dictionairy
References: <01c1ab1c$13174290$528e4991@pc1490>
Message-ID: <Xns91A8CB47036Ccliechtigmxnet@62.2.16.82>

"Arno Baan" <a.baan at fokkerspace.nl> wrote in
news:01c1ab1c$13174290$528e4991 at pc1490: 

> I really don't understand this, please help me.
> I've got the following problem, illustrated by the example code 
below.
> When I embed the python interpreter and I want to compile and run 
some
> simple python code. 
> It stops when I try to import a module.
> When I run the same code using PyRun_SimpleString it works the way I 
> expect it to. But because I need to retrieve computed values I need 
> the dict object.
> Can someone explain to me why I can't import modules when I use the 
> dict object??
> 
> #include<Python.h>
> #include<compile.h>
> #include<eval.h>
> #include<string>
> 
> void main(int argc, char * argv[]){
> PyObject * pcode, * pdict, * pmod;
> 
> string code1 = "print 'importing string'\nimport string\nprint 
'done'";
> Py_Initialize();
> 
> pdict = PyDict_New();
> pcode = Py_CompileString((char *)code1.c_str(), "<embed>",
> Py_file_input); 
> 
> (void) PyEval_EvalCode((PyCodeObject *)pcode, pdict, pdict);
> }
> 
> Output:
> importing string
> 
> Thanks a lot,
> 
> arno
> 

just a guess:
don't you need a "\n" at the end of the string above?

-- 
Chris <cliechti at gmx.net>



From mdk at mdk.com  Fri Feb 15 14:54:07 2002
From: mdk at mdk.com (MDK)
Date: Fri, 15 Feb 2002 14:54:07 -0500
Subject: Convert String to Dictionary question
References: <mailman.1013720546.5030.python-list@python.org> <a4j1h7$kker$1@ID-98166.news.dfncis.de>
Message-ID: <a4jp10$r5pc$1@ID-98166.news.dfncis.de>

"MDK" <mdk at mdk.com> wrote in message
news:a4j1h7$kker$1 at ID-98166.news.dfncis.de...
> Okay, I am conviced that I should use cPickle over evel().
>
> However, this brings me to another question.  Athough I use Python 2.2 I
> need to be sure that everything I do is compatable with 1.5.  Is there a
> chart somewhere that will show me, for example, that cPickle will work
with
> both versions?
>
> Thanks for all the replies.
>
>

Maybe cPickle won't work for me:

  File "C:\Python22\!MyTest\pipe.py", line 88, in put
    pic = cPickle.dumps(object,1)
cPickle.UnpickleableError: Cannot pickle <type 'array'> objects




From alet at unice.fr  Wed Feb 20 04:04:26 2002
From: alet at unice.fr (Jerome Alet)
Date: Wed, 20 Feb 2002 10:04:26 +0100
Subject: Python and GIS
References: <3C72F52F.5050901@sympatico.ca>
Message-ID: <3C73669A.991B1D02@unice.fr>

Ramkumar Kashyap wrote:
> I was wondering if there are any modules for GIS, Mapping, Layering, or
> hooks to Arcinfo, Mapinfo, etc.

May I suggest you my Python module and command line tool PDFMap, which
produces clickable maps in the PDF format ?

You can place objects on maps, they are scaled and oriented correctly,
and can link back to any web-enabled application if needed.

You can find it at http://sourceforge.net/projects/pdfmap/

hth.

Jerome Alet


From yvan.radenac at equant.com  Fri Feb  1 06:29:11 2002
From: yvan.radenac at equant.com (Yvan Radenac)
Date: 1 Feb 2002 03:29:11 -0800
Subject: I need your experience - classification and comparison of languages
References: <mailman.1011630571.24947.python-list@python.org> <b9660815.0201220007.45a98535@posting.google.com>
Message-ID: <b9660815.0202010329.27ba0caf@posting.google.com>

Thank you for your posts, it's the first time i use the news and my
english is far from perfect.
Thank you for all url you gave me.
It's a good beginning for my report.

I'm interresting in your experience as programmer, responsible of
projects, ....
As i will make a average of all results to obtain something the most
objective i can.

I found some old documents and comparisons (they are written between
1992 and 1997) or very subjective (like Java versus others by Sun,
...).

It seems that the languages the most used will be:
Ada (95), Beta, OCaml, OO COBOL,CLOS, C++, C#, Objective-C, Eiffel,
Java, Modula-3, O Oberon, Object Pascal (like Turbo-Pascal, Delphi,
...), Perl,
Python, Ruby, Sather?, Simula, Smalltalk.

I don't add to this list all the languages that have only a commercial
implementation, like VBA, ...
as i write a report about "oo languages and their public
implementation".

If you think i forget one or more oo common languages, let me know.

The result will be posted under the Free Documentation Licence from
GNU.

Of course, if you want to answer this table, thank you.

Regards
Yvan

Comparison:
----------
It's based on this scale table (to simplify):
Very bad|bad|Correct|Good|Very good|
--------|---|-------|----|---------|
  - -   | - |   O   |  + |   + +   |

Language|Readibility|Writability|Reliability|Cost
--------|-----------|-----------|-----------|----
        |           |           |           |


P.S.: a resume of the criterias, based on a course of The University
of Ottawa by Szpakowicz:
Comparison:
----------
Readability:
  - abstraction, support for generality: procedural abstraction, data
absraction.
  - absence of ambiguity (and of too much choice).
  - Orthogonality: no restrictions on combinations of concepts. For
example, can a procedure parametrer have ANY type? Can EVERYTHING be
evaluated?
  - Expressivity of control and data structures. (Exemples of low
expressive power: machine languages).
  - Appearance: style of comments, ...
Writability:
  - Abstraction and simplicy like readibility.
  - Expressivity, like readibility.
  - Modularity and tools for modularization, support for integrated
programmer's environments.
Reliability:
  - Safety for the programmer (type checking, error and exception
handling, unambiguous naming).
Cost:
  - Development time (ease of programming, availability of code).
  - Efficiency of implementation: how easy it is to build a language
processor.
  - Translation time and quality of object code.
  - Portability and standardization.


From akuchlin at dust.mems-exchange.org  Thu Feb 21 14:03:45 2002
From: akuchlin at dust.mems-exchange.org (A.M. Kuchling)
Date: 21 Feb 2002 19:03:45 GMT
Subject: Why Python is like BASIC (and why this is a good thing)
References: <slrna6qmd7.1so.rmunn@rmunn.dyndns.org> <de3fc1ef.0202201233.2228e5b1@posting.google.com> <slrna784si.m30.akuchlin@dust.mems-exchange.org> <de3fc1ef.0202201835.7f49f4be@posting.google.com>
Message-ID: <slrna7ah4h.mrg.akuchlin@dust.mems-exchange.org>

In article <de3fc1ef.0202201835.7f49f4be at posting.google.com>, 
   William Tanksley Google wrote:
> Good suggestion.  Where is it?

In the standard library for the past few versions.  (I think it was written
for some CP4E-related reason.)

--amk


From brian at sweetapp.com  Fri Feb 22 15:12:21 2002
From: brian at sweetapp.com (Brian Quinlan)
Date: Fri, 22 Feb 2002 12:12:21 -0800
Subject: XML RPC Woes
In-Reply-To: <20020222.142716.1960709859.1525@d57-47-79.home.cgocable.net>
Message-ID: <001e01c1bbdd$3d19ebc0$445d4540@Dell2>

Mark wrote:

> That was working just fine. However, all of a sudden I have not been
> able to get it to work at all. The server just keeps on returning "500
> Internal Error" now. I have investigated my own code as well as parts
of
> the SimpleXMLRPCServer and xmlrpclib (printing out values of variables
> that may be causing problems, etc).  The XML that is being sent by XML
> RPC to the server seems to be good. Does anybody have a clue what may
be
> causing this?

1. Have you enabled logging in your XML-RPC server? What output are you
getting?

2. Are you sure that your client is connecting to the correct URL?

3. Try subclassing SimpleXMLRPCServer, like so (code not tested):

class MyServer(SimpleXMLRPCServer):

    def handle_request(self):
        request, client_address = self.get_request()
        if self.verify_request(request, client_address):
            self.process_request(request, client_address)

This will cause the server to print exceptions and stop if they are
encountered.

Cheers,
Brian




From dima at xenon.spb.ru  Tue Feb  5 19:55:25 2002
From: dima at xenon.spb.ru (Dmitry Rozmanov)
Date: Wed, 6 Feb 2002 03:55:25 +0300
Subject: ASPI from Python
Message-ID: <a3putt$9sp$1@news.wplus.spb.ru>

Hi.

How can I interface to ASPI from python in Windows2000? Is it possible?

How can I talk with a scanner from python in Linux?

Thank you.

    ---Dmitry.





From sfurman at nameconnector.com  Thu Feb 14 17:00:35 2002
From: sfurman at nameconnector.com (Simrod Furman)
Date: Thu, 14 Feb 2002 17:00:35 -0500
Subject: xmlrpclib RPC calls sometimes hang 
Message-ID: <BFFEB31BC23CCD44B8924E139E1D87E4123C36@MAILBOX.nameconnector.com>

I'm writing a xmlrpclib client in Python 2.1 using xmlrpclib 0.9.9 on
Windows NT. I'm having a problem however, since the RPC call through
xmlrpclib is a blocking call, if there are any problems during the function
call such as a dropped modem or a problem on the XML-RPC server, the RPC
call wedges. As far as I can tell, xmlrpclib just hangs inside the RPC call.

I'm also using SSL from the M2Crypto library as a plug in the 'transport'
argument to xmlrpc.Server()

I've heard the suggestion to use timeout sockets from here: 

http://aspn.activestate.com/ASPN/Mail/Message/python-announce/935327

as a way to add this functionality in, but I think it would require me to
modify xmlrpclib.

Question 1: Is there a way to get xmlrpclib to timeout and throw an
exception after a specified amount of time?

Question 2: Has anyone tried xmlrpclib, SSL, and timeout sockets in Python
2.2?

--Simrod 

Simrod Furman
Parlance Corporation, home of the NameConnector Service
http://www.nameconnector.com



From brian_l at yahoo.com  Tue Feb 12 03:14:23 2002
From: brian_l at yahoo.com (Brian Lenihan)
Date: 12 Feb 2002 00:14:23 -0800
Subject: OT: (ambi)dexterity of Pythonistas
References: <b34d3835.0201300026.28e51376@posting.google.com> <a3d46d$tfn2@news1.gtech.com> <slrna5m2u6.vvj.quinn@schilling.ugcs.caltech.edu> <CuE68.17883$om6.426379@news1.tin.it> <uwuxwsqqm.fsf@ctwd0143.fitlinxx.com> <a3h3cn$cs6$1@newshost.accu.uu.nl> <bgw78.38375$KM2.1185647@atlpnn01.usenetserver.com> <mailman.1012934533.15278.python-list@python.org> <afe3b146.0202100036.f221c9a@posting.google.com> <ZND98.140690$KM2.6229113@atlpnn01.usenetserver.com>
Message-ID: <afe3b146.0202120014.1d38b556@posting.google.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message news:<ZND98.140690$KM2.6229113 at atlpnn01.usenetserver.com>...
> "Brian Lenihan" <brian_l at yahoo.com> wrote in message
> news:afe3b146.0202100036.f221c9a at posting.google.com...

> > I don't throw like a girl right-handed, so maybe I am actually
> > ambidextrous.
> >
> Not to mention gender-biased? :-)

Of course I'm gender biased.  I love girls.  I'm also sure there are
plenty of girls who can throw better than I can.  I left out the fact
that I'm also a switch-hitter because I thought it would only add to
the confusion.


From bkelley at wi.mit.edu  Tue Feb 19 16:15:32 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Tue, 19 Feb 2002 16:15:32 -0500
Subject: Editable strings anyone?
References: <7x1yfk45z8.fsf_-_@ruckus.brouhaha.com> <m37kpcuu31.fsf@mira.informatik.hu-berlin.de> <7x4rkgw79y.fsf@ruckus.brouhaha.com> <c76ff6fc.0202171321.4bd2e69c@posting.google.com>
Message-ID: <3C72C074.306@wi.mit.edu>

John Machin wrote:

> 
> You'll find one nuisance when working with 'c' arrays; it doesn't
> support coercing an ordinary string to an array -- you have to do it
> yourself. For example:
> 
> x = array.array('c', 'abc')
> x = x + 'def'
> 
> doesn't work. You have to do 
> 
> x = x  + array.array('c', 'def')
> 


I encountered the same problem and of course the solution was subclass 
array.array!  Unfortunately, array.array can't be subclassed.  (Python 
2.2)  In fact, I found arraymodule.c very confusing in this respect.

It turns out that array.array is a function that returns an instance of 
type array.ArrayType which isn't so bad except that it thinks that it is 
of type array.array

This got a little bit confusing when I initially tried to subclass 
array.array

 >>> x = array.array('c', 'adc')
 >>> type(x)
<type 'array.array'>
 >>> type(array.array)
<type 'builtin_function_or_method'>

 >>> class foo(array.array): pass
...
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
TypeError: cannot create 'builtin_function_or_method' instances

 >>> class foo(array.ArrayType): pass
...
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
TypeError: type 'array.array' is not an acceptable base type


What would it take to make array.array or array.ArrayType an acceptable 
base type for subclassing?


Brian Kelley



From jlh at cox.net  Thu Feb 21 00:55:45 2002
From: jlh at cox.net (Jeff Hinrichs)
Date: Thu, 21 Feb 2002 05:55:45 GMT
Subject: help - inserting values into database tables
References: <mailman.1014086793.5272.python-list@python.org> <3c742468$0$79564$6536d21b@news.bitstream.net>
Message-ID: <BZ%c8.4227$XT1.119424@news1.east.cox.net>

"mylinuxboxroot" <aries at bitstream.net> wrote in message
news:3c742468$0$79564$6536d21b at news.bitstream.net...
> Hey chajadan,
> Well I tried clicking on the ODBC icon in  control panel, I get the error
> saying
> I need to reinstall the ODBC Components......what a bummer.
> I don't have the install CD for windows 98.
http://www.microsoft.com/data/download.htm and download the mdac for your
system.  This will update the odbc files needed.  Have fun.




From eric.brunel at pragmadev.com  Tue Feb 12 10:55:41 2002
From: eric.brunel at pragmadev.com (Eric Brunel)
Date: Tue, 12 Feb 2002 15:55:41 +0000
Subject: Drawing in "Slow motion".
References: <3C692698.C9DAD820@inenco.no>
Message-ID: <a4ba6i$14bg$1@norfair.nerim.net>

Hi Hans,

Hans Kristian Ruud wrote:
> I have written a small script concerned with some drawing into a Canvas
> widget
> (The actual drawing procedure is triggered by an event bound to a
> Button).
> 
> My problem is that I would like the program to draw one line at a time;
> then
> pause, then draw the next line... and so on.
> 
> I have tried to insert pauses by calling time.seep,
> however, the drawing operations are realized only
> when the drawing procedure ends and program control is resumed by
> 
> root.mainloop()

Two solutions come to my mind:
- keep your time.sleep calls, but call the update_idletasks method on the 
canvas after each sleep. The call should redraw the screen.
- a bit trickier: use the delayed callback facility in Tk via the "after" 
method to draw while your application is running. This is a bit more 
complicated and you would certainly have to rewrite your code differently. 
But it's the only way to have a "reacting" application while the drawing 
takes place (i.e. user is able to click on buttons, call menus, etc...). 
Here is a simple example, just drawing random lines:

--------------------------------------------
import random
from Tkinter import *

w, h = 500, 500
x, y = random.randrange(w), random.randrange(h)

root = Tk()
cnv = Canvas(root, width=w, height=h)
cnv.pack(side=TOP)
Button(root, text='Enough', command=root.quit).pack(side=TOP)

def f(*whatever):
  global x, y
  oldx, oldy = x, y
  x, y = random.randrange(w), random.randrange(h)
  cnv.create_line(oldx, oldy, x, y)
  root.after(500, f)

root.after(500, f)
root.mainloop()
--------------------------------------------

HTH
 - eric -



From shalehperry at attbi.com  Tue Feb 26 13:36:48 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Tue, 26 Feb 2002 10:36:48 -0800 (PST)
Subject: Precompiled regular expressions slower?
In-Reply-To: <1014747651.11444.14.camel@hunsmac.mit.edu>
Message-ID: <XFMail.20020226103648.shalehperry@attbi.com>

On 26-Feb-2002 Peter Bienstman wrote:
>> Perhaps Peter should post the relevant code so we can see what is actually
>> happening for ourselves?
> 
> Here is the relevant portion of the code:
> 

After changing things as Jon Ribbens explains did you get a sufficient speed
boost?

My only other comment would be one of style and efficiency.  In the loop you
check the line against numerous regexs.  Would each line really match more than
one or two of those?  Perhaps a few strategic calls to continue may help.



From frederic.giacometti at arakne.com  Tue Feb 26 16:14:25 2002
From: frederic.giacometti at arakne.com (Frederic Giacometti)
Date: Tue, 26 Feb 2002 13:14:25 -0800
Subject: [ANN] JPE 2.0 The Java-python Extension
Message-ID: <3C7BFAB1.643BE736@arakne.com>

This is the official announcement of the second version of JPE.

JPE (The Java-Python Extension) is a production-grade seamless
integration of Python and Java.
JPE lets you call Java objects and classes from Python, and vice-versa,
within the same process space. JPE is reentrant (cross-language
callbacks), and supports multi-threading concurently in Java and Python.

The reference web site for JPE is http://jpe.sourceforge.net


As compared to its first release:

1) Build
The build system of JPE is based on a new build library (libplus):

  - Java and python are auto-detected from the command line, and
everything else is derived from there
(Java libraries, library paths...).

  - libplus generated its own makefiles, and runtime environment (env
variables) for the build and test phases; so there is no need to set-up
anything

  - everything is now plateform-independant, and the same scripts are
used on Win32 and posix (no more GNU make / cygwin make required on
windows)

  - JPE has been ported to posix environments. However, a shared python
library is required on these platforms. You can build your onw, or use
the branched version of Python 2.1, named 'pythonx', which is provided
for these purposes.

  - JPE is distributed in source form from CVS. Substantial efforts have
been invested to provide a config and build environment that makes it
straightforward to configure, build from source, and test
multi-component, multi-platform software systems. Hopefully, that can be
one step ahead from the current autoconf or distutil systems.
Easily extensible to support arbitrary software components and tools,
libplus currently supports C, C++, Python, Java, and swig wrappers.
Libplus generates configuration-specific makefiles that chain software
component dependencies together, and provides runtime software/tool
configuration information.
Litteratly, this is an 'open-source' project :))


2) New capabilities of JPE, for the 2d release:

- Java bean support: Jbean properties are accessed as regular python
attributes

- Cross-language native array access: JPE integrates the bufarray python
extension type; a layer on top of the python buffer interface that
provides type-safe access to contiguous fixed-length C arrays (actually,
99% of the situations in scientific C or Fortran libraries).
This integration permits to safely access arrays by shared memory
reference between Java, Python, and C; while hidding the native memory
format.


3) Acknowledgements:

Arakn?, and The Molecular Graphics Laboratory at The Scripps Research
Institute.


Benevolently yours,

Fr?d?ric Giacometti
fred at arakne.com





From threeseas at earthlink.net  Sun Feb  3 19:01:03 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Mon, 04 Feb 2002 00:01:03 GMT
Subject: Autocoding project proposal.
References: <83sn8rgmc7.fsf@panacea.canonical.org> <B87ACAC8.3B6F%jonathan@onegoodidea.com> <83it9mghng.fsf@panacea.canonical.org> <B87C2640.3CC2%jonathan@onegoodidea.com> <slrna5d2cq.g57.philh@comuno.freeserve.co.uk>
Message-ID: <1341.799T18T11762852threeseas@earthlink.net>

On 29-Jan-02 06:38:02 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Tue, 29 Jan 2002 10:13:20 +0000, Jonathan Hogg <jonathan at onegoodidea.com>
>wrote:
>>On 28/1/2002 23:49, in article 83it9mghng.fsf at panacea.canonical.org, "Kragen
>>Sitaker" <kragen at pobox.com> wrote:
>>
>>> These are problems, but they are not problems AppleScript solves.
>>
>>The only two sensible things I could get out of Timothy's manifesto was the
>>idea of GUI programs exposing their internal functionality through an
>>RPC-like mechanism, and that programming languages have a high entry cost
>>[...]
>>c.l.py isn't really the place for any of this

>I think c.l.py is an approriate place tob discuss the entry costs of
>programming and how programming languages can be made easier. Think of
>why python is the way it is.

Where else are you going to take a shot at trying to take down the
ultimate cost lower tool. The one that removes the need for putting up
with people like you.




---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From bkelley at wi.mit.edu  Wed Feb  6 11:11:49 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Wed, 06 Feb 2002 10:11:49 -0600
Subject: when does python free memory?
References: <10b3f29a.0202050840.7641923a@posting.google.com> <7934d084.0202051722.4367951@posting.google.com>
Message-ID: <3C6155C5.6090804@wi.mit.edu>

Andrae Muys wrote:

>reanes at banta-im.com (Robert Eanes) wrote in message news:<10b3f29a.0202050840.7641923a at posting.google.com>...
>
>>Hi-
>>Reading the docs and searching the list archives, it seemed that
>>python should release memory as soon as the last reference to an
>>object is deleted.  However, I can't seem to prove this in practice. 
>>Look at this example:
>>
>
>This is true for 1.5.2, however later versions of python may use a
>garbage collected store for some/all variables.  I don't know enough
>about python's gc to discuss this.  (ps. anyone got a good reference
>on a discussion of the current CPython (de-)allocator?)
>
><SNIP sample code demonstrating retained heap>
>
>>So apparently python knows that the memory formerly allocated to z is
>>available, but it doesn't tell the OS.  Is this the case, and if so,
>>is there any way around it?  I have an application that needs to run
>>80-150 python processes that can stay around for 15min to an hour, so
>>I'd like them not to hold onto the maximum amount of memory that they
>>use throughout the whole life of the process.
>>
Python running under windows does release memory back to the O/S so it 
is indeed a malloc/free issue.  I got over some of these issues running 
on Linux by running large memory consuming tasks in their own process 
and not their own thread.  Each process essentially had a parent process 
which controlled the I/O back to the scheduling program and a child 
process for doing the heavy lifting.

The child processes were started using calls popen2.  The drawbacks are 
that data has to be transferred back and forth through the read/write 
pipes or through some other mechanism and that the interaction with the 
child was very coarse grained.  Benefits included that the child process 
could be killed at any time and when it finished it gave memory back to 
the O/S, they included a simple callback mechanisms with a 
model-observer pattern for simple interactions and they didn't have the 
dreaded python-thread performance hit which sometimes was as high as 35% 
for gui based applications.

We eventually wrapped these up into a bunch of classes that we were 
derived from what we called LongRunningTasks that behaved similarly to 
threads but were able to be killed.

I gave a portion of a talk at the O'Reilly conference last year that 
covered a part of the issues involved.  Since then, I haven't seemed to 
generate enough interest for pursuing these psuedo-threads.  I spent a 
little bit of time porting them to windows as well.

Mitch Chapman wrote most of the code I used back at Bioreason, maybe I 
can convince him to help me formalize a package for doing this type of 
thing.

Brian Kelley
Whitehead Institute for Biomedical Research




From admin at salvador.venice.ca.us  Wed Feb  6 19:32:35 2002
From: admin at salvador.venice.ca.us (Salvador Peralta)
Date: Wed, 6 Feb 2002 16:32:35 -0800
Subject: Embedded Python Woes
Message-ID: <a3se8h$bnm$1@siamese.noc.ucla.edu>

Hi,

I'm trying to compile the following hello sir robin with gcc without 
much success.  Can someone please tell me where am I going wrong?  

#include<Python.h>
#include<pythonrun.h>

int main() {

Py_Initialize();
PyRun_SimpleString("x = brave + sir + robin");

}


$ gcc pytest.c -I/usr/include/python2.1/ 

/home/foo/tmp/ccpO93io.o: In function `main':
/home/foo/tmp/ccpO93io.o(.text+0x7): undefined reference to 
`Py_Initialize'/home/foo/tmp/ccpO93io.o(.text+0x14): undefined 
reference to `PyRun_SimpleString'
collect2: ld returned 1 exit status

$ pwd
/usr/include/python2.1
$ grep "PyRun_SimpleString\|Py_Initialize" *
pythonrun.h:DL_IMPORT(void) Py_Initialize(void);
pythonrun.h:DL_IMPORT(int) PyRun_SimpleString(char *);


From donod at home.com  Thu Feb 21 01:45:21 2002
From: donod at home.com (Don O'Donnell)
Date: Thu, 21 Feb 2002 06:45:21 GMT
Subject: Adding static typing to Python
References: <Pine.LNX.4.33.0202181937180.4907-100000@ns0.tbye.com> <mailman.1014094230.2291.python-list@python.org> <QUtc8.8357$Zp2.779989@atlpnn01.usenetserver.com>
Message-ID: <3C7497D5.34DDBA20@home.com>

On Mon, 18 Feb 2002 10:10:07 
Steve Holden wrote:

[...argument against static typing in Python...]

>    ... Pythonistas tend
> to delight in the "incidental generality" engendered by the language's
> robust polymorphism.



WOW!  Beautifully put, and how true!

I nominate this for Quote Of The Week.

Cheers,
Don


From fperez528 at yahoo.com  Mon Feb 18 10:54:30 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Mon, 18 Feb 2002 15:54:30 +0000
Subject: New to python
References: <yxfc8.145803$Ar5.5568410@typhoon.austin.rr.com>
Message-ID: <a4s0qp$bf2$1@peabody.colorado.edu>

Avi Homes wrote:

> Having problems with dictionaries..
> 
> I have one setup i.e.
> 
> ---
> Items{ 'IName': "empty", 'IDesc': "empty", 'IPrice': "empty"}
> temp=raw_input("Whats the name?")
> Items('IName')=temp

Items['Iname'] = temp

Dict/list/tuple elements in Python are accessed with [], not (). Use () only 
for function calls.

Welcome!

f.


From lee at leeHYPHENphillips.org.invalid  Wed Feb  6 16:39:38 2002
From: lee at leeHYPHENphillips.org.invalid (Lee Phillips)
Date: Wed, 06 Feb 2002 16:39:38 -0500
Subject: boolean values in appleevents
Message-ID: <060220021639381321%lee@leeHYPHENphillips.org.invalid>

   I am trying to learn how to send appleevents with python (because I
want to script applications on the Macintosh and I dislike
AppleScript). I've had good luck with what comes in
Mac/Demo/applescript in MacPython, except that I can't figure out how
to send boolean flags to applications. I guess neither can the authors
of the included example, makedisk.py, that tells Disk Copy to make a
disk image, because their argument, "leave_image_mounted=1", has the
same effect if you change it to "leave_image_mounted=0" : the default,
which is to leave the image mounted. 
   I know that a boolean value is not being sent, because when I
monitor the appleevents with the Capture AE control panel, I see this:

Process("Disk Copy").SendAE "ddsk,Crea,'----':?my disk image?, SvAs:fss
(?[long string]?), Moun:1"

where that bit at the end, "Moun:1" or "Moun:0", is ignored by the
receiver (Disk Copy), because it is expecting a boolean. If you send
the command from AppleScript, using

tell application "Disk Copy"
   activate
   create "my disk image" saving as "my disk image.img" with leave
                                  image mounted
end tell

the appleevent that gets sent is this:

Process("Disk Copy").SendAE "ddsk,Crea,'----':?my disk image?, SvAs:?my
disk image.img?, Moun:'true'(), &subj:'null'()" 

where "Moun:'true'()" or, if you say "without leave image mounted",
"Moun:'fals'()", actually changes what the receiver does. 

   Sorry about the the verboseness. Maybe I should just ask: does
anyone out there actually know how to send boolean values in
appleevents?


From infotechsys at pivot.net  Sun Feb 17 15:06:18 2002
From: infotechsys at pivot.net (infotechsys)
Date: Sun, 17 Feb 2002 15:06:18 -0500
Subject: Q. on make install.
Message-ID: <3C700D3A.10808@pivot.net>

Hi,
I'm in the process of installing Pythonw 2.1.2 on my
Redhat release 7.2. I don't what to remove 1.5.2 that
came with the Redhat release for fear of messing thier
code up. So, do I do a make altinstall and should I
set any env. variables? The reason for the upgrade
is to use GNUe software.
Thanks.
Wayne




From cce at clarkevans.com  Tue Feb 12 18:02:19 2002
From: cce at clarkevans.com (Clark C . Evans)
Date: Tue, 12 Feb 2002 18:02:19 -0500
Subject: Re-throwing an exception without loosing the stack trace
In-Reply-To: <15465.39126.250327.239174@beluga.mojam.com>; from skip@pobox.com on Tue, Feb 12, 2002 at 04:36:06PM -0600
References: <20020212174235.A17928@doublegemini.com> <15465.39126.250327.239174@beluga.mojam.com>
Message-ID: <20020212180219.A18142@doublegemini.com>

Some more detail as my previous post wasn't that clear.
I want to re-throw an exception and maintain the stack trace.

def very_nested():  
    ...
    barf
    ...

def somewhat_nested(): 
    ...
    very_nested()
    ..

def nested():
    try:
      ...
      somewhat_nested()
      ...
    except:  # requested incantation goes here
      database.close()
      raise 
    ...
    do stuff with database
    that is not closed
    ...

def main()
    ...
    nested()

When I run this, I want to know where "barf" came from.   
Right now, when I catch exceptions I get line 33 of
nested(), which is the raise statement --  I want
the line number of "barf".

I can't use finally since I want to continue 
using the database if it succeeds, but yes,
finally is close to what I want (the exception
passes through without having it's stack trace
screwed up)

Thanks!

Clark



From s.keim at laposte.net  Wed Feb 13 07:34:32 2002
From: s.keim at laposte.net (sebastien)
Date: 13 Feb 2002 04:34:32 -0800
Subject: UTC time conversion to internal time
References: <49ja8.17872$N31.902613@ozemail.com.au>
Message-ID: <7da8d8f8.0202130434.649576b0@posting.google.com>

"Colin Brown" <cbrown at met.co.nz> wrote in message news:<49ja8.17872$N31.902613 at ozemail.com.au>...
> I wish to convert UTC times to internal time representation (the inverse of
> time.gmtime). There does not seem to be any function to do this. Does anyone
> have a clean solution please.

in the calendar module:

timegm(tuple)
An unrelated but handy function that takes a time tuple such as
returned by the gmtime() function in the time module, and returns the
corresponding Unix timestamp value, assuming an epoch of 1970, and the
POSIX encoding. In fact, time.gmtime() and timegm() are each others'
inverse.

hope this will be helpfull


From mxp at dynalabs.de  Tue Feb 26 04:24:05 2002
From: mxp at dynalabs.de (Michael Piotrowski)
Date: Tue, 26 Feb 2002 10:24:05 +0100
Subject: Compiling for HPUX and AIX
References: <mailman.1014671543.6120.python-list@python.org>
Message-ID: <x6wux0ppe2.fsf@notus.mii.dynalabs.de>

"czrpb" <nanotech at europa.com> writes:

> That seemed to work, thanks!

Glad to hear that.

> Could I bother you to explain why and more importantly why Python
> itself can not figure this out?

Since we considered using Python for a project, I did repeated tests
on HP-UX during the beta phase of 2.2 and found several portability
problems in configure.  I submitted patches for some of them, and
AFAIK they have been accepted--the behavior on HP-UX had been much
worse.  However, since we eventually decided not to use Python, I
didn't have the time to look at the rest of the problems (they're in
the bug-tracking system on SourceForge).  The problem you hit might be
one of them.

>> I just tried building Python 2.2 on HP-UX and had no problems; here's
>> my build script:
>> 
>> ---------------------------------------------------------------------
>> #! /usr/bin/sh
>> 
>> unset LANG
>> export CCOPTS='+z'
>> 
>> make distclean
>> 
>> CC=cc ./configure --without-gcc --without-cxx \
>>  --prefix=/opt/python && make
>> ---------------------------------------------------------------------

This script is from my tests with the beta versions, so I've forgotten
the details ;-) Basically, from just looking at the script, I see two
possible problems:

a) The configure script *might* only work in the C locale (that's what
   the "unset LANG" is for)
b) The configure script *might* not recognize that you're not using
   gcc without you explicitly telling it (that's what the
   --without-gcc is for)

I would have to look closer to see if one or both of these points are
actually true; at least one of them must be, because otherwise there's
nothing special.  You might want to play a bit with these two options.

Now, I'm not actually using Python, but if you want, we could try to
fix the problem.

Greetings

-- 
Michael Piotrowski, M.A.                                  <mxp at dynalabs.de>


From sag at hydrosphere.com  Wed Feb 20 18:23:13 2002
From: sag at hydrosphere.com (Sue Giller)
Date: Wed, 20 Feb 2002 16:23:13 -0700
Subject: No space left on device
In-Reply-To: <A7B5C22F20EE3C4E95744E2F03F1887F55B1@MAYOR.retzlaff.com>
Message-ID: <20020220232308263.AAA94@mail.climatedata.com@SUEW2000>

Curiouser and curiouser

I don't think the problem is in the space on the drive.

I can get the program to run completely with no slowdown if I 
generate filenames without using the mktemp function
	fileName = base + number + extension

I can also get the program to run completely, with no slowdown, if I 
generate filenames, BUT don't make any files.

There seems to be some issue with the combination of using 
mktemp to get a unique filename (the slowdown is probably due to 
some sort of checking mktemp does for each name generation) and 
also creating a file.

My original code generated a series of temp files using one of three 
different seeds, but it failed also at the 21800 level.  So changing the 
seeds doesn't cure this.

I will try a test where I build a directory tree and see if that fixes 
things.  Currently, I am inclined to think there is something in 
mktemp and my computer (ram size or some such) that might be the 
problem.  I need to try this on a drive not on my main machine  as 
well.

Thanks for replies.



From graham_guttocks at yahoo.co.nz  Wed Feb 20 19:53:40 2002
From: graham_guttocks at yahoo.co.nz (Graham Guttocks)
Date: Wed, 20 Feb 2002 16:53:40 -0800 (PST)
Subject: why does popen2 silently ignore IOError?
Message-ID: <20020221005340.32346.qmail@web10304.mail.yahoo.com>

"Donn Cave" <donn at u.washington.edu> wrote:

> Where's the IOError?

No IOError, I just meant that IMO, an IOError should have been raised.

> in the general sense that it does not raise an exception in the case
> of a non-zero exit from the child process.

[...]

> I personally prefer to get an exception, with value from stderr
> output

This was what I was looking for.  It sounds like it isn't possible.

I thought what I wanted was simple.  I'm sending mail by opening a
pipe to /usr/sbin/sendmail -- if that fails for whatever reason (e.g,
sendmail program is missing) I want an exception raised with some
indication of the problem, instead of the current silent failure.

Is there any way to do this in Python?  Any example code to get me
started?

Regards,
Graham


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From logiplexsoftware at earthlink.net  Thu Feb 14 12:14:25 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Thu, 14 Feb 2002 09:14:25 -0800
Subject: Beginner question please help!
In-Reply-To: <28eed654.0202140843.510948d5@posting.google.com>
References: <28eed654.0202140843.510948d5@posting.google.com>
Message-ID: <20020214091425.48f92fbd.logiplexsoftware@earthlink.net>

On 14 Feb 2002 08:43:21 -0800
Michael Tu wrote:

> Hello,
> 
> I am trying to do the following exercise but after spending couple of
> hours without any luck, I think it is time to get some helps. Thank
> you in advance.
> 
> Write a compare function that return 1 if x > y, 0 if x == y, and -1
> if x < y.

Hm.  This sounds suspiciously like a homework assignment.  And an
especially easy assignment at that.  Keep trying ;)

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308



From tjreedy at home.com  Sat Feb 16 17:51:06 2002
From: tjreedy at home.com (Terry Reedy)
Date: Sat, 16 Feb 2002 22:51:06 GMT
Subject: Python Gnutella Library
References: <999cc599.0202161125.64272cb7@posting.google.com>
Message-ID: <unBb8.50670$X64.17280833@news1.rdc2.pa.home.com>

"Stephen Naicken" <stephennaicken at hotmail.com> wrote in message
news:999cc599.0202161125.64272cb7 at posting.google.com...
> I am currently working on dissertation project that makes use of
> Peer-to-Peer networks.  I have been trying to find a Gnutella
Library
> for Python but without success.  There are many Gnutella libraries
for
> Java (JTella) and other languages, but so far I have yet to find one
> for Python.

Go to google and enter 'Python Gnutella' and check thru at least the
first 20 entries.
There is both a server and a client/access module written in Python.

Terry J. Reedy





From gcordova at hebmex.com  Mon Feb 25 17:56:25 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Mon, 25 Feb 2002 16:56:25 -0600
Subject: Tkinter.createfilehandler analog for Windows?
Message-ID: <19650FF08A99D511B19E00B0D0F06323060B01@SRM201002>

> 
> The Tkinter.createfilehandler() method isn't supported on Windows,
> only Unix and Mac.  So, when confronted with moving a Python program
> that uses createfilehandler() to Windows, what do people do?  Use
> threads?  Use Tcl's channel C API (but that seems inaccessible from
> Python)?  Something completely different?
> 

What is TCL's channel api?

Sounds like something which can spawn an asynchronous
event through a C callback, which in turn could be tied
to a Python callback, which ... etc.

Sounds most useful.

-gustavo



From mikeb at mitre.org  Tue Feb 26 10:29:36 2002
From: mikeb at mitre.org (Mike Brenner)
Date: Tue, 26 Feb 2002 10:29:36 -0500
Subject: Python COM -- What is the name of the file?
Message-ID: <3C7BA9E0.F60FCEA@mitre.org>

Does anybody know how to get the name of the file an Excel Spreadsheet is in?

For example, to do a file-save-as in tab-delimited format, 
it is necessary to change "C:\data\stuff.xls"
to "C:\data\stuff.txt", but I don't know where to
get the "C:\data\stuff.xls" from in any given spreadsheet.

It would even be okay, if I could just get the name "stuff" from some part of the COM object.

(And, of course, hopefully the same technique will work in Microsoft Project, etc.)




From joshm at taconic.net  Sat Feb  9 15:55:09 2002
From: joshm at taconic.net (Joshua Muskovitz)
Date: Sat, 9 Feb 2002 15:55:09 -0500
Subject: Is this..... evil?
References: <Xns91B06961736C8RASXnewsDFE1@130.133.1.4>
Message-ID: <3c658a5a_3@corp.newsgroups.com>

My suggestion would be for each type of receiver to simply create its own
dict which maps message types (strings) to local functions.  If it's in the
dict, then you can assume its callable, and thus avoid the logic per
incoming message.  If you set up your constructors correctly, you could even
inherit the list of existing handlers from the base classes.  Thus class
BaseReceiver could create the dict which handles the MsgA and MsgB types.  A
derived class could add a MsgC handler to the dict, etc.  This also allows
derived classes to "overload" base handlers by simply replacing the function
reference in the dict.

something like (coded off the top of my head, if it don't run, sorry :-)

class Receiver:
    def __init__(self):
        self.msgFuncs = {}

    def message(self, msg):
        if self.msgFuncs.has_key(msg.type):
            self.msgFuncs[msg.type](msg)

class F(Receiver):
    def __init__(self):
        Receiver.__init__(self)
        self.msgFuncs['fighto'] = rec_fighto

    def rec_fighto(self, msg):
        ...

class S(Receiver):
    def __init(self):
        Receiver.__init__(self)
        self.msgFuncs['spello'] = rec_spello
        self.msgFuncs['fighto'] = rec_fighto
        self.msgFuncs['none'] = rec_none

    ...

There are better ways to handle the dict adds, of course, but you get the
idea.  The cool thing about this is that on a per receiver basis you can
have different classes of messages still handled by a single common handler
(just point them to the same internal function).

--
# Joshua Muskovitz
# joshm at taconic.net
def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
lyyrs('Hire me!  I need the work!')




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


From robin at jessikat.fsnet.co.uk  Mon Feb  4 19:04:48 2002
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 5 Feb 2002 00:04:48 +0000
Subject: Autocoding finally explained (was Re: READ THE WEB-SITE THAT THE KIKES TRY TO CENSOR)
References: <A042AF87D5C59948FB6368C76F958@news.wanadoo.fr>
 <mailman.1012843927.21386.python-list@python.org>
 <slrna5tsnh.3r0.philh@comuno.freeserve.co.uk>
Message-ID: <UGyE2IAgGyX8EwT+@jessikat.fsnet.co.uk>

In article <slrna5tsnh.3r0.philh at comuno.freeserve.co.uk>, phil hunt
<philh at comuno.freeserve.co.uk> writes
>On Mon, 4 Feb 2002 09:21:53 -0800, Cliff Wells <logiplexsoftware at earthlink.net> 
>wrote:
>>On Mon, 4 Feb 2002 14:35:23 GMT
>>charlie at news.wanadoo.fr wrote:
>>> [...]
>>> If you will behave Norris's road beneath cases, it will daily 
>>> learn the plate.  What will we excuse after Lionel converses the 
>>> ugly castle's ache?  If the filthy buckets can measure eventually, the 
>>> pathetic bandage may lift more camps.  It's very sad today, I'll 
>>> creep firmly or Nydia will change the wrinkles.  How doesn't 
>>> Peter kill fully?  Tomorrow, it covers a cup too dry alongside her 
>>> long lake.  
>>
>>Oh! So _that's_ what Rue meant by "autocoding"....
>
>It all makes perfect sense now!
>
aaaargghhhhh I feel a generative grammode appearoid murble murble
-- 
Robin Becker


From bkelley at wi.mit.edu  Fri Feb 15 13:50:35 2002
From: bkelley at wi.mit.edu (Brian Kelley)
Date: Fri, 15 Feb 2002 13:50:35 -0500
Subject: Python on Windows CE
References: <a4ev4g012jn@enews4.newsguy.com>
Message-ID: <3C6D587B.5040203@wi.mit.edu>

MegaHurts wrote:

> I've loaded the 1.5.2 Python for Windows CE
> onto my MobilePro with CE 2.0.
> 

I was using Python CE for my Vadem Clio a couple of years ago.  You 
certainly can write small applications with windows CE, I wrote the 
first prototype of a chemoinformatics package while sitting on a beach 
in fiji.

The best example of writing a gui is given by the taskbar demo which 
comes with the distribution, I believe this can be run standalone.

Unfortunately, 1.5.2 seems to be the end of the CE 2.0 distibution that 
Mark Hammond worked on a long time ago.

Brian



From gh_pythonlist at gmx.de  Tue Feb 19 14:41:49 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 20:41:49 +0100
Subject: Finding nonprintable characters?
In-Reply-To: <3C72A41F.20908@verio.net>
References: <3C72A41F.20908@verio.net>
Message-ID: <20020219194149.GA4533@lilith.hqd-internal>

Le 19/02/02 ? 12:14, VanL ?crivit:
> Hello,
> 
> I have a function
> 
> isBinary(filehandle)
> 
> that I'm not sure how to implement.  I've decided to define binary as 
> containing characters above \x80.

This algorithm only works for ASCII text. German, Spanish, French and
other texts are right out.

If you're on Unix, I'd suggest you interface the file command instead.

> But  what is the best way to do this?
> 
> 1. iterate through xreadline, so the whole thing doesn't get loaded into 
> memory?

You can read chunks of data with the read method of file-like objects.
read() takes an optional parameter with the lenght of the block. So
.read(1024) will read chunks of a lenght of 1024 bytes, for example.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.0 ?C      Wind: 6.8 m/s



From rhymes at myself.com  Sun Feb  3 05:12:13 2002
From: rhymes at myself.com (Lawrence Oluyede)
Date: Sun, 03 Feb 2002 11:12:13 +0100
Subject: curses module
References: <953o5u077ruthfqmljnauej8m4v8432qiu@4ax.com> <mailman.1012700705.5894.python-list@python.org> <a3j1a4$18e7ov$1@ID-7529.news.dfncis.de>
Message-ID: <f73q5uo64b3uhk6hbvb9u4834bvkaf8dr4@4ax.com>

On Sun, 3 Feb 2002 10:52:54 +0100, "Karl M. Syring" <syring at email.com>
wrote:

>But it shouldn't it build with pdcurses?

What is  pdcurses?

-- 
Lawrence "Rhymes" Oluyede (rhymes at myself.com)
"Per cambiare, per diventare un'altra
cosa dobbiamo prima sapere cosa siamo"
(Bruce Lee)


From codetube at yahoo.com  Mon Feb  4 02:46:57 2002
From: codetube at yahoo.com (CodeTube)
Date: 3 Feb 2002 23:46:57 -0800
Subject: Compiling Pythonwin.dsw
Message-ID: <5d65cf44.0202032346.63bc2913@posting.google.com>

Hello Everyone,

I seem to have a problem with compiling PythonWin. I get the message
"Unable to find win32ui.pyd. The module could not be found." when I
try to compile the project pythonwin.dsw.

Any ideas? Any help would be most appreciated.

Thank you,
CodeTube


From sholden at holdenweb.com  Wed Feb  6 13:27:29 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 6 Feb 2002 13:27:29 -0500
Subject: how to pass by reference??
References: <a3knt6$gr3$1@news.efn.org>
Message-ID: <jAe88.87233$KM2.2811767@atlpnn01.usenetserver.com>

"Spencer Doidge" <spencer at spencerdoidge.com> wrote ...
> I really tried to RTM first this time, but I failed.
> Can someone give me a hint where in the doc html's it explains how to pass
> by reference?

The same place it explains how to use arguments :-) -- by reference is the
only way to do it.

References to immutable objects, however, (strings, integers, etc.) cannot
be used to change the value, since the function argument is a *local copy*
of the references passed in. Binding that local copy to a new value will
have no effect on the argument that was passed in to the function.

> Specifically, I want to do the equivalent of this:
>
> char s[10];
>
> void foo(char *bfr)
> {
> bfr[0] = 'a';
> bfr[1] = 'b';
> return;
> }
>
> void main(void)
> {
> foo(s);
> return;
> }
>
Because strings are immutable there is no way to do this. You'd have to
return a new string value form the function and bind it to the name you
passed in as an argument.

regards
 Ste e
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From sholden at holdenweb.com  Wed Feb 20 08:26:35 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Feb 2002 08:26:35 -0500
Subject: Recursively traverse linked list -- help!
References: <3C71B585.90F644EB@kontact-no-spam-please.com> <6bCc8.26238$zJ3.1880837@typhoon.maine.rr.com> <7xu1scew8m.fsf@ruckus.brouhaha.com> <slrna76g9l.rag.yhatakka@kenny.evolvit.fi> <7xeljg7h7b.fsf@ruckus.brouhaha.com>
Message-ID: <RtNc8.17064$Zp2.1755158@atlpnn01.usenetserver.com>

"Paul Rubin" <phr-n2002a at nightsong.com> wrote in message
news:7xeljg7h7b.fsf at ruckus.brouhaha.com...
> yhatakka at dlc.nospam.fi (Yrj? Hatakka) writes:
> > >> def traverse(list):
> > >>     if list:
> > >>         print list[0]  # or do something here
> > >>         traverse(list[1:])
> > >>     else:
> > >>         print "done"
> > >>
> > >
> > > That's ok in Scheme but I don't think any current Python
implementations
> > > handle tail recursion "correctly".
> >
> > What do you mean? I tried above an it seems to work. When does it stop
> > working?
>
> By "correctly" I mean the code should be executed without pushing
> another recursive call onto the call stack for every list element.
> The recursive call should instead be compiled as a jump back to the
> top of the function.  I don't remember seeing it documented anywhere
> that Python does that.  Even if the current implementation happens to
> transform tail recursion like that, it's an implementation-specific
> optimization rather than part of the language definition like it is in
> Scheme.

Please remove your Scheme prejudices before treading in the Python world -
quotation marks notwithstanding. The remark you originally made was like an
obscure line of code that needed ten lines of comment to justify it :-)

All the tail recursion optimization does is speed things up slightly and
avoid stack exhaustion at high recursion depths. No wonder you felt the need
to quote "correctly".

For what it's worth, no current implementations of Python implement such
tail recursion optimization (AFAIK).

lots-of-better-ways-to-optimize-before-that-ly y'rs  - steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From anthony_barker at hotmail.com  Wed Feb 13 10:42:41 2002
From: anthony_barker at hotmail.com (Anthony_Barker)
Date: 13 Feb 2002 07:42:41 -0800
Subject: URLLIB2 problems
References: <899f842.0202121213.603e6617@posting.google.com>
Message-ID: <899f842.0202130742.2bc64e4c@posting.google.com>

I am still having problems getting through the proxy, I tried another method

 
import urllib2
 
# set up authentication info
authinfo = urllib2.HTTPBasicAuthHandler()
authinfo.add_password('realm', 'host', 'username', 'password')
#wasn't quite sure which realm it was so I just selected the server name
 
proxy_support = urllib2.ProxyHandler({"http" : "http://spamandeggs:8080"})
 
# build a new opener that adds authentication 
opener = urllib2.build_opener(proxy_support, authinfo)
 
# install it
urllib2.install_opener(opener)
 
f = urllib2.urlopen('http://www.python.org/')

...........fails with same network 101 error



anthony_barker at hotmail.com (Anthony_Barker) wrote in message news:<899f842.0202121213.603e6617 at posting.google.com>...
> I am having problems with python and our company's proxy server, which
> requires  authentication.
> 
> here is the code I am using:
> 
> import urllib2 
> proxy = urllib2.ProxyHandler({"http":"http://myname:mypass at proxyserver.company.com:8080/"})
> opener = urllib2.build_opener(proxy) 
> f = urllib2.urlopen('http://www.python.org') 
> 
> results
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
>     return _opener.open(url, data)
>   File "/usr/lib/python2.2/urllib2.py", line 322, in open
>     '_open', req)
>   File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
>     result = func(*args)
>   Fil

e "/usr/lib/python2.2/urllib2.py", line 785, in http_open
>     return self.do_open(httplib.HTTP, req)
>   File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
>     raise URLError(err)
> urllib2.URLError: <urlopen error (101, 'Network is unreachable')>
> 
> Under windows error is the same except for the last line (and line
> numbers)
> URLError: <urlopen error (10051, 'Network is unreachable')>
> 
> Both perl and wget work. wget only supports Basic Authentication, so I
> assume it is the same. Thinking it might be a network type issue I
> changed the domain names to IP addresses. (results follow)
> 
> Any help would be appreciated................
> 
> File "<stdin>", line 1, in ?
>   File "/usr/lib/python2.2/urllib2.py", line 138, in urlopen
>     return _opener.open(url, data)
>   File "/usr/lib/python2.2/urllib2.py", line 322, in open
>     '_open', req)
>   File "/usr/lib/python2.2/urllib2.py", line 301, in _call_chain
>     result = func(*args)
>   File "/usr/lib/python2.2/urllib2.py", line 785, in http_open
>     return self.do_open(httplib.HTTP, req)
>   File "/usr/lib/python2.2/urllib2.py", line 763, in do_open
>     raise URLError(err)
> urllib2.URLError: <urlopen error (111, 'Connection refused')>


From fumingw at yahoo.com  Thu Feb 21 16:57:17 2002
From: fumingw at yahoo.com (Fuming Wang)
Date: 21 Feb 2002 13:57:17 -0800
Subject: Python 2.2 on windows 98
References: <b0fd9226.0202201104.2c6a220@posting.google.com> <a50uti$ckg$1@norfair.nerim.net>
Message-ID: <b0fd9226.0202211357.2ab3ab41@posting.google.com>

"Gillou" <nospam at bigfoot.com> wrote in message news:<a50uti$ckg$1 at norfair.nerim.net>...
> Python for Windows has no readline support.
> You need to use python for cygwin (another distro) to have line recall in
> interactive mode.
> 

This worked!

Thanks,

Fuming Wang


From marklists at mceahern.com  Sun Feb  3 20:58:43 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 3 Feb 2002 17:58:43 -0800
Subject: Finding all of a classes methods?
In-Reply-To: <roy-CA3285.20323003022002@news1.panix.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBOEKGKAAA.marklists@mceahern.com>

Roy Smith:
> Is there any way to find all of a class's methods?

There's probably a better way than this, but it seems to work plus it might
give you some ideas...

#! /usr/bin/env python

import inspect

def runall(obj):
    all = inspect.getmembers(obj, inspect.ismethod)
    for methodTuple in all:
        methodTuple[1]()

class foo:

    def bar(self):
        print "bar"

    def bar2(self):
        print "bar2"

f = foo()
runall(f)





From python at chajadan.net  Tue Feb 26 03:18:48 2002
From: python at chajadan.net (chajadan)
Date: Tue, 26 Feb 2002 00:18:48 -0800
Subject: VPython--close causes crash!
Message-ID: <5.1.0.14.0.20020226001726.00a69050@63.118.78.184>

I downloaded VPython, and PyGeo, but whenever I close the VPython generated 
windows, it crashes all of Pythonwin.

Maybe this is a pythonwin problem, I'll try Idle--but anyone out there have 
any ideas?

--chajadan




From jgardn at alumni.washington.edu  Mon Feb  4 10:56:57 2002
From: jgardn at alumni.washington.edu (Jonathan Gardner)
Date: Tue, 5 Feb 2002 00:56:57 +0900
Subject: Stop feeding the trolls! Kill the autocoding thread!
In-Reply-To: <3c5e9cb9.17829734@news.t-online.de>
References: <83bsfkk31e.fsf@panacea.canonical.org> <83sn8rgmc7.fsf@panacea.canonical.org> <3c5e9cb9.17829734@news.t-online.de>
Message-ID: <200202041602.g14G2Oi28655@my.knctv.co.kr>

A quote of your post is below. Go ahead and reply, but don't post a copy to 
the list!

On Monday 04 February 2002 11:40 pm, Gerson Kurz wrote:
> On 27 Jan 2002 22:56:24 -0500, Kragen Sitaker <kragen at pobox.com>
>
> wrote:
> >COBOL has the same advantage, really.  Not that I'd want to use or
> >learn COBOL, but if I had to write a program to do some kind of simple
> >calculation and it needed to be readable by nontechnical people, COBOL
> >would be the tool for the job!
>
> Proof that you can write unintelligible code in a language that is as
> COBOLesque as can get:
>
> http://p-nand-q.com/sorted.htm
>
> Here is a fibonacci program in that language (note that everything is
> more or less english):
>
> This code uses the numbers twenty, seventyone, three, two, and one.
> This code sometimes goes to the second label if the first condition is
> true, sometimes goes to the fourth label if the second condition is
> true, always goes to the third label, always goes to the fifth label,
> and always goes to the first label.
> This code writes the eleventh number as a english cardinal.
> This code cannot read.
> This code uses the sums of the tenth number and the eleventh number,
> and of the second number and the fifth number.
> This code uses the condition
> that the second number is equal to the first number,
> and the condition that the second number is less than the third
> number.
> This code uses five labels.
> This code does not use any ordered differences.
> This code assigns the fourth number to the tenth number,
> the fifth number to the eleventh number,
> the fifth number to the second number,
> the second sum to the second number,
> the eleventh number to the fourth number,
> the eleventh number to the tenth number,
> and the first sum to the eleventh number.
> This code does not use any products.
> This code implements
> the third assignment,
> the first label,
> the first jump,
> the second jump,
> the third jump,
> the fourth label,
> the second assignment,
> the sixth assignment,
> the fourth jump,
> the third label,
> the fifth assignment,
> the seventh assignment,
> the first assignment,
> the fifth label,
> the first output,
> the fourth assignment,
> the fifth jump, and
> the second label.
> This code does not use any ratios.
> This code does not use any logical operations.
> Cool.



From cliechti at gmx.net  Fri Feb 15 15:56:45 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 15 Feb 2002 21:56:45 +0100
Subject: Basic Question
References: <m3y9huzd3a.fsf@mycomp.tky.hut.fi>
Message-ID: <Xns91B6DFCA18150cliechtigmxnet@62.2.16.82>

Karthikesh Raju <karthik at james.hut.fi> wrote in 
news:m3y9huzd3a.fsf at mycomp.tky.hut.fi:
> def randomList(n):
...
> in a file called randomList.py. Started python, imported randomList
> (>> import randomList) and then called this function (>>
> randomList(5)), got this un-understandable error:

>>> randomList.randomList(5)

the function is IN the module (which just has the same name in this 
case)
-- 
Chris <cliechti at gmx.net>



From charles at a3a.com.br  Fri Feb 22 14:30:05 2002
From: charles at a3a.com.br (Charles)
Date: Fri, 22 Feb 2002 15:30:05 -0400
Subject: ncurses
Message-ID: <02022215300500.03018@magneto>

Anybody know where can i find a GOOD tutorial or explantion
about ncurse (curses) python module ?

i need create a counter in linux console with a self refresh.
printing the value in same place, overwriting the old value.
like seconds in a clock.

Charles




From mdk at mdk.com  Wed Feb  6 08:32:56 2002
From: mdk at mdk.com (MDK)
Date: Wed, 6 Feb 2002 08:32:56 -0500
Subject: Is there a Python module that already does this?
Message-ID: <a3rbad$19u08k$1@ID-98166.news.dfncis.de>

I need to convert a list into a list of characters.

For example:

("cat",5,['dog',[3,3,1]],"zoo")

Would become:

('c','a','t',5,'d','o','g',3,3,1,'z','o','o')

Any information would be appreciated.

Thanks




From phr-n2002a at nightsong.com  Thu Feb 14 01:28:52 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 13 Feb 2002 22:28:52 -0800
Subject: python to lisp translation
References: <3C6A7825.714DF612@setf.de> <mailman.1013619443.24325.python-list@python.org>
Message-ID: <7x1yfomva3.fsf@ruckus.brouhaha.com>

Justin Sheehy <justin at iago.org> writes:
> > is there a python to lisp translator out there any where?
> 
> Not that I know of.
> 
> If there really was a good one (to either CL or Scheme) that would
> have the possibility of mollifying many of the performance-obsessed
> folks, since both of those languages have some good optimizing compilers.
> 
> However, it would be a seriously hard thing to do correctly and well.

I've been kicking the idea around for a while.  Do you see any
specific stumbling blocks?  I can see having to modify the underlying
Scheme implementation in some minor ways to get things like immutable
strings, but most of Python maps onto Scheme in a pretty natural way.


From threeseas at earthlink.net  Tue Feb  5 23:54:05 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:05 GMT
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net>
Message-ID: <1372.801T1070T13994028threeseas@earthlink.net>

On 03-Feb-02 19:47:37 Peter Seebach <seebs at plethora.net> wrote:
>In article <1173.799T632T11653818threeseas at earthlink.net>,
>Timothy Rue <threeseas at earthlink.net> wrote:
>>Don't like the idea of autocoding huh?

>Haven't been able to perceive whether or not you have an idea.  I have to
>say, you're talking to the wrong crowd.  You need people less mired in
>conventional ideas of software engineering.  Your best bet is almost
>certainly the C# people; they are certainly not mired in conventional ideas
>of software engineering.

Why would  be interested in a programming language by a company that has
been found guilty of Anti-trust? Not to mention that language is not
available anywhere near as much as something like Python.

It'd really be a good thing if the MS cronies state so in their sig.

This way I don't have to try and figure out why you have a problem knowing
the difference between what is idea and what is natural law physics. For
MS seems to think it own reality. It doesn't!

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From michael at rcp.co.uk  Mon Feb  4 11:30:15 2002
From: michael at rcp.co.uk (Michael Abbott)
Date: Mon, 4 Feb 2002 16:30:15 +0000 (UTC)
Subject: How can I write this C code in Python?
References: <e959e7b.0202020710.417801c8@posting.google.com>
Message-ID: <Xns91ABA7E2EB0C6michaelrcpcouk@194.238.50.13>

c at tech.usd253.org (Fred) wrote in 
news:e959e7b.0202020710.417801c8 at posting.google.com:

> Hello, I am just starting out with Python. I have a very basic book
> that is not helping me much. I have done some simple C programming in
> the past and I think if I could see the below C code done in Python it
> would get me over the beginners hump. I tried to do it in Python but
> got stuck right at the start, Python did not like it when I tried to
> say: for i = 1
> 
> Can anybody help here? Thanks!
> 
> Fred  c at tech.usd253.org
> 
> //Simple C code to create a times table.
> 
> #include <stdio.h>
> int main (void)
> 
> {
> 
> int i;
> int j;
> 
> for (i = 1; i < 13; i++)
>      {
>      for (j = 1; j < 13; j++)
>        printf("%4d", i*j);
>              printf("\n");
>    }
> return 0;
> }
> 

A direct translation is:

for i in range(1,13):
    for j in range(1,13):
        print '%4d' % (i*j),
    print



From logiplexsoftware at earthlink.net  Fri Feb  8 13:37:14 2002
From: logiplexsoftware at earthlink.net (Cliff Wells)
Date: Fri, 8 Feb 2002 10:37:14 -0800
Subject: Defending the Python lanuage...
In-Reply-To: <200202081311.g18DBQGo020260@ratthing-b246.strakt.com>
References: <3C62EE4A.5090305@destiny.com>
	<20020207140755.364d5d77.logiplexsoftware@earthlink.net>
	<200202081311.g18DBQGo020260@ratthing-b246.strakt.com>
Message-ID: <20020208103714.11a2ae01.logiplexsoftware@earthlink.net>

On Fri, 08 Feb 2002 14:11:26 +0100
Laura Creighton wrote:

> > If if.  No doubt things such as personality are going to be a huge
factor. 
> > I guess that sums up my assertion: the type of personality attracted to
> > programming is not the type to endure watching someone else code.  My
> > fingers would twitch themselves numb.  To have pair-programming work
you
> > would need two programmers who: 
> > 
> > 1.  have similar skill levels
> 
> This is incorrect.  

How dare you.   ;)

> It works great with people with radically different
> skill levels.  Of course, they start to converge, so they do not end
> up with different skill levels.  thats the point.  For example, people
> who pair program with me and are not aware of design patterns and
signature 
> based polymorphism will get that fixed in a hurry.  But what happens
> when Alex Martelli and I are pair programming is I learn so much stuff
> I never even heard about before.

No doubt.  This is in fact how I have trained people in the past - they sit
and watch me program with some toolkit and we discuss what I'm doing (I had
great success teaching someone with no programming experience whatsoever
how to integrate Python, wxPython and MySQL - in less than 3 months he was
writing on his own [and damn arrogant too, I might add]).  However, I think
a key point in his progress a a programmer was when he finally sat at his
own computer and worked independently of me.  I'm reminded of a Chinese
proverb I read in a calculus book:

I hear and I forget
I see and I remember
I do and I understand

Now you know Laura, that your posts are so rare and insightful that I am
loathe to disagree with you, -but- ;) my point is that while working
together is extremely beneficial for sharing ideas and information, I don't
see sharing the same console with another programmer for 90% of the time to
be as beneficial to productivity.  Sitting side-by-side at separate
consoles, yes, but not at the same console (except when there are
interesting discoveries/techniques to be shared or design questions to be
pondered). 

However, this discussion does give me the idea that pair-programming could
be very beneficial if programmers were rotated on some schedule, so that
one day they watched someone else and on another someone else watched them,
especially given a group of several programmers that could be rotated
through.  This way, programmers would get enough time doing development on
their own but still have a chance to share ideas with other programmers in
their group.  Too bad I'm not in a position to give it a try :P

> It is such a joy.

I'm glad to hear it =)

> > 2.  like and respect each other
> > 3.  have compatible programming philosophies
> 
> (you are spot on with those, however.)
> 
> Cliff, again you are up against 'American management is impossible.'
> Seriously consider visiting Sweden, for a vacation.  I think you would 
> like it here.

You'd better be careful what you ask for - you're starting to tempt me -
you could end up with some guy with a strange accent and stranger ideas
hanging around ;)  I expect I could learn quite a bit from you.


-Cliff

-- 
Cliff Wells, Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308  (800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan



From justin at iago.org  Tue Feb 19 14:45:57 2002
From: justin at iago.org (Justin Sheehy)
Date: Tue, 19 Feb 2002 14:45:57 -0500
Subject: Adding static typing to Python
In-Reply-To: <24c39b2c.0202190059.41874b92@posting.google.com> (ajeru@vknn.org's
 message of "19 Feb 2002 00:59:42 -0800")
References: <24c39b2c.0202181531.187fad4c@posting.google.com>
	<24c39b2c.0202190059.41874b92@posting.google.com>
Message-ID: <x7eljh6yru.fsf@ra.iago.org>

ajeru at vknn.org (Alexander Jerusalem) writes:

> And the final argument I have for static type checking is that it
> enables method dispatching based on parameter types. In a statically
> typed language you can create two methods that have the same name but
> differ on the parameter types. The correct method will be called for
> you depending on the type of the argument you pass in your call. That
> makes for a quite flexible way of extending a program. You can just
> add another method with the same name and another type to handle a
> special case without touching the existing methods.

I use parametric polymorphism quite a bit in languages like Java.

However, Python allows you to do something that is often even better,
which wouldn't work at all with forced static typing: 
signature-based polymorphism.  (I know that there is a better term for
this, but I can't think of it right now.)

Instead of dispatching on an object's type, you dispatch on its capabilities.

Sometimes this dispatch is implicit via a method call.

The encouragement of this pattern is a very strong point for Python, imo.

-Justin

 




From chrisabraham at yahoo.com  Tue Feb 19 13:27:51 2002
From: chrisabraham at yahoo.com (Chris Abraham)
Date: Tue, 19 Feb 2002 10:27:51 -0800 (PST)
Subject: User Group Mailing Lists for All!
Message-ID: <20020219182751.67791.qmail@web10001.mail.yahoo.com>

User Group Mailing Lists for All!

iMeme <http://www.imeme.net>, ZPugDC <http://www.zpug.org>, and beehive North
America <http://www.beehiveNA.com>  would like to extend our Mailman install
and our Domain Name (lists.zpug.org) to those of you who are:

* Running a Zope User Group or Zope Python User Group or a PIG. 
* Starting a Zope User Group or Zope Python User Group or a PIG. 
* Starting a regional virtual User Group or PIG -- list only. 
* Starting a cool Skin, Product, or other Zope or Python Project List. 
* Running a cool Skin, Product, or other Zope or Python Project List.
 
So, if you have these sorts of requirements and need and want a list such as
this, please let me know via email and we will hash it out.

Part two, of course, is the web presence -- we will be offering the equiv. of
boston.zpug.org, nyc.zpug.org, bay.zpug.org and sf.zpug.org -- but that  will
not be right now...

Thanks a lot for all the cool things I got to see and discuss at Python 
10

Cheers,

Chris


=====
-- 
chris abraham <chrisabraham at yahoo.com> +1 202-452-7442 tel
po box 15163 washington, dc 20003-0163 +1 703-995-0409 fax


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com



From laalien at hotmail.com  Tue Feb 12 10:27:31 2002
From: laalien at hotmail.com (Mark Zöltsch)
Date: Tue, 12 Feb 2002 16:27:31 +0100
Subject: searching backwards in a string
References: <mailman.1013525800.15169.python-list@python.org>
Message-ID: <a4bc9e$ope$06$1@news.t-online.com>

Hi

> Short answer:  Reverse the string first.

I'm new to this python stuff and to this newsgroup too..
but why you don't use 'rfind' instead of 'find'?
I think that's much easier...

somthing like:

import string

str = 'ac ab ab ac'
res = string.rfind(str, 'ac', 0, len(str))

print 'last position :', res



SEE yA
PS sorry for bad english and grammar





From usenet at thinkspot.net  Wed Feb 20 18:46:43 2002
From: usenet at thinkspot.net (Sheila King)
Date: Wed, 20 Feb 2002 15:46:43 -0800
Subject: Is there a bug in CGI-POST
References: <8f8ffe67.0202201345.66fa13ed@posting.google.com>
Message-ID: <a50gd3.3vse83f.1@kserver.org>

On 20 Feb 2002 13:45:54 -0800, sjoshi at ingr.com (Sunit Joshi) wrote in
comp.lang.python in article
<8f8ffe67.0202201345.66fa13ed at posting.google.com>:

> Hello All
> I'm trying use the cgi module and I have seen that if I use the POST
> method in the form then I donot get any form fields. This however
> works with the GET method. Is this known, or am'i doing something
> wrong here. I'm using Xitami Server on Win2000

You are using the FieldStorage from the cgi module, right? Do you have a
statement in your script, near the beginning, like

form = cgi.FieldStorage() 

??? (after importing the cgi module, of course...)


If so, then try this. 

Somewhere in your script, after the statement:

print "Content-type: text/html\n"

put this:

print "<br><pre>", str(form), "</pre><br>"

Now save it and run it again. Type in some stuff into your form and
click submit.

The contents of the FieldStorage object should be printed to your output
screen. You can see what is stored in the FieldStorage object, and it
should be the data that you typed into the form.

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

"When introducing your puppy to an adult cat,
restrain the puppy, not the cat." -- Gwen Bailey,
_The Perfect Puppy: How to Raise a Well-behaved Dog_



From maxm at mxm.dk  Mon Feb  4 11:34:39 2002
From: maxm at mxm.dk (Max M)
Date: Mon, 04 Feb 2002 16:34:39 +0000
Subject: Xython - XML-Formed Python
References: <mailman.1012751045.16555.python-list@python.org> <f72dac0d.0202040142.43e3a90e@posting.google.com>
Message-ID: <3C5EB81F.8070404@mxm.dk>

Alan Kennedy wrote:


> I don't understand what's wrong with
> 
> <script language="python" xml:space="preserve">
>     def messiah():
>         print "He's NOT the Messiah."
>         print "He's a very naughty boy."
> </script>

How about:

<script language="python" xml:space="preserve">
     def messiah():
         print '<sript language="JScript">'
         print 'document.write("He\'s NOT the Messiah.")'
         print 'document.write("He\'s a very naughty boy.")'
         print '</script>'
</script>

hmmm ... fishy

regards Max M



From feldspar at ix.netcom.com  Sun Feb 10 14:45:50 2002
From: feldspar at ix.netcom.com (Antaeus Feldspar)
Date: Sun, 10 Feb 2002 15:45:50 -0400
Subject: Is Parrot dead?
References: <B88C9617.532F%Roman.Fischer@teleweb.at>
Message-ID: <3C66CDEE.EDA48DC2@ix.netcom.com>

"Ing. Roman Fischer" wrote:
> 
> For a long time I do not find informations about Parrot (the programming
> language). O Reilly still estimates July 2001 for the "Programming in Parrot
> in a Nutshell"-book and even Amazon (which is not very shy to offer books
> long before they are available) does not have it in its offer. I also did
> not find any real complete definitions of this language.
> 
> So is this language dead?
> 
> Roman

I think I can clear up this confusion.  ^_^  It stems from a huge,
elaborate April Fool's joke in April of 2001, in which Guido van Rossum
and O'Reilly books and many other open-source leaders participated.  The
full story can be found here:
http://www.oreilly.com/news/parrotstory_0401.html .  

However, that is not the end of the story.  There wasn't ever a real
language named Parrot that combined the features of Perl and Python. 
However...  there is now a virtual machine in development that should be
able to run both Perl and Python code, as well as other interpreted
languages.  Since the lead developer of this project is the very man who
orchestrated the Parrot-the-programming-language hoax, Simon Cozens, the
obvious name for this virtual machine was Parrot.

	-jc


From philh at comuno.freeserve.co.uk  Sat Feb  9 19:26:54 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Sun, 10 Feb 2002 00:26:54 +0000
Subject: RFC -- HYGIENIC MACROS IN PYTHON
References: <tppu5u48fsmsclucfr5e0qd1musgnqk4vl@4ax.com> <slrna60fur.6sf.philh@comuno.freeserve.co.uk> <75s06ucg70ea2hkf0bk5lrtkgfscbpep6r@4ax.com> <slrna61344.8s1.philh@comuno.freeserve.co.uk> <dtl26uo7rquvbegrhndfak989ug8hpnuvn@4ax.com> <slrna65eh9.ahr.philh@comuno.freeserve.co.uk> <ugb66u8b7j7m5dm45dpm71dj27kqjcupp1@4ax.com> <slrna67k0u.for.philh@comuno.freeserve.co.uk> <br186u8qg3bu2hpsf0dcgs1ldih7gir4rg@4ax.com>
Message-ID: <slrna6bfie.o7n.philh@comuno.freeserve.co.uk>

On Fri, 08 Feb 2002 17:09:51 GMT, Courageous <jkraska at san.rr.com> wrote:
>On Fri, 8 Feb 2002 13:18:22 +0000, philh at comuno.freeserve.co.uk (phil hunt) wrote:
>
>>On Fri, 08 Feb 2002 01:42:51 GMT, Courageous <jkraska at san.rr.com> wrote:
>>>
>>>>It's just that being forced -- "for my own good" -- not to use certain
>>>>features because someone has decreed them intrinsically bad, doesn't
>>>>appeal to me at all.
>>>
>>>Sure. You realize that a hygienic macro doesn't prevent you from
>>>modifying things by consequence, right? All it does is require for
>>>you to mean it; this caters to Python's phenonomenon of least
>>>surprise. Imagine the following
>>>
>>>>>> i = 3
>>>>>> j = 4
>>>>>> mymacro i
>>>>>> print j
>>>5
>>>
>>>That would vex some folks.
>>
>>It would vex me in someone else's code unless there was a good reason for
>>it and it was well documented.
>>
>>> But that's exactly what a non hygienic
>>>macro can do.
>>
>>And it's exactly the behaviour I want a macro to be able to do.
>
>>defmacro incj:
>>   j = j + 1
>
>>In a non-hygenic macro system it is easy to see what underlying Python
>>code gets produced. In your system it isn't -- 
>
>Perhaps only because it hasn't been explained well enough. The only
>way the above macro definition could work without issuing an exception
>is if j already exists in the global namespace. Using current python
>forms, it would simply fail.

So I wouldn't be able to do that macro expansion. This means the macro
processor is conceptually more complex, and therefore more difficult
to learn and probably has more bugs in it too.


-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From loewis at informatik.hu-berlin.de  Thu Feb 28 04:23:14 2002
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Feb 2002 10:23:14 +0100
Subject: Source syntax escapes, new raw string representation. Was: Re: PEP 263 comments
References: <mailman.1014614501.21492.python-list@python.org> <a5jtqh$14l$0@216.39.172.122>
Message-ID: <j4k7sygdtp.fsf@informatik.hu-berlin.de>

bokr at oz.net (Bengt Richter) writes:

> Is this to be a *special purpose* (coding declaration)
> comment-context-limited escape mechanism, or an *open ended* one?

I'm not sure what an "open ended" escape mechanism is. Emacs allows
the definition of arbitrary variables in -*- lines; coding: is just
one of them.

> If special, I think Emacs can just as well adapt to Python as vice
> versa. ISTM ad hoc special escapes in comments are often the
> beginning of a convention for alternative out-of-band info, and
> should be considered in that light. Cf. HTML ;-/

I'm not sure I understand this reference, either. Do you consider the
encoding declaration in the META http-equiv tag as out-of-band, or the
one in the HTTP request? Or are you referring to something completely
different?


> Perhaps it is time to (re?)consider a standard Python mechanism for
> embedding OOB info at arbitrary places in Python source, analogous
> to XML's <![CDATA[ ... ]]> and <? ... ?> (BTW, xml CDATA has a
> HUGE**N wart (IMO) in that it is not nestable.

Processing instructions and CDATA sections are completely different
things. While I could see that processing instructions are some kind
of out-of-band information, CDATA sections are just a certain
well-defined way to present contents; you can easily transform them to
use the standard markup without losing "out-of-band" information -
they are in-band.

> I prefer orthogonal general purpose mechanisms to ad hoc syntactic
> escape warts, so I would prefer <? ... ?> as a vehicle for defining
> encoding. <?py blah blah?> could mean eval('blah blah') in some
> defined python environment

I had initially proposed the directive statement (PEP 244) as a
mechanism to also support encoding declarations; that was
rejected. Tim Peters objected to the introduction of a 'pragma'
statement at many occasions.

> In conjunction with a pythonic CDATA [1], it would permit wrapping a
> whole source file (or starting at the second line):

That sounds like a different technology. If you really want to propose
this, write a PEP.

Regards,
Martin


From kp87 at lycos.com  Fri Feb 15 01:38:38 2002
From: kp87 at lycos.com (kevin parks)
Date: 14 Feb 2002 22:38:38 -0800
Subject: sorting list of tuples by second (third...) tuple item
References: <3c6bbe96.12195968@news.easynews.net> <mailman.1013718563.21631.python-list@python.org>
Message-ID: <5e8bd451.0202142238.580f15c1@posting.google.com>

How would you sort something based on 2 criteria. Say you have a list
of lists (i hate tuples). first thing you want is everything sorted by
the first value and then everything that has the same value for [0]
then sorted by [1]



[['i3', 24.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i8', 82.25, 0.75, 101, 7, 6, 5, 0, 0, 0, 0, 0.02, 2, \
 ['i9', 9, 0, 255, \
 ['i8', 83.0, 1.0, 101, 1, 1, 5, 0, 0, 0, 0, 0.02, 2, \
 ['i3', 17.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i3', 25.5, 0.5, 101, 4, 3, 7, 0, 0, 0, 1, 0, 5, \
 ['i6', 2.0, 0.5, 85, 7, 4, 6, 0, 0, 0, -1, 0.02, 2, \
 ['i3', 13.5, 0.5, 101, 4, 3, 7, 0, 0, 0, 1, 0, 5, \
 ['i4', 5.5, 6.5996, 101, 7, 4, 5, 7, 4, 5, 1, 0, 50, 50, 6.5, 6.5,
0.08000002, \
 ['i9', 18.0, 6.0, 100, 1, 1, 3, 0, 0, 0, 0, 0, 5, \
 ['i2', 222.0, 16.0, 1.100001, 1.1699, 0.56995, 0.7800003, \
 ['i3', 26.5, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i9', 16.0, 0.5, 50, 7, 4, 3, 0, 0, 0, 0, 0, 5, \
 ['i6', 1.0, 0.5, 85, 1, 1, 7, 0, 0, 0, -1, 0.02, 2, \
 ['i3', 28.0, 0.5, 101, 4, 3, 7, 0, 0, 0, 1, 0, 5, \
 ['i9', 24.0, 1.5, 50, 1, 1, 4, 0, 0, 0, 0, 0, 5, \
 ['i7', 0.5, 0.5, 101, 1, 1, 6, 0, 0, 0, 1, 0.02, 2, \
 ['i6', 0.5, 0.5, 85, 7, 4, 6, 0, 0, 0, -1, 0.02, 2, \
 ['i8', 81.5, 0.75, 101, 7, 4, 4, 0, 0, 0, 0, 0.02, 2, \
 ['i6', 1.5, 0.5, 85, 7, 6, 7, 0, 0, 0, -1, 0.02, 2, \
 ['i9', 14.5, 1.5, 50, 1, 1, 4, 0, 0, 0, 0, 0, 5, \
 ['i9', 17.0, 1.0, 50, 1, 1, 4, 0, 0, 0, 0, 0, 5, \
 ['i3', 29.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i3', 12.0, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i3', 16.0, 0.5, 101, 4, 3, 7, 0, 0, 0, 1, 0, 5, \
 ['i1', 0, 0, 18.715278000000001, \
 ['i3', 14.5, 0.5, 101, 7, 6, 7, 0, 0, 0, -1, 0, 5, \
 ['i2', 181.4001, 0.1000001, 0.62, 0.63, 1.100001, 1.1699]]


so you would want the above list to have all the i1 lines together and
all the lines that start i2 together and then all the lines that have
the same starting value sorted by their second value, so that the
above list would start to sort perhaps like  this:


x = [['i1',0,0,18.715278], ['i2',181.4,0.1,0.62,0.63,1.10,1.17],
['i2',222.0,16.0,1.10,1.17,0.57,0.78],
['i3',12.000,0.500,101,7,6,7,0,0,0,-1,0,5],
['i3',13.500,0.500,101,4,3,7,0,0,0,1,0,5],
['i3',14.500,0.500,101,7,6,7,0,0,0,-1,0,5],
['i3',16.000,0.500,101,4,3,7,0,0,0,1,0,5],
['i3',17.000,0.500,101,7,6,7,0,0,0,-1,0,5],
['i3',24.000,0.500,101,7,6,7,0,0,0,-1,0,5],
['i3',25.500,0.500,101,4,3,7,0,0,0,1,0,5],
['i3',26.500,0.500,101,7,6,7,0,0,0,-1,0,5],
['i3',28.000,0.500,101,4,3,7,0,0,0,1,0,5],

.
.
.


? I have a stupid audio program that has a broken sort routine and
barfs if things are not presorted this way, even though it is supposed
to sort everything for you..

one odd thing to note is that the nested lists are not all the same
size.

inquiring minds want to know.

bets,

kevin


From eeskoe at yahoo.com  Sat Feb 16 18:38:18 2002
From: eeskoe at yahoo.com (Erika Skoe)
Date: Sat, 16 Feb 2002 17:38:18 -0600
Subject: Apache windows server : python module
References: <a4m865$80v$1@wanadoo.fr>
Message-ID: <a4mq7d$th2$1@news.doit.wisc.edu>

it looks lke mod_python itself isn't installed.  you'll get better answers,
however, from the mod_python mailing list.


"Aliocha" <aliocha_z at caramail.com> wrote in message
news:a4m865$80v$1 at wanadoo.fr...
> Hi,
>
> I have followed all indications to install python module for apache on the
> web page :
> (http://www.modpython.org/live/mod_python-2.7.6/doc-html/app-wininst.html)
> but when I start the server, this are the error lines :
>
> "Syntax error on line 631 of c:/program files/apache
> group/apache/conf/httpd:
> Invalid command 'PythonHandler', perhaps mis-spelled or defined by a
module
> included in the server configuration
> Note the errors or messages above, and press the <ESC> key to exit.
20..."
>
> Somebody have already met this problem ?
>
> Precisions :
> --------------------------------------------------------------------------
--
>
> The line which doesn't work in httpd.conf :
> It's the line "PythonHandler mptest" which redirect to "mptest.py".
> This file is in my "C:/Program Files/Apache Group/Apache/htdocs/python" .
> this file must print "Hello World"  in my browser.
>
>    <Directory "C:/Program Files/Apache Group/Apache/htdocs/python">
>     AddHandler python-program .py
>     PythonHandler mptest
>     PythonDebug on
>    </Directory>
>
> --------------------------------------------------------------------------
--
>
> Content of mptest.py :
>
> from mod_python import apache
>
> def handler(req):
>    req.content_type = "text/plain"
>    req.send_http_header()
>    req.write("Hello World!")
>    return apache.OK
>
> --------------------------------------------------------------------------
--
>
> Configuration :
>
> I have installed :
> - Apache : apache_1.3.22-win32-x86.exe
> - the python module files : mod_python-2.7.6.tgz
> - The dll of python module :  python2.2-apache1.3.22
> - Python executable : Python-2.2.exe
>
>
>




From synthespian at uol.com.br  Wed Feb 13 19:52:18 2002
From: synthespian at uol.com.br (synthespian)
Date: 13 Feb 2002 22:52:18 -0200
Subject: Opening random line from file with random.shuffle()
Message-ID: <87bses6g1p.fsf@uol.com.br>


Hello-

	I need to get a random line from a file. I read in thins NG about a "poorly" documented random.shuffle(), but I can't get it to work:

	


#!/usr/local/bin/python2.0

import random
file = open('/xxxx/yyy/zzz.txt')

allLines = file.readlines()

file.close()

getLine = random.shuffle(allLines)

print getLine,
string = str(getLine)

print string,


	All I get as output is
	"None None"

	Thanks for __any__ help!

	Regs
	Henry






----------------------
_________   1001101010
| ^   ^	|  /1111100110
| o   o | / 0010100111
|   v	|/  1110100111
| ----- |	
\_______/

______________________
Micro$oft-Free Human
100% Debian GNU/Linux 
----------------------
	KMFMS



From weeks at vitus.scs.agilent.com  Fri Feb  8 22:27:34 2002
From: weeks at vitus.scs.agilent.com (Greg Weeks)
Date: Sat, 9 Feb 2002 03:27:34 +0000 (UTC)
Subject: IOError: [Errno 4] Interrupted system call
Message-ID: <1013225254.610747@cswreg.cos.agilent.com>

In Python 1.5.2, I'm getting the message

    IOError: [Errno 4] Interrupted system call

in a context that doesn't make sense to me.  My program has two threads
(which won't appear useful out of context):

Thread 1: Fork a subprocess, wait for the subprocess to exit, and set
an Event.

Thread 2: Send a SIGTERM signal to the subprocess and wait for the Event to
be set.

I'm getting the error message from Thread 2 while waiting for the event.
The relevant portions of the stack trace are:

  <In my code: self.event.wait(10)>
  File "/opt/Python-1.5.2/Lib/threading.py", line 298, in wait
    self.__cond.wait(timeout)
  File "/opt/Python-1.5.2/Lib/threading.py", line 199, in wait
    _sleep(delay)

_sleep is just time.sleep.

What is strange about this is that the Python process should not be
receiving any signals.  Only the subprocess receives a signal.  So I
can't imagine what might be going on.  Any ideas?


Greg


PS: It is guaranteed that the fork has returned before SIGTERM is sent.

PS: The error is intermittent and occurs only if SIGTERM is sent very soon
after the fork has returned.


From news at chriswysocki.com  Wed Feb 13 02:03:14 2002
From: news at chriswysocki.com (Chris Wysocki)
Date: Wed, 13 Feb 2002 07:03:14 GMT
Subject: Python 2.2 <-> Solaris 2.6 installation errors
References: <mailman.1013548101.10897.python-list@python.org>
Message-ID: <i63k6uovaa07umml1b5bs28m88h8ffsira@4ax.com>

I'm having the exact same problem with Python 2.2 on Solaris 7 & 8
only in my case the problem seems to be with the struct module
(possibly others if I can get past the struct module).
After the 'make' and before running 'make install' I can start Python
and import the struct module and it appears fine.  Make test also
seems fine.  However when I run 'make install' I get the same error as
Jeff.  When I try to run Python after the failure the struct  module
is no longer available.

I'm using gcc 3.0.3 and all other required packages are the current
packages from Sunfreeware.

Didn't have any of these problems building previous versions of Python
on the same machines.

Any ideas, things to try?

Chris

On Tue, 12 Feb 2002 14:54:32 -0600, Jeff Bauer <jbauer at rubic.com>
wrote:

>I'm having problems installing Python 2.2 onto
>my Solaris 2.6 workstation.
>
>I am doing the boilerplate ...
>
>  ./configure
>  make
>  make install
>
>... but it bombs, possibly because distutils
>is looking for something that isn't there?
>
>Earlier versions of Python install fine on
>this machine.
>
>-Jeff
>
>
>skipping 'rgbimg' extension (up-to-date)
>skipping 'readline' extension (up-to-date)
>WARNING: removing "readline" since importing it failed
>Traceback (most recent call last):
>  File "./setup.py", line 793, in ?
>    main()
>  File "./setup.py", line 787, in main
>    scripts = ['Tools/scripts/pydoc']
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/core.py", 
>    line 138, in setup
>    dist.run_commands()
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
>    line 893, in run_commands
>    self.run_command(cmd)
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
>    line 913, in run_command
>    cmd_obj.run()
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/command/build.py", 
>    line 107, in run
>    self.run_command(cmd_name)
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/cmd.py", 
>    line 330, in run_command
>    self.distribution.run_command(command)
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/dist.py", 
>    line 913, in run_command
>    cmd_obj.run()
>  File
>"/export/home/jbauer/Python-2.2/Lib/distutils/command/build_ext.py", 
>    line 256, in run
>    self.build_extensions()
>  File "./setup.py", line 147, in build_extensions
>    build_ext.build_extensions(self)
>  File
>"/export/home/jbauer/Python-2.2/Lib/distutils/command/build_ext.py", 
>    line 383, in build_extensions
>    self.build_extension(ext)
>  File "./setup.py", line 179, in build_extension
>    for filename in self._built_objects:
>  File "/export/home/jbauer/Python-2.2/Lib/distutils/cmd.py", 
>    line 107, in __getattr__
>    raise AttributeError, attr
>AttributeError: _built_objects
>make: *** [sharedmods] Error 1



From skip at pobox.com  Tue Feb 26 12:49:20 2002
From: skip at pobox.com (Skip Montanaro)
Date: Tue, 26 Feb 2002 11:49:20 -0600
Subject: Can I import an arbitary file?
In-Reply-To: <9d5509fa.0202251811.11837b04@posting.google.com>
References: <9d5509fa.0202251811.11837b04@posting.google.com>
Message-ID: <15483.51872.120178.309101@12-248-41-177.client.attbi.com>

    Peter> What I want to do is import this file. The problem is, this file
    Peter> could be any name at all. What I want to do is parse my command
    Peter> line, and find out the name of my configuration file, and then
    Peter> import the file of that name. Is that possible?

Take a look at the execfile builtin function.  The import statement requires
the file have a basename that is a valid Python identifier, but execfile
doesn't.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From lac at strakt.com  Mon Feb  4 11:38:40 2002
From: lac at strakt.com (Laura Creighton)
Date: Mon, 04 Feb 2002 17:38:40 +0100
Subject: Question for the Python Community! 
In-Reply-To: Message from rony.steelandt@bucodi.com (Rony) 
   of "04 Feb 2002 08:02:43 PST." <b34d3835.0202040802.2cc7d0c0@posting.google.com> 
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>  <b34d3835.0202040802.2cc7d0c0@posting.google.com> 
Message-ID: <200202041638.g14GceGo011642@ratthing-b246.strakt.com>

The e-book I want is made of electronic reusable paper.
See: http://www.parc.xerox.com/dhl/projects/gyricon

Laura Creighton




From wall at adinet.com.uy  Tue Feb 12 10:35:49 2002
From: wall at adinet.com.uy (Walter Moreira)
Date: Tue, 12 Feb 2002 12:35:49 -0300
Subject: Problem building Python 2.2 docs
In-Reply-To: <7xeljrcz1o.fsf@ruckus.brouhaha.com>; from phr-n2002a@nightsong.com on Mon, Feb 11, 2002 at 04:40:51PM -0800
References: <7xeljrcz1o.fsf@ruckus.brouhaha.com>
Message-ID: <20020212123549.A1904@casa.parque>

On Mon, Feb 11, 2002 at 04:40:51PM -0800, Paul Rubin wrote:
> I just loaded a fresh Red Hat GNU/Linux 7.2 system on a newly
> formatted hard disc, and then installed Python 2.2 on it from the .tgz
> file.  Python 2.2 itself built and installed just fine.  When I typed
> "make" in the Doc directory though, a bunch of errors resulted, like:
> 
>      *** processing declarations ***
> 
>     Loading /usr/share/latex2html/versions/latin1.pl
> 
>      *** Fatal Error --- but easy to fix ***
> 
>     Cannot have '.' in file-name prefix, else dvips fails on images
>     Change the name from  /home/phr/Python-2.2/Doc/lib/lib.tex  and try again.
> 
>     *** Session transcript and error messages are in /home/phr/Python-2.2/Doc/html/lib/lib.how.
>     *** Exited with status 255.
> 

I could fix this by renaming the directory Python-2.2 to Python-2_2. Anyway, I
still cannot build the texinfo files. I'm waiting for the day that building
the docs doesn't need p*rl.

Walter



From scarblac at pino.selwerd.nl  Sun Feb 24 11:45:25 2002
From: scarblac at pino.selwerd.nl (Remco Gerlich)
Date: 24 Feb 2002 16:45:25 GMT
Subject: What values are considered false?
References: <e9e6dd1f.0202201727.9741615@posting.google.com> <3C746504.C2C70DF0@alcyone.com> <roy-630F36.23110820022002@reader1.panix.com> <roy-C72B26.23204520022002@reader1.panix.com> <3C751FEA.24CEFBBC@ford.com>
Message-ID: <slrna7i5v6.7kb.scarblac@pino.selwerd.nl>

Jeff Henry <jhenry19 at ford.com> wrote in comp.lang.python:
> Roy Smith wrote:
> > 
> > Roy Smith <roy at panix.com> wrote:
> > > PS -- In the lists of false things in python that have been mentioned on
> > > this thread, I don't think I've seen anybody mention {}, the empty
> > > dictionary.
> > 
> > Along those same lines, I half-expected the empty function to be false,
> > i.e.:
> > 
> > def foo():
> >    pass
> > 
> > if (foo):
> 
> Oops! I think you meant to say -
> 
> if (foo()):

No, he didn't. He thought the *function* should be false, not the function's
result. An interesting idea, but possibly hard to implement (write code to
decide if a function does nothing). Also, I haven't been able to come up
with any situation where it would be useful yet :-)

-- 
Remco Gerlich


From fredrik at pythonware.com  Wed Feb 13 15:00:38 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 20:00:38 GMT
Subject: redirect console output to widget?
References: <de6c123c.0202101612.72b27f8d@posting.google.com> <mailman.1013395757.3817.python-list@python.org> <de6c123c.0202131109.2708c2f4@posting.google.com>
Message-ID: <GBza8.25587$l93.4764652@newsb.telia.net>

Lei Chen wrote:
> I think you are right.  Is there an alternative to Tkinter that is
> thread safe?

Tkinter?

just run all UI code in the main thread, and let the writers
write to a Queue object; e.g.

from Tkinter import *

import thread # should use the threading module instead!
import Queue

import os

class ThreadSafeConsole(Text):
    def __init__(self, master, **options):
        Text.__init__(self, master, **options)
        self.queue = Queue.Queue()
        self.update_me()
    def write(self, line):
        self.queue.put(line)
    def clear(self):
        self.queue.put(None)
    def update_me(self):
        try:
            while 1:
                line = self.queue.get_nowait()
                if line is None:
                    widget.delete(1.0, END)
                else:
                    widget.insert(END, str(line))
                widget.see(END)
                widget.update_idletasks()
        except Queue.Empty:
            pass
        self.after(100, self.update_me)

#this function pipes input to an widget
def pipeToWidget(input, widget):
    widget.clear()
    while 1:
        line = input.readline()
        if not line:
            break
        widget.write(line)

def funcThread(widget):
    input = os.popen('dir', 'r')
    pipeToWidget(input, widget)

#uber-main
root = Tk()
widget = ThreadSafeConsole(root)
widget.pack(side=TOP, expand=YES, fill=BOTH)
thread.start_new(funcThread, (widget,))
thread.start_new(funcThread, (widget,))
thread.start_new(funcThread, (widget,))
thread.start_new(funcThread, (widget,))
thread.start_new(funcThread, (widget,))
root.mainloop()

</F>

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




From bryan at eevolved.com  Fri Feb  1 02:10:53 2002
From: bryan at eevolved.com (Bryan)
Date: Fri, 01 Feb 2002 02:10:53 -0500
Subject: Java vs Python Benchmarks: Java is faster
References: <Ldc68.67$8u4.28600@wagner.videotron.net> <1012540501.71692@yasure>
Message-ID: <Mdr68.1546$8u4.449344@wagner.videotron.net>

Donn Cave wrote:

> | - Interpreter intialization time has doubled in 2.2
> 
> Doubled!!?  Could that be some kind of site glitch?

It is possible, but I doubt it. I ran the tests ten times at various points 
during my day and consistently saw results that indicated that 2.2 takes 
twice as long to initialize the interpreter. Perform your own tests: 
perhaps I am overlooking something on my end.
-- 
Bryan Mongeau
http://www.eevolved.com/bryan
--
"A person starts to live when he can live outside himself."-- Einstein



From timr at probo.com  Tue Feb 19 00:23:24 2002
From: timr at probo.com (Tim Roberts)
Date: Mon, 18 Feb 2002 21:23:24 -0800
Subject: Operating system requirements
References: <3C71A556.3E7BF27E@xatt.netx>
Message-ID: <f5o37ugcp2dnidths56jkuk4cgdclr0oe7@4ax.com>

John Burke <jnxburxke at xatt.netx> wrote:

>What are the operating system requirements for python to run on a Windows
>pc. Thank you.

I think you just said it: a Windows PC.  It'll run in 95, 98, SE, ME, NT,
2K, or XP.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From risnes at alfanett.no  Fri Feb 15 16:44:07 2002
From: risnes at alfanett.no (André Larsen Risnes)
Date: Fri, 15 Feb 2002 21:44:07 GMT
Subject: entries between labels dynamically
References: <slrna6qqs9.b35.arturs@aph.waw.pdi.net>
	<20020215225431.7bdef34e.risnes@alfanett.no>
Message-ID: <20020215225632.2c84e41c.risnes@alfanett.no>

>         Label(f, ...)
          ^^^^^^^^^^^^^ Oops. Remove this 


From bokr at oz.net  Sun Feb 24 16:21:55 2002
From: bokr at oz.net (Bengt Richter)
Date: 24 Feb 2002 21:21:55 GMT
Subject: "pass by reference?"
References: <mailman.1014452738.24147.python-list@python.org> <3C7822E1.508@thinkware.se>
Message-ID: <a5blhj$k6n$0@216.39.172.122>

On Sun, 24 Feb 2002 00:16:49 +0100, Magnus =?ISO-8859-1?Q?Lyck=E5?= <magnus at thinkware.se> wrote:

[... an epistle covering the subject ...]

which I didn't see until now. I wonder if that is due to the difference between

Path: news.alt.net!news-out.cwix.com!newsfeed.cwix.com!opentransit.net!news.tele.dk!small.news.tele.dk!195.54.122.107!newsfeed1.bredband.com!bredband!newsspool1.bredband.com!news1.bredband.com.POSTED!not-for-mail
M

and

Path: news.alt.net!oz.net!usenet

Sorry for duplicating examples as if your post didn't exist (for me it didn't yet,
though sometimes I overlook something).

I could just have added a comment about __getval__, __update__, and ':='
(see other post of mine in this thread).

Regards,
Bengt Richter



From jbauer at rubic.com  Wed Feb 27 10:04:24 2002
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 27 Feb 2002 09:04:24 -0600
Subject: telnetlib...
Message-ID: <3C7CF578.F96584FB@rubic.com>

Vincent,

I've posted an example of using telnetlib on the ActiveState
Python Cookbook.

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52228

Jeff Bauer
Rubicon Research


Vincent A. Primavera wrote:
> Does anybody out there have any good telnetlib 
> examples they would be willing to share?



From mpoulin at verinet.com  Fri Feb  1 11:16:03 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Fri, 01 Feb 2002 11:16:03 -0500
Subject: Windows, RS232, and Python
References: <Pine.GSU.4.21.0202010910040.11090-100000@garcia.efn.org>
Message-ID: <3c5ada50$0$2204$7586b60c@news.frii.com>

In article <Pine.GSU.4.21.0202010910040.11090-100000 at garcia.efn.org>,
"Spencer Ernest Doidge" <spencer at efn.org> wrote:

> I want to have control of every aspect of RS232 transmitting and
> receiving in Win9x/NT with Python commands.
> 

Does this help? 
http://balder.prohosting.com/ibarona/en/python/uspp/uspp_en.html


From skip at pobox.com  Wed Feb 27 19:07:23 2002
From: skip at pobox.com (Skip Montanaro)
Date: Wed, 27 Feb 2002 18:07:23 -0600
Subject: regular expression help
In-Reply-To: <1014853112.982.15.camel@beast.l3.com>
References: <1014853112.982.15.camel@beast.l3.com>
Message-ID: <15485.29883.206252.555069@12-248-41-177.client.attbi.com>

    Drew> I'm having some issues doing a small but irritating regular
    Drew> expression match.

As Jamie Zawinski said probably more than once on comp.lang.emacs (and
Fredrik Lundh has reminded us here):

    Some people, when confronted with a problem, think "I know, I'll use
    regular expressions".  Now they have two problems.

So, why use regular expressions?  If your data is formatted as simply as you
suggest, something like this would work:

lines = """
Location:Home
        555-1212
        555-3434
        555-5656

Location:Work
        555-9999
""".split("\n")

phones = {}
for line in lines:
    if line.startswith("Location:"):
        where = line.split(":")[1]
        phones[where] = []
    elif line.startswith("\t"):
        phones[where].append(line.strip())
print phones

I'm sure there are cleaner solutions, but this does seem to collect phone
numbers together according to their location.

-- 
Skip Montanaro (skip at pobox.com - http://www.mojam.com/)



From ryan.phillips at csus.edu  Tue Feb 26 14:14:55 2002
From: ryan.phillips at csus.edu (Ryan Phillips)
Date: 26 Feb 2002 19:14:55 GMT
Subject: Coredumps with Python 2.1.1-3
References: <20020226.185425.1884661237.5746@ic.ac.uk>
Message-ID: <a5gmrf$n2o@dispatch.concentric.net>

"Phil Mayers" <p.mayers at ic.ac.uk> wrote in message
news:20020226.185425.1884661237.5746 at ic.ac.uk...
> I'm using the RPM for Python 2.1.1-3 from the Python website, and have
> just had a *really* weird problem. I was editing some code, and did a
> test run. It all worked. I turned to chat to a colleague again - turned
> back a few minutes later, scrolled one line up in the shell history, BOOM
> core dump with the following traceback:
>
> #0  0x806f9e1 in Py_GetProgramFullPath ()
> #1  0x80928d2 in PyTuple_GetSlice ()
> #2  0x806fa12 in Py_GetProgramFullPath ()
> #3  0x806fe0c in Py_GetProgramFullPath ()
> #4  0x807009a in Py_GetProgramFullPath ()
> #5  0x80700e9 in _PyGC_Insert ()
*snip*


Perhaps the problem is with the libc version.  Have you tried to get the
source and recompile python yourself?

-ryan




From phr-n2002a at nightsong.com  Sat Feb 23 06:36:04 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 23 Feb 2002 03:36:04 -0800
Subject: strings and sort()
References: <mailman.1014257380.19140.python-list@python.org> <7xsn7vbixg.fsf@ruckus.brouhaha.com> <m2vgco1mun.fsf@mycroft.actrix.gen.nz>
Message-ID: <7xn0y0fn17.fsf@ruckus.brouhaha.com>

Paul Foley <see at below> writes:
> >    (concat a b)   ; copy and concatenate two lists, like "a+b" in Python
> >    (nconcat a b)  ; concatenate in place, like "a.extend(b)" returning a
> 
> There's no "concat" in Common Lisp.  APPEND concatenates two (or more)
> lists [though it doesn't guarantee to copy the last one; it's possible
> for Lisp lists to share tail structure], but the destructive version
> of APPEND is called NCONC, not NAPPEND (there's no CONC, either).

Sorry, you're right, I met nconc and append.  It's been a while.


From wayne at mishre.com  Wed Feb 13 11:03:49 2002
From: wayne at mishre.com (Wayne Pierce)
Date: Wed, 13 Feb 2002 08:03:49 -0800
Subject: newbie asking for info about Python
References: <xkka8.2932$VX6.245554@weber.videotron.net>
Message-ID: <3C6A8E65.20701@mishre.com>

I've written the usual CGI stuff, done some client-side Windows 
development for end user and servers.  I wrote a program that secured 
Windows-based web servers in Python, a lot of shell scripting and 
parsing data.   I had a system of about 10K lines that would launch 
attacks at a network then parse the output and display the results to 
the end user.

I've also done some Java stuff with Jython, I'm currently working on a 
supply managment tool and a sales tool.

Joseph Youssef wrote:
> Hello, I know nothing about Python but before I start learning I would 
> like to know what Python is usually "used" for, I mean I don't think 
> it's made to write some kinda application or something, so what exaclty 
> is it used for?
> 
>  
> 
>  
> 


-- 
Wayne Pierce
web: http://www.mishre.com
email: wayne at mishre.com

"What you need to know."




From mal at lemburg.com  Tue Feb 26 15:40:03 2002
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 26 Feb 2002 21:40:03 +0100
Subject: Python and Oracle 9i
References: <b470cb58.0202250816.7839c1c4@posting.google.com> <Xns91C167290C907andy47halfcookedcom@203.109.252.31> <b470cb58.0202260701.53857ce0@posting.google.com>
Message-ID: <3C7BF2A3.34FB5EAC@lemburg.com>

Bob Lancaster wrote:
> 
> Unfortunately, many of our customers use Oracle, and we have to
> interface with a system which uses Oracle, so I probably have no other
> choice but to rewrite the database in such a way as to make DCOracle2
> happy. 

...or stick with mxODBC and use a decent ODBC driver for Oracle
like the ones from EasySoft, OpenLink or Data-Direct.

They cost money, but they also take the load off of you to write
a new Oracle interface each time Oracle bumps the patch level
of their DB ;-)

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/



From nhodgson at bigpond.net.au  Sun Feb  3 21:09:13 2002
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Mon, 04 Feb 2002 02:09:13 GMT
Subject: SilverCity 0.1.0
References: <mailman.1012764187.14672.python-list@python.org>
Message-ID: <d3m78.34171$Ni2.194073@news-server.bigpond.net.au>

Brian Quinlan:
> I have a Scintilla features request though:
> - expose the name of the lexer in the LexerModule class (it'd like to
> include the lexer's name in __repr__ and exceptions)

   Assuming you just mean moving it from private to public, OK, I'll change
that after 1.44 is released.

> What did you think about the higher-level API? i.e. lexing to events,
> the HTML generators, etc?

   Not my preferred way of working. I generally prefer writing navigational
code rather than  visitors or event handlers. Other people really like these
styles of interface and they are certainly popular for XML.

   Neil




From tim at vegeta.ath.cx  Sat Feb  2 20:40:12 2002
From: tim at vegeta.ath.cx (Tim Hammerquist)
Date: Sun, 03 Feb 2002 01:40:12 GMT
Subject: Guido's intuition (was Re: typetesting, adaptation, typeclasses, ...)
References: <yu99wuxv7n7r.fsf@europa.research.att.com> <mailman.1012678565.11760.python-list@python.org>
Message-ID: <slrna5p5ba.g91.tim@vegeta.ath.cx>

Tim Peters <tim.one at home.com> graced us by uttering:
> [Andrew Koenig]
>> It seems to be universal among programmers that they get hung up about
>> tiny syntax details until they get used to those details.  For example,
>> C and C++ programmers still argue about where to put the curly braces
>> in their programs.
> 
> That's because they don't have Guido to force them to do it the right way:
> 
> 	if (guido) {
> 		tab(indent);
> 	}

I was required to use BSD-style for school, but the "wasted" vertical
space offended my senses, so I settled on K&R. Never could understand
GNU or Whitesmith style, though. =)

BTW, less wasted v-space is one very attractive feature of Python.

My eyes just can't parse:

if ( some_condition() ) {
  {
    for (i = first_i();i <= 10; i++)
      {
        initialize_some_stuff();
        while ( some_other_condition() )
          {
            do_some_more_stuff()
          }
      }
  }

...yet I have no trouble parsing

if ( &some_condition ) {
    for $i (&first_i .. 10) {
        &initialize_some_stuff;
        &do_some_more_stuff while &some_other_condition;
    }
}

Imagine a "Lisp-style" indent!

if ( some_condition() ) {
    i = first_i();
    for (i = first_i(); i <= 10; i++) {
        initialize_some_stuff();
        while ( some_other_condition() ) {
            do_some_more_stuff() } } }

<Real-programmers-use-1TBS>-ly y'rs,
Tim Hammerquist
-- 
Note that by displacing from "I got confused" to "It got confused", the
programmer is not avoiding responsibility, but rather getting some analytical
distance in order to be able to consider the bug dispassionately.
    -- Jargon File 4.3.1


From cliechti at gmx.net  Mon Feb 25 11:49:24 2002
From: cliechti at gmx.net (Chris Liechti)
Date: 25 Feb 2002 17:49:24 +0100
Subject: Editors for Python
References: <d906b2c1.0202250818.57dcfcc7@posting.google.com>
Message-ID: <Xns91C0B5E261A53cliechtigmxnet@62.2.16.82>

mtaylor at lowebroadway.com (Moray Taylor) wrote in 
news:d906b2c1.0202250818.57dcfcc7 at posting.google.com:
> I am desperatly seeking a good python editor, so far JEdit is the 

i like SciTE (for linux/win32) which you can find under 
http://www.scintilla.org
(i use it as general editor and its also the editor component of 
pythonwin which i use for python programming )

-- 
Chris <cliechti at gmx.net>



From qvyht at removejippii.fi  Mon Feb 25 05:14:30 2002
From: qvyht at removejippii.fi (vekkuli ketkutin)
Date: Mon, 25 Feb 2002 10:14:30 GMT
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <3C794EEE.1F25B623@engcorp.com>
Message-ID: <3C7A0F28.5090505@removejippii.fi>


Peter Hansen wrote:

> Russell Nelson wrote:
> 
>>Henrik Motakef <henrik.motakef at web.de> writes:
>>
>>
>>>Gerhard H?ring <gh_pythonlist at gmx.de> writes:
>>>
>>>
>>>>Le 18/02/02 ? 15:28, Tim Peters ?crivit:
>>>>
>>>>>Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
>>>>>czignat?  Classifero en bragna!
>>>>>
>>>>Woas da Deifi wos des hoasn mog.
>>>>
>>>Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
>>>
>>Even better: we could switch languages within threads, to *really*
>>keep the monolinguists confused.
>>
> 
> Sed kiujn lingvojn ni devos uzi?
> 

El?pp?s ny pua h?p?tt?en esperanttuo, vuan uan 
liijan vaikkeeta miulle!



From QnickQm at alum.mit.edu  Thu Feb 28 10:27:12 2002
From: QnickQm at alum.mit.edu (Nick Mathewson)
Date: Thu, 28 Feb 2002 15:27:12 GMT
Subject: Peer review: Python encyphering script
References: <3c7c1a91.150875561@menuhin.netfront.net> <slrna7rdna.5m1.QnickQm@localhost.localdomain> <a5lfoo$38q$1@animus.fel.iae.nl>
Message-ID: <slrna7sj8i.8fp.QnickQm@localhost.localdomain>

In article <a5lfoo$38q$1 at animus.fel.iae.nl>, Carel Fellinger wrote:
> Nick Mathewson <QnickQm at alum.mit.edu> wrote:
>> In article <3c7c1a91.150875561 at menuhin.netfront.net>, A. Jones wrote:
> ...
>>>             if type(blurb[z]) == "<type 'list'>": #Stops nesting
>>> strings
> 
>> #It's bad style to rely in the string value of a type.  Intstead, you
> 
> Not only is it bad style, it does do what he thinks it does:)

Yeah, so it seems.  I should have resisted my "assume it somehow
works" instincts and checked whether type([]) had an __eq__ method
that did string coercion.

Apparently not; Python is not C++.

And-what-a-good-thing-that-is-ly Y'rs,
						Nick


From camarao at dcc.ufmg.br  Thu Feb 28 17:27:47 2002
From: camarao at dcc.ufmg.br (Carlos =?iso-8859-1?Q?Camar=E3o?=)
Date: Thu, 28 Feb 2002 18:27:47 -0400
Subject: SBLP'2002 deadline extension
Message-ID: <3C7EAEE3.37E0C795@dcc.ufmg.br>

The deadline for submissions to SBLP'2002 has been extended to 
March 7, 2002.

The full updated call for papers is included below. 

===============================================
                 SBLP'2002
VI Brazilian Symposium on Programming Languages

       http://www.inf.puc-rio.br/sblp

        Departamento de Inform?tica
                 PUC-Rio

         Rio de Janeiro, RJ, Brazil
             June 05-07, 2002
===============================================

Final Call for Papers
---------------------

The VI Brazilian Symposium on Programming Languages (SBLP'2002) will
be held at Pontif?cia Universidade Cat?lica do Rio de Janeiro
(PUC-Rio) in Rio de Janeiro, Brazil, on June 05-07, 2002. The
symposium is sponsored by the Brazilian Computer Society (SBC) andg
aims to provide a forum where researchers, educators, and
practitioners present and discuss the latest theoretical and practical
work on programming languages.

We invite authors to contribute to SBLP'2002 with the following
kinds of submission:

   * Technical research papers introducing original research results
     on topics such as (but not limited to):

        - Programming language design and implementation
        - Formal semantics of programming languages
        - Domain-specific programming languages
        - Visual programming languages
        - Programming languages for mobile, WWW, and network computing
        - Script languages
        - Design and implementation of programming tools and
          environments
        - New programming models
        - Theoretical foundations of programming languages 
        - Teaching programming languages

   * Experience reports on the use of programming languages, tools,
     and environments. The reports should clearly state the problems
     and difficulties identified on using those programming languages
     or technologies on projects, possibly giving solutions to those
     problems or hints to avoid them.

   * Tutorial proposals on subjects related to programming languages, 
     APIs, tools, environments or theories.

There will be also invited lectures, tutorials, panels, and working
meetings to closely group together professionals interested on
specific research topics.


Important dates
---------------

Deadline for submissions: March 7, 2002
Notification of acceptance: April 8, 2002
Camera-ready version: May 15, 2002

Conference chair
----------------

Edward Hermann Haeusler
Departamento de Inform?tica
Pontif?cia Universidade Cat?lica do Rio de Janeiro
Tel.: +55 (21) 3114-1500 R: 4345
email: hermann at inf.puc-rio.br

Program Chair
-------------

Carlos Camar?o 
Departamento de Ci?ncia da Computa??o
Universidade Federal de Minas Gerais
Tel: +55 (31) 3499 5889
email: camarao at dcc.ufmg.br

Submission details
-------------------

Contributions can be written in Portuguese, English or Spanish.
Those written in English can be selected for publication elsewhere
(probably Elsevier Electronic Notes in Theoretical Computer Science).

All papers and experience reports will be published in the conference
proceedings. Papers should have at most 14 pages and experience
reports at most 8 pages.

----
Tutorials will be published in a separate proceedings. They should
cover material of postgraduate level. The tutorial submission may be
only a proposal, containing:
  - title
  - author names
  - motivation, describing the relevance of the subject in the area
    of programming languages
  - abstract
  - description of the presentation (optional)

Tutorial proposals will be selected according to i) its technical
quality and ii) expected relevance and interest from the community.

The final tutorial should contain at most 30 pages.
----

Contributions should use the standard format of SBC (Brazilian
Computer Society). They are available in: 

  - LaTeX: http://www.sbc.org.br/templates/sbc-template-latex.zip
  - PDF/Adobe Acrobat:
http://www.sbc.org.br/templates/sbc-template-pdf.zip
  - Word: http://www.sbc.org.br/templates/sbc-template-dot.zip

Use of LaTeX is preferred. Examples of the use of the SBC format can
be found in these zip files. (You can also obtain them by visiting the
site http://www.sbc.org.br and following the links *eventos* and,
afterwards, *Informa??es ?teis*.)

For submission of contributions, fill out the required information at
http://www.dcc.ufmg.br/~camarao/sblp/forms/authpaper_reg.html
======================================================================


From db3l at fitlinxx.com  Mon Feb 18 19:10:49 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Feb 2002 19:10:49 -0500
Subject: distutils/package installers on windows
References: <6bd9f01b.0202160255.77244c9e@posting.google.com> <6bd9f01b.0202171918.a58611@posting.google.com> <u4rkezl0l.fsf@ctwd0143.fitlinxx.com> <a4rsa2$2grrk$1@ID-59885.news.dfncis.de>
Message-ID: <u1yfiz5yu.fsf@ctwd0143.fitlinxx.com>

"Thomas Heller" <theller at python.net> writes:

> The source code IS available:
> http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/python/distutils/
> in the misc directory. Guess I'll have to add a note about this
> to the source... There is a tricky symbolic link somewhere on SF
> which links parts of the distutils tree into the Python source tree.

Ah thanks - I saw the "../../misc" reference in bdist_wininst, but
couldn't find it in the main Python CVS source tree (project
python/python) - I didn't notice there was a separate top level
distutils tree (presumably from it's earlier incarnation as standalone
add-on).  Not sure about the symbolic link but it doesn't seem to be
reflected to a sandbox on a cvs checkout or update.

> I'm currently working on an bdist_wininst enhancement which takes an
> additional command line option (--install-script <script.py>).  This
> script would then be run (with sys.argv set to [script_name,
> '--installed']) after all the files have been copied, and it would
> again be run (with sys.argv set to [script_name, '--removed']) just
> before the deinstaller is about to remove all the files and
> directories.

Sounds like just the ticket.  I suppose it could still be a little
tricky to arrange the OPs request of shortcuts unless you could assume
the Win32 extensions would also be present, but still plenty of
flexibility.

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


From pgaherty at chambersharrap.co.uk  Tue Feb 12 10:56:02 2002
From: pgaherty at chambersharrap.co.uk (Patrick Gaherty)
Date: Tue, 12 Feb 2002 15:56:02 +0000
Subject: re.sub and empty groups
Message-ID: <3C693B12.1060504@chambersharrap.co.uk>

I'm trying something along the following lines:

line = re.sub(r'(a)(b)?(c)',r'\3\2\1',line)

and Python complains about empty groups when (b)? doesn't match.

What's the best way of getting around this?

Thanks in advance.

Patrick Gaherty



From gherman at europemail.com  Tue Feb 26 02:33:03 2002
From: gherman at europemail.com (Dinu Gherman)
Date: Tue, 26 Feb 2002 08:33:03 +0100
Subject: Paper "PDF Editing and Processing" at IPC 10?
References: <gherman-822C04.14540615022002@news.t-online.com> <Byhb8.10471$SJ3.374979@news1.east.cox.net>
Message-ID: <gherman-F98819.08330326022002@news.t-online.com>

In article <Byhb8.10471$SJ3.374979 at news1.east.cox.net>,
 "Jeff Hinrichs" <jlh at cox.net> wrote:

> I was at the presentation.  It was a system that parsed pdf files in to
> pythonic objects and structures and allowed for editing and saving existing
> pdf docs.  It was written completely in python.  The down side is that the
> presenter had no code to show or share and seemed quite taken aback when
> requests were made.
> 
> He said in a "few months" after the code had been "cleaned up."  So, it
> doesn't appear as though they will be releasing it.   Too bad, it was quite
> interesting.
> 
> -Jeff

In private communication Eric Deng told me about the same thing.
I don't think I really care, but I'm flabbergasted to see a paper 
being accepted without anybody having the possibility to verify 
or falsify its contents. I hope we'll do better for EPC 2002, the 
EuroPython conference in June: http://www.europython.org.

Regards,

Dinu


From paul at boddie.net  Mon Feb  4 21:28:27 2002
From: paul at boddie.net (Paul Boddie)
Date: 4 Feb 2002 20:28:27 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Feb  4)
Message-ID: <C587C06E0D2D15BE.08C92F8B9CB3521B.3849AAF8C8E92DF5@lp.airnews.net>

QOTW:  "[B]elieve it or not the big money is in tracking golf carts."  Larry

"Software basically sucks."  Tim Peters

"One less file is one less thing to get screwed up in the distribution,
IMHO."  Joshua Muskovitz


    Stackless Python returns with a 2.2 alpha release!
        http://groups.google.com/groups?selm=mailman.1012273506.27877.clpa-moderators%40python.org
    
    Brian represents Zope Corporation in releasing Zope 2.5.0.
        http://www.zope.org/Products/Zope/2.5.0/zope_250_release
    
    List comprehensions meet generators for the first time (in theory, at
    least) but nested scopes remain conspicuously absent.
        http://groups.google.com/groups?th=215e6e5a7bfd526
        http://groups.google.com/groups?th=df8b5e7709957eb7
    
    The dataflow.py toolkit provides dataflow-oriented programming in
    Python... using iterators and generators.
        http://groups.google.com/groups?selm=mailman.1012487991.6116.python-list%40python.org
    
    The print magazine "Py" is ready for preordering.
        http://groups.google.com/groups?selm=mailman.1012673764.20183.clpa-moderators%40python.org
    
    The O'Reilly Open Source Convention is looking for Python and Zope
    participants.
        http://groups.google.com/groups?selm=mailman.1012273503.27874.clpa-moderators%40python.org
    
    And the Ottawa Linux Symposium calls for papers.
        http://groups.google.com/groups?selm=mailman.1012273504.27875.clpa-moderators%40python.org
    
    The eGenix.com extensions (such as the popular mxDateTime and mxODBC)
    now have their own mailing list.
        http://groups.google.com/groups?selm=mailman.1012319341.25785.clpa-moderators%40python.org
    
    Herbrip is a Python implementation of an e-mail encryption system.
        http://groups.google.com/groups?selm=slrna5b46q.cfu.philh%40comuno.freeserve.co.uk
    
    PyReverse reaches 0.2.2 but, on the subject of CASE tools and class
    diagrams, is there anything in Python which understands XMI?
        http://groups.google.com/groups?selm=mailman.1012273502.27872.clpa-moderators%40python.org
        http://groups.google.com/groups?th=f061a3675d191d99
    
    SilverCity understands lexical details about a number of languages.
        http://groups.google.com/groups?selm=mailman.1012684745.32090.python-list%40python.org
    
    How can you find out which methods belong to an object? Use the
    inspect module, of course!
        http://groups.google.com/groups?th=38e85ff1b283359f
    
    Python is a useful tool in both vim and Emacs.
        http://www.vex.net/parnassus/apyllo.py?i=860466779
        http://groups.google.com/groups?selm=mailman.1012352629.27652.python-list%40python.org
    
    System administration often involves password management; for example,
    how can one change passwords on Windows?
        http://groups.google.com/groups?th=8cec4679a555ef35
    
    John DeWeese discovers Panda3D on SourceForge - it's a Python-driven
    rendering engine.
        http://groups.google.com/groups?selm=a3b24h%24kc6%241%40usc.ed
    

========================================================================

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

    Python.org's Python Language Website is the traditional
    center of Pythonia
        http://www.python.org
    Notice especially the master FAQ
        http://www.python.org/doc/FAQ.html

    PythonWare complements the digest you're reading with the
    daily python url
         http://www.pythonware.com/daily  
    Mygale is a news-gathering webcrawler that specializes in (new)
    World-Wide Web articles related to Python.
	 http://www.awaretek.com/nowak/mygale.html 
    While cosmetically similar, Mygale and the Daily Python-URL
    are utterly different in their technologies and generally in
    their results.

    comp.lang.python.announce announces new Python software.  Be
    sure to scan this newly-revitalized newsgroup at least weekly.
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python.announce

    Michael Hudson continued Andrew Kuchling's marvelous tradition
    of summarizing action on the python-dev mailing list once every
    other week, into July 2001.  Any volunteers to re-start this
    valuable series?
        http://starship.python.net/crew/mwh/summaries/
        http://www.amk.ca/python/dev

    The Vaults of Parnassus ambitiously collect Python resources
        http://www.vex.net/~x/parnassus/   

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

    The Python Software Foundation has replaced the Python Consortium
    as an independent nexus of activity
        http://www.python.org/psf/

    Cetus does much of the same
	http://www.cetus-links.org/oo_python.html

    Python FAQTS
        http://python.faqts.com/

    The old Python "To-Do List" now lives principally in a
    SourceForge reincarnation.
	http://sourceforge.net/tracker/?atid=355470&group_id=5470&func=browse
	http://python.sourceforge.net/peps/pep-0042.html

    *Py: the Journal of the Python Language*
        http://www.pyzine.com
     
    Links2Go is a new semi-automated link collection; it's impressive
    what AI can generate
        http://www.links2go.com/search?search=python

    Tenth International Python Conference 
        http://www.python10.org            

    Archive probing tricks of the trade:
        http://groups.google.com/groups?oi=djq&as_ugroup=comp.lang.python&num=100
        http://groups.google.com/groups?meta=site%3Dgroups%26group%3Dcomp.lang.python.*

Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://www.ddj.com/topics/pythonurl/         
  http://purl.org/thecliff/python/url.html (dormant)
or
  http://groups.google.com/groups?oi=djq&as_q=+Python-URL!&as_ugroup=comp.lang.python


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From BgPorter at NOartlogicSPAM.com  Fri Feb  8 09:53:12 2002
From: BgPorter at NOartlogicSPAM.com (Brett g Porter)
Date: Fri, 08 Feb 2002 14:53:12 GMT
Subject: encapsulation, typing & other questions
References: <mailman.1013176166.4689.python-list@python.org>
Message-ID: <sDR88.28793$p5.4892009@news1.rdc1.nj.home.com>

"Laura Creighton" <lac at strakt.com> wrote in message
news:mailman.1013176166.4689.python-list at python.org...
> Design Patterns in Python are a joy to write and use.  But please
> buy a copy of the Design Patterns Smalltalk Companion so your code
> does not end up as 'Thinking in C++, transliterated into python'.
> Python is incredibly more expressive than that.
>
Also don't forget to check out Bruce Eckel's "Thinking in Python" (in
progress, but still downloadable) and his "Thinking in Patterns in Java",
which is useful as a supplement to the GoF book.

http://www.mindview.net/Books


--
// Brett g Porter * Lead Engineer, Development Practices
// BgPorter at artlogic.com * www.artlogic.com
// Art & Logic, Inc. * software engineering and design
// Desktop * Embedded * Web




From phr-n2002a at nightsong.com  Thu Feb 21 07:05:14 2002
From: phr-n2002a at nightsong.com (Paul Rubin)
Date: 21 Feb 2002 04:05:14 -0800
Subject: high precision mathematics
References: <a4mvdl$cl7$1@cer.yubc.net> <a4n4gu$eel$1@peabody.colorado.edu> <gnju6uoug3dmp3i44sjqqjv4gctdlf1vch@4ax.com> <fq8p4a.ll.ln@127.0.0.1> <a4vgnh$3eoao$1@ID-7529.news.dfncis.de> <r2615a.ii.ln@127.0.0.1> <a527jt$42g58$1@ID-7529.news.dfncis.de> <7xpu2zdzhh.fsf@ruckus.brouhaha.com> <a52arn$4340r$3@ID-7529.news.dfncis.de> <7x8z9njjra.fsf@ruckus.brouhaha.com> <a52g60$4630o$2@ID-7529.news.dfncis.de> <7x7kp7krjl.fsf@ruckus.brouhaha.com> <a52ml7$42r9l$1@ID-7529.news.dfncis.de>
Message-ID: <7x8z9n6nwl.fsf@ruckus.brouhaha.com>

"Karl M. Syring" <syring at email.com> writes:
> > > > Not likely.
> > >
> > > You really should have a look at high precison libraries (e.g.
> > > http://www.hipilib.de). They will use rational numbers internally.
> >
> > Yes, but who actually uses those libraries, and for what?
> 
> The author of the Piologie library said somewhere that extended precision
> floating point arithmetic is completely useless, but it is fun.  Then there
> is the secret message coded into pi ... locate your name in pi
> (http://pi.nersc.gov)

High precision floating point and exact rational arithmetic have their
uses, but almost all ordinary numerical computing uses ordinary
floating point.  It's very unlikely that anyone would use exact
rational arithmetic for robot motion control.


From greg at cosc.canterbury.ac.nz  Thu Feb 28 23:28:31 2002
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Fri, 01 Mar 2002 17:28:31 +1300
Subject: Status of PEP's?
References: <mailman.1014944882.14890.python-list@python.org>
Message-ID: <3C7F036F.FF1EF7A9@cosc.canterbury.ac.nz>

James_Althoff at i2.com wrote:
> 
> def doSomethingFromTo(start,end):
> 
> The function has to test to see which of "start" and "end" is the smaller
> value in order to know whether to execute a for-loop with "<" or a for-loop
> with ">" (and the code specified in each separate for-loop has to be
> written redundantly).

This is no worse than what you would have to do to
achieve the same thing using range() or any other
existing facility.

If I were writing a function like that, I think I'd 
make the caller responsible for passing the arguments 
in the right order. But if you really want, you can
always sort out the argument order before entering
the loop, e.g.

   if start > end:
      start, end = end, start
   for start <= i < end:
      ...

No redundant for-loop code required.

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


From jonathan at onegoodidea.com  Mon Feb  4 18:51:29 2002
From: jonathan at onegoodidea.com (Jonathan Hogg)
Date: Mon, 04 Feb 2002 23:51:29 +0000
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <mailman.1012836247.15524.python-list@python.org> <7xu1sx41cl.fsf@ruckus.brouhaha.com> <3C5F04BF.3090806@yumpee.org> <7x4rkwdfx2.fsf@ruckus.brouhaha.com>
Message-ID: <B884CF01.43F6%jonathan@onegoodidea.com>

On 4/2/2002 22:46, in article 7x4rkwdfx2.fsf at ruckus.brouhaha.com, "Paul
Rubin" <phr-n2002a at nightsong.com> wrote:

> The program I'm hacking right now actually needs exact rational
> division, but that's another story.  The issue is there's no way to
> use integer division that works in both 2.1 and 3.0, without using
> something like divmod that would shock the unwary.

I'm sorry, the "unwary"? I presume by this you mean that your user base
contributes to your code, that you aren't the only programmer on the
project? Are the other coders having difficulty with reading or writing
1.5.2 forward compatible code?

The Arusha codebase (the largest Python project I'm involved in), runs to
10000 lines of Python - all of which operates fine [unknown bugs
notwithstanding ;-)] in 1.5.2 to 2.2 (though I had to hack some external
code that used the long-deprecated regex module).

> I bet less than 10% of users have 2.2 right now.  I'm curious about
> the ratio of 2.1 users to 1.5.2 users.  Do you have any idea what it
> is?  I don't know how I'd measure it.  There's no User-agent header
> collected at python.org every time someone runs a Python script.

The easy way for you to tell is to make two versions of your code available,
one written for 1.5.2 and one optimised for 2.1 upwards (where they may be
exactly the same version - I'm not suggesting you do any more work). You'll
be able to gauge from downloads/sales/however your users obtain your work,
which version they have.

At the moment, this honestly looks like a storm in a teacup to me. I'm sure
you can put a small disclaimer on your program saying "I cannot guarantee
this program will work in some unknown future time on an as yet unknown
version of the Python interpreter."

Jonathan



From jkraska at san.rr.com  Sun Feb 24 20:11:58 2002
From: jkraska at san.rr.com (Courageous)
Date: Mon, 25 Feb 2002 01:11:58 GMT
Subject: Silly link question: python22_d.dll
References: <fiii7us0tj0018c11lhlm9cc2ehlk1lr4j@4ax.com> <mailman.1014593379.10486.python-list@python.org>
Message-ID: <j74j7uomek4igopeoegk222pdm5u7v078p@4ax.com>

>In the meantime, you hit on the intended solution, which only Windows
>developers have the chutzpah to complain about.

:)

There is a school  of thought that binaries and documentation are
what make a system complete and not the source. While it's not
that big a deal at all, one shouldn't have to checkout and compile
all of Python in order to straightforwardly develop C Extensions.

As to the larger problem, I can't help ya. :)

C//



From martin at v.loewis.de  Sat Feb 16 16:26:24 2002
From: martin at v.loewis.de (Martin v. Loewis)
Date: 16 Feb 2002 22:26:24 +0100
Subject: XML and newlines
References: <3C6DBB49.C9C5ED0B@earthlink.net> <86n0yaw8u4.fsf@apo.hmotakef.int> <3C6E95FA.F95C914B@earthlink.net>
Message-ID: <m3bsepjeyn.fsf@mira.informatik.hu-berlin.de>

Hans Nowak <wurmy at earthlink.net> writes:

> It was part of a previous problem; the XML supposedly only contained \n,
> where \r\n was desirable. While investigating it, I stumbled upon a
> different, more serious problem-- no newlines at all.

This is all intentional. In content, the XML processor *must*
normalize newline characters, replacing different variants of newline
with #xA. In *attribute* only, newline is further normalized to space
characters.  So if you have content where newline characters are
relevant, don't put that into attributes. If you have data which are
that sensible to their binary representation, you may consider not
using XML at all, or chosing a binary-safe representation, such as
base64-encoded strings.

Regards,
Martin


From jon+usenet at unequivocal.co.uk  Tue Feb 26 13:02:41 2002
From: jon+usenet at unequivocal.co.uk (Jon Ribbens)
Date: Tue, 26 Feb 2002 18:02:41 -0000
Subject: Precompiled regular expressions slower?
References: <3c7bb531$0$3930$b45e6eb0@senator-bedfellow.mit.edu>
Message-ID: <slrna7njdv.bs0.jon+usenet@snowy.squish.net>

In article <3c7bb531$0$3930$b45e6eb0 at senator-bedfellow.mit.edu>,
  Peter Bienstman wrote:
> There are several reg ex in the loop. Since the Python version runs 2 to 3 
> times as slow as its Perl counterpart, I thought I'd try speeding up the 
> script by precompiling the reg ex:
> 
> re1 = compile(r"....") 
> ...
>   s = search(re1, line)

That's incorrect. The right syntax is:

  s = re1.search(line)

You should find this makes quite a big difference.


From jason at jorendorff.com  Mon Feb 25 04:50:57 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Mon, 25 Feb 2002 03:50:57 -0600
Subject: GUI (PyQt) & thread problem
In-Reply-To: <8b5d764c.0202250020.268f4a81@posting.google.com>
Message-ID: <HFEKILOLEFEFMKAECNDLCEOLDEAA.jason@jorendorff.com>

> Thank you for your reply. Do you have any pointers on how to implement
> this kind of functionality, i.e. when something happens in the worker
> thread (not GUI related), the GUI needs to be updated. Is it at all
> possible?

Yes.  There are two things you can try.  Both are documented for
C++ Qt here, briefly:
  http://doc.trolltech.com/3.0/threads.html


1.  When something happens in the worker thread, post an event
    about it.  A moment later, the GUI thread will pick up the
    event and handle it as normal.

This is the preferred technique and you can use it to post either
a normal basic event like a QPaintEvent (to cause a simple repaint)
or to post a special magic event of your own to any QObject of your
choosing.  You can be sure that when the target object
receives the event, it is in the GUI thread and can call GUI
functions.

The method you need for this is QThread.postEvent().
I can't tell if this function is available from PyQt or not.

2.  When something happens in the worker thread, have the
    worker acquire the Qt library mutex.  When you do this,
    your worker thread then may call whatever GUI functions
    it likes.  It then simply releases the mutex to turn control
    back over to the normal GUI thread.

Only do this if #1 doesn't work.

## Jason Orendorff    http://www.jorendorff.com/




From daniels at mindspring.com  Sat Feb 23 11:46:14 2002
From: daniels at mindspring.com (Alan Daniels)
Date: 23 Feb 2002 08:46:14 -0800
Subject: Newbie: Crazy, but Quick
References: <u7ag65cbc0q3e@corp.supernews.com> <CMfd8.5037$D6.14223@news.iol.ie> <u7b81629bo9o4a@corp.supernews.com>
Message-ID: <5d283efe.0202230846.1802e6a2@posting.google.com>

"Virginia O" <vodine at direct.ca> wrote in message news:<u7b81629bo9o4a at corp.supernews.com>...

> If it was upto me I'd just use a simple delimited text file :)

Or simpler yet, just store the data in one large Python data structure
(ex: a tuple of dictionaries, a tuple of tuples, etc), and just have
your program import it as needed.

You can always move to a true database later if this doesn't scale,
or if you one day need relational-style joins rather than simple
flat-file access.


From dsloyer at ieee.org  Thu Feb  7 00:53:13 2002
From: dsloyer at ieee.org (David Sloyer)
Date: 6 Feb 2002 21:53:13 -0800
Subject: scripted install of Python-2.2 and win32all?
Message-ID: <961e341e.0202062153.f976988@posting.google.com>

Is it possible for one to install Python-2.2 and win32all without any
user intervention?  I'd like to install these on a set of machines as
part of a script, and would like to do this automatically.  The
installation would take place on Windows XP and Win2K boxes.

I might like to select/de-select a couple of the check boxes, but
otherwise intend to accept the default values throughout.

Thanks!


From eppstein at ics.uci.edu  Thu Feb  7 13:01:14 2002
From: eppstein at ics.uci.edu (David Eppstein)
Date: Thu, 07 Feb 2002 10:01:14 -0800
Subject: Please stop responding to Rue
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <3c62a989.3448437947@news> <a3ucsk$12h8$1@nntp6.u.washington.edu>
Message-ID: <eppstein-9CBD10.10011407022002@news.service.uci.edu>

In article <a3ucsk$12h8$1 at nntp6.u.washington.edu>,
 "Donn Cave" <donn at u.washington.edu> wrote:

> So is sci.math choked with posts from people who normally seem
> to have their heads screwed on, endlessly replying this guy as
> if they sincerely think there's some purpose in it?

Yes.

Killfiling every post with subject containing the word "Autocoding", 
instead of just posts from the offender himself, seems to do a good job of 
cutting the noise here. In sci.math the subjects aren't consistent enough 
for that, so one has to use the "References:" header to killfile all 
follow-ups to the junk.
-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/


From gcordova at hebmex.com  Fri Feb  8 10:11:04 2002
From: gcordova at hebmex.com (Gustavo Cordova)
Date: Fri, 8 Feb 2002 09:11:04 -0600 
Subject: The Matrix of >Re: Autocoding evolves from........
Message-ID: <19650FF08A99D511B19E00B0D0F06323060A0C@SRM201002>

> [...stuff about the matrix cut out...]

You just might be close...

But *they* won't let you get any closer, though.

-gus



From m.wetherill at salfordNOSPAM.acNOSPAM.uk  Wed Feb 20 15:05:26 2002
From: m.wetherill at salfordNOSPAM.acNOSPAM.uk (Matt)
Date: Wed, 20 Feb 2002 20:05:26 -0000
Subject: COM file references
References: <a4vq9q$1jsn$1@mimas.salford.ac.uk> <jdKc8.1337$EYv.1260@news1.bloor.is>
Message-ID: <3c740189$0$29609$4c56ba96@master.news.zetnet.net>

many thanks Colin,
Getting there now :-)




From cbrown at met.co.nz  Thu Feb  7 14:23:02 2002
From: cbrown at met.co.nz (Colin Brown)
Date: Fri, 8 Feb 2002 08:23:02 +1300
Subject: EDF scheduler - paradigm shift
References: <eTJ78.14206$N31.668446@ozemail.com.au> <FJg88.14875$N31.722467@ozemail.com.au>
Message-ID: <GwA88.15383$N31.751759@ozemail.com.au>

The old grey matter has been busy and I have sorted out a few things:

1.    "sched" is obviously single-threaded, so the only time that sched can
see schedule changes is during the event handling period.
2.    Schedule entries made from another thread are potentially dangerous
possibly leading to data corruption.
3.    Although it may be possible to supply a timer function based on a
select (so that it can be interrupted), I suspect that the interrupted
currently active schedule entry would be actioned before its due time.
4.    Dummy "polling" schedule events could be used to periodically pick up
updated schedules.
5.    If the multi-threaded approach I outlined in my last example doesn't
fit the requirements then it is basically roll-your-own.

All that aside, I have rethought my program design and decided that an EDF
scheduler is not really necessary anyway. My mainframe background had
coloured my thinking (the mainline processing being an event handler). I
will now make the mainline simply be the scheduler and do all the real work
in my threads - a much cleaner approach.

Colin Brown
pyNZ





From mcherm at destiny.com  Mon Feb  4 09:38:34 2002
From: mcherm at destiny.com (Michael Chermside)
Date: Mon, 04 Feb 2002 09:38:34 -0500
Subject: Xython - XML-Formed Python
Message-ID: <3C5E9CEA.8060502@destiny.com>

[Description of plan to express python in XML]

> ... but working from an XML based language
> would be much more practical.
> 
> BTW - this isn't the first time this idea has been proposed.
>  


Indeed. The issue HAS come before, and I would urge anyone intending to 
work on it to use previous work as a starting point. In particular, I 
refer you to this:

http://mail.python.org/pipermail/python-list/2001-April/037330.html

which I think does a good job of summing up previous thinking on the 
subject.

-- Michael Chermside

;-)






From threeseas at earthlink.net  Tue Feb  5 23:54:17 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Wed, 06 Feb 2002 04:54:17 GMT
Subject: Autocoding - the resistance to expect from psuedo programmers
References: <922.799T2387T11875593threeseas@earthlink.net> <iJU5VJA6wdX8Ewk4@jessikat.fsnet.co.uk>
Message-ID: <1473.801T2981T14106343threeseas@earthlink.net>

On 03-Feb-02 19:56:26 Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>In article <922.799T2387T11875593threeseas at earthlink.net>, Timothy Rue
><threeseas at earthlink.net> writes
>>In reviewing over many messages in the autocoding and related threads and
>>seeing the evidence of contridictions, hypocracy, irrational and
>>illogical responses......
>>
>>Its damn clear that there is a need to remove the sort of people posing to
>>be professional who are in essence just good liars, abstraction
>>manipulators, from the industry.
>>
>>Who knows, maybe that stench will be enough to motivate me see that it
>>gets done, gets removed.
>>
>>Maybe I should be more appriciative of the stench if it so motivates me.
>>
>>Anyone want to add anything to this?
>>
>all your ideas are belong to you
>--
>Robin Becker


Cool example of reinvention. You must be a psuedo programmer.

Problem is it's not your idea and this that I'm on about is not ideas.
It's physics.

That's ok, I don't expect you to understand.....yet.

You need everybody and their brother before you reinvent what is not an
invention....


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From maxm at mxm.dk  Wed Feb 27 10:27:52 2002
From: maxm at mxm.dk (Max M)
Date: Wed, 27 Feb 2002 15:27:52 +0000
Subject: list vs. dict
References: <3C7CF0EE.3010009@vscht.cz>
Message-ID: <3C7CFAF8.1080108@mxm.dk>

Beda Kosata wrote:

> for a piece of code I need to store relatively large amount of records 
> (100-1000). Each record will contain few (10-20) pieces of information 
> (mostly strings or integers).


Unless those are very long strings this is a pretty small dataset.

> For convenience it would be better for me to make this records as dicts, 
> so I don't have to remember what data has what position in the list. 


First you should write it the way that is easiest for you (and others) 
to read and modify. That is the most important thing.

If, and only if speed, still is an issue then do you rewrite.

> However the most important for me is speed and I suppose that working 
> with lists is faster than with dicts. So my question is - how much do 
> lists and dicts differ in speed?


You really should test it with a simulated data set if it is that important.

> Does anyone have some experience?

It usiallu doesn't matter for business logic etc. but if it is 
computational intensive it might.

btw: if you really want to store it in a list you can get dict-like 
behaviour with list speed like this:

keys = (
     NAME, AGE, SEX
     ) = range(3)
def pseudoDict():
     return [None]*len(keys)

items = []

person = pseudoDict()
person[NAME] = 'Max M'
person[AGE]  = 36
person[SEX]  = 'Male'
items.append(person)

# or

person2 = []
person2.append('Somebody') # NAME
person2.append(29)         # AGE
person2.append('Male')     # SEX
items.append(person2)


for item in items:
     print item[NAME]


But I don't think it will be nessecary.

regards Max M



From gray_py at hotmail.com  Wed Feb 13 14:37:12 2002
From: gray_py at hotmail.com (f f)
Date: Wed, 13 Feb 2002 19:37:12 +0000
Subject: commands and Scientific.MPI problems
Message-ID: <F199JvlZGdOZofu45Xi00002485@hotmail.com>

I've been using Scientific.MPI lately and came across some strange 
behaviour. When working on cluster running MPICH 1.2.1..7 and running a 
script which uses commands.getstatusoutput() or commands.getoutput(), 
reduce/send/receive calls don't work. I get 'Internal error: msg packet 
discarded' errors coming from gmpriv.c:814. But when working on a cluster 
running MPICH 1.2.2 they work fine. Using os.getcwd() instead of 
commands.getstatusouput('pwd') works fine on both too.


_________________________________________________________________
Join the world?s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com




From db3l at fitlinxx.com  Mon Feb 25 23:13:25 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Feb 2002 23:13:25 -0500
Subject: size of socket.recv buffer
References: <a5elgo$5bu$1@news.cis.ohio-state.edu> <mailman.1014687041.32731.python-list@python.org>
Message-ID: <u664kx4m2.fsf@ctwd0143.fitlinxx.com>

"Mike C. Fletcher" <mcfletch at geocities.com> writes:

> OS-level buffer is now full, OS TCP stack signals server to stop sending 
> packets, as it can't deal with any new data until the OS-level buffer is 
> read/recv/cleared [I'm assuming TCP does this explicitly, I normally do 
> with my own networking code]

Actually, it's more a lack of a message than an explicit signal to
stop sending.  The two TCP stacks have a "window" (in terms of deltas
between the packet sequence numbers currently flowing between the
endpoints) that was negotiated between them as to how much data they
are willing to let the sender get ahead of the receiver.  Both ends
include sequence information in each packet (whether a data packet or
just an ACK).  As the receiver consumes data from the front of the
window, the sender is able to send more data on the tail end of the
window simply by knowing how far ahead in the sequence he is allowed
to be based on the window size.  So when the receiver begins to back
up (whether it lost earlier data, or can no longer queue up the
information higher up in the protocol stack), the front end of the
window won't move, and the sender is forced to wait since there is no
room in the window for it to push more data onto the network.

This is more efficient than an explicit "squelch" sort of command,
since you don't have to wait for a network round trip for the command
to be received (nor deal with failures to deliver that command) before
the data stops - and thus you avoid just having to retransmit data all
over again once the receiver does in fact have room to deal with it.
It also ensures that you always have a window size worth of data
already sent between sender and receiver even when the system is
blocked due to queueing, so that when things begin to open up, you
don't have to wait for another round trip delay to get new data on the
receiver (although you do to have the new data from the sender start
arriving at the receiver).

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


From peter at engcorp.com  Mon Feb 25 19:28:26 2002
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 25 Feb 2002 19:28:26 -0500
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <oqofim3dei.fsf@titan.progiciels-bpi.ca> <LNBBLJKPBEHFEDALKOLCEEGINNAA.tim.one@home.com> <mailman.1014064953.24595.python-list@python.org> <86vgcud0wb.fsf@apo.hmotakef.int> <m2wux85hud.fsf@desk.crynwr.com> <3C794EEE.1F25B623@engcorp.com> <3C7A0F28.5090505@removejippii.fi>
Message-ID: <3C7AD6AA.BF5FFCFB@engcorp.com>

vekkuli ketkutin wrote:
> 
> Peter Hansen wrote:
> >>>>>Blig matro en cordiolios, plur tanto lestanto forgolumnia.  Plachero
> >>>>>czignat?  Classifero en bragna!
> >>>>>
> >>>>Woas da Deifi wos des hoasn mog.
> >>>>
> >>>Und welche Sprache soll das jetzt sein? Ahh, Bergdeutsche...
> >>>
> >>Even better: we could switch languages within threads, to *really*
> >>keep the monolinguists confused.
> >
> > Sed kiujn lingvojn ni devos uzi?
> 
> El?pp?s ny pua h?p?tt?en esperanttuo, vuan uan
> liijan vaikkeeta miulle!

You got me! :-)  Mi ech ne povas trovi une dictionnaire
Suoma-Ingles to figure out kion vi diris!

Maybe this multilingual -- I mean polyglot -- thing wasn't
such a good idea after all....


From MaximilianScherr at T-Online.de  Tue Feb  5 14:50:11 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Tue, 05 Feb 2002 19:50:11 -0000
Subject: how can i check if a variable is set?
Message-ID: <a3pd1j+joko@eGroups.com>

how cna i check if a variable is set?
like:

if variable a doesn't exist then set it to 0
if it exists do nothing...




From jmarshal at mathworks.com  Mon Feb  4 09:02:14 2002
From: jmarshal at mathworks.com (Joshua Marshall)
Date: 4 Feb 2002 14:02:14 GMT
Subject: Backwards Compatibility of Python versions
References: <mailman.1012815072.4727.python-list@python.org> <7xsn8hjxzw.fsf@ruckus.brouhaha.com> <Saw78.38262$KM2.1175939@atlpnn01.usenetserver.com>
Message-ID: <a3m496$e1t$1@news.mathworks.com>

Steve Holden <sholden at holdenweb.com> wrote:

>> If I type "3//4" into 1.5.2 or 2.1, I get a SyntaxError exception, so
>> it isn't backwards compatible.
>>
> Erm, don't you seem to be using a different version of backwards
> compatibilty from most people? As far as I'm concerned, if I run version
> 1.5.2 of a language, and all my programs continue to run, then the new
> release provides backwards compatibility.

It is backwards-incompatible in that someone may be relying on the
SyntaxError being raised.  Though it's probably likely that there
aren't many such programs out there.


From emile at fenx.com  Sun Feb 17 14:30:12 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 17 Feb 2002 11:30:12 -0800
Subject: chomp equivalents
References: <a4p08q$2bi$1@eskinews.eskimo.com>
Message-ID: <a4p0mm$1q3ju$1@ID-11957.news.dfncis.de>

"Xeno Campanoli" <xeno at eskimo.com> wrote in message
news:a4p08q$2bi$1 at eskinews.eskimo.com...
> Is there a chomp equivalent in the "string" object?  rstrip doesn't
seem to
> take off the newline.
>

This is being asked more and more frequently.  Follow past discussions
at:

http://groups.google.com/groups?q=chomp+group:comp.lang.python&hl=en&sco
ring=d


HTH,
--

Emile van Sebille
emile at fenx.com

---------



From mcfletch at geocities.com  Sun Feb 10 03:40:33 2002
From: mcfletch at geocities.com (Mike C. Fletcher)
Date: Sun, 10 Feb 2002 03:40:33 -0500
Subject: [ZODB] getOID( object )?
References: <3C645383.7090908@geocities.com>
Message-ID: <3C663201.1090406@geocities.com>

Well, since I haven't heard another approach, I'm going with the 
sub-transaction approach to ensuring the object's storage:

def _storeEvent( self, event ):
     """Creates an oid for the event by storing it during a 
sub-transaction"""
     ### NOTE: this is _not_ thread-safe!!!
     # store the object with a key that's not a possible OID
     self._events[ 0 ] = event
     # end the previous sub-transaction, so that the object has an OID
     get_transaction().commit( 1 )
     self._events[ event.getID() ] = event
     del self._events[ 0 ]
     return event.getID()

Where getID is just returning _p_oid for the Persistence-based objects.

Just in case anyone else ever needs to do the same thing :) .

Enjoy,
Mike

Mike C. Fletcher wrote:
...
> I read in the developer's documentation that I can use _p_oid if the 
> object has already been stored, but I'm wondering if there's a way to 
> get the oid _before_ I store the object (since I want to store the index 
> information at the same time as the object itself).
> 
> Alternately, is there a way to force a Persistence.Persistence-based 
> object to be stored immediately (preferably without requiring that I 
> commit the whole transaction, as I'm not sure the indexing will work for 
> all objects, and may need to abort) so that I can rely on there being a 
> _p_oid available?  Could I use a sub-transaction, commit that and then 
> use the oid?  It seems a bit of overkill for something this trivial.
...
_______________________________________
   Mike C. Fletcher
   http://members.rogers.com/mcfletch/





From pricerbumanto at hotmail.com  Mon Feb 11 14:43:18 2002
From: pricerbumanto at hotmail.com (George Partlow)
Date: 11 Feb 2002 11:43:18 -0800
Subject: Pythonistas? <- Re: OT: Perl programmers?
References: <mailman.1013399900.24051.python-list@python.org>
Message-ID: <e5dfe6d0.0202111143.207079b5@posting.google.com>

<brueckd at tbye.com> wrote in message news:<mailman.1013399900.24051.python-list at python.org>...
> On Sun, 10 Feb 2002, Fernando P?rez wrote:
> 
> > David Eppstein wrote:
> >
> > >> Pythonista - Pie thun ee sta
> > >> I took it as a unintended play on Sandanista, which is a group of  "freedom
> > >> fighting" rebels in Nicaragua.
> > >
> > > I think it's a normal Spanish suffix, but it seems to be moving into
> > > English -- I've also seen "Gapatista" used by some Calif. activists
> > > <http://www.ics.uci.edu/~eppstein/pix/sv/j4p/Gapatista.html>
> >
> > Indeed, it's just the Spanish form of the 'ist' English suffix (pianist,
> > violinist, etc.) It simply means 'who performs an activity'.
> 
> Could be from a number of languages; I always assumed quasi-Esperanto:

That would have to have the emphasis on the "quasi"; in Esperanto we
don't have the letter "y" or the digraph "th" (the theta sound), so
the name of the snake is "pitono", giving (presumably) "Pitonisto"*...
which doesn't prevent someone writing in Esperanto in Usenet from
typing "Python-isto", of course (and then the reader gets to mentally
pronounce it however he or she wants to, or tends to pronounce
"Python" in his or her native language).  I vote for the "analogous to
Sandinista" explanation.

> 
> "-ist" - an individual professionally or avocationally occupied with the
> idea or activity defined by the root [1].
> 
> Of course, in either Spanish or Esperanto it would more commonly be
> "Pythonisto" anyway, so who knows.
> 
> -Dave
> 
> [1] http://www.webcom.com/~donh/esperanto.html

George

*Another possibility would be to try to reproduce the English
pronunciation of the name in Esperanto: "Pajtan".  At least this would
give something that's obviously the programming language rather than
the snake This seems like a bad choice, though, since it's not
especially recognizable when read, whereas most people know "Python"
when they read it, however they may pronounce it.


From James_Althoff at i2.com  Tue Feb 12 12:58:07 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Tue, 12 Feb 2002 09:58:07 -0800
Subject: Wow.
Message-ID: <OFFCDD190B.2291C739-ON88256B5E.00627517@i2.com>

[Meowbot]
>In a few weeks of skimming those threads i'd consistently
>wackyparsed

"wackyparse"  -- I like that.  Is it original?

(Passing on the "Meowbot" handle, though. <wink>)

Jim




From mwh at python.net  Mon Feb 11 06:32:50 2002
From: mwh at python.net (Michael Hudson)
Date: Mon, 11 Feb 2002 11:32:50 GMT
Subject: How did TimBL's speech at Python 10 go yesterday?
References: <mailman.1013178854.22715.python-list@python.org>
Message-ID: <u8za0i73k.fsf@python.net>

Michael Chermside <mcherm at destiny.com> writes:

> PS: I intentionally used URL in my comments, because if the URI is a
> URN instead of a URL then it can't possibly be "downloaded" as there's
> noplace to connect to. Just trying to keep the terminology straight.

There's something (unintentionally?) hilarious about this postscript.

Cheers,
M.

-- 
8. A programming language is low level when its programs require
   attention to the irrelevant.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From simon at ping-pong.budig  Tue Feb  5 09:03:43 2002
From: simon at ping-pong.budig (Simon Budig)
Date: 5 Feb 2002 15:03:43 +0100
Subject: Readline and unwanted filename completion
References: <3c599cab@si-nic.hrz.uni-siegen.de> <ur8o5xytl.fsf@python.net> <3c5b2c1f@si-nic.hrz.uni-siegen.de> <3c5c5606@si-nic.hrz.uni-siegen.de> <uit9dqxp9.fsf@python.net> <3c5eb2fd$1@si-nic.hrz.uni-siegen.de> <u4rkwql8p.fsf@python.net>
Message-ID: <3c5fe63f$1@si-nic.hrz.uni-siegen.de>

In article <u4rkwql8p.fsf at python.net>, Michael Hudson wrote:
[discussion of me about being lazy creating a sf account...  :-)   ]
> That's what I did when I submitted it:
> 
> https://sourceforge.net/tracker/index.php?func=detail&aid=513235&group_id=5470&atid=305470
> 
>:) you might want to type your email address into the "monitor" field.

Ok, did that   :-)

Btw - I already had submitted a Bug-Report via a friend:
https://sourceforge.net/tracker/index.php?func=detail&aid=511655&group_id=5470&atid=105470

You might want to link them as duplicate/related.

BTW: The stupid fixed width limit of SF messed the patch up.
Maybe creating an attachment is more useful...

Thanks for your help!

Bye,
	Simon

-- 
      Simon.Budig at unix-ag.org       http://www.home.unix-ag.org/simon/


From littlejohn.75 at free.fr  Fri Feb  1 01:20:55 2002
From: littlejohn.75 at free.fr (Francois Petitjean)
Date: Fri, 1 Feb 2002 07:20:55 +0100
Subject: confused by isinstance()
References: <3c5717bb$0$23518$626a54ce@news.free.fr> <a3b9ib$61c$1@panix2.panix.com>
Message-ID: <3c5a3352$0$4604$626a54ce@news.free.fr>

>You'll either need to post more code or reduce the code
Here is a trivial script (1) which highlights the problem :
#!/usr/bin/env python
"""Fake Foo python module to test isinstance"""

class Foo:
    def __init__(self):
        print "ctor " #,foo(self)

def foo(o):
    return isinstance(o,Foo)

def bar(o, modname='Foo'):
    """is o an instance of the class Foo in a module named modname"""
    mymod = __import__(modname)   # or import Foo as mymod
    cls = mymod.Foo
    print "bar(o) %s %s" % ( isinstance(o,cls), `cls`)
    print "names? %s" % ( cls.__name__ == o.__class__.__name__, )

if __name__ == '__main__':
    x = Foo()
    bar(x)
    print "foo(o) %s %s" % (foo(x), `Foo`)
    print "names? %s" % ( Foo.__name__ == x.__class__.__name__, )
    print `x`

the output: (python 2.1.1 Windows NT4)
$ ../Python21/python.exe Foo.py
ctor
bar(o) 0 <class Foo.Foo at 007CBABC>
names? 1
foo(o) 1 <class __main__.Foo at 007FCF9C>
names? 1
<__main__.Foo instance at 007FC2CC>

This, IMHO, shows the "insistance" of "isinstance" to be very picky about
its arguments. Or there is a pitfall in the handling of namespaces or
import semantics which I don't know?

Note that I am very aware that using isinstance or other type checking
should be avoided if possible. But, I want eagerly to understand.

(1) It is trivial as it has :
only one module
only one class
only one instance
and it uses only builtins and doesn't use 2.2 features.

TIA






From jdgorman at ieee.org  Wed Feb  6 00:23:22 2002
From: jdgorman at ieee.org (John Gorman)
Date: Wed, 06 Feb 2002 05:23:22 GMT
Subject: Which GUI toolkit to use?
References: <3c607ffd_3@corp.newsgroups.com>
Message-ID: <3C60BDF2.9040503@ieee.org>

Hi Joshua,
   This question has surfaced several times on c.l.p.

See, for instance, past postings at google/dejanews...
(http://groups.google.com/groups?q=python+gui+group%3Acomp.lang.python.*&btnG=Google+Search&hl=en)

Cameron Laird maintains a list that has some useful pointers.
http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html

Several of these work on multiple platforms.  I've had good success with 
wxPython code running successfully (w/o modification) on both Win2K and 
RH 7.2.

There are also gui builders available for wxPython/wxWindows 
(boa-constructor, wxDesigner) and pyQT (Black Adder) and pyGTk (glade). 
  As far as I know there are win32 versions of QT and GTk, so each of 
these three toolkits are accessible to Win platforms.

Cheers,
John Gorman


Joshua Muskovitz wrote:

> Specifics:
> 
> 1.  Needs to run on Windows platform.
> 2.  Needs to easily support completely custom paint (buttons, scrollers,
> listboxes, etc).
> 3.  Needs to handle simple input (editboxes, onLeftClick events).
> 
> The goal is to produce an app that runs full screen on a Win platform, and
> is interacted with via a touchscreen.  I would like to just paint the whole
> damn thing myself, but a widget manager will make input handling easier.
> There will be a lot of bitmaps making the whole thing pretty.
> 
> Tk seems like it is a pain to use, wxWindows seems like it doesn't make
> painting easy, pythonwin (win32ui) might be the way to go, ...
> 
> Decisions, decisions.  There wouldn't be anyone around here who has an
> opinion, is there?  :-)
> 
> --
> # Joshua Muskovitz
> # joshm at taconic.net
> def lyyrs(sig): return '-'.join(sig.split()+["ly y'rs"])
> lyyrs('Hire me!  I need the work!')
> 
> 
> 
> 
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----
> 



From dave at pythonapocrypha.com  Thu Feb 28 10:36:15 2002
From: dave at pythonapocrypha.com (Dave Brueck)
Date: 28 Feb 2002 07:36:15 -0800
Subject: Printing to a "log screen" in a multi-module QT app
References: <54abe19d.0202271213.5e6ef2b0@posting.google.com>
Message-ID: <477762c2.0202280736.6866626d@posting.google.com>

schneida at seiu.org (Anders Schneiderman) wrote in message news:<54abe19d.0202271213.5e6ef2b0 at posting.google.com>...
> I'm writing an application using QT.  Right now when I want to use the
> equivalent of "print", I call a method called "log" that's defined in
> the main class in my main module:
> 
> def log(self, msg):
>    try:
>       self.logArea.insertline(msg, -1)
>    except:
>       self.logArea.insertline(str(msg), -1)
> 
> To call it in other classes that are in other modules of the app, I
> pass "log" to the new class.  For ex:
> 
> class CodeEditor:
>    def __init__ (self, log, ...)
>        self.log = log
> 
> Is there a cleaner way to do this?  For example, is there a way to
> redefine the "print" command?  Whatever solution I come up with, the
> command to print has to be short;  "self.logObject.log('File
> successfully updated')" is too much of a pain.  Any suggestions?

Well, do you ever really need more than one logger? Most likely no, in
which case you don't really need to pass it to each class, just import
it at the beginning of each module. I usually have a program-wide
logger in one module like this:

logf = open('myapp.log','wt')
def log(*args):
   logf.write(' '.join(['%s' % x for x in args]) + '\n')
   logf.flush()

(yours would be slightly different obviously, but not much) and then
import it in each module like this:

from whatever import log

If you _really_ want to redefine 'print' you can (by setting
sys.stdout and sys.stderr to refer to your own file-like object (just
needs a 'write' method)) but that tends to be both confusing and
annoying so I usually use
the above method instead.

-Dave


From gimbo at ftech.net  Tue Feb 26 09:16:07 2002
From: gimbo at ftech.net (Andy Gimblett)
Date: Tue, 26 Feb 2002 14:16:07 +0000
Subject: Contributing patches (was Re: urllib and proxy)
In-Reply-To: <20020226100136.GA18498@andy.tynant.ftech.net>
References: <a5ecas$qiq$1@zeus.polsl.gliwice.pl> <20020226100136.GA18498@andy.tynant.ftech.net>
Message-ID: <20020226141607.GE18498@andy.tynant.ftech.net>

Earlier on today I posted to this list a modified urlopen() which
supports proxy specification.  How do I go about suggesting this as a
modification to the standard python library?

I've seen these:
    http://www.python.org/download/cvs.html
    http://www.python.org/patches/

and they seem to suggest that I can just use my existing SourceForge
account and submit a patch to the patch manager, and I presume then
someone examines it, decides if it's a good thing or not, and checks
it in or rejects it?

Is this how it works?

Just thought I'd ask here before blundering in.

Cheers,

Andy

-- 
Andy Gimblett - Programmer - Frontier Internet Services Limited
Tel: 029 20 820 044 Fax: 029 20 820 035 http://www.frontier.net.uk/
Statements made are at all times subject to Frontier's Terms and
Conditions of Business, which are available upon request.



From jason at jorendorff.com  Thu Feb 14 17:21:49 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 16:21:49 -0600
Subject: hex to int
In-Reply-To: <3C6C1E76.8AE06332@ix.netcom.com>
Message-ID: <HFEKILOLEFEFMKAECNDLEEKFDCAA.jason@jorendorff.com>

Antaeus Feldspar wrote:
> Bjorn Pettersen wrote:
> > >>> int('0xff', 16)
> > 255
> 
> In what version of Python do they introduce that?  In the documentation
> I have int() is listed as taking one and exactly one argument.  The code
> you list doesn't work for me.

Waaaaay back in Python 1.6.
  http://www.python.org/doc/1.6/lib/built-in-funcs.html
  http://www.python.org/doc/current/lib/built-in-funcs.html

Unfortunately, a lot of people are stuck with Python 1.5.2,
which will be three years old in April.  (sigh)

## Jason Orendorff    http://www.jorendorff.com/



From jstcyr at mediaone.net  Sat Feb 16 00:56:06 2002
From: jstcyr at mediaone.net (Jim St.Cyr)
Date: Sat, 16 Feb 2002 05:56:06 GMT
Subject: DCOracle syntax question
Message-ID: <Wvmb8.659$jj4.576986@typhoon.ne.ipsvc.net>

I have a query of the form

select * from table where key = 'key' and date = (select max(to_char(date))
from table where key = 'key')

Through SQL*Plus this executes fine.  When executing through DCOracle I get
an empty result.  I've tried both tag = 'query' and cursor.execute(query).
I think that either the subquery or the inline quotes is throwing things off
but am at a loss.

Any pointers would be appreciated.

Thanks.

Jim




From aahz at panix.com  Fri Feb  8 18:52:02 2002
From: aahz at panix.com (Aahz Maruch)
Date: 8 Feb 2002 15:52:02 -0800
Subject: OT: Perl programmers?
References: <a417rn$ge8$1@panix2.panix.com> <a40e2r.3vvegs9.1@kserver.org> <11DA76EA13D0CF8D.5EACD856ADF3065A.09057FA224243232@lp.airnews.net> <tbf14a.ep7.ln@wombat.san-francisco.ca.us>
Message-ID: <a41ob2$9k3$1@panix2.panix.com>

In article <tbf14a.ep7.ln at wombat.san-francisco.ca.us>,
Keith Keller <kkeller at speakeasy.net> wrote:
>
>Aahz, are you sorry you asked yet?  :)

Not really.  Someone referred to "Perlies" in an e-mail to me (which I
actually misread as "Perlers", because I'd heard that term before), and
when I asked another Perl programmer of my acquaintance, zie didn't much
care for "Perlers" (and had never heard of it, to boot).  So between all
the responses I've received in public and private, I think I'll stick
with "Perl programmers".

Thanks all!
-- 
                      --- Aahz  <*>  (Copyright 2002 by aahz at pobox.com)

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

"We should forget about small efficiencies, about 97% of the time.
Premature optimization is the root of all evil."  --Knuth


From jimd at vega.starshine.org  Thu Feb 28 04:41:18 2002
From: jimd at vega.starshine.org (Jim Dennis)
Date: 28 Feb 2002 09:41:18 GMT
Subject: SSL Server Sockets?
References: <a5ijos$1565$1@news.idiom.com>  <jw5f8.129$5vh.170404352@news.telia.no> <mailman.1014826795.30234.python-list@python.org>
Message-ID: <a5ktvu$ae8$1@news.idiom.com>

In article <mailman.1014826795.30234.python-list at python.org>, 
  Laura Creighton wrote:

>Martin Sj?gren (martin at strakt.com) made pyOpenSSL work for servers.  
>See http://pyopenssl.sourceforge.net There is a debian package.

>Laura Creighton

 Thank you, Laura.  I have fetched and installed this; and one 
 of my browser windows is waiting for me to read the docs.

 Pity.  There's sample code.



From sholden at holdenweb.com  Mon Feb 18 10:10:07 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 18 Feb 2002 10:10:07 -0500
Subject: Adding static typing to Python
References: <Pine.LNX.4.33.0202181937180.4907-100000@ns0.tbye.com> <mailman.1014094230.2291.python-list@python.org>
Message-ID: <QUtc8.8357$Zp2.779989@atlpnn01.usenetserver.com>

"Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> wrote ...
[sceptical remarks about static typing's value]
>
> Strong typing (in function prototypes) is for control freaks IMHO.  I used
> to be one, but it's just too much work.
>
A long time ago, in a galaxy far from here, I worked in a Computer Science
department where a new professor, with interests in formal methods, was
appointed. None of his group appeared to realise that in the real world the
real problem was in getting an accurate specification.

They had huge schemes for transforming a formal specification into an
implementation in some programming language, all of which completely ignored
the fact that the average user was totally incapable of expressing
themselves in any language which would transliterate into a formal
specification. They therefore had no adeqaute way of proving the correctness
of the *specification* even if they could guarantee its implementation. Ho,
hum.

Similarly, strong typing is useful when you want to ensure that particular
variables hold a limited range of types. This is s long way away from the
average Python programmers' interest [see sig block], since Pythonistas tend
to delight in the "incidental generality" engendered by the language's
robust polymorphism.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From emile at fenx.com  Sun Feb  3 14:34:10 2002
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 3 Feb 2002 11:34:10 -0800
Subject: Java vs Python Benchmarks: Java is faster
References: <mailman.1012761186.28271.python-list@python.org>
Message-ID: <a3k3li$1967ua$1@ID-11957.news.dfncis.de>

"Jason Orendorff" <jason at jorendorff.com> wrote in message
news:mailman.1012761186.28271.python-list at python.org...

>  5. Python still has SET_LINENO, which is basically a waste
>     of time.  (This is low-hanging fruit, but apparently the
>     team is too busy to grab it.)
>

Note the -O (and -OO) command line options to python already eliminate
this.

--

Emile van Sebille
emile at fenx.com

---------



From MaximilianScherr at T-Online.de  Mon Feb 11 16:38:08 2002
From: MaximilianScherr at T-Online.de (maximilianscherr)
Date: Mon, 11 Feb 2002 21:38:08 -0000
Subject: help in converting c code needed
Message-ID: <a49dk0+n16v@eGroups.com>

can anybody help me convert this c code to python:

int Hooks::Decompress (unsigned char * pdest, unsigned char * src2, 
int src_size)
{
    const unsigned char * psrc = src2;
    int leng = src_size;
    int dest_index = 0;
 bit_num = 8;
 treepos = 0;
 
    while(true)
    {
        if(bit_num == 8)
        {
            if(leng == 0)
            {
                return dest_index;
            }
            leng--;
            value = *psrc++;
            bit_num = 0;
            mask = 0x80;
        }
        if(value & mask)
            treepos = tree[treepos * 2];
        else
            treepos = tree[treepos * 2 + 1];
        mask >>= 1;
        bit_num++;
 
        if(treepos <= 0)
        {
            if(treepos == -256)
            {
                bit_num = 8;
                treepos = 0;
                continue;
            }
            pdest[dest_index++] = -treepos;
            treepos = 0;
        }
    }
}
 
int Hooks::Pack(unsigned char *pIn, unsigned char *pOut, int lengg)
{
 
 int actByte = 0;
 int bitByte = 0;
 int nrBits;
 unsigned int value;
 
 if (lengg<=0) return 0;
 
 while(lengg--)
 {
  nrBits = bit_table[*pIn][0];
  value = bit_table[*pIn++][1];
 
  while(nrBits--)
  {
   pOut[actByte] = (pOut[actByte] << 1) | (unsigned char)((value >> 
nrBits) & 0x1);
 
   bitByte = (bitByte + 1) & 0x07;
   if(!bitByte) actByte++;
  }
 }
 
 nrBits = bit_table[256][0];
 value = bit_table[256][1];
 
 while(nrBits--)
 {
  pOut[actByte] = (pOut[actByte] << 1) | (unsigned char)((value >> 
nrBits) & 0x1);
 
  bitByte = (bitByte + 1) & 0x07;
  if(!bitByte) actByte++;
 }
 
 if(bitByte)
 {
  while(bitByte < 8)
  {
   pOut[actByte] <<= 1;
   bitByte++;
  }
 
  actByte++;
 }
 
 return actByte;
}
 
 
class Hooks
{
public:
 int Hook(unsigned char *Data,bool who, int OldLen);
 void HackServerList(unsigned char *Data);
 int Pack(unsigned char *pIn, unsigned char *pOut, int lengg);
 
 int Decompress (unsigned char * pdest, unsigned char * src2, int 
src_size);
 int bit_num;
    int value;
    int mask;
    int treepos;
 bool InfoSent;
};
 
static int tree[512] =
{
    /*   0*/     1, 2,
    /*   1*/     3, 4,
    /*   2*/     5, 0,
    /*   3*/     6, 7,
    /*   4*/     8, 9,
    /*   5*/    10,   11,
    /*   6*/    12,   13,
    /*   7*/  -256,   14,
    /*   8*/    15,   16,
    /*   9*/    17,   18,
    /*  10*/    19,   20,
    /*  11*/    21,   22,
    /*  12*/    -1,   23,
    /*  13*/    24,   25,
    /*  14*/    26,   27,
    /*  15*/    28,   29,
    /*  16*/    30,   31,
    /*  17*/    32,   33,
    /*  18*/    34,   35,
    /*  19*/    36,   37,
    /*  20*/    38,   39,
    /*  21*/    40,  -64,
    /*  22*/    41,   42,
    /*  23*/    43,   44,
    /*  24*/    -6,   45,
    /*  25*/    46,   47,
    /*  26*/    48,   49,
    /*  27*/    50,   51,
    /*  28*/  -119,   52,
    /*  29*/   -32,   53,
    /*  30*/    54,  -14,
    /*  31*/    55,   -5,
    /*  32*/    56,   57,
    /*  33*/    58,   59,
    /*  34*/    60,   -2,
    /*  35*/    61,   62,
    /*  36*/    63,   64,
    /*  37*/    65,   66,
    /*  38*/    67,   68,
    /*  39*/    69,   70,
    /*  40*/    71,   72,
    /*  41*/   -51,   73,
    /*  42*/    74,   75,
    /*  43*/    76,   77,
    /*  44*/  -101, -111,
    /*  45*/    -4,  -97,
    /*  46*/    78,   79,
    /*  47*/  -110,   80,
    /*  48*/    81, -116,
    /*  49*/    82,   83,
    /*  50*/    84, -255,
    /*  51*/    85,   86,
    /*  52*/    87,   88,
    /*  53*/    89,   90,
    /*  54*/   -15,  -10,
    /*  55*/    91,   92,
    /*  56*/   -21,   93,
    /*  57*/  -117,   94,
    /*  58*/    95,   96,
    /*  59*/    97,   98,
    /*  60*/    99,  100,
    /*  61*/  -114,  101,
    /*  62*/  -105,  102,
    /*  63*/   -26,  103,
    /*  64*/   104,  105,
    /*  65*/   106,  107,
    /*  66*/   108,  109,
    /*  67*/   110,  111,
    /*  68*/   112,   -3,
    /*  69*/   113,   -7,
    /*  70*/   114, -131,
    /*  71*/   115, -144,
    /*  72*/   116,  117,
    /*  73*/   -20,  118,
    /*  74*/   119,  120,
    /*  75*/   121,  122,
    /*  76*/   123,  124,
    /*  77*/   125,  126,
    /*  78*/   127,  128,
    /*  79*/   129, -100 ,
    /*  80*/   130,   -8,
    /*  81*/   131,  132,
    /*  82*/   133,  134,
    /*  83*/  -120,  135,
    /*  84*/   136,  -31,
    /*  85*/   137,  138,
    /*  86*/  -109, -234,
    /*  87*/   139,  140,
    /*  88*/   141,  142,
    /*  89*/   143,  144,
    /*  90*/  -112,  145,
    /*  91*/   -19,  146,
    /*  92*/   147,  148,
    /*  93*/   149,  -66,
    /*  94*/   150, -145,
    /*  95*/   -13,  -65,
    /*  96*/   151,  152,
    /*  97*/   153,  154,
    /*  98*/   -30,  155,
    /*  99*/   156,  157,
    /* 100*/   -99,  158,
    /* 101*/   159,  160,
    /* 102*/   161,  162,
    /* 103*/   -23,  163,
    /* 104*/   -29,  164,
    /* 105*/   -11,  165,
    /* 106*/   166, -115,
    /* 107*/   167,  168,
    /* 108*/   169,  170,
    /* 109*/   -16,  171,
    /* 110*/   -34,  172,
    /* 111*/   173, -132,
    /* 112*/   174, -108,
    /* 113*/   175,  -22,
    /* 114*/   176,   -9,
    /* 115*/   177,  -84,
    /* 116*/   -17,  -37,
    /* 117*/   -28,  178,
    /* 118*/   179,  180,
    /* 119*/   181,  182,
    /* 120*/   183,  184,
    /* 121*/   185,  186,
    /* 122*/   187, -104,
    /* 123*/   188,  -78,
    /* 124*/   189,  -61,
    /* 125*/   -79, -178,
    /* 126*/   -59, -134,
    /* 127*/   190,  -25,
    /* 128*/   -83,  -18,
    /* 129*/   191,  -57,
    /* 130*/   -67,  192,
    /* 131*/   -98,  193,
    /* 132*/   -12,  -68,
    /* 133*/   194,  195,
    /* 134*/   -55, -128,
    /* 135*/   -24,  -50,
    /* 136*/   -70,  196,
    /* 137*/   -94,  -33,
    /* 138*/   197, -129,
    /* 139*/   -74,  198,
    /* 140*/   -82,  199,
    /* 141*/   -56,  -87,
    /* 142*/   -44,  200,
    /* 143*/  -248,  201,
    /* 144*/  -163,  -81,
    /* 145*/   -52, -123,
    /* 146*/   202, -113,
    /* 147*/   -48,  -41,
    /* 148*/  -122,  -40,
    /* 149*/   203,  -90,
    /* 150*/   -54,  204,
    /* 151*/   -86, -192,
    /* 152*/   205,  206,
    /* 153*/   207, -130,
    /* 154*/   -53,  208,
    /* 155*/  -133,  -45,
    /* 156*/   209,  210,
    /* 157*/   211,  -91,
    /* 158*/   212,  213,
    /* 159*/  -106,  -88,
    /* 160*/   214,  215,
    /* 161*/   216,  217,
    /* 162*/   218,  -49,
    /* 163*/   219,  220,
    /* 164*/   221,  222,
    /* 165*/   223,  224,
    /* 166*/   225,  226,
    /* 167*/   227, -102,
    /* 168*/  -160,  228,
    /* 169*/   -46,  229,
    /* 170*/  -127,  230,
    /* 171*/  -103,  231,
    /* 172*/   232,  233,
    /* 173*/   -60,  234,
    /* 174*/   235,  -76,
    /* 175*/   236, -121,
    /* 176*/   237,  -73,
    /* 177*/  -149,  238,
    /* 178*/   239, -107,
    /* 179*/   -35,  240,
    /* 180*/   -71,  -27,
    /* 181*/   -69,  241,
    /* 182*/   -89,  -77,
    /* 183*/   -62, -118,
    /* 184*/   -75,  -85,
    /* 185*/   -72,  -58,
    /* 186*/   -63,  -80,
    /* 187*/   242,  -42,
    /* 188*/  -150, -157,
    /* 189*/  -139, -236,
    /* 190*/  -126, -243,
    /* 191*/  -142, -214,
    /* 192*/  -138, -206,
    /* 193*/  -240, -146,
    /* 194*/  -204, -147,
    /* 195*/  -152, -201,
    /* 196*/  -227, -207,
    /* 197*/  -154, -209,
    /* 198*/  -153, -254,
    /* 199*/  -176, -156,
    /* 200*/  -165, -210,
    /* 201*/  -172, -185,
    /* 202*/  -195, -170,
    /* 203*/  -232, -211,
    /* 204*/  -219, -239,
    /* 205*/  -200, -177,
    /* 206*/  -175, -212,
    /* 207*/  -244, -143,
    /* 208*/  -246, -171,
    /* 209*/  -203, -221,
    /* 210*/  -202, -181,
    /* 211*/  -173, -250,
    /* 212*/  -184, -164,
    /* 213*/  -193, -218,
    /* 214*/  -199, -220,
    /* 215*/  -190, -249,
    /* 216*/  -230, -217,
    /* 217*/  -169, -216,
    /* 218*/  -191, -197,
    /* 219*/   -47,  243,
    /* 220*/   244,  245,
    /* 221*/   246,  247,
    /* 222*/  -148, -159,
    /* 223*/   248,  249,
    /* 224*/   -92,  -93,
    /* 225*/   -96, -225,
    /* 226*/  -151,  -95,
    /* 227*/   250,  251,
    /* 228*/  -241,  252,
    /* 229*/  -161,  -36,
    /* 230*/   253,  254,
    /* 231*/  -135,  -39,
    /* 232*/  -187, -124,
    /* 233*/   255, -251,
    /* 234*/  -162, -238,
    /* 235*/  -242,  -38,
    /* 236*/   -43, -125,
    /* 237*/  -215, -253,
    /* 238*/  -140, -208,
    /* 239*/  -137, -235,
    /* 240*/  -158, -237,
    /* 241*/  -136, -205,
    /* 242*/  -155, -141,
    /* 243*/  -228, -229,
    /* 244*/  -213, -168,
    /* 245*/  -224, -194,
    /* 246*/  -196, -226,
    /* 247*/  -183, -233,
    /* 248*/  -231, -167,
    /* 249*/  -174, -189,
    /* 250*/  -252, -166,
    /* 251*/  -198, -222,
    /* 252*/  -188, -179,
    /* 253*/  -223, -182,
    /* 254*/  -180, -186,
    /* 255*/  -245, -247,
};
 
static unsigned int bit_table[257][2] =
{
 0x02, 0x00,  0x05, 0x1F,  0x06, 0x22,  0x07, 0x34,  0x07, 0x75,  
0x06, 0x28,  0x06, 0x3B,  0x07, 0x32, 
 0x08, 0xE0,  0x08, 0x62,  0x07, 0x56,  0x08, 0x79,  0x09, 0x19D, 
0x08, 0x97,  0x06, 0x2A,  0x07, 0x57, 
 0x08, 0x71,  0x08, 0x5B,  0x09, 0x1CC, 0x08, 0xA7,  0x07, 0x25,  
0x07, 0x4F,  0x08, 0x66,  0x08, 0x7D, 
 0x09, 0x191, 0x09, 0x1CE,  0x07, 0x3F,  0x09, 0x90,  0x08, 0x59,  
0x08, 0x7B,  0x08, 0x91,  0x08, 0xC6, 
 0x06, 0x2D,  0x09, 0x186,  0x08, 0x6F,  0x09, 0x93,  0x0A, 0x1CC, 
0x08, 0x5A,  0x0A, 0x1AE, 0x0A, 0x1C0, 
 0x09, 0x148, 0x09, 0x14A,  0x09, 0x82,  0x0A, 0x19F,  0x09, 0x171, 
0x09, 0x120,  0x09, 0xE7,  0x0A, 0x1F3, 
 0x09, 0x14B, 0x09, 0x100, 0x09, 0x190, 0x06, 0x13,  0x09, 0x161, 
0x09, 0x125, 0x09, 0x133, 0x09, 0x195, 
 0x09, 0x173, 0x09, 0x1CA, 0x09, 0x86,  0x09, 0x1E9,  0x09, 0xDB,  
0x09, 0x1EC, 0x09, 0x8B,  0x09, 0x85, 
 0x05, 0x0A,  0x08, 0x96,  0x08, 0x9C,  0x09, 0x1C3,  0x09, 0x19C, 
0x09, 0x8F,  0x09, 0x18F, 0x09, 0x91, 
 0x09, 0x87,  0x09, 0xC6,  0x09, 0x177, 0x09, 0x89,  0x09, 0xD6,  
0x09, 0x8C,  0x09, 0x1EE, 0x09, 0x1EB, 
 0x09, 0x84,  0x09, 0x164,  0x09, 0x175, 0x09, 0x1CD,  0x08, 0x5E,  
0x09, 0x88,  0x09, 0x12B, 0x09, 0x172, 
 0x09, 0x10A, 0x09, 0x8D,  0x09, 0x13A, 0x09, 0x11C,  0x0A, 0x1E1, 
0x0A, 0x1E0,  0x09, 0x187, 0x0A, 0x1DC, 
 0x0A, 0x1DF, 0x07, 0x74,  0x09, 0x19F, 0x08, 0x8D,  0x08, 0xE4,  
0x07, 0x79,  0x09, 0xEA,  0x09, 0xE1, 
 0x08, 0x40,  0x07, 0x41,  0x09, 0x10B, 0x09, 0xB0,  0x08, 0x6A,  
0x08, 0xC1,  0x07, 0x71,  0x07, 0x78, 
 0x08, 0xB1,  0x09, 0x14C,  0x07, 0x43,  0x08, 0x76,  0x07, 0x66,  
0x07, 0x4D,  0x09, 0x8A,  0x06, 0x2F, 
 0x08, 0xC9,  0x09, 0xCE,  0x09, 0x149, 0x09, 0x160,  0x0A, 0x1BA,  
0x0A, 0x19E,  0x0A, 0x39F,  0x09, 0xE5, 
 0x09, 0x194,  0x09, 0x184,  0x09, 0x126,  0x07, 0x30,  0x08, 0x6C,  
0x09, 0x121,  0x09, 0x1E8,  0x0A, 0x1C1, 
 0x0A, 0x11D,  0x0A, 0x163,  0x0A, 0x385,  0x0A, 0x3DB,  0x0A, 
0x17D,  0x0A, 0x106,  0x0A, 0x397,  0x0A, 0x24E, 
 0x07, 0x2E,  0x08, 0x98,  0x0A, 0x33C,  0x0A, 0x32E,  0x0A, 0x1E9,  
0x09, 0xBF,  0x0A, 0x3DF,  0x0A, 0x1DD, 
 0x0A, 0x32D,  0x0A, 0x2ED,  0x0A, 0x30B,  0x0A, 0x107,  0x0A, 
0x2E8,  0x0A, 0x3DE,  0x0A, 0x125,  0x0A, 0x1E8, 
 0x09, 0xE9,  0x0A, 0x1CD,  0x0A, 0x1B5,  0x09, 0x165,  0x0A, 0x232,  
0x0A, 0x2E1,  0x0B, 0x3AE,  0x0B, 0x3C6, 
 0x0B, 0x3E2,  0x0A, 0x205,  0x0A, 0x29A,  0x0A, 0x248,  0x0A, 
0x2CD,  0x0A, 0x23B,  0x0B, 0x3C5,  0x0A, 0x251, 
 0x0A, 0x2E9,  0x0A, 0x252,  0x09, 0x1EA,  0x0B, 0x3A0,  0x0B, 
0x391,  0x0A, 0x23C,  0x0B, 0x392,  0x0B, 0x3D5, 
 0x0A, 0x233,  0x0A, 0x2CC,  0x0B, 0x390,  0x0A, 0x1BB,  0x0B, 
0x3A1,  0x0B, 0x3C4,  0x0A, 0x211,  0x0A, 0x203, 
 0x09, 0x12A,  0x0A, 0x231,  0x0B, 0x3E0,  0x0A, 0x29B,  0x0B, 
0x3D7,  0x0A, 0x202,  0x0B, 0x3AD,  0x0A, 0x213, 
 0x0A, 0x253,  0x0A, 0x32C,  0x0A, 0x23D,  0x0A, 0x23F,  0x0A, 
0x32F,  0x0A, 0x11C,  0x0A, 0x384,  0x0A, 0x31C, 
 0x0A, 0x17C,  0x0A, 0x30A,  0x0A, 0x2E0,  0x0A, 0x276,  0x0A, 
0x250,  0x0B, 0x3E3,  0x0A, 0x396,  0x0A, 0x18F, 
 0x0A, 0x204,  0x0A, 0x206,  0x0A, 0x230,  0x0A, 0x265,  0x0A, 
0x212,  0x0A, 0x23E,  0x0B, 0x3AC,  0x0B, 0x393, 
 0x0B, 0x3E1,  0x0A, 0x1DE,  0x0B, 0x3D6,  0x0A, 0x31D,  0x0B, 
0x3E5,  0x0B, 0x3E4,  0x0A, 0x207,  0x0B, 0x3C7, 
 0x0A, 0x277,  0x0B, 0x3D4,  0x08, 0xC0,  0x0A, 0x162,  0x0A, 0x3DA,  
0x0A, 0x124,  0x0A, 0x1B4,  0x0A, 0x264, 
 0x0A, 0x33D,  0x0A, 0x1D1,  0x0A, 0x1AF,  0x0A, 0x39E,  0x0A, 
0x24F,  0x0B, 0x373,  0x0A, 0x249,  0x0B, 0x372, 
 0x09, 0x167,  0x0A, 0x210,  0x0A, 0x23A,  0x0A, 0x1B8,  0x0B, 
0x3AF,  0x0A, 0x18E,  0x0A, 0x2EC,  0x07, 0x62, 
 0x04, 0x0D
 
};

?

Thanks, in advance




From vincent_a_primavera at netzero.net  Tue Feb 26 08:26:08 2002
From: vincent_a_primavera at netzero.net (Vincent A. Primavera)
Date: 26 Feb 2002 13:26:08 +0000
Subject: telnetlib...
Message-ID: <1014729969.11756.2.camel@del-vap-bos.ralphpill.com>

Hello,

	Does anybody out there have any good telnetlib examples they would be
willing to share?

	Thank you,

	Vincent A. Primavera.




From vodine at direct.ca  Fri Feb 15 11:53:32 2002
From: vodine at direct.ca (Virginia O)
Date: Fri, 15 Feb 2002 08:53:32 -0800
Subject: More license stuff
References: <u6oe6n32m9pjfb@corp.supernews.com> <3C6C566A.A51AD1E1@ccvcorp.com>
Message-ID: <u6qf3lffi10cd1@corp.supernews.com>

Thanks, Jeff.

Virginia

"Jeff Shannon" <jeff at ccvcorp.com> wrote in message
news:3C6C566A.A51AD1E1 at ccvcorp.com...


Virginia O wrote:

> I snagged this question from a thread below...
>
> >>OK, let's forget about the legal stuff for the moment.
>
>  More importantly, good manners. Let's say I'm wrapping up a Python
program
> with py2exe, and I'm going to distribute it. Which files would I be
expected
> to include as a matter of courtesy to the Python developers (and the
> developers of any additional modules that I use)? Just the licence files,
or
> is there anything else?
>
> Cheers,
> Simon Brunning
> >>>>>>>>>
>
> I also, being a newbie, don't understand what all this licensing jargon
> means in terms of distribution. If I write a program, and sell it, then
> what? I do understand I need to speak with a intellectual properties
lawyer,
> but I'd like to get a handle on things before I get started.'

The Python license explicitly states that this is fine.  First off, any code
you
write (whether in Python or in C or whatever) is yours to do with as you
please.  The Python license explicitly gives you permission to use code
modified
from the Python standard library, and to redistribute Python itself to
anyone
you want to.  You could presumably even charge money to redistribute
Python...
though, of course, since anyone can get it for free from the python.org
site,
it'd be a tough sell.  ;)  The only real restriction that the Python license
has, is that if something bad happens, you can't sue the PSF or the Python
dev
team over it.  So if that Python application dies and some client loses
thousands of dollars, well...  it's not the PSF's fault.  Whether or not
that
client can sue *you* depends on the license that *you* put on the code that
you
write, and whether that client agrees to it....

(Note, of course, standard disclaimers that I am not a lawyer nor an IP
expert;
this is merely an anecdotal layman's understanding of the license.)

Jeff Shannon
Technician/Programmer
Credit International







From mark at mceahern.com  Mon Feb 11 16:53:40 2002
From: mark at mceahern.com (Mark McEahern)
Date: Mon, 11 Feb 2002 13:53:40 -0800
Subject: subclass list
Message-ID: <NCBBLFCOHHDIKCAFGCFBKEEEKCAA.mark@mceahern.com>

Suppose I want to subclass list so that my subclass has optionally unique
values.  I'm using Python 2.2, so I start out like this:

class Bucket(list):

	def __init__(self, unique):
		self.unique = unique
		# Use a dict to keep track of unique values.
		if self.unique:
			self.keys = {}

Q:  How do I override append() so that I can determine whether to add the
value or not?

I tried:

    def append(self, value):
        if not self.unique or (self.unique and not
self.keys.has_key(value)):
            list.append(value)
            if self.unique:
                self.keys[value] = None

But then I get:

Traceback (most recent call last):
  File "./junk.py", line 31, in ?
    print getRandom(count=10)
  File "./junk.py", line 29, in getRandom
    bucket.append(random.randint(low, hi))
  File "./junk.py", line 22, in append
    list.append(value)
TypeError: descriptor 'append' requires a 'list' object but received a 'int'

Thanks,

// mark




From gmcm at hypernet.com  Fri Feb  1 08:47:03 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: 01 Feb 2002 13:47:03 GMT
Subject: Is thera a Metakit for Python 2.2  release ??
References: <11e94203.0201311433.519b556@posting.google.com>
Message-ID: <Xns91A8597CF9CEgmcmhypernetcom@199.171.54.215>

Jose Rivera wrote:

[snip]

>   File "D:\TOOLS\Python22\lib\metakit.py", line 12, in ?
>     from Mk4py import *
> ImportError: Module use of python21.dll conflicts with this version of
> Python.
> 
> It worked with Python 2.1, and I searched on the WEB and did not found
> a comment about a new release.

Official 2.2 builds should be appearing soon. Meanwhile, you can 
get Mk4py.dll for 2.2 (built from CVS; not quite official yet) from:

 http://www.mcmillan-inc.com/dnld/Mk4py2.2-cvs.zip

metakit.py not included (no changes required).

-- Gordon
http://www.mcmillan-inc.com/


From michael at stroeder.com  Wed Feb 20 17:08:26 2002
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 20 Feb 2002 23:08:26 +0100
Subject: HTTP state management without cookies?
References: <mailman.1014221673.28295.python-list@python.org>
Message-ID: <3C741E5A.E9EC1966@stroeder.com>

Geoffrey Talvola wrote:
> 
> Webware's WebKit app server has a different approach that doesn't require
> cookies and also doesn't require adding the session ID to all URL's.  It has
> an option to use "automatic path sessions".  If you turn on this option,
> then WebKit will redirect a request without a session ID like:
> 
> http://foo.bar/cgi-bin/WebKit.cgi/MyServlet
> 
> to:
> 
> http://foo.bar/cgi-bin/WebKit.cgi/_SID_=3678268432/MyServlet

It's just using PATH_INFO for the session ID which *is* part of the
URL.

I think people proposing use of Cookies were concerned about messing
up the bookmarks of the user with session IDs. It's much less of a
problem to produce the correct URLs within your app.

Ciao, Michael.


From James_Althoff at i2.com  Fri Feb  1 18:31:45 2002
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Fri, 1 Feb 2002 15:31:45 -0800
Subject: Defending the rodent (was Re: Defending the Python lanuage...)
Message-ID: <OF1C6BDF06.53914F74-ON88256B53.007F7BFC@i2.com>

[Alex Martelli]
>Now, if you type and edit fast, having your fingers stay close
>to the keyboard's home row has indisputable advantages wrt having
>them fly all over the place, to cursor keys or function keys or
>even worse separate devices like mice -- total travel distance and
>times for the fingers are greatly reduced.

Of course, if Douglas Englebart had had his way, this would all be moot.
Englebart -- who invented the computer mouse -- didn't like moving his hand
off the keyboard either.  So for his famous NLS system he devised a
one-handed "chording" keypad on which he typed (all characters) using his
left hand while manipulating the mouse with his right.  Decades later he
still maintained that this is *the* way to do it.  :-)

(QWERTY who?)

Jim




From gh_pythonlist at gmx.de  Tue Feb 19 04:51:10 2002
From: gh_pythonlist at gmx.de (Gerhard =?unknown-8bit?Q?H=E4ring?=)
Date: Tue, 19 Feb 2002 10:51:10 +0100
Subject: Python calling Java
In-Reply-To: <7xsn7yqmf5.fsf@ruckus.brouhaha.com>
References: <a4rjq4$4n8$1@web1.cup.hp.com> <ad052e5c.0202181707.69c22978@posting.google.com> <7xsn7yqmf5.fsf@ruckus.brouhaha.com>
Message-ID: <20020219095109.GD2611@lilith.hqd-internal>

Le 18/02/02 ? 17:40, Paul Rubin ?crivit:
> danb_83 at yahoo.com (Dan Bishop) writes:
> > > I have downloaded and tried out Jython, but it takes 5-6 seconds
> > > just to run a one-line 'print "hi"' script, and this is, of
> > > course, not acceptable for a web server.
> > 
> > Because you expect that running a ten-line script will take a minute?
> 
> Huh?  The 5-6 seconds is probably startup overhead so the ten-line
> script would take the same 5-6 seconds, which is already unacceptable
> for a web server.  Even 1 second is enough to make the server feel
> unresponsive.

A JVM just isn't suited for CGI scripts because of the startup overhead.
That's why I proposed to rewrite the whole stuff as Jython servlets.

Gerhard
-- 
This sig powered by Python!
Au?entemperatur in M?nchen: 4.3 ?C      Wind: 5.9 m/s



From op73418 at mail.telepac.pt  Wed Feb 27 17:18:14 2002
From: op73418 at mail.telepac.pt (Gonçalo Rodrigues)
Date: Wed, 27 Feb 2002 22:18:14 +0000
Subject: Status of PEP's?
References: <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p@4ax.com> <eppstein-546E75.11465727022002@news.service.uci.edu>
Message-ID: <ckmq7uo8l48cq9ig5qpgciu96gfc6sqb4r@4ax.com>

On Wed, 27 Feb 2002 11:46:57 -0800, David Eppstein
<eppstein at ics.uci.edu> wrote:

>In article <inaq7u8ca9rpgb21s5tgg3dgnqjerv7a0p at 4ax.com>,
> Gon?alo Rodrigues <op73418 at mail.telepac.pt> wrote:
>
>> Anyone knows how is the "acceptance status" of the following PEP's, that
>> is, is there any idea if they will be accepted for the next Python
>> realease:
>> 
>> PEP 276: Iterator for ints.
>> (I really love this one - as simple and "obvious" as it gets)
>
>I don't know an answer to your question, but in contrast to your opinion, 
>I really dislike this one.  It gives a very nonintuitive meaning for
>"x in y" when both are integers (equivalent to "x <= 0 < y"), and that 

Now you have me lost. As far as I understand the PEP only talks about
providing an iterator for integers. It does not talk about supporting
the in method for integers, which, as you state, would make up for a
confusing syntax.

>meaning would hold in all contexts, not just for-loops -- I would much 
>prefer that such expressions get flagged as likely coding error and that 
>more explicit syntax be required to create integer ranges.  I think there 
>should be a better syntax for for-loops over integers, but this isn't it.
>
>(See earlier discussion of PEP276 in this group for proposed alternatives, 
>I don't want to repeat the whole same discussion again here.)

I will do just that.

Thanx,
Gon?alo Rodrigues



From alanmk at hotmail.com  Thu Feb 28 10:34:08 2002
From: alanmk at hotmail.com (Alan Kennedy)
Date: 28 Feb 2002 07:34:08 -0800
Subject: Mod-python problem
References: <slrna7rdq5.2vf.bp@sam.localdomain>
Message-ID: <f72dac0d.0202280734.77bfa545@posting.google.com>

bp at sam.localdomain (Bob Parnes) wrote in message 
> Just in case I am missing something obvious, the following 
> is my test code, for a linux system:

Bob,

>From what I can see, your clTest class and cltest method are fine.
There are a few minor errors in the rest of the code though. I'll
repeat a complete version that works (for me anyway) at the bottom.

> #!/usr/bin/env python

While the shebang line is not a problem, it is not necessary for
mod_python. It is only necessary for CGI programs, so that the apache
daemon knows which interpreter to pass the script to.

> from mod_python import util

You're missing the following line here, see below for the reason why

from mod_python import apache

> 
> class clTest:
>     def cltest(self, req, formDict):

As I mentioned, these are fine, so I haven't repeated them.

> def main(req):

For mod_python, the default handler method is called "handler". You
can use other method names, but you have to explicitly mention this in
the httpd.conf, like so

#-- Extract from httpd.conf
<Directory /path/to/some/directory>
   AddHandler python-program .py
   PythonHandler formTest::main
</Directory>
#--

>     formDict = util.FieldStorage(req)
>     cl = clTest()

These two lines are fine.

>     return cl.cltest(req, formDict)

There are a couple of problems here.

1. The way to send output to the client is to use the "req.write"
method.
2. The return from the function should be a status value indicating
whether the request was successfully handled.

So this one line should be replaced with the two lines

    req.write(cl.cltest(req, formDict))
    return apache.OK

Note the "apache.OK" constant, which is why you need to import the
mod_python.apache module.

Also, you should really set a return content/MIME type for the output,
so the browser knows what to do with it. This is done like this

    req.content_type = "text/html"
    req.send_http_header()

So, the corrected version, which works for me, is as follows

# Script starts here

from mod_python import util
from mod_python import apache

class clTest:
    def cltest(self, req, formDict):
        res = '<html><head></head><body>'
        res += '<table align=center>'
        res += '<TR><TH colspan=2>Welcome to the \
            mod-python class test'
        res += '<TR><TD colspan=2>keys: %s' \
		    % formDict.keys()
        res += '<form method=post>'
        res += '<TR><TD>Name <TD><input type=text \
            name="name">'
        res += '<TR><TD>Phone <TD><input type=text \
            name="phone">'
        res += '<TR><TD colspan=2 align=center> \
            <input type=submit name=submit>'
        res += '</form></table></body></html>'
        return res

def handler(req):
    formDict = util.FieldStorage(req)
    cl = clTest()
    req.content_type = "text/html"
    req.send_http_header()
    req.write(cl.cltest(req, formDict))
    return apache.OK
    
# Script ends here

Hope that works.

Al.


From tim.one at home.com  Sun Feb 17 17:30:27 2002
From: tim.one at home.com (Tim Peters)
Date: Sun, 17 Feb 2002 17:30:27 -0500
Subject: Editable strings anyone?
In-Reply-To: <c76ff6fc.0202171321.4bd2e69c@posting.google.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECJNNAA.tim.one@home.com>

[John Machin]
> ...
> It's a pleasant surprise that the re module supports searching etc on
> 'c' arrays.

An even bigger surprise to most people is that the re module works on
mmap()'ed memory too (provided your platform supports the mmap module).

somebody-must-have-stayed-up-thinking-all-night-ly y'rs  - tim




From gmcm at hypernet.com  Fri Feb 15 09:05:21 2002
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 15 Feb 2002 09:05:21 -0500
Subject: ANN: Installer 5b2 (bug fix release)
Message-ID: <3C6CCF51.17132.3FBBB09@localhost>

Announcing version 5b2 of Installer:

 http://www.mcmillan-inc.com/installer_dnld.html

This is a minor bug fix release. Fixed:
 - Problems running in root directory of a drive
   on Windows
 - Error when used as Apache CGI on Linux

Platforms:
  Windows
  Linux
  Unix (relying on volunteers)

Python versions:
  tested with Python 1.5 to 2.2

What is it:
  The Installer package is any easy way to deploy
   Python apps to systems without Python installed 
   (or with incompatible Pythons installed).

   Installer support single-file and single-directory
   deployments. On Windows, Installer supports
   COM, including in-process COM servers.
 
License:
  MIT style.

Contact:
  gmcm at hypernet.com

<P><A HREF="http://www.mcmillan-
inc.com/installer_dnld.html">Installer 
Release 5b2</A>Easy Python app deployment. (15-Feb-
2002).




From marklists at mceahern.com  Sun Feb 10 12:55:37 2002
From: marklists at mceahern.com (Mark McEahern)
Date: Sun, 10 Feb 2002 09:55:37 -0800
Subject: box on soap
In-Reply-To: <3C66AEB9.4A1DF069@engcorp.com>
Message-ID: <NCBBLFCOHHDIKCAFGCFBIEPHKBAA.marklists@mceahern.com>

[Peter Hansen]
> ...Besides which, clicking
> on Mark's link brought up (in Netscape 4.7) a page which
> said "Directory Listing Denied:  This Virtual Directory does
> not allow contents to be listed."

The URL might not have wrapped properly.  I always have that problem.  Not
sure if that's why you couldn't get to it.

Cheers,

// mark




From henrik.motakef at web.de  Sun Feb 10 03:30:23 2002
From: henrik.motakef at web.de (Henrik Motakef)
Date: 10 Feb 2002 09:30:23 +0100
Subject: Animated graphs / wxPython
References: <a43rur$31f$1@newsg1.svr.pol.co.uk>
Message-ID: <86bsexzqlc.fsf@apo.hmotakef.int>

"Duncan Smith" <buzzard at urubu.freeserve.co.uk> writes:

> I need to frequently redraw a graph (ie. nodes and edges) to a wxPython
> panel.  ie. I remove and/or add edges to the graph and then redraw it.  I am
> currently using wxPython ellipses and lines for drawing the nodes and edges.

This isn't directly related to your question, but did you have a look
at the OGL library (included by default in wxPython, AFAIK)? Could
make things easier, scince it is all about diagrams of nodes and
connections between them.

hth
Henrik


From rony.steelandt at bucodi.com  Mon Feb  4 11:02:43 2002
From: rony.steelandt at bucodi.com (Rony)
Date: 4 Feb 2002 08:02:43 -0800
Subject: Question for the Python Community!
References: <_JU68.2091$3E5.208453@newsread2.prod.itd.earthlink.net>
Message-ID: <b34d3835.0202040802.2cc7d0c0@posting.google.com>

"Stede Troisi" <stede at computer.org> wrote in message news:<_JU68.2091$3E5.208453 at newsread2.prod.itd.earthlink.net>...

I'm using Python for quit qome years now and **think** i have the most
books available on the market... reason for that is that i'm
collecting just for personnal fun.

As for an e-book, for me it isn't a good solution for the moment for
following reasons :

- I don't like to read books on a PC
- The real e-books devices are still very expensive
- I couldn't put the e-book on my shelf or in my library :)

Remeber this answer is only MY answer ...

> I was thinking of doing an E-book (because I don't like the high price of
> regular books) on an introduction to programming and artificial intelligence
> in Python.
> 
> I was wondering if there would be interest in such a e-book and if it would
> sell?  I would love to here everyone's opinion.
> 
> Thanks,
> Stede


From starx at pacbell.net  Mon Feb 11 16:23:56 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 11 Feb 2002 21:23:56 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com> <suN98.25037$l93.4624653@newsb.telia.net> <mt0g6ukklms8q21qsrb2kkjoohblgq3bn8@4ax.com>
Message-ID: <Xns91B28848C4937RASXnewsDFE1@130.133.1.4>

Courageous || Mon 11 Feb 2002 09:41:35a:

> 
>>    "People tend to hold overly favorable views of their
>>    abilities in many social and intellectual domains. The
>>    authors suggest that this overestimation occurs, in
>>    part, because people who are unskilled in these
>>    domains suffer a dual burden: Not only do these
>>    people reach erroneous conclusions and make un-
>>    fortunate choices, but their incompetence robs them of the
>>    metacognitive ability to realize it" 
> 
> It's a wise enough observation. The concept of
> both individual and group assessment has been
> extensively studied. As it turns out, feeling one
> is better than other people on average is a normal,
> healthy aspect of human psychology. Only depressed
> people routinely rank themselves as average. Which
> is kind of funny; it appears that a healthy state
> of mind is somewhat delusional in this particular way.
> 
> With all that said, I know only one thing for sure:
> 
> I FUCKING ROCK AND YOU ALL SUCK!
> 
>:) :)

Considering the nature and meaning of you and i to depend on which person 
is speaking, i think we can all agree with that. :)

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From sholden at holdenweb.com  Fri Feb  8 18:12:02 2002
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 8 Feb 2002 18:12:02 -0500
Subject: [OT] PPT reader?
References: <mailman.1013181915.301.python-list@python.org> <00W88.108659$KM2.4246695@atlpnn01.usenetserver.com> <3C645B58.7274695A@bioreason.com>
Message-ID: <009801c1b0f6$08fc9960$0200000a@holdenweb.com>

----- Original Message -----
From: "Mitch Chapman" <Mitch.Chapman at bioreason.com>
Newsgroups: comp.lang.python
To: "Steve Holden" <sholden at holdenweb.com>
Sent: Friday, February 08, 2002 6:12 PM
Subject: Re: [OT] PPT reader?


> Steve Holden wrote:
> >     http://www.holdenweb.com/python/py10keynote.pdf
>
> Dangit.  This URL gives me 'HTTP/1.1 404 Object Not Found'.
>
> --
> Mitch Chapman
> Mitch.Chapman at bioreason.com
>
> Stuck on NT, without PPT
>
I deleted it! Go to

    http://www.python.org/doc/essays/ppt/

and you'll find Guido now provides PDF versions of the latest four
presentations.

regards
 Steve
--
Consulting, training, speaking: http://www.holdenweb.com/
Author, Python Web Programming: http://pydish.holdenweb.com/pwp/

"This is Python.  We don't care much about theory, except where it
intersects with useful practice."  Aahz Maruch on c.l.py






From shiver at yubc.net  Sat Feb 16 20:01:44 2002
From: shiver at yubc.net (I.J.)
Date: Sun, 17 Feb 2002 02:01:44 +0100
Subject: high precision mathematics
Message-ID: <a4mvdl$cl7$1@cer.yubc.net>

Is there a module for python that grants ussage of decimal numbers with more
than 14 decimal places, including all scientific functions to do with those
numbers?




From gaul at spam.utexas.edu  Tue Feb 12 07:55:23 2002
From: gaul at spam.utexas.edu (Andrew Gaul)
Date: Tue, 12 Feb 2002 12:55:23 +0000 (UTC)
Subject: Fun with lambda and map
References: <3c566a21.22129828@news.t-online.de> <slrna6i14c.sp.gaul@bifrost.cs.utexas.edu>
Message-ID: <slrna6i45q.sp.gaul@bifrost.cs.utexas.edu>

In article <slrna6i14c.sp.gaul at bifrost.cs.utexas.edu>, Andrew Gaul wrote:
> In article <3c566a21.22129828 at news.t-online.de>, Gerson Kurz wrote:
>> primes = lambda o:map(lambda a:filter(None,(map(lambda i:map(lambda x:a.__setitem__(x,0),range(2*i,o,i)),range(2,o)),a)[1])[1:],[range(o)])[0]
>> 
>> fibonacci = lambda x:map(lambda o:(map(lambda c:map(lambda l:o.__setslice__(l[0],l[1],l[2]),([o[2]+3,o[2]+4,[o[0]]],[0,3,[o[1],reduce(lambda x,o:x+o,o[:2]),o[2]+1]])),range(x)),o)[1],[[1,1,0]+range(x)])[0][3:]
>> 
>> (Note: These probably need python 2.2 to work because of the lambda
>> scope thing)
> 
> Here's a shorter implementation that works with both 1.5 and 2.2 that
> involves some cleverness:
> 
>     fibonacci = lambda z: map(lambda n, f=(lambda f,x,a,b: \
>     [x<=0 and b, x>0 and f(f,x-1,b,a+b)][x>0]): f(f,n,0,1), range(z))
> 
> It's not as nice as the Haskell version:
> 
>     fibonacci n = take n fib
>                   where fib = 1 : 1 : [ a+b | (a,b) <- zip fib (tail fib) ]

My apologies for following-up to my own post, but I've written something
more horrible:

    fibonacci = lambda z, f=(lambda f,x,l: \
    [x<=0 and l, x>0 and f(f,x-1,l+[l[-1]+l[-2]])][x>0]): f(f,z-2,[1,1])

Here's an explanation so that this post isn't totally frivolous:
Since we are trying to write the fibonacci in one line, it would seem
that recusion is impossible because any inner function will be a lambda
expression and thus nameless.  We can give this inner function a name
using a default parameter, but the inner lambda won't know about it
because the function name isn't bound yet.  If we pass the function name
as one of the parameters, we can use this to recurse.  There is a bit of
junk to emulate C's short-circuting ?: expression, which is documented
in the Python Cookbook
(http://aspn.activestate.com/ASPN/Python/Cookbook/Recipe/52310).

-- 
Andrew Gaul
http://gaul.org/


From bc90021 at bc90021redundantdot.net  Mon Feb 18 16:17:58 2002
From: bc90021 at bc90021redundantdot.net (bc90021)
Date: Mon, 18 Feb 2002 21:17:58 GMT
Subject: Newbie question...
References: <3C716B27.3090408@bc90021redundantdot.net>
Message-ID: <3C716F6C.7030708@bc90021redundantdot.net>

X-No-Archive: Yes

> error: failed dependencies:
>     rpmlib(PartialHardlinkSets) <= 4.0.4-1 is needed by python-2.2-7
>     libcrypto.so.3   is needed by python-2.2-7
>     libdb-4.0.so   is needed by python-2.2-7
>     libssl.so.3   is needed by python-2.2-7


I solved the "libdb-4.0.so" problem with the Rawhide DB4 RPM, so I'm 
continuing on the others...

bc90021



From philh at comuno.freeserve.co.uk  Wed Feb  6 08:34:07 2002
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Wed, 6 Feb 2002 13:34:07 +0000
Subject: Autocoding project proposal.
References: <un0yxkupb.fsf@synopsys.com> <2810.795T2138T13535728threeseas@earthlink.net> <slrna5hhm2.hij.sholden@pgrad.cs.usyd.edu.au> <1173.799T632T11653818threeseas@earthlink.net> <3c5dda29$0$79560$3c090ad1@news.plethora.net> <1372.801T1070T13994028threeseas@earthlink.net>
Message-ID: <slrna62c6f.afn.philh@comuno.freeserve.co.uk>

On Wed, 06 Feb 2002 04:54:05 GMT, Timothy Rue <threeseas at earthlink.net> wrote:
>On 03-Feb-02 19:47:37 Peter Seebach <seebs at plethora.net> wrote:
>>In article <1173.799T632T11653818threeseas at earthlink.net>,
>>Timothy Rue <threeseas at earthlink.net> wrote:
>>>Don't like the idea of autocoding huh?
>
>>Haven't been able to perceive whether or not you have an idea.  I have to
>>say, you're talking to the wrong crowd.  You need people less mired in
>>conventional ideas of software engineering.  Your best bet is almost
>>certainly the C# people; they are certainly not mired in conventional ideas
>>of software engineering.
>
>Why would  be interested in a programming language by a company that has
>been found guilty of Anti-trust? Not to mention that language is not
>available anywhere near as much as something like Python.

There is an open source implementation of C# as part of gcc, so I expect it runs
on pretty much everything.

-- 
===== Philip Hunt ===== philh at comuno.freeserve.co.uk =====
Herbivore, a zero-effort email encryption system. Details at:
<http://www.vision25.demon.co.uk/oss/herbivore/intro.html>





From jeff at ccvcorp.com  Tue Feb 19 14:44:57 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Tue, 19 Feb 2002 11:44:57 -0800
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <24c39b2c.0202190059.41874b92@posting.google.com>
Message-ID: <3C72AB39.28618CC3@ccvcorp.com>

Alexander Jerusalem wrote:

> And the final argument I have for static type checking is that it
> enables method dispatching based on parameter types. In a statically
> typed language you can create two methods that have the same name but
> differ on the parameter types. The correct method will be called for
> you depending on the type of the argument you pass in your call. That
> makes for a quite flexible way of extending a program. You can just
> add another method with the same name and another type to handle a
> special case without touching the existing methods.

In Python, you'd simply try to use the object in one way.  If that doesn't
work, catch the exception and try the next.  Then you can clearly see what
happens to different types of objects when you're applying the "same"
operation to them.  You don't need to use multiple methods to have
different behaviors, and at the same time, you don't have to add more
methods when you add a new type that you want to behave the same as an old
type...


> I'm not, however, completely satisfied with static type checking as
> well because I don't see why the compiler forces me to have a
> parameter comply with a static interface that has 10 methods when the
> method that uses the parameter acutally only uses one of those
> methods. I want the compiler to check if the parameter type I'm
> passing has what the method needs but nothing more.

This is exactly what Python does, except that it does it at runtime
instead of compile time; and really, it is *not* any slower to test this
than it is to try a couple of compile cycles.  Plus, as has been
previously noted, PyChecker will scan for most such errors if it's used...

Jeff Shannon
Technician/Programmer
Credit International




From bosahv at netscapenospam.net  Sun Feb 10 14:28:12 2002
From: bosahv at netscapenospam.net (Bo Vandenberg)
Date: Sun, 10 Feb 2002 19:28:12 GMT
Subject: python-mhash 1.0
References: <mailman.1013324724.16634.python-list@python.org>
Message-ID: <gRz98.356680$pa1.84576503@news3.rdc1.on.home.com>

Any chance there will be a windows Binary build of mhash sometime?

I'm curious, why was it removed?


A binary would make this _much_ easier to play with on WIN32.

Bo




From thomas.heller at ion-tof.com  Thu Feb 14 06:08:55 2002
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 14 Feb 2002 12:08:55 +0100
Subject: Problem freezing Win32 services
References: <a4g3s3$k0s$1@news.ost.eltele.no>
Message-ID: <a4g5sl$f89$1@ID-59885.news.dfncis.de>

"Alexander Wilkens" <alexw at fast.no.nospam> wrote in message news:a4g3s3$k0s$1 at news.ost.eltele.no...
> Hi,
>
> I'm attempting to freeze a Windows NT service (in Win2000) written in Python
> 2.2 using win32all 142 and the freeze util bundled with Python in the tools
> directory, and I've encountered some problems. Freezing console apps works
> fine.
I'm unsure about the state of freeze (at least for COM servers and services).
Maybe someone else can comment?

<shameless plug>
You could (should ;-) try out the newest py2exe release, which also can
create service exe-files: http://starship.python.net/crew/theller/py2exe/

Thomas





From aleax at aleax.it  Fri Feb  1 12:15:32 2002
From: aleax at aleax.it (Alex Martelli)
Date: Fri, 01 Feb 2002 17:15:32 GMT
Subject: Conventions for iterators
References: <mailman.1012555082.6918.python-list@python.org>
Message-ID: <U2A68.17631$3J1.540244@news2.tin.it>

Tim Peters wrote:
        ...
>>>> d = {1:1, 2:2}
        ...
> If you can't live with that, good!  Do the sensible thing and iterate over
> a *copy* (for k in d.keys(); for k, v in d.copy(); etc).

Presumably "for k, v in d.items()" -- "for k,v in d.copy()" would work only
if d's keys were all immutable sequences with two items [catching the
_timbot_ in a typo -- dies albo signanda lapillo...!-)].


Alex



From starx at pacbell.net  Fri Feb  8 22:36:11 2002
From: starx at pacbell.net (Philip Swartzleonard)
Date: 9 Feb 2002 03:36:11 GMT
Subject: Scope question
References: <a3u6aj$1b77q1$1@ID-98166.news.dfncis.de> <slrna664d7.d87.philh@comuno.freeserve.co.uk>
Message-ID: <Xns91AFC76818BBRASXnewsDFE1@130.133.1.4>

phil hunt || Thu 07 Feb 2002 03:45:43p:

> On Thu, 7 Feb 2002 10:26:02 -0500, MDK <mdk at mdk.com> wrote:
>>I have this:
>>
>># myprog.py
>>
>>from mymod import *
>>
>>def dosomething(x): dosomething(1000)
>>
>># End code
>>
>>I want the dosomething(1000) run from the function in mymod.  However,
>>instead it is running dosomething(x) in myprog.py. 
> 
> Do:
> 
> # myprog.py
> import mymod
> def dosomething(x):
>    mymod.dosomething(1000)
> 
>>import mymod
> 
> Thisi s the prefered way of doing things.

Yet another thing you can do, if you really want the import * too, is to 
just add 'import mymod' underneath it. This won't add anything but a 
name to the current namespace -- the module's already been initilized.

I wouldn't really reccommend it, but as i'm importing * from GL/GLU/GLUT 
and wx, i also import the modules themselfs so i can get there 
__dicts__, and filter all that crap out of the dictionary I pass to 
pycrust :).

I may stop doing this in the future though... i think i may make a 
module 'interface' and do 'form xxx import *'s in there, so then import 
it as 'i'. i.wxFrame, i.glutPostRedisplay, etc (it would also be nice if 
there was a sperate module of the GL's that didn't preface GL infront of 
everything. gl.vertex3d looks a lot better than gl.glVertex3d... :)

Ah.. goin' off on a tanget again...
Z'out

-- 
Philip Sw "Starweaver" [rasx] :: www.rubydragon.com


From NKluge at optonline.net  Sun Feb 17 14:25:19 2002
From: NKluge at optonline.net (Nicola Kluge)
Date: Sun, 17 Feb 2002 19:25:19 GMT
Subject: nested scopes
References: <CfTb8.129031$th4.36971458@news02.optonline.net>
Message-ID: <zsTb8.129039$th4.37009591@news02.optonline.net>

Nevermind, I found that it doesn't work in IDLE, but it works in cmdline
Python.






From joost_jacob at hotmail.com  Sun Feb 17 18:42:21 2002
From: joost_jacob at hotmail.com (J.Jacob)
Date: 17 Feb 2002 15:42:21 -0800
Subject: a portable text user interface ?
References: <13285ea2.0202080524.e2e7948@posting.google.com> <Xns91B074EB2FA5ahmebahhotmailcom@198.99.146.10> <Xgj98.24854$l93.4542796@newsb.telia.net> <13285ea2.0202100707.5fe84495@posting.google.com> <idN98.25036$l93.4624073@newsb.telia.net>
Message-ID: <13285ea2.0202171542.6282e493@posting.google.com>

I build a TextMap class for solving some of the problems, it is
available at
http://www.liacs.nl/~jjacob/textmap.py
It is (should be!) completely portable and has a demo when run
standalone.
It is only for output.  Input is not that hard with Tkinter.
The class can be used as a mixin class.  The .text attribute (type
Tkinter.Text) of the class can be used for adding functionality and
doing sophisticated stuff.
Any comments from people with Tkinter experience is very welcome since
i started studying Tkinter last week and i am sure there is a lot to
improve !
For example what is the most common way to bind e.g. <SHIFT-LEFTARROW>
to the horizontal scrollbar so it moves left, what key combo is most
common, does Tkinter have a key combo for that already ?


From db3l at fitlinxx.com  Tue Feb 12 22:34:25 2002
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Feb 2002 22:34:25 -0500
Subject: import images (wxPython)
References: <5kja8.21380$Cg5.1177018@news1.calgary.shaw.ca>
Message-ID: <uit92m4vy.fsf@ctwd0143.fitlinxx.com>

"Andrew" <anon at anon.com> writes:

> I was looking at some of the demo code and notice that they do an 'import
> images' in some files to get some predefined images.  Now I'm writing my own
> script and would like to make use of those images as well, but I get the
> following error...  ImportError: No module named images.
> I'm still learning python and am not 100% sure how the import statement
> works.  The images file is in the demo folder of the wxPython folder.  I
> tried things like import wxPython.demo.images, but that doesn't work either.
> Can somebody give me a hand?

That's just one of the modules that is part of the demo itself.  You
can't import it with wxPython.demo.images because the demo directory
of wxPython isn't designed as a Python package (it would have to have
an __init__.py file).  That is, the demo isn't really designed to be
included by other applications, at least not in terms of how it is
installed in the filesystem.

You could either include the demo directory in your sys.path (either
via an external definition of PYTHONPATH or by adding it to sys.path
in your code), or just make a copy of the images.py file and stick it
in your application directory and get it locally.

Or if you're just trying to borrow the images, you could just snarf
whichever functions you want from the file and include it in your
application :-)

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


From fperez528 at yahoo.com  Tue Feb 19 21:15:14 2002
From: fperez528 at yahoo.com (Fernando =?ISO-8859-1?Q?P=E9rez?=)
Date: Wed, 20 Feb 2002 02:15:14 +0000
Subject: Help with code
References: <a4votp$mmf$1@infocom.km.ua>
Message-ID: <a4vpik$ebl$1@peabody.colorado.edu>

Dmitry Savchuk wrote:

> class now:
> def?_init_(self):



it should be:
  def __init__(self):

Note the *two* underscores (__) on each side. Same goes for the __str__ 
method.

cheers,

f


From mwh at python.net  Tue Feb  5 04:56:48 2002
From: mwh at python.net (Michael Hudson)
Date: Tue, 5 Feb 2002 09:56:48 GMT
Subject: Inconsistence syntax in os module?
References: <3C5F731D.4060500@larsshack.org>
Message-ID: <u8za8qmf7.fsf@python.net>

Lars Kellogg-Stedman <lars at larsshack.org> writes:

> Howdy all,
> 
> I'm growing increasingly frustrated with the os module, in particular
> with its implementation of the system() and popen() calls.  The
> traditional (e.g., libc) implementation of both of these calls spawns
> a /bin/sh process to run the command, with the associated problems of
> shell metacharacter interpretation, the cost of an extra fork(), and
> so forth.

The os.system and os.popen functions call the libc system & popen
functions.  This is what most functions in os are; thin wrappers
around the libc functions of the same name.

If you want control, you know where to find it -- popen2, or using
fork & exec yourself.

Cheers,
M.

-- 
  Like most people, I don't always agree with the BDFL (especially
  when he wants to change things I've just written about in very 
  large books), ... 
         -- Mark Lutz, http://python.oreilly.com/news/python_0501.html


From chris.gonnerman at newcenturycomputers.net  Sun Feb 10 21:05:14 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Sun, 10 Feb 2002 20:05:14 -0600
Subject: Using pydoc
References: <a46upn$u92$2@gonzo.circuit.com>
Message-ID: <013101c1b2a0$9266b2c0$0101010a@local>

----- Original Message ----- 
From: "James Vahn" <jvahn+USENET at short.circuit.com>


> I've an existing (linux) script that begins "#!/usr/bin/python -i" and
> contains (amongst other things) the lines:
> 
>         import parted
>         import partition
> 
> Calls seem to be made like:
> 
>         drvobj = partition.Partition(drv)
>         drvobj.create_partition_table()
>         drvobj.commit_changes()
> 
> My question is:  How do I find documentation for those things?
> 
>     ~$ pydoc partition.Partition
>         no Python documentation found for 'partition.Partition'

Sounds like part of RedHat's "anaconda" (is that the right name?)
support/maintenance/configuration system.  As to where docs are,
I don't know.





From jhsh at maties.sun.ac.za  Tue Feb 26 06:45:08 2002
From: jhsh at maties.sun.ac.za (Jannie Hofmeyr)
Date: 26 Feb 2002 03:45:08 -0800
Subject: gnuplot woes, a second story
References: <mailman.1014696821.3069.python-list@python.org>
Message-ID: <f2730b27.0202260345.2d984578@posting.google.com>

chajadan <python at chajadan.net> wrote in message news:<mailman.1014696821.3069.python-list at python.org>...
> When I try to run gnuplot from my Python2.2 installation, I get an error 
> (listed below)
> 
> I'm on windows 98, I installed NumPy, I unzipped gp371w32.zip into my main 
> python directory, and I tried changing the command to the gnuplot program 
> FROM gnuplot_command = 'pgnuplot.exe' TO gnuplot_command = 
> 'c:\Python20\gp371w32\pgnuplot.exe'

Are you sure your path to gnuplot is correct? Shouldn't it be
'c:\Python22\...

I originally suggested changing the popen call to os, and it continues
to work for me. However, I cannot compare with your setup because I am
now under Win2000, using Python22, gnuplot compiled from cvs with the
latest pgnuplot.exe.

Jannie Hofmeyr


From python at chajadan.net  Wed Feb 13 06:00:39 2002
From: python at chajadan.net (chajadan)
Date: Wed, 13 Feb 2002 03:00:39 -0800
Subject: searching backwards in a string
Message-ID: <5.1.0.14.0.20020213025905.00a40b30@mail.chajadan.net>

The direction I'd say, should not be implied.

When I was beginning my ventures in programming, I would have said "Yes! 
Yes! It's so obviously implied, so let it be implied." but oddly enough, I 
now find myself seeing many non-trivial reasons to not imply the direction.

Besides, leaving a way to dictate the direction means you won't have to 
override the assumed direction that one weird day you realize you don't 
want it just to go in the direction ~it~ thinks is best ;)

--chajadan

At 10:33 PM 2/12/02 -0600, Skip Montanaro wrote:

>     Paul> here 50000 and -1 are the starting and ending locations.
>
>     Skip> If start > end isn't direction == -1 implied?
>
>     Paul> Nope.  I tried it and it didn't search backwards.
>
>That wasn't quite what I was asking.  Let me ask it a slightly different
>way.  Presuming the behavior can be changed so that searching in reverse
>works, shouldn't the ordering of start and end be sufficient to indicate the
>direction of the search?
>
>--
>Skip Montanaro (skip at pobox.com - http://www.mojam.com/)
>
>--
>http://mail.python.org/mailman/listinfo/python-list




From tjreedy at home.com  Tue Feb 12 10:55:55 2002
From: tjreedy at home.com (Terry Reedy)
Date: Tue, 12 Feb 2002 15:55:55 GMT
Subject: IPython 0.2.5
References: <a4ainp$gts$1@peabody.colorado.edu>
Message-ID: <fWaa8.39009$X64.14248751@news1.rdc2.pa.home.com>

"Fernando P?rez" <fperez528 at yahoo.com> wrote
> Portability: Linux (and other unices, including Mac OSX), Windows
XP. Should
> run fine on all WinNT, and probably also on Win9x (I can't test
that).

You wrote further on your web page: "Hopefully Windows users will
quickly migrate to XP (which is actually quite good in many respects)
and we'll bury Win95 and children were they should have been buried
years ago."

Don't hold your breath.  XP is a greater resource hog than ME is a
greather resource hog than W98 is a greater resource hog than W95, so
many (most?) users can't upgrade, even if they would want to pony up
the upgrade fee and run the risk (in some cases, certainty) of
disabling a hardware system and set of software than currently work.
(XP does NOT run everything that runs under Wxx.)

That said, I may download and see what happens on my Win98 system.

Terry J. Reedy





From jessw at loop.com  Mon Feb 18 01:27:37 2002
From: jessw at loop.com (Jesse W)
Date: Sun, 17 Feb 2002 22:27:37 -0800
Subject: IDLE detection code snippet
Message-ID: <3C702E59.7396.FDF741@localhost>

Dear IDLE-devl members,
	A problem that has been faced by people many times is that of 
detecting when IDLE is running.  I think this code is a solution to 
that problem.  Please try it out and see if it works consistently.

						Jesse Weinstein

def IDLEtest():
    """Returns 1 when IDLE is running, 0 else.
    Please let me know (through the IDLE-devl list) if you find
    a situation where this gives the wrong answer."""
    import sys
    try:
        if sys.stdin.__module__=='PyShell':
            return 1
        else:
            return 0
    except AttributeError:
        return 0




From chris.gonnerman at newcenturycomputers.net  Tue Feb 12 21:59:06 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 12 Feb 2002 20:59:06 -0600
Subject: newbie asking for info about Python
References: <xkka8.2932$VX6.245554@weber.videotron.net>
Message-ID: <00b701c1b43a$6e07eac0$0101010a@local>

I run Pygmy, which is an email client written in Python; CGI programs for the web, even entire web servers
(Zope for instance), pretty much anything can be done in Python.  Operating systems are about the only
thing it's not suitable for.
  ----- Original Message ----- 
  From: Joseph Youssef 
  Newsgroups: comp.lang.python
  To: python-list at python.org 
  Sent: Tuesday, February 12, 2002 8:21 PM
  Subject: newbie asking for info about Python


  Hello, I know nothing about Python but before I start learning I would like to know what Python is usually "used" for, I mean I don't think it's made to write some kinda application or something, so what exaclty is it used for?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20020212/f8d19520/attachment.html>

From fredrik at pythonware.com  Mon Feb 11 05:59:36 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 11 Feb 2002 10:59:36 GMT
Subject: Fear of autocoding!
References: <b03e80d.0202060227.368f6ea7@posting.google.com> <b34d3835.0202080025.1d69d198@posting.google.com> <7934d084.0202101718.404404d1@posting.google.com> <7643.807T2626T856092threeseas@earthlink.net> <b5ne6ug62r4ltpcuh4jmc3tg808kncf852@4ax.com>
Message-ID: <suN98.25037$l93.4624653@newsb.telia.net>

Courageous wrote:

> >Very good example of someone who is afraid of autocoding.
>
> You never did answer my question, by the way. If one has
> become delusional, how does one become aware of it?

for some reason, this whole thread reminds me of this little gem:

    Unskilled and Unaware of It: How Difficulties in
    Recognizing One's Own Incompetence Lead to
    Inflated Self-Assessments (Kruger & Dunning, 1999)

    "People tend to hold overly favorable views of their
    abilities in many social and intellectual domains. The
    authors suggest that this overestimation occurs, in
    part, because people who are unskilled in these
    domains suffer a dual burden: Not only do these
    people reach erroneous conclusions and make un-
    fortunate choices, but their incompetence robs
    them of the metacognitive ability to realize it"

    http://www.apa.org/journals/psp/psp7761121.html

of course, Rue is going to argue that this doesn't apply to him, but
to everyone else.  after all, he KNOWS he's smarter than everyone
else, and we're just too stupid to realize how stupid we are ;-)

</F>




From mpoulin at verinet.com  Sat Feb  2 10:43:13 2002
From: mpoulin at verinet.com (Marc Poulin)
Date: Sat, 02 Feb 2002 10:43:13 -0500
Subject: Python System Administration
References: <LwK68.7172$Xk4.381742@news1.east.cox.net>
Message-ID: <3c5c241a$0$2203$7586b60c@news.frii.com>

In article <LwK68.7172$Xk4.381742 at news1.east.cox.net>, "Andrew Replogle"
<replogle2 at cox.net> wrote:

> Anyone know of any books out there for python system administration or
> should I just stick to perl for these tasks?

I've been wondering the same thing.I searched Amazon and found

	Perl for System Administration by David N. Blank-Edelman
	Automating Windows With Perl by Scott McMahan 

but nothing for Python.

I guess we could just buy the Perl books and translate the code into Python.


From threeseas at earthlink.net  Wed Feb  6 23:28:54 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Thu, 07 Feb 2002 04:28:54 GMT
Subject: (Long) Re: Autocoding project proposal.
References: <8775.795T1852T13954726threeseas@earthlink.net> <uelk6750s.fsf@synopsys.com> <27327.798T2437T205791threeseas@earthlink.net> <uadup5dsf.fsf@synopsys.com> <4898.802T2975T395701threeseas@earthlink.net> <ug04ey1on.fsf@synopsys.com>
Message-ID: <1731.803T2524T145894threeseas@earthlink.net>

On 06-Feb-02 12:12:24 David Masterson <dmaster at synopsys.com> wrote:


>That is putting the cart in front of the horse.  We cannot create a
>tool to solve a problem until we know what the problem is.

That's not true. Not only has the problem been identified but it has been
pointed out that it is not a simple one time problem/solution matter.

The symbols of mathmatics weren't designed to solve one problem, they
were designed to solve all problems in that class of problems.

I have pointed out that programming is the act of automating complexity
that is made up of simpler things, in order to make use of complexity
simple for the users of that complexity.

The problems is that there is no general automation tool oriented to all
users. Yet the essence of programming boils down to two deminsional bit
flipping.

I've identified and defined the tool and it's natural components.

The problem is in getting an industry that has been programmed to depend
on some level of manual control over what the users can and can't do, to
let go of such control and make available such a natural automation tool
as identified, so that the user can do things for themselves, rather than
being dependant on the programmers and whether or not they have the
resources to get a programmer to do what they need done.

The problem is perhaps to prevasive for you to see?

---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From jeff at ccvcorp.com  Thu Feb  7 15:17:26 2002
From: jeff at ccvcorp.com (Jeff Shannon)
Date: Thu, 07 Feb 2002 12:17:26 -0800
Subject: threads killing other threads
References: <1013105970.135322@cswreg.cos.agilent.com>
Message-ID: <3C62E0D5.99C586DF@ccvcorp.com>


Greg Weeks wrote:

> I've occasionally wanted thread objects to have an exit() method, so that
> one thread could interrrupt and kill another.  This doesn't seem to be
> available, perhaps because POSIX threads don't support it.
>
> Is that correct?
>
> Is there some reason why threads killing threads would be undesirable or
> hard to implement?

I don't know the reasoning behind threads being unkillable (from outside),
but I do know the (apparently standard) way to accomplish what you want.

Create a global threading.Event object, named something descriptive like
Quit, and have every thread peek at it occasionally.  If the Quit event is
set, then the thread aborts.  If you need to be able to kill specific threads
individually, then you have a separate event object for each thread.

Of course, this *does* rely on the threads being well-behaved, and not
getting stuck in an infinite loop elsewhere, and so on....  but then, that's
pretty much a requirement anyhow.

Jeff Shannon
Technician/Programmer
Credit International




From threeseas at earthlink.net  Fri Feb 15 20:45:42 2002
From: threeseas at earthlink.net (Timothy Rue)
Date: Sat, 16 Feb 2002 01:45:42 GMT
Subject: Apology
References: <a40urp$o10$1@peabody.colorado.edu> <558.804T2090T11654864threeseas@earthlink.net> <3C68274C.F6F702F5@ccvcorp.com> <mailman.1013474362.26226.python-list@python.org> <slrna6ie79.2so.philh@comuno.freeserve.co.uk>
Message-ID: <2320.811T442T11913096threeseas@earthlink.net>

On 12-Feb-02 10:46:49 phil hunt <philh at comuno.freeserve.co.uk> wrote:
>On Mon, 11 Feb 2002 18:37:51 -0600, Chris Gonnerman
><chris.gonnerman at newcenturycomputers.net> wrote:
>>
>>Indeed.  He uses a nonstandard definition of "language" and applies
>>"physics"
>>to programming.  He can't program, you know; ten years on this project and
>>nothing to show for it.  If I had tried to develop something like this for
>>ten years and had nothing to show for it but a website, I'd give up.

>At least his code is all neatly lined up!

Chris's comprehension level is greatly lacking and you've shown your
superiority attitude.

Ten years???? it was Feburary 1988 when I identified the nine commands.

I never claimed to be a programmer and never wanted to make the VIC but
rather someone to use it. But I also don't, and never will want to be
abused by any programmer wanting to take credit and property rights to the
VIC.

The computer industry has alot of ethical problems in this area of taking
everything the can from the users, from their ideas, to their debugging
efforts (feedback), to their money in payment for what came from them, and
now rental....and always it is never quite right but lacking always... for
upgrading recycling of consumer entrapment abuse.

That's what evidence ..... well what is, is what is. You know what you
are.

It's good to know that such practice is going to soon come to an end.
.net is just another hype distraction. What would autocoding do to
distributed development, distributed computing and web services, when
these things are developer focus and autocoding goes beyond that to the
consumer, not to mention what all else the same tool set can be used for
to help the consumer do things by themselves?

There never will be enough programmers, but there is always exactly the
perfect number of user to do all the things a user needs to do for
themselves. Given the right freedom for the users, creating tools.


---
*3 S.E.A.S - Virtual Interaction Configuration (VIC) - VISION OF VISIONS!*
   *~ ~ ~      Advancing How we Perceive and Use the Tool of Computers!*
Timothy Rue      What's *DONE* in all we do?  *AI PK OI IP OP SF IQ ID KE*
Email @ mailto:timrue at mindspring.com      >INPUT->(Processing)->OUTPUT>v
Web @ http://www.mindspring.com/~timrue/  ^<--------<----9----<--------<



From clarence at netlojix.com  Mon Feb 18 18:25:06 2002
From: clarence at netlojix.com (Clarence Gardner)
Date: Mon, 18 Feb 2002 23:25:06 GMT
Subject: urllib2 module in Python2.2, with SSL
Message-ID: <m3gc8.149$Mk5.5661@newsfeed.avtel.net>

I'm having a problem with the HTTPS (or HTTPSConnection)
classes in urllib2.

If I use the built-in SSL sockets and do writes and reads
myself, everything works fine, but if I try to use to higher-
level functionality (to the same URL), I get

  File "/usr/local/lib/python2.2/httplib.py", line 728, in getreply
    response = self._conn.getresponse()
  File "/usr/local/lib/python2.2/httplib.py", line 572, in getresponse
    response = self.response_class(self.sock)
  File "/usr/local/lib/python2.2/httplib.py", line 98, in __init__
    self.fp = sock.makefile('rb', 0)
  File "/usr/local/lib/python2.2/httplib.py", line 607, in makefile
    buf = self.__ssl.read()
socket.sslerror: (5, 'EOF occurred in violation of protocol')


This seems to point to a possible problem in the makefile method 
of the FakeSocket class (used to make an SSL socket look like a
plain socket).

Has anyone used urllib2 functions with SSL successfully?

-- 
Clarence Gardner
Software Engineer
NetLojix Communications
clarence at netlojix.com


From peter at zack.com  Sat Feb  2 03:27:01 2002
From: peter at zack.com (Peter Traub)
Date: Sat, 02 Feb 2002 08:27:01 GMT
Subject: [NEWBIE] mod_python, DBMs, and persistence
Message-ID: <3C5BA30A.3ED29BE9@zack.com>

Hello all,

We have a Python CGI script that uses various DBM files having a few
thousand records (someday, maybe more).  Problem: It gets slower when
it's used with the bigger DBMs, and it gets really slow beyond about 2
users. There is no sophistication to this CGI (yet). It has to start
python and connect to the DBMs for every invocation. What is the best
way to remove this overhead? Use mod_python, fastcgi, or persistent CGI?
'Best' is defined as stability more than speed.

Also, how efficient is it to use DBMs as opposed to a DB such as MySQL?

And one final question... Would anyone run a business using mod_python?
fastCGI?

Many thanks in advance.

Peter




From fredrik at pythonware.com  Wed Feb 13 13:58:22 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 18:58:22 GMT
Subject: Coming of age...
References: <mailman.1013466258.26610.python-list@python.org>
Message-ID: <iHya8.25576$l93.4762435@newsb.telia.net>

Paul Prescod wrote:

> This is one more datapoint over a few I've collected in the last few
> weeks that somehow Python has crossed some kind of line from being a
> "like to try it someday" language to an "everybody cool is using it
> these days" language. Despite waiting for this moment for several years
> I am somewhat, er, nervous about what this means for the future...

turning into a "sued by kodak" language?

http://news.com.com/2100-1001-836322.html

from one of the patents mentioned in that article:

    "An object based data processing system including an extensible
    set of object types and a corresponding set of "object managers"
    wherein each object manager is a program for operating with the
    data stored in a corresponding type of object. The object managers
    in general support at least a standard set of operations. Any
    program can effect performance of these standard operations
    on objects of any type by making an "invocation" request. In
    response to an invocation request, object management services
    (which are available to all object managers) identifies and invokes
    an object manager that is suitable for performing the requested
    operation on the specified type of data. A mechanism is provided
    for linking data from one object into another object. An object
    catalog includes both information about objects and about links
    between objects."

replace "object manager" with "class", and "object catalog" with
module, and what do you get? ;-)

</F>




From shalehperry at attbi.com  Wed Feb 20 11:06:09 2002
From: shalehperry at attbi.com (Sean 'Shaleh' Perry)
Date: Wed, 20 Feb 2002 08:06:09 -0800 (PST)
Subject: two ideoms at one blow: line-reading and regexp-matching
In-Reply-To: <m2vgcs5hgf.fsf@desk.crynwr.com>
Message-ID: <XFMail.20020220080609.shalehperry@attbi.com>

On 20-Feb-2002 Russell Nelson wrote:
> There are two Python ideoms that bother me:
> 
>     while 1:
>         line = file.readline()
>         if not line: break
> 
> and:
> 
>     matched = re.match(r'egexp(.*)', line)
>     if matched:
>         print matched.group(1)
> 
> I'd really, really like it if you could do one(1) assignment within
> the boolean expression in a while and an if.  Like this:
> 

this is a design decision in the language.  By not allowing this style of
programming we do not have the ever so fun:

if foo = 2:
   handle_foo()

type bugs.  Sure there are ways around it.  But python's strength is that it is
easy to write good code and easy on the reader.  Hiding assignments in loops
and tests do not make it easier for the next reader.

With 2.x and later the first problem has largely gone away.  The new idiom is
more like:

for line in file.xreadlines():
    parse(line)

xreadlines() does not load the whole file into memory and is generally about as
efficient as you can get.

The second item does however make re's more annoying in python than they are in
other languages where re's are native to the language.



From Mitch.Chapman at bioreason.com  Tue Feb  5 18:37:50 2002
From: Mitch.Chapman at bioreason.com (Mitch Chapman)
Date: Tue, 05 Feb 2002 16:37:50 -0700
Subject: ZODB.FileStorage -- copy?  save as?
Message-ID: <3C606CCE.ACA1472D@bioreason.com>

I'd like to be able to do something like "save as" with
ZODB.FileStorage.FileStorage or a derivative thereof.  I'm having a 
hard time even identifying alternatives, mostly because I don't yet
understand the FileStorage implementation.

F'rinstance I don't know if/when it's safe to copy the associated 
files ('foo', 'foo.tmp'), whether or not one needs to copy the tmp
file, what the tmp file does, how one might attach an existing DB 
to a new FileStorage instance, etc.

Can anyone provide clues?  So far all I have found is the source 
code itself :) and 
	http://www.zope.org/Documentation/Developer/Models/ZODB/

Thanks for any help!
-- 
Mitch Chapman
Mitch.Chapman at bioreason.com


From phd at phd.pp.ru  Thu Feb 14 13:07:14 2002
From: phd at phd.pp.ru (Oleg Broytmann)
Date: Thu, 14 Feb 2002 21:07:14 +0300
Subject: zlib.so and libz.so and Python
In-Reply-To: <u6nuqr108os0ae@corp.supernews.com>; from DeepBleu@DeepBleu.org on Thu, Feb 14, 2002 at 11:59:40AM -0600
References: <u6niisr909gtd6@corp.supernews.com> <mailman.1013703624.30857.python-list@python.org> <u6nuqr108os0ae@corp.supernews.com>
Message-ID: <20020214210714.M7398@phd.pp.ru>

On Thu, Feb 14, 2002 at 11:59:40AM -0600, DeepBleu wrote:
> % ldd zlib.so
>          libz.so =>       /usr/openwin/lib/libz.so
> 

> % ldd zlib.so
>          libz.so =>       (file not found)

   export LD_LIBRARY_PATH=/usr/openwin/lib

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



From martin.franklin at westgeo.com  Tue Feb  5 11:50:25 2002
From: martin.franklin at westgeo.com (Martin Franklin)
Date: Tue, 5 Feb 2002 16:50:25 +0000
Subject: My own list object
References: <a3p01b$186eq8$1@ID-98166.news.dfncis.de> <a3p0ss$19ob7s$1@ID-11957.news.dfncis.de> <a3p190$19v1rc$1@ID-98166.news.dfncis.de>
Message-ID: <a3p2fg$49g$1@mail1.wg.waii.com>

MDK wrote:

> 
> "Emile van Sebille" <emile at fenx.com> wrote in message
> news:a3p0ss$19ob7s$1 at ID-11957.news.dfncis.de...
>>
>> "MDK" <mdk at mdk.com> wrote in message
>> news:a3p01b$186eq8$1 at ID-98166.news.dfncis.de...
>> > Hello,
>> >
>> > I am trying to create my own special type of list object.
>> >
>> > However, when I do x.append('yo') it uses Python's append instead of
>> the one
>> > from my class.
>> >
>> > How can I get it to use my append method?  I have tried def __append__
>> and
>> > then called my own class' function but that did not work.  I've played
>> with
>> > __getattr__ but that did not help.
>> >
>>
>> Call the method append and not __append__?
>>
>>
>> --
>>
>> Emile van Sebille
>> emile at fenx.com
> 
> Yes, I figured that to be the case because it should be transparant to the
> caller.  However, that is not the solution.
> 
> Thanks though.
> 
> 
> 

Post some code.  perhaps the answer is obvious...


for example


>>> class MyList:
...     def __init__(self):
...             self.list=[]          
...     def append(self, item):
...             self.list.append(item)
... 
>>> ml=MyList()
>>> ml.append(1)
>>> ml.list
[1]
>>> 

would apear to work?









From rich at worldsinfinite.com  Tue Feb  5 14:51:16 2002
From: rich at worldsinfinite.com (Rich Harkins)
Date: Tue, 5 Feb 2002 14:51:16 -0500
Subject: how can i check if a variable is set?
In-Reply-To: <a3pd1j+joko@eGroups.com>
References: <a3pd1j+joko@eGroups.com>
Message-ID: <200202051951.g15JpHV04104@localhost.localdomain>

How 'bout:

try:
	a			# Or _junk_unused_variable_=a
except NameError:
	a=0

Rich

On Tuesday 05 February 2002 02:50 pm, maximilianscherr wrote:
> how cna i check if a variable is set?
> like:
>
> if variable a doesn't exist then set it to 0
> if it exists do nothing...



From tim at worthy.demon.co.uk  Mon Feb 11 06:46:09 2002
From: tim at worthy.demon.co.uk (Tim Howarth)
Date: Mon, 11 Feb 2002 11:46:09 GMT
Subject: Long int to int
References: <dec413074b.tim@worthy.demon.co.uk> <7xeljsp90d.fsf@ruckus.brouhaha.com>
Message-ID: <69c619074b.tim@worthy.demon.co.uk>

In message <7xeljsp90d.fsf at ruckus.brouhaha.com>
          Paul Rubin <phr-n2002a at nightsong.com> wrote:

> Tim Howarth <tim at worthy.demon.co.uk> writes:
> > (Or worse, I can convert with eval
> > 
> > a = str(hex(a))[2:-1]
> > a = eval('0x'+a)
> 
>   a = eval('0x%x' % a)
> 
> might be a little more palatable.  

Nicer but I'm not sure that it works, I want

0xFFFFFFFFL
to become
-1

besides I grew up believeing eval==evil.


-- 
___
 |im    ---- ARM Powered ----


From timr at probo.com  Mon Feb 25 01:19:01 2002
From: timr at probo.com (Tim Roberts)
Date: Sun, 24 Feb 2002 22:19:01 -0800
Subject: comp.lang.python in English? (was Re: Why Python is like BASIC)
References: <mailman.1014418936.10783.python-list@python.org> <3C799833.AC08A401@cosc.canterbury.ac.nz> <3C79C1B7.AB9287D0@engcorp.com>
Message-ID: <bnlj7u8kteb8he8v9p403051ld9pd77cm2@4ax.com>

Peter Hansen <peter at engcorp.com> wrote:

>Greg Ewing wrote:
>> 
>> Dies Komputermaschine ist nicht fur gerfingerpoken
>> und mittengrabben. Ist easie schnappen der Springenwerken,
>> mit Poppenkorken und Spitzensparken. Die rubbernecken
>> Seitseer ist keepen der Hans in der Pocketsrelaxen,
>> watchen das Blinkenleits.
...
>
>ACHTUNG! ACHTUNG!
>
>Alles touristen und non-technishen looken peepers! Das machine is nicht 
>fur der fingerpoken und mittengrabben. Is easy schnappen der springenwerk, 
>blowenfusen und popencorken mit spitzen sparken. Das machine is diggen by 
>experten only. Is nicht fur gerwerken by das dummkopfen. Das rubbernecken 
>sightseeren keepen das cottenpiken hands in das pockets.  Relaxen und 
>watchen das blinkenlights.
>
>--------------------
>
>Apparently German, as Perl, has many ways of saying the same thing.

It's worth pointing out, just for those who might be thinking differently,
that the only actual German in the previous posting is the word
"ACHTUNG"...
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From claird at starbase.neosoft.com  Wed Feb 13 12:22:47 2002
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 13 Feb 2002 11:22:47 -0600
Subject: Python Gotcha with Octal Numbers
References: <mailman.1013611523.15053.python-list@python.org> <130220020751360934%jwbaxter@spamcop.net>
Message-ID: <2D484E1BB24678AC.1496996E81BD013F.73A9A3A001BEE345@lp.airnews.net>

In article <130220020751360934%jwbaxter at spamcop.net>,
John W. Baxter <jwbaxter at spamcop.net> wrote:
			.
			.
			.
>Did the elves at AT&T pick up the "leading 0 means octal" from
>something earlier, or did they invent this stupidity themselves?
			.
			.
			.
OK, folks, how far back *can* we trace this?  My
mind associates it with DEC systems, going back
to the '60s, but I couldn't find any confirmation
of that in a quick search.
-- 

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


From codetube at yahoo.com  Fri Feb  1 04:26:51 2002
From: codetube at yahoo.com (CodeTube)
Date: 1 Feb 2002 01:26:51 -0800
Subject: How to change the icon in window header of window developed using python?
Message-ID: <5d65cf44.0202010126.7b58df35@posting.google.com>

Hello Everyone,

Could some one please tell me how to change the icon (a red one saying
TK) in the window title bar of a window generated by python usging
tkinter.
I have developed an application using python and the gui using
Tkinter. However i am not able to change the red 'TK' icon to
something more customized.

Please HELP!!!!

CodeTube


From chris.gonnerman at newcenturycomputers.net  Wed Feb 20 07:57:16 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Wed, 20 Feb 2002 06:57:16 -0600
Subject: Adding static typing to Python
References: <24c39b2c.0202181531.187fad4c@posting.google.com> <ghp47u02vceavgohdod9c8vbs9u5po91bt@4ax.com> <24c39b2c.0202191646.776c1993@posting.google.com> <55367u4imqamlmv8ou707t40dk1jolrkmc@4ax.com> <24c39b2c.0202192354.765da777@posting.google.com>
Message-ID: <005501c1ba0e$a6c94000$0101010a@local>

----- Original Message -----
From: "Alexander Jerusalem" <ajeru at vknn.org>


> That's obviously not the opinion of the members of the "The Python
> type system SIG" including Guido van Rossum, otherwise they wouldn't
> consider adding static types as an optional feature.
>
> Alexander
>
>
> Daniel Klein <danielk at aracnet.com> wrote in message
news:<55367u4imqamlmv8ou707t40dk1jolrkmc at 4ax.com>...
> > The point is that it ceases to be the same language with the same
attractions
> > (or flaws) depending on how you look at it.
> >
> > Dan

Static types as an optional feature for function/method interfaces is
not a bad thing, and does not really change the language; but the key
word here is "optional."  If static typing of function arguments is
optional, I for one am not likely to use it, so the language hasn't
changed.

Making it mandatory, now, THAT changes the language.

Static typing of *all variables* (destroying type inference and
incidental generality) is another beast entirely; but if it were
possible without killing performance (for I'm not sure that it
would actually improve performance after all), if it's optional
then nothing is harmed.

Static typing is used by compilers to improve performance, I'm told;
but to do it with an interpreter would require implicit typechecking
by the interpreter core.  This would probably be faster than explicit
typechecking (as in, write it yourself) but I can't help but see it as
a performance loss vs. the usual method (no typechecks at all,
exceptions thrown for interface mismatches).





From p.mayers at ic.ac.uk  Wed Feb 20 11:30:39 2002
From: p.mayers at ic.ac.uk (Phil Mayers)
Date: Wed, 20 Feb 2002 16:30:39 +0000
Subject: Large integers from C
Message-ID: <20020220.163039.1504569917.30753@ic.ac.uk>

All,

I need to create Large (>300 bit) python Long's from C. I don't need (and
in fact, don't want) to use the features of the (apparently obselete?)
mpz module, or gmpy, or whatever. I simply want to say:

val = PyLong_FromOctets(char *v, int vlen);

...where v is a MSB- or LSB-first octet string, base 256. This function
or an equivalent appears to be completely lacking from the Python C
interface.

I just know someone is going to say "return the octet string and do this:

intval = reduce(lambda x,y: (x<<8)+ord(y), strval)

I've tried, and this was too slow (the whole point of the C module being
to speed this up).

Any ideas?

Cheers,
Phil


From fredrik at pythonware.com  Wed Feb 13 17:12:11 2002
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 13 Feb 2002 22:12:11 GMT
Subject: How to generate XML
References: <mailman.1013628862.14331.python-list@python.org>
Message-ID: <%wBa8.25609$l93.4772704@newsb.telia.net>

Michael Chermside wrote:
> >>> I'm amazed that people always start proposing difficult-to-use
> >>> libraries when this question comes up. I'd prefer to use
> >>>
> >>> print "<foo/>"
>
> I tend to agree with the original poster: simply writing your own print
> routines isn't difficult at all.

in my experience, it's not worth it: you always end up forgetting to
call cgi.escape at the right places, encode non-ascii strings properly,
make sure that your program doesn't explode when some function
returns a unicode string instead of a plain string, etc.

using print to write XML is not as bad as using RE to parse it, but
it's far from optimal.

Traceback (most recent call last):
  File "effbot.py", line 105, in ?
  File "SimpleXMLWriter.py", line 103, in end
    assert self.__tags, "unbalanced end(%s)" % tag
AssertionError: unbalanced end(F)




From wurmy at earthlink.net  Fri Feb 15 20:51:28 2002
From: wurmy at earthlink.net (Hans Nowak)
Date: Sat, 16 Feb 2002 01:51:28 GMT
Subject: XML and newlines
Message-ID: <3C6DBB49.C9C5ED0B@earthlink.net>

Howdy y'all,

Recently I've been dabbling with XML and the xml.dom.minidom 
module, with good results so far. Except for one problem I
encountered, that got me stumped. 

For my current project I need to store the values of an
instance's attributes in XML. We're using SQLXML to talk to
an SQL Server 2000 database, and changes in tables are made
by sending updategrams, in XML formats. To store an instance
of an object in the database, all its attributes need to be
captured in an XML "expression". This works well for all
attributes (mostly just strings and numbers) except one, a
long multiline string containing newlines.

I manage to store that string just fine; only when I retrieve
the record, all the newlines are gone. Since I wasn't sure
whether the newlines were sent in the first place, I used
the <?lb?> tag, as seen in the book "Learning XML". :-) So
that part of the XML may look like this:

  one
  <?lb?>
  two
  <?lb?>
  three

This at least stores the newlines, and when I get back the 
raw XML from the database (when retrieving a record), the
newlines are there. There's also a great deal of unwanted
tabs, but those are easy to get rid of. But as soon as I
parse the XML with xml.dom.minidom.parseString, all these
newlines are gone.

Right now I have a temporary solution that extracts the
desired data (including the newlines) from the raw XML string
by using a regular expression. This is not so elegant 
though, and I still have to manually convert things like
"&amp;" and friends. Is there a better way to do this? What
am I missing?

TIA,

-- 
Hans (base64.decodestring('d3VybXlAZWFydGhsaW5rLm5ldA==')) 
# decode for email address ;-)
The Pythonic Quarter:: http://www.awaretek.com/nowak/


From chris.gonnerman at newcenturycomputers.net  Tue Feb 12 08:44:18 2002
From: chris.gonnerman at newcenturycomputers.net (Chris Gonnerman)
Date: Tue, 12 Feb 2002 07:44:18 -0600
Subject: ActiveX Questions
References: <mailman.1013487810.369.python-list@python.org> <a4b3ok$1dm4mc$1@ID-7529.news.dfncis.de>
Message-ID: <001601c1b3cb$5f6879e0$0101010a@local>

----- Original Message -----
From: "Karl M. Syring" <syring at email.com>


> "Chris Gonnerman" <chris.gonnerman at newcenturycomputers.net> schrieb
> >
> > The control is called "REPORTOCX.ReportOcxCtrl.1" (why does the name
> > have to be such a pain to find?)  I can instantiate it:
> >
> > >>> o = win32com.client.Dispatch("REPORTOCX.ReportOcxCtrl.1")
> > >>>
> >
> > (I'm running 2.1.2 btw.)  When I try to use it's methods all I get is:
> >
> > >>> o.AboutBox()
> > Traceback (most recent call last):
> >   File "<interactive input>", line 1, in ?
> >   File "win32com\gen_py\E80AB8D2-EB1E-49EE-8C2A-57BB2E70302Ax0x1x0.py",
> line
> > 29, in AboutBox
> >     return self._oleobj_.InvokeTypes(0xfffffdd8, LCID, 1, (24, 0), (),)
> > com_error: (-2147418113, 'Unexpected failure', None, None)
> > >>>
>
> Looks like it is missing a DLL, probably one of MFC's (If it is on a bare
> bones Win95, there are more interesting modes of failure). Can't you
> inspect, what your OCX is trying to import?

Ur.  What's the Win32 equivalent of ldd (dumps linked dll's)?  I ran strings
on it and filtered out the .dll files (case-insensitive match) and came up
with this:

MFC42.DLL    <- got that
MSVCRT.dll   <- got that
KERNEL32.dll <- got that (system wouldn't run without it, right?)
USER32.dll   <- ditto
GDI32.dll    <- ditto
OLEAUT32.dll <- got that

What might I be missing?

I tried some other ActiveX controls with reasonable success, but a few
showed
the same error message.






From jason at jorendorff.com  Thu Feb 14 18:17:49 2002
From: jason at jorendorff.com (Jason Orendorff)
Date: Thu, 14 Feb 2002 17:17:49 -0600
Subject: Convert String to Dictionary question
In-Reply-To: <a4hacl$argt$1@ID-98166.news.dfncis.de>
Message-ID: <HFEKILOLEFEFMKAECNDLGEKLDCAA.jason@jorendorff.com>

MDK wrote:
> Pickling seems too slow.

Really?  What makes you say this?

In some random tests on my box, it seems that 
repr() is a teensy bit faster than cPickle.dumps(), and
eval() is several times slower than cPickle.loads().

## Jason Orendorff    http://www.jorendorff.com/



From lfini at arcetri.astro.it  Wed Feb  6 04:35:06 2002
From: lfini at arcetri.astro.it (Luca Fini)
Date: Wed, 6 Feb 2002 10:35:06 +0100
Subject: Saving Browser State without Cookies
In-Reply-To: <3c603127.18127765@news.t-online.de>
References: <mailman.1012938609.3651.python-list@python.org>
 <3c603127.18127765@news.t-online.de>
Message-ID: <Pine.LNX.4.44.0202061025240.4517-100000@localhost.localdomain>

On Tue, 5 Feb 2002, Gerson Kurz wrote:

> On Tue, 5 Feb 2002 12:48:29 -0700, "Orr, Steve" <sorr at rightnow.com>
> wrote:
> > 
> >I need to migrate a web application from PHP to Python and I have a
> >requirement to save state info WITHOUT using cookies where much of the state
> >data MUST not be visible to the end users. The PHP application allows users

I made a DB interface years ago which maintained state from one step to 
the following without using cookies. It was written in perl, but the 
technique has nothing to do with the language used.

It was based on the "HIDDEN" field of HTML Forms. If you define such a 
field in a form generated by your application, it is sent back by the 
browser when the form data are sent as any other field in the same form.

The HIDDEN field is not invisible to the user (he can look at it by 
showing the HTML source) but neither are cookies, so you solve the problem 
exactly in the same way as you do for cookies:  you might encode 
the status information as you like, perhaps using criptography.

I hope this is helpful,
					l.f.
 -- 
--------------------------------------------------------------------------
       -- )     Luca Fini                            Tel: +39 055 2752 307
 ___    |\      Osservatorio Astrofisico di Arcetri  Fax: +39 055 2752 292
/   |   | |-_   L.go E.Fermi, 5 +-----------------------------------------
(___|___//___)  50125 Firenze  /   WWW: http://www.arcetri.astro.it/~lfini
 (_)      (_)   Italia        / e-mail: lfini at arcetri.astro.it
-----------------------------+--------------------------------------------